Topical NSAIDs for acute musculoskeletal pain in adults
Transcript of Topical NSAIDs for acute musculoskeletal pain in adults
Topical NSAIDs for acute musculoskeletal pain in adults(Review)
Derry S Moore RA Gaskell H McIntyre M Wiffen PJ
This is a reprint of a Cochrane review prepared and maintained by The Cochrane Collaboration and published in The Cochrane Library2015 Issue 6
httpwwwthecochranelibrarycom
Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
T A B L E O F C O N T E N T S
1HEADER 1ABSTRACT 2PLAIN LANGUAGE SUMMARY 4SUMMARY OF FINDINGS FOR THE MAIN COMPARISON 6BACKGROUND 7OBJECTIVES 7METHODS
10RESULTS Figure 1 11Figure 2 13Figure 3 14Figure 4 16Figure 5 18Figure 6 20
25DISCUSSION 27AUTHORSrsquo CONCLUSIONS 28ACKNOWLEDGEMENTS 28REFERENCES 36CHARACTERISTICS OF STUDIES
102DATA AND ANALYSES Analysis 11 Comparison 1 Individual NSAID versus placebo Outcome 1 Clinical success 104Analysis 12 Comparison 1 Individual NSAID versus placebo Outcome 2 Local adverse events 106Analysis 21 Comparison 2 Diclofenac versus placebo (effect of formulation) Outcome 1 Clinical success 108Analysis 31 Comparison 3 Ibuprofen versus placebo (effect of formulation) Outcome 1 Clinical success 109Analysis 41 Comparison 4 Ketoprofen versus placebo Outcome 1 Clinical success 110Analysis 51 Comparison 5 All topical NSAIDs versus placebo Outcome 1 Local adverse events 111Analysis 52 Comparison 5 All topical NSAIDs versus placebo Outcome 2 Systemic adverse events 114Analysis 53 Comparison 5 All topical NSAIDs versus placebo Outcome 3 Adverse event withdrawals 116Analysis 61 Comparison 6 Topical NSAID versus active comparator Outcome 1 Clinical success - topical piroxicam vs
topical indomethacin 118Analysis 62 Comparison 6 Topical NSAID versus active comparator Outcome 2 Local adverse events - topical piroxicam
vs topical indomethacin 119119APPENDICES 142FEEDBACK 144WHATrsquoS NEW 144HISTORY 145CONTRIBUTIONS OF AUTHORS 145DECLARATIONS OF INTEREST 146SOURCES OF SUPPORT 146DIFFERENCES BETWEEN PROTOCOL AND REVIEW 146INDEX TERMS
iTopical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
[Intervention Review]
Topical NSAIDs for acute musculoskeletal pain in adults
Sheena Derry1 R Andrew Moore1 Helen Gaskell2 Mairead McIntyre3 Philip J Wiffen1
1Pain Research and Nuffield Department of Clinical Neurosciences (Nuffield Division of Anaesthetics) University of Oxford OxfordUK 2Department of Clinical Geratology Oxford University Hospitals NHS Trust John Radcliffe Hospital Oxford UK 3West HoeSurgery Plymouth UK
Contact address Sheena Derry Pain Research and Nuffield Department of Clinical Neurosciences (Nuffield Division of Anaesthetics)University of Oxford Pain Research Unit Churchill Hospital Oxford Oxfordshire OX3 7LE UK sheenaderryndcnoxacuk
Editorial group Cochrane Pain Palliative and Supportive Care GroupPublication status and date Edited (no change to conclusions) comment added to review published in Issue 11 2015Review content assessed as up-to-date 4 February 2015
Citation Derry S Moore RA Gaskell H McIntyre M Wiffen PJ Topical NSAIDs for acute musculoskeletal pain in adults CochraneDatabase of Systematic Reviews 2015 Issue 6 Art No CD007402 DOI 10100214651858CD007402pub3
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
A B S T R A C T
Background
Use of topical NSAIDs to treat acute musculoskeletal conditions has become widely accepted because they can provide pain reliefwithout associated systemic adverse events This review is an update of rsquoTopical NSAIDs for acute pain in adultsrsquo originally publishedin Issue 6 2010
Objectives
To determine the efficacy and safety of topically applied NSAIDs in acute musculoskeletal pain in adults
Search methods
We searched the Cochrane Register of Studies Online MEDLINE and EMBASE to February 2015 We sought unpublished studies byasking personal contacts and searching online clinical trial registers and manufacturers websites For the earlier review we also searchedour own in-house database and contacted manufacturers
Selection criteria
We included randomised double-blind active or placebo (inert carrier)-controlled trials in which treatments were administered toadults with acute pain resulting from strains sprains or sports or overuse-type injuries (twisted ankle for instance) There had to be atleast 10 participants in each treatment arm with application of treatment at least once daily
Data collection and analysis
Two review authors independently assessed studies for inclusion and extracted data We used numbers of participants achieving eachoutcome to calculate the risk ratio and numbers needed to treat for an additional beneficial outcome (NNT) or additional harmfuloutcome (NNH) compared with placebo or other active treatment We reported 95 confidence intervals (CI) We were particularlyinterested to compare different formulations (gel cream plaster) of individual NSAIDs
1Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Main results
For this update we added 14 new included studies (3489 participants) and excluded four studies We also identified 20 additionalreports of completed or ongoing studies that have not been published in full The earlier review included 47 studies
This update included 61 studies Most compared topical NSAIDs in the form of a gel spray or cream with a similar topical placebo5311 participants were treated with a topical NSAID 3470 with placebo and 220 with an oral NSAID This was a 63 increase inthe number of included participants over the previous version of this review We also identified a number of studies in clinical trialregistries with unavailable results amounting to about 5900 participants for efficacy and 5300 for adverse events
Formulations of topical diclofenac ibuprofen ketoprofen piroxicam and indomethacin demonstrated significantly higher rates ofclinical success (more participants with at least 50 pain relief ) than matching topical placebo (moderate or high quality data)Benzydamine did not Three drug and formulation combinations had NNTs for clinical success below 4 For diclofenac the Emulgelregformulation had the lowest NNT of 18 (95 CI 15 to 21) in two studies using at least 50 pain intensity reduction as the outcomeDiclofenac plasters other than Flectorreg also had a low NNT of 32 (26 to 42) based on good or excellent responses in some studiesKetoprofen gel had an NNT of 25 (20 to 34) from five studies in the 1980s some with less well defined outcomes Ibuprofen gelhad an NNT of 39 (27 to 67) from two studies with outcomes of marked improvement or complete remission All other drug andformulation combinations had NNT values above 4 indicating lesser efficacy
There were insufficient data to compare reliably individual topical NSAIDs with each other or the same oral NSAID
Local skin reactions were generally mild and transient and did not differ from placebo (high quality data) There were very few systemicadverse events (high quality data) or withdrawals due to adverse events (low quality data)
Authorsrsquo conclusions
Topical NSAIDs provided good levels of pain relief in acute conditions such as sprains strains and overuse injuries probably similar tothat provided by oral NSAIDs Gel formulations of diclofenac (as Emugelreg) ibuprofen and ketoprofen and some diclofenac patchesprovided the best effects Adverse events were usually minimal
Since the last version of this review the new included studies have provided additional information In particular information ontopical diclofenac is greatly expanded The present review supports the previous review in concluding that topical NSAIDs are effectivein providing pain relief and goes further to demonstrate that certain formulations mainly gel formulations of diclofenac ibuprofenand ketoprofen provide the best results Large amounts of unpublished data have been identified and this could influence results inupdates of this review
P L A I N L A N G U A G E S U M M A R Y
Topical non-steroidal anti-inflammatory drugs for acute musculoskeletal pain in adults
Acute musculoskeletal pain describes conditions like a sprained ankle or a muscle pull These usually get better over two or three weekswithout treatment but can be very painful while they last
Topical non-steroidal anti-inflammatory drugs (NSAIDs) are applied to unbroken skin where it hurts as gels creams sprays or plastersTopical NSAIDs penetrate the skin enter tissues or joints and reduce processes causing pain in the tissue Drug levels in the bloodwith topical NSAIDs are very much lower than with the same drug taken by mouth This minimises the risk of harmful effects
We searched medical databases for clinical trials comparing topical NSAIDs with placebo (creams or gels that do not contain a medicine)or other medicines in adults aged 16 years or older with musculoskeletal pain (typically sports injuries) The evidence is current toFebruary 2015
This review is an update of rsquoTopical NSAIDs for acute pain in adultsrsquo originally published in Issue 6 2010 We identified 14 new studiesto add to the 47 studies included in the earlier review We also identified 14 studies in a clinical trial registry that are completed andthree short reports from meetings for which we could not find full details (about 4500 participants) Three more studies are ongoing(almost 900 participants)
The 61 included studies involving 8386 participants were generally of high-quality They tested a number of different topical drugsmostly against a topical placebo (carrier without the NSAID) with application at least once a day We were interested in participants
2Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
having good pain reduction (by about half ) around seven days after treatment started At later times most people are expected to getbetter even without treatment
We looked at particular formulations of individual drugs Gel formulations of diclofenac and ketoprofen were among the most effectivealong with ibuprofen gel and diclofenac plaster For diclofenac and ketoprofen gels 7 or 8 people out of 10 with a painful strain sprainor muscle pull had much reduced pain after seven days compared with only 2 or 3 out of 10 with placebo (high quality data) OtherNSAIDs and formulations were better than placebo but not by as much Because both topical NSAIDs and topical placebo are rubbedinto the skin in these studies we know that any effect is not just from rubbing
About 1 in 20 people experienced a mild and short-lived side effect like redness at the application site This was the same for topicalNSAID and topical placebo (high quality data) Side effects like a stomach upset or feeling sick were uncommon with no differencebetween topical NSAID and topical placebo (high quality data) There were no serious side effects
3Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
S U M M A R Y O F F I N D I N G S F O R T H E M A I N C O M P A R I S O N [Explanation]
Topical NSAIDs compared with topical placebo for acute musculoskeletal pain in adults
Patient or population adults with strains sprains or muscle pullSettings communityIntervention topical NSAID (topical diclofenac ibuprofen and ketoprofen gels only shown here for efficacy)Comparison topical placebo
Outcomes Probable outcome withintervention
Probable outcome withcomparator
RR NNT NNTp or NNH(95 CI)
No of studies partici-pants
Quality of the evidence(GRADE)
Comments
Topical diclofenac gel
(as Emulgel)
Clinical success (eg50 reduction in pain)
780 in 1000 200 in 1000 RR34 (27 to 55)NNT18 (15 to 21)
2 studies314 participants
High Consistent results in 2moderately sized recentstudies of high quality
Topical ibuprofen gel
Clinical success (eg
50 reduction in pain)
420 in 1000 160 in 1000 RR27 (17 to 42)NNT39 (27 to 67)
2 studies241 participants
Moderate Modest effect size andnumbers of participants
Topical ketoprofen gel
Clinical success (eg
50 reduction in pain)
720 in 1000 330 in 1000 RR22 (17 to 28)NNT25 (20 to 34)
5 studies348 participants
Moderate Modest effect size andnumbers of participantsbut studies small withnone recent
All topical NSAIDs
Local adverse events
46 in 1000 50 in 1000 RR10 (080 to 12)NNH not calculated
42 studies6125 participants
High Large number of studiesand participants with con-sistent results
All topical NSAIDs
Systemic adverse
events
32 in 1000 35 in 1000 RR10 (07 to 13)NNH not calculated
38 studies5372 participants
High Large number of studiesand participants with con-sistent results
All topical NSAIDs
Withdrawals - adverse
events
11 in 1000 11 in 1000 RR10 (07 to 17)NNH not calculated
42 studies5790 participants
High Large number of studiesand participants with con-sistent results4
To
pica
lN
SA
IDs
for
acu
tem
uscu
losk
ele
talp
ain
inad
ults
(Revie
w)
Co
pyrig
ht
copy2015
Th
eC
och
ran
eC
olla
bo
ratio
nP
ub
lished
by
Joh
nW
iley
ampS
on
sL
td
Serious adverse events 1 in total 0 in total Not calculated All data Low Small numbers of events
GRADE Working Group grades of evidenceHigh quality Further research is very unlikely to change our confidence in the estimate of effectModerate quality Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimateLow quality Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimateVery low quality We are very uncertain about the estimate
CI confidence interval RR risk ratio NNT number needed to treat for an additional beneficial outcome NNTp number needed to treatto prevent an event happening NNH number needed to treat for an additional harmful outcome
xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
5To
pica
lN
SA
IDs
for
acu
tem
uscu
losk
ele
talp
ain
inad
ults
(Revie
w)
Co
pyrig
ht
copy2015
Th
eC
och
ran
eC
olla
bo
ratio
nP
ub
lished
by
Joh
nW
iley
ampS
on
sL
td
B A C K G R O U N D
This review is an update of a review originally published in Issue 62010 on rsquoTopical NSAIDs for acute pain in adultsrsquo (Massey 2010)We have changed the title to specify musculoskeletal pain becausetopical NSAIDs are not normally used to treat visceral pain orheadache We felt that the new title better reflected the content ofthe review
The use of topical NSAIDs for pain relief has been a controversialsubject in analgesic practice In some parts of the world (includingmuch of Western Europe) they have been available for many yearsare widely available without prescription widely advertised usedextensively and evidence for their use is considered adequate Inother parts of the world they were regarded as little more thanplacebo with any apparent effect attributed to the process of rub-bing at the site of the affected area In some places (for examplethe US) their use was almost unknown until the mid-2010s InEngland 52 million prescriptions for topical NSAIDs were dis-pensed in 2013 (PACT 2014) mainly for formulations of ibupro-fen (245 million) piroxicam (118 million) and diclofenac (127million)
There is good evidence for the efficacy of topical NSAIDs in acuteand chronic musculoskeletal pain (Mason 2004a Mason 2004bMoore 1998a) In the US the Food and Drug Administration li-censed topical nonsteroidal products in 2007 and in England theNational Institute for Health and Care Excellence (NICE) recom-mended topical therapies as first line treatment in its guidelinesfor osteoarthritis in 2008 (NICE 2008) Earlier reviews of topicalanalgesics covered studies investigating the underlying science toexplain biological plausibility in addition to clinical trials (Anon2005 Moore 2008a)
This review is one of a series on topical analgesics including topicalcapsaicin at low and high doses (Derry 2012a Derry 2013) andtopical NSAIDs in chronic pain conditions (Derry 2012b) andsalicylate-containing rubefacients (Derry 2014)
Description of the condition
Acute pain is usually defined as pain of less than three monthsrsquo du-ration It is often associated with injury including trauma surgerymusculoskeletal injuries such as strains sprains and over-use in-juries or soft tissue injuries such as muscle soreness or cramps
Description of the intervention
Clinicians prescribe NSAIDs on a routine basis for a range of mildto moderate pain NSAIDs are the most commonly prescribedanalgesic medications worldwide and their efficacy for treatingacute pain has been well demonstrated (Moore 2003) They re-versibly inhibit cyclooxygenase (prostaglandin endoperoxide syn-thase) the enzyme mediating production of prostaglandins and
thromboxane A2 (FitzGerald 2001) Prostaglandins mediate a va-riety of physiological functions such as maintenance of the gastricmucosal barrier regulation of renal blood flow and regulation ofendothelial tone They also play an important role in inflamma-tory and nociceptive processes However relatively little is knownabout the mechanism of action of this class of compounds asidefrom their ability to inhibit cyclooxygenase-dependent prostanoidformation (Hawkey 1999)NSAIDs taken orally or intravenously are transported to all partsof the body in the blood and relatively high blood concentrationsare needed to achieve effective tissue concentrations at the site ofthe pain and inflammation These high concentrations throughoutthe body can give rise to a number of adverse events that canbe unpleasant (for example dyspepsia) or potentially serious (forexample gastrointestinal bleeding)A topical medication is a one applied to body surfaces such asthe skin or mucous membranes to treat ailments A large rangeof types of topical formulation may be used including but notlimited to creams foams gels lotions ointments and plastersThe exact formulation of a topical medication is often determinedby the speed of drug absorption required The need may be forslow absorption into the circulation to maintain low drug concen-trations and perhaps avoiding extensive first pass metabolism inthe liver plasters containing drug reservoirs may be used for thisas with transdermal opioids or contraceptive steroids For rapidabsorption the formulation is enhanced by substances to improveor assist skin penetration perhaps only to generate high concen-trations in tissues rather than in the blood gel formulations areuseful for this purpose which is why they are sometimes used fortopical NSAIDs
Topical NSAIDs
Topical NSAIDs are formulated for direct application to thepainful site and to produce a local pain-relieving effect whileavoiding body-wide distribution of the drug at physiologically ac-tive levels (McPherson 2013) This method of application (dos-ing) necessarily limits their use to more superficial painful condi-tions such as sprains strains and muscle or tendon soreness Theywould not for example be indicated for deep visceral pain orheadaches They are also not appropriate for use on broken skinso would not be used on open wounds (accidental or surgical)
How the intervention might work
For a topical formulation to be effective it must first penetratethe skin Only when the drug has entered the lower layers ofthe skin can it be absorbed by the blood and transported to thesite of action or penetrate deeper into areas where inflammationoccurs Individual drugs have different degrees of penetration Abalance between lipid and aqueous solubility is needed to optimisepenetration and use of prodrug esters has been suggested as a way
6Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
of enhancing permeability Formulation is also crucial to goodskin penetration Experiments with artificial membranes or humanepidermis suggest that creams are generally less effective than gelsor sprays but newer formulations such as microemulsions mayhave greater potentialOnce the drug has reached the site of action it must be presentat a sufficiently high concentration to inhibit cyclooxygenase en-zymes thereby reducing prostaglandin synthesis This in turn re-duces inflammation and relieves pain It is probable that in acuteconditions topical NSAIDs exert their action primarily by lo-cal reduction of symptoms independent of any systemic uptakeand delivery Tissue levels of NSAIDs applied topically certainlyreach levels high enough to inhibit cyclooxygenase-2 (Anon 2005Haroutiunian 2010 Moore 2008a) However plasma concentra-tions found after topical administration are only a fraction (usu-ally much less than 5) of the levels found in plasma followingoral administration Topical application can potentially limit sys-temic adverse events by increasing local effects and minimisingsystemic concentrations of the drug We know that the incidenceof upper gastrointestinal bleeding is low with chronic use of topi-cal NSAIDs (Evans 1995) although it has been reported partic-ularly in people with risk factors (Zimmerman 1995) We haveno certain knowledge of lower effects on cardiovascular events orrenal failure both of which have been associated with oral NSAIDuse
Why it is important to do this review
Since the last review in 2010 a number of new studies have beenpublished nearly all of which investigated various formulations ofdiclofenac These new studies are generally of higher quality thanmany of the earlier ones in this review and have the potential toinfluence the strength of its conclusions substantially Moreoverthe additional information allows for analysis based not only on aparticular drug but also on the formulation of that drug This canprovide better insight into whether formulation affects efficacy oftopical NSAIDs in acute musculoskeletal painAn updated review of evidence for topical NSAIDs was needed toinform choices made by consumers prescribers and commission-ers (purchasers of healthcare)
O B J E C T I V E S
To determine the efficacy and safety of topically applied NSAIDsin acute musculoskeletal pain in adults
M E T H O D S
Criteria for considering studies for this review
Types of studies
Randomised controlled double-blind studies comparing topicalNSAIDs with placebo (inert carrier) or other active treatment foracute pain with at least 10 participants per treatment arm andoutcomes close to seven days (minimum three days) We excludedstudies published only as short abstracts (which report insufficientdata to assess methods) or studying experimentally induced pain(which does not correlate well with clinical pain) Because a cross-over design is not appropriate for self limiting conditions such assprains strains and contusions we only considered parallel-groupdesigns
Types of participants
Adults (aged 16 years or more) with acute musculoskeletal pain ofat least moderate intensity resulting mainly from strains sprainsor sports injuries Typically for sports injuries the injury wouldhave occurred within 24 or 48 hours
Types of interventions
Included studies had at least one treatment arm using a topicalNSAID and a comparator arm using placebo (inert carrier withoutNSAID or other active treatment) The topical NSAID had to beapplied at least once daily We did not include salicylates in thisreview as they are no longer classified as topical NSAIDs and arecovered in a separate review (Derry 2014)
Types of outcome measures
We sought information on participant characteristics includingage sex and condition treated
Primary outcomes
The primary outcome was rsquoclinical successrsquo defined as at least a50 reduction in pain or equivalent measure such as a rsquovery goodrsquoor rsquoexcellentrsquo global assessment of treatment or rsquononersquo or rsquoslightrsquopain on rest or movement measured on a categorical scale (Moore1998a) We used the following hierarchy of outcomes to extractdata for the primary outcome
bull Participant reported reduction in pain of at least 50bull Participant reported global assessment of treatmentbull Pain on movementbull Pain at rest or spontaneous painbull Undefined rsquoimprovementrsquo
We used only participant reported outcomes of efficacy and notphysician or investigator reported outcomes
7Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Secondary outcomes
bull Numbers of participants with adverse events local andsystemic
bull Numbers of withdrawals all cause lack of efficacy andadverse events
We anticipated that outcomes would be reported after differentdurations of treatment and extracted data reported as close toseven days as possible with a minimum of three days We also ex-tracted data for outcomes reported after longer durations of treat-ment We anticipated that reporting of adverse events would varybetween studies with regard to the terminology used method ofascertainment and categories reported (for example occurring inat least 5 of participants or where there is a statistically signifi-cant difference between treatment groups) We took care to iden-tify these details where relevant
Search methods for identification of studies
Electronic searches
We searched the following databases without language restrictionbull CENTRAL (The Cochrane Library) Issue 4 2009 for the
original review and the Cochrane Register of Studies Online(CRSO) to 3 February 2015 for this update
bull MEDLINE (via Ovid) from inception to December 2009for the original review and from 2008 to 3 February 2015 forthis update
bull EMBASE (via Ovid) from inception to December 2009for the original review and from 2008 to 3 February 2015 forthis update
bull Oxford Pain Relief Database for the original review (Jadad1996a) This resource is no longer being updated
See Appendix 1 for the CENTRAL search strategy Appendix 2 forthe MEDLINE search strategy and Appendix 3 for the EMBASEsearch strategy
Searching other resources
We searched the reference lists of review articles and includedstudies We have previously asked manufacturers for details ofunpublished studies but did not make new requestsWe searched two clinical trial registries (clinicaltrialsgov (clinicaltrialsgov) and the World Health Organization In-ternational Clinical Trials Registry Platform (appswhointtrialsearch)) and asked personal contacts about ongoing and un-published studies
Data collection and analysis
Selection of studies
We screened the titles and abstracts of studies identified by thesearches to eliminate those that clearly did not satisfy the inclu-sion criteria and obtained full reports of the remaining studies todetermine inclusion in the review
Data extraction and management
Review authors were not blinded to the authorsrsquo names and in-stitutions journal of publication or study results at any stage ofthe review Two review authors independently selected the stud-ies for inclusion assessed methodological quality and risk of biasand extracted data We resolved disagreements and uncertaintiesthrough discussionWe abstracted information on participants interventions and out-comes from the original reports into a standard data extractionform One review author entered data suitable for meta-analysisinto Review Manager 5 (RevMan 2014) and another review au-thor checked it
Assessment of risk of bias in included studies
We used the Oxford Quality Score as the basis for inclusion lim-iting inclusion to studies that were randomised and double-blindas a minimum (Jadad 1996b)Two review authors independently assessed risk of bias for eachstudy using the criteria outlined in the Cochrane Handbook forSystematic Reviews of Interventions with any disagreements resolvedby discussion (Higgins 2011) We assessed the following for eachstudy
bull Random sequence generation (checking for possibleselection bias) We assessed the method used to generate theallocation sequence as low risk of bias (any truly randomprocess for example random number table computer randomnumber generator) unclear risk of bias (method used to generatesequence was not clearly stated) We excluded studies using anon-random process that were therefore at high risk of bias (forexample odd or even date of birth hospital or clinic recordnumber)
bull Allocation concealment (checking for possible selectionbias) The method used to conceal allocation to interventionsbefore assignment determines whether intervention allocationcould have been foreseen in advance of or during recruitmentor changed after assignment We assessed the methods as lowrisk of bias (for example telephone or central randomisationconsecutively numbered sealed opaque envelopes) unclear riskof bias (method was not clearly stated) We excluded studies thatdid not conceal allocation and were therefore at high risk of bias(for example open list)
8Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
bull Blinding of outcome assessment (checking for possibledetection bias) We assessed the methods used to blind studyparticipants and outcome assessors from knowledge of whichintervention a participant received We assessed the methods aslow risk of bias (study stated that it was blinded and describedthe method used to achieve blinding for example identicaltubes containing gel or identical plasters matched in appearanceand smell) unclear risk of bias (study stated that it was blindedbut did not provide an adequate description of how blinding wasachieved) We excluded studies that were not double-blind andwere therefore at high risk of bias
bull Incomplete outcome data (checking for possible attritionbias due to the amount nature and handling of incompleteoutcome data) We assessed the methods used to deal withincomplete data as low risk of bias (less than 10 ofparticipants did not complete the study or used rsquobaselineobservation carried forwardrsquo analysis or both) unclear risk ofbias (used rsquolast observation carried forwardrsquo analysis) or high riskof bias (used rsquocompleterrsquo analysis)
bull Size (checking for possible biases confounded by small size)Small studies have been shown to overestimate treatment effectsprobably due to methodological weaknesses (Dechartres 2013Nuumlesch 2010) We assessed studies as at low risk of bias if theyhad at least 200 participants at unclear risk if they had 50 to 200participants and at high risk if they had fewer than 50participants
Measures of treatment effect
We used risk ratio (RR) to establish statistical difference and num-bers needed to treat for an additional beneficial outcome (NNT)with 95 confidence intervals (CI) We pooled percentages as ab-solute measures of benefit or harmWe used the following terms to describe adverse outcomes in termsof harm or prevention of harm
bull When significantly fewer adverse outcomes occurred withtreatment than with control (placebo or active) we used the termthenumber needed to treat to prevent one event (NNTp)
bull When significantly more adverse outcomes occurred withtreatment compared with control (placebo or active) we usedthe term the number needed to treat for an additional harmfuloutcome or cause one event (NNH)
Unit of analysis issues
Randomisation was to the individual participant
Dealing with missing data
Wherever possible we used intention-to-treat (ITT) analysis wherethe ITT population consists of participants who were randomisedapplied at least one dose of the assigned study medication andprovided at least one post-baseline assessment We assigned miss-ing participants zero improvement
We also looked for information about methods of imputation formissing data
Assessment of heterogeneity
We examined heterogeneity visually using LrsquoAbbeacute plots (LrsquoAbbeacute1987) a visual method for assessing differences in results of indi-vidual studies and with the I2 statistic
Assessment of reporting biases
The aim of this review was to use dichotomous outcomes of knownutility and of value to patients (Moore 2013) The review did notdepend on what the authors of the original studies chose to reportor not Studies that did not report dichotomous results but onlyaverage pain data did not contribute to analysesWe assessed publication bias using a method designed to detect theamount of unpublished data with a null effect required to makeany result clinically irrelevant (usually taken to mean an NNT of10 or higher Moore 2008b)
Data synthesis
We pooled data only for comparisons and outcomes where therewere at least two studies and 200 participants (Moore 1998b)When two active treatment arms were compared with a placeboarm we took care to avoid double counting of participants in theplacebo arm if both active groups contributed to an analysis wesplit the placebo group between themWe calculated RRs with 95 CIs using the fixed-effect model(Morris 1995) A statistically significant benefit of topical NSAIDover control was assumed when the lower limit of the 95 CI ofthe RR was greater than one A statistically significant benefit ofcontrol over active treatment was assumed when the upper limitof the 95 CI was less than one We calculated NNTs with 95CIs using the pooled number of events by the method of Cookand Sackett (Cook 1995)Statistically significant differences between NNTs for differenttopical NSAIDs were tested using the z test (Tramer 1997) wherethere were sufficient data to do so and where the studies weresufficiently similar in types of participant outcome and durationto make such comparisons sensible
Subgroup analysis and investigation of heterogeneity
We carried out separate analyses for individual NSAIDs andwhere the data permitted for different formulations of individualNSAIDs
Sensitivity analysis
The earlier review included sensitivity analyses for various factorsthat are now covered by the assessment of risk of bias
9Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
R E S U L T S
Description of studies
Results of the search
New searches for this update identified 20 publications that wereexamined in further detail to determine inclusion status We alsoidentified 20 additional studies in clinical trial registries See Figure1
10Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Figure 1 Study flow diagram
11Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
We identified 14 new studies (11 publications three unpub-lished reports) satisfying our inclusion criteria (Costantino2011 Coudreuse 2010 Gonzaacutelez de Vega 2013 Hofman2000 Klainguti 2010 Kuehl 2011 Li 2013 NCT01255423NCT01272934 NCT01272947 Predel 2012 Predel 2013aPredel 2013b Saillant 1998) one post hoc analysis of a study thatwas included in the earlier review (Mueller 2010 in Predel 2004)and an additional publication and a pooled analysis that includednew data for another study from the earlier review (Lionberger2011 pooled analysis in Joussellin 2003)
Included studies
All except one of the new studies compared diclofenac withplacebo A number of different formulations were used includ-ing diclofenac epolamine (DHEP) with or without heparin (Flec-toparin Tissugel or Flector EP Tissugel) applied as a plaster (orpatch) diclofenac diethylamine (DDEA) applied as a gel and di-clofenac with lethicin applied as a spray gel The remaining studycompared diclofenac gel with traumeel a ldquofixed combination ofplant and mineral extractsrdquo applied as a gel or an ointmentThere were 47 studies in the original review 14 new studies wereincluded making a total of 61 studies in this updated reviewAll used a parallel group design Forty-four compared a topicalNSAID with placebo 13 a topical NSAID with an active com-parator (a different topical NSAID an oral NSAID the same top-ical NSAID in a different formulation or a compound of plantand mineral extracts) and four had both placebo and active com-parators In total 5311 participants were treated with a topicalNSAID 3470 with placebo and 220 with an oral NSAID Top-ical NSAIDs used were benzydamine diclofenac etofenamatefelbinac fentiazac flunoxaprophen flurbiprofen ibuprofen in-domethacin ketoprofen ketorolac lysine clonixinate meclofe-namic acid naproxen niflumic acid and piroxicam They wereapplied as creams gels sprays foams or plasters (patches) Topicalplacebos were the inert carriers without the active NSAID OralNSAIDs used were ibuprofen (as tablets) and indomethacin (ascapsules)Most studies enrolled participants who had sprains strains andcontusions usually as a result of sports injuries and treatmentwas started within a few hours or days Other studies enrolledparticipants with overuse-type injuries such as tendinitis and acutelow back pain where pain had been present for days or weeks butless than three monthsParticipants were treated for at least five days and up to threeweeks with most studies lasting seven to 14 days Participantswere usually assessed in clinic at intervals during treatment andsometimes also at home using daily patient diaries We used out-comes closest to seven days because many of these injuries are self
limiting with differences between active treatment and placebobeing diminished or lost after longer intervalsMost studies reported dichotomous outcomes suitable for a re-sponder analysis although group mean change (for pain or phys-ical function for example) was usually the primary outcomesHowever the definition of response varied both in the parametermeasured (for example pain pain on movement patient globalevaluation of treatment) and in the scale used to measure it (forexample a 3- 4- or 5-point scale for patient global evaluation)Details of included studies are in the Characteristics of includedstudies tableWe identified 14 completed but apparently unpublished studies ina clinical trial registry for which no results have been posted (4403participants NCT00351104 NCT00352625 NCT00426985NCT00640705 NCT00640939 NCT00680472NCT00680784 NCT00765700 NCT00869063NCT00869180 NCT00931866 NCT01874626NCT01957215 NCT02324270) We have placed these underCharacteristics of studies awaiting classification We also identi-fied three conference abstracts that relate to completed studies thatdo not appear to have been published but that may satisfy ourinclusion criteria One is likely to be the same study as one of theincluded studies identified in a clinical trial registry (Pallay 2013in NCT01272947) one relates to another study identified in theclinical trial registry that is awaiting classification (Ekman 2010in NCT00765700) while we could find no published or unpub-lished reports of the other Sarzi-Puttini 2014)We also identified three ongoing studies with an estimated en-rolment of 880 participants (NCT01945034 NCT02100670NCT02290821) Details are in the Characteristics of ongoingstudies table
Excluded studies
For the original review 25 studies were excluded after readingthe full paper For this update we excluded four new studies (Cesarone 2008 Coulibaly 2009 Kuwabara 2013 Vinciguerra2008) Details are in the Characteristics of excluded studies table
Risk of bias in included studies
All studies were randomised and double-blind One study scored25 (Sinniger 1981) 23 scored 35 23 scored 45 and 14 scored55 for methodological quality using the Oxford Quality Scale Abreakdown of the scores for individual studies is reported in theCharacteristics of included studies tableComments on potential biases in individual studies are reported inthe rsquoRisk of biasrsquo section of the Characteristics of included studiestable The findings are displayed in Figure 2 and Figure 3
12Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Figure 2 Risk of bias summary review authorsrsquo judgements about each risk of bias item for each includedstudy
13Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Figure 3 Methodological quality graph review authorsrsquo judgements about each methodological quality
item presented as percentages across all included studies
Allocation
All the studies were randomised but only 17 adequately describedthe method used to generate the random sequence Thirteen stud-ies adequately described the method used to conceal allocation ofthe sequence No studies were at high risk of bias for this item
Blinding
All studies were double-blind and 38 adequately described themethod used to maintain the blinding No studies were at highrisk of bias for this item
Incomplete outcome data
Thirty-six studies included all participants in the primary analysisor provided sufficient data to allow missing participants to be in-cluded as non-responders and were judged at low risk of bias Wejudged four studies to be at high risk of attrition bias (Campbell1994 Kuehl 2011 Maziegraveres 2005a Russell 1991) Three unpub-lished studies contributed only to adverse event analyses and ac-counted for all participants for these outcomes (NCT01255423NCT01272934 NCT01272947)
Other potential sources of bias
We judged one study that included more than 200 participantsin each treatment arm to be at low risk of bias from size butthis study had a very high attrition rate (see rsquoIncomplete outcome
data (attrition bias)rsquo) and did not report all the efficacy outcomesmeasured (Kuehl 2011) We judged 27 studies to be at high riskbecause they included fewer than 50 participants per treatmentarm
Effects of interventions
See Summary of findings for the main comparison TopicalNSAIDs compared with topical placebo for acute musculoskeletalpain in adultsThree studies did not contribute data suitable for analysis of at leastone outcome (Gonzaacutelez de Vega 2013 Gualdi 1987 Hoffmann2012)
1 Topical NSAID versus placebo
Details of efficacy outcomes in individual studies are in Appendix4 and of adverse events and withdrawals in Appendix 5 Appendix6 has details of the concentration of topical products the amountapplied the frequency of application and an estimation of thedaily dose of topical NSAID applied Not all studies providedsufficient information to allow calculation of daily dose appliedFor example for topical diclofenac the estimated doses appliedvaried between about 60 and 280 mg for topical ketoprofen 100to about 450 mg for topical ibuprofen 300 to 800 mg
Participants with clinical success
14Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Topical diclofenac versus placebo
Ten studies contributed to this analysis (Coudreuse 2010Joussellin 2003 Klainguti 2010 Li 2013 Predel 2004 Predel2012 Predel 2013a Predel 2013b Rowbotham 2003) of whichone (Predel 2012) had two active treatment arms A total of 1074participants were treated with topical diclofenac and 976 withplacebo (Analysis 11 Figure 4)
15Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Figure 4 Forest plot of comparison 2 Individual NSAID versus placebo outcome 21 Clinical success
16Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
bull The proportion of participants experiencing successfultreatment with topical diclofenac was 74 (8001074 range39 to 100)
bull The proportion of participants experiencing successfultreatment with placebo was 47 (461976 range 8 to 82)
bull The RR for treatment compared with placebo was 16(95 CI 15 to 17)
bull The NNT for successful treatment was 37 (32 to 43) Forevery four participants treated with topical diclofenac one wouldexperience successful treatment who would not have done sowith placebo
Effect of formulation
The effects of formulation are shown in Analysis 21 and Figure5
17Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Figure 5 LrsquoAbbeacute plot of clinical success in studies of topical diclofenac versus topical placebo The size ofthe symbol is proportional to the size of the study (inset scale) Dark blue Emulgel light blue spraygel red
Flector pink other patch or plaster
bull Four studies used a Flectorreg plaster (1030 participantsJoussellin 2003 Li 2013 Rowbotham 2003 Saillant 1998) TheRR for treatment compared with placebo was 15 (14 to 17)and the NNT was 47 (37 to 65)
bull Three studies used other makes of plaster (474 participantsCoudreuse 2010 Klainguti 2010 Predel 2004) The RR fortreatment compared with placebo was 16 (14 to 18) and theNNT was 32 (26 to 42)
bull Two studies used Voltaren Emulgel (314 participants Predel2012 Predel 2013b) The RR for treatment compared withplacebo was 38 (27 to 55) and the NNT was 18 (15 to 21)
bull One study used a spray gel (232 participants Predel2013a) The RR for treatment compared with placebo was 12(105 to 13) and the NNT was 80 (48 to 24)
Diclofenac as the gel formulation Emulgel was statistically more ef-ficacious than the plaster formulation as Flector plaster (z = 6360P value lt 000001)
Topical ibuprofen versus placebo
Five studies contributed to this analysis (Billigmann 1996Campbell 1994 Dreiser 1988 Machen 2002 Ramesh 1983) Atotal of 218 participants were treated with topical ibuprofen and218 with placebo (Analysis 11)
bull The proportion of participants experiencing successfultreatment with topical ibuprofen was 55 (120218 range 31to 81)
18Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
bull The proportion of participants experiencing successfultreatment with placebo was 33 (73218 range 13 to 76)
bull The RR of treatment compared with placebo was 16 (13to 20)
bull The NNT for successful treatment was 46 (33 to 80) Forevery five participants treated with topical ibuprofen one wouldexperience successful treatment who would not have done sowith placebo
Effect of formulation
The effects of formulation are shown in Analysis 31bull Three studies used cream formulations (195 participants
Campbell 1994 Dreiser 1988 Ramesh 1983) Although this isjust below our threshold for pooled analysis we have includedthis analysis for completeness and the results should beinterpreted with caution The RR for treatment compared withplacebo was 13 (103 to 16) and the NNT was 64 (34 to 41)
bull Two studies used gel formulations (241 participantsBilligmann 1996 Machen 2002) The RR for treatmentcompared with placebo was 27 (17 to 42) and the NNT was39 (27 to 67)
There was no statistically significant difference between the geland cream formulations (z = 1160 P value = 0246)
Topical ketoprofen versus placebo
Seven studies contributed to this analysis (Airaksinen 1993Dreiser 1989 Julien 1989 Kockelbergh 1985 Maziegraveres 2005bMaziegraveres 2005a Noret 1987) A total of 346 participants weretreated with topical ketoprofen and 337 with placebo (Analysis11)
bull The proportion of participants experiencing successfultreatment with topical ketoprofen was 73 (251346 range57 to 89)
bull The proportion of participants experiencing successfultreatment with placebo was 47 (157337 range 17 to 73)
bull The RR of treatment compared with placebo was 16 (14to 18)
bull The NNT for successful treatment was 39 (30 to 53) Forevery four participants treated with topical ketoprofen onewould experience successful treatment who would not have doneso with placebo
Effect of formulation
The effects of formulation are shown in Analysis 41 and Figure6
19Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Figure 6 LrsquoAbbeacute plot of clinical success in studies of topical ketoprofen versus topical placebo The size ofthe symbol is proportional to the size of the study (inset scale) Light blue ketoprofen gel pink ketoprofen
plaster
Two studies used a plaster formulation (335 participants Maziegraveres2005b Maziegraveres 2005a) The RR for treatment compared withplacebo was 12 (104 to 14) and the NNT was 82 (45 to 47)Five studies used gel formulations (348 participants Airaksinen1993 Dreiser 1989 Julien 1989 Kockelbergh 1985 Noret 1987)The RR for treatment compared with placebo was 22 (17 to 28)and the NNT was 25 (20 to 34)Ketoprofen as a gel formulation was statistically more efficaciousthan a plaster formulation (z = 3860 P value = 000014)
Topical piroxicam versus placebo
Three studies contributed to this analysis (Aoki 1984 Fujimaki1985 Russell 1991) A total of 255 participants were treated withtopical piroxicam and 249 with placebo (Analysis 11)
bull The proportion of participants experiencing successfultreatment with topical piroxicam was 68 (179255 range 53to 79)
bull The proportion of participants experiencing successfultreatment with placebo was 47 (118249 range 45 to 49)
bull The RR of treatment compared with placebo was 15 (13to 17)
bull The NNT for successful treatment was 44 (32 to 69) Forevery four participants treated with topical piroxicam one wouldexperience successful treatment who would not have done sowith placebo
Topical indomethacin versus placebo
20Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Three studies contributed to this analysis ( kermark 1990 Aoki1984 Fujimaki 1985) A total of 168 participants were treatedwith topical indomethacin and 173 with placebo (Analysis 11)
bull The proportion of participants experiencing successfultreatment with topical indomethacin was 58 (97168 range54 to 64)
bull The proportion of participants experiencing successfultreatment with placebo was 46 (79173 range 25 to 49)
bull The RR of treatment compared with placebo was 13 (103to 16)
bull The NNT for successful treatment was 83 (44 to 65) Forevery eight participants treated with topical indomethacin onewould experience successful treatment who would not have doneso with placebo
Topical benzydamine versus placebo
Three studies contributed to this analysis (Chatterjee 1977 Haig1986 Linde 1985) A total of 96 participants were treated withtopical benzydamine and 97 with placebo (Analysis 11) Al-though this is just below our threshold for pooled analysis we haveincluded this analysis for completeness and the results should beinterpreted with caution
bull The proportion of participants experiencing successfultreatment with topical benzydamine was 77 (7496 range70 to 86)
bull The proportion of participants experiencing successfultreatment with placebo was 67 (6597 range 48 to 80)
bull The RR of treatment compared with placebo was 12 (096to 14) There was no statistically significant difference betweentreatments (Figure 4)
Results for participants with clinical success with individual topi-cal NSAIDs where there were adequate data for analysis are sum-marised below in rsquoSummary of results Arsquo and Analysis 11 Analysis21 Analysis 31 and Analysis 41
Summary of results A Participants with clinical success
Comparison Studies Participants NSAID()
Placebo () Relative benefit (95CI)
NNT (95 CI)
Diclofenac -Flector plaster
4 1030 63 41 15 (14 to 17) 47 (37 to 65)
Diclofenac -other plaster
3 474 88 57 16 (14 to 18) 32 (26 to 42)
Diclofenac -Emulgel
2 314 78 20 38 (27 to 55) 18 (15 to 21)
Diclofenac -other gel
1 232 94 82 12 (11 to 13) 80 (48 to 24)
Ibuprofen -cream
3 195 71 56 13 (103 to 16) 64 (34 to 41)
Ibuprofen - gel 2 241 42 16 27 (17 to 42) 39 (27 to 67)
Ketoprofen -plaster
2 335 73 60 12 (104 to 14) 82 (45 to 47)
Ketoprofen - gel 5 348 72 33 22 (17 to 28) 25 (20 to 34)
Piroxicam 3 504 70 47 15 (13 to 17) 44 (32 to 69)
Indomethacin 3 341 58 46 13 (103 to 16) 83 (44 to 65)
21Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Benzydamine 3 193 77 67 12 (096 to 14) not calculated
Results for these two comparisons are derived from very smallamounts of data and are provided here for completeness Theyshould be interpreted with caution
Local adverse events
Local adverse events were irritation of the area to which the topicalNSAID was applied including redness or erythema and itch orpruritus Where reported these were usually described as mild andtransient
All topical NSAIDs versus placebo
Forty-two studies contributed to this analysis of which three com-pared two different drugs with placebo (Aoki 1984 Diebshlag1990 Fujimaki 1985) Three studies had two treatment armscomparing different formulations or application regimens for di-clofenac with placebo which have been combined for this analysis
(Costantino 2011 Klainguti 2010 Predel 2012) In total 3619participants were treated with topical NSAIDs and 3121 withplacebo (Analysis 51)
bull The proportion of participants experiencing a local adverseevent with a topical NSAID was 43 (1553619 range 0 to33)
bull The proportion of participants experiencing a local adverseevent with placebo was 46 (1453121 range 0 to 32)
bull The RR of topical NSAID compared with placebo was 098(080 to 12)
bull There was no significant difference between treatmentgroups so the NNH was not calculated
Individual topical NSAIDs versus placebo
Results for local adverse events with individual topical NSAIDswhere there were adequate data for analysis are in Summary ofresults B and Analysis 12
Summary of results B Participants with local adverse events
Comparison Studies Participants NSAID()
Placebo()
RR(95 CI)
NNH(95 CI)
All NSAIDs 42 6740 43 46 098 (080 to 12) Not calculated
Diclofenac 15 3271 31 43 078 (056 to 11) Not calculated
Ketoprofen 8 852 11 95 12 (083 to 17) Not calculated
Piroxicam 3 522 23 54 042 (017 to 11) Not calculated
Felbinac 3 397 30 15 19 (049 to 75) Not calculated
Indomethacin 3 354 63 22 27 (091 to 77) Not calculated
Ibuprofen 3 321 10 43 23 (098 to 54) Not calculated
Systemic adverse events
All topical NSAIDs versus placebo
Thirty-six studies contributed data on systemic adverse eventsof which three compared two different drugs with placebo (Aoki1984 Diebshlag 1990 Fujimaki 1985) Two studies had two treat-
22Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
ment arms comparing different formulations or application regi-mens for diclofenac with placebo which have been combined forthis analysis (Klainguti 2010 Predel 2012) In total 2956 partic-ipants were treated with a topical NSAID and 2620 with placebo(Analysis 52)
bull Twenty-three studies reported no systemic adverse events inany arm of the study
bull The proportion of participants experiencing a systemicadverse event with a topical NSAID was 31 (922956)
bull The proportion of participants experiencing a systemicadverse event with placebo was 35 (912620)
bull The RR of topical NSAID compared with placebo was 096(073 to 13)
bull There was no significant difference between treatmentgroups so the NNH was not calculated
A further six studies did not report the occurrence or oth-erwise of systemic adverse events (Billigmann 1996 Julien1989 Kockelbergh 1985 Noret 1987 Ramesh 1983 Vecchiet1991) while two studies did not report numbers of participantswith systemic adverse events ( kermark 1990 Auclair 1989)Costantino 2011 reported that there were no systemic gastroin-testinal adverse events Two studies reported only on total adverseevents without distinguishing between local and systemic events(NCT01255423 NCT01272934)
Serious adverse events
Two studies reported serious adverse events In Hoffmann 2012one participant experienced three serious adverse events none ofwhich was judged to be related to the study medication (diclofenacplaster) In NCT01272934 one participant using diclofenac gelruptured the ligaments of the wrist There was no statement aboutlikely relationship to the study medication but this seems unlikely
Adverse event withdrawals
Forty-two studies reported data relating to adverse event with-drawals of which three compared two different drugs with placebo(Aoki 1984 Diebshlag 1990 Fujimaki 1985) Two studies hadtwo treatment arms comparing different formulations or applica-tion regimens for diclofenac with placebo which have been com-bined for this analysis (Klainguti 2010 Predel 2012) In total3365 participants received a topical NSAID and 3040 placebo(Analysis 53)
bull Forty-four comparisons reported no adverse eventwithdrawals
bull The proportion of participants withdrawing from the studydue to an adverse event after treatment with a topical NSAIDwas 098 (333365)
bull The proportion of participants withdrawing from the studydue to an adverse event after treatment with placebo was 099(303040)
bull The RR of topical NSAID compared to placebo was 10(064 to 16)
bull There was no significant difference between treatmentgroups so the NNH was not calculated
Four studies did not specifically mention adverse event with-drawals (Haig 1986 Hoffmann 2012 Klainguti 2010 Vecchiet1991) while one reported that one participant withdrew with mildpruritus but did not state the treatment arm (Joussellin 2003)Ten studies specifically reported withdrawals due to lack of efficacy(Dreiser 1989 Dreiser 1994 Kuehl 2011 Machen 2002 Maziegraveres2005b Maziegraveres 2005a Noret 1987 Predel 2013a Russell 1991Thorling 1990) (Appendix 5) Numbers of participants withdraw-ing were generally low with rates of 6 or less except in Kuehl2011 where the rate was 10 with active treatment (diclofenacplaster) and 12 with placebo We did not carry out any analysisbecause the outcome was inconsistently reported
2 Topical NSAID versus active comparator
Details of efficacy outcomes in individual studies are in Appendix4 and of adverse events and withdrawals in Appendix 5
Participants with clinical success
Topical NSAID versus oral NSAID
bull kermark 1990 compared indomethacin spray withindomethacin capsules with response rates of 55 (1222) withspray and 23 (522) with capsules
bull Hosie 1993 compared felbinac foam with ibuprofen tabletswith response rates of 64 (81127) with felbinac foam and72 (96133) with ibuprofen tablets
bull Whitefield 2002 compared ibuprofen gel with ibuprofentablets with response rates of 60 (3050) with gel and 54(3650) with tablets
There were insufficient data for meta-analysis for any one of thesecomparisons felbinac is not known to be better than placebo
Topical NSAID versus different formulation of the same
topical NSAID
bull Fioravanti 1999 compared DHEP (diclofenac) gelformulated with and without lecithin with response rates of70 (3550) in both treatment arms
bull Mahler 2003 compared DHEP (diclofenac) gel formulatedwith and without lecithin with response rates of 89 (8292)with lecithin and 70 (6288) without lecithin
bull Gallacchi 1990 compared topical diclofenac formulated asFlectorreg gel and Emugelreg with response rates of 76 (1925)in both treatment arms
23Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
bull Governali 1995 compared topical ketoprofen cream withgel with response rates of 93 (1415) with cream and 27 (415) with gel
There were insufficient data for analysis
Topical NSAID versus different topical NSAID
Eight studies compared one topical NSAID versus at least oneother topical NSAID piroxicam versus indomethacin (Aoki1984 Fujimaki 1985 Sugioka 1984) ibuprofen versus ketopro-fen (Curioni 1985 Picchio 1981) ketoprofen versus etofena-mate (Curioni 1985 Tonutti 1994) ibuprofen versus etofenamate(Curioni 1985) ketorolac versus etofenamate (Diebshlag 1990)and diclofenac versus lysine clonixinate (Hofman 2000)There were sufficient data to compare only piroxicam with in-domethacin (Aoki 1984 Fujimaki 1985 Sugioka 1984 Analysis61)
bull The proportion of participants experiencing clinical successwith topical piroxicam was 56 (185330 range 49 to 78)
bull The proportion of participants experiencing clinical successwith topical indomethacin was 45 (140311 range 33 to64)
bull The RR of piroxicam compared with indomethacin was 12(11 to 14)
bull The NNT for successful treatment was 91 (53 to 30) Forevery nine participants treated with topical piroxicam onewould experience a clinical success who would not haveexperienced one with topical indomethacin
Topical NSAID versus different topical intervention
One study compared diclofenac gel with a herbal product calledTraumeel gel under double-blind conditions with response ratesfor being pain-free at seven days of 8137 (58) with diclofenacgel and 7140 (50) with Traumeel gel (Gonzaacutelez de Vega 2013)
Local adverse events
Topical NSAID versus oral NSAID
Two studies comparing a topical NSAID with an oral NSAIDprovided data on local adverse events ( kermark 1990 Hosie1993) There were five events with topical NSAID and three withoral NSAID which were too few for analysis
Topical NSAID versus different topical NSAID
All nine studies comparing one topical NSAID with at least oneother reported on local adverse events with a total of 48 events in1005 participants (48) There were sufficient data to compare
only piroxicam with indomethacin (Aoki 1984 Fujimaki 1985Sugioka 1984 Analysis 62)
bull The proportion of participants experiencing local adverseevents with topical piroxicam was 21 (7340 range 12 to28)
bull The proportion of participants experiencing local adverseevents with topical indomethacin was 10 (33331 range 29to 15)
bull The RR of piroxicam compared with indomethacin was021 (009 to 047)
bull The NNT to prevent a local adverse event was 13 (87 to23) For every thirteen participants treated with topicalpiroxicam one would not experience a local adverse event whowould have experienced one with topical indomethacin
Topical NSAID versus different topical intervention
Gonzaacutelez de Vega 2013 reported that adverse events were infre-quent and mild to moderate in intensity but did not distinguishbetween local and systemic events Numbers of participants expe-riencing any adverse event were 8147 with diclofenac gel and 14148 with Traumeel gel
Systemic adverse events
kermark 1990 reported numbers of events rather than numbersof participants with events while Tonutti 1994 and Whitefield2002 reported no adverse events attributable to the study medi-cation and Fioravanti 1999 Gallacchi 1990 Gualdi 1987 andSugioka 1984 did not mention systemic adverse events Gonzaacutelezde Vega 2013 did not distinguish between local and systemicevents In the remaining studies a total of 16 events were reportedin topical NSAID treatment arms (797 participants 2) and 11with ibuprofen tablets (134 participants 8)
Serious adverse events
No serious adverse events were reported in any treatment arm
Withdrawals
The only withdrawals reported due to adverse events were in stud-ies with placebo treatment arms ( kermark 1990 Fujimaki 1985)and have been reviewedThree studies reported withdrawals due to lack of efficacy (Gonzaacutelez de Vega 2013 Hofman 2000 Tonutti 1994) (Appendix5) There were insufficient data for analysisSome studies reported exclusions from analysis (efficacy or safetyor both) following randomisation mainly due to protocol vio-lations or loss to follow-up (Appendix 5) There is no reason tobelieve these exclusions would introduce systematic bias and thenumbers involved were not likely to influence results
24Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
D I S C U S S I O N
This updated review of topical NSAIDs for acute musculoskele-tal pain in adults differs from previous reviews Previously dataallowed only for comparison of individual topical NSAIDs withplacebo irrespective of formulation With a substantial amount ofnew data for diclofenac it is possible to distinguish effects of for-mulation for individual NSAIDs Because formulation chemistrycan substantially affect the rate and total amount of drug accessingsubcutaneous injured tissues the effect of formulation may be asimportant as the individual NSAID used Drug and formulationshould thus be considered together when assessing efficacy andthis is now possible
We have also included an assessment of the daily dose of NSAIDapplied to the skin This involved having information on the con-centration of NSAID in the preparation the amount used andthe frequency of use Not all studies reported all three but an esti-mation of topical doses applied was possible It varied by factors ofthree or four for each topical NSAID However for topical formu-lations the key issue is less the dose applied but the amount thatpenetrates locally (producing analgesic effect) and the amount en-tering the systemic circulation (producing potential harms) Bothwill depend on the exact formulation of the topical agent andwhether there is occlusion (Moore 2008a)
Summary of main results
This review included 61 studies comparing a topical NSAID withplacebo another topical NSAID or an oral NSAID In total5311 participants were treated with a topical NSAID 3470 withplacebo and 220 with an oral NSAID There were 63 moreparticipants than in the previous version of this review Conditionstreated were sprains strains and contusions mainly resulting fromsports injuries and overuse injuries such as tendinitisFormulations of topical diclofenac ibuprofen ketoprofen piroxi-cam and indomethacin demonstrated significantly higher rates ofclinical success than matching topical placebo lacking the NSAIDbenzydamine did not Three drug and formulation combinationshad NNTs for clinical success below 4 For diclofenac Emulgelreghad the lowest NNT of 18 (15 to 21) in two studies using atleast 50 pain intensity reduction as the outcome (high qualityevidence) Diclofenac plasters other than Flectorreg also had a lowNNT of 32 (26 to 42) based on good or excellent responses inrelatively recent studies (high quality evidence) Ketoprofen gelhad an NNT of 25 (20 to 34) from five studies in the 1980ssome with less well defined outcomes (moderate quality evidence)Ibuprofen gel had an NNT of 39 (27 to 67) from two stud-ies with outcomes of marked improvement or complete remission(moderate quality evidence) All other drug and formulation com-binations had NNT values above 4 indicating lesser efficacyThese results are better than alternative topical products that mightbe used for acute musculoskeletal pain There is no evidence tosupport the use of topical salicylate rubefacients (Derry 2014)
Treatment with a topical NSAID was not associated with anincrease in local adverse events (skin reactions) compared withplacebo (inert carrier) or in withdrawals due to adverse events(high quality evidence) The inert carrier was sometimes associ-ated with mild skin irritation but this rarely led to cessation oftreatment and quickly resolved Systemic adverse events were un-common and did not differ between topical NSAID and placebo(high quality evidence) Two participants experienced serious ad-verse events with diclofenac plaster and diclofenac gel but it isunlikely that these were related to the study medicationsThere were insufficient data directly comparing a topical NSAIDwith the same oral NSAID to draw conclusions about efficacyBased on very limited data for oral NSAIDs there were fewersystemic adverse events with topical than oral treatment Therewere sufficient data only for topical piroxicam compared withtopical indomethacin to compare one topical agent with anotherThese limited data suggested that piroxicam was more effectivethan indomethacin and was less likely to cause local adverse eventsIt is worth noting here that topical indomethacin did not givesignificantly better pain relief than placebo in two of the threestudies in this analysis
Overall completeness and applicability ofevidence
There is a tension between pooling studies to produce analyseswith larger numbers and the subsequent large increases in clinicaland statistical heterogeneity on the one hand and using the ap-proach of clinical homogeneity with subsequent smaller numbersof participants on the other hand Previous reviews have takenthe former approach that is useful in demonstrating that topicalNSAIDs rsquoworkrsquo by being significantly better than placebo Becauseof the substantial increase in the amount of data available in thisreview we have chosen to seek greater clinical homogeneity thisproduces results that are more relevant to patient and prescriberchoiceThere were too few studies comparing one topical NSAID versusanother or versus the same oral NSAID to allow meaningful directcomparisons between individual drugs or routes of administrationThe conditions treated in these studies are representative of thoselikely to be suitable for acute treatment with topical NSAIDs Themean age of participants in individual studies ranged from 25 to57 years and the nature of recruitment in many studies meant thatparticipants were actively engaged in sporting activities Neverthe-less older people in their 60s to 80s were also included in somestudies and the low levels of predominantly mild adverse eventsmeans that this route of administration of NSAIDs is suitable forall age groups able to manage the application processInformation from other sources mainly randomised studies last-ing 12 weeks or more in older populations with arthritis tendto confirm this A systematic review of topical NSAIDs in olderadults was difficult to interpret but suggested that the range of
25Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
withdrawal rates in these studies was similar with topical and oralNSAIDs (Makris 2010) It also claimed potentiation of warfarineffects but that was with topical salicylate not an NSAID Incontrast a pooled safety analysis of topical diclofenac in peopleaged 75 years or older reported minimal changes with a mean re-duction in haemoglobin of less than 1 gL with topical diclofenacand a mean systolic blood pressure reduction of almost 4 mm Hg(Roth 2012) Two large randomised 12-week studies comparingtopical with oral diclofenac in arthritis reported lower rates of gas-trointestinal adverse events with topical than oral especially severeevents but larger reductions in haemoglobin with oral diclofenac(Simon 2009 Tugwell 2004)The available evidence was limited by numbers to comment on rarebut potentially serious adverse events One example is the potentialfor photo-sensitivity reactions with topical ketoprofen Currentadvice from the Medicines and Healthcare products RegulatoryAgency in the UK is to avoid direct sunlight ultraviolet (UV) rayssunlamps and sunbeds while using topical ketoprofen and to seea healthcare professional or go to hospital if they experience a skinreaction to sunlight sunlamps or sunbeds (MHRA 2009)
Quality of the evidence
All included studies were both randomised and double-blind nonewas considered at high risk of methodological bias Many werecarried out in the 1980s and 1990s when methodological rigorand detailed reporting were not given such high priority and stud-ies did not always report details of the randomisation treatmentallocation and blinding processes More recent studies often didreport methodological details and tended to be larger (see Figure4 for a comparison of quality of reporting for different dates andNSAIDs) Our primary outcome of clinical success was not alwayswell-defined and was measured using different scales but againmore recent studies tended to report outcomes betterThe studies were conducted in different conditions with some-what different outcome definitions and duration and with differ-ent topical NSAIDs and formulations Moreover the small size ofmany of the studies is likely to result in considerable chance varia-tion (Counsell 1994 Moore 1998b) These factors would accountfor the high I2 values seen in several analyses Despite these sourcesof potential clinical heterogeneity most studies showed benefit oftopical NSAID over placeboThe design of studies to be able to demonstrate analgesic sensi-tivity is important in self limiting conditions such as strains andsprains Too long a duration and the condition results in sponta-neous resolution of painful symptoms while too short a durationmay be inadequate to show any effect The decision by trialists toconcentrate on outcomes closest to seven days of treatment appearsto be prudent and has been adopted in this and previous reviewsThere are potential differences in response to treatment betweenstrains and sprains and overuse-type injuries such as tendinitis
and future reviews may examine this At the present time thereare too few existing trials to explore any differences adequatelyBaseline pain may be a cause for concern Seven studies didnot report baseline pain levels (Billigmann 1996 Curioni 1985Haig 1986 NCT01255423 NCT01272934 NCT01272947Sinniger 1981) and a further 11 reported either mean levels ofless than moderate pain or a significant proportion of individualswith less than moderate pain ( kermark 1990 Aoki 1984 Auclair1989 Diebshlag 1990 Fujimaki 1985 Jenoure 1997 Linde 1985Picchio 1981 Ramesh 1983 Sugioka 1984 Whitefield 2002)using recognised scales Insufficient pain at baseline compromisesthe ability of a study to demonstrate any improvement All thenewly added studies reported baseline pain to be of at least mod-erate intensity
Potential biases in the review process
There has been greater interest in topical NSAIDs in recent yearsmainly because lower systemic drug levels reduce the risk of trou-blesome and severe adverse events particularly in the gastroin-testinal tract and renal and cardiovascular systems Most of theattention has been in chronic conditions such as osteoarthritiswith few studies in acute painful conditions Low levels of seriousadverse events with topical NSAIDs has been noted previously(Evans 1995) and the near absence of serious adverse events inthis review is unlikely to be due to any biases in the review processOne potential bias is that clinical trials for topical NSAIDs may nothave been published One previous review did find previously un-published trials (Moore 1998a) but a subsequent attempt that in-cluded extensive contacts with pharmaceutical companies revealedno additional data (Mason 2004a) While some old unpublishedstudies of topical NSAIDs in acute painful conditions may existthey constitute an unknown number of studies and participantswhose results are unknown and are likely to remain unknownFurthermore their relevance to current clinical practice may belimited as better formulations are developed New systems of trialregistration mean that we know what recent studies have been doneor are ongoing the number of studies and participants is knowneven if their results remain unknown We identified in Clinicaltri-alsgov three unpublished studies (612 participants) with adverseevent data but no dichotomous efficacy data 14 completed un-published studies (4403 participants) with no results posted andthree ongoing studies (880 participants)For the main topical NSAIDs of interest and where most informa-tion exists about 4200 participants in this review provided data onefficacy for diclofenac ibuprofen ketoprofen indomethacin andbenzydamine compared with placebo For efficacy there are un-known results from almost 5900 participants in studies known tohave been done but essentially unpublished Almost 6500 partic-ipants in this review provided information on local adverse eventsfor topical NSAIDs compared with placebo For local adverseevents the unknown results from known studies represents almost
26Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
5300 participants It is clear that identified unpublished but un-available study data amounts to a further potentially large increasein knowledge over and above the 60 increase in numbers ofparticipants already included in this updated reviewBased on efficacy data on known and available study results un-published trials showing no difference between any topical NSAIDand topical placebo and involving 5500 participants would haveto exist in order for the NNT to be as high as 9 at which pointthe effectiveness of topical NSAIDs would become clinically irrel-evant (Moore 2006) This amount of unpublished negative data isobviously available and while a negative result in all the identifiedstudies is unlikely knowledge would be greatly served by havingthese unpublished trial results availableWe have not yet attempted to obtain results from these clinicaltrials from the trial sponsors because this takes a considerableamount of time and may not be successful Moreover the studieswere spilt between nine different sponsors Cerimon Pharmaceu-ticals (five studies) Novartis (four studies) Endo Pharmaceuticals(three studies) GlaxoSmithKline (two studies) Hisamatsu Phar-maceutical (two studies) and one each from Actavis Pfizer Im-primis Pharmaceutical and Strategic Science amp Technologies
Agreements and disagreements with otherstudies or reviews
A review published in 2004 included some of the studies in thisreview and reported an NNT for all topical NSAIDs combinedof 38 (34 to 44) for clinical success equivalent to half pain reliefat seven days (Mason 2004a) That review found no differencebetween topical NSAID and placebo for local adverse events as didthis review In turn the Mason review was in broad agreement withthe original systematic review on topical NSAIDs (Moore 1998a)To our knowledge no previous review assembled sufficient trialdata to analyse results by both drug and formulation as was donehere
A U T H O R S rsquo C O N C L U S I O N S
Implications for practice
For people with acute musculoskeletal pain
Topical NSAIDs can provide good levels of pain relief in acuteconditions such as sprains strains and overuse injuries proba-bly similar to that provided by oral NSAIDs Gel formulations ofdiclofenac (as Emulgelreg) ibuprofen and ketoprofen and somediclofenac patches provide the best effects Adverse events are usu-ally minimal with topical NSAIDs
For clinicians
Topical diclofenac ibuprofen or ketoprofen gels provide goodpain relief for painful acute musculoskeletal conditions and arebetter tolerated than oral formulations These drugs and formu-lations are more likely to be cost effective than alternative topicalpreparations such as topical rubefacients
For policy makers
Topical NSAIDs are not associated with an increased incidence oflocal skin reactions compared with the inert carrier and while thecarrier may cause mild transient irritation it is rarely troublesomeTopical NSAIDs do not cause systemic (mainly gastrointestinal)problems commonly seen with oral NSAIDs making them partic-ularly useful for individuals unable to tolerate oral administrationor for whom it is contraindicated
For funders
Topical diclofenac ibuprofen or ketoprofen gels should be consid-ered for initial treatment of acute musculoskeletal painful condi-tions where there are no contraindications such as damaged skinThese drugs and formulations are more likely to be cost effectivethan alternative topical preparations such as topical rubefacientsBecause formulations of topical NSAIDs are likely to change overtime the relevant trials performed and reported in or before the1990s must be limited and may be questionable Funders mightwish to consider asking pharmaceutical companies without recenttrial evidence for their products to produce it
Implications for research
General
The general thrust of these findings is that gel formulations of top-ical diclofenac ibuprofen and ketoprofen work best but for somedrugs (ketoprofen for instance) studies were pre-1990 Thesestudies may not be relevant to products available now Becauseformulation can have a significant effect on efficacy formulationchanges should be accompanied by relevant randomised trials
Design
The design of the trials is generally good and the sports injurymodel appears to be reliable and reproducible Modern studieshave ensured that participants entering the trials have at least mod-erate pain and this helps sensitivity to detect an analgesic responseMajor changes to the design of these trials would not appear to beneeded
27Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Measurement (endpoints)
A major issue is not in the measurement of pain as most studiesespecially modern studies have used standard pain intensity andpain relief scales However reporting of average pain changes isinadequate and the use of responder analyses (at least 50 painintensity reduction or people experiencing mild or no pain) ispreferred
Comparison between active treatments
Indirect comparisons with placebo are probably as informative asuse of an active comparator
A C K N O W L E D G E M E N T S
The Oxford Pain Relief Trust the NHS Cochrane CollaborationProgramme Grant Scheme and the NIHR Biomedical ResearchCentre Programme provided support for the earlier review
The National Institute for Health Research (NIHR) is the largestsingle funder of the Cochrane Pain Palliative and SupportiveCare Review Group Disclaimer the views and opinions expressedherein are those of the authors and do not necessarily reflect thoseof the NIHR National Health Service (NHS) or the Departmentof Health
R E F E R E N C E S
References to studies included in this review
Airaksinen 1993 published data onlyAiraksinen O Venaumllaumlinen J Pietilaumlinen T Ketoprofen 25gel versus placebo gel in the treatment of acute soft tissueinjuries International Journal of Clinical PharmacologyTherapy and Toxicology 199331(11)561ndash3
kermark 1990 published data only
kermark C Forsskaringhl B Topical indomethacin in overuseinjuries in athletes A randomized double blind studycomparing Elmetacinreg with oral indomethacin andplacebo International Journal of Sports Medicine 199011(5)393ndash6
Aoki 1984 published data only
Aoki T Numajiri M Yamamoto M A well controlledcomparative study of piroxicam gel indomethacin geland placebo gel in the treatment of trauma JapanesePharmacology and Therapeutics 198412(12)101ndash17
Auclair 1989 published data onlyAuclair J Georges M Grapton X Gryp L DrsquoHooghe MMeisser RG et al A double-blind controlled multi-centerstudy of percutaneous niflumic acid gel and placebo in thetreatment of Achilles heel tendinitis Current TherapeuticResearch 198946(4)782ndash8
Billigmann 1996 published data onlyBilligmann PW Treatment of ankle distortion withibuprofen gel [Therapie von Sprunggelenksndashdistosionenmit ibuprofenndashmikrogel] Therapiewoche 199646(21)1187ndash92
Campbell 1994 published data onlyCampbell J Dunn T Evaluation of topical ibuprofen creamin the treatment of acute ankle sprains Journal of Accidentand Emergency Medicine 199411(3)178ndash82
Chatterjee 1977 published data onlyChatterjee DS A double-blind clinical study withbenzydamine 3 cream on soft tissue injuries in anoccupational health centre Journal of International MedicalResearch 19775(6)450ndash8
Costantino 2011 published data only
Costantino C Kwarecki J Samokhin AV Mautone GRovati S Diclofenac epolamine plus heparin plasterversus diclofenac epolamine plaster in mild to moderateankle sprain a randomized double-blind parallel-groupplacebo-controlled multicentre phase III trial ClinicalDrug Investigation 201131(1)15ndash26 [DOI 10216511585890-000000000-00000]
Coudreuse 2010 published data onlyCoudreuse JM de Vathaire F Effect of a plaster containingDHEP and heparin in acute ankle sprains with oedemaa randomized double-blind placebo-controlled clinicalstudy Current Medical Research and Opinion 201026(9)2221ndash8 [DOI 101185030079952010508020]
Curioni 1985 published data only
Curioni BG Di Domenica F Daolio P Spignoli GEvaluation of ibuproxam gel in traumatology [Valutazionedellrsquoefficacia terapeutica di ibudros gel in traumatologia]Clinica Europea 198524(3)456ndash60
Diebshlag 1990 published data only
Diebschlag W Nocker W Bullingham R A double-blindstudy of the efficacy of topical ketorolac tromethamine gelin the treatment of ankle sprain in comparison to placeboand etofenamate Journal of Clinical Pharmacology 199030
(1)82ndash9
Dreiser 1988 published data onlyDreiser RL Clinical trial of efficacy and tolerability oftopical ibuprofen in the treatment of tendinitis Le JournalInternational De Meacutedecine 198811915ndash31
Dreiser 1989 published data only
Dreiser RL Clinical trial - Fatsum gel FG-6 Supplied byMenarini 1989
Dreiser 1990 published data onlyDreiser RE Charlot J Lopez A Ditisheim A Clinicalevaluation of niflumic acid gel in the treatment ofuncomplicated ankle sprains Current Medical Research andOpinion 199012(2)93ndash9
28Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Dreiser 1994 published data only
Dreiser RL Roche R de Sahb R Thomas F Leutenegger EFlurbiprofen local action transcutaneous (LAT tm) clinicalevaluation in the treatment of acute ankle sprains EuropeanJournal of Rheumatology and Inflammation 199414(4)9ndash13
Fioravanti 1999 published data onlyFioravanti A Cicero MR Nerucci F Manopulo RMarcolongo R Double-blind controlled clinical study of theefficacy and tolerability of diclofenac-N-(2-hydroxyethyl)-pyrrolidine lecithin gel compared with diclofenac-N-(2-hydroxyethyl)-pyrrolidine gel in patients with periand extraarticular inflammatory diseases Drugs underExperimental and Clinical Research 199925(5)235ndash40
Fujimaki 1985 published data only
Fujimaki E et al Clinical evaluation of piroxicam gelversus indomethacin gel and placebo in the treatment ofmuscle pain a double-blind multicenter study JapanesePharmacology and Therapeutics 198512(12)119ndash37
Gallacchi 1990 published data only
Gallacchi G Mautone G Lauldi P Topical treatmentwith diclofenac hydroxyethylpyrrilidine (Flector gel 1)Clinical Trials Journal 199027(1)58ndash64
Gonzaacutelez de Vega 2013 published data only
Gonzaacutelez de Vega C Speed C Wolfarth B Gonzaacutelez JTraumeel vs diclofenac for reducing pain and improvingankle mobility after acute ankle sprain a multicentrerandomised blinded controlled and non-inferiority trialInternational Journal of Clinical Practice 201367(10)979ndash89 [DOI 101111ijcp12219]
Governali 1995 published data only
Governali E Casalini D A controlled clinical studyon ketoprofen gel 5 versus ketoprofen ointment 1in patients with post-traumatic lesions [Ricerda clinicacontrollata tra ketoprofene gel 5 e ketoprofene crema1 in pazienti con postumi di lesioni traumatiche]Riabilitazione 199528(1)61ndash9
Gualdi 1987 published data only
Gualdi A Bonollo L Martini A Forgione A Non-steroidal anti-inflammatory drugs for topical therapy intraumatology a double-blind study with flunoxaprofen andketoprofen [Antinflammatori no steroidi per uso topicoin traumatologia studio clinico con flunoxaprofene echetoprofene] Riforma Medica 1987102(10)401ndash4
Haig 1986 published data only
Haig G Portable thermogram technique for topicallyapplied benzydamine cream in acute soft-tissue injuriesInternation Journal of Tissue Reactions 19868(2)145ndash7
Hoffmann 2012 published data only
Hoffmann P Kopa ka P Gugliotta B Rovati S Efficacyand tolerability of DHEP-heparin plaster in reducingpain in mild-to-moderate muscle contusions a double-blind randomized trial Current Medical Researchand Opinion 201228(8)1313ndash21 [DOI 101185030079952012709182]
Hofman 2000 published data only
Hofman J Nasswetter G Cayetti LM Lysine clonixinate gelin soft tissue injuries Controlled randomized prospectivedouble-blind clinical trial with diclofenac [Clonixinatode lisina gel en lesiones de tejidos blandos ensayo clinicoprospectivo doble ciego al azar controlado con diclofenac]Prensa Medica Argentina 200087(5)513ndash20
Hosie 1993 published data only
Hosie GAC The topical NSAID felbinac versus oralibuprofen a comparison of efficacy in the treatment ofacute lower back injury British Journal of Clinical Research199345ndash17
Jenoure 1997 published data only
Jenoure PJ Rostan A Gremion G Meier JL Grossen RBielinki R et al Multicentre double-blind controlledclinical study on the efficacy of diclofenac epolamineTissugel plaster in patients with epicondylitis [Studiomulticentrico controllato in doppio cieco su diclofenacTissugel plaster in pazienti con epicondilite] MedicinaDello Sport 197750(3)285ndash92
Joussellin 2003 published data onlylowast Joussellin E Flector Tissugel for the treatment ofpainful ankle sprains [Flector Tissugel dans le traitementdes entorses douloureuses de la cheville] Journal deTraumatologie du Sport 2003201S5ndash9Lionberger DR Joussellin E Lanzarotti A Yanchick JMagelli M Diclofenac epolamine topical patch relieves painassociated with ankle sprain Journal of Pain Research 2011447ndash53 [DOI 102147JPRS15380]Lionberger DR Joussellin E Yanchick J Magelli MLanzarotti A Pooled analysis of clinical trial data evaluatingthe safety and effectiveness of diclofenac epolamine topicalpatch 13 for the treatment of acute ankle sprain OpenAccess Journal of Sports Medicine 2011275ndash84 [DOI102147OAJSMS17048]
Julien 1989 published data onlyJulien D Clinical trial - Fastum gel FG-8 Supplied byMenarini 1989
Klainguti 2010 published data onlyKlainguti A Forgacs A Berkes I Castellacci E A plastercontaining DHEP and heparin for mild to moderatecontusions and sprains with haematoma a double-blind randomized study Current Medical Researchand Opinion 201026(9)2243ndash51 [DOI 101185030079952010508022]
Kockelbergh 1985 published data onlyKockelbergh M Verspeelt P Caloine R Dermaux F Localanti inflammatory treatment with a ketoprofen gel currentclinical findings Journal Belge de Medecine Physique et deRehabilitation 19858(4)205ndash13
Kuehl 2011 published data only
Kuehl K Carr W Yanchick J Magelli M Rovati SAnalgesic efficacy and safety of the diclofenac epolaminetopical patch 13 (DETP) in minor soft tissue injuryInternational Journal of Sports Medicine 201132(8)635ndash43[DOI 101055s-0031-1275359]
29Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Li 2013 published data only
Li C Frangione V Rovati S Zheng Q Diclofenacepolamine medicated plaster in the treatment of minor softtissue injuries a multicenter randomized controlled trialCurrent Medical Research and Opinion 201329(9)1137ndash46[DOI 101185030079952013816669]
Linde 1985 published data only
Linde F Hvass I Juumlrgensen U Madsen F Treatment ofsprained ankles with 5 benzydamine creme A double-blind study [Ankelforstuvninger behandlet med benzydamin5 creme] Ugeskrift for Laeger 1985148(1)12ndash3
Machen 2002 published data onlyMachen J Whitefield M Efficacy of a proprietary ibuprofengel in soft tissue injuries a randomised double blindplacebo controlled study International Journal of ClinicalPharmacology 200256(2)102ndash6
Mahler 2003 published data only
Mahler P Mahler F Duruz H Ramazzina M Liguori VMautone G Double-blind randomized controlled studyon the efficacy and safety of a novel diclofenac epolaminegel formulated with lecithin for the treatment of sprainsstrains and contusions Drugs under Experimental andClinical Research 200329(1)45ndash52
Maziegraveres 2005a published data onlyMaziegraveres B Rouanet S Velicy J Scarsi C Reiner V Topicalketoprofen patch (100 mg) for the treatment of anklesprain a randomized double-blind placebo-controlledstudy American Journal of Sports Medicine 200533(4)515ndash23 [DOI 1011770363546504268135]
Maziegraveres 2005b published data onlyMaziegraveres B Rouanet S Guillon Y Scarsi C Reiner VTopical ketoprofen patch in the treatment of tendinitisa randomized double blind placebo controlled studyJournal of Rheumatology 200532(8)1563ndash70
McLatchie 1989 published data onlyMcLatchie GR McDonald M Lawrence GF RogmansD Lisai P Hibberd M Soft tissue trauma a randomisedcontrolled trial of the topical application of felbinac a newNSAID British Journal of Clinical Practice 198943(8)277ndash80
Morris 1991 published data only
Morris WD Scott HV Peters WA Ketelbey JW Felbinactopical gel for acute soft tissue sports injuries New ZealandJournal of Sports Medicine 19911945ndash7
NCT01255423 unpublished data only
Novartis (Sponsors) A randomized double-blind multi-center placebo-controlled parallel group study to evaluatethe efficacy and safety of diclofenac sodium topical gel(DSG) 1 applied 4 times daily in subjects with acuteankle sprain clinicaltrialsgovct2showNCT01255423(accessed 3 February 2015) 2012 [CTG NCT01255423]
NCT01272934 unpublished data only
Novartis (Sponsors) A randomized double-blind multi-center placebo-controlled parallel group study to evaluatethe efficacy and safety of diclofenac sodium topical gel
(DSG) 1 applied four times daily in subjects withacute ankle sprain clinicaltrialsgovct2showresultsNCT01272934 (accessed 3 February 2015) 2013 [CTGNCT01272934]
NCT01272947 unpublished data onlylowast Novartis (Sponsors) A randomized double-blindmulti-center placebo-controlled parallel group study toevaluate the efficacy and safety of diclofenac sodium topicalgel (DSG) 1 applied four times daily in subjects withacute blunt soft tissue injuriescontusions of the limbsclinicaltrialsgovct2showresultsNCT01272947 (accessed3 February 2015) 2012 [CTG NCT01272947]Pallay R Pabst H Giannetti B Burnett I Monnet JDiclofenac sodium topical gel (DSG) 1 for acute softtissue injuriescontusions of the limbs 2013 AnnualAssembly of the American Academy of Physical Medicineand Rehabilitation National Harbor MD United States2013 Vol 5 (9 Suppl 1)S204
Noret 1987 published data only
Noret A Roty V Allington N Hauters P Zuinen C PoelsR Ketoprofen gel as topical treatment for sport injuriesActa Therapeutica 198713367ndash78
Parrini 1992 published data onlyParrini M Cabitza P Arrigo A Vanasia M Efficacy andtolerability of ketoprofen lysine salt foam for topical use inthe treatment of traumatic pathologies of the locomotorapparatus [Efficacia e tollerabilita del ketoprofene sale dilisina schiuma per uso topico nel trattamento di alcunepatologie traumatiche dellrsquoapparato locomotore] La ClinicaTerapeutica 1992141(9)199ndash204
Picchio 1981 published data only
Picchio AA Volta S Longoni A Controlled clinical trial ofibuprofen for topical use in sport injuries [Studio clinicocontrollato sullrsquoimpiego dellrsquoibuprofen per uso topico intraumatologica sportiva] Medicina dello Sport 198134403ndash6
Predel 2004 published data only
Mueller EA Kirch W Reiter S Extent and time courseof pain intensity upon treatment with a topical diclofenacsodium patch versus placebo in acute traumatic injury basedon a validated end point post hoc analysis of a randomizedplacebo-controlled trial Expert Opinion on Pharmacotherapy201011(4)493ndash8 [DOI 10151714656560903535898]lowast Predel HG Koll R Pabst H Dieter R Gallacchi GGiannetti B et al Diclofenac patch for topical treatmentof acute impact injuries a randomised double blindplacebo controlled multicentre study British Journal ofSports Medicine 200438(3)318ndash23 [DOI 101136bjsm2003005017]
Predel 2012 published data onlyPredel HG Hamelsky S Gold M Giannetti B Efficacy andsafety of diclofenac diethylamine 232 gel in acute anklesprain Medicine and Science in Sports and Exercise 201244
(9)1629ndash36 [DOI 101249MSS0b013e318257ed41]
30Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Predel 2013a published data only
Predel HG Giannetti B Seigfried B Novellini R MenkeG A randomized double-blind placebo-controlledmulticentre study to evaluate the efficacy and safetyof diclofenac 4 spray gel in the treatment of acuteuncomplicated ankle sprain Journal of InternationalMedical Research 201341(4)1187ndash202 [DOI 1011770300060513487639]
Predel 2013b published data onlyPredel HG Giannetti B Pabst H Schaefer A Hug AMBurnett I Efficacy and safety of diclofenac diethylamine116 gel in acute neck pain a randomized double-blindplacebo-controlled study BMC Musculoskeletal Disorders201314250 [DOI 1011861471-2474-14-250]
Ramesh 1983 published data only
Ramesh N Steuber U Ibuprofen in a cream vehicle foraccidental and sport injuries [Dolgit creme bei unfallndash undsportverletzungen] Therapiewoche 1983334563ndash70
Rowbotham 2003 published data only
Rowbotham M Galer B Block J Backonja M FlectorTissugel efficacy and safety in minor sport injuries Avs controlled clinical trial Journal of Sports Traumatology200320IS15ndashIS20
Russell 1991 published data only
Russell AL Piroxicam 05 topical gel compared to placeboin the treatment of acute soft tissue injuries a double-blind study comparing efficacy and safety Clinical andInvestigative Medicine 199114(1)35ndash43
Saillant 1998 published data only
Lionberger DR Joussellin E Yanchick J Magelli MLanzarotti A Pooled analysis of clinical trial data evaluatingthe safety and effectiveness of diclofenac epolamine topicalpatch 13 for the treatment of acute ankle sprain OpenAccess Journal of Sports Medicine 2011275ndash84 [DOI102147OAJSMS17048]lowast Saillant G Study comparing the efficacy and toleranceof Flector Tissugelreg to that of a placebo in the treatmentof benign ankle sprains [Eacutetude comparant lrsquoefficaciteacute et latoleacuterance de FlectorTissugelreg agrave celles drsquoun placebo dans letraitement des entorses beacutenignes de la cheville] Medicinedu Sport 1998721ndash5
Sanguinetti 1989 published data onlySanguinetti C Treatment of soft tissue injury with BPAAgel Results of an Italian multicenter study vs placebo[Trattemento con BPAA gel dei traumi dei tessuti moli]Clinica Terapeutica 1989130(5)255ndash258
Sinniger 1981 published data onlySinniger M Blanchard P Controlled clinical trial withFentiazac cream in sport microtraumatology Journal ofInternational Medical Research 19819(4)300ndash2
Spacca 2005 published data onlySpacca G Cacchio A Forgcs A Monteforte PRovetta G Analgesic efficacy of a lecithin-vehiculateddiclofenac epolamine gel in shoulder periarthritis andlateral epicondylitis a placebo-controlled multicenter
randomized double-blind clinical trial Drugs underExperimental and Clinical Research 200531(4)147ndash54
Sugioka 1984 published data only
Sugioka Y Multicenter clinical evaluation of piroxicam gelvs indomethacin gel in the treatment of non-traumaticdiseases of tendon or muscle Japanese Pharmacology andTherapeutics 198412139ndash53
Thorling 1990 published data only
Thorling J Linden B Berg R Sandahl A A double blindcomparison of naproxen gel and placebo in the treatment ofsoft tissue injuries Current Medical Research and Opinion199012242ndash8
Tonutti 1994 published data only
Tonutti A The use of ketoprofen gel 5 (orudis gel) intraumatology controlled double-blind study vs etofenamate[Utilizzazione del ketoprofene gel 5 (orudis gel) nellapratica traumatologica studio in doppio cieco controllatoverso etofenamato] Ortopedia e Traumatologia Oggi 199414(3)119ndash25
Vecchiet 1991 published data only
Vecchiet L Colozzi A Effects of meclofenamic acid in thetreatment of lesions deriving from minor traumatologyClinical Journal of Pain 19917 Suppl 1S54ndash9
Whitefield 2002 published data only
Whitefield M OrsquoKane CJA Anderson S Comparativeefficacy of a proprietary topical ibuprofen gel and oralibuprofen in acute soft tissue injuries a randomised doubleblind study Journal of Clinical Pharmacy and Therapeutics200227(6)409ndash17
References to studies excluded from this review
Ambrus 1987 published data only
Ambrus P Boumlhmer D Mobilat ointment in acute sprains[Mobilat Salbe bei akuten Distorsionen] Fortschritte derMedizin 1987105(13)259ndash62
Anon 1993 unpublished data only
Anon Comparative clinical efficacy of Oruvail piroxicamand diclofenac gels in soft tissue injury Unpublished
Ascherl 1982 published data onlyAscherl R Schlemmer H Blumel G Lechner F Theeffectiveness of etofenamate in minor sports injuries ofthe knee and ankle joint A double blind study [DieWirksamkeit der Etofenamat in kleineren Sportverletzungenam Knie und Sprunggelenk Eine DoppelblindndashStudie]Fortschritte der Medizin 1982100(37)1729ndash34
Bagliani 1976 published data only
Bagliani A Montalbetti L Topical treatment ofthrombophlebitis with feprazone and benzydamineControlled clinical study [Il trattamento topico ditromboflebite con feprazone e benzidamina Studio clinicocontrollato] Minerva Medica 197667(14)880ndash4
Baracchi 1982 published data only
Baracchi G Messina Denaro S Piscini S Experience ofthe topical use of isobutylfenylproprionic acid (Ibuprofen)in traumatic inflammation A double-blind comparison
31Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
with placebo [Esperienza sullrsquoimpiego topico dellrsquoacidoisobutilndashfenilndashpropionico (ibuprofen) nella flogositraumatica Confronto a doppia cecita con placebo]Gazzetta Medica Italiana 1982141(12)691ndash4
Boumlhmer 1995 published data onlyBoumlhmer D Ambrus P Treatment of muscular injurieswith diclofenac-diethylammonium emugel [Behandlungvon muskelverletzungen mit diclofenacndashdiethylammoniumemugel] Sportverletzung Sportschaden 1995994ndash5
Burnham 1998 published data only
Burnham R Gregg R Healy P Steadward R Theeffectiveness of topical diclofenac for lateral epicondylitisClinical Journal of Sport Medicine 19988(2)78ndash81
Cesarone 2008 published data onlyCesarone MR Belcaro G Pellegrini L Ledda A VinciguerraG Ricci A et al Treatment of ankle sprain in patientswith vascular diseases of the lower limbs MinervaCardioangiologica 200856 (5 Suppl)39ndash46
Coulibaly 2009 published data only
Coulibaly SK Courau S Staiger C The sprained anklecomparative study of a comfreyroot extract ointment versusa diclofenac gel [Entorses de la cheville eacutetude comparativeentre un extrait agrave base de racines de consoude et lediclofeacutenac] Phytotheacuterpie 20097147ndash9 [DOI 101007s10298-009-0383-x]
Diebschlag 1985 published data only
Diebschlag W Benzydamine cream in post-traumaticoedema International Journal of Tissue Reactions 19857(3)219ndash23
Diebschlag 1986 published data onlyDiebschlag W Diclofenac in blunt traumatic ankle jointswelling Volumetric monitoring in a placebo controlleddouble blind trial [Diclofenac in stumpfen traumatischenSprunggelenk Schwellung Volumetrische Uumlberwachungin einer Placebondashkontrollierten DoppelblindndashStudie]Fortschritte der Medizin 1986104(21)437ndash40
Diebschlag 1992 published data onlyDiebschlag W Nocker W Lehmacher W Treatment of acutesprains of the ankle joint A comparison of the effectivenessand tolerance of two gel preparations containingindomethacin [Die Behandlung der akuten Verstauchungendes Sprunggelenks Ein Vergleich der Wirksamkeitund Vertraumlglichkeit von zwei GelndashZubereitungen mitIndometacin] Fortschritte der Medizin 1992110(6)64ndash72
Fantato 1971 published data only
Fantato S De Gregorio M Clinical evaluation of topicalbenzydamine in traumatology Arzneimittel-Forschung197121(10)1530ndash5
Galer 2000 published data onlyGaler BS Rowbotham M Perander J Devers A FriedmanE Topical diclofenac patch relieves minor sports injury painresults of a multicenter controlled clinical trial Journal ofPain and Symptom Management 200019(4)287ndash94
Hallmeier 1986 published data only
Hallmeier B Michelbach B Etofenamate under tapebandages - a controlled study [Etofenamat untertapendashverbaumlnden] Medizinische Welt 198637(43)1344ndash8
Hallmeier 1988 published data onlyHallmeier B Efficacy and tolerance of etofenamate anddiclofenac in acute sports injuries [Wirksamkeit undVertraumlglichkeit von Etofenamat und Diclofenac bei akutenSportverletzungen] Rheuma 19888183ndash6
Kaneko 1999 published data only
Kaneko M Shimojo H Saito H Onuma Y YamashitaK Clinical evaluation of felbinac patch (SELSPOT) onpost-traumatic disease clinical comparative study versuscommercially available patch Japanese Pharmacology andTherapeutics 19992775ndash85
Kockelbergh 1985b published data only
Kockelbergh M Verspeelt P Caloine R Dermaux F Localanti-inflammatory treatment with a ketoprofen gel currentclinical findings [Traitement antindashinflammatoire local parun gel de keacutetoprofegravene donneacutees cliniques reacutecentes] JournalBelge de Medecine Physique et de Rehabilitation 19858(4)205-13 (study 2)
Kuwabara 2013 published data only
Kuwabara Y Hamamoto H Hikake S Miwa Y Arandomized multi-center double-blind placebo-controlledphase IIIII trial to evaluate the eftopical patch in thetreatment of pain Journal of Pain 2013 Vol 14 (4 Suppl1)S73
Lee 1991 published data only
Lee EH Lee PY Ngai AT Chiu EH Treatment of acute softtissue trauma with a topical non-steroidal anti-inflammatorydrug (biphenylacetic acid 3 gel) Singapore MedicalJournal 199132(4)238ndash41
Link 1996 published data onlyLink R Balint G Pavlik G Otto J Krause W Topicaltreatment of soft tissue rheumatism and athletic injuriesEffectiveness and tolerance of a new ketoprofen gel[Topische Behandlung von WeichteilndashRheumatismus undSportverletzungen Wirksamkeit und Vertraumlglichkeit einesneuen Ketoprofen Gel] Fortschritte der Medizin 1996114(25)311ndash4
May 2007 published data onlyMay JJ Lovell G Hopkins WG Effectiveness of1 diclofenac gel in the treatment of wrist extensortenosynovitis in long distance kayakers Journal of Scienceand Medicine in Sport 200710(1)59ndash65
Oakland 1993 published data only
Oakland C Rapier C A comparison of the efficacy of thetopical NSAID felbinac and ultrasound in the treatmentof acute ankle injuries British Journal of Clinical Research1993489ndash96
Odaglia 1987 unpublished data only
Odaglia G Sereni G Sports minor traumatology results ofa double-blind controlled clinical study ketoprofen (fastumgel 25) versus placebo Menarini unpublished data
32Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Picardi 1993 published data only
Picardi E De Iasio R Efficacy of percutaneous anti-inflammatory drugs (NSAIDs) in swimmers and waterpoloplayers [Efficacia dei farmaci antinfiammatori (FANS) pervia percutanea in atleti di nuoto e pallanuoto] ClinicaTerapeutica 1993143(6)507ndash9
Taboada 1992 published data onlyTaboada A Controlled trial of piroxicam gel associated withultrasound in acute disturbances of the locomotive system[Experienca controlada con gel de piroxicam asociado aultrasonidos en afecciones agudas del aparato locomotor]Prensa Medica Argentina 199279(10)630ndash2
Vanderstraeten 1990 published data only
Vanderstraeten G Schuermans P Study on the effect ofetofenamate 10 cream in comparison with an oral NSAIDin strains and sprains due to sports injuries Acta BelgicaMedica Physica 199013(3)139ndash41
Vinciguerra 2008 published data only
Vinciguerra G Belcaro G Cesarone MR Errichi BM DiRenzo A Errichi S et al Management of uncomplicatedankle sprains with topical or oral ketoprofen treatmentA registry study Minerva Cardioangiologica 200856 (5
Suppl)47ndash53
Von Klug 1977 published data onlyVon Klug H Experience with a locally applied anti-inflammatory drug [Erfahrungen mit einem localanwendbaren antirheumatikum] Arzneimittel-ForschungDrug Research 1977271350ndash4
References to studies awaiting assessment
NCT00351104 published data onlyEndo Pharmaceuticals (Sponsors) A randomized double-blind placebo-controlled parallel group phase III study ofthe efficacy tolerability and safety of ketoprofen topicalpatch 20 (KTP) in the treatment of pain associated withgrade 1 or grade 2 ankle sprain or strain clinicaltrialsgovct2showNCT00351104 (accessed 3 February 2015) 2008[CTG NCT00351104]
NCT00352625 unpublished data only
Endo Pharmaceuticals (Sponsors) A randomized double-blind placebo-controlled parallel group phase III study ofthe efficacy tolerability and safety of ketoprofen topicalpatch 20 (KTP) in the treatment of pain associatedwith tendonitis or bursitis of the shoulder elbow or kneeclinicaltrialsgovct2showNCT00352625 (accessed 3February 2015) 2008 [CTG NCT00352625]
NCT00426985 unpublished data only
Endo Pharmaceuticals (Sponsors) A randomized double-blind placebo-controlled parallel group phase iii study ofthe efficacy tolerability and safety of ketoprofen topicalpatch 20 (KTP) in the treatment of pain associatedwith tendonitis or bursitis of the shoulder elbow or kneeclinicaltrialsgovct2showNCT00426985 (accessed 3February 2015) 2008 [CTG NCT00426985]
NCT00640705 unpublished data only
Cerimon Pharmaceuticals (Sponsors) A randomizeddouble-blind placebo-controlled study of the efficacyand safety of a diclofenac sodium patch for the topicaltreatment of pain due to mild to moderate ankle sprainclinicaltrialsgovct2showNCT00640705 (accessed 3February 2015) 2008 [CTG NCT00640705]
NCT00640939 unpublished data onlyCerimon Pharmaceuticals (Sponsors) A randomizeddouble-blind placebo-controlled study of the efficacy andsafety of a diclofenac sodium patch for the topical treatmentof pain due to mild to moderate tendonitis or bursitisclinicaltrialsgovct2showNCT00640939 (accessed 3February 2015) 2008 [CTG NCT00640939]
NCT00680472 unpublished data onlyHisamitsu Pharmaceutical Co Inc (Sponsors) Arandomized multicenter double-blind placebo-controlledtwo-week study to assess the efficacy and safety of HKT-500 in subjects with acute shoulder pain clinicaltrialsgovct2showNCT00680472 (accessed 3 February 2015) 2015[CTG NCT00680472]
NCT00680784 unpublished data only
Hisamitsu Pharmaceutical Co Inc (Sponsors) Arandomized multicenter double-blind placebo-controlledstudy to evaluate the efficacy and safety of HKT-500 inthe treatment of pain associated with Grade I or Grade IIankle sprain clinicaltrialsgovct2showNCT00680784(accessed 3 February 2015) 2015 [CTG NCT00680784]
NCT00765700 unpublished data onlyEkman E Skrepnik S Jones M Lawson K Schupp JEfficacy and safety of ketoprofen 10 cream in acute softtissue injuries (phase 3 study tdlp-110-001) Proceedings ofthe 13th World Congress on Pain 2010 Aug 29 - Sept 2Montreal CanadaImprimis Pharmaceuticals Inc (Sponsors) A randomizedmulticenter double-blind placebo-controlled parallel-group phase 3 study to assess the efficacy and safetyof Ketotransdeltrade (Ketoprofen topical cream 10) inthe treatment of pain associated with mild to moderateacute soft tissue injury clinicaltrialsgovct2showNCT00765700 (accessed 3 February 2015) 2013 [CTGNCT00765700]
NCT00869063 unpublished data onlyCerimon Pharmaceuticals (Sponsors) A randomizeddouble-blind placebo-controlled study of the efficacy andsafety of a diclofenac sodium patch for the topical treatmentof acute pain due to mild to moderate wrist sprain strainor contusion clinicaltrialsgovct2showNCT00869063(accessed 3 February 2015) 2010 [CTG NCT00869063]
NCT00869180 unpublished data onlyCerimon Pharmaceuticals (Sponsors) A randomizeddouble-blind placebo-controlled study of the efficacy andsafety of a diclofenac sodium patch for the topical treatmentof acute pain due to mild to moderate ankle sprainclinicaltrialsgovct2showNCT00869180 (accessed 3February 2015) 2010 [CTG NCT00869180]
33Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT00931866 unpublished data only
Cerimon Pharmaceuticals (Sponsors) A randomizeddouble-blind placebo-controlled study of the efficacy andsafety of a diclofenac sodium patch for the topical treatmentof acute pain due to mild to moderate soft tissue injuriesclinicaltrialsgovct2showNCT00931866 (accessed 3February 2015) 2010 [CTG NCT00931866]
NCT01874626 unpublished data only
Strategic Science amp Technologies LLC (Sponsors) Arandomized double-blind placebo-controlled multi-dosepivotal study to determine the efficacy and safety of SST0225 a topical ibuprofen cream in the treatment of painassociated with acute ankle sprain clinicaltrialsgovct2showNCT01874626 (accessed 3 February 2015) 2014[CTG NCT01874626]
NCT01957215 unpublished data only
GlaxoSmithKline (Sponsors) A clinical study to assess theefficacy of pain relief of topical indomethacin patch overplacebo in ankle sprain patients clinicaltrialsgovct2showNCT01957215 (accessed 3 February 2015) 2015 [CTGNCT01957215]
NCT02324270 unpublished data only
Actavis Inc (Sponsors) Randomized double-blindmultiple-center placebo-controlled study comparing thesafety and efficacy of generic diclofenac epolamine toFlectorreg patch in the treatment of acute pain due to minorankle sprain clinicaltrialsgovct2showNCT02324270(accessed 3 February 2015) 2014 [CTG NCT02324270]
Sarzi-Puttini 2014 published data only
Sarzi-Puttini P Atzeni F Damiani C Casale R BarbagalloM Cazzola M A double-blind randomised parallelgroup active controlled multicentre study to assess thetherapeutic non-inferiority of SKP-021 a 03 ketoprofenpatch versus diclofenac sodium patch in patients withacute inflammatory musculoskeletal injuries Proceeding ofthe Annual European Congress of Rheumatology of theEuropean League Against Rheumatism EULAR 2014 June11-14 Paris France 201473
References to ongoing studies
NCT01945034 unpublished data onlyPfizer (Sponsors) Placebo-controlled double-blindevaluation of the efficacy and safety of ibuprofen 5 topicalgel for the treatment of ankle sprain clinicaltrialsgovct2showNCT01945034 (accessed 3 February 2015) 2014[CTG NCT01945034]
NCT02100670 unpublished data onlyGlaxoSmithKline (Sponsors) A clinical study to assessthe efficacy and onset of pain relief of topical MFC51123diclofenac-menthol gel versus controls in ankle sprainclinicaltrialsgovct2showNCT02100670 (accessed 3February 2015) 2014 [ClinicalTiralsgov NCT02100670]
NCT02290821 unpublished data only
Novartis (Sponsors) A randomized double-blind placebo-controlled parallel group study to evaluate the efficacy andsafety of diclofenac sodium topical gel (DSG) 1 applied
four times daily in subjects with acute blunt soft tissueinjuriescontusions of the limbs clinicaltrialsgovct2showNCT02290821 (accessed 3 February 2015) 2015 [CTGNCT02290821]
Additional references
Anon 2005
Anon Topical analgesics a review of reviews and a bit ofperspective wwwjr2oxacukBandolierExtraforbandoTopextra3pdf 2005
Cook 1995
Cook RJ Sackett DL The number needed to treat aclinically useful measure of treatment effect BMJ 1995310
(6977)452ndash4
Counsell 1994Counsell CE Clarke MJ Slattery J Sandercock PA Themiracle of DICE therapy for acute stroke fact or fictionalproduct of subgroup analysis BMJ 1994309(6970)1677ndash81
Dechartres 2013Dechartres A Trinquart L Boutron I Ravaud P Influenceof trial sample size on treatment effect estimates meta-epidemiological study BMJ 2013346f2304 [DOI101136bmjf2304]
Derry 2012a
Derry S Moore RA Topical capsaicin (low concentration)for chronic neuropathic pain in adults Cochrane Databaseof Systematic Reviews 2012 Issue 9 [DOI 10100214651858CD010111]
Derry 2012bDerry S Moore RA Rabbie R Topical NSAIDs forchronic musculoskeletal pain in adults Cochrane Databaseof Systematic Reviews 2012 Issue 9 [DOI 10100214651858CD007400pub2]
Derry 2013Derry S Sven-Rice A Cole P Tan T Moore RA Topicalcapsaicin (high concentration) for chronic neuropathic painin adults Cochrane Database of Systematic Reviews 2013Issue 2 [DOI 10100214651858CD007393pub3]
Derry 2014
Derry S Matthews PR Wiffen PJ Moore RASalicylate-containing rubefacients for acute and chronicmusculoskeletal pain in adults Cochrane Database ofSystematic Reviews 2014 Issue 11 [DOI 10100214651858CD007403pub3]
Evans 1995Evans JM McMahon AD McGilchrist MM White GMurray FE McDevitt DG et al Topical non-steroidal anti-inflammatory drugs and admission to hospital for uppergastrointestinal bleeding and perforation a record linkagecase-control study BMJ 1995311(6996)22ndash6
FitzGerald 2001
FitzGerald GA Patrono C The coxibs selective inhibitorsof cyclooxygenase-2 New England Journal of Medicine2001345(6)433ndash42
34Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Haroutiunian 2010
Haroutiunian S Drennan DA Lipman AG Topical NSAIDtherapy for musculoskeletal pain Pain Medicine 201011
(4)535ndash49 [DOI 101111j1526-4637201000809x]
Hawkey 1999
Hawkey CJ Cox-2 inhibitors Lancet 1999353(9149)307ndash14
Higgins 2011
Higgins JPT Altman DG Sterne JAC Chapter 8 Assessingrisk of bias in included studies In Higgins JPT GreenS (editors) Cochrane Handbook for Systematic Reviewsof Interventions Version 510 [updated March 2011]The Cochrane Collaboration 2011 Available fromwwwcochrane-handbookorg
Jadad 1996a
Jadad AR Carroll D Moore A McQuay H Developing adatabase of published reports of randomised clinical trials inpain research Pain 199666(2-3)239ndash46
Jadad 1996b
Jadad AR Moore RA Carroll D Jenkinson C ReynoldsDJ Gavaghan DJ et al Assessing the quality of reports ofrandomized clinical trials is blinding necessary ControlledClinical Trials 199617(1)1ndash12
LrsquoAbbeacute 1987
LrsquoAbbeacute KA Detsky AS OrsquoRourke K Meta-analysis inclinical research Annals of Internal Medicine 1987107(2)224ndash33
Makris 2010Makris UE Kohler MJ Fraenkel L Adverse effects oftopical nonsteroidal antiinflammatory drugs in older adultswith osteoarthritis a systematic literature review Journalof Rheumatology 201037(6)1236ndash43 [DOI 103899jrheum090935]
Mason 2004a
Mason L Moore RA Edwards JE Derry S McQuay HJTopical NSAIDs for acute pain a meta-analysis BMCFamily Practice 2004510
Mason 2004bMason L Moore RA Edwards JE Derry S McQuayHJ Topical NSAIDs for chronic musculoskeletal painsystematic review and meta-analysis BMC MusculoskeletalDisorders 2004528
McPherson 2013McPherson ML Cimino NM Topical NSAIDformulations Pain Medicine 201314 Suppl 1S35ndash9[DOI 101111pme12288]
MHRA 2009
Medicines and Healthcare Products Regulatory AgencyTopical ketoprofen reminder on risk of photo-sensitivityreactions Drug Safety Update 2009 June Vol 2 issue 116
Moore 1998aMoore RA Tramegraver MR Carroll D Wiffen PJ McQuayHJ Quantitative systematic review of topically applied non-
steroidal anti-inflammatory drugs BMJ 1998316(7128)333ndash8
Moore 1998bMoore RA Gavaghan D Tramegraver MR Collins SL McQuayHJ Size is everything - large amounts of information areneeded to overcome random effects in estimating directionand magnitude of treatment effects Pain 199878(3)209ndash16
Moore 2003
Moore RA Edwards J Barden J McQuay HJ BandolierrsquosLittle Book of Pain Oxford Oxford University Press 2003[ISBN 0ndash19ndash263247ndash7]
Moore 2006
Moore RA Barden J Derry S McQuay HJ Managingpotential publication bias In McQuay HJ Kalso EMoore RA editor(s) Systematic Reviews in Pain ResearchMethodology Refined Seattle IASP Press 200615ndash23[ISBN 978ndash0ndash931092ndash69ndash5]
Moore 2008aMoore RA Derry S McQuay HJ Topical agents in thetreatment of rheumatic pain Rheumatic Diseases Clinics ofNorth America 200834(2)415ndash32
Moore 2008bMoore RA Barden J Derry S McQuay HJ Managingpotential publication bias In McQuay HJ Kalso EMoore RA editor(s) Systematic Reviews in Pain ResearchMethodology Refined Seattle IASP Press 200815ndash24[ISBN 978ndash0ndash931092ndash69ndash5]
Moore 2013
Moore RA Straube S Aldington D Pain measures andcut-offs - rsquono worse than mild painrsquo as a simple universaloutcome Anaesthesia 201368(4)400ndash12 [DOI 101111anae12148]
Morris 1995
Morris JA Gardner MJ Calculating confidence intervalsfor relative risk odds ratios and standardised ratios andrates In Gardner MJ Altman DG editor(s) Statistics withConfidence - Confidence Intervals and Statistical GuidelinesLondon British Medical Journal 199550ndash63
NICE 2008National Institute for Health and Care ExcellenceOsteoarthritis The care and management of osteoarthritisin adults 2008 wwwniceorgukCG059fullguideline
Nuumlesch 2010Nuumlesch E Trelle S Reichenbach S Rutjes AW TschannenB Altman DG et al Small study effects in meta-analysesof osteoarthritis trials meta-epidemiological study BMJ2010341c3515 [DOI 101136bmjc3515]
PACT 2014
Prescribing and Primary Care team Health and SocialCare Information Centre Prescrption Cost AnalysisEngland 2013 Health and Social Care Information CentreThe NHS Information Centre 2014298ndash301 [ISBN978ndash1ndash78386ndash089ndash0]
35Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Quane 1998
Quane PA Graham GG Zeigler JB Pharmacology ofbenzydamine Inflammopharmacology 19986(2)95ndash107
RevMan 2014The Nordic Cochrane Centre The Cochrane CollaborationReview Manager (RevMan) 53 Copenhagen The NordicCochrane Centre The Cochrane Collaboration 2014
Roth 2012Roth SH Fuller P Pooled safety analysis of diclofenacsodium topical solution 15 (ww) in the treatment ofosteoarthritis in patients aged 75 years or older ClinicalInterventions in Aging 20127127ndash37 [DOI 102147CIAS30884]
Simon 2009
Simon LS Grierson LM Naseer Z Bookman AA ZevShainhouse J Efficacy and safety of topical diclofenaccontaining dimethyl sulfoxide (DMSO) compared withthose of topical placebo DMSO vehicle and oral diclofenacfor knee osteoarthritis Pain 2009143(3)238ndash45 [DOI101016jpain200903008]
Tramer 1997
Tramer MR Reynolds DJ Moore RA McQuay HJ Impact
of covert duplicate publication on meta-analysis a casestudy BMJ 1997315(7109)635ndash40
Tugwell 2004Tugwell PS Wells GA Shainhouse JZ Equivalence studyof a topical diclofenac solution (pennsaid) compared withoral diclofenac in symptomatic treatment of osteoarthritisof the knee a randomized controlled trial Journal ofRheumatology 200431(10)2002ndash12
Zimmerman 1995
Zimmerman J Siguencia J Tsvang E Upper gastrointestinalhemorrhage associated with cutaneous application ofdiclofenac gel American Journal of Gastroenterology 199590(11)2032ndash4 [PUBMED 7485017]
References to other published versions of this review
Massey 2010
Massey T Derry S Moore RA McQuay HJ TopicalNSAIDs for acute pain in adults Cochrane Databaseof Systematic Reviews 2010 Issue 6 [DOI 10100214651858CD007402pub2]
lowast Indicates the major publication for the study
36Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
C H A R A C T E R I S T I C S O F S T U D I E S
Characteristics of included studies [ordered by study ID]
Airaksinen 1993
Methods R DB PC parallel groupsGel applied to the painful area twice daily for 7 daysAssessment at baseline 3 7 days
Participants Minor soft tissue injuries (lt 7 days)N = 56M 45 F 11Age not reportedMean baseline pain at rest 25-26 mm
Interventions Ketoprofen gel 2 x 5 g (125 mg) daily n = 29Placebo gel n = 27Rescue medication paracetamol 500 mgNo other treatment allowed
Outcomes PGE 5-point scale but reported as ldquoimprovedrdquo or ldquosame or worserdquo (responder = ldquoim-provedrdquo)Improvement in pain with movement 100 mm VAS reported as group meanAdverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Unclear risk No mention of early withdrawals ormethod of imputation
Size High risk lt 50 participants per treatment group
37Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Aoki 1984
Methods R DB PC AC parallel groupsGel applied to affected area 3 or 4 times daily with no occlusion for 7 daysAssessment at baseline 3 7 days
Participants Acute orthopaedic trauma (contusion distortion fracture lt 7 days)N = 252 (203 analysed for efficacy)M 98 F 105Age range 8 to 86 years 13 younger than 20 yearsBaseline pain mild in 35Exclusions 23 protocol violations 26 reasons ldquonot relatedrdquo to drug Equally distributedbetween groups
Interventions Piroxicam gel 05 1 g 3 to 4 x daily n = 84Indomethacin gel 1 1 g 3 to 4 x daily n = 84Placebo gel n = 84No other medication or initiation of physical therapy allowed
Outcomes PGE 5-point scale (responder = ldquobetterrdquo and ldquomuch betterrdquo)Adverse eventsWithdrawals and exclusions
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Low risk ldquokey code sealed until end of studyrdquo
Blinding (performance bias and detectionbias)All outcomes
Low risk Gels in ldquoidentical tubesrdquo
Incomplete outcome data (attrition bias)All outcomes
Unclear risk gt 10 withdrawals ldquounrelated to treat-mentrdquo and for ldquoprotocol violationsrdquo Nofurther details but no significant differ-ences between groups
Size Unclear risk 50 to 200 participants per treatment group
38Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Auclair 1989
Methods R DB PC parallel groupsGel massaged into skin over affected heel 3 times daily after cleaning with soap and waterfor up to 21 daysAssessment at baseline 7 21 days
Participants Acute Achilles heel tendinitis (not associated with continuous pain at rest or gt 1 monthhistory)N = 243 (227 analysed for efficacy)MF not reportedMean age 29 yearsBaseline pain ~ 10 had lt 26 mm on palpation of tendon ~ 30 had mild or no painon dorsiflexion of footExclusions failure to meet inclusion criteria major protocol violations failure to takestudy medication for full duration
Interventions Niflumic acid gel 25 3 x 5 g daily n = 117Placebo gel n = 110No other analgesics and anti-inflammatories physiotherapy or supportive measures al-lowed
Outcomes PGE 5-point scale (responder = ldquogoodrdquo or ldquovery goodrdquo)Pain improved or disappeared on dorsiflexionAdverse eventsWithdrawals and exclusions
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Unclear risk 10 excluded for ldquofailing to meet entry cri-teria and protocol violationsrdquo No furtherdetails
Size Unclear risk 50 to 200 participants per treatment group
39Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Billigmann 1996
Methods R DB PC parallel groupsGel applied 3 times daily with rubbingAssessed at baseline 3 5 7 days
Participants Distortion of ankle jointN = 160M and FAge 18+ yearsBaseline pain not reported
Interventions Ibuprofen microgel 5 3 x 10 cm (= 200 mg) daily n = 80Placebo gel n = 80
Outcomes Pain with movement VAS (responder = decreased by 20)Complete remissionAdverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Unclear risk Insufficient data to assess
Size Unclear risk 50 to 200 participants per treatment group
Campbell 1994
Methods R DB PC parallel groupsCream applied 4 times daily for 7 days (up to 14 days optional)Self assessed using daily diary for 7 days and up to 14 days
Participants Acute ankle sprain (lt 24 hours no fracture)N = 100 (51 analysed)M 33 F 18Mean age 29 years
40Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Campbell 1994 (Continued)
Baseline pain at rest gt 35 mm on walking 80 mmExclusions did not return diaries protocol exclusions (25 ibuprofen 24 placebo)
Interventions Ibuprofen cream 5 (Proflex) 4 x 4rdquo (10 cm) daily n = 26Placebo cream n = 25Advised to use rest and regular icing for 48 hours then walking and exerciseRescue medication paracetamol
Outcomes Improvement in walking ability 4-point scale (responder = ldquoimprovementrdquo)Pain on walking 100 mm VAS (mean data)Withdrawals and exclusions
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Low risk Randomisation carried out by sponsorTubes dispensed by hospital pharmacy whoheld the codes
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical creamrdquo
Incomplete outcome data (attrition bias)All outcomes
High risk gt 40 lost to follow-up Approximatelyequal between groups
Size High risk lt 50 participants per treatment arm as anal-ysed
Chatterjee 1977
Methods R DB PC parallel groupsCream applied to site of injury 3 times daily for 6 daysAssessment at baseline 2 6 days
Participants Soft tissue injuries (recent)N = 51MF not reportedAge not reportedBaseline pain on passive movement moderate or severe in all but 3 participants
41Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Chatterjee 1977 (Continued)
Interventions Benzydamine HCl cream 3 3 x daily n = 25Placebo cream n = 25(5 active 6 placebo participants also received ultrasound)No other topical agent allowed
Outcomes Pain on passive movement 4-point scale (responder = ldquoabsentrdquo or ldquoslightrdquo)Tenderness with pressure 4-point scale (responder = ldquoabsentrdquo or ldquoslightrdquo)Adverse eventsWithdrawals and exclusions
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk ldquopredetermined randomised schedulerdquo
Allocation concealment (selection bias) Low risk Sealed copy of schedule held by investiga-tor and duplicate copy kept by clinical trialco-ordinator Looked at only in event ofadverse reaction (not necessary)
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoindistinguishable in appearance andconsistencyrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk 2 lost to follow-up no withdrawals dueto adverse events Responder analysis
Size Unclear risk lt 50 participants per treatment arm
Costantino 2011
Methods R DB PC multicentre parallel groupPlaster applied daily for 7 daysPI assessment daily overall treatment efficacy and tolerability assessed at 3 and 7 days
Participants Grade I or II ankle sprain (lt 48 hours) with lateral external ligament involvementPI on movement ge 50100 oedema ge 20 mm difference between anklesN = 430M 249 F 175 (for analysis)Mean age 35 yearsBaseline PI on movement 72100 (SD 12)
42Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Costantino 2011 (Continued)
Interventions DHEPhep n = 142DHEP n = 146Placebo n = 142Rescue medication paracetamol to maximum 3 g dailyNo other treatments allowed
Outcomes Mean change in PI on movement from baseline to 3 daysMean reduction in oedema at 3 daysTolerability at 3 daysRescue medicationAdverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk Computer-generated
Allocation concealment (selection bias) Unclear risk ldquosealed envelopesrdquo
Blinding (performance bias and detectionbias)All outcomes
Low risk Shape colour size and application methodidentical for all plasters
Incomplete outcome data (attrition bias)All outcomes
Unclear risk Very few withdrawals but used LOCF
Size Unclear risk 50 to 200 participants per treatment arm
Coudreuse 2010
Methods R DB PC multicentre parallel groupPlaster applied daily for 7 daysPI assessed twice daily over 3 days then day 7 Overall treatment efficacy and tolerabilityassessed on days 3 7
Participants Ankle sprain (lt 48 hours) with lateral external ligament involvementPI on movement ge 50100 oedema ge 20 mm difference between anklesN = 240 (233 for analysis)M 148 F 86 6 unknown
Interventions DHEPhep n = 120Placebo n = 120Rescue medication paracetamol to maximum 4 g daily
43Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Coudreuse 2010 (Continued)
No other treatments allowed
Outcomes Global efficacy at 7 days 4-point scale (responder = ldquoexcellentrdquo or ldquogoodrdquo)Mean change in PI on movement at 6 hours and 7 daysMean change in oedema at 3 and 7 daysTolerabilityAdverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk Computer-generated
Allocation concealment (selection bias) Unclear risk ldquosealed envelopesrdquo
Blinding (performance bias and detectionbias)All outcomes
Low risk Appearance and odour identical for all plas-ters
Incomplete outcome data (attrition bias)All outcomes
Low risk ITT analysis lt 5 excluded for missingdata equal between groups
Size Unclear risk 50 to 200 participants per treatment arm
Curioni 1985
Methods R DB PC AC parallel groupsGel rubbed into affected area until absorbed twice daily for 10 daysAssessed at baseline and daily to 10 days
Participants Acute soft tissue injuriesN = 60M 33 F 27Median age 33 yearsBaseline pain not given
Interventions Ibuproxam gel 10 n = 20Ketoprofen gel n = 20Etofenamate gel n = 20
Outcomes PGE 4-point scale (ldquogoodrdquo or ldquoexcellentrdquo)Resolution of symptomsAdverse events
44Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Curioni 1985 (Continued)
Withdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk Medication supplied in identical tubes
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
Diebshlag 1990
Methods R DB PC AC parallel groupsGel applied 3 times daily without occlusion for 14 daysAssessment at baseline 2 3 4 8 15 days
Participants Ankle sprain (lt 24 hours)N = 37M 24 F 13Mean age 28 yearsBaseline pain slight to moderate
Interventions Ketorolac gel 2 3 x 3 g daily n = 13Etofenamate gel 5 3 x 3 g daily n = 12Placebo gel n = 12Rescue medication paracetamolNo other analgesic or anti-inflammatory medication ice packs or physiotherapy allowed
Outcomes Reduction in PI 100 mm VAS and 4-point scale (responder = ldquoimprovedrdquo)Adverse eventsWithdrawals and exclusions
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
45Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Diebshlag 1990 (Continued)
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Low risk ldquoMedication assignment supplied in asealed enveloperdquo Opened only if seriousparticipant event necessitation treatmentdisclosure occurred (not necessary)
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical appearancerdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
Dreiser 1988
Methods R DB PC parallel groupsCream applied 3 times dailyAssessment at baseline 7 days
Participants Acute tendinitis (lt 1 month)N = 64M 35 F 25Mean age 36 yearsBaseline spontaneous pain ge 60 mm
Interventions Ibuprofen cream 5 3 x 4 cm daily n = 32 (3 x 10 cm for large joints)Placebo cream n = 32No other topical systemic or physical treatment allowed
Outcomes PGE scale not reported (responder = ldquoimprovementrdquo or ldquocomplete relief rdquo)Improvement in pain VAS (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35Oxford Validity Score 1016
Risk of bias
Bias Authorsrsquo judgement Support for judgement
46Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Dreiser 1988 (Continued)
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Low risk Missing participants added back usingBOCF
Size High risk lt 50 participants per treatment arm
Dreiser 1989
Methods R DB PC parallel groupsGel applied twice daily to affected area with light massage then covered with standardcompressAssessed at baseline 3 7 days
Participants Uncomplicated recent ankle sprainN = 60M 36 F 24Mean age 33 yearsMean baseline pain 54 mm
Interventions Ketoprofen gel 25 2 x 5 cm daily n = 30Placebo gel n = 30No concomitant therapy other than simple oral analgesia allowed
Outcomes PGE 3-point scale (responder = ldquobetterrdquo)Improvement in pain VAS (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk ldquodrawing lotsrdquo
Allocation concealment (selection bias) Unclear risk Not described
47Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Dreiser 1989 (Continued)
Blinding (performance bias and detectionbias)All outcomes
Low risk Treatments ldquoidentical in every way exceptthat placebo did not contain active princi-plerdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participant per treatment arm
Dreiser 1990
Methods R DB PC parallel groupsGel lightly massaged into skin over affected area 3 times daily then covered with standardcompressAssessed at baseline 3 7 days
Participants Uncomplicated ankle sprain (lt 4 days)N = 60 (59 analysed)M 29 F 29 (not stated for 1 participant)Mean age 33 yearsBaseline pain ge moderately severeExclusions 1 participant had only moderate pain at baseline
Interventions Niflumic acid gel 25 3 x 5 g daily n = 30Placebo gel n = 30Concomitant treatment with systemic NSAIDs local therapies or physiotherapy werenot allowed
Outcomes PGE 4-point scale (responder = ldquocuredrdquo or ldquoimprovedrdquo)Improvement in pain VAS (mean data)Adverse eventsWithdrawals and exclusions
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
48Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Dreiser 1990 (Continued)
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
Dreiser 1994
Methods R DB PC parallel groupsPatch applied twice dailyAssessed at baseline 3 7 days
Participants Traumatic ankle sprain (lt 2 days)N = 131M 84 F 47Mean age 34 yearsBaseline pain ge 50 mm
Interventions Flurbiprofen patch 2 x 40 mg daily n = 65Placebo patch n = 66Rescue medication paracetamol Ice or light restraint allowedExclusions 1 from flurbiprofen group for protocol violation
Outcomes PGE 4-point scale (responder = ldquogoodrdquo or ldquovery goodrdquo)Improvement in pain VAS (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk Placebo patch was ldquonon-medicated (butotherwise identical)rdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
49Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Fioravanti 1999
Methods R DB AC parallel groupsGel lightly massaged into skin 3 times daily and kept dry for 6 to 8 hoursAssessed at baseline 3 10 days
Participants Peri and extra-articular inflammatory diseasesN = 100M 32 F 68Mean age 49 yearsBaseline spontaneous pain ge 40 mm
Interventions DHEP lecithin gel 3 x 5 g (= 65 mg) daily n = 50DHEP gel 3 x 5 g (= 65 mg) daily n = 50
Outcomes PGE 4-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo)Pain on movement meanAdverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size Unclear risk 50 participants per treatment arm
Fujimaki 1985
Methods R DB PC AC parallel groupsGel applied to affected area 3 or 4 times daily with no occlusion for up to 14 daysAssessed at baseline 7 14 days
Participants Muscle pain or inflammation in neck shoulder back chest and upper and lower ex-tremities or a combinationN = 271 (247 analysed)M 97 F 149
50Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Fujimaki 1985 (Continued)
Age lt 20 to 89 yearsBaseline pain mostly mild to moderateExclusions 24 due to protocol violations loss to follow-up
Interventions Piroxicam gel 05 1 g 3 to 4 x daily n = 92Indomethacin gel 1 1 g 3 to 4 x daily n = 90Placebo gel n = 89No concomitant oral or topical analgesic or anti-inflammatory medication allowed Nophysical therapy initiated after start of study
Outcomes PGE 5-point scale (responder = ldquobetterrdquo or ldquomuch betterrdquo)Physician rated improvement 5-point scale (responder = ldquomarked improvementrdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total =45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Low risk Cartons numbered randomly and numbersheld in a key code until study completion
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical tubesrdquo packed in numbered car-ton Gel bases slightly different in appear-ance so dispensing physician did not haveaccess to them
Incomplete outcome data (attrition bias)All outcomes
Unclear risk ~ 8 excluded from analysis for unknownreasons or lost to or inadequate follow-upApproximately equal between groups
Size Unclear risk 50 to 200 participants per treatment arm
Gallacchi 1990
Methods R DB AC parallel groupsGel applied to affected area 4 times daily with light massage for 14 daysAssessment at baseline 7 14 days
Participants Painful inflammatory conditionsN = 50M 20 F 30Mean age 50 years
51Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Gallacchi 1990 (Continued)
Baseline pain ge moderate severity
Interventions Diclofenac gel 1 2 g 4 x daily n = 25 (Flector)Diclofenac sodium 1 2 g 4 x daily n = 25 (Voltaren Emugel)No other medication that could interfere with test drugs allowed
Outcomes PGE 5-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo)Improvement in pain on pressure 4-point scale (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
Gonzaacutelez de Vega 2013
Methods R DB (for gels) SB (for ointment) AC multicentre parallel groups2 g (~ 6 cm of gel) applied over injured area x 3 daily for 14 daysAssessments at 0 4 7 14 days of treatment and day 42 follow-up
Participants Grade I or II ankle sprain (lt 24 hours) with lateral external ligament involvement PIon weight bearing ge 30100N = 449M = 308 F = 112 (for analysis)Mean age 28 years
Interventions Traumeel gel 3 x 2 g daily n = 140Traumeel ointment n = 143 (not analysed in this review)Diclofenac gel 1 3 x 2 g daily n = 137Rescue medication paracetamol to maximum 2 g daily
52Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Gonzaacutelez de Vega 2013 (Continued)
Outcomes Pain-free on day 7Global efficacy 5-point scale (responder = ldquogoodrdquo and ldquovery goodrdquo) on day 14Normal function on day 14 yes or no (responder = ldquoyesrdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 (gels) W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk Computer-generated
Allocation concealment (selection bias) Low risk Central allocation kits assigned in orderreceived and used envelopes (no furtherdetails)
Blinding (performance bias and detectionbias)All outcomes
Low risk For gel comparison ldquoidentical containersrdquo
Incomplete outcome data (attrition bias)All outcomes
Unclear risk Used LOCF Most withdrawals due to earlyrecovery (within 14 days) approximatelyequal between groups
Size Unclear risk 50 to 200 participants per treatment arm
Governali 1995
Methods R DB AC parallel groupsGel or cream applied 3 times daily for up to 14 daysAssessed at baseline 7 14 days
Participants Soft tissue injuries + 2 fracturesN = 30M = 21 F = 9Median age 38 yearsMean baseline pain on movement moderate to severe (28 scale 0 to 4)
Interventions Ketoprofen gel 5 3 x 2 to 3 g daily n = 15Ketoprofen cream 1 3 x 2 to 3 g daily n = 15
Outcomes PGE 5-point scale (responder = ldquogoodrdquo and ldquoexcellentrdquo)Adverse eventsWithdrawals
53Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Governali 1995 (Continued)
Notes Oxford Quality Score R1 DB1 W1 Total =35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Treatments were given in identical tubesand measurements made by blinded ob-servers but one was a cream and the othera gel
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
Gualdi 1987
Methods R DB AC parallel groupsGel applied twice daily for 10 daysAssessed at baseline 4 7 10 days
Participants Soft tissue injuriesN = 60M = 37 F = 23Mean age 32 years (range 13 to 78)Mean baseline pain on movement moderate to severe (22 scale 0 to 3)
Interventions Flunoxaprofen gel 2 x 3 to 5 cm daily n = 30Ketoprofen gel 2 x 3 to 5 cm daily n = 30
Outcomes Improvement in pain on pressure (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
54Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Gualdi 1987 (Continued)
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Unclear risk Insufficient data
Size High risk lt 50 participants per treatment arm
Haig 1986
Methods R DB PC parallel groupsCream applied lightly to affected area 6 times daily for 6 daysAssessed at baseline 2 4 6 days
Participants Soft tissue injuries (lt 24 hours)N = 43MF not reportedAge not reportedBaseline pain not reported
Interventions Benzydamine cream 3 6 x daily n = 21Placebo cream n = 22
Outcomes Pain on movement 4-point scale (responder = ldquoimprovedrdquo)Adverse events
Notes Oxford Quality Score R1 DB2 W0 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not reported
Allocation concealment (selection bias) Unclear risk Not reported
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquomatching placebordquo
55Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Haig 1986 (Continued)
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants apparently included inanalysis
Size High risk lt 50 participants per treatment arm
Hoffmann 2012
Methods R DB PC multicentre parallel groupPlaster applied x 1 daily (minimum of 20 consecutive hours of application per day) for14 daysPI assessed daily overall efficacy assessed at 7 and 14 days
Participants Unilateral mild to moderate muscle contusion of upper or lower limb (lt 72 hours) PIge 50100 superficial haematoma le 10 x 14 cm at injured siteN = 354M 126 F 228Mean age 39 years
Interventions DHEPhep plaster x 1 daily n = 121DHEP plaster x 1 daily n = 115Placebo plaster n = 118Rescue medication paracetamol 500 mg (no limit reported)
Outcomes PGE 5-point scale (responder = excellent or good) at 14 daysMean reduction in PI at 3 and 8 daysRescue medicationAdverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Method not reported
Allocation concealment (selection bias) Unclear risk Method not reported ldquoMedication packedand labelled in order to render all partici-pants and personnel fully blinded to treat-ment administeredrdquo
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoIndistinguishable regarding appearanceshape colour size and odourrdquo
56Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Hoffmann 2012 (Continued)
Incomplete outcome data (attrition bias)All outcomes
Unclear risk No responder analysis reported and impu-tation method not mentioned
Size Unclear risk 50 to 200 participants per treatment arm
Hofman 2000
Methods R DB AC parallel groupsGel applied to affected region 4 times daily with gentle massageAssessed at baseline 8 days in clinic and daily diary
Participants Soft tissue articular pain (le 15 days)N = 142M 19 F 123Mean age 57 yearsMean baseline PI moderate to severe
Interventions Diclofenac sodium gel 1 4 x 2 cm daily n = 69Lysine clonixinate gel 5 4 x 2 cm daily n = 73(2 cm = 225 mg)No other analgesic local treatment (including immobilisation bandaging) or acupunc-tureRescue mediation allowed after 2 applications if needed
Outcomes PGE 3-point scale (ldquogoodrdquo)PI participant diary (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk Diclofenac gel repackaged to maintain DBwith lysine clonixinate gel Minor differ-ences between gels only apparent when di-rectly compared
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
57Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Hofman 2000 (Continued)
Size Unclear risk 50 to 200 participants per treatment arm
Hosie 1993
Methods R DB AC parallel groupsFoam (approximately the size of a golf ball) applied and 1 tablet taken 3 times daily for7 days and up to 14 daysAssessed at baseline 7 14 (if necessary) days
Participants Acute lower back injury (lt 1 month)N = 287 (261 analysed for efficacy)M 151 F 136Mean age 37 years (range 18 to 63)Most participants had moderate to severe pain on movement one had noneExclusions 25 lost to follow up one assessed at 14 days but not 7 days
Interventions Felbinac foam 3 3 x 2 g daily + placebo tablets 3 x 1 daily n = 140 (127 analysed forefficacy)Ibuprofen tablets 3 x 400 mg daily + placebo foam 3 x 2 g daily n = 147 (134 analysedfor efficacy but one had no pain at baseline)No other oral injectable or topical analgesic or anti-inflammatory medication Ongoingphysiotherapy to continue without change
Outcomes Pain on movement 5-point scale (responder = ldquononerdquo or ldquomildrdquo)Spontaneous pain 5-point scale (responder = ldquononerdquo or ldquomildrdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquodouble dummyrdquo
Incomplete outcome data (attrition bias)All outcomes
Unclear risk Imputation method not mentioned
Size Unclear risk 50 to 200 participants per treatment arm
58Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Jenoure 1997
Methods R DB PC parallel groupsPlaster applied to skin over affected area twice daily and kept in place with an elasticbandageAssessed at baseline 7 14 days and after further 14 days without treatment
Participants Humero-radial epicondyl pain (tendinopathic) - nearly all tennis elbowN = 85M 54 F 31Mean age 45 yearsBaseline pain ldquomildrdquo in ~ 10 of placebo group and 29 of active group
Interventions DHEP plaster (Tissugel) x 2 daily n = 44Placebo plaster x 2 daily n = 41
Outcomes Pain on pressure 5-point scale (responder = ldquononerdquo or ldquomildrdquo)Spontaneous pain 5-point scale (responder = ldquono painrdquo)Adverse events
Notes Oxford Quality Score R1 DB2 W0 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not reported
Allocation concealment (selection bias) Unclear risk Not reported
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical characteristicsrdquo
Incomplete outcome data (attrition bias)All outcomes
Unclear risk No useable data
Size High risk lt 50 participants per treatment arm
Joussellin 2003
Methods R DB PC parallel groupsPlaster applied to skin over affected area once dailyAssessed at baseline 1 2 3 7 days
Participants Ankle sprain (lt 48 hours)N = 134M 72 F 62Age range 18 to 65 years
59Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Joussellin 2003 (Continued)
Baseline spontaneous pain ge 50 mm
Interventions DHEP plaster (Flector Tissugel 1) x 1 daily n = 68Placebo plaster x 1 daily n = 66Rescue medication paracetamol
Outcomes PGE 4-point scale (responder = ldquoexcellentrdquo)Pain on movement VAS (mean)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoIdentical in size appearance and usedsame formula as active patch without ac-tive ingredientrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in responder anal-ysis
Size Unclear risk 50 to 200 participants per treatment arm
Julien 1989
Methods R DB PC parallel groupsGel applied to affected area twice daily with light massageAssessed at baseline 3 7 days in clinic and daily diary
Participants TendinitisN = 60M 29 F 31Mean age 41 yearsBaseline pain gt 50 mm
Interventions Ketoprofen gel 25 2 x 5 cm (= 50 mg) daily n = 30Placebo gel 2 x 5 cm daily n = 30No concomitant therapy other than simple analgesia
60Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Julien 1989 (Continued)
Outcomes PGE 4-point scale (responder = ldquoimprovedrdquo or ldquorecoveredrdquo)Pain on movement 4-point scale (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB1 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk Random numbers table
Allocation concealment (selection bias) Unclear risk Randomisation code supplied by Menarinilaboratories remote from allocation
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in responder anal-ysis
Size High risk lt 50 participants per treatment arm
Klainguti 2010
Methods R DB PC multicentre parallel groupPlaster applied x 1 daily (minimum 20 consecutive hours of application daily) for 10daysPI assessed daily Overall treatment efficacy and tolerability assessed at days 3 and 7
Participants Unilateral mild to moderate muscle contusion or strain of upper or lower limb (lt 72hours) superficial haematoma le 140 cm2 PI ge 40100N = 185M 90 F 95Mean age 39 years
Interventions DHEPhep plaster x 1 daily n = 65DHEP plaster x 1 daily n = 61Placebo plaster x 1 daily n = 59
Outcomes Overall treatment efficacy at 3 days 5-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo)Resolution of haematoma at 10 days yes or noRescue medicationAdverse eventsWithdrawals
61Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Klainguti 2010 (Continued)
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk Computer-generated
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoIdentical in size appearance and odourrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk Withdrawals lt 5 All participants in-cluded in analysis
Size Unclear risk 50 to 200 participants per treatment arm
Kockelbergh 1985
Methods R DB PC parallel groupsGel applied twice dailyAssessed at baseline 3 7 days
Participants Acute soft tissue trauma (lt 24 hours)N = 74M 60 F 14Mean age 27 yearsBaseline pain gt 65 mm
Interventions Ketoprofen gel 25 2 x 5 cm (= 15 mg) daily n = 38Placebo gel 2 x 5 cm daily n = 36No concomitant treatmentRescue medication glafenine
Outcomes PGE 3-point scale (responder = ldquogoodrdquo)Spontaneous pain 100 mm VAS (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
62Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Kockelbergh 1985 (Continued)
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in responder anal-ysis
Size High risk lt 50 participants per treatment arm
Kuehl 2011
Methods R DB PC multicentre parallel groupsPlaster applied x 2 daily for 14 days or until resolutionPI assessed daily Overall treatment efficacy assessed at 14 days
Participants Mild or moderate sprain strain or contusion (lt 7 days) PI ge 510N = 418M 206 F 212Mean age 39 years
Interventions DHEP patch x 2 daily n = 207Placebo patch x 2 daily n = 211Rescue medication no analgesic allowed (or icewrapping) use = discontinuation
Outcomes Resolution of injury (VAS le 210)Overall tolerability 5-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk Computer-generated
Allocation concealment (selection bias) Unclear risk Not described
63Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Kuehl 2011 (Continued)
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical in appearance with same contentexcept diclofenacrdquo
Incomplete outcome data (attrition bias)All outcomes
High risk Used LOCF high withdrawal rate (43)Did not report all efficacy outcomes mea-sured
Size Low risk gt 200 participants per treatment arm
Li 2013
Methods R DB PC multicentre parallel groupsPlaster applied x 2 daily for 7 daysPI assessed daily overall treatment efficacy and tolerability assessed at 7 days
Participants Mild or moderate ankle or knee sprain muscle strain or contusion (lt 72 hours) PI ge
50100N = 384M 144 F 240Mean age 42 years
Interventions DHEP plaster x 2 daily n = 192Placebo plaster x 2 daily n = 192Rescue medication paracetamol to maximum 2 g daily
Outcomes ge 50 reduction in PI at 7 daysOverall treatment efficacy 5-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo)Overall tolerability 5-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo)Rescue medicationAdverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk Computer-generated
Allocation concealment (selection bias) Unclear risk Method not fully described ldquosealed en-velopesrdquo ldquoconcealed from investigators andparticipantsrdquo
64Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Li 2013 (Continued)
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical in texture size color and odorrdquo
Incomplete outcome data (attrition bias)All outcomes
Unclear risk Withdrawals 2 Methods states LOCFbut appears to describe BOCF for with-drawals
Size Unclear risk 50 to 200 participants per treatment arm
Linde 1985
Methods R DB PC parallel groupsCream applied x 3 daily for 5 days with elastic support for the first 3 daysAssessed at baseline 4 8 days
Participants Sprained ankle (lt 24 hours)N = 100M 58 F 42Mean age 28 yearsBaseline pain all participants had ldquowalking painrdquo
Interventions Benzydamine 3 cream x 3 daily n = 50Placebo gel x 3 daily n = 50
Outcomes Pain on movement responder = ldquofree of walking painrdquoAdverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35Paper describes a benzydamine cream and a placebo gel
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described Paper describes a benzy-damine cream and a placebo gel
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in responder anal-ysis
65Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Linde 1985 (Continued)
Size Unclear risk 50 participants per treatment arm
Machen 2002
Methods R DB PC parallel groupsGel gently (ldquominimal rubrdquo not vigorously) massaged into skin over affected site untilabsorbed 3 times daily until symptoms disappeared or for maximum of 7 daysAssessment at baseline and once daily using diary cards to 7 days
Participants Soft tissue injury (lt 2 weeks and untreated)N = 85 (81 analysed)M 42 F 39Mean age 41 yearsBaseline pain gt 50 mm4 placebo participants lost to follow-up
Interventions Ibuprofen gel 5 x 3 daily n = 40Placebo gel x 3 daily n = 41Initiation of other medication or physiotherapy not allowed during study
Outcomes PGE 5-point scale (responder = ldquomarked improvementrdquo or ldquocomplete clearancerdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk Gels had similar physical characteristicsand were supplied in identical tubes
Incomplete outcome data (attrition bias)All outcomes
Low risk lt 5 missing participants Remaining par-ticipants included in responder analysis
Size High risk lt 50 participants per treatment arm
66Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Mahler 2003
Methods R DB AC parallel groupsGel applied with gentle massage to affected area 3 times daily without occlusion for 10daysAssessed at baseline 3 10 days in clinic and daily patient diary
Participants First-degree ankle or knee sprains first-degree muscle strains and mild-to-moderatecontusionsn = 100M 69 F 31Mean age 32 yearsMean baseline pain with activity ge 65 mm
Interventions DHEP lethicin gel 3 x 5 g (= 65 mg) daily n = 52DHEP gel 3 x 5 g (= 65 mg) daily n = 48All participants treated with ice at site of inflammation for first 48 hours but no immo-bilisation allowedRescue medication paracetamol 500 mg if strictly necessary
Outcomes PGE 4-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo)Pain on movement 100-mm VAS (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 5
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk ldquoComputer-generated randomization listrdquo
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk Pharmaceutically inert colouring agentsadded to reference formulation so that gelswere indistinguishable
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in responder anal-ysis
Size High risk lt 50 participants in each treatment arm
67Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Maziegraveres 2005a
Methods R DB PC parallel groupsNew patch applied directly to skin over painful area each morningAssessed at baseline 3 7 14 days
Participants Symptomatic tendonitis in upper or lower limbs not requiring surgery (le 15 days)N = 172M 72 F 100Mean age 46 yearsBaseline pain with activity ge 40 mm
Interventions Ketoprofen patch 100 mg x 1 daily n = 87Placebo patch x 1 daily n = 85No analgesic or steroid by any route or other topical medication or physical therapyallowedRescue medication permitted but not within 12 hours of assessment
Outcomes PGE 4-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk ldquocomputer generated global randomizationcoderdquo
Allocation concealment (selection bias) Low risk ldquoThe randomization list and code en-velopes were prepared by the company ap-pointed for clinical supplies packaging Therandom code was disclosed only after studycompletion and database closurerdquo
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquothe same indistinguishable patch with noingredientrdquo
Incomplete outcome data (attrition bias)All outcomes
High risk LOCF withdrawals 12 All participantsincluded in analysis
Size Unclear risk 50 to 200 participants per treatment arm
68Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Maziegraveres 2005b
Methods R DB PC parallel groupsNew patch applied directly to skin over painful area each morningAssessed at baseline 3 7 14 days
Participants Painful benign ankle sprain (le 48 hours)N = 163M 83 F 80Mean age 37 yearsBaseline spontaneous pain ge 50 mm
Interventions Ketoprofen patch 100 mg x 1 daily n = 81Placebo patch x 1 daily n = 82No analgesic or steroid by any route or other topical medication or physical therapyallowedRescue medication permitted but not within 12 hours of assessment
Outcomes PGE 4-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk ldquocomputer-generated global randomiza-tion coderdquo
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk The same transdermal patch with no activeingredient
Incomplete outcome data (attrition bias)All outcomes
Unclear risk LOCF withdrawals 5 to 6 All partic-ipants included in analysis
Size Unclear risk 50 to 200 participants per treatment arm
McLatchie 1989
Methods R DB PC parallel groupsGel applied to injured site 3 times daily for 7 daysAssessment at baseline 4 7 days at clinic daily patient diary
Participants Acute soft tissue injury (lt 48 hours)N = 231
69Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
McLatchie 1989 (Continued)
M 143 F 88Mean age 33 yearsBaseline pain moderate to severe
Interventions Felbinac gel 3 3 x 3 cm daily n = 118Placebo gel 3 x 3 cm daily n = 113Rescue medication paracetamol
Outcomes Patient diary mean changeAdverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquotubes identical in all aspectsrdquo
Incomplete outcome data (attrition bias)All outcomes
Unclear risk No usable efficacy data
Size Unclear risk 50 to 200 participants per treatment arm
Morris 1991
Methods R DB PC parallel groupsGel applied to site of injury 3 times daily for 7 daysAssessed at baseline 7 days at clinic and daily patient diary
Participants Acute soft tissue injury (lt 3 days)N = 100 (84 analysed for efficacy)M 70 F 14Mean age 25 yearsBaseline pain moderate to severeExclusions 1 participant in placebo group lost to follow-up 15 protocol violations
Interventions Felbinac gel 3 3 x 1 cm daily n = 41Placebo gel n = 43Ice joint immobilisation bandaging and compression allowed
70Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Morris 1991 (Continued)
No concomitant oral NSAID occlusive dressing physiotherapy or liniments allowedRescue medication paracetamol
Outcomes PGE 5-point scale (responder = ldquogoodrdquo and ldquovery goodrdquo)Change in PI patient diary 10 cm VAS (mean data)Adverse eventsWithdrawals and exclusions
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Low risk ldquoRandomisation was undertaken at theproduction facility and a sealed copy of thelist supplied to the investigator for refer-ence only in defined circumstancesrdquo
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical tubes and outer boxesrdquo ldquoplacebowas a similarly constituted gelrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All relevant participants included in analy-sis
Size High risk lt 50 participants per treatment arm
NCT01255423
Methods R DB PC parallel group studyGel applied 4 x daily
Participants Acute lateral ankle sprain Grade I to II le 12 hoursN = 206Mean (plusmn SD) age 31 years (plusmn 13)M 87 F 119Baseline PI not reported
Interventions Diclofenac sodium gel 1 x 4 daily n = 104Placebo gel n = 102
Outcomes Mean PI on movementAdverse events
71Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT01255423 (Continued)
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Method of blinding not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants accounted for
Size Unclear risk 50 to 200 participants per treatment arm
NCT01272934
Methods R DB PC parallel group studyGel applied 4 x daily
Participants Acute lateral ankle sprain Grade I to II le 12 hoursN = 205M 101 F 104Mean (plusmn SD) age 32 years (plusmn 11)
Baseline PI not reported
Interventions Diclofenac sodium gel 1 x 4 daily n = 102Placebo gel x 4 daily n = 103
Outcomes Mean PI on movementAdverse events
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
72Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT01272934 (Continued)
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Method of blinding not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants accounted for
Size Unclear risk 50 to 200 participants per treatment arm
NCT01272947
Methods R DB PC parallel groupsGel applied 4 x daily
Participants Acute blunt soft tissue injuriescontusions of the limbs lt 3 hoursN = 204M 101 F 103Mean (plusmn SD) age 30 years (plusmn 11)
Baseline PI not reported
Interventions Diclofenac sodium gel 1 x 4 daily n = 104Placebo gel x 4 daily n = 100
Outcomes Mean PI on movementAdverse events
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Method of blinding not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants accounted for
73Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT01272947 (Continued)
Size Unclear risk 50 to 200 participants per treatment arm
Noret 1987
Methods R DB PC parallel groupsGel applied twice daily for 7 daysAssessment at baseline 3 8 days
Participants Minor sports injuries (lt 24 hours)N = 98 (93 analysed)M 71 F 27Mean age 29 yearsBaseline pain gt 60 mm
Interventions Ketoprofen gel 25 2 x 5 cm daily (= 15 mg) n = 48Placebo gel n = 45No other treatment given
Outcomes PGE 4-point scale (responder = ldquogoodrdquo and ldquoexcellentrdquo)Spontaneous pain 100 mm VAS (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Low risk ldquoallocated according to a randomization listand a corresponding code in a sealed enve-loperdquo
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
74Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Parrini 1992
Methods R DB PC parallel groupsFoam (the size of a walnut or a 1-second spray) applied with massage 3 x daily for 7days
Participants Articular trauma strains distortionsN = 169M 94 F 75Mean age 37 yearsMean baseline pain on movement 31 (scale 1 to 4)
Interventions Ketoprofen foam 15 3 x 2 g (= 600 mg) daily n = 83Placebo foam 3 x 2 g n = 86
Outcomes Pain on movement 4-point scale (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB1 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk ldquopatients were randomised according to themethod of random numbersrdquo [translated]
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Unclear risk No usable efficacy data
Size Unclear risk 50 to 200 participants per treatment arm
Picchio 1981
Methods R DB AC parallel groupsCream applied with slight massage until completely absorbed 3 times daily for up to 16daysAssessed at baseline 4 8 12 16 days
Participants Acute sports injuriesN = 40M 24 F 16Mean age 22 years (range 12 to 46)Most participants had mild to moderate baseline pain (12 and 9 with slight pain on
75Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Picchio 1981 (Continued)
movement)
Interventions Ibuprofen gel 10 3 x daily n = 20Ketoprofen gel 1 3 x daily n = 20
Outcomes Pain on movement (responder = ldquononerdquo)Adverse events
Notes Oxford Quality Score R1 DB2 W0 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk ldquotubes were identical in appearancerdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
Predel 2004
Methods R DB PC parallel groupsNew patch applied to injured area twice daily for 7 days Contact of patch with humidityor water to be avoidedAssessment at baseline 3 7 days
Participants Traumatic blunt soft tissue injuries (lt 3 hours no treatment)N = 120M 73 F 47Mean age 32 yearsBaseline pain gt 60 mm
Interventions Diclofenac sodium patch 2 x daily (140 mg per patch) n = 60Placebo patch 2 x daily n = 60NSAIDs analgesics psychotropic agents other topical preparations and bandages notallowed
Outcomes PGE 4-point scale (responder = rdquogoodldquo and rdquoexcellentldquo)Pain on movement 10 cm VAS (mean data)
76Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Predel 2004 (Continued)
Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk rdquocomputer generated block randomisationlist
Allocation concealment (selection bias) Low risk An independent statistician produced ran-domisation list and an independent con-tract research organisation packaged med-ication according to list Nobody else hadaccess to the randomisation list until thedatabase was closed
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoThe placebo patch was visually indistin-guishable from the active patchrdquo To avoidunblinding due to different smell anystudy nurse involved with medication wasnot involved in outcome assessment
Incomplete outcome data (attrition bias)All outcomes
Low risk Withdrawals lt 1 All participants in-cluded in analysis
Size Unclear risk 50 to 200 participants per treatment arm
Predel 2012
Methods R DB PC parallel groupsTreatment for 7 daysAssessed at baseline 1 3 5 and 8 days
Participants Grade I or II ankle sprain with lateral external ligament involvement lt 12 hoursPI on movement ge 50100N = 242M 152 F 90Mean age 32 years
Mean baseline pain on movement = 75100
Interventions Diclofenac gel (Voltaren Emulgel 232) 2 x 5 cm + 1 x 5 cm placebo gel daily n = 80Diclofenac gel (Voltaren Emulgel 232) 3 x 5 cm daily n = 80Placebo gel 3 x 5 cm daily n = 82Rescue medication paracetamol (to maximum 2 g daily)
77Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Predel 2012 (Continued)
No ice or bandages after randomisation
Outcomes ge 50 reduction in PI on movement at 5 daysPGE 5-point scale (responder = ldquogoodrdquo or rsquovery goodldquo)Patient satisfaction 5-point scale (responder = rdquogoodldquo or rsquovery goodrdquo or ldquoexcellentrdquo)Rescue medicationAdverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical in composition appearance tex-ture and smellrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk LOCF but withdrawals lt 3 in all treat-ment arms
Size Unclear risk 50 to 200 participants per treatment arm
Predel 2013a
Methods R DB PC parallel groupsTreatment for 14 daysAssessed at baseline 3 7 10 and 14 days (plusmn 1 day)
Participants Uncomplicated one-sided ankle sprain with swelling ge 12 mm (2 to 18 hours)PI on movement tenderness joint mobility (scale 0 to 3) summed as ge 5 and le 7N = 232M 126 F 106Mean age 29 years
Interventions Diclofenac 4 spray gel 3 x 4 to 5 sprays daily (96 to 120 mg diclofenac sodium) n =118Placebo spray gel 3 x 4 to 5 sprays daily n = 114Rescue medication paracetamol 500 mg (maximum 10 tablets per week)No ice or bandaging allowed
78Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Predel 2013a (Continued)
Outcomes None or slight PI on movement at 3 and 7 daysPGE 4-point scale (responder = ldquovery goodrdquo at 14 days)gt 50 reduction in swelling at 10 daysAdverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk ldquoplacebo (vehicle only no active ingredi-ent)rdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk LOCF for full or early cure Withdrawalslt 2
Size Unclear risk 50 to 200 participants per treatment arm
Predel 2013b
Methods R DB PC parallel groups2 g gel applied with fingertips over affected area and massaged in for 1 minute Treatmentfor 5 daysAssessed at baseline 2 3 5 days
Participants Uncomplicated neck pain originating from cervical joints and accompanying soft tissues(ge 12 hours but lt 3 months) PI ge 50100N = 72M 33 F 39Mean age 34 years
Interventions Diclofenac gel (Voltaren Emulgel) 4 x 2 g daily n = 36Placebo gel 4 x 2 g daily n = 36Rescue medication paracetamol up to 2 g dailyNo concomitant therapies allowed
Outcomes ge 50 decrease in PI on movement after 48 hoursPGE 5-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo at 5 daysAdverse eventsWithdrawals
79Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Predel 2013b (Continued)
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk Sponsor ldquoproduced randomisation listrdquo
Allocation concealment (selection bias) Low risk Automated remote system
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical in packaging labelling scheduleof administration appearance and odourrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk No missing data
Size High risk lt 50 participants per treatment arm
Ramesh 1983
Methods R DB PC parallel groupsCream applied to painful area and rubbed into skin over a large area for up to 10 daysAssessment at baseline 3 7 10 days
Participants Strains sprains contusions compressionsN = 80M 42 F 38Age 11 to 81 yearsBaseline pain 5 ibuprofen 2 placebo participants had no or slight pain
Interventions Ibuprofen cream 5 3 to 4 x 5 to 10 cm daily n = 40Placebo cream 3 to 4 x 5 to 10 cm daily n = 40Adjuvant therapy was not administered
Outcomes Pain on movement 4-point scale (responder = ldquononerdquo or ldquoslightrdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
80Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Ramesh 1983 (Continued)
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Low risk Randomisation key in sealed envelopeavailable for emergencies but opened onlyafter completion
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical appearance and odourrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
Rowbotham 2003
Methods R DB PC parallel groupsNew patches applied to the affected painful area for 12 consecutive hours twice dailyfor up to 14 daysAssessed at baseline 14 days in clinic and daily patient diary
Participants Minor sports injuries (sprains sprains contusions lt 72 hours)N = 372M 253 F 119Mean age 33 yearsBaseline pain at rest ge 510
Interventions DHEP patch (Flector Tissuegel) 2 x daily (equivalent to diclofenac sodium 140 mg perpatch) n = 191Placebo patch 2 x daily n = 181
Outcomes PGE 5-point scale (responder = ldquogoodrdquo and ldquoexcellentrdquo)Pain resolved lt moderate for 2 daysSpontaneous pain 10 cm VAS (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
81Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Rowbotham 2003 (Continued)
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk ldquoSystegraveme identiquerdquo without diclofenac
Incomplete outcome data (attrition bias)All outcomes
Unclear risk No information about imputation All par-ticipants included in analysis
Size Unclear risk 50 to 200 participants per treatment arm
Russell 1991
Methods R DB PC parallel groupsAffected area washed with soap and water and dried then gel applied and carefullyrubbed into skin 4 times daily for at least 7 daysAssessed at baseline 4 8 15 (if necessary) days at clinic and daily patient diary
Participants Acute soft tissue injuries (recent not recurrent)N = 214 (200 analysed)M 95 F 105Mean age 40 yearsBaseline pain gt 65 mm
Interventions Piroxicam gel 05 4 x 5 mg daily n = 100Placebo gel n = 100No other NSAIDs or analgesic drugs including liniments containing salicylates allowedAncillary therapy at the discretion of the investigator
Outcomes PGE 4-point scale (responder = ldquogoodrdquo and ldquoexcellentrdquo)Spontaneous pain mean reductionAdverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk ldquocomputer generated randomization coderdquo
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical base formulationrdquo
82Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Russell 1991 (Continued)
Incomplete outcome data (attrition bias)All outcomes
High risk High withdrawal rate (8 with piroxicam50 with placebo) No information aboutimputation
Size Unclear risk 50 to 100 participants per treatment group
Saillant 1998
Methods R DB PC parallel groupsTreatment for 7 daysAssessed at baseline 1 2 3 and 7 days
Participants Ankle sprain (lt 48 hours)N = 140M 72 F 62Age 18 to 65 yearsBaseline spontaneous pain ge 50 mm
Interventions (1) DHEP plaster (Flector Tissugel 1) 1 x daily n = 70(2) Placebo plaster 1 x daily n = 70Rescue medication paracetamol (to maximum 3 g daily)Ice allowed
Outcomes ge 30 decrease in PI at 7 daysPGE 4-point scale (responder = ldquoexcellentrdquo)No or low pain on passive stretchSingle foot leaning OK without painWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk Identical in size appearance and used sameformula as active patch without active in-gredient
Incomplete outcome data (attrition bias)All outcomes
Unclear risk Used LOCF for primary outcome With-drawals lt 7 and equal between groupsAll participants included in analysis
83Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Saillant 1998 (Continued)
Size Unclear risk 50 to 200 participants per treatment arm
Sanguinetti 1989
Methods R DB PC parallel groupsGel applied 3 times daily for 7 consecutive daysAssessment at baseline 7 days
Participants Soft tissue trauma (lt 48 hours)N = 82M = 47 F = 35Mean age 34 yearsBaseline pain moderate to severe
Interventions Felbinac gel 3 3 x daily n = 42Placebo gel 3 x daily n = 40No other NSAID steroid other topical application allowedRescue medication paracetamol felbinac is an active metabolite of the NSAID fenbufen
Outcomes PGE scale not reported (responder = ldquogoodrdquo and ldquovery goodrdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoindistinguishable in appearance colour orodourrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
84Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Sinniger 1981
Methods R DB PC parallel groupsCream applied 2 to 3 times daily with gentle massage or if massage not possible (toopainful) with protective dressingAssessment at baseline 5 10 days
Participants Minor soft tissue injuriesN = 20M 11 F 9Mean age 40 yearsBaseline pain not reported
Interventions Fentiazac cream 5 2 to 3 x daily n = 10Placebo cream 2 to 3 x daily n = 10All participants told to restNo other local and systemic treatments allowedRescue medication analgesic if actually needed
Outcomes Pain relief scale not reported (responder = total pain relief ) improvement in pain on movement pain scale not reported (mean data)Adverse events
Notes Oxford Quality Score R1 DB1 W0 Total = 25
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
85Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Spacca 2005
Methods R DB PC parallel groupsGel applied 3 times daily with gentle massage until complete absorption for up to 10daysAssessment at baseline 10 days in clinic and daily patient diary
Participants Shoulder periarthritis or lateral epicondylitis (lt 5 days)N = 155M 74 F 81Mean age 51 yearsBaseline pain with activity gt 70 mm
Interventions DHEP lecithin gel (Effigel) 3 x 5 g daily n = 79Placebo gel 3 x 5 g daily n = 76Rescue medication (paracetamol) allowed if pain unbearableNo other analgesic or anti-inflammatory drug allowed
Outcomes Improvement in pain 100 mm VAS (mean data)Adverse events
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Unclear risk No usable efficacy data
Size Unclear risk 50 to 200 participants per treatment arm
Sugioka 1984
Methods R DB AC parallel groupsGel applied to affected area 3 to 4 times daily without occlusion for 14 daysAssessed at baseline 7 14 days
Participants Non-traumatic diseases of muscle or tendonN = 366 (340 analysed for efficacy)M 115 F 202 (completers)
86Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Sugioka 1984 (Continued)
Age range 12 to 84 years (most 30 to 70)Baseline pain on movement ldquononerdquo or ldquomildrdquo in about 13 of participantsExclusions for protocol violations 8 piroxicam 18 indomethacin
Interventions Piroxicam gel 05 3 to 4 x 1 g daily n = 183Indomethacin gel 1 3 to 4 x 1 g daily n = 183No concomitant anti-inflammatory or analgesic drug including steroids or initiationof physical therapy allowed
Outcomes PGE 5-point scale (responder = ldquobetterrdquo or ldquomuch betterrdquo)Pain on movement 4-point scale (responder = ldquoreducedrdquo or ldquodisappearedrdquo)
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Low risk Key code sealed and retained until end ofstudy
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoboth packages were of the same appear-ance and indistinguishablerdquo and investiga-tors did not see contents
Incomplete outcome data (attrition bias)All outcomes
Low risk All relevant participants included in analy-sis
Size Unclear risk 50 to 200 participants per treatment arm
Thorling 1990
Methods R DB PC parallel groupsParticipants given specific instructions on how to apply gel (not reported) to affectedarea 2 to 6 times daily as requiredAssessment at baseline 3 7 days in clinic
Participants Soft tissue injuries (lt 48 hours)N = 120M 85 F 35Mean age 27 yearsBaseline pain moderate to severe
87Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Thorling 1990 (Continued)
Interventions Naproxen gel 10 2 to 6 x daily n = 60Placebo gel 2 to 6 x daily n = 60Rescue medication paracetamol 500 mg
Outcomes PGE 5-point scale (responder = ldquogoodrdquo and ldquovery goodrdquo)Pain on passive movement 4-point scale (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquosupplied in unmarked tubesrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size Unclear risk 50 to 200 participants per treatment arm
Tonutti 1994
Methods R DB AC parallel groupsGel applied 3 times daily for 2 to 3 weeksAssessed at baseline and intervals of 7 days
Participants Muscle or joint traumaN = 30M 20 F 10Mean age 34 years1 participant had injury of mild severity Mean baseline pain on active movement 28(scale 0 to 4)
Interventions Ketoprofen gel 5 3 x 2 to 3 g daily n = 15Etofenamate gel 5 3 x 2 to 3 g n = 15No concomitant treatment with NSAID aspirin steroid or physical therapy
88Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Tonutti 1994 (Continued)
Outcomes PGE 4-point scale (responder = ldquogoodrdquo and ldquoexcellentrdquo)Pain on movement 5-point scale (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not reported
Allocation concealment (selection bias) Unclear risk Not reported
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquothe two drugs were packed in indistin-guishable tubesrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
Vecchiet 1991
Methods R DB PC parallel groupsGel applied to the skin on and around painful area and gently rubbed in until absorbedtwice daily for up to 10 daysAssessed at baseline 5 10 days
Participants Soft tissue trauma (minor sports injuries)N = 60M 60Mean age 25 yearsMean baseline pain on active movement moderate
Interventions Meclofenamic acid gel 5 2 x 10 cm daily (= 4 g) n = 30Placebo gel 2 x 10 cm daily n = 30Both groups treated with ice rest and bandage for first 48 hours before starting testtreatmentRescue medication paracetamol
Outcomes PGE 4-point scale (responder = ldquogoodrdquo and ldquoexcellentrdquo)Pain on movement 4-point scale (mean data)Withdrawals
89Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Vecchiet 1991 (Continued)
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
Whitefield 2002
Methods R DB (double dummy) AC parallel groupsGel applied to affected site with gentle massage and 1 tablet taken 3 times daily for atleast 7 daysAssessed at baseline 7 14 (if necessary) days in clinic and daily patient diary
Participants Soft tissue injuries (lt 24 hours)N = 100M 95 F 5Mean age 26 years (range 18 to 50)Mean baseline pain on movement 22 cm
Interventions Ibuprofen gel 5 + placebo tablet 3 x daily n = 50Ibuprofen 400 mg tablet + placebo gel 3 x daily n = 50No other medication or physical therapy was prescribed and no other analgesics wereallowed
Outcomes PGE 3-point scale (responder = ldquoexcellentrdquo)Change in condition of injury site 5-point scale (responder = ldquocompletely betterrdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
90Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Whitefield 2002 (Continued)
Random sequence generation (selectionbias)
Unclear risk Not reported
Allocation concealment (selection bias) Unclear risk Not reported
Blinding (performance bias and detectionbias)All outcomes
Low risk Placebo tablets Identical in appearance toactive tablets Active and placebo gels hadsimilar physical characteristics and weresupplied in identical tubes
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size Unclear risk 50 participants per treatment arm
kermark 1990
Methods R DB (double dummy) PC AC parallel groupsSpray applied to affected area and capsules taken 3 times daily for 2 weeksAssessment at baseline 3 or 4 7 and 14 days
Participants Superficial overuse sports injuries (symptom onset 74 weeks)N = 70M 44 F 18 (completers)Mean age 30 yearsBaseline pain on palpation mostly slight to moderate
Interventions Elmetacin spray (indomethacin 1) 3 to 5 x 05 to 15 ml daily + placebo capsules n= 23Indomethacin capsules 3 x 25 mg daily + placebo spray n = 23Placebo spray and capsules n = 24Rescue medication paracetamol
Outcomes No pain on palpation (= responder)Participant improvement 100 mm VAS (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk ldquorandom number coderdquo
91Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
kermark 1990 (Continued)
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical in appearancerdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk Responder analysis lt 5 withdrawals
Size High risk lt 50 participants per treatment group
AC active control BOCF baseline observation carried forward DB double-blind DHEP diclofenac hydroxyethylpyrrolidine ordiclofenac epolamine F female HCl hydrochloride hep heparin LOCF last observation carried forward M male N numberof participants in study n number of participants in treatment arm PC placebo control PGE Participant Global Evaluation PIpain intensity R randomised VAS visual analogue scale W withdrawals
Characteristics of excluded studies [ordered by study ID]
Study Reason for exclusion
Ambrus 1987 No usable dichotomous data
Anon 1993 Not double-blind
Ascherl 1982 No usable dichotomous data
Bagliani 1976 Not an RCT
Baracchi 1982 No usable data
Burnham 1998 lt10 participantstreatment arm in first period of crossover study
Boumlhmer 1995 Active control invalid
Cesarone 2008 Not an RCT
Coulibaly 2009 Not double-blind
Diebschlag 1985 No usable dichotomous data
Diebschlag 1986 Inappropriate randomisation
Diebschlag 1992 No usable dichotomous data
92Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Fantato 1971 No usable dichotomous data
Galer 2000 No usable data
Hallmeier 1986 Not double-blind
Hallmeier 1988 Not double-blind
Kaneko 1999 Inappropriate randomisation - quasi-randomised
Kockelbergh 1985b Treatment not applied daily
Kuwabara 2013 Used NSAID-lidocaine combination (conference abstract)
Lee 1991 Not an RCT
Link 1996 No usable dichotomous data
May 2007 No usable dichotomous data
Oakland 1993 Inappropriate comparator
Odaglia 1987 Not an RCT
Picardi 1993 Not an RCT
Taboada 1992 Dose and duration of treatment unclear
Vanderstraeten 1990 Not double-blind
Vinciguerra 2008 Not an RCT
Von Klug 1977 Chronic and acute outcomes combined
RCT randomised controlled trial
Characteristics of studies awaiting assessment [ordered by study ID]
NCT00351104
Methods R DB PC parallel groups
Participants Ankle sprain or strain Grade I or II Age ge 18 yearsN = 220
93Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT00351104 (Continued)
Interventions Ketoprofen patch 20 x 1 dailyPlacebo patch x 1 dailyTreatment for 21 days
Outcomes Assessments days 3 7 21PI (0-10) at rest and during activitiesPGEFunctionRescue medication
Notes Sponsor Endo PharmaceuticalsEstimated completion February 2007 No results
NCT00352625
Methods R DB PC parallel groups
Participants Shoulder elbow or knee tendonitis or bursitis Age ge 18 yearsN = 330
Interventions Ketoprofen patch 20 x 1 dailyPlacebo patch x 1 dailyTreatment for 21 days
Outcomes Assessments days 3 7 21PI (0-10) at rest and during activitiesPGEFunctionRescue medication
Notes Sponsor Endo PharmaceuticalsEstimated completion February 2007 No results
NCT00426985
Methods R DB PC parallel groups
Participants Shoulder elbow or knee tendonitis or bursitis Age ge 18 years
Interventions Ketoprofen patch 20 x 1 dailyPlacebo patchTreatment for 21 days
Outcomes PI during activityPI at restPGERescue medication
94Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT00426985 (Continued)
Safety
Notes Study terminated May 2008 but ldquosufficient number of subjects accrued to conduct analysisrdquo No results
NCT00640705
Methods R DB PC parallel groups
Participants Ankle sprain or strain Grade I or II le 48 hours PI 510 to 910 Age 18 to 75 yearsN = 170
Interventions Diclofenac sodium patch (15 mg) x 1 dailyPlaceboTreatment for 7 days
Outcomes EfficacySafety
Notes Study terminated July 2008 (Sponsor decision - no further details) No results posted
NCT00640939
Methods R DB PC parallel groups
Participants Shoulder elbow or knee tendonitis or bursitis Age 18 to 75 yearsBaseline pain mild to moderate but states 510 to 910N = 308
Interventions Diclofenac sodium patch 1 (15 mg) x 1 dailyPlacebo patch x 1 dailyDuration of treatment not reported probably 14 days
Outcomes EfficacySafety
Notes Sponsor Cerimon PharmaceuticalsEstimated completion April 2008 No results
NCT00680472
Methods R DB PC parallel groups
Participants Acute unilateral shoulder pain requiring treatment for ge 2 weeks Age ge 18 yearsN = 368
95Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT00680472 (Continued)
Interventions Ketoprofen patch (HKT-500) x 1 dailyPlacebo x 1 dailyTreatment for 14 days
Outcomes PainSafety
Notes Sponsor Hisamitsu Pharmaceutical Co IncEstimated completion October 2008 No results
NCT00680784
Methods R DB PC parallel groups
Participants Acute benign ankle sprain Grade I-II le 48 hours Age ge 18 yearsN = 260
Interventions Ketoprofen patch (HKT-500) x 1 dailyPlacebo patch x 1 dailyTreatment for 14 days
Outcomes PainSafety
Notes Sponsor Hisamitsu Pharmaceutical Co IncEstimated completion November 2008 No results
NCT00765700
Methods R DB PC parallel groups
Participants Acute sprain or strain of upper and lower extremities mild to moderate injuries le 72 hours Age 18 to 70 yearsN = 364
Interventions Ketoprofen cream 10 1 g x 3 dailyPlacebo cream x 3 dailyTreatment for 7 days
Outcomes PainSafety
Notes Sponsor Imprimis Pharmaceuticals IncEstimated completion September 2009 No results
96Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT00869063
Methods R DB PC parallel groups
Participants Wrist sprain strain or contusion (mild to moderate) ldquorecentrdquo Age 17 to 75 yearsN = 214
Interventions Diclofenac sodium patch 1 x 1 dailyPlacebo patch x 1 dailyTreatment for 7 days
Outcomes Change in PI during activity at 3 and 7 daysSafety
Notes Sponsor Cerimon PharmaceuticalsEstimated completion September 2009 No results
NCT00869180
Methods R DB PC parallel groups
Participants Unilateral ankle sprain (mild or moderate) ldquorecentrdquo Age 17 to 75 yearsN = 219
Interventions Diclofenac sodium patch x 1 dailyPlacebo patch x 1 dailyTreatment for 7 days
Outcomes Change in PI during activity at 3 and 7 days
Notes Sponsor Cerimon PharmaceuticalsEstimated completion August 2009 No results
NCT00931866
Methods R DB PC parallel groups
Participants Unilateral soft tissue injury between mid-bicep to wrist or mid-thigh to ankle (mild to moderate) ldquorecentrdquo Age 18to 75 yearsN = 407
Interventions Diclofenac sodium patch x 1 dailyPlacebo patch x 1 dailyTreatment for 14 days
Outcomes Change in PI during activity at 7 and 14 daysPIPGERescue medicationSafety
97Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT00931866 (Continued)
Notes Sponsor Cerimon PharmaceuticalsEstimated completion December 2009 No results
NCT01874626
Methods R DB PC parallel groups
Participants Acute ankle sprain Grade I or II le 24 hours Age ge 16 years Baseline PI ge 510N = 305
Interventions Ibuprofen cream 200 mg in 27 g cream x 4 dailyPlacebo cream x 4 dailyTreatment for 7 days
Outcomes PI on movement daily to 7 daysPGE 7 daysSystemic and local adverse events
Notes Estimated completion January 2014 No results
NCT01957215
Methods R DB PC parallel groups
Participants Acute lateral ankle sprain Grade I-II le 24 hours PI ge 510 Age 18 to 65 yearsN = 270
Interventions Indomethacin patch x 2 dailyPlacebo patch x 2 dailyDuration of treatment not reported probably 7 days
Outcomes PI on movement dailyPR dailyPGERescue medication
Notes Estimated completion September 2014 No results
NCT02324270
Methods R DB PC and AC parallel groups
Participants Uncomplicated acute ankle sprain Grade I or II lt 48 hours PI gt 50100 Age 18 to 65 yearsN = 658
98Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT02324270 (Continued)
Interventions Diclofenac epolamine (Flector) 13 patchGeneric diclofenac epolamine 13 patchPlacebo patchDuration of treatment not reported
Outcomes Bioequivalence studyChange from baseline in PI (VAS) at 3 daysApplication site reactions
Notes Estimated completion December 2014 No results
Sarzi-Puttini 2014
Methods R DB AC parallel group non-inferiority studyDuration 7 days
Participants Acute musculoskeletal injury (mainly muscular joint tendon)N = 697M 271 F 426Mean age 52 years
Interventions Ketoprofen (SKP-021) patch ketoprofen 30 mgDiclofenac (Voltadola) patch diclofenac sodium 140 mgPatches applied twice daily for 7 days
Outcomes ge 50 reduction in PI from baseline to end of study (100 mm VAS)Patient overall ratingClinical symptoms (4-point scale)Time to responseAdverse events skin reactionsSerious adverse events
Notes ldquoThe analysis of the data of this trial showed that the two formulations were equally effective and well tolerated inthe treatment of acute musculoskeletal injuriesrdquo
AC active-controlled DB double-blind F female M male N number of participants in study PC placebo-controlled PGE PatientGlobal Evaluation of treatment PI pain intensity R randomised VAS visual analogue scale
99Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Characteristics of ongoing studies [ordered by study ID]
NCT01945034
Trial name or title Placebo-controlled double-blind evaluation of the efficacy and safety of ibuprofen 5 topical gel for thetreatment of ankle sprain
Methods R DB PC parallel groups
Participants Ankle sprain Grade I or II age ge 12 yearsEstimated enrolment 280
Interventions Ibuprofen gel 5 x 2 dailyIbuprofen gel 5 x 3 dailyPlacebo gelTreatment for 7 days with additional 3 days as needed
Outcomes PI on weight bearing and rest at 7 daysPGE at 10 daysRescue medicationSafety
Starting date November 2013
Contact information Pfizer
Notes Estimated completion February 2015Includes participants aged ge 12 years
NCT02100670
Trial name or title A clinical study to assess the efficacy and onset of pain relief of topical MFC51123 diclofenac-menthol gelversus controls in ankle sprain
Methods R DB PC and AC parallel groups
Participants Acute lateral ankle sprain Grade I or II le 24 hours PI ge 510 Age 16 to 65 yearsEstimated enrolment 400
Interventions Diclofenac sodium 1 + methanol 3 gelDiclofenac sodium 1 + methanol 009 gelMethanol 3 gelPlacebo + 009 methanol gelTreatment for 10 days with 4 g gel x 4 daily
Outcomes PRPID on movement (0 to 10 days)PGETime to complete recoveryAdverse events
100Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT02100670 (Continued)
Starting date November 2013
Contact information GSKClinicalSupportHDgskcom
Notes Estimated completion November 2014
NCT02290821
Trial name or title A randomised double-blind placebo-controlled parallel group study to evaluate the efficacy and safety ofdiclofenac sodium topical gel (DSG) 1 applied four times daily in subjects with acute blunt soft tissueinjuriescontusions of the limbs
Methods R DB PC parallel groups
Participants Acute blunt soft tissue injuries or contusions of the limbs le 6 hours (ldquofreshrdquo) Age ge 16 yearsEstimated enrolment 200
Interventions Diclofenac sodium gel 1 x 4 dailyPlacebo gelDuration of treatment not reported
Outcomes PainSafety
Starting date December 2014
Contact information Novartis
Notes Estimated completion August 2015
AC active control DB double-blind PC placebo-controlled PGE Patient Global Evaluation of treatment PI pain intensity PIDpain intensity difference PR pain relief R randomised
101Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
D A T A A N D A N A L Y S E S
Comparison 1 Individual NSAID versus placebo
Outcome or subgroup titleNo ofstudies
No ofparticipants Statistical method Effect size
1 Clinical success 29 Risk Ratio (M-H Fixed 95 CI) Subtotals only11 Diclofenac 10 2050 Risk Ratio (M-H Fixed 95 CI) 160 [149 172]12 Ibuprofen 5 436 Risk Ratio (M-H Fixed 95 CI) 164 [133 201]13 Ketoprofen 7 683 Risk Ratio (M-H Fixed 95 CI) 156 [137 177]14 Piroxicam 3 504 Risk Ratio (M-H Fixed 95 CI) 148 [127 173]15 Indomethacin 3 341 Risk Ratio (M-H Fixed 95 CI) 126 [103 155]16 Benzydamine 3 193 Risk Ratio (M-H Fixed 95 CI) 115 [096 138]
2 Local adverse events 33 Risk Ratio (M-H Fixed 95 CI) Subtotals only21 Diclofenac 15 3271 Risk Ratio (M-H Fixed 95 CI) 078 [056 110]22 Ibuprofen 3 321 Risk Ratio (M-H Fixed 95 CI) 230 [098 543]23 Ketoprofen 8 852 Risk Ratio (M-H Fixed 95 CI) 119 [083 170]24 Piroxicam 3 522 Risk Ratio (M-H Fixed 95 CI) 042 [017 108]25 Felbinac 3 397 Risk Ratio (M-H Fixed 95 CI) 191 [049 750]26 Indomethacin 3 354 Risk Ratio (M-H Fixed 95 CI) 265 [091 773]
Comparison 2 Diclofenac versus placebo (effect of formulation)
Outcome or subgroup titleNo ofstudies
No ofparticipants Statistical method Effect size
1 Clinical success 10 2050 Risk Ratio (M-H Fixed 95 CI) 160 [149 172]11 Plaster - Flector 4 1030 Risk Ratio (M-H Fixed 95 CI) 153 [136 171]12 Plaster - other 3 474 Risk Ratio (M-H Fixed 95 CI) 155 [137 175]13 Gel - Emulgel 2 314 Risk Ratio (M-H Fixed 95 CI) 384 [268 550]14 Gel - other 1 232 Risk Ratio (M-H Fixed 95 CI) 115 [105 127]
Comparison 3 Ibuprofen versus placebo (effect of formulation)
Outcome or subgroup titleNo ofstudies
No ofparticipants Statistical method Effect size
1 Clinical success 5 436 Risk Ratio (M-H Fixed 95 CI) 164 [133 201]11 Cream 3 195 Risk Ratio (M-H Fixed 95 CI) 128 [103 159]12 Gel 2 241 Risk Ratio (M-H Fixed 95 CI) 266 [169 421]
102Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Comparison 4 Ketoprofen versus placebo
Outcome or subgroup titleNo ofstudies
No ofparticipants Statistical method Effect size
1 Clinical success 7 683 Risk Ratio (M-H Fixed 95 CI) 156 [137 177]11 Plaster 2 335 Risk Ratio (M-H Fixed 95 CI) 121 [104 140]12 Gel 5 348 Risk Ratio (M-H Fixed 95 CI) 219 [174 275]
Comparison 5 All topical NSAIDs versus placebo
Outcome or subgroup titleNo ofstudies
No ofparticipants Statistical method Effect size
1 Local adverse events 42 6740 Risk Ratio (M-H Fixed 95 CI) 098 [080 121]2 Systemic adverse events 36 5576 Risk Ratio (M-H Fixed 95 CI) 096 [073 126]3 Adverse event withdrawals 42 6405 Risk Ratio (M-H Fixed 95 CI) 101 [064 159]
Comparison 6 Topical NSAID versus active comparator
Outcome or subgroup titleNo ofstudies
No ofparticipants Statistical method Effect size
1 Clinical success - topicalpiroxicam vs topicalindomethacin
3 641 Risk Ratio (M-H Fixed 95 CI) 124 [107 144]
2 Local adverse events -topical piroxicam vs topicalindomethacin
3 671 Risk Ratio (M-H Fixed 95 CI) 021 [009 047]
103Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Analysis 11 Comparison 1 Individual NSAID versus placebo Outcome 1 Clinical success
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 1 Individual NSAID versus placebo
Outcome 1 Clinical success
Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
1 Diclofenac
Coudreuse 2010 99117 82116 175 120 [ 104 138 ]
Joussellin 2003 3668 2466 52 146 [ 099 215 ]
Klainguti 2010 5662 4659 100 116 [ 099 136 ]
Li 2013 173192 114192 242 152 [ 134 172 ]
Predel 2004 5560 560 11 1100 [ 474 2555 ]
Predel 2012 (1) 5980 941 25 336 [ 186 607 ]
Predel 2012 (2) 5780 841 22 365 [ 193 690 ]
Predel 2013a 111118 93114 201 115 [ 105 127 ]
Predel 2013b 3636 736 16 487 [ 257 923 ]
Rowbotham 2003 75191 48181 104 148 [ 110 200 ]
Saillant 1998 4370 2570 53 172 [ 119 248 ]
Subtotal (95 CI) 1074 976 1000 160 [ 149 172 ]Total events 800 (NSAID) 461 (Placebo)
Heterogeneity Chi2 = 12176 df = 10 (Plt000001) I2 =92
Test for overall effect Z = 1278 (P lt 000001)
2 Ibuprofen
Billigmann 1996 2580 1080 136 250 [ 129 486 ]
Campbell 1994 2126 1925 264 106 [ 080 142 ]
Dreiser 1988 2632 1232 164 217 [ 134 349 ]
Machen 2002 2540 941 121 285 [ 152 532 ]
Ramesh 1983 2340 2340 314 100 [ 069 146 ]
Subtotal (95 CI) 218 218 1000 164 [ 133 201 ]Total events 120 (NSAID) 73 (Placebo)
Heterogeneity Chi2 = 2102 df = 4 (P = 000031) I2 =81
Test for overall effect Z = 465 (P lt 000001)
3 Ketoprofen
Airaksinen 1993 2429 1427 91 160 [ 107 238 ]
Dreiser 1989 1830 530 32 360 [ 154 844 ]
02 05 1 2 5
Favours placebo Favours NSAID
(Continued )
104Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
( Continued)Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Julien 1989 1830 630 38 300 [ 138 650 ]
Kockelbergh 1985 3038 2236 142 129 [ 095 176 ]
Mazi res 2005a 7281 6082 376 121 [ 104 141 ]
Mazi res 2005b 5087 4185 262 119 [ 090 158 ]
Noret 1987 3951 947 59 399 [ 218 733 ]
Subtotal (95 CI) 346 337 1000 156 [ 137 177 ]Total events 251 (NSAID) 157 (Placebo)
Heterogeneity Chi2 = 3104 df = 6 (P = 000002) I2 =81
Test for overall effect Z = 683 (P lt 000001)
4 Piroxicam
Aoki 1984 5672 3367 286 158 [ 120 207 ]
Fujimaki 1985 4483 4082 337 109 [ 080 147 ]
Russell 1991 79100 45100 377 176 [ 138 223 ]
Subtotal (95 CI) 255 249 1000 148 [ 127 173 ]Total events 179 (NSAID) 118 (Placebo)
Heterogeneity Chi2 = 624 df = 2 (P = 004) I2 =68
Test for overall effect Z = 499 (P lt 000001)
5 Indomethacin
Aoki 1984 4164 3367 413 130 [ 096 176 ]
Fujimaki 1985 4482 4082 513 110 [ 082 148 ]
kermark 1990 1222 624 74 218 [ 099 481 ]
Subtotal (95 CI) 168 173 1000 126 [ 103 155 ]Total events 97 (NSAID) 79 (Placebo)
Heterogeneity Chi2 = 269 df = 2 (P = 026) I2 =26
Test for overall effect Z = 221 (P = 0027)
6 Benzydamine
Chatterjee 1977 2125 1225 185 175 [ 112 272 ]
Haig 1986 1821 1322 196 145 [ 098 214 ]
Linde 1985 3550 4050 618 088 [ 070 110 ]
Subtotal (95 CI) 96 97 1000 115 [ 096 138 ]Total events 74 (NSAID) 65 (Placebo)
Heterogeneity Chi2 = 1033 df = 2 (P = 001) I2 =81
Test for overall effect Z = 152 (P = 013)
02 05 1 2 5
Favours placebo Favours NSAID
(1) Three times daily application
(2) Twice daily application
105Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Analysis 12 Comparison 1 Individual NSAID versus placebo Outcome 2 Local adverse events
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 1 Individual NSAID versus placebo
Outcome 2 Local adverse events
Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
1 Diclofenac
Coudreuse 2010 1117 7116 102 014 [ 002 113 ]
Jenoure 1997 144 141 15 093 [ 006 1442 ]
Joussellin 2003 168 366 44 032 [ 003 303 ]
Klainguti 2010 0126 159 30 016 [ 001 381 ]
Kuehl 2011 16207 12211 173 136 [ 066 280 ]
Li 2013 4192 3192 44 133 [ 030 588 ]
NCT01255423 1104 3102 44 033 [ 003 309 ]
NCT01272934 1102 0103 07 303 [ 012 7350 ]
NCT01272947 0104 0100 Not estimable
Predel 2012 (1) 1160 182 19 051 [ 003 809 ]
Predel 2013a 1120 4116 59 024 [ 003 213 ]
Predel 2013b 036 036 Not estimable
Rowbotham 2003 27191 31181 463 083 [ 051 133 ]
Saillant 1998 070 070 Not estimable
Spacca 2005 079 076 Not estimable
Subtotal (95 CI) 1720 1551 1000 078 [ 056 110 ]Total events 54 (NSAID) 66 (Placebo)
Heterogeneity Chi2 = 944 df = 10 (P = 049) I2 =00
Test for overall effect Z = 142 (P = 016)
2 Ibuprofen
Billigmann 1996 1180 480 573 275 [ 091 827 ]
Machen 2002 440 241 283 205 [ 040 1057 ]
Ramesh 1983 140 140 143 100 [ 006 1544 ]
Subtotal (95 CI) 160 161 1000 230 [ 098 543 ]Total events 16 (NSAID) 7 (Placebo)
Heterogeneity Chi2 = 048 df = 2 (P = 079) I2 =00
Test for overall effect Z = 190 (P = 0057)
3 Ketoprofen
002 01 1 10 50
Favours NSAID Favours placebo
(Continued )
106Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
( Continued)
Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Airaksinen 1993 529 427 99 116 [ 035 389 ]
Dreiser 1989 030 230 60 020 [ 001 400 ]
Julien 1989 130 030 12 300 [ 013 7083 ]
Kockelbergh 1985 138 136 24 095 [ 006 1459 ]
Mazi res 2005a 1281 682 142 202 [ 080 513 ]
Mazi res 2005b 2987 2785 651 105 [ 068 161 ]
Noret 1987 151 047 12 277 [ 012 6636 ]
Parrini 1992 083 086 Not estimable
Subtotal (95 CI) 429 423 1000 119 [ 083 170 ]Total events 49 (NSAID) 40 (Placebo)
Heterogeneity Chi2 = 357 df = 6 (P = 073) I2 =00
Test for overall effect Z = 095 (P = 034)
4 Piroxicam
Aoki 1984 179 274 147 047 [ 004 506 ]
Fujimaki 1985 183 282 143 049 [ 005 534 ]
Russell 1991 4102 10102 710 040 [ 013 123 ]
Subtotal (95 CI) 264 258 1000 042 [ 017 108 ]Total events 6 (NSAID) 14 (Placebo)
Heterogeneity Chi2 = 003 df = 2 (P = 098) I2 =00
Test for overall effect Z = 180 (P = 0072)
5 Felbinac
McLatchie 1989 3118 2113 666 144 [ 024 844 ]
Morris 1991 041 043 Not estimable
Sanguinetti 1989 342 140 334 286 [ 031 2634 ]
Subtotal (95 CI) 201 196 1000 191 [ 049 750 ]Total events 6 (NSAID) 3 (Placebo)
Heterogeneity Chi2 = 023 df = 1 (P = 063) I2 =00
Test for overall effect Z = 093 (P = 035)
6 Indomethacin
Aoki 1984 270 274 440 106 [ 015 730 ]
Fujimaki 1985 582 282 452 250 [ 050 1252 ]
kermark 1990 422 024 108 978 [ 056 17191 ]
Subtotal (95 CI) 174 180 1000 265 [ 091 773 ]Total events 11 (NSAID) 4 (Placebo)
Heterogeneity Chi2 = 167 df = 2 (P = 043) I2 =00
Test for overall effect Z = 179 (P = 0073)
002 01 1 10 50
Favours NSAID Favours placebo
(1) DHEP + heparin and DHEP combined
107Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Analysis 21 Comparison 2 Diclofenac versus placebo (effect of formulation) Outcome 1 Clinical success
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 2 Diclofenac versus placebo (effect of formulation)
Outcome 1 Clinical success
Study or subgroup Experimental Control Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
1 Plaster - Flector
Joussellin 2003 3668 2466 52 146 [ 099 215 ]
Li 2013 173192 114192 242 152 [ 134 172 ]
Rowbotham 2003 75191 48181 104 148 [ 110 200 ]
Saillant 1998 4370 2570 53 172 [ 119 248 ]
Subtotal (95 CI) 521 509 451 153 [ 136 171 ]Total events 327 (Experimental) 211 (Control)
Heterogeneity Chi2 = 051 df = 3 (P = 092) I2 =00
Test for overall effect Z = 720 (P lt 000001)
2 Plaster - other
Coudreuse 2010 99117 82116 175 120 [ 104 138 ]
Klainguti 2010 5662 4659 100 116 [ 099 136 ]
Predel 2004 5560 560 11 1100 [ 474 2555 ]
Subtotal (95 CI) 239 235 285 155 [ 137 175 ]Total events 210 (Experimental) 133 (Control)
Heterogeneity Chi2 = 4658 df = 2 (Plt000001) I2 =96
Test for overall effect Z = 699 (P lt 000001)
3 Gel - Emulgel
Predel 2012 (1) 116160 1782 48 350 [ 227 540 ]
Predel 2013b 3636 736 16 487 [ 257 923 ]
Subtotal (95 CI) 196 118 64 384 [ 268 550 ]Total events 152 (Experimental) 24 (Control)
Heterogeneity Chi2 = 070 df = 1 (P = 040) I2 =00
Test for overall effect Z = 734 (P lt 000001)
4 Gel - other
Predel 2013a 111118 93114 201 115 [ 105 127 ]
Subtotal (95 CI) 118 114 201 115 [ 105 127 ]Total events 111 (Experimental) 93 (Control)
02 05 1 2 5
Favours placebo Favours diclofenac
(Continued )
108Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
( Continued)Study or subgroup Experimental Control Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Heterogeneity not applicable
Test for overall effect Z = 284 (P = 00045)
Total (95 CI) 1074 976 1000 160 [ 149 172 ]Total events 800 (Experimental) 461 (Control)
Heterogeneity Chi2 = 12176 df = 9 (Plt000001) I2 =93
Test for overall effect Z = 1278 (P lt 000001)
Test for subgroup differences Chi2 = 5003 df = 3 (P = 000) I2 =94
02 05 1 2 5
Favours placebo Favours diclofenac
(1) twice daily and three times daily applications combined
Analysis 31 Comparison 3 Ibuprofen versus placebo (effect of formulation) Outcome 1 Clinical success
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 3 Ibuprofen versus placebo (effect of formulation)
Outcome 1 Clinical success
Study or subgroup Ibuprofen Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
1 Cream
Campbell 1994 2126 1925 264 106 [ 080 142 ]
Dreiser 1988 2632 1232 164 217 [ 134 349 ]
Ramesh 1983 2340 2340 314 100 [ 069 146 ]
Subtotal (95 CI) 98 97 742 128 [ 103 159 ]Total events 70 (Ibuprofen) 54 (Placebo)
Heterogeneity Chi2 = 791 df = 2 (P = 002) I2 =75
Test for overall effect Z = 222 (P = 0026)
2 Gel
Billigmann 1996 2580 1080 136 250 [ 129 486 ]
Machen 2002 2540 941 121 285 [ 152 532 ]
Subtotal (95 CI) 120 121 258 266 [ 169 421 ]Total events 50 (Ibuprofen) 19 (Placebo)
001 01 1 10 100
Favours placebo Favours ibuprofen
(Continued )
109Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
( Continued)Study or subgroup Ibuprofen Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Heterogeneity Chi2 = 008 df = 1 (P = 078) I2 =00
Test for overall effect Z = 420 (P = 0000026)
Total (95 CI) 218 218 1000 164 [ 133 201 ]Total events 120 (Ibuprofen) 73 (Placebo)
Heterogeneity Chi2 = 2102 df = 4 (P = 000031) I2 =81
Test for overall effect Z = 465 (P lt 000001)
Test for subgroup differences Chi2 = 805 df = 1 (P = 000) I2 =88
001 01 1 10 100
Favours placebo Favours ibuprofen
Analysis 41 Comparison 4 Ketoprofen versus placebo Outcome 1 Clinical success
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 4 Ketoprofen versus placebo
Outcome 1 Clinical success
Study or subgroup Diclofenac Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
1 Plaster
Mazi res 2005a 7281 6082 376 121 [ 104 141 ]
Mazi res 2005b 5087 4185 262 119 [ 090 158 ]
Subtotal (95 CI) 168 167 638 121 [ 104 140 ]Total events 122 (Diclofenac) 101 (Placebo)
Heterogeneity Chi2 = 002 df = 1 (P = 090) I2 =00
Test for overall effect Z = 249 (P = 0013)
2 Gel
Airaksinen 1993 2429 1427 91 160 [ 107 238 ]
Dreiser 1989 1830 530 32 360 [ 154 844 ]
Julien 1989 1830 630 38 300 [ 138 650 ]
Kockelbergh 1985 3038 2236 142 129 [ 095 176 ]
Noret 1987 3951 947 59 399 [ 218 733 ]
001 01 1 10 100
Favours placebo Favours diclofenac
(Continued )
110Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
( Continued)Study or subgroup Diclofenac Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Subtotal (95 CI) 178 170 362 219 [ 174 275 ]Total events 129 (Diclofenac) 56 (Placebo)
Heterogeneity Chi2 = 1937 df = 4 (P = 000067) I2 =79
Test for overall effect Z = 674 (P lt 000001)
Total (95 CI) 346 337 1000 156 [ 137 177 ]Total events 251 (Diclofenac) 157 (Placebo)
Heterogeneity Chi2 = 3104 df = 6 (P = 000002) I2 =81
Test for overall effect Z = 683 (P lt 000001)
Test for subgroup differences Chi2 = 1859 df = 1 (P = 000) I2 =95
001 01 1 10 100
Favours placebo Favours diclofenac
Analysis 51 Comparison 5 All topical NSAIDs versus placebo Outcome 1 Local adverse events
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 5 All topical NSAIDs versus placebo
Outcome 1 Local adverse events
Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Airaksinen 1993 529 427 26 116 [ 035 389 ]
Aoki 1984 (1) 270 137 08 106 [ 010 1128 ]
Aoki 1984 (2) 179 137 09 047 [ 003 728 ]
Auclair 1989 5123 6116 39 079 [ 025 251 ]
Billigmann 1996 1180 480 26 275 [ 091 827 ]
Costantino 2011 (3) 3287 6142 51 025 [ 006 097 ]
Coudreuse 2010 1117 7116 45 014 [ 002 113 ]
Diebshlag 1990 (4) 113 06 04 150 [ 007 3229 ]
Diebshlag 1990 (5) 112 06 04 162 [ 008 3466 ]
Dreiser 1989 030 230 16 020 [ 001 400 ]
002 01 1 10 50
Favours NSAID Favours placebo
(Continued )
111Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
( Continued)Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Dreiser 1990 030 330 22 014 [ 001 265 ]
Dreiser 1994 265 066 03 508 [ 025 10373 ]
Fujimaki 1985 (6) 183 141 09 049 [ 003 770 ]
Fujimaki 1985 (7) 582 141 09 250 [ 030 2070 ]
Haig 1986 021 022 Not estimable
Jenoure 1997 144 141 07 093 [ 006 1442 ]
Joussellin 2003 168 366 19 032 [ 003 303 ]
Julien 1989 130 030 03 300 [ 013 7083 ]
Klainguti 2010 (8) 0126 159 13 016 [ 001 381 ]
Kockelbergh 1985 138 136 07 095 [ 006 1459 ]
Kuehl 2011 16207 12211 76 136 [ 066 280 ]
Li 2013 4192 3192 19 133 [ 030 588 ]
Linde 1985 050 050 Not estimable
Machen 2002 440 241 13 205 [ 040 1057 ]
Mazi res 2005a 1281 682 38 202 [ 080 513 ]
Mazi res 2005b 2987 2785 174 105 [ 068 161 ]
McLatchie 1989 3118 2113 13 144 [ 024 844 ]
Morris 1991 041 043 Not estimable
NCT01255423 1104 3102 19 033 [ 003 309 ]
NCT01272934 1102 0103 03 303 [ 012 7350 ]
NCT01272947 0104 0100 Not estimable
Noret 1987 151 047 03 277 [ 012 6636 ]
Parrini 1992 083 086 Not estimable
Predel 2012 (9) 1160 182 08 051 [ 003 809 ]
Predel 2013a 1120 4116 26 024 [ 003 213 ]
Predel 2013b 036 036 Not estimable
Ramesh 1983 140 140 06 100 [ 006 1544 ]
Rowbotham 2003 27191 31181 203 083 [ 051 133 ]
Russell 1991 4102 10102 64 040 [ 013 123 ]
Saillant 1998 070 070 Not estimable
Sanguinetti 1989 342 140 07 286 [ 031 2634 ]
002 01 1 10 50
Favours NSAID Favours placebo
(Continued )
112Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
( Continued)Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Sinniger 1981 010 010 Not estimable
Spacca 2005 079 076 Not estimable
Thorling 1990 160 060 03 300 [ 012 7220 ]
kermark 1990 422 024 03 978 [ 056 17191 ]
Total (95 CI) 3619 3121 1000 098 [ 080 121 ]Total events 155 (NSAID) 145 (Placebo)
Heterogeneity Chi2 = 3354 df = 35 (P = 054) I2 =00
Test for overall effect Z = 017 (P = 087)
Test for subgroup differences Not applicable
002 01 1 10 50
Favours NSAID Favours placebo
(1) indomethacin
(2) piroxicam
(3) DHEP + heparin and DHEP combined
(4) ketorolac
(5) etofenamate
(6) piroxicam
(7) indomethacin
(8) DHEP + heparin and DHEP combined
(9) Twice and three times daily application combined
113Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Analysis 52 Comparison 5 All topical NSAIDs versus placebo Outcome 2 Systemic adverse events
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 5 All topical NSAIDs versus placebo
Outcome 2 Systemic adverse events
Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Airaksinen 1993 129 027 05 280 [ 012 6593 ]
Aoki 1984 (1) 079 037 Not estimable
Aoki 1984 (2) 070 037 Not estimable
Campbell 1994 126 025 05 289 [ 012 6775 ]
Chatterjee 1977 025 025 Not estimable
Coudreuse 2010 1117 0116 05 297 [ 012 7227 ]
Diebshlag 1990 (3) 013 06 Not estimable
Diebshlag 1990 (4) 012 06 Not estimable
Dreiser 1988 032 032 Not estimable
Dreiser 1989 030 030 Not estimable
Dreiser 1990 030 030 Not estimable
Dreiser 1994 065 066 Not estimable
Fujimaki 1985 (5) 083 041 Not estimable
Fujimaki 1985 (6) 182 041 07 152 [ 006 3647 ]
Haig 1986 021 022 Not estimable
Hoffmann 2012 0115 0118 Not estimable
Joussellin 2003 168 066 05 291 [ 012 7025 ]
Klainguti 2010 (7) 2126 059 07 236 [ 012 4844 ]
Kuehl 2011 15207 23211 236 066 [ 036 124 ]
Li 2013 10192 4192 41 250 [ 080 783 ]
Linde 1985 050 050 Not estimable
Machen 2002 040 041 Not estimable
Mazi res 2005a 1187 785 73 154 [ 062 377 ]
Mazi res 2005b 1381 1482 144 094 [ 047 187 ]
McLatchie 1989 0118 0113 Not estimable
Morris 1991 041 043 Not estimable
001 01 1 10 100
Favours topical NSAID Favours placebo
(Continued )
114Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
( Continued)Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
NCT01272947 2104 2100 21 096 [ 014 670 ]
Parrini 1992 083 086 Not estimable
Predel 2004 060 060 Not estimable
Predel 2012 (8) 3160 382 41 051 [ 011 248 ]
Predel 2013a 6120 8116 84 073 [ 026 203 ]
Predel 2013b 036 136 16 033 [ 001 792 ]
Rowbotham 2003 21191 22181 234 090 [ 052 159 ]
Russell 1991 4102 7102 73 057 [ 017 189 ]
Saillant 1998 070 070 Not estimable
Sanguinetti 1989 042 040 Not estimable
Sinniger 1981 010 010 Not estimable
Spacca 2005 079 076 Not estimable
Thorling 1990 060 060 Not estimable
Total (95 CI) 2956 2620 1000 096 [ 073 126 ]Total events 92 (NSAID) 91 (Placebo)
Heterogeneity Chi2 = 946 df = 15 (P = 085) I2 =00
Test for overall effect Z = 030 (P = 076)
Test for subgroup differences Not applicable
001 01 1 10 100
Favours topical NSAID Favours placebo
(1) piroxicam
(2) indomethacin
(3) ketorolac
(4) etofenamate
(5) piroxicam
(6) indomethacin
(7) DHEP + heparin and DHEP combined
(8) Twice and three times daily application combined
115Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Analysis 53 Comparison 5 All topical NSAIDs versus placebo Outcome 3 Adverse event withdrawals
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 5 All topical NSAIDs versus placebo
Outcome 3 Adverse event withdrawals
Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Airaksinen 1993 029 027 Not estimable
Aoki 1984 072 034 Not estimable
Aoki 1984 064 033 Not estimable
Auclair 1989 1123 0116 14 283 [ 012 6879 ]
Billigmann 1996 280 080 14 500 [ 024 10253 ]
Campbell 1994 026 025 Not estimable
Chatterjee 1977 025 025 Not estimable
Costantino 2011 0188 0142 Not estimable
Coudreuse 2010 0117 0116 Not estimable
Diebshlag 1990 013 06 Not estimable
Diebshlag 1990 012 06 Not estimable
Dreiser 1988 032 032 Not estimable
Dreiser 1989 030 230 69 020 [ 001 400 ]
Dreiser 1990 030 130 42 033 [ 001 787 ]
Dreiser 1994 065 066 Not estimable
Fujimaki 1985 083 041 Not estimable
Fujimaki 1985 482 041 18 455 [ 025 8261 ]
Julien 1989 030 030 Not estimable
Kockelbergh 1985 038 036 Not estimable
Kuehl 2011 4207 9211 248 045 [ 014 145 ]
Li 2013 2192 0192 14 500 [ 024 10347 ]
Linde 1985 050 050 Not estimable
Machen 2002 040 041 Not estimable
Mazi res 2005a 987 685 169 147 [ 055 394 ]
Mazi res 2005b 381 082 14 709 [ 037 13503 ]
McLatchie 1989 0118 0113 Not estimable
001 01 1 10 100
Favours topical NSAID Favours placebo
(Continued )
116Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
( Continued)Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Morris 1991 041 043 Not estimable
NCT01255423 0104 0102 Not estimable
NCT01272934 0102 1103 41 034 [ 001 817 ]
NCT01272947 0104 0100 Not estimable
Noret 1987 148 045 14 282 [ 012 6740 ]
Parrini 1992 060 060 Not estimable
Predel 2004 160 060 14 300 [ 012 7220 ]
Predel 2012 2160 182 37 103 [ 009 1114 ]
Predel 2013a 1120 1116 28 097 [ 006 1527 ]
Predel 2013b 036 036 Not estimable
Ramesh 1983 140 140 28 100 [ 006 1544 ]
Rowbotham 2003 0191 0181 Not estimable
Russell 1991 1102 8102 222 013 [ 002 098 ]
Saillant 1998 070 070 Not estimable
Sanguinetti 1989 042 040 Not estimable
Sinniger 1981 010 010 Not estimable
Spacca 2005 079 076 Not estimable
Thorling 1990 060 060 Not estimable
kermark 1990 122 024 13 326 [ 014 7610 ]
Total (95 CI) 3365 3040 1000 101 [ 064 159 ]Total events 33 (NSAID) 30 (Placebo)
Heterogeneity Chi2 = 1502 df = 16 (P = 052) I2 =00
Test for overall effect Z = 004 (P = 097)
Test for subgroup differences Not applicable
001 01 1 10 100
Favours topical NSAID Favours placebo
117Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Analysis 61 Comparison 6 Topical NSAID versus active comparator Outcome 1 Clinical success - topical
piroxicam vs topical indomethacin
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 6 Topical NSAID versus active comparator
Outcome 1 Clinical success - topical piroxicam vs topical indomethacin
Study or subgroup Piroxicam Indomethacin Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Aoki 1984 5672 4164 301 121 [ 097 151 ]
Fujimaki 1985 4483 4482 307 099 [ 074 131 ]
Sugioka 1984 85175 55165 392 146 [ 112 190 ]
Total (95 CI) 330 311 1000 124 [ 107 144 ]Total events 185 (Piroxicam) 140 (Indomethacin)
Heterogeneity Chi2 = 390 df = 2 (P = 014) I2 =49
Test for overall effect Z = 278 (P = 00055)
Test for subgroup differences Not applicable
01 02 05 1 2 5 10
Favours indomethacin Favours piroxicam
118Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Analysis 62 Comparison 6 Topical NSAID versus active comparator Outcome 2 Local adverse events -
topical piroxicam vs topical indomethacin
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 6 Topical NSAID versus active comparator
Outcome 2 Local adverse events - topical piroxicam vs topical indomethacin
Study or subgroup Piroxicam Indomethacin Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Aoki 1984 179 270 64 044 [ 004 478 ]
Fujimaki 1985 183 582 152 020 [ 002 165 ]
Sugioka 1984 5178 26179 784 019 [ 008 049 ]
Total (95 CI) 340 331 1000 021 [ 009 047 ]Total events 7 (Piroxicam) 33 (Indomethacin)
Heterogeneity Chi2 = 041 df = 2 (P = 081) I2 =00
Test for overall effect Z = 382 (P = 000013)
Test for subgroup differences Not applicable
002 01 1 10 50
Favours piroxicam Favours indomethacin
A P P E N D I C E S
Appendix 1 CENTRAL search strategy (via CRSO) for 2015 update
1 MESH DESCRIPTOR Anti-Inflammatory Agents Non-Steroidal EXPLODE ALL TREES (13419)2 (bufexamac OR bufexine OR calmaderm OR ekzemase OR dicoflenac OR solaraze OR pennsaid OR voltarol OR emugel OR
voltarene OR voltarol OR optha OR voltaren OR etofenamate OR afrolate OR algesalona OR bayro OR deiron OR etofen ORflexium OR flogoprofen OR rheuma-gel OR rheumon OR traumalix OR traumon OR zenavan OR felbinac OR dolinac OR flexfreeOR napageln OR target OR traxam OR fentiazac OR domureuma OR fentiazaco OR norvedan OR riscalon OR fepradinol ORdalgen OR flexidol OR cocresol OR rangozona OR reuflodol OR pinazone OR zepelin OR flufenamic OR dignodolin OR rheumaOR lindofluid OR sastridex OR lunoxaprofen OR priaxim OR flubiprofen OR fenomel OR ocufen OR ocuflur OR ldquoTrans ActLATrdquo OR tulip OR ibuprofen OR cuprofen OR ldquodeep relief rdquo OR fenbid OR ibu-cream OR ibugel OR ibuleve OR ibumousse ORibuspray OR ldquonurofen gelrdquo OR proflex OR motrin OR advil OR radian OR ralgex OR ibutop OR indomethacin OR indocin ORindospray OR isonixin OR nixyn OR ketoprofen OR tiloket OR oruvail OR powergel OR solpaflex OR ketorolac OR acular ORtrometamol OR meclofenamic OR naproxen OR naprosyn OR niflumic OR actol OR flunir OR niflactol topico OR niflugel ORnifluril OR oxyphenbutazone OR californit OR diflamil OR otone OR tanderil OR piketoprofen OR calmatel OR triparsean ORpiroxicam OR feldene OR pranoprofen OR oftalar OR pranox OR suxibuzone OR danilon OR flamilon OR ufenamate OR fenazolOR flector OR benzydamine) TIABKY (25220)
3 1 OR 2 (32484)4 MESH DESCRIPTOR Administration Topical EXPLODE ALL TREES (2169)
119Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
5 (topical OR cutaneous OR dermal OR transcutaneous OR transdermal OR percutaneous OR skin OR massage ORembrocation OR gel OR ointment OR aerosol OR cream OR cregraveme OR lotion OR mouse OR foam OR liniment OR spray OR rubOR balm OR salve OR emulsion OR oil OR patch OR plaster)TIABKY (67940)
6 4 OR 5 (70486)7 MESH DESCRIPTOR Athletic Injuries EXPLODE ALL TREES (411)8 (strain OR sprain OR contusion OR distortion OR compression OR ldquosports injurrdquo OR ldquosoft tissue injurrdquo OR tendnitis OR
ldquomuscle painrdquo OR periarthritis OR epicondylitis OR tenosynovitis)TIABKY (9158)9 7 OR 8 (9448)
10 MESH DESCRIPTOR pain EXPLODE ALL TREES (29943)11 (pain OR analgesi)TIABKY (74815)12 10 OR 11 (80041)13 3 AND 6 AND 9 AND 12 (110)
Appendix 2 MEDLINE search strategy via Ovid (for 2015 update)
1 exp Anti-inflammatory Agents non-steroidal (162888)2 (bufexamac OR bufexine OR calmaderm OR ekzemase OR diclofenac OR solaraze OR pennsaid OR voltarol OR emugel OR
voltarene OR voltarol OR optha OR voltaren OR etofenamate OR afrolate OR algesalona OR bayro OR deiron OR etofen ORflexium OR flogoprofen OR rheuma-gel OR rheumon OR traumalix OR traumon OR zenavan OR felbinac OR dolinac OR flexfreeOR napageln OR target OR traxam OR fentiazac OR domureuma OR fentiazaco OR norvedan OR riscalon OR fepradinol ORdalgen OR flexidol OR cocresol OR rangozona OR reuflodol OR pinazone OR zepelin OR flufenamic OR dignodolin OR rheumaOR lindofluid OR sastridex OR lunoxaprofen OR priaxim OR flubiprofen OR fenomel OR ocufen OR ocuflur OR ldquoTrans ActLATrdquo OR tulip OR ibuprofen OR cuprofen OR ldquodeep relief rdquo OR fenbid OR ibu-cream OR ibugel OR ibuleve OR ibumousse ORibuspray OR ldquonurofen gelrdquo OR proflex OR motrin OR advil OR radian OR ralgex OR ibutop OR indomethacin OR indocin ORindospray OR isonixin OR nixyn OR ketoprofen OR tiloket OR oruvail OR powergel OR solpaflex OR ketorolac OR acular ORtrometamol OR meclofenamic OR naproxen OR naprosyn OR niflumic OR actol OR flunir OR niflactol topico OR niflugel ORnifluril OR oxyphenbutazone OR californit OR diflamil OR otone OR tanderil OR piketoprofen OR calmatel OR triparsean ORpiroxicam OR feldene OR pranoprofen OR oftalar OR pranox OR suxibuzone OR danilon OR flamilon OR ufenamate OR fenazolOR flector OR benzydamine)mp (558284)
3 1 OR 2 (664691)4 exp Administration Topical (69697)5 (topical OR cutaneous OR dermal OR transcutaneous OR transdermal OR percutaneous OR skin OR massage OR
embrocation OR gel OR ointment OR aerosol OR cream OR cregraveme OR lotion OR mouse OR foam OR liniment OR spray OR rubOR balm OR salve OR emulsion OR oil OR patch OR plaster)mp (1982395)
6 4 OR 5 (1997720)7 exp Athletic Injuries (21464)8 (strain OR sprain OR contusion OR distortion OR compression OR ldquosports injurrdquo OR ldquosoft tissue injurrdquo OR tendnitis OR
ldquomuscle painrdquo OR periarthritis OR epicondylitis OR tenosynovitis)mp (420133)9 7 OR 8 (439228)
10 Pain (113906)11 (pain OR analgesi)mp (585249)12 10 or 11 (585249)13 randomized controlled trialpt (401171)14 randomizedab (296222)15 placeboab (155341)16 drug therapyfs (1789858)17 randomlyab (207517)18 trialab (308477)19 groupsab (1318386)20 OR13-1921 3 AND 6 AND 9 AND 12 AND 20 (139)22 limit 21 to yr=ldquo2008-Currentrdquo (56)
120Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Appendix 3 EMBASE search strategy via Ovid (for 2015 update)
1 exp Anti-inflammatory Agents non-steroidal (452266)2 (bufexamac OR bufexine OR calmaderm OR ekzemase OR dicoflenac OR solaraze OR pennsaid OR voltarol OR emugel OR
voltarene OR voltarol OR optha OR voltaren OR etofenamate OR afrolate OR algesalona OR bayro OR deiron OR etofen ORflexium OR flogoprofen OR rheuma-gel OR rheumon OR traumalix OR traumon OR zenavan OR felbinac OR dolinac OR flexfreeOR napageln OR target OR traxam OR fentiazac OR domureuma OR fentiazaco OR norvedan OR riscalon OR fepradinol ORdalgen OR flexidol OR cocresol OR rangozona OR reuflodol OR pinazone OR zepelin OR flufenamic OR dignodolin OR rheumaOR lindofluid OR sastridex OR lunoxaprofen OR priaxim OR flubiprofen OR fenomel OR ocufen OR ocuflur OR ldquoTrans ActLATrdquo OR tulip OR ibuprofen OR cuprofen OR ldquodeep relief rdquo OR fenbid OR ibu-cream OR ibugel OR ibuleve OR ibumousse ORibuspray OR ldquonurofen gelrdquo OR proflex OR motrin OR advil OR radian OR ralgex OR ibutop OR indomethacin OR indocin ORindospray OR isonixin OR nixyn OR ketoprofen OR tiloket OR oruvail OR powergel OR solpaflex OR ketorolac OR acular ORtrometamol OR meclofenamic OR naproxen OR naprosyn OR niflumic OR actol OR flunir OR niflactol topico OR niflugel ORnifluril OR oxyphenbutazone OR californit OR diflamil OR otone OR tanderil OR piketoprofen OR calmatel OR triparsean ORpiroxicam OR feldene OR pranoprofen OR oftalar OR pranox OR suxibuzone OR danilon OR flamilon OR ufenamate OR fenazolOR flector OR benzydamine)mp (790435)
3 1 OR 2 (1108683)4 exp Administration Topical (68490)5 (topical OR cutaneous OR dermal OR transcutaneous OR transdermal OR percutaneous OR skin OR massage OR
embrocation OR gel OR ointment OR aerosol OR cream OR cregraveme OR lotion OR mouse OR foam OR liniment OR spray OR rubOR balm OR salve OR emulsion OR oil OR patch OR plaster)mp (3208844)
6 4 OR 5 (3208844)7 exp Athletic Injuries (24675)8 (strain OR sprain OR contusion OR distortion OR compression OR ldquosports injurrdquo OR ldquosoft tissue injurrdquo OR tendnitis OR
ldquomuscle painrdquo OR periarthritis OR epicondylitis OR tenosynovitis)mp (804322)9 7 OR 8 (825164)
10 Pain (216406)11 (pain OR analgesi)mp (971593)12 10 OR 11 (971593)13 clinical trialssh (841252)14 controlled clinical trialssh (389335)15 randomized controlled trialsh (357169)16 double-blind proceduresh (118945)17 (clin adj25 trial)ab (331002)18 ((doubl or trebl or tripl) adj25 (blind or mask))ab (141726)19 placeboab (203390)20 randomab (906176)21 OR13-20 (1731380)22 3 AND 6 AND 9 AND 10 AND 19 (439)23 limit 22 to yr=ldquo2008-Currentrdquo (204)
Appendix 4 Summary of outcomes efficacy
Study ID Treatment Clinical response Other response
Airaksinen 1993 (1) Ketoprofen gel 2 x 5 g (125mg) daily n = 29(2) Placebo gel n = 27
PGE rdquoimprovedldquo at 7 days(1) 2429(2) 1427
No additional data
121Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
kermark 1990 (1) Indomethacin spray 1 (El-metacin) 3-5 x 05-15 mL dailyn = 23(2) Indomethacin capsules 3 x 25mg daily n = 23(3) Placebo spray and capsules n= 24
No pain on palpation at 7 days(1) 1222(2) 522(3) 624
Patient assessment of improve-ment at 7 days (scale 0 to 100)(1) 57(2) 49(3) 30
Aoki 1984 (1) Piroxicam gel 5 3 or 4 x 1 gdaily n = 84(2) Indomethacin gel 1 3 or 4x 1 g daily n = 84(3) Placebo gel n = 84
PGE (5 point) rdquobetter or muchbetterldquo at 7 days(1) 5672(2) 4164(3) 3367
Pain on movement rdquoreducedldquo orrdquodisappearedldquo at 7 days(1) 4861(2) 3860(3) 3563
Auclair 1989 (1) Niflumic acid gel 25 3 x 5g daily n = 117(2) Placebo gel n = 110
PGE (5 point) rdquogood or verygoodldquo at 7 days(1) 69117(2) 54110
Pain on palpation rdquoimprovedldquo at7 days(1) 69117(2) 53110
Billigmann 1996 (1) Ibuprofen microgel 5 3 x200 mg daily n = 80(2) Placebo gel n = 80
Complete remission(1) 2580(2) 1080
Improvement in pain with move-ment of 20 at 7 days(1) 6580(2) 5580
Campbell 1994 (1) Ibuprofen cream 5 (Proflex)4 x 4rdquo daily n = 26(2) Placebo cream n = 25
Improvement in walking ability (4point) at 7 days(1) 2126(2) 1925
No additional data
Chatterjee 1977 (1) Benzydamine HCl cream 33 x daily n = 25(2) Placebo cream n = 25
Pain on movement ldquoabsentslightrdquoat 6 days(1) 2125(2) 1225
Tenderness with pressure ldquoabsentslightrdquo at 6 days(1) 2125(2) 1225
Costantino 2011 (1) DHEP-heparin plaster (Flec-torparin) x 1 daily n = 142(2) DHEP plaster (Flector) x 1daily n = 146(3) Placebo plaster n = 142
No responder analysis for efficacyOverall treatment efficacy not re-ported
Mean reduction from baseline forPI on movement at 3 days (fromgraph)(1) 24 mm(2) 19 mm(3) 14 mm
Coudreuse 2010 (1) DHEP-heparin plaster (Flec-torparin) n = 120(2) Placebo n = 120
PGE ldquogoodrdquo or ldquoexcellentrdquo at 7days(1) 99117(2) 82116These are results for physician-re-ported judgement - not partici-pant
Mean reduction from baseline forPI on movement over 6 hours(1) about 30(2) about 20Overall greater in DHEP-heparingroup than placebo over 7 days
122Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Curioni 1985 (1) Ibuproxam n = 20(2) Ketoprofen n = 20(3) Etofenamate n = 20
Resolution of symptoms by 7 days(1) 1520(2) 1320(3) 1320
PGE ldquogoodrdquo or ldquoexcellentrdquo at 10days(1) 1920(2) not reported(3) 1620
Diebshlag 1990 (1) Ketorolac gel 2 3 x 3 g dailyn = 13(2) Etofenamate gel 5 3 x 3 gdaily n = 12(3) Placebo gel n = 12
Improvement in pain at 7 days(1) 1213(2) 1012(3) 912
No additional data
Dreiser 1988 (1) Ibuprofen cream 5 3 x 4 cmdaily n = 32 (3 x 10 cm for largejoints)(2) Placebo cream n = 32
PGE ldquoimprovementrdquo or ldquocom-plete relief rdquo at 7 days(1) 2632(2) 1232
(1) significantly better than (2)for mean improvement in sponta-neous pain movement pain restpain tenderness to pressure (VAS)
Dreiser 1989 (1) Ketoprofen gel 25 2 x 5 cmdaily n = 30(2) Placebo gel n = 30
PGE (3 point) ldquobetterrdquo at 7 days(1) 1830(2) 530
(1) significantly better than (2) formean improvement in pain (restand movement) (VAS)
Dreiser 1990 (1) Niflumic acid gel 25 3 x 5g daily n = 30(2) Placebo gel n = 30
PGE (4 point) ldquocuredrdquo or ldquoim-provedrdquo at 7 days(1) 2330(2) 1030
(1) significantly better than (2) formean improvement in pain (VAS)
Dreiser 1994 (1) Flurbiprofen plaster 2 x 40 mgdaily n = 65(2) Placebo plaster n = 66
PGE (4 point) ldquogoodrdquo or ldquoverygoodrdquo at 7 days(1) 4865(2) 4166
(1) significantly better than (2)for mean improvement in sponta-neous pain but not pain on move-ment or palpation (VAS)
Fioravanti 1999 (1) DHEP lecithin gel 3 x 5 g (=65 mg) daily n = 50(2) DHEP gel 3 x 5 g (= 65 mg)daily n = 50
PGE (4 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 10 days(1) 3550(2) 3550
(1) significantly better than (2)for mean improvement in spon-taneous pain at 7 days but notfor pain on movement at 10 days(VAS)
Fujimaki 1985 (1) Piroxicam gel 05 3 or 4 x 1g daily n = 92(2) Indomethacin gel 1 3 or 4 x1 g daily n = 90(3) Placebo gel n = 89
PGE (5 point) ldquobetterrdquo or ldquomuchbetterrdquo at end of treatment at 14days(1) 4483(2) 4482(3) 4082
No additional data
Gallacchi 1990 (1) Diclofenac hy-droxyethylpyrrolidine gel 1 4 x2 g daily n = 25 (Flector gel)(2) Diclofenac sodium 1 4 x 2 g
PGE (5 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 14 days(1) 1925(2) 1925
No significant difference betweengroups for pain on applied pres-sure at 7 and 14 days
123Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
daily n = 25 (Voltaren Emugel)
Gonzaacutelez de Vega 2013 (1) Traumeel gel 3 x 2 g daily n =140(2) Traumeel ointment n = 143(3) Diclofenac gel 1 3 x 2 g dailyn = 137
Pain-free at 7 days(1) 7140(3) 8137PGE (5 point) ldquogoodrdquo or ldquoverygoodrdquo(1) 128140(3) 127137
Normal function (5 point) scoreof 0 or 1 at 14 days(1) 133140(3) 131137
Governali 1995 (1) Ketoprofen gel 5 3 x 2-3 gdaily n = 15(2) Ketoprofen cream 1 3 x 2-3 g daily n = 15
PGE (5 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 7 days(1) 1415(2) 415
No additional data
Gualdi 1987 (1) Flunaxaprofen gel 2 x 3-5 cmdaily n = 30(2) Ketoprofen gel 2 x 3-5 cmdaily n = 30
No dichotomous data No significant difference betweengroups for pain on movement at 7days
Haig 1986 (1) Benzydamine cream 3 6 xdaily n = 21(2) Placebo cream n = 22
Pain on movement ldquoimprovedrdquo by6 days(1) 1821(2) 1322
No additional data
Hoffmann 2012 (1) DHEP-heparin plaster (Flec-torparin) x 1 daily n = 121(2) DHEP (Flector) plaster x 1daily n = 115(3) Placebo plaster n = 118
PGE (5 point) ldquogoodrdquo or ldquoexcel-lentrdquo over 80 in all groups
Mean reduction from baseline inPI (100 mm VAS)Day 3(1) 191 mm(2) 114 mm(3) 52 mmDay 8(1) 443 mm(2) 372 mm(3) 296 mm
Hofman 2000 (1) Diclofenac sodium gel 1 4x 2 cm daily n = 69(2) Lysine clonixinate gel 5 4 x2 cm (225 mg) daily n = 73
PGE (3 point) at 8 days ldquogoodrdquo(1) 3869(2) 3673
No significant difference betweentreatments for any pain outcomes
Hosie 1993 (1) Felbinac foam 3 3 x 2 g daily+ placebo tablets 3 x 1 daily n =140 (127 analysed for efficacy)(2) Ibuprofen tablets 3 x 400 mgdaily + placebo foam 3 x 2 g dailyn = 147 (134 analysed for efficacy)
Pain on movement ldquononerdquo orldquomildrdquo at 7 days(1) 81127(2) 96133
Spontaneous pain ldquononerdquo orldquomildrdquo at 7 days(1) 99127(2) 108134
124Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Jenoure 1997 (1) DHEP plaster (Tissugel) 2 xdaily n = 44(2) Placebo plaster 2 x daily n =41
Baseline pain in two groups notbalanced and data in table and fig-ure do not agree so efficacy out-comes not used
No additional data
Joussellin 2003 (1) DHEP plaster (Flector Tis-sugel 1) 1 x daily n = 68(2) Placebo plaster 1 x daily n =66
PGE (4 point) ldquoexcellentrdquo at 7 days(1) 3668(2) 2466ge 30 reduction in PI at 7 days(1) 2568(2) 1166
(1) significantly better than (2) formean pain on movement at 6 days
Julien 1989 (1) Ketoprofen gel 25 2 x 5 cm(= 50 mg) daily n = 30(2) Placebo gel n = 30
PGE (4 point) ldquorecoveredrdquo at 7days(1) 1830(2) 630
PGE (4 point) ldquorecoveredrdquo or ldquoim-provedrdquo at 7 days(1) 2530(2) 1330
Klainguti 2010 (1) DHEP-heparin plaster (Flec-torparin) x 1 daily n = 62(2) DHEP plaster (Flector) x 1daily n = 61(3) Placebo n = 59
Overall treatment efficacy (partic-ipant and investigator) (5 point)ldquogoodrdquo or ldquoexcellentrdquo at 3 days(1) 5662(2) not reported(3) 4659
Mean reduction in PI at day 3(from graph)(1) 33 mm(2) Not reported(3) 24 mm
Kockelbergh 1985 (1) Ketoprofen gel 25 2 x 5 cm(= 15 mg) daily n = 38(2) Placebo gel n = 36
PGE (3 point) ldquogoodrdquo at 7 days(1) 3038(2) 2236
(1) and (2) slightly better than (3)for mean spontaneous pain at 7days
Kuehl 2011 (1) DETP 13 plaster 2 x dailyn = 207(2) Placebo plaster n = 211
No dichotomous data Percentage reduction from base-line at last application(1) 73(2) 62
Li 2013 (1) DHEP plaster (Flector) 2 xdaily n = 192(2) Placebo plaster 2 x daily n =192
ge 50 pain reduction at day 7(posthoc analysis)(1) 173192(2) 114192
Overall treatment efficacy (5point) ldquogoodrdquo or ldquoexcellentrdquo(1) 161192(2) 81192Participant and investigator ratingMean plusmn SD reduction in PI onmovement at 7 days(1) 538 plusmn 170(2) 370 plusmn 183
Linde 1985 (1) Benzydamine 3 cream 3 xdaily n = 50(2) Placebo gel n = 50
No pain on movement (walking)at 8 days(1) 3550(2) 4050
No additional data
125Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Machen 2002 (1) Ibuprofen gel 5 3 x daily n= 40(2) Placebo gel n = 41
PGE (5 point) ldquomarked improve-mentrdquo or ldquocomplete clearancerdquo at7 days(1) 2540(2) 941
Clinically meaningful (ge 30 mm)pain relief at day 7(1) 3040(2) 1641
Mahler 2003 (1) DHEP + lethicin gel 3 x 5 gdaily n = 52(2) DHEP gel 3 x 5 g daily n = 48
PGE (4 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 10 days(1) 4952(2) 3948
Mean reduction in pain on move-ment at 3 and 10 days significantlygreater with (1) than (2)
Maziegraveres 2005b (1) Ketoprofen plaster 100 mg x1 daily n = 81(2) Placebo plaster n = 82
PGE (4 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 14 days(1) 7281(2) 6082
All mean efficacy measures im-proved more for (1) than (2) mostwere statistically significant
Maziegraveres 2005a (1) Ketoprofen plaster 100 mg x1 daily n = 87(2) Placebo plaster n = 85
PGE (4 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 14 days(1) 5087(2) 4185
All mean efficacy measures im-proved more for (1) than (2) mostwere statistically significant
McLatchie 1989 (1) Felbinac gel 3 3 x 3 cm dailyn = 118(2) Placebo gel n = 113
No dichotomous data Patient daily self-assessment formean pain on rest movement atnight interference with normaland leisure activities show betterefficacy for (1) than (2) from day2 (VAS)
Morris 1991 (1) Felbinac gel 3 3 x 1 cm dailyn = 41(2) Placebo gel n = 43
PGE (5 point) ldquogoodrdquo or ldquoverygoodrdquo at 7 days(1) 2341(2) 2743
(1) better than (2) for meanimprovement in symptoms andsporting function at 7 days
NCT01255423 (1) Diclofenac sodium gel 1 x 4daily n = 104(2) Placebo gel x 4 daily n = 100
No dichotomous data VAS (mean plusmn SD) at 72 hours(baseline PI not reported)(1) 374 plusmn 252(2) 388 plusmn 241
NCT01272934 (1) Diclofenac sodium gel 1 x 4daily n = 104(2) Placebo gel x 4 daily n = 100
No dichotomous data VAS (mean plusmn SD) at 72 hours(baseline PI not reported)(1) 256 plusmn 159(2) 612 plusmn 166
NCT01272947 (1) Diclofenac sodium gel 1 x 4daily n = 104(2) Placebo gel x 4 daily n = 100
No dichotomous data VAS (mean plusmn SD) at 24 hours(baseline PI not reported)(1) 331 plusmn 214(2) 654 plusmn 169
126Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Noret 1987 (1) Ketoprofen gel 25 2 x 5 cm(75 mg) daily n = 48(2) Placebo gel n = 45
PGE (4 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 8 days(1) 3948(2) 945
Decrease in mean spontaneouspain significantly greater in (1)than (2) by 3 days
Parrini 1992 (1) Ketoprofen foam 15 3 x 2 g(200 mg) daily n = 83(2) Placebo foam n = 86
No dichotomous data Mean pain on movement andpressure significantly decreased by7 days in (1) compared with (2)
Picchio 1981 (1) Ibuprofen gel 10 3 x daily n = 20(2) Ketoprofen gel 1 3 x daily n= 20
No pain on movement at 8 days(1) 320(2) 020
Spontaneous pain ldquononerdquo at 8days(1) 620(2) 020
Predel 2004 (1) Diclofenac sodium plaster 2 xdaily (140 mgplaster) n = 60(2) Placebo plaster n = 60
PGE (4 point) ldquogoodrdquo ldquoexcellentrdquoat 7 days(1) 5560(2) 560
(1) better than (2) for reductionin tenderness pain and speed ofpain reduction
Predel 2012 (1) Diclofenac gel (VoltarenEmulgel 232) 2 x 5 cm daily n= 80(2)Diclofenac gel (Voltaren Emulgel232 ) 3 x 5 cm daily n = 80(3) Placebo gel n = 82
ge 50 red in PI on movement at5 days(1) 5780(2) 5980(3) 1782
PGE efficacy (5 point) ldquogoodrdquo orldquovery goodrdquo at 8 days(1) 6880(2) 7380(3) 2482
Predel 2013a (1) Diclofenac 4 spray gel 4 or5 sprays 3 x daily (96-120 mg di-clofenac sodium) n = 118(2) Placebo spray gel n = 114
None or slight PI on movement at7-8 days(1) 111118(2) 93114
None or slight PI on movement at3-4 days(1) 76118(2) 58114
Predel 2013b (1) Diclofenac gel (VoltarenEmulgel) 4 x 2 g daily n = 36(2) Placebo gel 4 x 2 g daily n =36
PGE (5 point) ldquogoodrdquo or ldquoexcel-lentrdquo(1) 3636(2) 736
ge 50 reduction in PI on move-ment after 48 hours(1) 3436(2) 336
Ramesh 1983 (1) Ibuprofen cream 5 3 or 4 x5-10 cm daily n = 40(2) Placebo cream n = 40
Pain on movement (4 point)ldquononerdquo or ldquoslightrdquo at 7 days(1) 2340(2) 2340
Physician global assessment at 10days ldquogoodrdquo(1) 2940(2) 1640
Rowbotham 2003 (1) Diclofenac epolamine plaster(Flector Tissuegel) 2 x daily n =191(2) Placebo plaster n = 181
Pain intensity le 210 for 2 daysor 4 consecutive evaluations by 7days(1) 75191(2) 48181
Mean pain on rest significantlybetter with (1) than (2) after 7 days
127Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Russell 1991 (1) Piroxicam gel 05 4 x 5 mgdaily n = 100(2) Placebo gel n = 100
PGE (4 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 8 days(1) 79100(2) 45100
Statistically greater reduction inmean pain on movement at 8 dayswith (1) than (2)
Saillant 1998 (1) DHEP plaster (Flector Tis-sugel 1) 1 x daily n = 70(2) Placebo plaster 1 x daily n =70
PGE (4 point) ldquoexcellentrdquo at 7 days(1) 4370(2) 2570
ge 30 decrease in PI at 7 days(1) 6470(2) 5070
Sanguinetti 1989 (1) Felbinac gel 3 3 x daily n =42(2) Placebo gel n = 40
PGE ldquogoodrdquo or ldquovery goodrdquo at 7days(1) 3442(2) 1140
(1) better than (2) by 2 days
Sinniger 1981 (1) Fentiazac cream 5 2 or 3 xdaily n = 10(2) Placebo cream n = 10
Complete pain relief within 10days(1) 710(2) 110
Improvement in active pain onmovement at 5 days(1) 67(2) 32
Spacca 2005 (1) DHEP lecithin gel (Effigel) 3x 5 g daily n = 79(2) Placebo gel n = 76
No dichotomous data Mean pain scores improved morerapidly in (1) than (2) - statisticallysignificant at 3 and 6 days
Sugioka 1984 (1) Piroxicam gel 05 3 or 4 x 1g daily n = 183(2) Indomethacin gel 1 3 or 4 x1 g daily n = 183
PGE (5 point) ldquobetterrdquo or ldquomuchbetterrdquo at 14 days(1) 85175(2) 55165
Pain on movement ldquoreducedrdquo orldquodisappearedrdquo at 7 days(1) 77175(2) 63165
Thorling 1990 (1) Naproxen gel 10 2-6 x dailyn = 60(2) Placebo gel n = 60
PGE (5 point) ldquogoodrdquo or ldquoverygoodrdquo at 7 days(1) 3860(2) 2760
Participants using naproxen im-proved more rapidly and had sig-nificantly lower severity scores byday 3
Tonutti 1994 (1) Ketoprofen gel 5 3 x 2-3 gdaily n = 15(2) Etofenamate gel 5 3 x 2-3g n = 15
PGE (4 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 7 days(1) 1015(2) 1115
Significant reductions in pain onmovement by 7 days in bothgroups
Vecchiet 1991 (1) Meclofenamic acid gel 5 2 x10 cm daily (2 g) n = 30(2) Placebo n = 30
PGE (4 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 10 days(1) 3030(2) 1930
(1) significantly better than (2)for mean improvement in sponta-neous pain movement pain func-tional restriction
Whitefield 2002 (1) Ibuprofen gel 5 + placebotablet 3 x daily n = 50(2) Ibuprofen 400 mg tablet +placebo gel 3 x daily n = 50
Participant satisfied at 7 days(1) 3050(2) 3650
ldquoCompletely betterrdquo at 14 days(1) 2450(2) 3050
128Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
DHEP diclofenac epolamine HCl hydrochloride n number PGE participant global evaluation PI pain intensity SD standarddeviation VAS visual analogue scale
Appendix 5 Summary of outcomes adverse events and withdrawals
Study ID Treatment Local AEs Systemic AEs Serious AEs Withdrawals
Airaksinen 1993 (1) Ketoprofen gel 2x 5 g (125 mg) dailyn = 29(2) Placebo gel n =27
(1) 529(2) 427
(1) 129 (nausea af-ter paracetamol)(2) 027
None AE noneOther nonereported
kermark 1990 (1)Indomethacin spray1 (Elmetacin) 3-5 x 05-15 mL dailyn = 23(2) Indomethacincapsules 3 x 25 mgdaily n = 23(3) Placebo sprayand capsules n = 24
(1) 422(2) 022(3) 024
No usable data -reported for eventsnot participants
None reported AE (1) 1 (2) 1 (3)0Lost to follow-up(1) 1 (2) 2 (3) 3
Aoki 1984 (1) Piroxicam gel5 3 or 4 x 1 gdaily n = 84(2) Indomethacingel 1 3 or 4 x 1 gdaily n = 84(3) Placebo gel n =84
(1) 179(2) 270(3) 274
None None reported AE none23 excluded for pro-tocol violations (1)7 (2) 7 (3) 926 withdrew for rea-sons unrelated totreatment (1) 5 (2)13 (3) 8
Auclair 1989 (1) Niflumic acidgel 25 3 x 5 gdaily n = 117(2) Placebo gel n =110
All AEs(1) 5123(2) 6116Most commonly cu-taneous eruptions
No usable data None reported AE (1) 1123 (2) 011626 excluded fromefficacy analysis fornot meeting entrycriteria and protocolviolations
Billigmann 1996 (1) Ibuprofen mi-crogel 5 3 x 200mg daily n = 80(2) Placebo gel n =
(1) 1180(2) 480
None reported None reported AE (1) 280 (aller-gic rash dermatitis)No reason given (1)380 (2) 580
129Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
80 Symptom-free (1)180 (2) 180
Campbell 1994 (1) Ibuprofen cream5 (Proflex) 4 x 4rdquodaily n = 26(2) Placebo cream n= 25
No data (1) 126 (headache)(2) 025
No data AE noneExclusions 493 presented late 2missing forms 1 ap-peared twice 43 didnot return diaries
Chatterjee 1977 (1) BenzydamineHCl cream 3 3 xdaily n = 25(2) Placebo cream n= 25
None None None AE none1 participant lostto follow-up (groupnot reported)
Costantino 2011 (1) DHEP-hep-arin plaster (Flector-parin) x 1 daily n =142(2) DHEP plaster(Flector) x 1 daily n= 146(3) Placebo plastern = 142
Possibly or probablydrug-related(1) 2142(2) 1145(3) 6142
All mild and goneby 14 days ex-cept 1 (moderate) inplacebo group
Not reportedNo systemic GIevents
None AE noneOther(1) 0142(2) 2146 (proto-col deviation lost tofollow-up)(3) 2142 (lost tofollow-up)
Coudreuse 2010 (1) DHEP-hep-arin plaster (Flector-parin) n = 120(2) Placebo n = 120
(1) 1117(2) 7116
All AEs mild ormoderate resolvedspontaneously
(1) 1117(2) 0116
None AE(1) 1117 (increasedoedema)(2) 0116Other(1) 6117 (recovery3 no reason 3)(2) 3116 (recovery1 no reason 2)
Curioni 1985 (1) Ibuproxam n =20(2) Ketoprofen n =20(3) Etofenamate n =20
None None None AE noneOther none
Diebshlag 1990 (1) Ketorolac gel2 3 x 3 g daily n =13(2) Etofenamate gel5 3 x 3 g daily n =12
(1) 113(2) 112(3) 012
None None AE none1 ketorolac partic-ipant did not at-tend 15 day follow-up due to car acci-
130Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
(3) Placebo gel n =12
dent
Dreiser 1988 (1) Ibuprofen cream5 3 x 4 cm dailyn = 32 (3 x 10 cmfor large joints)(2) Placebo cream n= 32
No usable data None Not reported AE none4 placebo partici-pants lost to follow-up
Dreiser 1989 (1) Ketoprofen gel25 2 x 5 cm dailyn = 30(2) Placebo gel n =30
(1) 030(2) 230
None None reported AE (2) 230 (intol-erance)LoE (1) 130 (2) 130
Dreiser 1990 (1) Niflumic acidgel 25 3 x 5 gdaily n = 30(2) Placebo gel n =30
(1) 030(2) 330
None None AE (2) 130 (ery-thema)Exclusion 1 from(2) from efficacyanalysis for inade-quate baseline pain
Dreiser 1994 (1)Flurbiprofen plaster2 x 40 mg daily n =65(2) Placebo plastern = 66
(1) 265(2) 066
None None AE 0(1) 165 excludedfrom efficacy analy-sis for protocol vio-lation(2) 266 (1 LoE 1cured)
Fioravanti 1999 (1) DHEP lecithingel 3 x 5 g (= 65 mg)daily n = 50(2) DHEP gel 3 x 5g (= 65 mg) daily n= 50
(1) 050(2) 150
No data None reported AE noneOther none
Fujimaki 1985 (1) Piroxicam gel 05 3 or 4 x 1 g dailyn = 92(2) Indomethacingel 1 3 or 4 x 1 gdaily n = 90(3) Placebo gel n =89
(1) 183(2) 582(3) 282
(1) 083(2) 182 (nauseaand vomiting)(3) 082
None AE (1) 0 (2) 4 (3)0Unknown reasons(1) 2 (2) 1Did not return af-ter 1st visitirregularvisits (1) 6 (2) 6(3) 7
Gallacchi 1990 (1) Diclofenac hy-droxyethylpyrro-lidine gel 1 4 x 2
No AEs None None AE noneOther none
131Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
g daily n = 25 (Flec-tor gel)(2) Di-clofenac sodium 14 x 2 g daily n = 25(Voltaren Emugel)
Gonzaacutelez de Vega2013
(1) Traumeel gel 3 x2 g daily n = 140(2) Traumeel oint-ment n = 143(3) Diclofenac gel1 3 x 2 g daily n =137
No usable data
Infrequent mild tomoderate
No usable data None AE noneOther(1) 11140(2) 12137
Governali 1995 (1) Ketoprofen gel5 3 x 2-3 g dailyn = 15(2) Ketoprofencream 1 3 x 2-3 gdaily n = 15
No side effects None None AE noneOther none
Gualdi 1987 (1) Flunaxaprofengel 2 x 3-5 cm dailyn = 30(2) Ketoprofen gel 2x 3-5 cm daily n =30
(1) 130(2) 330
No data None reported AE noneOther none
Haig 1986 (1) Benzydaminecream 3 6 x dailyn = 21(2) Placebo cream n= 22
No AEs reported None None reported AE none reportedOther no data
Hoffmann 2012 (1) DHEP-hep-arin plaster (Flector-parin) x 1 daily n =121(2) DHEP (Flector)plaster x 1 daily n =115(3) Placebo plastern = 118
Most AEs were mi-nor local reactions(eg pruritus anderythema) in areaof plaster of mildto moderate inten-sity and resolvedwithout interrupt-ing treatment
No treatment-related systemic AEsrecorded
(2) 1 participanthad 3 SAEs nonejudged related tostudy medication
AE noneExclusions(1) 5121(2) 10115(3) 7118Excluded fromper protocol analysisdue to poor compli-ance or personal de-cision
Hofman 2000 (1) Diclofenacsodium gel 1 4 x2 cm daily n = 69(2) Lysine clonixi-
(1) 158(2) 161
None None AE noneLoE (1) 9 (2) 8
132Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
nate gel 5 4 x 2cm (225 mg) dailyn = 73
Hosie 1993 (1) Felbinac foam3 3 x 2 g daily+ placebo tablets 3x 1 daily n = 140(127 analysed for ef-ficacy)(2) Ibupro-fen tablets 3 x 400mg daily + placebofoam 3 x 2 g daily n= 147 (134 analysedfor efficacy)
(1) 1127(2) 3134
GI events (1) 14127 (2) 11134For (1) more mildnone definitely drugrelated for(2) definitely relatedto study drug
None AE noneExclusions (1) 13(2) 13 did not returnfor 7 day follow-up
Jenoure 1997 (1) DHEP plaster(Tissugel) 2 x dailyn = 44(2) Placebo plaster 2x daily n = 41
(1) 144(2) 141
No data None reported AE none reportedOther nonereported
Joussellin 2003 (1) DHEPplaster (Flector Tis-sugel 1) 1 x dailyn = 68(2) Placebo plaster 1x daily n = 66
(1) 168 (pruritus)(2) 366 (pruritus 2burning 1)
All AEs mild ormoderate
(1) 168 (allergic re-action)(2) 066
None reported AE(1) 066(2) 166Other(1) 366(2) 266
Julien 1989 (1) Ketoprofen gel25 2 x 5 cm (= 50mg) daily n = 30(2) Placebo gel n =30
(1) 130(2) 030
Not reported None AE noneOther none
Klainguti 2010 (1) DHEP-hep-arin plaster (Flector-parin) x 1 daily n =62(2) DHEP plaster(Flector) x 1 daily n= 61(3) Placebo n = 59
(1) 062(2) 061(3) 159All AEs mild in na-ture and resolvedspontaneously
(1) 165 (facial in-fection)(2) 161 (abdominalpain)(3) 059
None AE noneOther(1) 365(2) 161(3) 359
Kockelbergh 1985a (1) Ketoprofen gel25 2 x 5 cm (= 15mg) daily n = 38(2) Placebo gel n =
(1) 138(2) 126
Not reported None AE noneOther none
133Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
36
Kuehl 2011 (1) DETP 13plaster 2 x daily n =207(2) Placebo plastern = 211
(1) 16207(2) 12211
(1) 15207(2) 23211
None AE(1) 4207(2) 9211LoE(1) 21207(2) 25211Other(1) 58207(2) 62211
Li 2013 (1) DHEP plaster(Flector) 2 x daily n= 192(2) Placebo plaster 2x daily n = 192
(1) 4192(2) 3192
(1) 10192(2) 4192
None AE(1) 2192(2) 0192Other(1) 5192(2) 1192
Linde 1985 (1) Benzydamine3 cream 3 x dailyn = 50(2) Placebo gel n =50
(1) 440(2) 241
None None AE none(1) 6 (2) 6 excludedfrom 1st assessment(1) 3 (2) 4 ex-cluded from final as-sessment
Machen 2002 (1) Ibuprofen gel5 3 x daily n = 40(2) Placebo gel n =41
(1) 440(2) 241
None None AE none(1) 1 LoE 1 proto-col violation(2) 4 LoE
Mahler 2003 (1) DHEP + lethicingel 3 x 5 g daily n =52(2) DHEP gel 3 x 5g daily n = 48
(1) 152(2) 048
(1) 152(2) 048
None AE none5 lost to follow-up
Maziegraveres 2005b (1) Ketoprofen plas-ter 100 mg x 1 dailyn = 81(2) Placebo plastern = 82
At 21 days(1) 1281(2) 682
(1) 1381(2) 1482
None AE (1) 381(1) 781 (1 LoE 6cured)(2) 782 (5 LoE 2cured)
Maziegraveres 2005a (1) Ketoprofen plas-ter 100 mg x 1 dailyn = 87(2) Placebo plastern = 85
At 21 days(1) 2987(2) 2785
(1) 1187(2) 785
None AE (1) 987 (2) 685(1) 687 (2 LoE 4cured)(2) 585 (4 LoE 1cured)
134Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
McLatchie 1989 (1) Felbinac gel 33 x 3 cm daily n =118(2) Placebo gel n =113
(1) 3118(2) 2113Mild transient localirritation
None reported None AE noneOther none
Morris 1991 (1) Felbinac gel 33 x 1 cm daily n =41(2) Placebo gel n =43
None None None AE none(1) 4 (protocol vio-lations)(2) 1 (lost to follow-up)Exclusions 11 fromefficacy analysis be-cause evaluated by4 different investi-gators
NCT01255423 (1) Diclofenacsodium gel 1 x 4daily n = 104(2) Placebo gel x 4daily n = 100
(1) 1104(2) 3102
Total AE(1) 11104(2) 8102
None None
NCT01272934 (1) Diclofenacsodium gel 1 x 4daily n = 104(2) Placebo gel x 4daily n = 100
(1) 1102(2) 0103
Total AE (excludingSAE)(1) 6102(2) 3103
(1) 0102(2) 1103 (rupturedligaments in wrist)
AE see SAE
NCT01272947 (1) Diclofenacsodium gel 1 x 4daily n = 104(2) Placebo gel x 4daily n = 100
None (1) 2104(2) 2100
None None
Noret 1987 (1) Ketoprofen gel25 2 x 5 cm (75mg) daily n = 48(2) Placebo gel n =45
(1) 151(2) 047
None reported Not reported AE (1) 151 (skinallergy)(1) 1 LoE 1 unre-lated to trial(2) 1 LoE 1 unre-lated to trial
Parrini 1992 (1) Ketopro-fen foam 15 3 x 2g (200 mg) daily n= 83(2) Placebo foam n= 86
None None None AE noneOther none
135Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Picchio 1981 (1) Ibuprofen gel10 3 x daily n =20(2) Ketoprofen gel1 3 x daily n = 20
None None None AE noneOther not reported
Predel 2004 (1) Diclofenacsodium plaster 2 xdaily (140 mgplas-ter) n = 60(2) Placebo plastern = 60
12 participants ex-pe-rienced 16 mild AEswith no differencesbetween groups
None None AE (1) 160Other none
Predel 2012 (1) Diclofenac gel(Voltaren Emulgel232) 2 x 5 cmdaily n = 80(2) Diclofenac gel(Voltaren Emulgel232) 3 x 5 cmdaily n = 80(3) Placebo gel n =82
(1) 080(2) 180(3) 182All mild to moder-ate
(1) 280(2) 180(3) 382
None AE(1) 080(2) 080(3) 182Other(1) 180(2) 280(3) 282(protocol violationslost to follow-up)
Predel 2013a (1) Diclofenac 4spray gel 4 or 5sprays 3 x daily (96-120 mg diclofenacsodium) n = 118(2) Placebo spraygel n = 114
(1) 1118(2) 4114
(1) 5118(2) 4114All AEs mild re-versible
None AE(1) 1118(2) 1114Other(1) 3118(2) 43114
Predel 2013b (1) Diclofenac gel(Voltaren Emulgel)4 x 2 g daily n = 36(2) Placebo gel 4 x 2g daily n = 36
None (1) 036(2) 136
None AE noneOther none
Ramesh 1983 (1) Ibuprofen cream5 3 or 4 x 5-10 cmdaily n = 40(2) Placebo cream n= 40
(1) 140(2) 140
None reported Not reported AE (1) 140 (2) 140Other none
Rowbotham 2003 (1) DHEP plaster(Flector Tissuegel) 2x daily n = 191(2) Placebo plastern = 181
(1) 27191 (pruritis14)(2) 31181 (pruritis21)
(1) 21191(2) 22181
None re-ported (ldquovast major-ity mildrdquo)
AE none(1) 3191 (2) 4181(did not finish trialand complete dailydiaries)
136Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Russell 1991 (1) Piroxicam gel 05 4 x 5 mg daily n= 100(2) Placebo gel n =100
(1) 4102(2) 10102
GI or CNS events(1) 4 (2) 7Any AE (1) 7102(2) 15102
None reported AE (1) 1102 (2) 8102Other (1) 6 LoE 1ldquootherrdquo(2) 42 LoEExclusions7 did not complywith study medica-tion schedule 6 lostto follow-up 1 pro-tocol violation
Saillant 1998 (1) DHEPplaster (Flector Tis-sugel 1) 1 x dailyn = 70(2) Placebo plaster 1x daily n = 70
None None None AE noneOther(1) 570(2) 570
Sanguinetti 1989 (1) Felbinac gel 33 x daily n = 42(2) Placebo gel n =40
(1) 342(2) 140
None None reported AE noneOther nonereported
Sinniger 1981 (1) Fentiazac cream5 2 or 3 x daily n= 10(2) Placebo cream n= 10
ldquoNo untoward sideeffectsrdquo
None None AE noneOther nonereported
Spacca 2005 (1) DHEP lecithingel (Effigel) 3 x 5 gdaily n = 79(2) Placebo gel n =76
ldquoNo signsof cutaneous irrita-tion or sensitisationobservedrdquo
No AEs observed None AE noneOther nonereported
Sugioka 1984 (1) Piroxicam gel 05 3 or 4 x 1 g dailyn = 183(2) Indomethacingel 1 3 or 4 x 1 gdaily n = 183
(1) 5178(2) 26179
None reported None reported AE none reportedExclusions due toprotocol violations(1) 8 (2) 18Withdrawals(1) 11 (2) 12
Thorling 1990 (1) Naproxen gel10 2-6 x daily n =60(2) Placebo gel n =60
(1) 160(2) 060
None None AE none(1) 1 LoE 1 proto-col violation(2) 1 participant re-quest
137Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Tonutti 1994 (1) Ketoprofen gel5 3 x 2-3 g dailyn = 15(2) Etofenamate gel5 3 x 2-3 g n =15
None No AEs attributableto the medication
None AE NoneLoE (1) 1 (2) 2
Vecchiet 1991 (1) Meclofe-namic acid gel 5 2x 10 cm daily (2 g)n = 30(2) Placebo n = 30
Tolerabil-ity excellent or goodin nearly all partici-pants
No data None AE none reported(2) 5 lost to follow-up
Whitefield 2002 (1) Ibuprofen gel5 + placebo tablet3 x daily n = 50(2) Ibuprofen 400mg tablet + placebogel 3 x daily n = 50
No data 6 AEs reportednone judged relatedto study medication
None reported AE noneRecovered (1) 3 (2)2LoE (2) 1Lost to follow-up(1) 1 (2) 1
AE adverse event CNS central nervous system DHEP diclofenac epolamine GI gastrointestinal HCl hydrochloride LoE lackof efficacy n number SAE serious adverse event
Appendix 6 Concentration amount and frequency of dosing
Study Drug Concentration Quantity Frequency Estimated daily dose of topi-cal NSAID
Joussellin 2003 Diclofenac 1 Plaster 1 180 mg epolamine salt 140 mgNa salt
Li 2013 Diclofenac 1 Plaster 2 360 mg epolamine salt 280 mgNa salt
Rowbotham 2003 Diclofenac 1 Plaster 2 360 mg epolamine salt 280 mgNa salt
Saillant 1998 Diclofenac 1 Plaster 1 180 mg epolamine salt 140 mgNa salt
Coudreuse 2010 Diclofenac 1 Plaster 1 180 mg epolamine salt 140 mgNa salt
Klainguti 2010 Diclofenac 1 Plaster 1 180 mg epolamine salt 140 mgNa salt
138Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Predel 2004 Diclofenac - Plaster 2 280 mg Na salt
Predel 2012 Diclofenac 232 5 cm ribbon~ 2 g 2 or 3 92-138 mg (Na equiv) as di-ethylamine salt
Predel 2013b Diclofenac 116 2 g 4 92 mg (Na equiv) as diethy-lamine salt
Predel 2013a Diclofenac 4 4-5 sprays 3 96-120 mg Na salt
Costantino 2011 Diclofenac 1 Plaster 1 180 mg epolamine salt 140 mgNa salt
Fioravanti 1999 Diclofenac - 5 g 3 195 mg epolamine salt
Gallacchi 1990 Diclofenac 1 2 g 4 80 mg
Gonzaacutelez de Vega2013
Diclofenac 1 6 cm 2g 3 60 mg
Hoffmann 2012 Diclofenac 1 Plaster 1 180 mg epolamine salt 140 mgNa salt
Hofman 2000 Diclofenac 1 2 cm 4 probably 30-40 mg Na salt
Jenoure 1997 Diclofenac 1 Plaster 2 360 mg epolamine salt 280 mgNa salt
Mahler 2003 Diclofenac - 5 g 3 195 mg epolamine salt
NCT01255423 Diclofenac 1 - 4 -
NCT01272934 Diclofenac 1 - 4 -
NCT01272947 Diclofenac 1 - 4 -
Spacca 2005 Diclofenac - 5 g 3 195 mg epolamine salt
Campbell 1994 Ibuprofen 5 4 inch ribbon 4 Assume 800 mg
Dreiser 1988 Ibuprofen 5 4 cm ribbon(10 cm for largerjoints)
4 Assume up-800 mg
Ramesh 1983 Ibuprofen 5 5-10 cm ribbon 3-4 Assume 300-800 mg
Billigmann 1996 Ibuprofen 5 10 cm 4 g gel 3 600 mg
139Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Machen 2002 Ibuprofen 5 - 3 -
Picchio 1981 Ibuprofen 10 - 3 -
Whitefield 2002 Ibuprofen 5 - 3 -
Maziegraveres 2005b Ketoprofen - Plaster 1 100 mg
Maziegraveres 2005a Ketoprofen - Plaster 1 100 mg
Airaksinen 1993 Ketoprofen - 5 g 2 125 mg
Dreiser 1989 Ketoprofen 25 5 cm 2 100 mg
Julien 1989 Ketoprofen 25 5 cm 2 100 mg
Noret 1987 Ketoprofen 25 5 cm 75 g 2 375 mg
Curioni 1985 Ketoprofen No details 2 -
Governali 1995 Ketoprofen 5 2-3 g 3 300-450 mg
Governali 1995 Ketoprofen 1 2-3 g 3 60-90 mg
Gualdi 1987 Ketoprofen - 3-5 cm 2 -
Kockelbergh 1985 Ketoprofen 25 5 cm 2 100 mg
Parrini 1992 Ketoprofen 150 2 g 3 600 mg
Picchio 1981 Ketoprofen 1 - 3 -
Tonutti 1994 Ketoprofen 5 2-3 g 3 300-450 mg
kermark 1990 Indomethacin 1 05-15 mL 3-5 12-60 mg
Aoki 1984 Indomethacin 1 1 g 3-4 30-40 mg
Fujimaki 1985 Indomethacin 1 1 g 3-4 30-40 mg
Sugioka 1984 Indomethacin 1 1 g 3-4 30-40 mg
Aoki 1984 Piroxicam 05 1 g 3-4 15-20 mg
Fujimaki 1985 Piroxicam 05 1 g 3-4 15-20 mg
140Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Russell 1991 Piroxicam 05 5 mg 4 20 mg
Sugioka 1984 Piroxicam 05 1 g 3-4 15-20 mg
Chatterjee 1977 Benzydamine 3 - 3 -
Haig 1986 Benzydamine 3 - 6 -
Linde 1985 Benzydamine 3 - 3 -
Auclair 1989 Niflumic acid 25 10 cm 5 g 3 375 mg
Dreiser 1990 Niflumic acid 25 10 cm 5 g 3 375 mg
Curioni 1985 Ibuproxam gel 10 - 2 -
Curioni 1985 Etofenamate No details - 2 -
Diebshlag 1990 Etofenamate 5 3 g 3 450 mg
Tonutti 1994 Etofenamate 5 2-3 g 3 300-450 mg
Diebshlag 1990 Ketorolac 2 3 g 3 360 mg
Dreiser 1994 Flurbiprofen Patch 2 80 mg
Gualdi 1987 Flunoxaprofen - 3-5 cm 2 -
Hofman 2000 Lysine clonixinate 5 2 cm 4 90 mg
Hosie 1993 Felbinac 3 2 g 3 180 mg
McLatchie 1989 Felbinac 3 3 cm 3 -
Morris 1991 Felbinac 3 1 cm 3 -
Sanguinetti 1989 Felbinac 3 - 3 -
Sinniger 1981 Fentiazac 5 Varied according toinvolved areas
2-3 -
Thorling 1990 Naproxen 10 - 2-6 -
Vecchiet 1991 Meclofenamic acid 5 10 cm 4 g 2 400 mg
equiv equivalent Na sodium
141Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
F E E D B A C K
Query on formulations of topical NSAIDs particularly DMSO from Dr Chrubasik 11 April 2012
Summary
Dr Chrubasik highlighted this letter to the Editor postgradmedorgdoi103810pgm2011092482
DMSO [dimethyl sulphoxide] but also other additives eg nonivamide (which is a capsaicinoid added as drug enhancer) may contributeto the overall effect of topical NSAIDs Nonivamide certainly contributes to the analgesic effect and to adverse events (heat sensationburning pruritus etc) This has not been considered in the Cochrane review by Massey T Derry S Moore RA McQuay HJ TopicalNSAIDs for acute pain in adults but Dr Chrubasik believes should be done otherwise the effect size of the NSAID topicals is favoured
Reply
We have been asked by Dr Chrubasik to comment on a letter (Roth 2011) about the formulations of topical NSAIDs particularly howDMSO and other penetration enhancers can affect efficacy estimates or adverse event reporting in osteoarthritis It was suggested thatthe review of Topical NSAIDs for acute pain in adults did not consider this resulting in a bias towards the topical NSAIDThere are a number of points to be made here
1 Penetration enhancers are used in formulations of topical products to encourage local absorption through the skin and produce ahigh local concentration Topical NSAIDs use penetration enhancers and the result is high local concentration in joints for instancebut low systemic concentrations (Moore 2008) That is how they work Formulation is an important part of medicinal chemistry as awhole not just for topical agents
2 In our analysis of topical NSAIDs we were aware that a range of properties are or have been ascribed to the analgesia resultingfrom application of topical agents and which could contribute to overestimation of treatment effect of topical NSAID These includefeelings of heat or cold and even the act of rubbing itself For that reason we have chosen to include only double-blind studies wherethe placebo agent is identical to the active with the exception of course of the NSAID So heat cold rubbing and penetrationenhancers should be identical as best we can judge That leaves only the NSAID itself to provide any additional analgesic effect andit is that which we measure This is analogous for example to use of acupuncture say where the better studies show no differencebetween ldquotruerdquo acupuncture and ldquoshamrdquo acupuncture performed at nonspecific sites but better than non-treatment controls Theargument that we should only use high quality studies to evaluate evidence about pain interventions is well made
3 Overestimation of analgesic effect because of effects of enhancers themselves would be better made in direct comparisons oftopical and oral NSAIDs where local or even systemic effects would not be balanced in the oral study arm However our reviewconcentrated on placebo-controlled studies and had few studies with active controls Moreover the real test would be in chronicrather than acute conditions with long duration (12 week) outcomes using current best evidence rules (Moore 2010) includingimputation (Moore 2012) In their response to Rothrsquos letter the authors of the original review of products available in the USA showrather similar effect sizes of oral diclofenac and topical diclofenac with different penetration enhancers (Barthel 2011) in such studies
4 The Roth letter sought to differentiate between topical diclofenac preparations based on the penetration enhancers used Thatdifferent formulations may have different effect sizes is a fair point to make Two of the studies in our review of topical NSAIDs inacute conditions used diclofenac sodium 1 gel comparing it with either diclofenac epolamine gel (Gallacchi 1990 50 participants)or lysine clonixinate gel (Hofman 2000 142 participants) no difference between formulations was demonstrated It is difficult tomake any judgement for topical NSAIDs in acute conditions due to the relatively small number and particularly the small size ofstudies We did an analysis by drug and this showed that some topical NSAIDs were consistently beneficial irrespective offormulation while others had little or no efficacy This fits in with some theoretical considerations of molecular architecture andtissue penetration (Moore 2008) In chronic pain where there are larger studies and much more data we have considered formulation(Derry in preparation)
References
Roth SH Letter to the editor The importance of differentiating between topical NSAIDs Postgraduate Medicine 2011123251-2[103810pgm2011092482]Moore RA Derry S McQuay HJ Topical agents in the treatment of rheumatic pain Rheumatic Diseases Clinics of North America200834415-32
142Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Moore RA Eccleston C Derry S Wiffen P Bell RF Straube S et al ACTINPAIN Writing Group of the IASP Special Interest Groupon Systematic Reviews in Pain Relief Cochrane Pain Palliative and Supportive Care Systematic Review Group Editors ldquoEvidencerdquo inchronic pain - establishing best practice in the reporting of systematic reviews Pain 2010150386-9Moore RA Straube S Eccleston C Derry S Aldington D Wiffen P et al Estimate at your peril imputation methods for patientwithdrawal can bias efficacy outcomes in chronic pain trials using responder analyses Pain 2012153265-8Barthel HR Axford-Gatley RA Response to Roth Letter to the Editor Postgraduate Medicine 2011123253-4 [103810pgm2011092483]
Contributors
Feedback from Sigrun ChrubasikAuthors involved with responding Andrew Moore Sheena DerryFeedback Editor Kate Seers
Query submitted by Peter C Goslashtzsche 3 November 2015
Summary
Date of Submission 03-Nov-2015Name Peter C GoslashtzscheEmail Address pcgcochranedkAffiliation Nordic Cochrane CentreRole DirectorComment The authors found that the results were missing from 5900 patients Furthermore there was extreme heterogeneity in theirmeta-analyses eg I square was 92 for the diclofenac trials which were the most common ones and there was extreme funnel plotasymmetry with the largest trials showing the smallest effects (the authors didnrsquot show funnel plots but I constructed one for diclofenac)Moreover the trials were industry funded of relatively poor quality and the authors analysed published data not data from clinicalstudy reports and did not try to obtain all the missing trials and data from the manufacturersThe authors concluded that topical NSAIDs are effective in providing pain relief but also cautioned that the large amounts of unpublisheddata ldquocould influence results in updates of this reviewrdquo They certainly could I believe it is plain wrong to perform meta-analyses on theauthorsrsquo data When I most recently reviewed this area for the BMJ in 2010 I concluded that we donrsquot know whether topical NSAIDsare beneficial (1)1 Goslashtzsche PC NSAIDs Clin Evid (Online) 2010 Jun 28I agree with the conflict of interest statement belowI certify that I have no affiliations with or involvement in any organization or entity with a financial interest in the subject matter ofmy feedback
Reply
Response submitted by authors Sheena Derry and Professor Andrew Moore on 4 November 2015Peter has made these same points elsewhere [PubMed Commons PubPeer gt Cochrane Database Syst Rev October 2015 httpspubpeercompublicationsA9E5BEA36549727357F9FD14CC2537]As a preliminary it is important to stress that the number of trials available in pain has increased over time and newer trials are typicallybetter and larger This means we can move from answering the simple question of whether an intervention works to more importantand relevant questions such as how well the intervention works and for drugs particularly to examine effects of dose and formulationFormulation is now recognised to have profound effects For oral analgesics for example fast acting formulations demonstrate up todouble the analgesic effect for a given dose as our recently updated overview points out (Cochrane Database Syst Rev 2015 Sep 289CD008659 see also Pain 2014 Jan155(1)14-21)What was true for oral analgesics in acute pain is now true also for topical NSAIDs in acute pain where manufacturers have beenputting a lot of effort into trying to produce new formulations that work better This updated review sought to examine the influenceof formulation and drug although dose is somewhat more difficult There are good reasons why formulation may be important(Rheumatic Disease Clinics 2008 Vol 34 Issue 2 p 415-432)
143Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Our searches did identify a large number of unpublished trials We deplore this but are powerless to change things For a previousreview of topical NSAIDs we contacted all identified manufacturers and asked for published or unpublished data The yield was small(as others have found) but some unpublished studies were brought into the public domain Waiting for all studies and clinical trialreports of all studies would take forever and could probably never be achieved We believe that many of these unpublished studiesrelate to drugs andor formulations that have never been manufactured commercially While these would be of interest in determiningwhat does and doesnrsquot work and to direct future research they would probably have little clinical relevance because these formulationsare unlikely to come to marketPeterrsquos main issue appears to be heterogeneity Tests for heterogeneity are problematical anyway (Pain 2000 85415-24) and the I squareof 92 that he quotes was for all topical diclofenac formulations combined While we do give an overall summary for diclofenac wedemonstrated in the review that the different diclofenac formulations produced different results from one another using LrsquoAbbeacute plotsand in our detailed analyses showing large variations in efficacy between them In the circumstance a high I square for all combined(clinical heterogeneity) is to be expected but it is not relevant The bulk of the studies on diclofenac were published in the last fiveyears were of decent quality and moderate to large size There were older data for ketoprofen but again these showed major differencesbetween formulations Differences between formulations are highlighted throughout the reviewTrying to determine publication bias using funnel plots or other measures is something of a lost cause (Journal of Clinical Epidemiology200053207-16) It is especially so with small numbers of trials (Journal of Clinical Epidemiology 2000 53 477-484) and makingsense of funnel plots is anything but easy for most people (Journal of Clinical Epidemiology 2005 58 894-901) A useless methodseems an odd choice to make to criticise our reviewThere are very good scientific reasons why drug and formulation may play a big part in the efficacy of topical NSAIDs This is alsothe case for oral analgesics used in acute pain where formulation improvements generating rapid absorption confers greater efficacySimple lumping strategies may have been permissible in past systematic review methodology but a more forensic approach is needednow and in the future This is what we have attempted to do in this latest reviewComparisons with Peterrsquos 2010 review seem inappropriate since that review was based on other reviews that are now out of date
Contributors
Feedback from Peter C GoslashtzscheAuthors involved with responding Andrew Moore Sheena DerryFeedback Editor Kate Seers
W H A T rsquo S N E W
Last assessed as up-to-date 4 February 2015
Date Event Description
4 November 2015 Feedback has been incorporated Feedback has been incorporated See Feedback 2
H I S T O R Y
Protocol first published Issue 4 2008
Review first published Issue 6 2010
144Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Date Event Description
13 October 2015 Amended Small errors found in Summary Table B for total num-ber of participants with piroxicam and indomethacinand in text for NNT with diclofenac (other gel) Nochange to conclusions
6 October 2015 Amended Small error found in Summary Table B percentagesfor NSAID and placebo were the wrong way round forcomparison of all NSAIDs and placebo RR and textwere correct No change to results or conclusions
13 February 2015 New citation required but conclusions have notchanged
Conclusions not changed Results remain essentiallythe same but the focus has changed to examine drugand formulation combination
3 February 2015 New search has been performed Title changed from Topical NSAIDs for acute pain inadults to Topical NSAIDs for acute musculoskeletalpain in adults to increase specificityNew searches run and new studies identified Fourteennew included studies using diclofenac (3489 new par-ticipants a 63 increase) four new excluded studiesFifteen studies awaiting classification (completed noresults available but likely to satisfy inclusion criteria)
23 May 2014 Amended Error in data reported for clinical success in Hosie 1993was brought to our attention and has been corrected
12 June 2012 Feedback has been incorporated We have incorporated feedback received from DrSigrun Chrubasik and the authorrsquos response on DMSOand other additives
C O N T R I B U T I O N S O F A U T H O R S
For the earlier review Tom Massey and SD identified studies and carried out data extraction analysis and drafting RAM and HJMwere involved in planning acted as adjudicators and were involved with writing
For this update SD and RAM carried out searches data extraction and analysis All authors were involved with writing the review
145Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
D E C L A R A T I O N S O F I N T E R E S T
SD has no conflicts relating to this review or any similar product
RAM has no conflicts relating to this review or any similar product
PW has no conflicts relating to this review or any similar product
HG has no conflicts relating to this review or any similar product
MM has no conflicts relating to this review or any similar product
S O U R C E S O F S U P P O R T
Internal sources
bull Oxford Pain Relief Trust UKGeneral institutional support
External sources
bull No sources of support supplied
D I F F E R E N C E S B E T W E E N P R O T O C O L A N D R E V I E W
For this update in 2015 we have changed the title to specify musculoskeletal pain because topical NSAIDs are not normally used totreat visceral pain or headache We felt that the new title better reflected the content of the review We have also changed the focus ofthe review from pooled analysis of all topical NSAIDs and all studies of a particular NSAID to an examination of individual drug andits formulation This makes the review much more relevant We have expanded the rsquoRisk of biasrsquo assessment and added a rsquoSummaryof findingsrsquo table and PRISMA flow chart We have removed a number of sensitivity analyses because they were not appropriate giventhe current information on the impact of formulation on efficacy The sensitivity analyses have been superseded by the rsquoRisk of biasrsquoassessment and taken into account in the rsquoSummary of findingsrsquo tables
An earlier review in 2004 chose to exclude studies using benzydamine on the grounds that it was no longer considered to be an NSAID(Mason 2004a) Although the protocol for this review stated that we would not include benzydamine after further consultation wenow believe that it should be classified as an NSAID albeit with a different mode of action which is not fully understood (Quane1998) Thus we have reinstated studies using topical benzydamine
I N D E X T E R M S
Medical Subject Headings (MeSH)
Acute Disease Administration Topical Anti-Inflammatory Agents Non-Steroidal [lowastadministration amp dosage adverse effects] AthleticInjuries [drug therapy] Pain [lowastdrug therapy] Randomized Controlled Trials as Topic Sprains and Strains [drug therapy]
146Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
MeSH check words
Adult Humans
147Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
T A B L E O F C O N T E N T S
1HEADER 1ABSTRACT 2PLAIN LANGUAGE SUMMARY 4SUMMARY OF FINDINGS FOR THE MAIN COMPARISON 6BACKGROUND 7OBJECTIVES 7METHODS
10RESULTS Figure 1 11Figure 2 13Figure 3 14Figure 4 16Figure 5 18Figure 6 20
25DISCUSSION 27AUTHORSrsquo CONCLUSIONS 28ACKNOWLEDGEMENTS 28REFERENCES 36CHARACTERISTICS OF STUDIES
102DATA AND ANALYSES Analysis 11 Comparison 1 Individual NSAID versus placebo Outcome 1 Clinical success 104Analysis 12 Comparison 1 Individual NSAID versus placebo Outcome 2 Local adverse events 106Analysis 21 Comparison 2 Diclofenac versus placebo (effect of formulation) Outcome 1 Clinical success 108Analysis 31 Comparison 3 Ibuprofen versus placebo (effect of formulation) Outcome 1 Clinical success 109Analysis 41 Comparison 4 Ketoprofen versus placebo Outcome 1 Clinical success 110Analysis 51 Comparison 5 All topical NSAIDs versus placebo Outcome 1 Local adverse events 111Analysis 52 Comparison 5 All topical NSAIDs versus placebo Outcome 2 Systemic adverse events 114Analysis 53 Comparison 5 All topical NSAIDs versus placebo Outcome 3 Adverse event withdrawals 116Analysis 61 Comparison 6 Topical NSAID versus active comparator Outcome 1 Clinical success - topical piroxicam vs
topical indomethacin 118Analysis 62 Comparison 6 Topical NSAID versus active comparator Outcome 2 Local adverse events - topical piroxicam
vs topical indomethacin 119119APPENDICES 142FEEDBACK 144WHATrsquoS NEW 144HISTORY 145CONTRIBUTIONS OF AUTHORS 145DECLARATIONS OF INTEREST 146SOURCES OF SUPPORT 146DIFFERENCES BETWEEN PROTOCOL AND REVIEW 146INDEX TERMS
iTopical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
[Intervention Review]
Topical NSAIDs for acute musculoskeletal pain in adults
Sheena Derry1 R Andrew Moore1 Helen Gaskell2 Mairead McIntyre3 Philip J Wiffen1
1Pain Research and Nuffield Department of Clinical Neurosciences (Nuffield Division of Anaesthetics) University of Oxford OxfordUK 2Department of Clinical Geratology Oxford University Hospitals NHS Trust John Radcliffe Hospital Oxford UK 3West HoeSurgery Plymouth UK
Contact address Sheena Derry Pain Research and Nuffield Department of Clinical Neurosciences (Nuffield Division of Anaesthetics)University of Oxford Pain Research Unit Churchill Hospital Oxford Oxfordshire OX3 7LE UK sheenaderryndcnoxacuk
Editorial group Cochrane Pain Palliative and Supportive Care GroupPublication status and date Edited (no change to conclusions) comment added to review published in Issue 11 2015Review content assessed as up-to-date 4 February 2015
Citation Derry S Moore RA Gaskell H McIntyre M Wiffen PJ Topical NSAIDs for acute musculoskeletal pain in adults CochraneDatabase of Systematic Reviews 2015 Issue 6 Art No CD007402 DOI 10100214651858CD007402pub3
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
A B S T R A C T
Background
Use of topical NSAIDs to treat acute musculoskeletal conditions has become widely accepted because they can provide pain reliefwithout associated systemic adverse events This review is an update of rsquoTopical NSAIDs for acute pain in adultsrsquo originally publishedin Issue 6 2010
Objectives
To determine the efficacy and safety of topically applied NSAIDs in acute musculoskeletal pain in adults
Search methods
We searched the Cochrane Register of Studies Online MEDLINE and EMBASE to February 2015 We sought unpublished studies byasking personal contacts and searching online clinical trial registers and manufacturers websites For the earlier review we also searchedour own in-house database and contacted manufacturers
Selection criteria
We included randomised double-blind active or placebo (inert carrier)-controlled trials in which treatments were administered toadults with acute pain resulting from strains sprains or sports or overuse-type injuries (twisted ankle for instance) There had to be atleast 10 participants in each treatment arm with application of treatment at least once daily
Data collection and analysis
Two review authors independently assessed studies for inclusion and extracted data We used numbers of participants achieving eachoutcome to calculate the risk ratio and numbers needed to treat for an additional beneficial outcome (NNT) or additional harmfuloutcome (NNH) compared with placebo or other active treatment We reported 95 confidence intervals (CI) We were particularlyinterested to compare different formulations (gel cream plaster) of individual NSAIDs
1Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Main results
For this update we added 14 new included studies (3489 participants) and excluded four studies We also identified 20 additionalreports of completed or ongoing studies that have not been published in full The earlier review included 47 studies
This update included 61 studies Most compared topical NSAIDs in the form of a gel spray or cream with a similar topical placebo5311 participants were treated with a topical NSAID 3470 with placebo and 220 with an oral NSAID This was a 63 increase inthe number of included participants over the previous version of this review We also identified a number of studies in clinical trialregistries with unavailable results amounting to about 5900 participants for efficacy and 5300 for adverse events
Formulations of topical diclofenac ibuprofen ketoprofen piroxicam and indomethacin demonstrated significantly higher rates ofclinical success (more participants with at least 50 pain relief ) than matching topical placebo (moderate or high quality data)Benzydamine did not Three drug and formulation combinations had NNTs for clinical success below 4 For diclofenac the Emulgelregformulation had the lowest NNT of 18 (95 CI 15 to 21) in two studies using at least 50 pain intensity reduction as the outcomeDiclofenac plasters other than Flectorreg also had a low NNT of 32 (26 to 42) based on good or excellent responses in some studiesKetoprofen gel had an NNT of 25 (20 to 34) from five studies in the 1980s some with less well defined outcomes Ibuprofen gelhad an NNT of 39 (27 to 67) from two studies with outcomes of marked improvement or complete remission All other drug andformulation combinations had NNT values above 4 indicating lesser efficacy
There were insufficient data to compare reliably individual topical NSAIDs with each other or the same oral NSAID
Local skin reactions were generally mild and transient and did not differ from placebo (high quality data) There were very few systemicadverse events (high quality data) or withdrawals due to adverse events (low quality data)
Authorsrsquo conclusions
Topical NSAIDs provided good levels of pain relief in acute conditions such as sprains strains and overuse injuries probably similar tothat provided by oral NSAIDs Gel formulations of diclofenac (as Emugelreg) ibuprofen and ketoprofen and some diclofenac patchesprovided the best effects Adverse events were usually minimal
Since the last version of this review the new included studies have provided additional information In particular information ontopical diclofenac is greatly expanded The present review supports the previous review in concluding that topical NSAIDs are effectivein providing pain relief and goes further to demonstrate that certain formulations mainly gel formulations of diclofenac ibuprofenand ketoprofen provide the best results Large amounts of unpublished data have been identified and this could influence results inupdates of this review
P L A I N L A N G U A G E S U M M A R Y
Topical non-steroidal anti-inflammatory drugs for acute musculoskeletal pain in adults
Acute musculoskeletal pain describes conditions like a sprained ankle or a muscle pull These usually get better over two or three weekswithout treatment but can be very painful while they last
Topical non-steroidal anti-inflammatory drugs (NSAIDs) are applied to unbroken skin where it hurts as gels creams sprays or plastersTopical NSAIDs penetrate the skin enter tissues or joints and reduce processes causing pain in the tissue Drug levels in the bloodwith topical NSAIDs are very much lower than with the same drug taken by mouth This minimises the risk of harmful effects
We searched medical databases for clinical trials comparing topical NSAIDs with placebo (creams or gels that do not contain a medicine)or other medicines in adults aged 16 years or older with musculoskeletal pain (typically sports injuries) The evidence is current toFebruary 2015
This review is an update of rsquoTopical NSAIDs for acute pain in adultsrsquo originally published in Issue 6 2010 We identified 14 new studiesto add to the 47 studies included in the earlier review We also identified 14 studies in a clinical trial registry that are completed andthree short reports from meetings for which we could not find full details (about 4500 participants) Three more studies are ongoing(almost 900 participants)
The 61 included studies involving 8386 participants were generally of high-quality They tested a number of different topical drugsmostly against a topical placebo (carrier without the NSAID) with application at least once a day We were interested in participants
2Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
having good pain reduction (by about half ) around seven days after treatment started At later times most people are expected to getbetter even without treatment
We looked at particular formulations of individual drugs Gel formulations of diclofenac and ketoprofen were among the most effectivealong with ibuprofen gel and diclofenac plaster For diclofenac and ketoprofen gels 7 or 8 people out of 10 with a painful strain sprainor muscle pull had much reduced pain after seven days compared with only 2 or 3 out of 10 with placebo (high quality data) OtherNSAIDs and formulations were better than placebo but not by as much Because both topical NSAIDs and topical placebo are rubbedinto the skin in these studies we know that any effect is not just from rubbing
About 1 in 20 people experienced a mild and short-lived side effect like redness at the application site This was the same for topicalNSAID and topical placebo (high quality data) Side effects like a stomach upset or feeling sick were uncommon with no differencebetween topical NSAID and topical placebo (high quality data) There were no serious side effects
3Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
S U M M A R Y O F F I N D I N G S F O R T H E M A I N C O M P A R I S O N [Explanation]
Topical NSAIDs compared with topical placebo for acute musculoskeletal pain in adults
Patient or population adults with strains sprains or muscle pullSettings communityIntervention topical NSAID (topical diclofenac ibuprofen and ketoprofen gels only shown here for efficacy)Comparison topical placebo
Outcomes Probable outcome withintervention
Probable outcome withcomparator
RR NNT NNTp or NNH(95 CI)
No of studies partici-pants
Quality of the evidence(GRADE)
Comments
Topical diclofenac gel
(as Emulgel)
Clinical success (eg50 reduction in pain)
780 in 1000 200 in 1000 RR34 (27 to 55)NNT18 (15 to 21)
2 studies314 participants
High Consistent results in 2moderately sized recentstudies of high quality
Topical ibuprofen gel
Clinical success (eg
50 reduction in pain)
420 in 1000 160 in 1000 RR27 (17 to 42)NNT39 (27 to 67)
2 studies241 participants
Moderate Modest effect size andnumbers of participants
Topical ketoprofen gel
Clinical success (eg
50 reduction in pain)
720 in 1000 330 in 1000 RR22 (17 to 28)NNT25 (20 to 34)
5 studies348 participants
Moderate Modest effect size andnumbers of participantsbut studies small withnone recent
All topical NSAIDs
Local adverse events
46 in 1000 50 in 1000 RR10 (080 to 12)NNH not calculated
42 studies6125 participants
High Large number of studiesand participants with con-sistent results
All topical NSAIDs
Systemic adverse
events
32 in 1000 35 in 1000 RR10 (07 to 13)NNH not calculated
38 studies5372 participants
High Large number of studiesand participants with con-sistent results
All topical NSAIDs
Withdrawals - adverse
events
11 in 1000 11 in 1000 RR10 (07 to 17)NNH not calculated
42 studies5790 participants
High Large number of studiesand participants with con-sistent results4
To
pica
lN
SA
IDs
for
acu
tem
uscu
losk
ele
talp
ain
inad
ults
(Revie
w)
Co
pyrig
ht
copy2015
Th
eC
och
ran
eC
olla
bo
ratio
nP
ub
lished
by
Joh
nW
iley
ampS
on
sL
td
Serious adverse events 1 in total 0 in total Not calculated All data Low Small numbers of events
GRADE Working Group grades of evidenceHigh quality Further research is very unlikely to change our confidence in the estimate of effectModerate quality Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimateLow quality Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimateVery low quality We are very uncertain about the estimate
CI confidence interval RR risk ratio NNT number needed to treat for an additional beneficial outcome NNTp number needed to treatto prevent an event happening NNH number needed to treat for an additional harmful outcome
xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
5To
pica
lN
SA
IDs
for
acu
tem
uscu
losk
ele
talp
ain
inad
ults
(Revie
w)
Co
pyrig
ht
copy2015
Th
eC
och
ran
eC
olla
bo
ratio
nP
ub
lished
by
Joh
nW
iley
ampS
on
sL
td
B A C K G R O U N D
This review is an update of a review originally published in Issue 62010 on rsquoTopical NSAIDs for acute pain in adultsrsquo (Massey 2010)We have changed the title to specify musculoskeletal pain becausetopical NSAIDs are not normally used to treat visceral pain orheadache We felt that the new title better reflected the content ofthe review
The use of topical NSAIDs for pain relief has been a controversialsubject in analgesic practice In some parts of the world (includingmuch of Western Europe) they have been available for many yearsare widely available without prescription widely advertised usedextensively and evidence for their use is considered adequate Inother parts of the world they were regarded as little more thanplacebo with any apparent effect attributed to the process of rub-bing at the site of the affected area In some places (for examplethe US) their use was almost unknown until the mid-2010s InEngland 52 million prescriptions for topical NSAIDs were dis-pensed in 2013 (PACT 2014) mainly for formulations of ibupro-fen (245 million) piroxicam (118 million) and diclofenac (127million)
There is good evidence for the efficacy of topical NSAIDs in acuteand chronic musculoskeletal pain (Mason 2004a Mason 2004bMoore 1998a) In the US the Food and Drug Administration li-censed topical nonsteroidal products in 2007 and in England theNational Institute for Health and Care Excellence (NICE) recom-mended topical therapies as first line treatment in its guidelinesfor osteoarthritis in 2008 (NICE 2008) Earlier reviews of topicalanalgesics covered studies investigating the underlying science toexplain biological plausibility in addition to clinical trials (Anon2005 Moore 2008a)
This review is one of a series on topical analgesics including topicalcapsaicin at low and high doses (Derry 2012a Derry 2013) andtopical NSAIDs in chronic pain conditions (Derry 2012b) andsalicylate-containing rubefacients (Derry 2014)
Description of the condition
Acute pain is usually defined as pain of less than three monthsrsquo du-ration It is often associated with injury including trauma surgerymusculoskeletal injuries such as strains sprains and over-use in-juries or soft tissue injuries such as muscle soreness or cramps
Description of the intervention
Clinicians prescribe NSAIDs on a routine basis for a range of mildto moderate pain NSAIDs are the most commonly prescribedanalgesic medications worldwide and their efficacy for treatingacute pain has been well demonstrated (Moore 2003) They re-versibly inhibit cyclooxygenase (prostaglandin endoperoxide syn-thase) the enzyme mediating production of prostaglandins and
thromboxane A2 (FitzGerald 2001) Prostaglandins mediate a va-riety of physiological functions such as maintenance of the gastricmucosal barrier regulation of renal blood flow and regulation ofendothelial tone They also play an important role in inflamma-tory and nociceptive processes However relatively little is knownabout the mechanism of action of this class of compounds asidefrom their ability to inhibit cyclooxygenase-dependent prostanoidformation (Hawkey 1999)NSAIDs taken orally or intravenously are transported to all partsof the body in the blood and relatively high blood concentrationsare needed to achieve effective tissue concentrations at the site ofthe pain and inflammation These high concentrations throughoutthe body can give rise to a number of adverse events that canbe unpleasant (for example dyspepsia) or potentially serious (forexample gastrointestinal bleeding)A topical medication is a one applied to body surfaces such asthe skin or mucous membranes to treat ailments A large rangeof types of topical formulation may be used including but notlimited to creams foams gels lotions ointments and plastersThe exact formulation of a topical medication is often determinedby the speed of drug absorption required The need may be forslow absorption into the circulation to maintain low drug concen-trations and perhaps avoiding extensive first pass metabolism inthe liver plasters containing drug reservoirs may be used for thisas with transdermal opioids or contraceptive steroids For rapidabsorption the formulation is enhanced by substances to improveor assist skin penetration perhaps only to generate high concen-trations in tissues rather than in the blood gel formulations areuseful for this purpose which is why they are sometimes used fortopical NSAIDs
Topical NSAIDs
Topical NSAIDs are formulated for direct application to thepainful site and to produce a local pain-relieving effect whileavoiding body-wide distribution of the drug at physiologically ac-tive levels (McPherson 2013) This method of application (dos-ing) necessarily limits their use to more superficial painful condi-tions such as sprains strains and muscle or tendon soreness Theywould not for example be indicated for deep visceral pain orheadaches They are also not appropriate for use on broken skinso would not be used on open wounds (accidental or surgical)
How the intervention might work
For a topical formulation to be effective it must first penetratethe skin Only when the drug has entered the lower layers ofthe skin can it be absorbed by the blood and transported to thesite of action or penetrate deeper into areas where inflammationoccurs Individual drugs have different degrees of penetration Abalance between lipid and aqueous solubility is needed to optimisepenetration and use of prodrug esters has been suggested as a way
6Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
of enhancing permeability Formulation is also crucial to goodskin penetration Experiments with artificial membranes or humanepidermis suggest that creams are generally less effective than gelsor sprays but newer formulations such as microemulsions mayhave greater potentialOnce the drug has reached the site of action it must be presentat a sufficiently high concentration to inhibit cyclooxygenase en-zymes thereby reducing prostaglandin synthesis This in turn re-duces inflammation and relieves pain It is probable that in acuteconditions topical NSAIDs exert their action primarily by lo-cal reduction of symptoms independent of any systemic uptakeand delivery Tissue levels of NSAIDs applied topically certainlyreach levels high enough to inhibit cyclooxygenase-2 (Anon 2005Haroutiunian 2010 Moore 2008a) However plasma concentra-tions found after topical administration are only a fraction (usu-ally much less than 5) of the levels found in plasma followingoral administration Topical application can potentially limit sys-temic adverse events by increasing local effects and minimisingsystemic concentrations of the drug We know that the incidenceof upper gastrointestinal bleeding is low with chronic use of topi-cal NSAIDs (Evans 1995) although it has been reported partic-ularly in people with risk factors (Zimmerman 1995) We haveno certain knowledge of lower effects on cardiovascular events orrenal failure both of which have been associated with oral NSAIDuse
Why it is important to do this review
Since the last review in 2010 a number of new studies have beenpublished nearly all of which investigated various formulations ofdiclofenac These new studies are generally of higher quality thanmany of the earlier ones in this review and have the potential toinfluence the strength of its conclusions substantially Moreoverthe additional information allows for analysis based not only on aparticular drug but also on the formulation of that drug This canprovide better insight into whether formulation affects efficacy oftopical NSAIDs in acute musculoskeletal painAn updated review of evidence for topical NSAIDs was needed toinform choices made by consumers prescribers and commission-ers (purchasers of healthcare)
O B J E C T I V E S
To determine the efficacy and safety of topically applied NSAIDsin acute musculoskeletal pain in adults
M E T H O D S
Criteria for considering studies for this review
Types of studies
Randomised controlled double-blind studies comparing topicalNSAIDs with placebo (inert carrier) or other active treatment foracute pain with at least 10 participants per treatment arm andoutcomes close to seven days (minimum three days) We excludedstudies published only as short abstracts (which report insufficientdata to assess methods) or studying experimentally induced pain(which does not correlate well with clinical pain) Because a cross-over design is not appropriate for self limiting conditions such assprains strains and contusions we only considered parallel-groupdesigns
Types of participants
Adults (aged 16 years or more) with acute musculoskeletal pain ofat least moderate intensity resulting mainly from strains sprainsor sports injuries Typically for sports injuries the injury wouldhave occurred within 24 or 48 hours
Types of interventions
Included studies had at least one treatment arm using a topicalNSAID and a comparator arm using placebo (inert carrier withoutNSAID or other active treatment) The topical NSAID had to beapplied at least once daily We did not include salicylates in thisreview as they are no longer classified as topical NSAIDs and arecovered in a separate review (Derry 2014)
Types of outcome measures
We sought information on participant characteristics includingage sex and condition treated
Primary outcomes
The primary outcome was rsquoclinical successrsquo defined as at least a50 reduction in pain or equivalent measure such as a rsquovery goodrsquoor rsquoexcellentrsquo global assessment of treatment or rsquononersquo or rsquoslightrsquopain on rest or movement measured on a categorical scale (Moore1998a) We used the following hierarchy of outcomes to extractdata for the primary outcome
bull Participant reported reduction in pain of at least 50bull Participant reported global assessment of treatmentbull Pain on movementbull Pain at rest or spontaneous painbull Undefined rsquoimprovementrsquo
We used only participant reported outcomes of efficacy and notphysician or investigator reported outcomes
7Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Secondary outcomes
bull Numbers of participants with adverse events local andsystemic
bull Numbers of withdrawals all cause lack of efficacy andadverse events
We anticipated that outcomes would be reported after differentdurations of treatment and extracted data reported as close toseven days as possible with a minimum of three days We also ex-tracted data for outcomes reported after longer durations of treat-ment We anticipated that reporting of adverse events would varybetween studies with regard to the terminology used method ofascertainment and categories reported (for example occurring inat least 5 of participants or where there is a statistically signifi-cant difference between treatment groups) We took care to iden-tify these details where relevant
Search methods for identification of studies
Electronic searches
We searched the following databases without language restrictionbull CENTRAL (The Cochrane Library) Issue 4 2009 for the
original review and the Cochrane Register of Studies Online(CRSO) to 3 February 2015 for this update
bull MEDLINE (via Ovid) from inception to December 2009for the original review and from 2008 to 3 February 2015 forthis update
bull EMBASE (via Ovid) from inception to December 2009for the original review and from 2008 to 3 February 2015 forthis update
bull Oxford Pain Relief Database for the original review (Jadad1996a) This resource is no longer being updated
See Appendix 1 for the CENTRAL search strategy Appendix 2 forthe MEDLINE search strategy and Appendix 3 for the EMBASEsearch strategy
Searching other resources
We searched the reference lists of review articles and includedstudies We have previously asked manufacturers for details ofunpublished studies but did not make new requestsWe searched two clinical trial registries (clinicaltrialsgov (clinicaltrialsgov) and the World Health Organization In-ternational Clinical Trials Registry Platform (appswhointtrialsearch)) and asked personal contacts about ongoing and un-published studies
Data collection and analysis
Selection of studies
We screened the titles and abstracts of studies identified by thesearches to eliminate those that clearly did not satisfy the inclu-sion criteria and obtained full reports of the remaining studies todetermine inclusion in the review
Data extraction and management
Review authors were not blinded to the authorsrsquo names and in-stitutions journal of publication or study results at any stage ofthe review Two review authors independently selected the stud-ies for inclusion assessed methodological quality and risk of biasand extracted data We resolved disagreements and uncertaintiesthrough discussionWe abstracted information on participants interventions and out-comes from the original reports into a standard data extractionform One review author entered data suitable for meta-analysisinto Review Manager 5 (RevMan 2014) and another review au-thor checked it
Assessment of risk of bias in included studies
We used the Oxford Quality Score as the basis for inclusion lim-iting inclusion to studies that were randomised and double-blindas a minimum (Jadad 1996b)Two review authors independently assessed risk of bias for eachstudy using the criteria outlined in the Cochrane Handbook forSystematic Reviews of Interventions with any disagreements resolvedby discussion (Higgins 2011) We assessed the following for eachstudy
bull Random sequence generation (checking for possibleselection bias) We assessed the method used to generate theallocation sequence as low risk of bias (any truly randomprocess for example random number table computer randomnumber generator) unclear risk of bias (method used to generatesequence was not clearly stated) We excluded studies using anon-random process that were therefore at high risk of bias (forexample odd or even date of birth hospital or clinic recordnumber)
bull Allocation concealment (checking for possible selectionbias) The method used to conceal allocation to interventionsbefore assignment determines whether intervention allocationcould have been foreseen in advance of or during recruitmentor changed after assignment We assessed the methods as lowrisk of bias (for example telephone or central randomisationconsecutively numbered sealed opaque envelopes) unclear riskof bias (method was not clearly stated) We excluded studies thatdid not conceal allocation and were therefore at high risk of bias(for example open list)
8Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
bull Blinding of outcome assessment (checking for possibledetection bias) We assessed the methods used to blind studyparticipants and outcome assessors from knowledge of whichintervention a participant received We assessed the methods aslow risk of bias (study stated that it was blinded and describedthe method used to achieve blinding for example identicaltubes containing gel or identical plasters matched in appearanceand smell) unclear risk of bias (study stated that it was blindedbut did not provide an adequate description of how blinding wasachieved) We excluded studies that were not double-blind andwere therefore at high risk of bias
bull Incomplete outcome data (checking for possible attritionbias due to the amount nature and handling of incompleteoutcome data) We assessed the methods used to deal withincomplete data as low risk of bias (less than 10 ofparticipants did not complete the study or used rsquobaselineobservation carried forwardrsquo analysis or both) unclear risk ofbias (used rsquolast observation carried forwardrsquo analysis) or high riskof bias (used rsquocompleterrsquo analysis)
bull Size (checking for possible biases confounded by small size)Small studies have been shown to overestimate treatment effectsprobably due to methodological weaknesses (Dechartres 2013Nuumlesch 2010) We assessed studies as at low risk of bias if theyhad at least 200 participants at unclear risk if they had 50 to 200participants and at high risk if they had fewer than 50participants
Measures of treatment effect
We used risk ratio (RR) to establish statistical difference and num-bers needed to treat for an additional beneficial outcome (NNT)with 95 confidence intervals (CI) We pooled percentages as ab-solute measures of benefit or harmWe used the following terms to describe adverse outcomes in termsof harm or prevention of harm
bull When significantly fewer adverse outcomes occurred withtreatment than with control (placebo or active) we used the termthenumber needed to treat to prevent one event (NNTp)
bull When significantly more adverse outcomes occurred withtreatment compared with control (placebo or active) we usedthe term the number needed to treat for an additional harmfuloutcome or cause one event (NNH)
Unit of analysis issues
Randomisation was to the individual participant
Dealing with missing data
Wherever possible we used intention-to-treat (ITT) analysis wherethe ITT population consists of participants who were randomisedapplied at least one dose of the assigned study medication andprovided at least one post-baseline assessment We assigned miss-ing participants zero improvement
We also looked for information about methods of imputation formissing data
Assessment of heterogeneity
We examined heterogeneity visually using LrsquoAbbeacute plots (LrsquoAbbeacute1987) a visual method for assessing differences in results of indi-vidual studies and with the I2 statistic
Assessment of reporting biases
The aim of this review was to use dichotomous outcomes of knownutility and of value to patients (Moore 2013) The review did notdepend on what the authors of the original studies chose to reportor not Studies that did not report dichotomous results but onlyaverage pain data did not contribute to analysesWe assessed publication bias using a method designed to detect theamount of unpublished data with a null effect required to makeany result clinically irrelevant (usually taken to mean an NNT of10 or higher Moore 2008b)
Data synthesis
We pooled data only for comparisons and outcomes where therewere at least two studies and 200 participants (Moore 1998b)When two active treatment arms were compared with a placeboarm we took care to avoid double counting of participants in theplacebo arm if both active groups contributed to an analysis wesplit the placebo group between themWe calculated RRs with 95 CIs using the fixed-effect model(Morris 1995) A statistically significant benefit of topical NSAIDover control was assumed when the lower limit of the 95 CI ofthe RR was greater than one A statistically significant benefit ofcontrol over active treatment was assumed when the upper limitof the 95 CI was less than one We calculated NNTs with 95CIs using the pooled number of events by the method of Cookand Sackett (Cook 1995)Statistically significant differences between NNTs for differenttopical NSAIDs were tested using the z test (Tramer 1997) wherethere were sufficient data to do so and where the studies weresufficiently similar in types of participant outcome and durationto make such comparisons sensible
Subgroup analysis and investigation of heterogeneity
We carried out separate analyses for individual NSAIDs andwhere the data permitted for different formulations of individualNSAIDs
Sensitivity analysis
The earlier review included sensitivity analyses for various factorsthat are now covered by the assessment of risk of bias
9Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
R E S U L T S
Description of studies
Results of the search
New searches for this update identified 20 publications that wereexamined in further detail to determine inclusion status We alsoidentified 20 additional studies in clinical trial registries See Figure1
10Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Figure 1 Study flow diagram
11Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
We identified 14 new studies (11 publications three unpub-lished reports) satisfying our inclusion criteria (Costantino2011 Coudreuse 2010 Gonzaacutelez de Vega 2013 Hofman2000 Klainguti 2010 Kuehl 2011 Li 2013 NCT01255423NCT01272934 NCT01272947 Predel 2012 Predel 2013aPredel 2013b Saillant 1998) one post hoc analysis of a study thatwas included in the earlier review (Mueller 2010 in Predel 2004)and an additional publication and a pooled analysis that includednew data for another study from the earlier review (Lionberger2011 pooled analysis in Joussellin 2003)
Included studies
All except one of the new studies compared diclofenac withplacebo A number of different formulations were used includ-ing diclofenac epolamine (DHEP) with or without heparin (Flec-toparin Tissugel or Flector EP Tissugel) applied as a plaster (orpatch) diclofenac diethylamine (DDEA) applied as a gel and di-clofenac with lethicin applied as a spray gel The remaining studycompared diclofenac gel with traumeel a ldquofixed combination ofplant and mineral extractsrdquo applied as a gel or an ointmentThere were 47 studies in the original review 14 new studies wereincluded making a total of 61 studies in this updated reviewAll used a parallel group design Forty-four compared a topicalNSAID with placebo 13 a topical NSAID with an active com-parator (a different topical NSAID an oral NSAID the same top-ical NSAID in a different formulation or a compound of plantand mineral extracts) and four had both placebo and active com-parators In total 5311 participants were treated with a topicalNSAID 3470 with placebo and 220 with an oral NSAID Top-ical NSAIDs used were benzydamine diclofenac etofenamatefelbinac fentiazac flunoxaprophen flurbiprofen ibuprofen in-domethacin ketoprofen ketorolac lysine clonixinate meclofe-namic acid naproxen niflumic acid and piroxicam They wereapplied as creams gels sprays foams or plasters (patches) Topicalplacebos were the inert carriers without the active NSAID OralNSAIDs used were ibuprofen (as tablets) and indomethacin (ascapsules)Most studies enrolled participants who had sprains strains andcontusions usually as a result of sports injuries and treatmentwas started within a few hours or days Other studies enrolledparticipants with overuse-type injuries such as tendinitis and acutelow back pain where pain had been present for days or weeks butless than three monthsParticipants were treated for at least five days and up to threeweeks with most studies lasting seven to 14 days Participantswere usually assessed in clinic at intervals during treatment andsometimes also at home using daily patient diaries We used out-comes closest to seven days because many of these injuries are self
limiting with differences between active treatment and placebobeing diminished or lost after longer intervalsMost studies reported dichotomous outcomes suitable for a re-sponder analysis although group mean change (for pain or phys-ical function for example) was usually the primary outcomesHowever the definition of response varied both in the parametermeasured (for example pain pain on movement patient globalevaluation of treatment) and in the scale used to measure it (forexample a 3- 4- or 5-point scale for patient global evaluation)Details of included studies are in the Characteristics of includedstudies tableWe identified 14 completed but apparently unpublished studies ina clinical trial registry for which no results have been posted (4403participants NCT00351104 NCT00352625 NCT00426985NCT00640705 NCT00640939 NCT00680472NCT00680784 NCT00765700 NCT00869063NCT00869180 NCT00931866 NCT01874626NCT01957215 NCT02324270) We have placed these underCharacteristics of studies awaiting classification We also identi-fied three conference abstracts that relate to completed studies thatdo not appear to have been published but that may satisfy ourinclusion criteria One is likely to be the same study as one of theincluded studies identified in a clinical trial registry (Pallay 2013in NCT01272947) one relates to another study identified in theclinical trial registry that is awaiting classification (Ekman 2010in NCT00765700) while we could find no published or unpub-lished reports of the other Sarzi-Puttini 2014)We also identified three ongoing studies with an estimated en-rolment of 880 participants (NCT01945034 NCT02100670NCT02290821) Details are in the Characteristics of ongoingstudies table
Excluded studies
For the original review 25 studies were excluded after readingthe full paper For this update we excluded four new studies (Cesarone 2008 Coulibaly 2009 Kuwabara 2013 Vinciguerra2008) Details are in the Characteristics of excluded studies table
Risk of bias in included studies
All studies were randomised and double-blind One study scored25 (Sinniger 1981) 23 scored 35 23 scored 45 and 14 scored55 for methodological quality using the Oxford Quality Scale Abreakdown of the scores for individual studies is reported in theCharacteristics of included studies tableComments on potential biases in individual studies are reported inthe rsquoRisk of biasrsquo section of the Characteristics of included studiestable The findings are displayed in Figure 2 and Figure 3
12Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Figure 2 Risk of bias summary review authorsrsquo judgements about each risk of bias item for each includedstudy
13Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Figure 3 Methodological quality graph review authorsrsquo judgements about each methodological quality
item presented as percentages across all included studies
Allocation
All the studies were randomised but only 17 adequately describedthe method used to generate the random sequence Thirteen stud-ies adequately described the method used to conceal allocation ofthe sequence No studies were at high risk of bias for this item
Blinding
All studies were double-blind and 38 adequately described themethod used to maintain the blinding No studies were at highrisk of bias for this item
Incomplete outcome data
Thirty-six studies included all participants in the primary analysisor provided sufficient data to allow missing participants to be in-cluded as non-responders and were judged at low risk of bias Wejudged four studies to be at high risk of attrition bias (Campbell1994 Kuehl 2011 Maziegraveres 2005a Russell 1991) Three unpub-lished studies contributed only to adverse event analyses and ac-counted for all participants for these outcomes (NCT01255423NCT01272934 NCT01272947)
Other potential sources of bias
We judged one study that included more than 200 participantsin each treatment arm to be at low risk of bias from size butthis study had a very high attrition rate (see rsquoIncomplete outcome
data (attrition bias)rsquo) and did not report all the efficacy outcomesmeasured (Kuehl 2011) We judged 27 studies to be at high riskbecause they included fewer than 50 participants per treatmentarm
Effects of interventions
See Summary of findings for the main comparison TopicalNSAIDs compared with topical placebo for acute musculoskeletalpain in adultsThree studies did not contribute data suitable for analysis of at leastone outcome (Gonzaacutelez de Vega 2013 Gualdi 1987 Hoffmann2012)
1 Topical NSAID versus placebo
Details of efficacy outcomes in individual studies are in Appendix4 and of adverse events and withdrawals in Appendix 5 Appendix6 has details of the concentration of topical products the amountapplied the frequency of application and an estimation of thedaily dose of topical NSAID applied Not all studies providedsufficient information to allow calculation of daily dose appliedFor example for topical diclofenac the estimated doses appliedvaried between about 60 and 280 mg for topical ketoprofen 100to about 450 mg for topical ibuprofen 300 to 800 mg
Participants with clinical success
14Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Topical diclofenac versus placebo
Ten studies contributed to this analysis (Coudreuse 2010Joussellin 2003 Klainguti 2010 Li 2013 Predel 2004 Predel2012 Predel 2013a Predel 2013b Rowbotham 2003) of whichone (Predel 2012) had two active treatment arms A total of 1074participants were treated with topical diclofenac and 976 withplacebo (Analysis 11 Figure 4)
15Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Figure 4 Forest plot of comparison 2 Individual NSAID versus placebo outcome 21 Clinical success
16Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
bull The proportion of participants experiencing successfultreatment with topical diclofenac was 74 (8001074 range39 to 100)
bull The proportion of participants experiencing successfultreatment with placebo was 47 (461976 range 8 to 82)
bull The RR for treatment compared with placebo was 16(95 CI 15 to 17)
bull The NNT for successful treatment was 37 (32 to 43) Forevery four participants treated with topical diclofenac one wouldexperience successful treatment who would not have done sowith placebo
Effect of formulation
The effects of formulation are shown in Analysis 21 and Figure5
17Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Figure 5 LrsquoAbbeacute plot of clinical success in studies of topical diclofenac versus topical placebo The size ofthe symbol is proportional to the size of the study (inset scale) Dark blue Emulgel light blue spraygel red
Flector pink other patch or plaster
bull Four studies used a Flectorreg plaster (1030 participantsJoussellin 2003 Li 2013 Rowbotham 2003 Saillant 1998) TheRR for treatment compared with placebo was 15 (14 to 17)and the NNT was 47 (37 to 65)
bull Three studies used other makes of plaster (474 participantsCoudreuse 2010 Klainguti 2010 Predel 2004) The RR fortreatment compared with placebo was 16 (14 to 18) and theNNT was 32 (26 to 42)
bull Two studies used Voltaren Emulgel (314 participants Predel2012 Predel 2013b) The RR for treatment compared withplacebo was 38 (27 to 55) and the NNT was 18 (15 to 21)
bull One study used a spray gel (232 participants Predel2013a) The RR for treatment compared with placebo was 12(105 to 13) and the NNT was 80 (48 to 24)
Diclofenac as the gel formulation Emulgel was statistically more ef-ficacious than the plaster formulation as Flector plaster (z = 6360P value lt 000001)
Topical ibuprofen versus placebo
Five studies contributed to this analysis (Billigmann 1996Campbell 1994 Dreiser 1988 Machen 2002 Ramesh 1983) Atotal of 218 participants were treated with topical ibuprofen and218 with placebo (Analysis 11)
bull The proportion of participants experiencing successfultreatment with topical ibuprofen was 55 (120218 range 31to 81)
18Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
bull The proportion of participants experiencing successfultreatment with placebo was 33 (73218 range 13 to 76)
bull The RR of treatment compared with placebo was 16 (13to 20)
bull The NNT for successful treatment was 46 (33 to 80) Forevery five participants treated with topical ibuprofen one wouldexperience successful treatment who would not have done sowith placebo
Effect of formulation
The effects of formulation are shown in Analysis 31bull Three studies used cream formulations (195 participants
Campbell 1994 Dreiser 1988 Ramesh 1983) Although this isjust below our threshold for pooled analysis we have includedthis analysis for completeness and the results should beinterpreted with caution The RR for treatment compared withplacebo was 13 (103 to 16) and the NNT was 64 (34 to 41)
bull Two studies used gel formulations (241 participantsBilligmann 1996 Machen 2002) The RR for treatmentcompared with placebo was 27 (17 to 42) and the NNT was39 (27 to 67)
There was no statistically significant difference between the geland cream formulations (z = 1160 P value = 0246)
Topical ketoprofen versus placebo
Seven studies contributed to this analysis (Airaksinen 1993Dreiser 1989 Julien 1989 Kockelbergh 1985 Maziegraveres 2005bMaziegraveres 2005a Noret 1987) A total of 346 participants weretreated with topical ketoprofen and 337 with placebo (Analysis11)
bull The proportion of participants experiencing successfultreatment with topical ketoprofen was 73 (251346 range57 to 89)
bull The proportion of participants experiencing successfultreatment with placebo was 47 (157337 range 17 to 73)
bull The RR of treatment compared with placebo was 16 (14to 18)
bull The NNT for successful treatment was 39 (30 to 53) Forevery four participants treated with topical ketoprofen onewould experience successful treatment who would not have doneso with placebo
Effect of formulation
The effects of formulation are shown in Analysis 41 and Figure6
19Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Figure 6 LrsquoAbbeacute plot of clinical success in studies of topical ketoprofen versus topical placebo The size ofthe symbol is proportional to the size of the study (inset scale) Light blue ketoprofen gel pink ketoprofen
plaster
Two studies used a plaster formulation (335 participants Maziegraveres2005b Maziegraveres 2005a) The RR for treatment compared withplacebo was 12 (104 to 14) and the NNT was 82 (45 to 47)Five studies used gel formulations (348 participants Airaksinen1993 Dreiser 1989 Julien 1989 Kockelbergh 1985 Noret 1987)The RR for treatment compared with placebo was 22 (17 to 28)and the NNT was 25 (20 to 34)Ketoprofen as a gel formulation was statistically more efficaciousthan a plaster formulation (z = 3860 P value = 000014)
Topical piroxicam versus placebo
Three studies contributed to this analysis (Aoki 1984 Fujimaki1985 Russell 1991) A total of 255 participants were treated withtopical piroxicam and 249 with placebo (Analysis 11)
bull The proportion of participants experiencing successfultreatment with topical piroxicam was 68 (179255 range 53to 79)
bull The proportion of participants experiencing successfultreatment with placebo was 47 (118249 range 45 to 49)
bull The RR of treatment compared with placebo was 15 (13to 17)
bull The NNT for successful treatment was 44 (32 to 69) Forevery four participants treated with topical piroxicam one wouldexperience successful treatment who would not have done sowith placebo
Topical indomethacin versus placebo
20Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Three studies contributed to this analysis ( kermark 1990 Aoki1984 Fujimaki 1985) A total of 168 participants were treatedwith topical indomethacin and 173 with placebo (Analysis 11)
bull The proportion of participants experiencing successfultreatment with topical indomethacin was 58 (97168 range54 to 64)
bull The proportion of participants experiencing successfultreatment with placebo was 46 (79173 range 25 to 49)
bull The RR of treatment compared with placebo was 13 (103to 16)
bull The NNT for successful treatment was 83 (44 to 65) Forevery eight participants treated with topical indomethacin onewould experience successful treatment who would not have doneso with placebo
Topical benzydamine versus placebo
Three studies contributed to this analysis (Chatterjee 1977 Haig1986 Linde 1985) A total of 96 participants were treated withtopical benzydamine and 97 with placebo (Analysis 11) Al-though this is just below our threshold for pooled analysis we haveincluded this analysis for completeness and the results should beinterpreted with caution
bull The proportion of participants experiencing successfultreatment with topical benzydamine was 77 (7496 range70 to 86)
bull The proportion of participants experiencing successfultreatment with placebo was 67 (6597 range 48 to 80)
bull The RR of treatment compared with placebo was 12 (096to 14) There was no statistically significant difference betweentreatments (Figure 4)
Results for participants with clinical success with individual topi-cal NSAIDs where there were adequate data for analysis are sum-marised below in rsquoSummary of results Arsquo and Analysis 11 Analysis21 Analysis 31 and Analysis 41
Summary of results A Participants with clinical success
Comparison Studies Participants NSAID()
Placebo () Relative benefit (95CI)
NNT (95 CI)
Diclofenac -Flector plaster
4 1030 63 41 15 (14 to 17) 47 (37 to 65)
Diclofenac -other plaster
3 474 88 57 16 (14 to 18) 32 (26 to 42)
Diclofenac -Emulgel
2 314 78 20 38 (27 to 55) 18 (15 to 21)
Diclofenac -other gel
1 232 94 82 12 (11 to 13) 80 (48 to 24)
Ibuprofen -cream
3 195 71 56 13 (103 to 16) 64 (34 to 41)
Ibuprofen - gel 2 241 42 16 27 (17 to 42) 39 (27 to 67)
Ketoprofen -plaster
2 335 73 60 12 (104 to 14) 82 (45 to 47)
Ketoprofen - gel 5 348 72 33 22 (17 to 28) 25 (20 to 34)
Piroxicam 3 504 70 47 15 (13 to 17) 44 (32 to 69)
Indomethacin 3 341 58 46 13 (103 to 16) 83 (44 to 65)
21Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Benzydamine 3 193 77 67 12 (096 to 14) not calculated
Results for these two comparisons are derived from very smallamounts of data and are provided here for completeness Theyshould be interpreted with caution
Local adverse events
Local adverse events were irritation of the area to which the topicalNSAID was applied including redness or erythema and itch orpruritus Where reported these were usually described as mild andtransient
All topical NSAIDs versus placebo
Forty-two studies contributed to this analysis of which three com-pared two different drugs with placebo (Aoki 1984 Diebshlag1990 Fujimaki 1985) Three studies had two treatment armscomparing different formulations or application regimens for di-clofenac with placebo which have been combined for this analysis
(Costantino 2011 Klainguti 2010 Predel 2012) In total 3619participants were treated with topical NSAIDs and 3121 withplacebo (Analysis 51)
bull The proportion of participants experiencing a local adverseevent with a topical NSAID was 43 (1553619 range 0 to33)
bull The proportion of participants experiencing a local adverseevent with placebo was 46 (1453121 range 0 to 32)
bull The RR of topical NSAID compared with placebo was 098(080 to 12)
bull There was no significant difference between treatmentgroups so the NNH was not calculated
Individual topical NSAIDs versus placebo
Results for local adverse events with individual topical NSAIDswhere there were adequate data for analysis are in Summary ofresults B and Analysis 12
Summary of results B Participants with local adverse events
Comparison Studies Participants NSAID()
Placebo()
RR(95 CI)
NNH(95 CI)
All NSAIDs 42 6740 43 46 098 (080 to 12) Not calculated
Diclofenac 15 3271 31 43 078 (056 to 11) Not calculated
Ketoprofen 8 852 11 95 12 (083 to 17) Not calculated
Piroxicam 3 522 23 54 042 (017 to 11) Not calculated
Felbinac 3 397 30 15 19 (049 to 75) Not calculated
Indomethacin 3 354 63 22 27 (091 to 77) Not calculated
Ibuprofen 3 321 10 43 23 (098 to 54) Not calculated
Systemic adverse events
All topical NSAIDs versus placebo
Thirty-six studies contributed data on systemic adverse eventsof which three compared two different drugs with placebo (Aoki1984 Diebshlag 1990 Fujimaki 1985) Two studies had two treat-
22Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
ment arms comparing different formulations or application regi-mens for diclofenac with placebo which have been combined forthis analysis (Klainguti 2010 Predel 2012) In total 2956 partic-ipants were treated with a topical NSAID and 2620 with placebo(Analysis 52)
bull Twenty-three studies reported no systemic adverse events inany arm of the study
bull The proportion of participants experiencing a systemicadverse event with a topical NSAID was 31 (922956)
bull The proportion of participants experiencing a systemicadverse event with placebo was 35 (912620)
bull The RR of topical NSAID compared with placebo was 096(073 to 13)
bull There was no significant difference between treatmentgroups so the NNH was not calculated
A further six studies did not report the occurrence or oth-erwise of systemic adverse events (Billigmann 1996 Julien1989 Kockelbergh 1985 Noret 1987 Ramesh 1983 Vecchiet1991) while two studies did not report numbers of participantswith systemic adverse events ( kermark 1990 Auclair 1989)Costantino 2011 reported that there were no systemic gastroin-testinal adverse events Two studies reported only on total adverseevents without distinguishing between local and systemic events(NCT01255423 NCT01272934)
Serious adverse events
Two studies reported serious adverse events In Hoffmann 2012one participant experienced three serious adverse events none ofwhich was judged to be related to the study medication (diclofenacplaster) In NCT01272934 one participant using diclofenac gelruptured the ligaments of the wrist There was no statement aboutlikely relationship to the study medication but this seems unlikely
Adverse event withdrawals
Forty-two studies reported data relating to adverse event with-drawals of which three compared two different drugs with placebo(Aoki 1984 Diebshlag 1990 Fujimaki 1985) Two studies hadtwo treatment arms comparing different formulations or applica-tion regimens for diclofenac with placebo which have been com-bined for this analysis (Klainguti 2010 Predel 2012) In total3365 participants received a topical NSAID and 3040 placebo(Analysis 53)
bull Forty-four comparisons reported no adverse eventwithdrawals
bull The proportion of participants withdrawing from the studydue to an adverse event after treatment with a topical NSAIDwas 098 (333365)
bull The proportion of participants withdrawing from the studydue to an adverse event after treatment with placebo was 099(303040)
bull The RR of topical NSAID compared to placebo was 10(064 to 16)
bull There was no significant difference between treatmentgroups so the NNH was not calculated
Four studies did not specifically mention adverse event with-drawals (Haig 1986 Hoffmann 2012 Klainguti 2010 Vecchiet1991) while one reported that one participant withdrew with mildpruritus but did not state the treatment arm (Joussellin 2003)Ten studies specifically reported withdrawals due to lack of efficacy(Dreiser 1989 Dreiser 1994 Kuehl 2011 Machen 2002 Maziegraveres2005b Maziegraveres 2005a Noret 1987 Predel 2013a Russell 1991Thorling 1990) (Appendix 5) Numbers of participants withdraw-ing were generally low with rates of 6 or less except in Kuehl2011 where the rate was 10 with active treatment (diclofenacplaster) and 12 with placebo We did not carry out any analysisbecause the outcome was inconsistently reported
2 Topical NSAID versus active comparator
Details of efficacy outcomes in individual studies are in Appendix4 and of adverse events and withdrawals in Appendix 5
Participants with clinical success
Topical NSAID versus oral NSAID
bull kermark 1990 compared indomethacin spray withindomethacin capsules with response rates of 55 (1222) withspray and 23 (522) with capsules
bull Hosie 1993 compared felbinac foam with ibuprofen tabletswith response rates of 64 (81127) with felbinac foam and72 (96133) with ibuprofen tablets
bull Whitefield 2002 compared ibuprofen gel with ibuprofentablets with response rates of 60 (3050) with gel and 54(3650) with tablets
There were insufficient data for meta-analysis for any one of thesecomparisons felbinac is not known to be better than placebo
Topical NSAID versus different formulation of the same
topical NSAID
bull Fioravanti 1999 compared DHEP (diclofenac) gelformulated with and without lecithin with response rates of70 (3550) in both treatment arms
bull Mahler 2003 compared DHEP (diclofenac) gel formulatedwith and without lecithin with response rates of 89 (8292)with lecithin and 70 (6288) without lecithin
bull Gallacchi 1990 compared topical diclofenac formulated asFlectorreg gel and Emugelreg with response rates of 76 (1925)in both treatment arms
23Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
bull Governali 1995 compared topical ketoprofen cream withgel with response rates of 93 (1415) with cream and 27 (415) with gel
There were insufficient data for analysis
Topical NSAID versus different topical NSAID
Eight studies compared one topical NSAID versus at least oneother topical NSAID piroxicam versus indomethacin (Aoki1984 Fujimaki 1985 Sugioka 1984) ibuprofen versus ketopro-fen (Curioni 1985 Picchio 1981) ketoprofen versus etofena-mate (Curioni 1985 Tonutti 1994) ibuprofen versus etofenamate(Curioni 1985) ketorolac versus etofenamate (Diebshlag 1990)and diclofenac versus lysine clonixinate (Hofman 2000)There were sufficient data to compare only piroxicam with in-domethacin (Aoki 1984 Fujimaki 1985 Sugioka 1984 Analysis61)
bull The proportion of participants experiencing clinical successwith topical piroxicam was 56 (185330 range 49 to 78)
bull The proportion of participants experiencing clinical successwith topical indomethacin was 45 (140311 range 33 to64)
bull The RR of piroxicam compared with indomethacin was 12(11 to 14)
bull The NNT for successful treatment was 91 (53 to 30) Forevery nine participants treated with topical piroxicam onewould experience a clinical success who would not haveexperienced one with topical indomethacin
Topical NSAID versus different topical intervention
One study compared diclofenac gel with a herbal product calledTraumeel gel under double-blind conditions with response ratesfor being pain-free at seven days of 8137 (58) with diclofenacgel and 7140 (50) with Traumeel gel (Gonzaacutelez de Vega 2013)
Local adverse events
Topical NSAID versus oral NSAID
Two studies comparing a topical NSAID with an oral NSAIDprovided data on local adverse events ( kermark 1990 Hosie1993) There were five events with topical NSAID and three withoral NSAID which were too few for analysis
Topical NSAID versus different topical NSAID
All nine studies comparing one topical NSAID with at least oneother reported on local adverse events with a total of 48 events in1005 participants (48) There were sufficient data to compare
only piroxicam with indomethacin (Aoki 1984 Fujimaki 1985Sugioka 1984 Analysis 62)
bull The proportion of participants experiencing local adverseevents with topical piroxicam was 21 (7340 range 12 to28)
bull The proportion of participants experiencing local adverseevents with topical indomethacin was 10 (33331 range 29to 15)
bull The RR of piroxicam compared with indomethacin was021 (009 to 047)
bull The NNT to prevent a local adverse event was 13 (87 to23) For every thirteen participants treated with topicalpiroxicam one would not experience a local adverse event whowould have experienced one with topical indomethacin
Topical NSAID versus different topical intervention
Gonzaacutelez de Vega 2013 reported that adverse events were infre-quent and mild to moderate in intensity but did not distinguishbetween local and systemic events Numbers of participants expe-riencing any adverse event were 8147 with diclofenac gel and 14148 with Traumeel gel
Systemic adverse events
kermark 1990 reported numbers of events rather than numbersof participants with events while Tonutti 1994 and Whitefield2002 reported no adverse events attributable to the study medi-cation and Fioravanti 1999 Gallacchi 1990 Gualdi 1987 andSugioka 1984 did not mention systemic adverse events Gonzaacutelezde Vega 2013 did not distinguish between local and systemicevents In the remaining studies a total of 16 events were reportedin topical NSAID treatment arms (797 participants 2) and 11with ibuprofen tablets (134 participants 8)
Serious adverse events
No serious adverse events were reported in any treatment arm
Withdrawals
The only withdrawals reported due to adverse events were in stud-ies with placebo treatment arms ( kermark 1990 Fujimaki 1985)and have been reviewedThree studies reported withdrawals due to lack of efficacy (Gonzaacutelez de Vega 2013 Hofman 2000 Tonutti 1994) (Appendix5) There were insufficient data for analysisSome studies reported exclusions from analysis (efficacy or safetyor both) following randomisation mainly due to protocol vio-lations or loss to follow-up (Appendix 5) There is no reason tobelieve these exclusions would introduce systematic bias and thenumbers involved were not likely to influence results
24Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
D I S C U S S I O N
This updated review of topical NSAIDs for acute musculoskele-tal pain in adults differs from previous reviews Previously dataallowed only for comparison of individual topical NSAIDs withplacebo irrespective of formulation With a substantial amount ofnew data for diclofenac it is possible to distinguish effects of for-mulation for individual NSAIDs Because formulation chemistrycan substantially affect the rate and total amount of drug accessingsubcutaneous injured tissues the effect of formulation may be asimportant as the individual NSAID used Drug and formulationshould thus be considered together when assessing efficacy andthis is now possible
We have also included an assessment of the daily dose of NSAIDapplied to the skin This involved having information on the con-centration of NSAID in the preparation the amount used andthe frequency of use Not all studies reported all three but an esti-mation of topical doses applied was possible It varied by factors ofthree or four for each topical NSAID However for topical formu-lations the key issue is less the dose applied but the amount thatpenetrates locally (producing analgesic effect) and the amount en-tering the systemic circulation (producing potential harms) Bothwill depend on the exact formulation of the topical agent andwhether there is occlusion (Moore 2008a)
Summary of main results
This review included 61 studies comparing a topical NSAID withplacebo another topical NSAID or an oral NSAID In total5311 participants were treated with a topical NSAID 3470 withplacebo and 220 with an oral NSAID There were 63 moreparticipants than in the previous version of this review Conditionstreated were sprains strains and contusions mainly resulting fromsports injuries and overuse injuries such as tendinitisFormulations of topical diclofenac ibuprofen ketoprofen piroxi-cam and indomethacin demonstrated significantly higher rates ofclinical success than matching topical placebo lacking the NSAIDbenzydamine did not Three drug and formulation combinationshad NNTs for clinical success below 4 For diclofenac Emulgelreghad the lowest NNT of 18 (15 to 21) in two studies using atleast 50 pain intensity reduction as the outcome (high qualityevidence) Diclofenac plasters other than Flectorreg also had a lowNNT of 32 (26 to 42) based on good or excellent responses inrelatively recent studies (high quality evidence) Ketoprofen gelhad an NNT of 25 (20 to 34) from five studies in the 1980ssome with less well defined outcomes (moderate quality evidence)Ibuprofen gel had an NNT of 39 (27 to 67) from two stud-ies with outcomes of marked improvement or complete remission(moderate quality evidence) All other drug and formulation com-binations had NNT values above 4 indicating lesser efficacyThese results are better than alternative topical products that mightbe used for acute musculoskeletal pain There is no evidence tosupport the use of topical salicylate rubefacients (Derry 2014)
Treatment with a topical NSAID was not associated with anincrease in local adverse events (skin reactions) compared withplacebo (inert carrier) or in withdrawals due to adverse events(high quality evidence) The inert carrier was sometimes associ-ated with mild skin irritation but this rarely led to cessation oftreatment and quickly resolved Systemic adverse events were un-common and did not differ between topical NSAID and placebo(high quality evidence) Two participants experienced serious ad-verse events with diclofenac plaster and diclofenac gel but it isunlikely that these were related to the study medicationsThere were insufficient data directly comparing a topical NSAIDwith the same oral NSAID to draw conclusions about efficacyBased on very limited data for oral NSAIDs there were fewersystemic adverse events with topical than oral treatment Therewere sufficient data only for topical piroxicam compared withtopical indomethacin to compare one topical agent with anotherThese limited data suggested that piroxicam was more effectivethan indomethacin and was less likely to cause local adverse eventsIt is worth noting here that topical indomethacin did not givesignificantly better pain relief than placebo in two of the threestudies in this analysis
Overall completeness and applicability ofevidence
There is a tension between pooling studies to produce analyseswith larger numbers and the subsequent large increases in clinicaland statistical heterogeneity on the one hand and using the ap-proach of clinical homogeneity with subsequent smaller numbersof participants on the other hand Previous reviews have takenthe former approach that is useful in demonstrating that topicalNSAIDs rsquoworkrsquo by being significantly better than placebo Becauseof the substantial increase in the amount of data available in thisreview we have chosen to seek greater clinical homogeneity thisproduces results that are more relevant to patient and prescriberchoiceThere were too few studies comparing one topical NSAID versusanother or versus the same oral NSAID to allow meaningful directcomparisons between individual drugs or routes of administrationThe conditions treated in these studies are representative of thoselikely to be suitable for acute treatment with topical NSAIDs Themean age of participants in individual studies ranged from 25 to57 years and the nature of recruitment in many studies meant thatparticipants were actively engaged in sporting activities Neverthe-less older people in their 60s to 80s were also included in somestudies and the low levels of predominantly mild adverse eventsmeans that this route of administration of NSAIDs is suitable forall age groups able to manage the application processInformation from other sources mainly randomised studies last-ing 12 weeks or more in older populations with arthritis tendto confirm this A systematic review of topical NSAIDs in olderadults was difficult to interpret but suggested that the range of
25Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
withdrawal rates in these studies was similar with topical and oralNSAIDs (Makris 2010) It also claimed potentiation of warfarineffects but that was with topical salicylate not an NSAID Incontrast a pooled safety analysis of topical diclofenac in peopleaged 75 years or older reported minimal changes with a mean re-duction in haemoglobin of less than 1 gL with topical diclofenacand a mean systolic blood pressure reduction of almost 4 mm Hg(Roth 2012) Two large randomised 12-week studies comparingtopical with oral diclofenac in arthritis reported lower rates of gas-trointestinal adverse events with topical than oral especially severeevents but larger reductions in haemoglobin with oral diclofenac(Simon 2009 Tugwell 2004)The available evidence was limited by numbers to comment on rarebut potentially serious adverse events One example is the potentialfor photo-sensitivity reactions with topical ketoprofen Currentadvice from the Medicines and Healthcare products RegulatoryAgency in the UK is to avoid direct sunlight ultraviolet (UV) rayssunlamps and sunbeds while using topical ketoprofen and to seea healthcare professional or go to hospital if they experience a skinreaction to sunlight sunlamps or sunbeds (MHRA 2009)
Quality of the evidence
All included studies were both randomised and double-blind nonewas considered at high risk of methodological bias Many werecarried out in the 1980s and 1990s when methodological rigorand detailed reporting were not given such high priority and stud-ies did not always report details of the randomisation treatmentallocation and blinding processes More recent studies often didreport methodological details and tended to be larger (see Figure4 for a comparison of quality of reporting for different dates andNSAIDs) Our primary outcome of clinical success was not alwayswell-defined and was measured using different scales but againmore recent studies tended to report outcomes betterThe studies were conducted in different conditions with some-what different outcome definitions and duration and with differ-ent topical NSAIDs and formulations Moreover the small size ofmany of the studies is likely to result in considerable chance varia-tion (Counsell 1994 Moore 1998b) These factors would accountfor the high I2 values seen in several analyses Despite these sourcesof potential clinical heterogeneity most studies showed benefit oftopical NSAID over placeboThe design of studies to be able to demonstrate analgesic sensi-tivity is important in self limiting conditions such as strains andsprains Too long a duration and the condition results in sponta-neous resolution of painful symptoms while too short a durationmay be inadequate to show any effect The decision by trialists toconcentrate on outcomes closest to seven days of treatment appearsto be prudent and has been adopted in this and previous reviewsThere are potential differences in response to treatment betweenstrains and sprains and overuse-type injuries such as tendinitis
and future reviews may examine this At the present time thereare too few existing trials to explore any differences adequatelyBaseline pain may be a cause for concern Seven studies didnot report baseline pain levels (Billigmann 1996 Curioni 1985Haig 1986 NCT01255423 NCT01272934 NCT01272947Sinniger 1981) and a further 11 reported either mean levels ofless than moderate pain or a significant proportion of individualswith less than moderate pain ( kermark 1990 Aoki 1984 Auclair1989 Diebshlag 1990 Fujimaki 1985 Jenoure 1997 Linde 1985Picchio 1981 Ramesh 1983 Sugioka 1984 Whitefield 2002)using recognised scales Insufficient pain at baseline compromisesthe ability of a study to demonstrate any improvement All thenewly added studies reported baseline pain to be of at least mod-erate intensity
Potential biases in the review process
There has been greater interest in topical NSAIDs in recent yearsmainly because lower systemic drug levels reduce the risk of trou-blesome and severe adverse events particularly in the gastroin-testinal tract and renal and cardiovascular systems Most of theattention has been in chronic conditions such as osteoarthritiswith few studies in acute painful conditions Low levels of seriousadverse events with topical NSAIDs has been noted previously(Evans 1995) and the near absence of serious adverse events inthis review is unlikely to be due to any biases in the review processOne potential bias is that clinical trials for topical NSAIDs may nothave been published One previous review did find previously un-published trials (Moore 1998a) but a subsequent attempt that in-cluded extensive contacts with pharmaceutical companies revealedno additional data (Mason 2004a) While some old unpublishedstudies of topical NSAIDs in acute painful conditions may existthey constitute an unknown number of studies and participantswhose results are unknown and are likely to remain unknownFurthermore their relevance to current clinical practice may belimited as better formulations are developed New systems of trialregistration mean that we know what recent studies have been doneor are ongoing the number of studies and participants is knowneven if their results remain unknown We identified in Clinicaltri-alsgov three unpublished studies (612 participants) with adverseevent data but no dichotomous efficacy data 14 completed un-published studies (4403 participants) with no results posted andthree ongoing studies (880 participants)For the main topical NSAIDs of interest and where most informa-tion exists about 4200 participants in this review provided data onefficacy for diclofenac ibuprofen ketoprofen indomethacin andbenzydamine compared with placebo For efficacy there are un-known results from almost 5900 participants in studies known tohave been done but essentially unpublished Almost 6500 partic-ipants in this review provided information on local adverse eventsfor topical NSAIDs compared with placebo For local adverseevents the unknown results from known studies represents almost
26Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
5300 participants It is clear that identified unpublished but un-available study data amounts to a further potentially large increasein knowledge over and above the 60 increase in numbers ofparticipants already included in this updated reviewBased on efficacy data on known and available study results un-published trials showing no difference between any topical NSAIDand topical placebo and involving 5500 participants would haveto exist in order for the NNT to be as high as 9 at which pointthe effectiveness of topical NSAIDs would become clinically irrel-evant (Moore 2006) This amount of unpublished negative data isobviously available and while a negative result in all the identifiedstudies is unlikely knowledge would be greatly served by havingthese unpublished trial results availableWe have not yet attempted to obtain results from these clinicaltrials from the trial sponsors because this takes a considerableamount of time and may not be successful Moreover the studieswere spilt between nine different sponsors Cerimon Pharmaceu-ticals (five studies) Novartis (four studies) Endo Pharmaceuticals(three studies) GlaxoSmithKline (two studies) Hisamatsu Phar-maceutical (two studies) and one each from Actavis Pfizer Im-primis Pharmaceutical and Strategic Science amp Technologies
Agreements and disagreements with otherstudies or reviews
A review published in 2004 included some of the studies in thisreview and reported an NNT for all topical NSAIDs combinedof 38 (34 to 44) for clinical success equivalent to half pain reliefat seven days (Mason 2004a) That review found no differencebetween topical NSAID and placebo for local adverse events as didthis review In turn the Mason review was in broad agreement withthe original systematic review on topical NSAIDs (Moore 1998a)To our knowledge no previous review assembled sufficient trialdata to analyse results by both drug and formulation as was donehere
A U T H O R S rsquo C O N C L U S I O N S
Implications for practice
For people with acute musculoskeletal pain
Topical NSAIDs can provide good levels of pain relief in acuteconditions such as sprains strains and overuse injuries proba-bly similar to that provided by oral NSAIDs Gel formulations ofdiclofenac (as Emulgelreg) ibuprofen and ketoprofen and somediclofenac patches provide the best effects Adverse events are usu-ally minimal with topical NSAIDs
For clinicians
Topical diclofenac ibuprofen or ketoprofen gels provide goodpain relief for painful acute musculoskeletal conditions and arebetter tolerated than oral formulations These drugs and formu-lations are more likely to be cost effective than alternative topicalpreparations such as topical rubefacients
For policy makers
Topical NSAIDs are not associated with an increased incidence oflocal skin reactions compared with the inert carrier and while thecarrier may cause mild transient irritation it is rarely troublesomeTopical NSAIDs do not cause systemic (mainly gastrointestinal)problems commonly seen with oral NSAIDs making them partic-ularly useful for individuals unable to tolerate oral administrationor for whom it is contraindicated
For funders
Topical diclofenac ibuprofen or ketoprofen gels should be consid-ered for initial treatment of acute musculoskeletal painful condi-tions where there are no contraindications such as damaged skinThese drugs and formulations are more likely to be cost effectivethan alternative topical preparations such as topical rubefacientsBecause formulations of topical NSAIDs are likely to change overtime the relevant trials performed and reported in or before the1990s must be limited and may be questionable Funders mightwish to consider asking pharmaceutical companies without recenttrial evidence for their products to produce it
Implications for research
General
The general thrust of these findings is that gel formulations of top-ical diclofenac ibuprofen and ketoprofen work best but for somedrugs (ketoprofen for instance) studies were pre-1990 Thesestudies may not be relevant to products available now Becauseformulation can have a significant effect on efficacy formulationchanges should be accompanied by relevant randomised trials
Design
The design of the trials is generally good and the sports injurymodel appears to be reliable and reproducible Modern studieshave ensured that participants entering the trials have at least mod-erate pain and this helps sensitivity to detect an analgesic responseMajor changes to the design of these trials would not appear to beneeded
27Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Measurement (endpoints)
A major issue is not in the measurement of pain as most studiesespecially modern studies have used standard pain intensity andpain relief scales However reporting of average pain changes isinadequate and the use of responder analyses (at least 50 painintensity reduction or people experiencing mild or no pain) ispreferred
Comparison between active treatments
Indirect comparisons with placebo are probably as informative asuse of an active comparator
A C K N O W L E D G E M E N T S
The Oxford Pain Relief Trust the NHS Cochrane CollaborationProgramme Grant Scheme and the NIHR Biomedical ResearchCentre Programme provided support for the earlier review
The National Institute for Health Research (NIHR) is the largestsingle funder of the Cochrane Pain Palliative and SupportiveCare Review Group Disclaimer the views and opinions expressedherein are those of the authors and do not necessarily reflect thoseof the NIHR National Health Service (NHS) or the Departmentof Health
R E F E R E N C E S
References to studies included in this review
Airaksinen 1993 published data onlyAiraksinen O Venaumllaumlinen J Pietilaumlinen T Ketoprofen 25gel versus placebo gel in the treatment of acute soft tissueinjuries International Journal of Clinical PharmacologyTherapy and Toxicology 199331(11)561ndash3
kermark 1990 published data only
kermark C Forsskaringhl B Topical indomethacin in overuseinjuries in athletes A randomized double blind studycomparing Elmetacinreg with oral indomethacin andplacebo International Journal of Sports Medicine 199011(5)393ndash6
Aoki 1984 published data only
Aoki T Numajiri M Yamamoto M A well controlledcomparative study of piroxicam gel indomethacin geland placebo gel in the treatment of trauma JapanesePharmacology and Therapeutics 198412(12)101ndash17
Auclair 1989 published data onlyAuclair J Georges M Grapton X Gryp L DrsquoHooghe MMeisser RG et al A double-blind controlled multi-centerstudy of percutaneous niflumic acid gel and placebo in thetreatment of Achilles heel tendinitis Current TherapeuticResearch 198946(4)782ndash8
Billigmann 1996 published data onlyBilligmann PW Treatment of ankle distortion withibuprofen gel [Therapie von Sprunggelenksndashdistosionenmit ibuprofenndashmikrogel] Therapiewoche 199646(21)1187ndash92
Campbell 1994 published data onlyCampbell J Dunn T Evaluation of topical ibuprofen creamin the treatment of acute ankle sprains Journal of Accidentand Emergency Medicine 199411(3)178ndash82
Chatterjee 1977 published data onlyChatterjee DS A double-blind clinical study withbenzydamine 3 cream on soft tissue injuries in anoccupational health centre Journal of International MedicalResearch 19775(6)450ndash8
Costantino 2011 published data only
Costantino C Kwarecki J Samokhin AV Mautone GRovati S Diclofenac epolamine plus heparin plasterversus diclofenac epolamine plaster in mild to moderateankle sprain a randomized double-blind parallel-groupplacebo-controlled multicentre phase III trial ClinicalDrug Investigation 201131(1)15ndash26 [DOI 10216511585890-000000000-00000]
Coudreuse 2010 published data onlyCoudreuse JM de Vathaire F Effect of a plaster containingDHEP and heparin in acute ankle sprains with oedemaa randomized double-blind placebo-controlled clinicalstudy Current Medical Research and Opinion 201026(9)2221ndash8 [DOI 101185030079952010508020]
Curioni 1985 published data only
Curioni BG Di Domenica F Daolio P Spignoli GEvaluation of ibuproxam gel in traumatology [Valutazionedellrsquoefficacia terapeutica di ibudros gel in traumatologia]Clinica Europea 198524(3)456ndash60
Diebshlag 1990 published data only
Diebschlag W Nocker W Bullingham R A double-blindstudy of the efficacy of topical ketorolac tromethamine gelin the treatment of ankle sprain in comparison to placeboand etofenamate Journal of Clinical Pharmacology 199030
(1)82ndash9
Dreiser 1988 published data onlyDreiser RL Clinical trial of efficacy and tolerability oftopical ibuprofen in the treatment of tendinitis Le JournalInternational De Meacutedecine 198811915ndash31
Dreiser 1989 published data only
Dreiser RL Clinical trial - Fatsum gel FG-6 Supplied byMenarini 1989
Dreiser 1990 published data onlyDreiser RE Charlot J Lopez A Ditisheim A Clinicalevaluation of niflumic acid gel in the treatment ofuncomplicated ankle sprains Current Medical Research andOpinion 199012(2)93ndash9
28Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Dreiser 1994 published data only
Dreiser RL Roche R de Sahb R Thomas F Leutenegger EFlurbiprofen local action transcutaneous (LAT tm) clinicalevaluation in the treatment of acute ankle sprains EuropeanJournal of Rheumatology and Inflammation 199414(4)9ndash13
Fioravanti 1999 published data onlyFioravanti A Cicero MR Nerucci F Manopulo RMarcolongo R Double-blind controlled clinical study of theefficacy and tolerability of diclofenac-N-(2-hydroxyethyl)-pyrrolidine lecithin gel compared with diclofenac-N-(2-hydroxyethyl)-pyrrolidine gel in patients with periand extraarticular inflammatory diseases Drugs underExperimental and Clinical Research 199925(5)235ndash40
Fujimaki 1985 published data only
Fujimaki E et al Clinical evaluation of piroxicam gelversus indomethacin gel and placebo in the treatment ofmuscle pain a double-blind multicenter study JapanesePharmacology and Therapeutics 198512(12)119ndash37
Gallacchi 1990 published data only
Gallacchi G Mautone G Lauldi P Topical treatmentwith diclofenac hydroxyethylpyrrilidine (Flector gel 1)Clinical Trials Journal 199027(1)58ndash64
Gonzaacutelez de Vega 2013 published data only
Gonzaacutelez de Vega C Speed C Wolfarth B Gonzaacutelez JTraumeel vs diclofenac for reducing pain and improvingankle mobility after acute ankle sprain a multicentrerandomised blinded controlled and non-inferiority trialInternational Journal of Clinical Practice 201367(10)979ndash89 [DOI 101111ijcp12219]
Governali 1995 published data only
Governali E Casalini D A controlled clinical studyon ketoprofen gel 5 versus ketoprofen ointment 1in patients with post-traumatic lesions [Ricerda clinicacontrollata tra ketoprofene gel 5 e ketoprofene crema1 in pazienti con postumi di lesioni traumatiche]Riabilitazione 199528(1)61ndash9
Gualdi 1987 published data only
Gualdi A Bonollo L Martini A Forgione A Non-steroidal anti-inflammatory drugs for topical therapy intraumatology a double-blind study with flunoxaprofen andketoprofen [Antinflammatori no steroidi per uso topicoin traumatologia studio clinico con flunoxaprofene echetoprofene] Riforma Medica 1987102(10)401ndash4
Haig 1986 published data only
Haig G Portable thermogram technique for topicallyapplied benzydamine cream in acute soft-tissue injuriesInternation Journal of Tissue Reactions 19868(2)145ndash7
Hoffmann 2012 published data only
Hoffmann P Kopa ka P Gugliotta B Rovati S Efficacyand tolerability of DHEP-heparin plaster in reducingpain in mild-to-moderate muscle contusions a double-blind randomized trial Current Medical Researchand Opinion 201228(8)1313ndash21 [DOI 101185030079952012709182]
Hofman 2000 published data only
Hofman J Nasswetter G Cayetti LM Lysine clonixinate gelin soft tissue injuries Controlled randomized prospectivedouble-blind clinical trial with diclofenac [Clonixinatode lisina gel en lesiones de tejidos blandos ensayo clinicoprospectivo doble ciego al azar controlado con diclofenac]Prensa Medica Argentina 200087(5)513ndash20
Hosie 1993 published data only
Hosie GAC The topical NSAID felbinac versus oralibuprofen a comparison of efficacy in the treatment ofacute lower back injury British Journal of Clinical Research199345ndash17
Jenoure 1997 published data only
Jenoure PJ Rostan A Gremion G Meier JL Grossen RBielinki R et al Multicentre double-blind controlledclinical study on the efficacy of diclofenac epolamineTissugel plaster in patients with epicondylitis [Studiomulticentrico controllato in doppio cieco su diclofenacTissugel plaster in pazienti con epicondilite] MedicinaDello Sport 197750(3)285ndash92
Joussellin 2003 published data onlylowast Joussellin E Flector Tissugel for the treatment ofpainful ankle sprains [Flector Tissugel dans le traitementdes entorses douloureuses de la cheville] Journal deTraumatologie du Sport 2003201S5ndash9Lionberger DR Joussellin E Lanzarotti A Yanchick JMagelli M Diclofenac epolamine topical patch relieves painassociated with ankle sprain Journal of Pain Research 2011447ndash53 [DOI 102147JPRS15380]Lionberger DR Joussellin E Yanchick J Magelli MLanzarotti A Pooled analysis of clinical trial data evaluatingthe safety and effectiveness of diclofenac epolamine topicalpatch 13 for the treatment of acute ankle sprain OpenAccess Journal of Sports Medicine 2011275ndash84 [DOI102147OAJSMS17048]
Julien 1989 published data onlyJulien D Clinical trial - Fastum gel FG-8 Supplied byMenarini 1989
Klainguti 2010 published data onlyKlainguti A Forgacs A Berkes I Castellacci E A plastercontaining DHEP and heparin for mild to moderatecontusions and sprains with haematoma a double-blind randomized study Current Medical Researchand Opinion 201026(9)2243ndash51 [DOI 101185030079952010508022]
Kockelbergh 1985 published data onlyKockelbergh M Verspeelt P Caloine R Dermaux F Localanti inflammatory treatment with a ketoprofen gel currentclinical findings Journal Belge de Medecine Physique et deRehabilitation 19858(4)205ndash13
Kuehl 2011 published data only
Kuehl K Carr W Yanchick J Magelli M Rovati SAnalgesic efficacy and safety of the diclofenac epolaminetopical patch 13 (DETP) in minor soft tissue injuryInternational Journal of Sports Medicine 201132(8)635ndash43[DOI 101055s-0031-1275359]
29Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Li 2013 published data only
Li C Frangione V Rovati S Zheng Q Diclofenacepolamine medicated plaster in the treatment of minor softtissue injuries a multicenter randomized controlled trialCurrent Medical Research and Opinion 201329(9)1137ndash46[DOI 101185030079952013816669]
Linde 1985 published data only
Linde F Hvass I Juumlrgensen U Madsen F Treatment ofsprained ankles with 5 benzydamine creme A double-blind study [Ankelforstuvninger behandlet med benzydamin5 creme] Ugeskrift for Laeger 1985148(1)12ndash3
Machen 2002 published data onlyMachen J Whitefield M Efficacy of a proprietary ibuprofengel in soft tissue injuries a randomised double blindplacebo controlled study International Journal of ClinicalPharmacology 200256(2)102ndash6
Mahler 2003 published data only
Mahler P Mahler F Duruz H Ramazzina M Liguori VMautone G Double-blind randomized controlled studyon the efficacy and safety of a novel diclofenac epolaminegel formulated with lecithin for the treatment of sprainsstrains and contusions Drugs under Experimental andClinical Research 200329(1)45ndash52
Maziegraveres 2005a published data onlyMaziegraveres B Rouanet S Velicy J Scarsi C Reiner V Topicalketoprofen patch (100 mg) for the treatment of anklesprain a randomized double-blind placebo-controlledstudy American Journal of Sports Medicine 200533(4)515ndash23 [DOI 1011770363546504268135]
Maziegraveres 2005b published data onlyMaziegraveres B Rouanet S Guillon Y Scarsi C Reiner VTopical ketoprofen patch in the treatment of tendinitisa randomized double blind placebo controlled studyJournal of Rheumatology 200532(8)1563ndash70
McLatchie 1989 published data onlyMcLatchie GR McDonald M Lawrence GF RogmansD Lisai P Hibberd M Soft tissue trauma a randomisedcontrolled trial of the topical application of felbinac a newNSAID British Journal of Clinical Practice 198943(8)277ndash80
Morris 1991 published data only
Morris WD Scott HV Peters WA Ketelbey JW Felbinactopical gel for acute soft tissue sports injuries New ZealandJournal of Sports Medicine 19911945ndash7
NCT01255423 unpublished data only
Novartis (Sponsors) A randomized double-blind multi-center placebo-controlled parallel group study to evaluatethe efficacy and safety of diclofenac sodium topical gel(DSG) 1 applied 4 times daily in subjects with acuteankle sprain clinicaltrialsgovct2showNCT01255423(accessed 3 February 2015) 2012 [CTG NCT01255423]
NCT01272934 unpublished data only
Novartis (Sponsors) A randomized double-blind multi-center placebo-controlled parallel group study to evaluatethe efficacy and safety of diclofenac sodium topical gel
(DSG) 1 applied four times daily in subjects withacute ankle sprain clinicaltrialsgovct2showresultsNCT01272934 (accessed 3 February 2015) 2013 [CTGNCT01272934]
NCT01272947 unpublished data onlylowast Novartis (Sponsors) A randomized double-blindmulti-center placebo-controlled parallel group study toevaluate the efficacy and safety of diclofenac sodium topicalgel (DSG) 1 applied four times daily in subjects withacute blunt soft tissue injuriescontusions of the limbsclinicaltrialsgovct2showresultsNCT01272947 (accessed3 February 2015) 2012 [CTG NCT01272947]Pallay R Pabst H Giannetti B Burnett I Monnet JDiclofenac sodium topical gel (DSG) 1 for acute softtissue injuriescontusions of the limbs 2013 AnnualAssembly of the American Academy of Physical Medicineand Rehabilitation National Harbor MD United States2013 Vol 5 (9 Suppl 1)S204
Noret 1987 published data only
Noret A Roty V Allington N Hauters P Zuinen C PoelsR Ketoprofen gel as topical treatment for sport injuriesActa Therapeutica 198713367ndash78
Parrini 1992 published data onlyParrini M Cabitza P Arrigo A Vanasia M Efficacy andtolerability of ketoprofen lysine salt foam for topical use inthe treatment of traumatic pathologies of the locomotorapparatus [Efficacia e tollerabilita del ketoprofene sale dilisina schiuma per uso topico nel trattamento di alcunepatologie traumatiche dellrsquoapparato locomotore] La ClinicaTerapeutica 1992141(9)199ndash204
Picchio 1981 published data only
Picchio AA Volta S Longoni A Controlled clinical trial ofibuprofen for topical use in sport injuries [Studio clinicocontrollato sullrsquoimpiego dellrsquoibuprofen per uso topico intraumatologica sportiva] Medicina dello Sport 198134403ndash6
Predel 2004 published data only
Mueller EA Kirch W Reiter S Extent and time courseof pain intensity upon treatment with a topical diclofenacsodium patch versus placebo in acute traumatic injury basedon a validated end point post hoc analysis of a randomizedplacebo-controlled trial Expert Opinion on Pharmacotherapy201011(4)493ndash8 [DOI 10151714656560903535898]lowast Predel HG Koll R Pabst H Dieter R Gallacchi GGiannetti B et al Diclofenac patch for topical treatmentof acute impact injuries a randomised double blindplacebo controlled multicentre study British Journal ofSports Medicine 200438(3)318ndash23 [DOI 101136bjsm2003005017]
Predel 2012 published data onlyPredel HG Hamelsky S Gold M Giannetti B Efficacy andsafety of diclofenac diethylamine 232 gel in acute anklesprain Medicine and Science in Sports and Exercise 201244
(9)1629ndash36 [DOI 101249MSS0b013e318257ed41]
30Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Predel 2013a published data only
Predel HG Giannetti B Seigfried B Novellini R MenkeG A randomized double-blind placebo-controlledmulticentre study to evaluate the efficacy and safetyof diclofenac 4 spray gel in the treatment of acuteuncomplicated ankle sprain Journal of InternationalMedical Research 201341(4)1187ndash202 [DOI 1011770300060513487639]
Predel 2013b published data onlyPredel HG Giannetti B Pabst H Schaefer A Hug AMBurnett I Efficacy and safety of diclofenac diethylamine116 gel in acute neck pain a randomized double-blindplacebo-controlled study BMC Musculoskeletal Disorders201314250 [DOI 1011861471-2474-14-250]
Ramesh 1983 published data only
Ramesh N Steuber U Ibuprofen in a cream vehicle foraccidental and sport injuries [Dolgit creme bei unfallndash undsportverletzungen] Therapiewoche 1983334563ndash70
Rowbotham 2003 published data only
Rowbotham M Galer B Block J Backonja M FlectorTissugel efficacy and safety in minor sport injuries Avs controlled clinical trial Journal of Sports Traumatology200320IS15ndashIS20
Russell 1991 published data only
Russell AL Piroxicam 05 topical gel compared to placeboin the treatment of acute soft tissue injuries a double-blind study comparing efficacy and safety Clinical andInvestigative Medicine 199114(1)35ndash43
Saillant 1998 published data only
Lionberger DR Joussellin E Yanchick J Magelli MLanzarotti A Pooled analysis of clinical trial data evaluatingthe safety and effectiveness of diclofenac epolamine topicalpatch 13 for the treatment of acute ankle sprain OpenAccess Journal of Sports Medicine 2011275ndash84 [DOI102147OAJSMS17048]lowast Saillant G Study comparing the efficacy and toleranceof Flector Tissugelreg to that of a placebo in the treatmentof benign ankle sprains [Eacutetude comparant lrsquoefficaciteacute et latoleacuterance de FlectorTissugelreg agrave celles drsquoun placebo dans letraitement des entorses beacutenignes de la cheville] Medicinedu Sport 1998721ndash5
Sanguinetti 1989 published data onlySanguinetti C Treatment of soft tissue injury with BPAAgel Results of an Italian multicenter study vs placebo[Trattemento con BPAA gel dei traumi dei tessuti moli]Clinica Terapeutica 1989130(5)255ndash258
Sinniger 1981 published data onlySinniger M Blanchard P Controlled clinical trial withFentiazac cream in sport microtraumatology Journal ofInternational Medical Research 19819(4)300ndash2
Spacca 2005 published data onlySpacca G Cacchio A Forgcs A Monteforte PRovetta G Analgesic efficacy of a lecithin-vehiculateddiclofenac epolamine gel in shoulder periarthritis andlateral epicondylitis a placebo-controlled multicenter
randomized double-blind clinical trial Drugs underExperimental and Clinical Research 200531(4)147ndash54
Sugioka 1984 published data only
Sugioka Y Multicenter clinical evaluation of piroxicam gelvs indomethacin gel in the treatment of non-traumaticdiseases of tendon or muscle Japanese Pharmacology andTherapeutics 198412139ndash53
Thorling 1990 published data only
Thorling J Linden B Berg R Sandahl A A double blindcomparison of naproxen gel and placebo in the treatment ofsoft tissue injuries Current Medical Research and Opinion199012242ndash8
Tonutti 1994 published data only
Tonutti A The use of ketoprofen gel 5 (orudis gel) intraumatology controlled double-blind study vs etofenamate[Utilizzazione del ketoprofene gel 5 (orudis gel) nellapratica traumatologica studio in doppio cieco controllatoverso etofenamato] Ortopedia e Traumatologia Oggi 199414(3)119ndash25
Vecchiet 1991 published data only
Vecchiet L Colozzi A Effects of meclofenamic acid in thetreatment of lesions deriving from minor traumatologyClinical Journal of Pain 19917 Suppl 1S54ndash9
Whitefield 2002 published data only
Whitefield M OrsquoKane CJA Anderson S Comparativeefficacy of a proprietary topical ibuprofen gel and oralibuprofen in acute soft tissue injuries a randomised doubleblind study Journal of Clinical Pharmacy and Therapeutics200227(6)409ndash17
References to studies excluded from this review
Ambrus 1987 published data only
Ambrus P Boumlhmer D Mobilat ointment in acute sprains[Mobilat Salbe bei akuten Distorsionen] Fortschritte derMedizin 1987105(13)259ndash62
Anon 1993 unpublished data only
Anon Comparative clinical efficacy of Oruvail piroxicamand diclofenac gels in soft tissue injury Unpublished
Ascherl 1982 published data onlyAscherl R Schlemmer H Blumel G Lechner F Theeffectiveness of etofenamate in minor sports injuries ofthe knee and ankle joint A double blind study [DieWirksamkeit der Etofenamat in kleineren Sportverletzungenam Knie und Sprunggelenk Eine DoppelblindndashStudie]Fortschritte der Medizin 1982100(37)1729ndash34
Bagliani 1976 published data only
Bagliani A Montalbetti L Topical treatment ofthrombophlebitis with feprazone and benzydamineControlled clinical study [Il trattamento topico ditromboflebite con feprazone e benzidamina Studio clinicocontrollato] Minerva Medica 197667(14)880ndash4
Baracchi 1982 published data only
Baracchi G Messina Denaro S Piscini S Experience ofthe topical use of isobutylfenylproprionic acid (Ibuprofen)in traumatic inflammation A double-blind comparison
31Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
with placebo [Esperienza sullrsquoimpiego topico dellrsquoacidoisobutilndashfenilndashpropionico (ibuprofen) nella flogositraumatica Confronto a doppia cecita con placebo]Gazzetta Medica Italiana 1982141(12)691ndash4
Boumlhmer 1995 published data onlyBoumlhmer D Ambrus P Treatment of muscular injurieswith diclofenac-diethylammonium emugel [Behandlungvon muskelverletzungen mit diclofenacndashdiethylammoniumemugel] Sportverletzung Sportschaden 1995994ndash5
Burnham 1998 published data only
Burnham R Gregg R Healy P Steadward R Theeffectiveness of topical diclofenac for lateral epicondylitisClinical Journal of Sport Medicine 19988(2)78ndash81
Cesarone 2008 published data onlyCesarone MR Belcaro G Pellegrini L Ledda A VinciguerraG Ricci A et al Treatment of ankle sprain in patientswith vascular diseases of the lower limbs MinervaCardioangiologica 200856 (5 Suppl)39ndash46
Coulibaly 2009 published data only
Coulibaly SK Courau S Staiger C The sprained anklecomparative study of a comfreyroot extract ointment versusa diclofenac gel [Entorses de la cheville eacutetude comparativeentre un extrait agrave base de racines de consoude et lediclofeacutenac] Phytotheacuterpie 20097147ndash9 [DOI 101007s10298-009-0383-x]
Diebschlag 1985 published data only
Diebschlag W Benzydamine cream in post-traumaticoedema International Journal of Tissue Reactions 19857(3)219ndash23
Diebschlag 1986 published data onlyDiebschlag W Diclofenac in blunt traumatic ankle jointswelling Volumetric monitoring in a placebo controlleddouble blind trial [Diclofenac in stumpfen traumatischenSprunggelenk Schwellung Volumetrische Uumlberwachungin einer Placebondashkontrollierten DoppelblindndashStudie]Fortschritte der Medizin 1986104(21)437ndash40
Diebschlag 1992 published data onlyDiebschlag W Nocker W Lehmacher W Treatment of acutesprains of the ankle joint A comparison of the effectivenessand tolerance of two gel preparations containingindomethacin [Die Behandlung der akuten Verstauchungendes Sprunggelenks Ein Vergleich der Wirksamkeitund Vertraumlglichkeit von zwei GelndashZubereitungen mitIndometacin] Fortschritte der Medizin 1992110(6)64ndash72
Fantato 1971 published data only
Fantato S De Gregorio M Clinical evaluation of topicalbenzydamine in traumatology Arzneimittel-Forschung197121(10)1530ndash5
Galer 2000 published data onlyGaler BS Rowbotham M Perander J Devers A FriedmanE Topical diclofenac patch relieves minor sports injury painresults of a multicenter controlled clinical trial Journal ofPain and Symptom Management 200019(4)287ndash94
Hallmeier 1986 published data only
Hallmeier B Michelbach B Etofenamate under tapebandages - a controlled study [Etofenamat untertapendashverbaumlnden] Medizinische Welt 198637(43)1344ndash8
Hallmeier 1988 published data onlyHallmeier B Efficacy and tolerance of etofenamate anddiclofenac in acute sports injuries [Wirksamkeit undVertraumlglichkeit von Etofenamat und Diclofenac bei akutenSportverletzungen] Rheuma 19888183ndash6
Kaneko 1999 published data only
Kaneko M Shimojo H Saito H Onuma Y YamashitaK Clinical evaluation of felbinac patch (SELSPOT) onpost-traumatic disease clinical comparative study versuscommercially available patch Japanese Pharmacology andTherapeutics 19992775ndash85
Kockelbergh 1985b published data only
Kockelbergh M Verspeelt P Caloine R Dermaux F Localanti-inflammatory treatment with a ketoprofen gel currentclinical findings [Traitement antindashinflammatoire local parun gel de keacutetoprofegravene donneacutees cliniques reacutecentes] JournalBelge de Medecine Physique et de Rehabilitation 19858(4)205-13 (study 2)
Kuwabara 2013 published data only
Kuwabara Y Hamamoto H Hikake S Miwa Y Arandomized multi-center double-blind placebo-controlledphase IIIII trial to evaluate the eftopical patch in thetreatment of pain Journal of Pain 2013 Vol 14 (4 Suppl1)S73
Lee 1991 published data only
Lee EH Lee PY Ngai AT Chiu EH Treatment of acute softtissue trauma with a topical non-steroidal anti-inflammatorydrug (biphenylacetic acid 3 gel) Singapore MedicalJournal 199132(4)238ndash41
Link 1996 published data onlyLink R Balint G Pavlik G Otto J Krause W Topicaltreatment of soft tissue rheumatism and athletic injuriesEffectiveness and tolerance of a new ketoprofen gel[Topische Behandlung von WeichteilndashRheumatismus undSportverletzungen Wirksamkeit und Vertraumlglichkeit einesneuen Ketoprofen Gel] Fortschritte der Medizin 1996114(25)311ndash4
May 2007 published data onlyMay JJ Lovell G Hopkins WG Effectiveness of1 diclofenac gel in the treatment of wrist extensortenosynovitis in long distance kayakers Journal of Scienceand Medicine in Sport 200710(1)59ndash65
Oakland 1993 published data only
Oakland C Rapier C A comparison of the efficacy of thetopical NSAID felbinac and ultrasound in the treatmentof acute ankle injuries British Journal of Clinical Research1993489ndash96
Odaglia 1987 unpublished data only
Odaglia G Sereni G Sports minor traumatology results ofa double-blind controlled clinical study ketoprofen (fastumgel 25) versus placebo Menarini unpublished data
32Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Picardi 1993 published data only
Picardi E De Iasio R Efficacy of percutaneous anti-inflammatory drugs (NSAIDs) in swimmers and waterpoloplayers [Efficacia dei farmaci antinfiammatori (FANS) pervia percutanea in atleti di nuoto e pallanuoto] ClinicaTerapeutica 1993143(6)507ndash9
Taboada 1992 published data onlyTaboada A Controlled trial of piroxicam gel associated withultrasound in acute disturbances of the locomotive system[Experienca controlada con gel de piroxicam asociado aultrasonidos en afecciones agudas del aparato locomotor]Prensa Medica Argentina 199279(10)630ndash2
Vanderstraeten 1990 published data only
Vanderstraeten G Schuermans P Study on the effect ofetofenamate 10 cream in comparison with an oral NSAIDin strains and sprains due to sports injuries Acta BelgicaMedica Physica 199013(3)139ndash41
Vinciguerra 2008 published data only
Vinciguerra G Belcaro G Cesarone MR Errichi BM DiRenzo A Errichi S et al Management of uncomplicatedankle sprains with topical or oral ketoprofen treatmentA registry study Minerva Cardioangiologica 200856 (5
Suppl)47ndash53
Von Klug 1977 published data onlyVon Klug H Experience with a locally applied anti-inflammatory drug [Erfahrungen mit einem localanwendbaren antirheumatikum] Arzneimittel-ForschungDrug Research 1977271350ndash4
References to studies awaiting assessment
NCT00351104 published data onlyEndo Pharmaceuticals (Sponsors) A randomized double-blind placebo-controlled parallel group phase III study ofthe efficacy tolerability and safety of ketoprofen topicalpatch 20 (KTP) in the treatment of pain associated withgrade 1 or grade 2 ankle sprain or strain clinicaltrialsgovct2showNCT00351104 (accessed 3 February 2015) 2008[CTG NCT00351104]
NCT00352625 unpublished data only
Endo Pharmaceuticals (Sponsors) A randomized double-blind placebo-controlled parallel group phase III study ofthe efficacy tolerability and safety of ketoprofen topicalpatch 20 (KTP) in the treatment of pain associatedwith tendonitis or bursitis of the shoulder elbow or kneeclinicaltrialsgovct2showNCT00352625 (accessed 3February 2015) 2008 [CTG NCT00352625]
NCT00426985 unpublished data only
Endo Pharmaceuticals (Sponsors) A randomized double-blind placebo-controlled parallel group phase iii study ofthe efficacy tolerability and safety of ketoprofen topicalpatch 20 (KTP) in the treatment of pain associatedwith tendonitis or bursitis of the shoulder elbow or kneeclinicaltrialsgovct2showNCT00426985 (accessed 3February 2015) 2008 [CTG NCT00426985]
NCT00640705 unpublished data only
Cerimon Pharmaceuticals (Sponsors) A randomizeddouble-blind placebo-controlled study of the efficacyand safety of a diclofenac sodium patch for the topicaltreatment of pain due to mild to moderate ankle sprainclinicaltrialsgovct2showNCT00640705 (accessed 3February 2015) 2008 [CTG NCT00640705]
NCT00640939 unpublished data onlyCerimon Pharmaceuticals (Sponsors) A randomizeddouble-blind placebo-controlled study of the efficacy andsafety of a diclofenac sodium patch for the topical treatmentof pain due to mild to moderate tendonitis or bursitisclinicaltrialsgovct2showNCT00640939 (accessed 3February 2015) 2008 [CTG NCT00640939]
NCT00680472 unpublished data onlyHisamitsu Pharmaceutical Co Inc (Sponsors) Arandomized multicenter double-blind placebo-controlledtwo-week study to assess the efficacy and safety of HKT-500 in subjects with acute shoulder pain clinicaltrialsgovct2showNCT00680472 (accessed 3 February 2015) 2015[CTG NCT00680472]
NCT00680784 unpublished data only
Hisamitsu Pharmaceutical Co Inc (Sponsors) Arandomized multicenter double-blind placebo-controlledstudy to evaluate the efficacy and safety of HKT-500 inthe treatment of pain associated with Grade I or Grade IIankle sprain clinicaltrialsgovct2showNCT00680784(accessed 3 February 2015) 2015 [CTG NCT00680784]
NCT00765700 unpublished data onlyEkman E Skrepnik S Jones M Lawson K Schupp JEfficacy and safety of ketoprofen 10 cream in acute softtissue injuries (phase 3 study tdlp-110-001) Proceedings ofthe 13th World Congress on Pain 2010 Aug 29 - Sept 2Montreal CanadaImprimis Pharmaceuticals Inc (Sponsors) A randomizedmulticenter double-blind placebo-controlled parallel-group phase 3 study to assess the efficacy and safetyof Ketotransdeltrade (Ketoprofen topical cream 10) inthe treatment of pain associated with mild to moderateacute soft tissue injury clinicaltrialsgovct2showNCT00765700 (accessed 3 February 2015) 2013 [CTGNCT00765700]
NCT00869063 unpublished data onlyCerimon Pharmaceuticals (Sponsors) A randomizeddouble-blind placebo-controlled study of the efficacy andsafety of a diclofenac sodium patch for the topical treatmentof acute pain due to mild to moderate wrist sprain strainor contusion clinicaltrialsgovct2showNCT00869063(accessed 3 February 2015) 2010 [CTG NCT00869063]
NCT00869180 unpublished data onlyCerimon Pharmaceuticals (Sponsors) A randomizeddouble-blind placebo-controlled study of the efficacy andsafety of a diclofenac sodium patch for the topical treatmentof acute pain due to mild to moderate ankle sprainclinicaltrialsgovct2showNCT00869180 (accessed 3February 2015) 2010 [CTG NCT00869180]
33Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT00931866 unpublished data only
Cerimon Pharmaceuticals (Sponsors) A randomizeddouble-blind placebo-controlled study of the efficacy andsafety of a diclofenac sodium patch for the topical treatmentof acute pain due to mild to moderate soft tissue injuriesclinicaltrialsgovct2showNCT00931866 (accessed 3February 2015) 2010 [CTG NCT00931866]
NCT01874626 unpublished data only
Strategic Science amp Technologies LLC (Sponsors) Arandomized double-blind placebo-controlled multi-dosepivotal study to determine the efficacy and safety of SST0225 a topical ibuprofen cream in the treatment of painassociated with acute ankle sprain clinicaltrialsgovct2showNCT01874626 (accessed 3 February 2015) 2014[CTG NCT01874626]
NCT01957215 unpublished data only
GlaxoSmithKline (Sponsors) A clinical study to assess theefficacy of pain relief of topical indomethacin patch overplacebo in ankle sprain patients clinicaltrialsgovct2showNCT01957215 (accessed 3 February 2015) 2015 [CTGNCT01957215]
NCT02324270 unpublished data only
Actavis Inc (Sponsors) Randomized double-blindmultiple-center placebo-controlled study comparing thesafety and efficacy of generic diclofenac epolamine toFlectorreg patch in the treatment of acute pain due to minorankle sprain clinicaltrialsgovct2showNCT02324270(accessed 3 February 2015) 2014 [CTG NCT02324270]
Sarzi-Puttini 2014 published data only
Sarzi-Puttini P Atzeni F Damiani C Casale R BarbagalloM Cazzola M A double-blind randomised parallelgroup active controlled multicentre study to assess thetherapeutic non-inferiority of SKP-021 a 03 ketoprofenpatch versus diclofenac sodium patch in patients withacute inflammatory musculoskeletal injuries Proceeding ofthe Annual European Congress of Rheumatology of theEuropean League Against Rheumatism EULAR 2014 June11-14 Paris France 201473
References to ongoing studies
NCT01945034 unpublished data onlyPfizer (Sponsors) Placebo-controlled double-blindevaluation of the efficacy and safety of ibuprofen 5 topicalgel for the treatment of ankle sprain clinicaltrialsgovct2showNCT01945034 (accessed 3 February 2015) 2014[CTG NCT01945034]
NCT02100670 unpublished data onlyGlaxoSmithKline (Sponsors) A clinical study to assessthe efficacy and onset of pain relief of topical MFC51123diclofenac-menthol gel versus controls in ankle sprainclinicaltrialsgovct2showNCT02100670 (accessed 3February 2015) 2014 [ClinicalTiralsgov NCT02100670]
NCT02290821 unpublished data only
Novartis (Sponsors) A randomized double-blind placebo-controlled parallel group study to evaluate the efficacy andsafety of diclofenac sodium topical gel (DSG) 1 applied
four times daily in subjects with acute blunt soft tissueinjuriescontusions of the limbs clinicaltrialsgovct2showNCT02290821 (accessed 3 February 2015) 2015 [CTGNCT02290821]
Additional references
Anon 2005
Anon Topical analgesics a review of reviews and a bit ofperspective wwwjr2oxacukBandolierExtraforbandoTopextra3pdf 2005
Cook 1995
Cook RJ Sackett DL The number needed to treat aclinically useful measure of treatment effect BMJ 1995310
(6977)452ndash4
Counsell 1994Counsell CE Clarke MJ Slattery J Sandercock PA Themiracle of DICE therapy for acute stroke fact or fictionalproduct of subgroup analysis BMJ 1994309(6970)1677ndash81
Dechartres 2013Dechartres A Trinquart L Boutron I Ravaud P Influenceof trial sample size on treatment effect estimates meta-epidemiological study BMJ 2013346f2304 [DOI101136bmjf2304]
Derry 2012a
Derry S Moore RA Topical capsaicin (low concentration)for chronic neuropathic pain in adults Cochrane Databaseof Systematic Reviews 2012 Issue 9 [DOI 10100214651858CD010111]
Derry 2012bDerry S Moore RA Rabbie R Topical NSAIDs forchronic musculoskeletal pain in adults Cochrane Databaseof Systematic Reviews 2012 Issue 9 [DOI 10100214651858CD007400pub2]
Derry 2013Derry S Sven-Rice A Cole P Tan T Moore RA Topicalcapsaicin (high concentration) for chronic neuropathic painin adults Cochrane Database of Systematic Reviews 2013Issue 2 [DOI 10100214651858CD007393pub3]
Derry 2014
Derry S Matthews PR Wiffen PJ Moore RASalicylate-containing rubefacients for acute and chronicmusculoskeletal pain in adults Cochrane Database ofSystematic Reviews 2014 Issue 11 [DOI 10100214651858CD007403pub3]
Evans 1995Evans JM McMahon AD McGilchrist MM White GMurray FE McDevitt DG et al Topical non-steroidal anti-inflammatory drugs and admission to hospital for uppergastrointestinal bleeding and perforation a record linkagecase-control study BMJ 1995311(6996)22ndash6
FitzGerald 2001
FitzGerald GA Patrono C The coxibs selective inhibitorsof cyclooxygenase-2 New England Journal of Medicine2001345(6)433ndash42
34Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Haroutiunian 2010
Haroutiunian S Drennan DA Lipman AG Topical NSAIDtherapy for musculoskeletal pain Pain Medicine 201011
(4)535ndash49 [DOI 101111j1526-4637201000809x]
Hawkey 1999
Hawkey CJ Cox-2 inhibitors Lancet 1999353(9149)307ndash14
Higgins 2011
Higgins JPT Altman DG Sterne JAC Chapter 8 Assessingrisk of bias in included studies In Higgins JPT GreenS (editors) Cochrane Handbook for Systematic Reviewsof Interventions Version 510 [updated March 2011]The Cochrane Collaboration 2011 Available fromwwwcochrane-handbookorg
Jadad 1996a
Jadad AR Carroll D Moore A McQuay H Developing adatabase of published reports of randomised clinical trials inpain research Pain 199666(2-3)239ndash46
Jadad 1996b
Jadad AR Moore RA Carroll D Jenkinson C ReynoldsDJ Gavaghan DJ et al Assessing the quality of reports ofrandomized clinical trials is blinding necessary ControlledClinical Trials 199617(1)1ndash12
LrsquoAbbeacute 1987
LrsquoAbbeacute KA Detsky AS OrsquoRourke K Meta-analysis inclinical research Annals of Internal Medicine 1987107(2)224ndash33
Makris 2010Makris UE Kohler MJ Fraenkel L Adverse effects oftopical nonsteroidal antiinflammatory drugs in older adultswith osteoarthritis a systematic literature review Journalof Rheumatology 201037(6)1236ndash43 [DOI 103899jrheum090935]
Mason 2004a
Mason L Moore RA Edwards JE Derry S McQuay HJTopical NSAIDs for acute pain a meta-analysis BMCFamily Practice 2004510
Mason 2004bMason L Moore RA Edwards JE Derry S McQuayHJ Topical NSAIDs for chronic musculoskeletal painsystematic review and meta-analysis BMC MusculoskeletalDisorders 2004528
McPherson 2013McPherson ML Cimino NM Topical NSAIDformulations Pain Medicine 201314 Suppl 1S35ndash9[DOI 101111pme12288]
MHRA 2009
Medicines and Healthcare Products Regulatory AgencyTopical ketoprofen reminder on risk of photo-sensitivityreactions Drug Safety Update 2009 June Vol 2 issue 116
Moore 1998aMoore RA Tramegraver MR Carroll D Wiffen PJ McQuayHJ Quantitative systematic review of topically applied non-
steroidal anti-inflammatory drugs BMJ 1998316(7128)333ndash8
Moore 1998bMoore RA Gavaghan D Tramegraver MR Collins SL McQuayHJ Size is everything - large amounts of information areneeded to overcome random effects in estimating directionand magnitude of treatment effects Pain 199878(3)209ndash16
Moore 2003
Moore RA Edwards J Barden J McQuay HJ BandolierrsquosLittle Book of Pain Oxford Oxford University Press 2003[ISBN 0ndash19ndash263247ndash7]
Moore 2006
Moore RA Barden J Derry S McQuay HJ Managingpotential publication bias In McQuay HJ Kalso EMoore RA editor(s) Systematic Reviews in Pain ResearchMethodology Refined Seattle IASP Press 200615ndash23[ISBN 978ndash0ndash931092ndash69ndash5]
Moore 2008aMoore RA Derry S McQuay HJ Topical agents in thetreatment of rheumatic pain Rheumatic Diseases Clinics ofNorth America 200834(2)415ndash32
Moore 2008bMoore RA Barden J Derry S McQuay HJ Managingpotential publication bias In McQuay HJ Kalso EMoore RA editor(s) Systematic Reviews in Pain ResearchMethodology Refined Seattle IASP Press 200815ndash24[ISBN 978ndash0ndash931092ndash69ndash5]
Moore 2013
Moore RA Straube S Aldington D Pain measures andcut-offs - rsquono worse than mild painrsquo as a simple universaloutcome Anaesthesia 201368(4)400ndash12 [DOI 101111anae12148]
Morris 1995
Morris JA Gardner MJ Calculating confidence intervalsfor relative risk odds ratios and standardised ratios andrates In Gardner MJ Altman DG editor(s) Statistics withConfidence - Confidence Intervals and Statistical GuidelinesLondon British Medical Journal 199550ndash63
NICE 2008National Institute for Health and Care ExcellenceOsteoarthritis The care and management of osteoarthritisin adults 2008 wwwniceorgukCG059fullguideline
Nuumlesch 2010Nuumlesch E Trelle S Reichenbach S Rutjes AW TschannenB Altman DG et al Small study effects in meta-analysesof osteoarthritis trials meta-epidemiological study BMJ2010341c3515 [DOI 101136bmjc3515]
PACT 2014
Prescribing and Primary Care team Health and SocialCare Information Centre Prescrption Cost AnalysisEngland 2013 Health and Social Care Information CentreThe NHS Information Centre 2014298ndash301 [ISBN978ndash1ndash78386ndash089ndash0]
35Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Quane 1998
Quane PA Graham GG Zeigler JB Pharmacology ofbenzydamine Inflammopharmacology 19986(2)95ndash107
RevMan 2014The Nordic Cochrane Centre The Cochrane CollaborationReview Manager (RevMan) 53 Copenhagen The NordicCochrane Centre The Cochrane Collaboration 2014
Roth 2012Roth SH Fuller P Pooled safety analysis of diclofenacsodium topical solution 15 (ww) in the treatment ofosteoarthritis in patients aged 75 years or older ClinicalInterventions in Aging 20127127ndash37 [DOI 102147CIAS30884]
Simon 2009
Simon LS Grierson LM Naseer Z Bookman AA ZevShainhouse J Efficacy and safety of topical diclofenaccontaining dimethyl sulfoxide (DMSO) compared withthose of topical placebo DMSO vehicle and oral diclofenacfor knee osteoarthritis Pain 2009143(3)238ndash45 [DOI101016jpain200903008]
Tramer 1997
Tramer MR Reynolds DJ Moore RA McQuay HJ Impact
of covert duplicate publication on meta-analysis a casestudy BMJ 1997315(7109)635ndash40
Tugwell 2004Tugwell PS Wells GA Shainhouse JZ Equivalence studyof a topical diclofenac solution (pennsaid) compared withoral diclofenac in symptomatic treatment of osteoarthritisof the knee a randomized controlled trial Journal ofRheumatology 200431(10)2002ndash12
Zimmerman 1995
Zimmerman J Siguencia J Tsvang E Upper gastrointestinalhemorrhage associated with cutaneous application ofdiclofenac gel American Journal of Gastroenterology 199590(11)2032ndash4 [PUBMED 7485017]
References to other published versions of this review
Massey 2010
Massey T Derry S Moore RA McQuay HJ TopicalNSAIDs for acute pain in adults Cochrane Databaseof Systematic Reviews 2010 Issue 6 [DOI 10100214651858CD007402pub2]
lowast Indicates the major publication for the study
36Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
C H A R A C T E R I S T I C S O F S T U D I E S
Characteristics of included studies [ordered by study ID]
Airaksinen 1993
Methods R DB PC parallel groupsGel applied to the painful area twice daily for 7 daysAssessment at baseline 3 7 days
Participants Minor soft tissue injuries (lt 7 days)N = 56M 45 F 11Age not reportedMean baseline pain at rest 25-26 mm
Interventions Ketoprofen gel 2 x 5 g (125 mg) daily n = 29Placebo gel n = 27Rescue medication paracetamol 500 mgNo other treatment allowed
Outcomes PGE 5-point scale but reported as ldquoimprovedrdquo or ldquosame or worserdquo (responder = ldquoim-provedrdquo)Improvement in pain with movement 100 mm VAS reported as group meanAdverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Unclear risk No mention of early withdrawals ormethod of imputation
Size High risk lt 50 participants per treatment group
37Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Aoki 1984
Methods R DB PC AC parallel groupsGel applied to affected area 3 or 4 times daily with no occlusion for 7 daysAssessment at baseline 3 7 days
Participants Acute orthopaedic trauma (contusion distortion fracture lt 7 days)N = 252 (203 analysed for efficacy)M 98 F 105Age range 8 to 86 years 13 younger than 20 yearsBaseline pain mild in 35Exclusions 23 protocol violations 26 reasons ldquonot relatedrdquo to drug Equally distributedbetween groups
Interventions Piroxicam gel 05 1 g 3 to 4 x daily n = 84Indomethacin gel 1 1 g 3 to 4 x daily n = 84Placebo gel n = 84No other medication or initiation of physical therapy allowed
Outcomes PGE 5-point scale (responder = ldquobetterrdquo and ldquomuch betterrdquo)Adverse eventsWithdrawals and exclusions
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Low risk ldquokey code sealed until end of studyrdquo
Blinding (performance bias and detectionbias)All outcomes
Low risk Gels in ldquoidentical tubesrdquo
Incomplete outcome data (attrition bias)All outcomes
Unclear risk gt 10 withdrawals ldquounrelated to treat-mentrdquo and for ldquoprotocol violationsrdquo Nofurther details but no significant differ-ences between groups
Size Unclear risk 50 to 200 participants per treatment group
38Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Auclair 1989
Methods R DB PC parallel groupsGel massaged into skin over affected heel 3 times daily after cleaning with soap and waterfor up to 21 daysAssessment at baseline 7 21 days
Participants Acute Achilles heel tendinitis (not associated with continuous pain at rest or gt 1 monthhistory)N = 243 (227 analysed for efficacy)MF not reportedMean age 29 yearsBaseline pain ~ 10 had lt 26 mm on palpation of tendon ~ 30 had mild or no painon dorsiflexion of footExclusions failure to meet inclusion criteria major protocol violations failure to takestudy medication for full duration
Interventions Niflumic acid gel 25 3 x 5 g daily n = 117Placebo gel n = 110No other analgesics and anti-inflammatories physiotherapy or supportive measures al-lowed
Outcomes PGE 5-point scale (responder = ldquogoodrdquo or ldquovery goodrdquo)Pain improved or disappeared on dorsiflexionAdverse eventsWithdrawals and exclusions
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Unclear risk 10 excluded for ldquofailing to meet entry cri-teria and protocol violationsrdquo No furtherdetails
Size Unclear risk 50 to 200 participants per treatment group
39Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Billigmann 1996
Methods R DB PC parallel groupsGel applied 3 times daily with rubbingAssessed at baseline 3 5 7 days
Participants Distortion of ankle jointN = 160M and FAge 18+ yearsBaseline pain not reported
Interventions Ibuprofen microgel 5 3 x 10 cm (= 200 mg) daily n = 80Placebo gel n = 80
Outcomes Pain with movement VAS (responder = decreased by 20)Complete remissionAdverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Unclear risk Insufficient data to assess
Size Unclear risk 50 to 200 participants per treatment group
Campbell 1994
Methods R DB PC parallel groupsCream applied 4 times daily for 7 days (up to 14 days optional)Self assessed using daily diary for 7 days and up to 14 days
Participants Acute ankle sprain (lt 24 hours no fracture)N = 100 (51 analysed)M 33 F 18Mean age 29 years
40Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Campbell 1994 (Continued)
Baseline pain at rest gt 35 mm on walking 80 mmExclusions did not return diaries protocol exclusions (25 ibuprofen 24 placebo)
Interventions Ibuprofen cream 5 (Proflex) 4 x 4rdquo (10 cm) daily n = 26Placebo cream n = 25Advised to use rest and regular icing for 48 hours then walking and exerciseRescue medication paracetamol
Outcomes Improvement in walking ability 4-point scale (responder = ldquoimprovementrdquo)Pain on walking 100 mm VAS (mean data)Withdrawals and exclusions
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Low risk Randomisation carried out by sponsorTubes dispensed by hospital pharmacy whoheld the codes
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical creamrdquo
Incomplete outcome data (attrition bias)All outcomes
High risk gt 40 lost to follow-up Approximatelyequal between groups
Size High risk lt 50 participants per treatment arm as anal-ysed
Chatterjee 1977
Methods R DB PC parallel groupsCream applied to site of injury 3 times daily for 6 daysAssessment at baseline 2 6 days
Participants Soft tissue injuries (recent)N = 51MF not reportedAge not reportedBaseline pain on passive movement moderate or severe in all but 3 participants
41Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Chatterjee 1977 (Continued)
Interventions Benzydamine HCl cream 3 3 x daily n = 25Placebo cream n = 25(5 active 6 placebo participants also received ultrasound)No other topical agent allowed
Outcomes Pain on passive movement 4-point scale (responder = ldquoabsentrdquo or ldquoslightrdquo)Tenderness with pressure 4-point scale (responder = ldquoabsentrdquo or ldquoslightrdquo)Adverse eventsWithdrawals and exclusions
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk ldquopredetermined randomised schedulerdquo
Allocation concealment (selection bias) Low risk Sealed copy of schedule held by investiga-tor and duplicate copy kept by clinical trialco-ordinator Looked at only in event ofadverse reaction (not necessary)
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoindistinguishable in appearance andconsistencyrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk 2 lost to follow-up no withdrawals dueto adverse events Responder analysis
Size Unclear risk lt 50 participants per treatment arm
Costantino 2011
Methods R DB PC multicentre parallel groupPlaster applied daily for 7 daysPI assessment daily overall treatment efficacy and tolerability assessed at 3 and 7 days
Participants Grade I or II ankle sprain (lt 48 hours) with lateral external ligament involvementPI on movement ge 50100 oedema ge 20 mm difference between anklesN = 430M 249 F 175 (for analysis)Mean age 35 yearsBaseline PI on movement 72100 (SD 12)
42Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Costantino 2011 (Continued)
Interventions DHEPhep n = 142DHEP n = 146Placebo n = 142Rescue medication paracetamol to maximum 3 g dailyNo other treatments allowed
Outcomes Mean change in PI on movement from baseline to 3 daysMean reduction in oedema at 3 daysTolerability at 3 daysRescue medicationAdverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk Computer-generated
Allocation concealment (selection bias) Unclear risk ldquosealed envelopesrdquo
Blinding (performance bias and detectionbias)All outcomes
Low risk Shape colour size and application methodidentical for all plasters
Incomplete outcome data (attrition bias)All outcomes
Unclear risk Very few withdrawals but used LOCF
Size Unclear risk 50 to 200 participants per treatment arm
Coudreuse 2010
Methods R DB PC multicentre parallel groupPlaster applied daily for 7 daysPI assessed twice daily over 3 days then day 7 Overall treatment efficacy and tolerabilityassessed on days 3 7
Participants Ankle sprain (lt 48 hours) with lateral external ligament involvementPI on movement ge 50100 oedema ge 20 mm difference between anklesN = 240 (233 for analysis)M 148 F 86 6 unknown
Interventions DHEPhep n = 120Placebo n = 120Rescue medication paracetamol to maximum 4 g daily
43Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Coudreuse 2010 (Continued)
No other treatments allowed
Outcomes Global efficacy at 7 days 4-point scale (responder = ldquoexcellentrdquo or ldquogoodrdquo)Mean change in PI on movement at 6 hours and 7 daysMean change in oedema at 3 and 7 daysTolerabilityAdverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk Computer-generated
Allocation concealment (selection bias) Unclear risk ldquosealed envelopesrdquo
Blinding (performance bias and detectionbias)All outcomes
Low risk Appearance and odour identical for all plas-ters
Incomplete outcome data (attrition bias)All outcomes
Low risk ITT analysis lt 5 excluded for missingdata equal between groups
Size Unclear risk 50 to 200 participants per treatment arm
Curioni 1985
Methods R DB PC AC parallel groupsGel rubbed into affected area until absorbed twice daily for 10 daysAssessed at baseline and daily to 10 days
Participants Acute soft tissue injuriesN = 60M 33 F 27Median age 33 yearsBaseline pain not given
Interventions Ibuproxam gel 10 n = 20Ketoprofen gel n = 20Etofenamate gel n = 20
Outcomes PGE 4-point scale (ldquogoodrdquo or ldquoexcellentrdquo)Resolution of symptomsAdverse events
44Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Curioni 1985 (Continued)
Withdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk Medication supplied in identical tubes
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
Diebshlag 1990
Methods R DB PC AC parallel groupsGel applied 3 times daily without occlusion for 14 daysAssessment at baseline 2 3 4 8 15 days
Participants Ankle sprain (lt 24 hours)N = 37M 24 F 13Mean age 28 yearsBaseline pain slight to moderate
Interventions Ketorolac gel 2 3 x 3 g daily n = 13Etofenamate gel 5 3 x 3 g daily n = 12Placebo gel n = 12Rescue medication paracetamolNo other analgesic or anti-inflammatory medication ice packs or physiotherapy allowed
Outcomes Reduction in PI 100 mm VAS and 4-point scale (responder = ldquoimprovedrdquo)Adverse eventsWithdrawals and exclusions
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
45Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Diebshlag 1990 (Continued)
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Low risk ldquoMedication assignment supplied in asealed enveloperdquo Opened only if seriousparticipant event necessitation treatmentdisclosure occurred (not necessary)
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical appearancerdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
Dreiser 1988
Methods R DB PC parallel groupsCream applied 3 times dailyAssessment at baseline 7 days
Participants Acute tendinitis (lt 1 month)N = 64M 35 F 25Mean age 36 yearsBaseline spontaneous pain ge 60 mm
Interventions Ibuprofen cream 5 3 x 4 cm daily n = 32 (3 x 10 cm for large joints)Placebo cream n = 32No other topical systemic or physical treatment allowed
Outcomes PGE scale not reported (responder = ldquoimprovementrdquo or ldquocomplete relief rdquo)Improvement in pain VAS (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35Oxford Validity Score 1016
Risk of bias
Bias Authorsrsquo judgement Support for judgement
46Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Dreiser 1988 (Continued)
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Low risk Missing participants added back usingBOCF
Size High risk lt 50 participants per treatment arm
Dreiser 1989
Methods R DB PC parallel groupsGel applied twice daily to affected area with light massage then covered with standardcompressAssessed at baseline 3 7 days
Participants Uncomplicated recent ankle sprainN = 60M 36 F 24Mean age 33 yearsMean baseline pain 54 mm
Interventions Ketoprofen gel 25 2 x 5 cm daily n = 30Placebo gel n = 30No concomitant therapy other than simple oral analgesia allowed
Outcomes PGE 3-point scale (responder = ldquobetterrdquo)Improvement in pain VAS (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk ldquodrawing lotsrdquo
Allocation concealment (selection bias) Unclear risk Not described
47Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Dreiser 1989 (Continued)
Blinding (performance bias and detectionbias)All outcomes
Low risk Treatments ldquoidentical in every way exceptthat placebo did not contain active princi-plerdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participant per treatment arm
Dreiser 1990
Methods R DB PC parallel groupsGel lightly massaged into skin over affected area 3 times daily then covered with standardcompressAssessed at baseline 3 7 days
Participants Uncomplicated ankle sprain (lt 4 days)N = 60 (59 analysed)M 29 F 29 (not stated for 1 participant)Mean age 33 yearsBaseline pain ge moderately severeExclusions 1 participant had only moderate pain at baseline
Interventions Niflumic acid gel 25 3 x 5 g daily n = 30Placebo gel n = 30Concomitant treatment with systemic NSAIDs local therapies or physiotherapy werenot allowed
Outcomes PGE 4-point scale (responder = ldquocuredrdquo or ldquoimprovedrdquo)Improvement in pain VAS (mean data)Adverse eventsWithdrawals and exclusions
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
48Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Dreiser 1990 (Continued)
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
Dreiser 1994
Methods R DB PC parallel groupsPatch applied twice dailyAssessed at baseline 3 7 days
Participants Traumatic ankle sprain (lt 2 days)N = 131M 84 F 47Mean age 34 yearsBaseline pain ge 50 mm
Interventions Flurbiprofen patch 2 x 40 mg daily n = 65Placebo patch n = 66Rescue medication paracetamol Ice or light restraint allowedExclusions 1 from flurbiprofen group for protocol violation
Outcomes PGE 4-point scale (responder = ldquogoodrdquo or ldquovery goodrdquo)Improvement in pain VAS (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk Placebo patch was ldquonon-medicated (butotherwise identical)rdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
49Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Fioravanti 1999
Methods R DB AC parallel groupsGel lightly massaged into skin 3 times daily and kept dry for 6 to 8 hoursAssessed at baseline 3 10 days
Participants Peri and extra-articular inflammatory diseasesN = 100M 32 F 68Mean age 49 yearsBaseline spontaneous pain ge 40 mm
Interventions DHEP lecithin gel 3 x 5 g (= 65 mg) daily n = 50DHEP gel 3 x 5 g (= 65 mg) daily n = 50
Outcomes PGE 4-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo)Pain on movement meanAdverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size Unclear risk 50 participants per treatment arm
Fujimaki 1985
Methods R DB PC AC parallel groupsGel applied to affected area 3 or 4 times daily with no occlusion for up to 14 daysAssessed at baseline 7 14 days
Participants Muscle pain or inflammation in neck shoulder back chest and upper and lower ex-tremities or a combinationN = 271 (247 analysed)M 97 F 149
50Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Fujimaki 1985 (Continued)
Age lt 20 to 89 yearsBaseline pain mostly mild to moderateExclusions 24 due to protocol violations loss to follow-up
Interventions Piroxicam gel 05 1 g 3 to 4 x daily n = 92Indomethacin gel 1 1 g 3 to 4 x daily n = 90Placebo gel n = 89No concomitant oral or topical analgesic or anti-inflammatory medication allowed Nophysical therapy initiated after start of study
Outcomes PGE 5-point scale (responder = ldquobetterrdquo or ldquomuch betterrdquo)Physician rated improvement 5-point scale (responder = ldquomarked improvementrdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total =45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Low risk Cartons numbered randomly and numbersheld in a key code until study completion
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical tubesrdquo packed in numbered car-ton Gel bases slightly different in appear-ance so dispensing physician did not haveaccess to them
Incomplete outcome data (attrition bias)All outcomes
Unclear risk ~ 8 excluded from analysis for unknownreasons or lost to or inadequate follow-upApproximately equal between groups
Size Unclear risk 50 to 200 participants per treatment arm
Gallacchi 1990
Methods R DB AC parallel groupsGel applied to affected area 4 times daily with light massage for 14 daysAssessment at baseline 7 14 days
Participants Painful inflammatory conditionsN = 50M 20 F 30Mean age 50 years
51Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Gallacchi 1990 (Continued)
Baseline pain ge moderate severity
Interventions Diclofenac gel 1 2 g 4 x daily n = 25 (Flector)Diclofenac sodium 1 2 g 4 x daily n = 25 (Voltaren Emugel)No other medication that could interfere with test drugs allowed
Outcomes PGE 5-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo)Improvement in pain on pressure 4-point scale (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
Gonzaacutelez de Vega 2013
Methods R DB (for gels) SB (for ointment) AC multicentre parallel groups2 g (~ 6 cm of gel) applied over injured area x 3 daily for 14 daysAssessments at 0 4 7 14 days of treatment and day 42 follow-up
Participants Grade I or II ankle sprain (lt 24 hours) with lateral external ligament involvement PIon weight bearing ge 30100N = 449M = 308 F = 112 (for analysis)Mean age 28 years
Interventions Traumeel gel 3 x 2 g daily n = 140Traumeel ointment n = 143 (not analysed in this review)Diclofenac gel 1 3 x 2 g daily n = 137Rescue medication paracetamol to maximum 2 g daily
52Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Gonzaacutelez de Vega 2013 (Continued)
Outcomes Pain-free on day 7Global efficacy 5-point scale (responder = ldquogoodrdquo and ldquovery goodrdquo) on day 14Normal function on day 14 yes or no (responder = ldquoyesrdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 (gels) W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk Computer-generated
Allocation concealment (selection bias) Low risk Central allocation kits assigned in orderreceived and used envelopes (no furtherdetails)
Blinding (performance bias and detectionbias)All outcomes
Low risk For gel comparison ldquoidentical containersrdquo
Incomplete outcome data (attrition bias)All outcomes
Unclear risk Used LOCF Most withdrawals due to earlyrecovery (within 14 days) approximatelyequal between groups
Size Unclear risk 50 to 200 participants per treatment arm
Governali 1995
Methods R DB AC parallel groupsGel or cream applied 3 times daily for up to 14 daysAssessed at baseline 7 14 days
Participants Soft tissue injuries + 2 fracturesN = 30M = 21 F = 9Median age 38 yearsMean baseline pain on movement moderate to severe (28 scale 0 to 4)
Interventions Ketoprofen gel 5 3 x 2 to 3 g daily n = 15Ketoprofen cream 1 3 x 2 to 3 g daily n = 15
Outcomes PGE 5-point scale (responder = ldquogoodrdquo and ldquoexcellentrdquo)Adverse eventsWithdrawals
53Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Governali 1995 (Continued)
Notes Oxford Quality Score R1 DB1 W1 Total =35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Treatments were given in identical tubesand measurements made by blinded ob-servers but one was a cream and the othera gel
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
Gualdi 1987
Methods R DB AC parallel groupsGel applied twice daily for 10 daysAssessed at baseline 4 7 10 days
Participants Soft tissue injuriesN = 60M = 37 F = 23Mean age 32 years (range 13 to 78)Mean baseline pain on movement moderate to severe (22 scale 0 to 3)
Interventions Flunoxaprofen gel 2 x 3 to 5 cm daily n = 30Ketoprofen gel 2 x 3 to 5 cm daily n = 30
Outcomes Improvement in pain on pressure (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
54Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Gualdi 1987 (Continued)
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Unclear risk Insufficient data
Size High risk lt 50 participants per treatment arm
Haig 1986
Methods R DB PC parallel groupsCream applied lightly to affected area 6 times daily for 6 daysAssessed at baseline 2 4 6 days
Participants Soft tissue injuries (lt 24 hours)N = 43MF not reportedAge not reportedBaseline pain not reported
Interventions Benzydamine cream 3 6 x daily n = 21Placebo cream n = 22
Outcomes Pain on movement 4-point scale (responder = ldquoimprovedrdquo)Adverse events
Notes Oxford Quality Score R1 DB2 W0 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not reported
Allocation concealment (selection bias) Unclear risk Not reported
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquomatching placebordquo
55Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Haig 1986 (Continued)
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants apparently included inanalysis
Size High risk lt 50 participants per treatment arm
Hoffmann 2012
Methods R DB PC multicentre parallel groupPlaster applied x 1 daily (minimum of 20 consecutive hours of application per day) for14 daysPI assessed daily overall efficacy assessed at 7 and 14 days
Participants Unilateral mild to moderate muscle contusion of upper or lower limb (lt 72 hours) PIge 50100 superficial haematoma le 10 x 14 cm at injured siteN = 354M 126 F 228Mean age 39 years
Interventions DHEPhep plaster x 1 daily n = 121DHEP plaster x 1 daily n = 115Placebo plaster n = 118Rescue medication paracetamol 500 mg (no limit reported)
Outcomes PGE 5-point scale (responder = excellent or good) at 14 daysMean reduction in PI at 3 and 8 daysRescue medicationAdverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Method not reported
Allocation concealment (selection bias) Unclear risk Method not reported ldquoMedication packedand labelled in order to render all partici-pants and personnel fully blinded to treat-ment administeredrdquo
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoIndistinguishable regarding appearanceshape colour size and odourrdquo
56Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Hoffmann 2012 (Continued)
Incomplete outcome data (attrition bias)All outcomes
Unclear risk No responder analysis reported and impu-tation method not mentioned
Size Unclear risk 50 to 200 participants per treatment arm
Hofman 2000
Methods R DB AC parallel groupsGel applied to affected region 4 times daily with gentle massageAssessed at baseline 8 days in clinic and daily diary
Participants Soft tissue articular pain (le 15 days)N = 142M 19 F 123Mean age 57 yearsMean baseline PI moderate to severe
Interventions Diclofenac sodium gel 1 4 x 2 cm daily n = 69Lysine clonixinate gel 5 4 x 2 cm daily n = 73(2 cm = 225 mg)No other analgesic local treatment (including immobilisation bandaging) or acupunc-tureRescue mediation allowed after 2 applications if needed
Outcomes PGE 3-point scale (ldquogoodrdquo)PI participant diary (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk Diclofenac gel repackaged to maintain DBwith lysine clonixinate gel Minor differ-ences between gels only apparent when di-rectly compared
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
57Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Hofman 2000 (Continued)
Size Unclear risk 50 to 200 participants per treatment arm
Hosie 1993
Methods R DB AC parallel groupsFoam (approximately the size of a golf ball) applied and 1 tablet taken 3 times daily for7 days and up to 14 daysAssessed at baseline 7 14 (if necessary) days
Participants Acute lower back injury (lt 1 month)N = 287 (261 analysed for efficacy)M 151 F 136Mean age 37 years (range 18 to 63)Most participants had moderate to severe pain on movement one had noneExclusions 25 lost to follow up one assessed at 14 days but not 7 days
Interventions Felbinac foam 3 3 x 2 g daily + placebo tablets 3 x 1 daily n = 140 (127 analysed forefficacy)Ibuprofen tablets 3 x 400 mg daily + placebo foam 3 x 2 g daily n = 147 (134 analysedfor efficacy but one had no pain at baseline)No other oral injectable or topical analgesic or anti-inflammatory medication Ongoingphysiotherapy to continue without change
Outcomes Pain on movement 5-point scale (responder = ldquononerdquo or ldquomildrdquo)Spontaneous pain 5-point scale (responder = ldquononerdquo or ldquomildrdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquodouble dummyrdquo
Incomplete outcome data (attrition bias)All outcomes
Unclear risk Imputation method not mentioned
Size Unclear risk 50 to 200 participants per treatment arm
58Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Jenoure 1997
Methods R DB PC parallel groupsPlaster applied to skin over affected area twice daily and kept in place with an elasticbandageAssessed at baseline 7 14 days and after further 14 days without treatment
Participants Humero-radial epicondyl pain (tendinopathic) - nearly all tennis elbowN = 85M 54 F 31Mean age 45 yearsBaseline pain ldquomildrdquo in ~ 10 of placebo group and 29 of active group
Interventions DHEP plaster (Tissugel) x 2 daily n = 44Placebo plaster x 2 daily n = 41
Outcomes Pain on pressure 5-point scale (responder = ldquononerdquo or ldquomildrdquo)Spontaneous pain 5-point scale (responder = ldquono painrdquo)Adverse events
Notes Oxford Quality Score R1 DB2 W0 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not reported
Allocation concealment (selection bias) Unclear risk Not reported
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical characteristicsrdquo
Incomplete outcome data (attrition bias)All outcomes
Unclear risk No useable data
Size High risk lt 50 participants per treatment arm
Joussellin 2003
Methods R DB PC parallel groupsPlaster applied to skin over affected area once dailyAssessed at baseline 1 2 3 7 days
Participants Ankle sprain (lt 48 hours)N = 134M 72 F 62Age range 18 to 65 years
59Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Joussellin 2003 (Continued)
Baseline spontaneous pain ge 50 mm
Interventions DHEP plaster (Flector Tissugel 1) x 1 daily n = 68Placebo plaster x 1 daily n = 66Rescue medication paracetamol
Outcomes PGE 4-point scale (responder = ldquoexcellentrdquo)Pain on movement VAS (mean)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoIdentical in size appearance and usedsame formula as active patch without ac-tive ingredientrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in responder anal-ysis
Size Unclear risk 50 to 200 participants per treatment arm
Julien 1989
Methods R DB PC parallel groupsGel applied to affected area twice daily with light massageAssessed at baseline 3 7 days in clinic and daily diary
Participants TendinitisN = 60M 29 F 31Mean age 41 yearsBaseline pain gt 50 mm
Interventions Ketoprofen gel 25 2 x 5 cm (= 50 mg) daily n = 30Placebo gel 2 x 5 cm daily n = 30No concomitant therapy other than simple analgesia
60Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Julien 1989 (Continued)
Outcomes PGE 4-point scale (responder = ldquoimprovedrdquo or ldquorecoveredrdquo)Pain on movement 4-point scale (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB1 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk Random numbers table
Allocation concealment (selection bias) Unclear risk Randomisation code supplied by Menarinilaboratories remote from allocation
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in responder anal-ysis
Size High risk lt 50 participants per treatment arm
Klainguti 2010
Methods R DB PC multicentre parallel groupPlaster applied x 1 daily (minimum 20 consecutive hours of application daily) for 10daysPI assessed daily Overall treatment efficacy and tolerability assessed at days 3 and 7
Participants Unilateral mild to moderate muscle contusion or strain of upper or lower limb (lt 72hours) superficial haematoma le 140 cm2 PI ge 40100N = 185M 90 F 95Mean age 39 years
Interventions DHEPhep plaster x 1 daily n = 65DHEP plaster x 1 daily n = 61Placebo plaster x 1 daily n = 59
Outcomes Overall treatment efficacy at 3 days 5-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo)Resolution of haematoma at 10 days yes or noRescue medicationAdverse eventsWithdrawals
61Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Klainguti 2010 (Continued)
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk Computer-generated
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoIdentical in size appearance and odourrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk Withdrawals lt 5 All participants in-cluded in analysis
Size Unclear risk 50 to 200 participants per treatment arm
Kockelbergh 1985
Methods R DB PC parallel groupsGel applied twice dailyAssessed at baseline 3 7 days
Participants Acute soft tissue trauma (lt 24 hours)N = 74M 60 F 14Mean age 27 yearsBaseline pain gt 65 mm
Interventions Ketoprofen gel 25 2 x 5 cm (= 15 mg) daily n = 38Placebo gel 2 x 5 cm daily n = 36No concomitant treatmentRescue medication glafenine
Outcomes PGE 3-point scale (responder = ldquogoodrdquo)Spontaneous pain 100 mm VAS (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
62Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Kockelbergh 1985 (Continued)
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in responder anal-ysis
Size High risk lt 50 participants per treatment arm
Kuehl 2011
Methods R DB PC multicentre parallel groupsPlaster applied x 2 daily for 14 days or until resolutionPI assessed daily Overall treatment efficacy assessed at 14 days
Participants Mild or moderate sprain strain or contusion (lt 7 days) PI ge 510N = 418M 206 F 212Mean age 39 years
Interventions DHEP patch x 2 daily n = 207Placebo patch x 2 daily n = 211Rescue medication no analgesic allowed (or icewrapping) use = discontinuation
Outcomes Resolution of injury (VAS le 210)Overall tolerability 5-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk Computer-generated
Allocation concealment (selection bias) Unclear risk Not described
63Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Kuehl 2011 (Continued)
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical in appearance with same contentexcept diclofenacrdquo
Incomplete outcome data (attrition bias)All outcomes
High risk Used LOCF high withdrawal rate (43)Did not report all efficacy outcomes mea-sured
Size Low risk gt 200 participants per treatment arm
Li 2013
Methods R DB PC multicentre parallel groupsPlaster applied x 2 daily for 7 daysPI assessed daily overall treatment efficacy and tolerability assessed at 7 days
Participants Mild or moderate ankle or knee sprain muscle strain or contusion (lt 72 hours) PI ge
50100N = 384M 144 F 240Mean age 42 years
Interventions DHEP plaster x 2 daily n = 192Placebo plaster x 2 daily n = 192Rescue medication paracetamol to maximum 2 g daily
Outcomes ge 50 reduction in PI at 7 daysOverall treatment efficacy 5-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo)Overall tolerability 5-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo)Rescue medicationAdverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk Computer-generated
Allocation concealment (selection bias) Unclear risk Method not fully described ldquosealed en-velopesrdquo ldquoconcealed from investigators andparticipantsrdquo
64Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Li 2013 (Continued)
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical in texture size color and odorrdquo
Incomplete outcome data (attrition bias)All outcomes
Unclear risk Withdrawals 2 Methods states LOCFbut appears to describe BOCF for with-drawals
Size Unclear risk 50 to 200 participants per treatment arm
Linde 1985
Methods R DB PC parallel groupsCream applied x 3 daily for 5 days with elastic support for the first 3 daysAssessed at baseline 4 8 days
Participants Sprained ankle (lt 24 hours)N = 100M 58 F 42Mean age 28 yearsBaseline pain all participants had ldquowalking painrdquo
Interventions Benzydamine 3 cream x 3 daily n = 50Placebo gel x 3 daily n = 50
Outcomes Pain on movement responder = ldquofree of walking painrdquoAdverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35Paper describes a benzydamine cream and a placebo gel
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described Paper describes a benzy-damine cream and a placebo gel
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in responder anal-ysis
65Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Linde 1985 (Continued)
Size Unclear risk 50 participants per treatment arm
Machen 2002
Methods R DB PC parallel groupsGel gently (ldquominimal rubrdquo not vigorously) massaged into skin over affected site untilabsorbed 3 times daily until symptoms disappeared or for maximum of 7 daysAssessment at baseline and once daily using diary cards to 7 days
Participants Soft tissue injury (lt 2 weeks and untreated)N = 85 (81 analysed)M 42 F 39Mean age 41 yearsBaseline pain gt 50 mm4 placebo participants lost to follow-up
Interventions Ibuprofen gel 5 x 3 daily n = 40Placebo gel x 3 daily n = 41Initiation of other medication or physiotherapy not allowed during study
Outcomes PGE 5-point scale (responder = ldquomarked improvementrdquo or ldquocomplete clearancerdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk Gels had similar physical characteristicsand were supplied in identical tubes
Incomplete outcome data (attrition bias)All outcomes
Low risk lt 5 missing participants Remaining par-ticipants included in responder analysis
Size High risk lt 50 participants per treatment arm
66Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Mahler 2003
Methods R DB AC parallel groupsGel applied with gentle massage to affected area 3 times daily without occlusion for 10daysAssessed at baseline 3 10 days in clinic and daily patient diary
Participants First-degree ankle or knee sprains first-degree muscle strains and mild-to-moderatecontusionsn = 100M 69 F 31Mean age 32 yearsMean baseline pain with activity ge 65 mm
Interventions DHEP lethicin gel 3 x 5 g (= 65 mg) daily n = 52DHEP gel 3 x 5 g (= 65 mg) daily n = 48All participants treated with ice at site of inflammation for first 48 hours but no immo-bilisation allowedRescue medication paracetamol 500 mg if strictly necessary
Outcomes PGE 4-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo)Pain on movement 100-mm VAS (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 5
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk ldquoComputer-generated randomization listrdquo
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk Pharmaceutically inert colouring agentsadded to reference formulation so that gelswere indistinguishable
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in responder anal-ysis
Size High risk lt 50 participants in each treatment arm
67Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Maziegraveres 2005a
Methods R DB PC parallel groupsNew patch applied directly to skin over painful area each morningAssessed at baseline 3 7 14 days
Participants Symptomatic tendonitis in upper or lower limbs not requiring surgery (le 15 days)N = 172M 72 F 100Mean age 46 yearsBaseline pain with activity ge 40 mm
Interventions Ketoprofen patch 100 mg x 1 daily n = 87Placebo patch x 1 daily n = 85No analgesic or steroid by any route or other topical medication or physical therapyallowedRescue medication permitted but not within 12 hours of assessment
Outcomes PGE 4-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk ldquocomputer generated global randomizationcoderdquo
Allocation concealment (selection bias) Low risk ldquoThe randomization list and code en-velopes were prepared by the company ap-pointed for clinical supplies packaging Therandom code was disclosed only after studycompletion and database closurerdquo
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquothe same indistinguishable patch with noingredientrdquo
Incomplete outcome data (attrition bias)All outcomes
High risk LOCF withdrawals 12 All participantsincluded in analysis
Size Unclear risk 50 to 200 participants per treatment arm
68Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Maziegraveres 2005b
Methods R DB PC parallel groupsNew patch applied directly to skin over painful area each morningAssessed at baseline 3 7 14 days
Participants Painful benign ankle sprain (le 48 hours)N = 163M 83 F 80Mean age 37 yearsBaseline spontaneous pain ge 50 mm
Interventions Ketoprofen patch 100 mg x 1 daily n = 81Placebo patch x 1 daily n = 82No analgesic or steroid by any route or other topical medication or physical therapyallowedRescue medication permitted but not within 12 hours of assessment
Outcomes PGE 4-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk ldquocomputer-generated global randomiza-tion coderdquo
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk The same transdermal patch with no activeingredient
Incomplete outcome data (attrition bias)All outcomes
Unclear risk LOCF withdrawals 5 to 6 All partic-ipants included in analysis
Size Unclear risk 50 to 200 participants per treatment arm
McLatchie 1989
Methods R DB PC parallel groupsGel applied to injured site 3 times daily for 7 daysAssessment at baseline 4 7 days at clinic daily patient diary
Participants Acute soft tissue injury (lt 48 hours)N = 231
69Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
McLatchie 1989 (Continued)
M 143 F 88Mean age 33 yearsBaseline pain moderate to severe
Interventions Felbinac gel 3 3 x 3 cm daily n = 118Placebo gel 3 x 3 cm daily n = 113Rescue medication paracetamol
Outcomes Patient diary mean changeAdverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquotubes identical in all aspectsrdquo
Incomplete outcome data (attrition bias)All outcomes
Unclear risk No usable efficacy data
Size Unclear risk 50 to 200 participants per treatment arm
Morris 1991
Methods R DB PC parallel groupsGel applied to site of injury 3 times daily for 7 daysAssessed at baseline 7 days at clinic and daily patient diary
Participants Acute soft tissue injury (lt 3 days)N = 100 (84 analysed for efficacy)M 70 F 14Mean age 25 yearsBaseline pain moderate to severeExclusions 1 participant in placebo group lost to follow-up 15 protocol violations
Interventions Felbinac gel 3 3 x 1 cm daily n = 41Placebo gel n = 43Ice joint immobilisation bandaging and compression allowed
70Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Morris 1991 (Continued)
No concomitant oral NSAID occlusive dressing physiotherapy or liniments allowedRescue medication paracetamol
Outcomes PGE 5-point scale (responder = ldquogoodrdquo and ldquovery goodrdquo)Change in PI patient diary 10 cm VAS (mean data)Adverse eventsWithdrawals and exclusions
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Low risk ldquoRandomisation was undertaken at theproduction facility and a sealed copy of thelist supplied to the investigator for refer-ence only in defined circumstancesrdquo
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical tubes and outer boxesrdquo ldquoplacebowas a similarly constituted gelrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All relevant participants included in analy-sis
Size High risk lt 50 participants per treatment arm
NCT01255423
Methods R DB PC parallel group studyGel applied 4 x daily
Participants Acute lateral ankle sprain Grade I to II le 12 hoursN = 206Mean (plusmn SD) age 31 years (plusmn 13)M 87 F 119Baseline PI not reported
Interventions Diclofenac sodium gel 1 x 4 daily n = 104Placebo gel n = 102
Outcomes Mean PI on movementAdverse events
71Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT01255423 (Continued)
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Method of blinding not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants accounted for
Size Unclear risk 50 to 200 participants per treatment arm
NCT01272934
Methods R DB PC parallel group studyGel applied 4 x daily
Participants Acute lateral ankle sprain Grade I to II le 12 hoursN = 205M 101 F 104Mean (plusmn SD) age 32 years (plusmn 11)
Baseline PI not reported
Interventions Diclofenac sodium gel 1 x 4 daily n = 102Placebo gel x 4 daily n = 103
Outcomes Mean PI on movementAdverse events
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
72Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT01272934 (Continued)
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Method of blinding not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants accounted for
Size Unclear risk 50 to 200 participants per treatment arm
NCT01272947
Methods R DB PC parallel groupsGel applied 4 x daily
Participants Acute blunt soft tissue injuriescontusions of the limbs lt 3 hoursN = 204M 101 F 103Mean (plusmn SD) age 30 years (plusmn 11)
Baseline PI not reported
Interventions Diclofenac sodium gel 1 x 4 daily n = 104Placebo gel x 4 daily n = 100
Outcomes Mean PI on movementAdverse events
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Method of blinding not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants accounted for
73Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT01272947 (Continued)
Size Unclear risk 50 to 200 participants per treatment arm
Noret 1987
Methods R DB PC parallel groupsGel applied twice daily for 7 daysAssessment at baseline 3 8 days
Participants Minor sports injuries (lt 24 hours)N = 98 (93 analysed)M 71 F 27Mean age 29 yearsBaseline pain gt 60 mm
Interventions Ketoprofen gel 25 2 x 5 cm daily (= 15 mg) n = 48Placebo gel n = 45No other treatment given
Outcomes PGE 4-point scale (responder = ldquogoodrdquo and ldquoexcellentrdquo)Spontaneous pain 100 mm VAS (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Low risk ldquoallocated according to a randomization listand a corresponding code in a sealed enve-loperdquo
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
74Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Parrini 1992
Methods R DB PC parallel groupsFoam (the size of a walnut or a 1-second spray) applied with massage 3 x daily for 7days
Participants Articular trauma strains distortionsN = 169M 94 F 75Mean age 37 yearsMean baseline pain on movement 31 (scale 1 to 4)
Interventions Ketoprofen foam 15 3 x 2 g (= 600 mg) daily n = 83Placebo foam 3 x 2 g n = 86
Outcomes Pain on movement 4-point scale (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB1 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk ldquopatients were randomised according to themethod of random numbersrdquo [translated]
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Unclear risk No usable efficacy data
Size Unclear risk 50 to 200 participants per treatment arm
Picchio 1981
Methods R DB AC parallel groupsCream applied with slight massage until completely absorbed 3 times daily for up to 16daysAssessed at baseline 4 8 12 16 days
Participants Acute sports injuriesN = 40M 24 F 16Mean age 22 years (range 12 to 46)Most participants had mild to moderate baseline pain (12 and 9 with slight pain on
75Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Picchio 1981 (Continued)
movement)
Interventions Ibuprofen gel 10 3 x daily n = 20Ketoprofen gel 1 3 x daily n = 20
Outcomes Pain on movement (responder = ldquononerdquo)Adverse events
Notes Oxford Quality Score R1 DB2 W0 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk ldquotubes were identical in appearancerdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
Predel 2004
Methods R DB PC parallel groupsNew patch applied to injured area twice daily for 7 days Contact of patch with humidityor water to be avoidedAssessment at baseline 3 7 days
Participants Traumatic blunt soft tissue injuries (lt 3 hours no treatment)N = 120M 73 F 47Mean age 32 yearsBaseline pain gt 60 mm
Interventions Diclofenac sodium patch 2 x daily (140 mg per patch) n = 60Placebo patch 2 x daily n = 60NSAIDs analgesics psychotropic agents other topical preparations and bandages notallowed
Outcomes PGE 4-point scale (responder = rdquogoodldquo and rdquoexcellentldquo)Pain on movement 10 cm VAS (mean data)
76Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Predel 2004 (Continued)
Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk rdquocomputer generated block randomisationlist
Allocation concealment (selection bias) Low risk An independent statistician produced ran-domisation list and an independent con-tract research organisation packaged med-ication according to list Nobody else hadaccess to the randomisation list until thedatabase was closed
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoThe placebo patch was visually indistin-guishable from the active patchrdquo To avoidunblinding due to different smell anystudy nurse involved with medication wasnot involved in outcome assessment
Incomplete outcome data (attrition bias)All outcomes
Low risk Withdrawals lt 1 All participants in-cluded in analysis
Size Unclear risk 50 to 200 participants per treatment arm
Predel 2012
Methods R DB PC parallel groupsTreatment for 7 daysAssessed at baseline 1 3 5 and 8 days
Participants Grade I or II ankle sprain with lateral external ligament involvement lt 12 hoursPI on movement ge 50100N = 242M 152 F 90Mean age 32 years
Mean baseline pain on movement = 75100
Interventions Diclofenac gel (Voltaren Emulgel 232) 2 x 5 cm + 1 x 5 cm placebo gel daily n = 80Diclofenac gel (Voltaren Emulgel 232) 3 x 5 cm daily n = 80Placebo gel 3 x 5 cm daily n = 82Rescue medication paracetamol (to maximum 2 g daily)
77Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Predel 2012 (Continued)
No ice or bandages after randomisation
Outcomes ge 50 reduction in PI on movement at 5 daysPGE 5-point scale (responder = ldquogoodrdquo or rsquovery goodldquo)Patient satisfaction 5-point scale (responder = rdquogoodldquo or rsquovery goodrdquo or ldquoexcellentrdquo)Rescue medicationAdverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical in composition appearance tex-ture and smellrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk LOCF but withdrawals lt 3 in all treat-ment arms
Size Unclear risk 50 to 200 participants per treatment arm
Predel 2013a
Methods R DB PC parallel groupsTreatment for 14 daysAssessed at baseline 3 7 10 and 14 days (plusmn 1 day)
Participants Uncomplicated one-sided ankle sprain with swelling ge 12 mm (2 to 18 hours)PI on movement tenderness joint mobility (scale 0 to 3) summed as ge 5 and le 7N = 232M 126 F 106Mean age 29 years
Interventions Diclofenac 4 spray gel 3 x 4 to 5 sprays daily (96 to 120 mg diclofenac sodium) n =118Placebo spray gel 3 x 4 to 5 sprays daily n = 114Rescue medication paracetamol 500 mg (maximum 10 tablets per week)No ice or bandaging allowed
78Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Predel 2013a (Continued)
Outcomes None or slight PI on movement at 3 and 7 daysPGE 4-point scale (responder = ldquovery goodrdquo at 14 days)gt 50 reduction in swelling at 10 daysAdverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk ldquoplacebo (vehicle only no active ingredi-ent)rdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk LOCF for full or early cure Withdrawalslt 2
Size Unclear risk 50 to 200 participants per treatment arm
Predel 2013b
Methods R DB PC parallel groups2 g gel applied with fingertips over affected area and massaged in for 1 minute Treatmentfor 5 daysAssessed at baseline 2 3 5 days
Participants Uncomplicated neck pain originating from cervical joints and accompanying soft tissues(ge 12 hours but lt 3 months) PI ge 50100N = 72M 33 F 39Mean age 34 years
Interventions Diclofenac gel (Voltaren Emulgel) 4 x 2 g daily n = 36Placebo gel 4 x 2 g daily n = 36Rescue medication paracetamol up to 2 g dailyNo concomitant therapies allowed
Outcomes ge 50 decrease in PI on movement after 48 hoursPGE 5-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo at 5 daysAdverse eventsWithdrawals
79Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Predel 2013b (Continued)
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk Sponsor ldquoproduced randomisation listrdquo
Allocation concealment (selection bias) Low risk Automated remote system
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical in packaging labelling scheduleof administration appearance and odourrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk No missing data
Size High risk lt 50 participants per treatment arm
Ramesh 1983
Methods R DB PC parallel groupsCream applied to painful area and rubbed into skin over a large area for up to 10 daysAssessment at baseline 3 7 10 days
Participants Strains sprains contusions compressionsN = 80M 42 F 38Age 11 to 81 yearsBaseline pain 5 ibuprofen 2 placebo participants had no or slight pain
Interventions Ibuprofen cream 5 3 to 4 x 5 to 10 cm daily n = 40Placebo cream 3 to 4 x 5 to 10 cm daily n = 40Adjuvant therapy was not administered
Outcomes Pain on movement 4-point scale (responder = ldquononerdquo or ldquoslightrdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
80Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Ramesh 1983 (Continued)
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Low risk Randomisation key in sealed envelopeavailable for emergencies but opened onlyafter completion
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical appearance and odourrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
Rowbotham 2003
Methods R DB PC parallel groupsNew patches applied to the affected painful area for 12 consecutive hours twice dailyfor up to 14 daysAssessed at baseline 14 days in clinic and daily patient diary
Participants Minor sports injuries (sprains sprains contusions lt 72 hours)N = 372M 253 F 119Mean age 33 yearsBaseline pain at rest ge 510
Interventions DHEP patch (Flector Tissuegel) 2 x daily (equivalent to diclofenac sodium 140 mg perpatch) n = 191Placebo patch 2 x daily n = 181
Outcomes PGE 5-point scale (responder = ldquogoodrdquo and ldquoexcellentrdquo)Pain resolved lt moderate for 2 daysSpontaneous pain 10 cm VAS (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
81Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Rowbotham 2003 (Continued)
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk ldquoSystegraveme identiquerdquo without diclofenac
Incomplete outcome data (attrition bias)All outcomes
Unclear risk No information about imputation All par-ticipants included in analysis
Size Unclear risk 50 to 200 participants per treatment arm
Russell 1991
Methods R DB PC parallel groupsAffected area washed with soap and water and dried then gel applied and carefullyrubbed into skin 4 times daily for at least 7 daysAssessed at baseline 4 8 15 (if necessary) days at clinic and daily patient diary
Participants Acute soft tissue injuries (recent not recurrent)N = 214 (200 analysed)M 95 F 105Mean age 40 yearsBaseline pain gt 65 mm
Interventions Piroxicam gel 05 4 x 5 mg daily n = 100Placebo gel n = 100No other NSAIDs or analgesic drugs including liniments containing salicylates allowedAncillary therapy at the discretion of the investigator
Outcomes PGE 4-point scale (responder = ldquogoodrdquo and ldquoexcellentrdquo)Spontaneous pain mean reductionAdverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk ldquocomputer generated randomization coderdquo
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical base formulationrdquo
82Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Russell 1991 (Continued)
Incomplete outcome data (attrition bias)All outcomes
High risk High withdrawal rate (8 with piroxicam50 with placebo) No information aboutimputation
Size Unclear risk 50 to 100 participants per treatment group
Saillant 1998
Methods R DB PC parallel groupsTreatment for 7 daysAssessed at baseline 1 2 3 and 7 days
Participants Ankle sprain (lt 48 hours)N = 140M 72 F 62Age 18 to 65 yearsBaseline spontaneous pain ge 50 mm
Interventions (1) DHEP plaster (Flector Tissugel 1) 1 x daily n = 70(2) Placebo plaster 1 x daily n = 70Rescue medication paracetamol (to maximum 3 g daily)Ice allowed
Outcomes ge 30 decrease in PI at 7 daysPGE 4-point scale (responder = ldquoexcellentrdquo)No or low pain on passive stretchSingle foot leaning OK without painWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk Identical in size appearance and used sameformula as active patch without active in-gredient
Incomplete outcome data (attrition bias)All outcomes
Unclear risk Used LOCF for primary outcome With-drawals lt 7 and equal between groupsAll participants included in analysis
83Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Saillant 1998 (Continued)
Size Unclear risk 50 to 200 participants per treatment arm
Sanguinetti 1989
Methods R DB PC parallel groupsGel applied 3 times daily for 7 consecutive daysAssessment at baseline 7 days
Participants Soft tissue trauma (lt 48 hours)N = 82M = 47 F = 35Mean age 34 yearsBaseline pain moderate to severe
Interventions Felbinac gel 3 3 x daily n = 42Placebo gel 3 x daily n = 40No other NSAID steroid other topical application allowedRescue medication paracetamol felbinac is an active metabolite of the NSAID fenbufen
Outcomes PGE scale not reported (responder = ldquogoodrdquo and ldquovery goodrdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoindistinguishable in appearance colour orodourrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
84Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Sinniger 1981
Methods R DB PC parallel groupsCream applied 2 to 3 times daily with gentle massage or if massage not possible (toopainful) with protective dressingAssessment at baseline 5 10 days
Participants Minor soft tissue injuriesN = 20M 11 F 9Mean age 40 yearsBaseline pain not reported
Interventions Fentiazac cream 5 2 to 3 x daily n = 10Placebo cream 2 to 3 x daily n = 10All participants told to restNo other local and systemic treatments allowedRescue medication analgesic if actually needed
Outcomes Pain relief scale not reported (responder = total pain relief ) improvement in pain on movement pain scale not reported (mean data)Adverse events
Notes Oxford Quality Score R1 DB1 W0 Total = 25
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
85Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Spacca 2005
Methods R DB PC parallel groupsGel applied 3 times daily with gentle massage until complete absorption for up to 10daysAssessment at baseline 10 days in clinic and daily patient diary
Participants Shoulder periarthritis or lateral epicondylitis (lt 5 days)N = 155M 74 F 81Mean age 51 yearsBaseline pain with activity gt 70 mm
Interventions DHEP lecithin gel (Effigel) 3 x 5 g daily n = 79Placebo gel 3 x 5 g daily n = 76Rescue medication (paracetamol) allowed if pain unbearableNo other analgesic or anti-inflammatory drug allowed
Outcomes Improvement in pain 100 mm VAS (mean data)Adverse events
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Unclear risk No usable efficacy data
Size Unclear risk 50 to 200 participants per treatment arm
Sugioka 1984
Methods R DB AC parallel groupsGel applied to affected area 3 to 4 times daily without occlusion for 14 daysAssessed at baseline 7 14 days
Participants Non-traumatic diseases of muscle or tendonN = 366 (340 analysed for efficacy)M 115 F 202 (completers)
86Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Sugioka 1984 (Continued)
Age range 12 to 84 years (most 30 to 70)Baseline pain on movement ldquononerdquo or ldquomildrdquo in about 13 of participantsExclusions for protocol violations 8 piroxicam 18 indomethacin
Interventions Piroxicam gel 05 3 to 4 x 1 g daily n = 183Indomethacin gel 1 3 to 4 x 1 g daily n = 183No concomitant anti-inflammatory or analgesic drug including steroids or initiationof physical therapy allowed
Outcomes PGE 5-point scale (responder = ldquobetterrdquo or ldquomuch betterrdquo)Pain on movement 4-point scale (responder = ldquoreducedrdquo or ldquodisappearedrdquo)
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Low risk Key code sealed and retained until end ofstudy
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoboth packages were of the same appear-ance and indistinguishablerdquo and investiga-tors did not see contents
Incomplete outcome data (attrition bias)All outcomes
Low risk All relevant participants included in analy-sis
Size Unclear risk 50 to 200 participants per treatment arm
Thorling 1990
Methods R DB PC parallel groupsParticipants given specific instructions on how to apply gel (not reported) to affectedarea 2 to 6 times daily as requiredAssessment at baseline 3 7 days in clinic
Participants Soft tissue injuries (lt 48 hours)N = 120M 85 F 35Mean age 27 yearsBaseline pain moderate to severe
87Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Thorling 1990 (Continued)
Interventions Naproxen gel 10 2 to 6 x daily n = 60Placebo gel 2 to 6 x daily n = 60Rescue medication paracetamol 500 mg
Outcomes PGE 5-point scale (responder = ldquogoodrdquo and ldquovery goodrdquo)Pain on passive movement 4-point scale (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquosupplied in unmarked tubesrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size Unclear risk 50 to 200 participants per treatment arm
Tonutti 1994
Methods R DB AC parallel groupsGel applied 3 times daily for 2 to 3 weeksAssessed at baseline and intervals of 7 days
Participants Muscle or joint traumaN = 30M 20 F 10Mean age 34 years1 participant had injury of mild severity Mean baseline pain on active movement 28(scale 0 to 4)
Interventions Ketoprofen gel 5 3 x 2 to 3 g daily n = 15Etofenamate gel 5 3 x 2 to 3 g n = 15No concomitant treatment with NSAID aspirin steroid or physical therapy
88Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Tonutti 1994 (Continued)
Outcomes PGE 4-point scale (responder = ldquogoodrdquo and ldquoexcellentrdquo)Pain on movement 5-point scale (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not reported
Allocation concealment (selection bias) Unclear risk Not reported
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquothe two drugs were packed in indistin-guishable tubesrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
Vecchiet 1991
Methods R DB PC parallel groupsGel applied to the skin on and around painful area and gently rubbed in until absorbedtwice daily for up to 10 daysAssessed at baseline 5 10 days
Participants Soft tissue trauma (minor sports injuries)N = 60M 60Mean age 25 yearsMean baseline pain on active movement moderate
Interventions Meclofenamic acid gel 5 2 x 10 cm daily (= 4 g) n = 30Placebo gel 2 x 10 cm daily n = 30Both groups treated with ice rest and bandage for first 48 hours before starting testtreatmentRescue medication paracetamol
Outcomes PGE 4-point scale (responder = ldquogoodrdquo and ldquoexcellentrdquo)Pain on movement 4-point scale (mean data)Withdrawals
89Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Vecchiet 1991 (Continued)
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
Whitefield 2002
Methods R DB (double dummy) AC parallel groupsGel applied to affected site with gentle massage and 1 tablet taken 3 times daily for atleast 7 daysAssessed at baseline 7 14 (if necessary) days in clinic and daily patient diary
Participants Soft tissue injuries (lt 24 hours)N = 100M 95 F 5Mean age 26 years (range 18 to 50)Mean baseline pain on movement 22 cm
Interventions Ibuprofen gel 5 + placebo tablet 3 x daily n = 50Ibuprofen 400 mg tablet + placebo gel 3 x daily n = 50No other medication or physical therapy was prescribed and no other analgesics wereallowed
Outcomes PGE 3-point scale (responder = ldquoexcellentrdquo)Change in condition of injury site 5-point scale (responder = ldquocompletely betterrdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
90Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Whitefield 2002 (Continued)
Random sequence generation (selectionbias)
Unclear risk Not reported
Allocation concealment (selection bias) Unclear risk Not reported
Blinding (performance bias and detectionbias)All outcomes
Low risk Placebo tablets Identical in appearance toactive tablets Active and placebo gels hadsimilar physical characteristics and weresupplied in identical tubes
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size Unclear risk 50 participants per treatment arm
kermark 1990
Methods R DB (double dummy) PC AC parallel groupsSpray applied to affected area and capsules taken 3 times daily for 2 weeksAssessment at baseline 3 or 4 7 and 14 days
Participants Superficial overuse sports injuries (symptom onset 74 weeks)N = 70M 44 F 18 (completers)Mean age 30 yearsBaseline pain on palpation mostly slight to moderate
Interventions Elmetacin spray (indomethacin 1) 3 to 5 x 05 to 15 ml daily + placebo capsules n= 23Indomethacin capsules 3 x 25 mg daily + placebo spray n = 23Placebo spray and capsules n = 24Rescue medication paracetamol
Outcomes No pain on palpation (= responder)Participant improvement 100 mm VAS (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk ldquorandom number coderdquo
91Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
kermark 1990 (Continued)
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical in appearancerdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk Responder analysis lt 5 withdrawals
Size High risk lt 50 participants per treatment group
AC active control BOCF baseline observation carried forward DB double-blind DHEP diclofenac hydroxyethylpyrrolidine ordiclofenac epolamine F female HCl hydrochloride hep heparin LOCF last observation carried forward M male N numberof participants in study n number of participants in treatment arm PC placebo control PGE Participant Global Evaluation PIpain intensity R randomised VAS visual analogue scale W withdrawals
Characteristics of excluded studies [ordered by study ID]
Study Reason for exclusion
Ambrus 1987 No usable dichotomous data
Anon 1993 Not double-blind
Ascherl 1982 No usable dichotomous data
Bagliani 1976 Not an RCT
Baracchi 1982 No usable data
Burnham 1998 lt10 participantstreatment arm in first period of crossover study
Boumlhmer 1995 Active control invalid
Cesarone 2008 Not an RCT
Coulibaly 2009 Not double-blind
Diebschlag 1985 No usable dichotomous data
Diebschlag 1986 Inappropriate randomisation
Diebschlag 1992 No usable dichotomous data
92Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Fantato 1971 No usable dichotomous data
Galer 2000 No usable data
Hallmeier 1986 Not double-blind
Hallmeier 1988 Not double-blind
Kaneko 1999 Inappropriate randomisation - quasi-randomised
Kockelbergh 1985b Treatment not applied daily
Kuwabara 2013 Used NSAID-lidocaine combination (conference abstract)
Lee 1991 Not an RCT
Link 1996 No usable dichotomous data
May 2007 No usable dichotomous data
Oakland 1993 Inappropriate comparator
Odaglia 1987 Not an RCT
Picardi 1993 Not an RCT
Taboada 1992 Dose and duration of treatment unclear
Vanderstraeten 1990 Not double-blind
Vinciguerra 2008 Not an RCT
Von Klug 1977 Chronic and acute outcomes combined
RCT randomised controlled trial
Characteristics of studies awaiting assessment [ordered by study ID]
NCT00351104
Methods R DB PC parallel groups
Participants Ankle sprain or strain Grade I or II Age ge 18 yearsN = 220
93Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT00351104 (Continued)
Interventions Ketoprofen patch 20 x 1 dailyPlacebo patch x 1 dailyTreatment for 21 days
Outcomes Assessments days 3 7 21PI (0-10) at rest and during activitiesPGEFunctionRescue medication
Notes Sponsor Endo PharmaceuticalsEstimated completion February 2007 No results
NCT00352625
Methods R DB PC parallel groups
Participants Shoulder elbow or knee tendonitis or bursitis Age ge 18 yearsN = 330
Interventions Ketoprofen patch 20 x 1 dailyPlacebo patch x 1 dailyTreatment for 21 days
Outcomes Assessments days 3 7 21PI (0-10) at rest and during activitiesPGEFunctionRescue medication
Notes Sponsor Endo PharmaceuticalsEstimated completion February 2007 No results
NCT00426985
Methods R DB PC parallel groups
Participants Shoulder elbow or knee tendonitis or bursitis Age ge 18 years
Interventions Ketoprofen patch 20 x 1 dailyPlacebo patchTreatment for 21 days
Outcomes PI during activityPI at restPGERescue medication
94Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT00426985 (Continued)
Safety
Notes Study terminated May 2008 but ldquosufficient number of subjects accrued to conduct analysisrdquo No results
NCT00640705
Methods R DB PC parallel groups
Participants Ankle sprain or strain Grade I or II le 48 hours PI 510 to 910 Age 18 to 75 yearsN = 170
Interventions Diclofenac sodium patch (15 mg) x 1 dailyPlaceboTreatment for 7 days
Outcomes EfficacySafety
Notes Study terminated July 2008 (Sponsor decision - no further details) No results posted
NCT00640939
Methods R DB PC parallel groups
Participants Shoulder elbow or knee tendonitis or bursitis Age 18 to 75 yearsBaseline pain mild to moderate but states 510 to 910N = 308
Interventions Diclofenac sodium patch 1 (15 mg) x 1 dailyPlacebo patch x 1 dailyDuration of treatment not reported probably 14 days
Outcomes EfficacySafety
Notes Sponsor Cerimon PharmaceuticalsEstimated completion April 2008 No results
NCT00680472
Methods R DB PC parallel groups
Participants Acute unilateral shoulder pain requiring treatment for ge 2 weeks Age ge 18 yearsN = 368
95Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT00680472 (Continued)
Interventions Ketoprofen patch (HKT-500) x 1 dailyPlacebo x 1 dailyTreatment for 14 days
Outcomes PainSafety
Notes Sponsor Hisamitsu Pharmaceutical Co IncEstimated completion October 2008 No results
NCT00680784
Methods R DB PC parallel groups
Participants Acute benign ankle sprain Grade I-II le 48 hours Age ge 18 yearsN = 260
Interventions Ketoprofen patch (HKT-500) x 1 dailyPlacebo patch x 1 dailyTreatment for 14 days
Outcomes PainSafety
Notes Sponsor Hisamitsu Pharmaceutical Co IncEstimated completion November 2008 No results
NCT00765700
Methods R DB PC parallel groups
Participants Acute sprain or strain of upper and lower extremities mild to moderate injuries le 72 hours Age 18 to 70 yearsN = 364
Interventions Ketoprofen cream 10 1 g x 3 dailyPlacebo cream x 3 dailyTreatment for 7 days
Outcomes PainSafety
Notes Sponsor Imprimis Pharmaceuticals IncEstimated completion September 2009 No results
96Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT00869063
Methods R DB PC parallel groups
Participants Wrist sprain strain or contusion (mild to moderate) ldquorecentrdquo Age 17 to 75 yearsN = 214
Interventions Diclofenac sodium patch 1 x 1 dailyPlacebo patch x 1 dailyTreatment for 7 days
Outcomes Change in PI during activity at 3 and 7 daysSafety
Notes Sponsor Cerimon PharmaceuticalsEstimated completion September 2009 No results
NCT00869180
Methods R DB PC parallel groups
Participants Unilateral ankle sprain (mild or moderate) ldquorecentrdquo Age 17 to 75 yearsN = 219
Interventions Diclofenac sodium patch x 1 dailyPlacebo patch x 1 dailyTreatment for 7 days
Outcomes Change in PI during activity at 3 and 7 days
Notes Sponsor Cerimon PharmaceuticalsEstimated completion August 2009 No results
NCT00931866
Methods R DB PC parallel groups
Participants Unilateral soft tissue injury between mid-bicep to wrist or mid-thigh to ankle (mild to moderate) ldquorecentrdquo Age 18to 75 yearsN = 407
Interventions Diclofenac sodium patch x 1 dailyPlacebo patch x 1 dailyTreatment for 14 days
Outcomes Change in PI during activity at 7 and 14 daysPIPGERescue medicationSafety
97Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT00931866 (Continued)
Notes Sponsor Cerimon PharmaceuticalsEstimated completion December 2009 No results
NCT01874626
Methods R DB PC parallel groups
Participants Acute ankle sprain Grade I or II le 24 hours Age ge 16 years Baseline PI ge 510N = 305
Interventions Ibuprofen cream 200 mg in 27 g cream x 4 dailyPlacebo cream x 4 dailyTreatment for 7 days
Outcomes PI on movement daily to 7 daysPGE 7 daysSystemic and local adverse events
Notes Estimated completion January 2014 No results
NCT01957215
Methods R DB PC parallel groups
Participants Acute lateral ankle sprain Grade I-II le 24 hours PI ge 510 Age 18 to 65 yearsN = 270
Interventions Indomethacin patch x 2 dailyPlacebo patch x 2 dailyDuration of treatment not reported probably 7 days
Outcomes PI on movement dailyPR dailyPGERescue medication
Notes Estimated completion September 2014 No results
NCT02324270
Methods R DB PC and AC parallel groups
Participants Uncomplicated acute ankle sprain Grade I or II lt 48 hours PI gt 50100 Age 18 to 65 yearsN = 658
98Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT02324270 (Continued)
Interventions Diclofenac epolamine (Flector) 13 patchGeneric diclofenac epolamine 13 patchPlacebo patchDuration of treatment not reported
Outcomes Bioequivalence studyChange from baseline in PI (VAS) at 3 daysApplication site reactions
Notes Estimated completion December 2014 No results
Sarzi-Puttini 2014
Methods R DB AC parallel group non-inferiority studyDuration 7 days
Participants Acute musculoskeletal injury (mainly muscular joint tendon)N = 697M 271 F 426Mean age 52 years
Interventions Ketoprofen (SKP-021) patch ketoprofen 30 mgDiclofenac (Voltadola) patch diclofenac sodium 140 mgPatches applied twice daily for 7 days
Outcomes ge 50 reduction in PI from baseline to end of study (100 mm VAS)Patient overall ratingClinical symptoms (4-point scale)Time to responseAdverse events skin reactionsSerious adverse events
Notes ldquoThe analysis of the data of this trial showed that the two formulations were equally effective and well tolerated inthe treatment of acute musculoskeletal injuriesrdquo
AC active-controlled DB double-blind F female M male N number of participants in study PC placebo-controlled PGE PatientGlobal Evaluation of treatment PI pain intensity R randomised VAS visual analogue scale
99Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Characteristics of ongoing studies [ordered by study ID]
NCT01945034
Trial name or title Placebo-controlled double-blind evaluation of the efficacy and safety of ibuprofen 5 topical gel for thetreatment of ankle sprain
Methods R DB PC parallel groups
Participants Ankle sprain Grade I or II age ge 12 yearsEstimated enrolment 280
Interventions Ibuprofen gel 5 x 2 dailyIbuprofen gel 5 x 3 dailyPlacebo gelTreatment for 7 days with additional 3 days as needed
Outcomes PI on weight bearing and rest at 7 daysPGE at 10 daysRescue medicationSafety
Starting date November 2013
Contact information Pfizer
Notes Estimated completion February 2015Includes participants aged ge 12 years
NCT02100670
Trial name or title A clinical study to assess the efficacy and onset of pain relief of topical MFC51123 diclofenac-menthol gelversus controls in ankle sprain
Methods R DB PC and AC parallel groups
Participants Acute lateral ankle sprain Grade I or II le 24 hours PI ge 510 Age 16 to 65 yearsEstimated enrolment 400
Interventions Diclofenac sodium 1 + methanol 3 gelDiclofenac sodium 1 + methanol 009 gelMethanol 3 gelPlacebo + 009 methanol gelTreatment for 10 days with 4 g gel x 4 daily
Outcomes PRPID on movement (0 to 10 days)PGETime to complete recoveryAdverse events
100Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT02100670 (Continued)
Starting date November 2013
Contact information GSKClinicalSupportHDgskcom
Notes Estimated completion November 2014
NCT02290821
Trial name or title A randomised double-blind placebo-controlled parallel group study to evaluate the efficacy and safety ofdiclofenac sodium topical gel (DSG) 1 applied four times daily in subjects with acute blunt soft tissueinjuriescontusions of the limbs
Methods R DB PC parallel groups
Participants Acute blunt soft tissue injuries or contusions of the limbs le 6 hours (ldquofreshrdquo) Age ge 16 yearsEstimated enrolment 200
Interventions Diclofenac sodium gel 1 x 4 dailyPlacebo gelDuration of treatment not reported
Outcomes PainSafety
Starting date December 2014
Contact information Novartis
Notes Estimated completion August 2015
AC active control DB double-blind PC placebo-controlled PGE Patient Global Evaluation of treatment PI pain intensity PIDpain intensity difference PR pain relief R randomised
101Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
D A T A A N D A N A L Y S E S
Comparison 1 Individual NSAID versus placebo
Outcome or subgroup titleNo ofstudies
No ofparticipants Statistical method Effect size
1 Clinical success 29 Risk Ratio (M-H Fixed 95 CI) Subtotals only11 Diclofenac 10 2050 Risk Ratio (M-H Fixed 95 CI) 160 [149 172]12 Ibuprofen 5 436 Risk Ratio (M-H Fixed 95 CI) 164 [133 201]13 Ketoprofen 7 683 Risk Ratio (M-H Fixed 95 CI) 156 [137 177]14 Piroxicam 3 504 Risk Ratio (M-H Fixed 95 CI) 148 [127 173]15 Indomethacin 3 341 Risk Ratio (M-H Fixed 95 CI) 126 [103 155]16 Benzydamine 3 193 Risk Ratio (M-H Fixed 95 CI) 115 [096 138]
2 Local adverse events 33 Risk Ratio (M-H Fixed 95 CI) Subtotals only21 Diclofenac 15 3271 Risk Ratio (M-H Fixed 95 CI) 078 [056 110]22 Ibuprofen 3 321 Risk Ratio (M-H Fixed 95 CI) 230 [098 543]23 Ketoprofen 8 852 Risk Ratio (M-H Fixed 95 CI) 119 [083 170]24 Piroxicam 3 522 Risk Ratio (M-H Fixed 95 CI) 042 [017 108]25 Felbinac 3 397 Risk Ratio (M-H Fixed 95 CI) 191 [049 750]26 Indomethacin 3 354 Risk Ratio (M-H Fixed 95 CI) 265 [091 773]
Comparison 2 Diclofenac versus placebo (effect of formulation)
Outcome or subgroup titleNo ofstudies
No ofparticipants Statistical method Effect size
1 Clinical success 10 2050 Risk Ratio (M-H Fixed 95 CI) 160 [149 172]11 Plaster - Flector 4 1030 Risk Ratio (M-H Fixed 95 CI) 153 [136 171]12 Plaster - other 3 474 Risk Ratio (M-H Fixed 95 CI) 155 [137 175]13 Gel - Emulgel 2 314 Risk Ratio (M-H Fixed 95 CI) 384 [268 550]14 Gel - other 1 232 Risk Ratio (M-H Fixed 95 CI) 115 [105 127]
Comparison 3 Ibuprofen versus placebo (effect of formulation)
Outcome or subgroup titleNo ofstudies
No ofparticipants Statistical method Effect size
1 Clinical success 5 436 Risk Ratio (M-H Fixed 95 CI) 164 [133 201]11 Cream 3 195 Risk Ratio (M-H Fixed 95 CI) 128 [103 159]12 Gel 2 241 Risk Ratio (M-H Fixed 95 CI) 266 [169 421]
102Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Comparison 4 Ketoprofen versus placebo
Outcome or subgroup titleNo ofstudies
No ofparticipants Statistical method Effect size
1 Clinical success 7 683 Risk Ratio (M-H Fixed 95 CI) 156 [137 177]11 Plaster 2 335 Risk Ratio (M-H Fixed 95 CI) 121 [104 140]12 Gel 5 348 Risk Ratio (M-H Fixed 95 CI) 219 [174 275]
Comparison 5 All topical NSAIDs versus placebo
Outcome or subgroup titleNo ofstudies
No ofparticipants Statistical method Effect size
1 Local adverse events 42 6740 Risk Ratio (M-H Fixed 95 CI) 098 [080 121]2 Systemic adverse events 36 5576 Risk Ratio (M-H Fixed 95 CI) 096 [073 126]3 Adverse event withdrawals 42 6405 Risk Ratio (M-H Fixed 95 CI) 101 [064 159]
Comparison 6 Topical NSAID versus active comparator
Outcome or subgroup titleNo ofstudies
No ofparticipants Statistical method Effect size
1 Clinical success - topicalpiroxicam vs topicalindomethacin
3 641 Risk Ratio (M-H Fixed 95 CI) 124 [107 144]
2 Local adverse events -topical piroxicam vs topicalindomethacin
3 671 Risk Ratio (M-H Fixed 95 CI) 021 [009 047]
103Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Analysis 11 Comparison 1 Individual NSAID versus placebo Outcome 1 Clinical success
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 1 Individual NSAID versus placebo
Outcome 1 Clinical success
Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
1 Diclofenac
Coudreuse 2010 99117 82116 175 120 [ 104 138 ]
Joussellin 2003 3668 2466 52 146 [ 099 215 ]
Klainguti 2010 5662 4659 100 116 [ 099 136 ]
Li 2013 173192 114192 242 152 [ 134 172 ]
Predel 2004 5560 560 11 1100 [ 474 2555 ]
Predel 2012 (1) 5980 941 25 336 [ 186 607 ]
Predel 2012 (2) 5780 841 22 365 [ 193 690 ]
Predel 2013a 111118 93114 201 115 [ 105 127 ]
Predel 2013b 3636 736 16 487 [ 257 923 ]
Rowbotham 2003 75191 48181 104 148 [ 110 200 ]
Saillant 1998 4370 2570 53 172 [ 119 248 ]
Subtotal (95 CI) 1074 976 1000 160 [ 149 172 ]Total events 800 (NSAID) 461 (Placebo)
Heterogeneity Chi2 = 12176 df = 10 (Plt000001) I2 =92
Test for overall effect Z = 1278 (P lt 000001)
2 Ibuprofen
Billigmann 1996 2580 1080 136 250 [ 129 486 ]
Campbell 1994 2126 1925 264 106 [ 080 142 ]
Dreiser 1988 2632 1232 164 217 [ 134 349 ]
Machen 2002 2540 941 121 285 [ 152 532 ]
Ramesh 1983 2340 2340 314 100 [ 069 146 ]
Subtotal (95 CI) 218 218 1000 164 [ 133 201 ]Total events 120 (NSAID) 73 (Placebo)
Heterogeneity Chi2 = 2102 df = 4 (P = 000031) I2 =81
Test for overall effect Z = 465 (P lt 000001)
3 Ketoprofen
Airaksinen 1993 2429 1427 91 160 [ 107 238 ]
Dreiser 1989 1830 530 32 360 [ 154 844 ]
02 05 1 2 5
Favours placebo Favours NSAID
(Continued )
104Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
( Continued)Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Julien 1989 1830 630 38 300 [ 138 650 ]
Kockelbergh 1985 3038 2236 142 129 [ 095 176 ]
Mazi res 2005a 7281 6082 376 121 [ 104 141 ]
Mazi res 2005b 5087 4185 262 119 [ 090 158 ]
Noret 1987 3951 947 59 399 [ 218 733 ]
Subtotal (95 CI) 346 337 1000 156 [ 137 177 ]Total events 251 (NSAID) 157 (Placebo)
Heterogeneity Chi2 = 3104 df = 6 (P = 000002) I2 =81
Test for overall effect Z = 683 (P lt 000001)
4 Piroxicam
Aoki 1984 5672 3367 286 158 [ 120 207 ]
Fujimaki 1985 4483 4082 337 109 [ 080 147 ]
Russell 1991 79100 45100 377 176 [ 138 223 ]
Subtotal (95 CI) 255 249 1000 148 [ 127 173 ]Total events 179 (NSAID) 118 (Placebo)
Heterogeneity Chi2 = 624 df = 2 (P = 004) I2 =68
Test for overall effect Z = 499 (P lt 000001)
5 Indomethacin
Aoki 1984 4164 3367 413 130 [ 096 176 ]
Fujimaki 1985 4482 4082 513 110 [ 082 148 ]
kermark 1990 1222 624 74 218 [ 099 481 ]
Subtotal (95 CI) 168 173 1000 126 [ 103 155 ]Total events 97 (NSAID) 79 (Placebo)
Heterogeneity Chi2 = 269 df = 2 (P = 026) I2 =26
Test for overall effect Z = 221 (P = 0027)
6 Benzydamine
Chatterjee 1977 2125 1225 185 175 [ 112 272 ]
Haig 1986 1821 1322 196 145 [ 098 214 ]
Linde 1985 3550 4050 618 088 [ 070 110 ]
Subtotal (95 CI) 96 97 1000 115 [ 096 138 ]Total events 74 (NSAID) 65 (Placebo)
Heterogeneity Chi2 = 1033 df = 2 (P = 001) I2 =81
Test for overall effect Z = 152 (P = 013)
02 05 1 2 5
Favours placebo Favours NSAID
(1) Three times daily application
(2) Twice daily application
105Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Analysis 12 Comparison 1 Individual NSAID versus placebo Outcome 2 Local adverse events
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 1 Individual NSAID versus placebo
Outcome 2 Local adverse events
Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
1 Diclofenac
Coudreuse 2010 1117 7116 102 014 [ 002 113 ]
Jenoure 1997 144 141 15 093 [ 006 1442 ]
Joussellin 2003 168 366 44 032 [ 003 303 ]
Klainguti 2010 0126 159 30 016 [ 001 381 ]
Kuehl 2011 16207 12211 173 136 [ 066 280 ]
Li 2013 4192 3192 44 133 [ 030 588 ]
NCT01255423 1104 3102 44 033 [ 003 309 ]
NCT01272934 1102 0103 07 303 [ 012 7350 ]
NCT01272947 0104 0100 Not estimable
Predel 2012 (1) 1160 182 19 051 [ 003 809 ]
Predel 2013a 1120 4116 59 024 [ 003 213 ]
Predel 2013b 036 036 Not estimable
Rowbotham 2003 27191 31181 463 083 [ 051 133 ]
Saillant 1998 070 070 Not estimable
Spacca 2005 079 076 Not estimable
Subtotal (95 CI) 1720 1551 1000 078 [ 056 110 ]Total events 54 (NSAID) 66 (Placebo)
Heterogeneity Chi2 = 944 df = 10 (P = 049) I2 =00
Test for overall effect Z = 142 (P = 016)
2 Ibuprofen
Billigmann 1996 1180 480 573 275 [ 091 827 ]
Machen 2002 440 241 283 205 [ 040 1057 ]
Ramesh 1983 140 140 143 100 [ 006 1544 ]
Subtotal (95 CI) 160 161 1000 230 [ 098 543 ]Total events 16 (NSAID) 7 (Placebo)
Heterogeneity Chi2 = 048 df = 2 (P = 079) I2 =00
Test for overall effect Z = 190 (P = 0057)
3 Ketoprofen
002 01 1 10 50
Favours NSAID Favours placebo
(Continued )
106Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
( Continued)
Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Airaksinen 1993 529 427 99 116 [ 035 389 ]
Dreiser 1989 030 230 60 020 [ 001 400 ]
Julien 1989 130 030 12 300 [ 013 7083 ]
Kockelbergh 1985 138 136 24 095 [ 006 1459 ]
Mazi res 2005a 1281 682 142 202 [ 080 513 ]
Mazi res 2005b 2987 2785 651 105 [ 068 161 ]
Noret 1987 151 047 12 277 [ 012 6636 ]
Parrini 1992 083 086 Not estimable
Subtotal (95 CI) 429 423 1000 119 [ 083 170 ]Total events 49 (NSAID) 40 (Placebo)
Heterogeneity Chi2 = 357 df = 6 (P = 073) I2 =00
Test for overall effect Z = 095 (P = 034)
4 Piroxicam
Aoki 1984 179 274 147 047 [ 004 506 ]
Fujimaki 1985 183 282 143 049 [ 005 534 ]
Russell 1991 4102 10102 710 040 [ 013 123 ]
Subtotal (95 CI) 264 258 1000 042 [ 017 108 ]Total events 6 (NSAID) 14 (Placebo)
Heterogeneity Chi2 = 003 df = 2 (P = 098) I2 =00
Test for overall effect Z = 180 (P = 0072)
5 Felbinac
McLatchie 1989 3118 2113 666 144 [ 024 844 ]
Morris 1991 041 043 Not estimable
Sanguinetti 1989 342 140 334 286 [ 031 2634 ]
Subtotal (95 CI) 201 196 1000 191 [ 049 750 ]Total events 6 (NSAID) 3 (Placebo)
Heterogeneity Chi2 = 023 df = 1 (P = 063) I2 =00
Test for overall effect Z = 093 (P = 035)
6 Indomethacin
Aoki 1984 270 274 440 106 [ 015 730 ]
Fujimaki 1985 582 282 452 250 [ 050 1252 ]
kermark 1990 422 024 108 978 [ 056 17191 ]
Subtotal (95 CI) 174 180 1000 265 [ 091 773 ]Total events 11 (NSAID) 4 (Placebo)
Heterogeneity Chi2 = 167 df = 2 (P = 043) I2 =00
Test for overall effect Z = 179 (P = 0073)
002 01 1 10 50
Favours NSAID Favours placebo
(1) DHEP + heparin and DHEP combined
107Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Analysis 21 Comparison 2 Diclofenac versus placebo (effect of formulation) Outcome 1 Clinical success
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 2 Diclofenac versus placebo (effect of formulation)
Outcome 1 Clinical success
Study or subgroup Experimental Control Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
1 Plaster - Flector
Joussellin 2003 3668 2466 52 146 [ 099 215 ]
Li 2013 173192 114192 242 152 [ 134 172 ]
Rowbotham 2003 75191 48181 104 148 [ 110 200 ]
Saillant 1998 4370 2570 53 172 [ 119 248 ]
Subtotal (95 CI) 521 509 451 153 [ 136 171 ]Total events 327 (Experimental) 211 (Control)
Heterogeneity Chi2 = 051 df = 3 (P = 092) I2 =00
Test for overall effect Z = 720 (P lt 000001)
2 Plaster - other
Coudreuse 2010 99117 82116 175 120 [ 104 138 ]
Klainguti 2010 5662 4659 100 116 [ 099 136 ]
Predel 2004 5560 560 11 1100 [ 474 2555 ]
Subtotal (95 CI) 239 235 285 155 [ 137 175 ]Total events 210 (Experimental) 133 (Control)
Heterogeneity Chi2 = 4658 df = 2 (Plt000001) I2 =96
Test for overall effect Z = 699 (P lt 000001)
3 Gel - Emulgel
Predel 2012 (1) 116160 1782 48 350 [ 227 540 ]
Predel 2013b 3636 736 16 487 [ 257 923 ]
Subtotal (95 CI) 196 118 64 384 [ 268 550 ]Total events 152 (Experimental) 24 (Control)
Heterogeneity Chi2 = 070 df = 1 (P = 040) I2 =00
Test for overall effect Z = 734 (P lt 000001)
4 Gel - other
Predel 2013a 111118 93114 201 115 [ 105 127 ]
Subtotal (95 CI) 118 114 201 115 [ 105 127 ]Total events 111 (Experimental) 93 (Control)
02 05 1 2 5
Favours placebo Favours diclofenac
(Continued )
108Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
( Continued)Study or subgroup Experimental Control Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Heterogeneity not applicable
Test for overall effect Z = 284 (P = 00045)
Total (95 CI) 1074 976 1000 160 [ 149 172 ]Total events 800 (Experimental) 461 (Control)
Heterogeneity Chi2 = 12176 df = 9 (Plt000001) I2 =93
Test for overall effect Z = 1278 (P lt 000001)
Test for subgroup differences Chi2 = 5003 df = 3 (P = 000) I2 =94
02 05 1 2 5
Favours placebo Favours diclofenac
(1) twice daily and three times daily applications combined
Analysis 31 Comparison 3 Ibuprofen versus placebo (effect of formulation) Outcome 1 Clinical success
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 3 Ibuprofen versus placebo (effect of formulation)
Outcome 1 Clinical success
Study or subgroup Ibuprofen Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
1 Cream
Campbell 1994 2126 1925 264 106 [ 080 142 ]
Dreiser 1988 2632 1232 164 217 [ 134 349 ]
Ramesh 1983 2340 2340 314 100 [ 069 146 ]
Subtotal (95 CI) 98 97 742 128 [ 103 159 ]Total events 70 (Ibuprofen) 54 (Placebo)
Heterogeneity Chi2 = 791 df = 2 (P = 002) I2 =75
Test for overall effect Z = 222 (P = 0026)
2 Gel
Billigmann 1996 2580 1080 136 250 [ 129 486 ]
Machen 2002 2540 941 121 285 [ 152 532 ]
Subtotal (95 CI) 120 121 258 266 [ 169 421 ]Total events 50 (Ibuprofen) 19 (Placebo)
001 01 1 10 100
Favours placebo Favours ibuprofen
(Continued )
109Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
( Continued)Study or subgroup Ibuprofen Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Heterogeneity Chi2 = 008 df = 1 (P = 078) I2 =00
Test for overall effect Z = 420 (P = 0000026)
Total (95 CI) 218 218 1000 164 [ 133 201 ]Total events 120 (Ibuprofen) 73 (Placebo)
Heterogeneity Chi2 = 2102 df = 4 (P = 000031) I2 =81
Test for overall effect Z = 465 (P lt 000001)
Test for subgroup differences Chi2 = 805 df = 1 (P = 000) I2 =88
001 01 1 10 100
Favours placebo Favours ibuprofen
Analysis 41 Comparison 4 Ketoprofen versus placebo Outcome 1 Clinical success
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 4 Ketoprofen versus placebo
Outcome 1 Clinical success
Study or subgroup Diclofenac Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
1 Plaster
Mazi res 2005a 7281 6082 376 121 [ 104 141 ]
Mazi res 2005b 5087 4185 262 119 [ 090 158 ]
Subtotal (95 CI) 168 167 638 121 [ 104 140 ]Total events 122 (Diclofenac) 101 (Placebo)
Heterogeneity Chi2 = 002 df = 1 (P = 090) I2 =00
Test for overall effect Z = 249 (P = 0013)
2 Gel
Airaksinen 1993 2429 1427 91 160 [ 107 238 ]
Dreiser 1989 1830 530 32 360 [ 154 844 ]
Julien 1989 1830 630 38 300 [ 138 650 ]
Kockelbergh 1985 3038 2236 142 129 [ 095 176 ]
Noret 1987 3951 947 59 399 [ 218 733 ]
001 01 1 10 100
Favours placebo Favours diclofenac
(Continued )
110Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
( Continued)Study or subgroup Diclofenac Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Subtotal (95 CI) 178 170 362 219 [ 174 275 ]Total events 129 (Diclofenac) 56 (Placebo)
Heterogeneity Chi2 = 1937 df = 4 (P = 000067) I2 =79
Test for overall effect Z = 674 (P lt 000001)
Total (95 CI) 346 337 1000 156 [ 137 177 ]Total events 251 (Diclofenac) 157 (Placebo)
Heterogeneity Chi2 = 3104 df = 6 (P = 000002) I2 =81
Test for overall effect Z = 683 (P lt 000001)
Test for subgroup differences Chi2 = 1859 df = 1 (P = 000) I2 =95
001 01 1 10 100
Favours placebo Favours diclofenac
Analysis 51 Comparison 5 All topical NSAIDs versus placebo Outcome 1 Local adverse events
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 5 All topical NSAIDs versus placebo
Outcome 1 Local adverse events
Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Airaksinen 1993 529 427 26 116 [ 035 389 ]
Aoki 1984 (1) 270 137 08 106 [ 010 1128 ]
Aoki 1984 (2) 179 137 09 047 [ 003 728 ]
Auclair 1989 5123 6116 39 079 [ 025 251 ]
Billigmann 1996 1180 480 26 275 [ 091 827 ]
Costantino 2011 (3) 3287 6142 51 025 [ 006 097 ]
Coudreuse 2010 1117 7116 45 014 [ 002 113 ]
Diebshlag 1990 (4) 113 06 04 150 [ 007 3229 ]
Diebshlag 1990 (5) 112 06 04 162 [ 008 3466 ]
Dreiser 1989 030 230 16 020 [ 001 400 ]
002 01 1 10 50
Favours NSAID Favours placebo
(Continued )
111Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
( Continued)Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Dreiser 1990 030 330 22 014 [ 001 265 ]
Dreiser 1994 265 066 03 508 [ 025 10373 ]
Fujimaki 1985 (6) 183 141 09 049 [ 003 770 ]
Fujimaki 1985 (7) 582 141 09 250 [ 030 2070 ]
Haig 1986 021 022 Not estimable
Jenoure 1997 144 141 07 093 [ 006 1442 ]
Joussellin 2003 168 366 19 032 [ 003 303 ]
Julien 1989 130 030 03 300 [ 013 7083 ]
Klainguti 2010 (8) 0126 159 13 016 [ 001 381 ]
Kockelbergh 1985 138 136 07 095 [ 006 1459 ]
Kuehl 2011 16207 12211 76 136 [ 066 280 ]
Li 2013 4192 3192 19 133 [ 030 588 ]
Linde 1985 050 050 Not estimable
Machen 2002 440 241 13 205 [ 040 1057 ]
Mazi res 2005a 1281 682 38 202 [ 080 513 ]
Mazi res 2005b 2987 2785 174 105 [ 068 161 ]
McLatchie 1989 3118 2113 13 144 [ 024 844 ]
Morris 1991 041 043 Not estimable
NCT01255423 1104 3102 19 033 [ 003 309 ]
NCT01272934 1102 0103 03 303 [ 012 7350 ]
NCT01272947 0104 0100 Not estimable
Noret 1987 151 047 03 277 [ 012 6636 ]
Parrini 1992 083 086 Not estimable
Predel 2012 (9) 1160 182 08 051 [ 003 809 ]
Predel 2013a 1120 4116 26 024 [ 003 213 ]
Predel 2013b 036 036 Not estimable
Ramesh 1983 140 140 06 100 [ 006 1544 ]
Rowbotham 2003 27191 31181 203 083 [ 051 133 ]
Russell 1991 4102 10102 64 040 [ 013 123 ]
Saillant 1998 070 070 Not estimable
Sanguinetti 1989 342 140 07 286 [ 031 2634 ]
002 01 1 10 50
Favours NSAID Favours placebo
(Continued )
112Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
( Continued)Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Sinniger 1981 010 010 Not estimable
Spacca 2005 079 076 Not estimable
Thorling 1990 160 060 03 300 [ 012 7220 ]
kermark 1990 422 024 03 978 [ 056 17191 ]
Total (95 CI) 3619 3121 1000 098 [ 080 121 ]Total events 155 (NSAID) 145 (Placebo)
Heterogeneity Chi2 = 3354 df = 35 (P = 054) I2 =00
Test for overall effect Z = 017 (P = 087)
Test for subgroup differences Not applicable
002 01 1 10 50
Favours NSAID Favours placebo
(1) indomethacin
(2) piroxicam
(3) DHEP + heparin and DHEP combined
(4) ketorolac
(5) etofenamate
(6) piroxicam
(7) indomethacin
(8) DHEP + heparin and DHEP combined
(9) Twice and three times daily application combined
113Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Analysis 52 Comparison 5 All topical NSAIDs versus placebo Outcome 2 Systemic adverse events
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 5 All topical NSAIDs versus placebo
Outcome 2 Systemic adverse events
Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Airaksinen 1993 129 027 05 280 [ 012 6593 ]
Aoki 1984 (1) 079 037 Not estimable
Aoki 1984 (2) 070 037 Not estimable
Campbell 1994 126 025 05 289 [ 012 6775 ]
Chatterjee 1977 025 025 Not estimable
Coudreuse 2010 1117 0116 05 297 [ 012 7227 ]
Diebshlag 1990 (3) 013 06 Not estimable
Diebshlag 1990 (4) 012 06 Not estimable
Dreiser 1988 032 032 Not estimable
Dreiser 1989 030 030 Not estimable
Dreiser 1990 030 030 Not estimable
Dreiser 1994 065 066 Not estimable
Fujimaki 1985 (5) 083 041 Not estimable
Fujimaki 1985 (6) 182 041 07 152 [ 006 3647 ]
Haig 1986 021 022 Not estimable
Hoffmann 2012 0115 0118 Not estimable
Joussellin 2003 168 066 05 291 [ 012 7025 ]
Klainguti 2010 (7) 2126 059 07 236 [ 012 4844 ]
Kuehl 2011 15207 23211 236 066 [ 036 124 ]
Li 2013 10192 4192 41 250 [ 080 783 ]
Linde 1985 050 050 Not estimable
Machen 2002 040 041 Not estimable
Mazi res 2005a 1187 785 73 154 [ 062 377 ]
Mazi res 2005b 1381 1482 144 094 [ 047 187 ]
McLatchie 1989 0118 0113 Not estimable
Morris 1991 041 043 Not estimable
001 01 1 10 100
Favours topical NSAID Favours placebo
(Continued )
114Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
( Continued)Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
NCT01272947 2104 2100 21 096 [ 014 670 ]
Parrini 1992 083 086 Not estimable
Predel 2004 060 060 Not estimable
Predel 2012 (8) 3160 382 41 051 [ 011 248 ]
Predel 2013a 6120 8116 84 073 [ 026 203 ]
Predel 2013b 036 136 16 033 [ 001 792 ]
Rowbotham 2003 21191 22181 234 090 [ 052 159 ]
Russell 1991 4102 7102 73 057 [ 017 189 ]
Saillant 1998 070 070 Not estimable
Sanguinetti 1989 042 040 Not estimable
Sinniger 1981 010 010 Not estimable
Spacca 2005 079 076 Not estimable
Thorling 1990 060 060 Not estimable
Total (95 CI) 2956 2620 1000 096 [ 073 126 ]Total events 92 (NSAID) 91 (Placebo)
Heterogeneity Chi2 = 946 df = 15 (P = 085) I2 =00
Test for overall effect Z = 030 (P = 076)
Test for subgroup differences Not applicable
001 01 1 10 100
Favours topical NSAID Favours placebo
(1) piroxicam
(2) indomethacin
(3) ketorolac
(4) etofenamate
(5) piroxicam
(6) indomethacin
(7) DHEP + heparin and DHEP combined
(8) Twice and three times daily application combined
115Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Analysis 53 Comparison 5 All topical NSAIDs versus placebo Outcome 3 Adverse event withdrawals
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 5 All topical NSAIDs versus placebo
Outcome 3 Adverse event withdrawals
Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Airaksinen 1993 029 027 Not estimable
Aoki 1984 072 034 Not estimable
Aoki 1984 064 033 Not estimable
Auclair 1989 1123 0116 14 283 [ 012 6879 ]
Billigmann 1996 280 080 14 500 [ 024 10253 ]
Campbell 1994 026 025 Not estimable
Chatterjee 1977 025 025 Not estimable
Costantino 2011 0188 0142 Not estimable
Coudreuse 2010 0117 0116 Not estimable
Diebshlag 1990 013 06 Not estimable
Diebshlag 1990 012 06 Not estimable
Dreiser 1988 032 032 Not estimable
Dreiser 1989 030 230 69 020 [ 001 400 ]
Dreiser 1990 030 130 42 033 [ 001 787 ]
Dreiser 1994 065 066 Not estimable
Fujimaki 1985 083 041 Not estimable
Fujimaki 1985 482 041 18 455 [ 025 8261 ]
Julien 1989 030 030 Not estimable
Kockelbergh 1985 038 036 Not estimable
Kuehl 2011 4207 9211 248 045 [ 014 145 ]
Li 2013 2192 0192 14 500 [ 024 10347 ]
Linde 1985 050 050 Not estimable
Machen 2002 040 041 Not estimable
Mazi res 2005a 987 685 169 147 [ 055 394 ]
Mazi res 2005b 381 082 14 709 [ 037 13503 ]
McLatchie 1989 0118 0113 Not estimable
001 01 1 10 100
Favours topical NSAID Favours placebo
(Continued )
116Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
( Continued)Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Morris 1991 041 043 Not estimable
NCT01255423 0104 0102 Not estimable
NCT01272934 0102 1103 41 034 [ 001 817 ]
NCT01272947 0104 0100 Not estimable
Noret 1987 148 045 14 282 [ 012 6740 ]
Parrini 1992 060 060 Not estimable
Predel 2004 160 060 14 300 [ 012 7220 ]
Predel 2012 2160 182 37 103 [ 009 1114 ]
Predel 2013a 1120 1116 28 097 [ 006 1527 ]
Predel 2013b 036 036 Not estimable
Ramesh 1983 140 140 28 100 [ 006 1544 ]
Rowbotham 2003 0191 0181 Not estimable
Russell 1991 1102 8102 222 013 [ 002 098 ]
Saillant 1998 070 070 Not estimable
Sanguinetti 1989 042 040 Not estimable
Sinniger 1981 010 010 Not estimable
Spacca 2005 079 076 Not estimable
Thorling 1990 060 060 Not estimable
kermark 1990 122 024 13 326 [ 014 7610 ]
Total (95 CI) 3365 3040 1000 101 [ 064 159 ]Total events 33 (NSAID) 30 (Placebo)
Heterogeneity Chi2 = 1502 df = 16 (P = 052) I2 =00
Test for overall effect Z = 004 (P = 097)
Test for subgroup differences Not applicable
001 01 1 10 100
Favours topical NSAID Favours placebo
117Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Analysis 61 Comparison 6 Topical NSAID versus active comparator Outcome 1 Clinical success - topical
piroxicam vs topical indomethacin
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 6 Topical NSAID versus active comparator
Outcome 1 Clinical success - topical piroxicam vs topical indomethacin
Study or subgroup Piroxicam Indomethacin Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Aoki 1984 5672 4164 301 121 [ 097 151 ]
Fujimaki 1985 4483 4482 307 099 [ 074 131 ]
Sugioka 1984 85175 55165 392 146 [ 112 190 ]
Total (95 CI) 330 311 1000 124 [ 107 144 ]Total events 185 (Piroxicam) 140 (Indomethacin)
Heterogeneity Chi2 = 390 df = 2 (P = 014) I2 =49
Test for overall effect Z = 278 (P = 00055)
Test for subgroup differences Not applicable
01 02 05 1 2 5 10
Favours indomethacin Favours piroxicam
118Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Analysis 62 Comparison 6 Topical NSAID versus active comparator Outcome 2 Local adverse events -
topical piroxicam vs topical indomethacin
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 6 Topical NSAID versus active comparator
Outcome 2 Local adverse events - topical piroxicam vs topical indomethacin
Study or subgroup Piroxicam Indomethacin Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Aoki 1984 179 270 64 044 [ 004 478 ]
Fujimaki 1985 183 582 152 020 [ 002 165 ]
Sugioka 1984 5178 26179 784 019 [ 008 049 ]
Total (95 CI) 340 331 1000 021 [ 009 047 ]Total events 7 (Piroxicam) 33 (Indomethacin)
Heterogeneity Chi2 = 041 df = 2 (P = 081) I2 =00
Test for overall effect Z = 382 (P = 000013)
Test for subgroup differences Not applicable
002 01 1 10 50
Favours piroxicam Favours indomethacin
A P P E N D I C E S
Appendix 1 CENTRAL search strategy (via CRSO) for 2015 update
1 MESH DESCRIPTOR Anti-Inflammatory Agents Non-Steroidal EXPLODE ALL TREES (13419)2 (bufexamac OR bufexine OR calmaderm OR ekzemase OR dicoflenac OR solaraze OR pennsaid OR voltarol OR emugel OR
voltarene OR voltarol OR optha OR voltaren OR etofenamate OR afrolate OR algesalona OR bayro OR deiron OR etofen ORflexium OR flogoprofen OR rheuma-gel OR rheumon OR traumalix OR traumon OR zenavan OR felbinac OR dolinac OR flexfreeOR napageln OR target OR traxam OR fentiazac OR domureuma OR fentiazaco OR norvedan OR riscalon OR fepradinol ORdalgen OR flexidol OR cocresol OR rangozona OR reuflodol OR pinazone OR zepelin OR flufenamic OR dignodolin OR rheumaOR lindofluid OR sastridex OR lunoxaprofen OR priaxim OR flubiprofen OR fenomel OR ocufen OR ocuflur OR ldquoTrans ActLATrdquo OR tulip OR ibuprofen OR cuprofen OR ldquodeep relief rdquo OR fenbid OR ibu-cream OR ibugel OR ibuleve OR ibumousse ORibuspray OR ldquonurofen gelrdquo OR proflex OR motrin OR advil OR radian OR ralgex OR ibutop OR indomethacin OR indocin ORindospray OR isonixin OR nixyn OR ketoprofen OR tiloket OR oruvail OR powergel OR solpaflex OR ketorolac OR acular ORtrometamol OR meclofenamic OR naproxen OR naprosyn OR niflumic OR actol OR flunir OR niflactol topico OR niflugel ORnifluril OR oxyphenbutazone OR californit OR diflamil OR otone OR tanderil OR piketoprofen OR calmatel OR triparsean ORpiroxicam OR feldene OR pranoprofen OR oftalar OR pranox OR suxibuzone OR danilon OR flamilon OR ufenamate OR fenazolOR flector OR benzydamine) TIABKY (25220)
3 1 OR 2 (32484)4 MESH DESCRIPTOR Administration Topical EXPLODE ALL TREES (2169)
119Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
5 (topical OR cutaneous OR dermal OR transcutaneous OR transdermal OR percutaneous OR skin OR massage ORembrocation OR gel OR ointment OR aerosol OR cream OR cregraveme OR lotion OR mouse OR foam OR liniment OR spray OR rubOR balm OR salve OR emulsion OR oil OR patch OR plaster)TIABKY (67940)
6 4 OR 5 (70486)7 MESH DESCRIPTOR Athletic Injuries EXPLODE ALL TREES (411)8 (strain OR sprain OR contusion OR distortion OR compression OR ldquosports injurrdquo OR ldquosoft tissue injurrdquo OR tendnitis OR
ldquomuscle painrdquo OR periarthritis OR epicondylitis OR tenosynovitis)TIABKY (9158)9 7 OR 8 (9448)
10 MESH DESCRIPTOR pain EXPLODE ALL TREES (29943)11 (pain OR analgesi)TIABKY (74815)12 10 OR 11 (80041)13 3 AND 6 AND 9 AND 12 (110)
Appendix 2 MEDLINE search strategy via Ovid (for 2015 update)
1 exp Anti-inflammatory Agents non-steroidal (162888)2 (bufexamac OR bufexine OR calmaderm OR ekzemase OR diclofenac OR solaraze OR pennsaid OR voltarol OR emugel OR
voltarene OR voltarol OR optha OR voltaren OR etofenamate OR afrolate OR algesalona OR bayro OR deiron OR etofen ORflexium OR flogoprofen OR rheuma-gel OR rheumon OR traumalix OR traumon OR zenavan OR felbinac OR dolinac OR flexfreeOR napageln OR target OR traxam OR fentiazac OR domureuma OR fentiazaco OR norvedan OR riscalon OR fepradinol ORdalgen OR flexidol OR cocresol OR rangozona OR reuflodol OR pinazone OR zepelin OR flufenamic OR dignodolin OR rheumaOR lindofluid OR sastridex OR lunoxaprofen OR priaxim OR flubiprofen OR fenomel OR ocufen OR ocuflur OR ldquoTrans ActLATrdquo OR tulip OR ibuprofen OR cuprofen OR ldquodeep relief rdquo OR fenbid OR ibu-cream OR ibugel OR ibuleve OR ibumousse ORibuspray OR ldquonurofen gelrdquo OR proflex OR motrin OR advil OR radian OR ralgex OR ibutop OR indomethacin OR indocin ORindospray OR isonixin OR nixyn OR ketoprofen OR tiloket OR oruvail OR powergel OR solpaflex OR ketorolac OR acular ORtrometamol OR meclofenamic OR naproxen OR naprosyn OR niflumic OR actol OR flunir OR niflactol topico OR niflugel ORnifluril OR oxyphenbutazone OR californit OR diflamil OR otone OR tanderil OR piketoprofen OR calmatel OR triparsean ORpiroxicam OR feldene OR pranoprofen OR oftalar OR pranox OR suxibuzone OR danilon OR flamilon OR ufenamate OR fenazolOR flector OR benzydamine)mp (558284)
3 1 OR 2 (664691)4 exp Administration Topical (69697)5 (topical OR cutaneous OR dermal OR transcutaneous OR transdermal OR percutaneous OR skin OR massage OR
embrocation OR gel OR ointment OR aerosol OR cream OR cregraveme OR lotion OR mouse OR foam OR liniment OR spray OR rubOR balm OR salve OR emulsion OR oil OR patch OR plaster)mp (1982395)
6 4 OR 5 (1997720)7 exp Athletic Injuries (21464)8 (strain OR sprain OR contusion OR distortion OR compression OR ldquosports injurrdquo OR ldquosoft tissue injurrdquo OR tendnitis OR
ldquomuscle painrdquo OR periarthritis OR epicondylitis OR tenosynovitis)mp (420133)9 7 OR 8 (439228)
10 Pain (113906)11 (pain OR analgesi)mp (585249)12 10 or 11 (585249)13 randomized controlled trialpt (401171)14 randomizedab (296222)15 placeboab (155341)16 drug therapyfs (1789858)17 randomlyab (207517)18 trialab (308477)19 groupsab (1318386)20 OR13-1921 3 AND 6 AND 9 AND 12 AND 20 (139)22 limit 21 to yr=ldquo2008-Currentrdquo (56)
120Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Appendix 3 EMBASE search strategy via Ovid (for 2015 update)
1 exp Anti-inflammatory Agents non-steroidal (452266)2 (bufexamac OR bufexine OR calmaderm OR ekzemase OR dicoflenac OR solaraze OR pennsaid OR voltarol OR emugel OR
voltarene OR voltarol OR optha OR voltaren OR etofenamate OR afrolate OR algesalona OR bayro OR deiron OR etofen ORflexium OR flogoprofen OR rheuma-gel OR rheumon OR traumalix OR traumon OR zenavan OR felbinac OR dolinac OR flexfreeOR napageln OR target OR traxam OR fentiazac OR domureuma OR fentiazaco OR norvedan OR riscalon OR fepradinol ORdalgen OR flexidol OR cocresol OR rangozona OR reuflodol OR pinazone OR zepelin OR flufenamic OR dignodolin OR rheumaOR lindofluid OR sastridex OR lunoxaprofen OR priaxim OR flubiprofen OR fenomel OR ocufen OR ocuflur OR ldquoTrans ActLATrdquo OR tulip OR ibuprofen OR cuprofen OR ldquodeep relief rdquo OR fenbid OR ibu-cream OR ibugel OR ibuleve OR ibumousse ORibuspray OR ldquonurofen gelrdquo OR proflex OR motrin OR advil OR radian OR ralgex OR ibutop OR indomethacin OR indocin ORindospray OR isonixin OR nixyn OR ketoprofen OR tiloket OR oruvail OR powergel OR solpaflex OR ketorolac OR acular ORtrometamol OR meclofenamic OR naproxen OR naprosyn OR niflumic OR actol OR flunir OR niflactol topico OR niflugel ORnifluril OR oxyphenbutazone OR californit OR diflamil OR otone OR tanderil OR piketoprofen OR calmatel OR triparsean ORpiroxicam OR feldene OR pranoprofen OR oftalar OR pranox OR suxibuzone OR danilon OR flamilon OR ufenamate OR fenazolOR flector OR benzydamine)mp (790435)
3 1 OR 2 (1108683)4 exp Administration Topical (68490)5 (topical OR cutaneous OR dermal OR transcutaneous OR transdermal OR percutaneous OR skin OR massage OR
embrocation OR gel OR ointment OR aerosol OR cream OR cregraveme OR lotion OR mouse OR foam OR liniment OR spray OR rubOR balm OR salve OR emulsion OR oil OR patch OR plaster)mp (3208844)
6 4 OR 5 (3208844)7 exp Athletic Injuries (24675)8 (strain OR sprain OR contusion OR distortion OR compression OR ldquosports injurrdquo OR ldquosoft tissue injurrdquo OR tendnitis OR
ldquomuscle painrdquo OR periarthritis OR epicondylitis OR tenosynovitis)mp (804322)9 7 OR 8 (825164)
10 Pain (216406)11 (pain OR analgesi)mp (971593)12 10 OR 11 (971593)13 clinical trialssh (841252)14 controlled clinical trialssh (389335)15 randomized controlled trialsh (357169)16 double-blind proceduresh (118945)17 (clin adj25 trial)ab (331002)18 ((doubl or trebl or tripl) adj25 (blind or mask))ab (141726)19 placeboab (203390)20 randomab (906176)21 OR13-20 (1731380)22 3 AND 6 AND 9 AND 10 AND 19 (439)23 limit 22 to yr=ldquo2008-Currentrdquo (204)
Appendix 4 Summary of outcomes efficacy
Study ID Treatment Clinical response Other response
Airaksinen 1993 (1) Ketoprofen gel 2 x 5 g (125mg) daily n = 29(2) Placebo gel n = 27
PGE rdquoimprovedldquo at 7 days(1) 2429(2) 1427
No additional data
121Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
kermark 1990 (1) Indomethacin spray 1 (El-metacin) 3-5 x 05-15 mL dailyn = 23(2) Indomethacin capsules 3 x 25mg daily n = 23(3) Placebo spray and capsules n= 24
No pain on palpation at 7 days(1) 1222(2) 522(3) 624
Patient assessment of improve-ment at 7 days (scale 0 to 100)(1) 57(2) 49(3) 30
Aoki 1984 (1) Piroxicam gel 5 3 or 4 x 1 gdaily n = 84(2) Indomethacin gel 1 3 or 4x 1 g daily n = 84(3) Placebo gel n = 84
PGE (5 point) rdquobetter or muchbetterldquo at 7 days(1) 5672(2) 4164(3) 3367
Pain on movement rdquoreducedldquo orrdquodisappearedldquo at 7 days(1) 4861(2) 3860(3) 3563
Auclair 1989 (1) Niflumic acid gel 25 3 x 5g daily n = 117(2) Placebo gel n = 110
PGE (5 point) rdquogood or verygoodldquo at 7 days(1) 69117(2) 54110
Pain on palpation rdquoimprovedldquo at7 days(1) 69117(2) 53110
Billigmann 1996 (1) Ibuprofen microgel 5 3 x200 mg daily n = 80(2) Placebo gel n = 80
Complete remission(1) 2580(2) 1080
Improvement in pain with move-ment of 20 at 7 days(1) 6580(2) 5580
Campbell 1994 (1) Ibuprofen cream 5 (Proflex)4 x 4rdquo daily n = 26(2) Placebo cream n = 25
Improvement in walking ability (4point) at 7 days(1) 2126(2) 1925
No additional data
Chatterjee 1977 (1) Benzydamine HCl cream 33 x daily n = 25(2) Placebo cream n = 25
Pain on movement ldquoabsentslightrdquoat 6 days(1) 2125(2) 1225
Tenderness with pressure ldquoabsentslightrdquo at 6 days(1) 2125(2) 1225
Costantino 2011 (1) DHEP-heparin plaster (Flec-torparin) x 1 daily n = 142(2) DHEP plaster (Flector) x 1daily n = 146(3) Placebo plaster n = 142
No responder analysis for efficacyOverall treatment efficacy not re-ported
Mean reduction from baseline forPI on movement at 3 days (fromgraph)(1) 24 mm(2) 19 mm(3) 14 mm
Coudreuse 2010 (1) DHEP-heparin plaster (Flec-torparin) n = 120(2) Placebo n = 120
PGE ldquogoodrdquo or ldquoexcellentrdquo at 7days(1) 99117(2) 82116These are results for physician-re-ported judgement - not partici-pant
Mean reduction from baseline forPI on movement over 6 hours(1) about 30(2) about 20Overall greater in DHEP-heparingroup than placebo over 7 days
122Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Curioni 1985 (1) Ibuproxam n = 20(2) Ketoprofen n = 20(3) Etofenamate n = 20
Resolution of symptoms by 7 days(1) 1520(2) 1320(3) 1320
PGE ldquogoodrdquo or ldquoexcellentrdquo at 10days(1) 1920(2) not reported(3) 1620
Diebshlag 1990 (1) Ketorolac gel 2 3 x 3 g dailyn = 13(2) Etofenamate gel 5 3 x 3 gdaily n = 12(3) Placebo gel n = 12
Improvement in pain at 7 days(1) 1213(2) 1012(3) 912
No additional data
Dreiser 1988 (1) Ibuprofen cream 5 3 x 4 cmdaily n = 32 (3 x 10 cm for largejoints)(2) Placebo cream n = 32
PGE ldquoimprovementrdquo or ldquocom-plete relief rdquo at 7 days(1) 2632(2) 1232
(1) significantly better than (2)for mean improvement in sponta-neous pain movement pain restpain tenderness to pressure (VAS)
Dreiser 1989 (1) Ketoprofen gel 25 2 x 5 cmdaily n = 30(2) Placebo gel n = 30
PGE (3 point) ldquobetterrdquo at 7 days(1) 1830(2) 530
(1) significantly better than (2) formean improvement in pain (restand movement) (VAS)
Dreiser 1990 (1) Niflumic acid gel 25 3 x 5g daily n = 30(2) Placebo gel n = 30
PGE (4 point) ldquocuredrdquo or ldquoim-provedrdquo at 7 days(1) 2330(2) 1030
(1) significantly better than (2) formean improvement in pain (VAS)
Dreiser 1994 (1) Flurbiprofen plaster 2 x 40 mgdaily n = 65(2) Placebo plaster n = 66
PGE (4 point) ldquogoodrdquo or ldquoverygoodrdquo at 7 days(1) 4865(2) 4166
(1) significantly better than (2)for mean improvement in sponta-neous pain but not pain on move-ment or palpation (VAS)
Fioravanti 1999 (1) DHEP lecithin gel 3 x 5 g (=65 mg) daily n = 50(2) DHEP gel 3 x 5 g (= 65 mg)daily n = 50
PGE (4 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 10 days(1) 3550(2) 3550
(1) significantly better than (2)for mean improvement in spon-taneous pain at 7 days but notfor pain on movement at 10 days(VAS)
Fujimaki 1985 (1) Piroxicam gel 05 3 or 4 x 1g daily n = 92(2) Indomethacin gel 1 3 or 4 x1 g daily n = 90(3) Placebo gel n = 89
PGE (5 point) ldquobetterrdquo or ldquomuchbetterrdquo at end of treatment at 14days(1) 4483(2) 4482(3) 4082
No additional data
Gallacchi 1990 (1) Diclofenac hy-droxyethylpyrrolidine gel 1 4 x2 g daily n = 25 (Flector gel)(2) Diclofenac sodium 1 4 x 2 g
PGE (5 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 14 days(1) 1925(2) 1925
No significant difference betweengroups for pain on applied pres-sure at 7 and 14 days
123Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
daily n = 25 (Voltaren Emugel)
Gonzaacutelez de Vega 2013 (1) Traumeel gel 3 x 2 g daily n =140(2) Traumeel ointment n = 143(3) Diclofenac gel 1 3 x 2 g dailyn = 137
Pain-free at 7 days(1) 7140(3) 8137PGE (5 point) ldquogoodrdquo or ldquoverygoodrdquo(1) 128140(3) 127137
Normal function (5 point) scoreof 0 or 1 at 14 days(1) 133140(3) 131137
Governali 1995 (1) Ketoprofen gel 5 3 x 2-3 gdaily n = 15(2) Ketoprofen cream 1 3 x 2-3 g daily n = 15
PGE (5 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 7 days(1) 1415(2) 415
No additional data
Gualdi 1987 (1) Flunaxaprofen gel 2 x 3-5 cmdaily n = 30(2) Ketoprofen gel 2 x 3-5 cmdaily n = 30
No dichotomous data No significant difference betweengroups for pain on movement at 7days
Haig 1986 (1) Benzydamine cream 3 6 xdaily n = 21(2) Placebo cream n = 22
Pain on movement ldquoimprovedrdquo by6 days(1) 1821(2) 1322
No additional data
Hoffmann 2012 (1) DHEP-heparin plaster (Flec-torparin) x 1 daily n = 121(2) DHEP (Flector) plaster x 1daily n = 115(3) Placebo plaster n = 118
PGE (5 point) ldquogoodrdquo or ldquoexcel-lentrdquo over 80 in all groups
Mean reduction from baseline inPI (100 mm VAS)Day 3(1) 191 mm(2) 114 mm(3) 52 mmDay 8(1) 443 mm(2) 372 mm(3) 296 mm
Hofman 2000 (1) Diclofenac sodium gel 1 4x 2 cm daily n = 69(2) Lysine clonixinate gel 5 4 x2 cm (225 mg) daily n = 73
PGE (3 point) at 8 days ldquogoodrdquo(1) 3869(2) 3673
No significant difference betweentreatments for any pain outcomes
Hosie 1993 (1) Felbinac foam 3 3 x 2 g daily+ placebo tablets 3 x 1 daily n =140 (127 analysed for efficacy)(2) Ibuprofen tablets 3 x 400 mgdaily + placebo foam 3 x 2 g dailyn = 147 (134 analysed for efficacy)
Pain on movement ldquononerdquo orldquomildrdquo at 7 days(1) 81127(2) 96133
Spontaneous pain ldquononerdquo orldquomildrdquo at 7 days(1) 99127(2) 108134
124Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Jenoure 1997 (1) DHEP plaster (Tissugel) 2 xdaily n = 44(2) Placebo plaster 2 x daily n =41
Baseline pain in two groups notbalanced and data in table and fig-ure do not agree so efficacy out-comes not used
No additional data
Joussellin 2003 (1) DHEP plaster (Flector Tis-sugel 1) 1 x daily n = 68(2) Placebo plaster 1 x daily n =66
PGE (4 point) ldquoexcellentrdquo at 7 days(1) 3668(2) 2466ge 30 reduction in PI at 7 days(1) 2568(2) 1166
(1) significantly better than (2) formean pain on movement at 6 days
Julien 1989 (1) Ketoprofen gel 25 2 x 5 cm(= 50 mg) daily n = 30(2) Placebo gel n = 30
PGE (4 point) ldquorecoveredrdquo at 7days(1) 1830(2) 630
PGE (4 point) ldquorecoveredrdquo or ldquoim-provedrdquo at 7 days(1) 2530(2) 1330
Klainguti 2010 (1) DHEP-heparin plaster (Flec-torparin) x 1 daily n = 62(2) DHEP plaster (Flector) x 1daily n = 61(3) Placebo n = 59
Overall treatment efficacy (partic-ipant and investigator) (5 point)ldquogoodrdquo or ldquoexcellentrdquo at 3 days(1) 5662(2) not reported(3) 4659
Mean reduction in PI at day 3(from graph)(1) 33 mm(2) Not reported(3) 24 mm
Kockelbergh 1985 (1) Ketoprofen gel 25 2 x 5 cm(= 15 mg) daily n = 38(2) Placebo gel n = 36
PGE (3 point) ldquogoodrdquo at 7 days(1) 3038(2) 2236
(1) and (2) slightly better than (3)for mean spontaneous pain at 7days
Kuehl 2011 (1) DETP 13 plaster 2 x dailyn = 207(2) Placebo plaster n = 211
No dichotomous data Percentage reduction from base-line at last application(1) 73(2) 62
Li 2013 (1) DHEP plaster (Flector) 2 xdaily n = 192(2) Placebo plaster 2 x daily n =192
ge 50 pain reduction at day 7(posthoc analysis)(1) 173192(2) 114192
Overall treatment efficacy (5point) ldquogoodrdquo or ldquoexcellentrdquo(1) 161192(2) 81192Participant and investigator ratingMean plusmn SD reduction in PI onmovement at 7 days(1) 538 plusmn 170(2) 370 plusmn 183
Linde 1985 (1) Benzydamine 3 cream 3 xdaily n = 50(2) Placebo gel n = 50
No pain on movement (walking)at 8 days(1) 3550(2) 4050
No additional data
125Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Machen 2002 (1) Ibuprofen gel 5 3 x daily n= 40(2) Placebo gel n = 41
PGE (5 point) ldquomarked improve-mentrdquo or ldquocomplete clearancerdquo at7 days(1) 2540(2) 941
Clinically meaningful (ge 30 mm)pain relief at day 7(1) 3040(2) 1641
Mahler 2003 (1) DHEP + lethicin gel 3 x 5 gdaily n = 52(2) DHEP gel 3 x 5 g daily n = 48
PGE (4 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 10 days(1) 4952(2) 3948
Mean reduction in pain on move-ment at 3 and 10 days significantlygreater with (1) than (2)
Maziegraveres 2005b (1) Ketoprofen plaster 100 mg x1 daily n = 81(2) Placebo plaster n = 82
PGE (4 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 14 days(1) 7281(2) 6082
All mean efficacy measures im-proved more for (1) than (2) mostwere statistically significant
Maziegraveres 2005a (1) Ketoprofen plaster 100 mg x1 daily n = 87(2) Placebo plaster n = 85
PGE (4 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 14 days(1) 5087(2) 4185
All mean efficacy measures im-proved more for (1) than (2) mostwere statistically significant
McLatchie 1989 (1) Felbinac gel 3 3 x 3 cm dailyn = 118(2) Placebo gel n = 113
No dichotomous data Patient daily self-assessment formean pain on rest movement atnight interference with normaland leisure activities show betterefficacy for (1) than (2) from day2 (VAS)
Morris 1991 (1) Felbinac gel 3 3 x 1 cm dailyn = 41(2) Placebo gel n = 43
PGE (5 point) ldquogoodrdquo or ldquoverygoodrdquo at 7 days(1) 2341(2) 2743
(1) better than (2) for meanimprovement in symptoms andsporting function at 7 days
NCT01255423 (1) Diclofenac sodium gel 1 x 4daily n = 104(2) Placebo gel x 4 daily n = 100
No dichotomous data VAS (mean plusmn SD) at 72 hours(baseline PI not reported)(1) 374 plusmn 252(2) 388 plusmn 241
NCT01272934 (1) Diclofenac sodium gel 1 x 4daily n = 104(2) Placebo gel x 4 daily n = 100
No dichotomous data VAS (mean plusmn SD) at 72 hours(baseline PI not reported)(1) 256 plusmn 159(2) 612 plusmn 166
NCT01272947 (1) Diclofenac sodium gel 1 x 4daily n = 104(2) Placebo gel x 4 daily n = 100
No dichotomous data VAS (mean plusmn SD) at 24 hours(baseline PI not reported)(1) 331 plusmn 214(2) 654 plusmn 169
126Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Noret 1987 (1) Ketoprofen gel 25 2 x 5 cm(75 mg) daily n = 48(2) Placebo gel n = 45
PGE (4 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 8 days(1) 3948(2) 945
Decrease in mean spontaneouspain significantly greater in (1)than (2) by 3 days
Parrini 1992 (1) Ketoprofen foam 15 3 x 2 g(200 mg) daily n = 83(2) Placebo foam n = 86
No dichotomous data Mean pain on movement andpressure significantly decreased by7 days in (1) compared with (2)
Picchio 1981 (1) Ibuprofen gel 10 3 x daily n = 20(2) Ketoprofen gel 1 3 x daily n= 20
No pain on movement at 8 days(1) 320(2) 020
Spontaneous pain ldquononerdquo at 8days(1) 620(2) 020
Predel 2004 (1) Diclofenac sodium plaster 2 xdaily (140 mgplaster) n = 60(2) Placebo plaster n = 60
PGE (4 point) ldquogoodrdquo ldquoexcellentrdquoat 7 days(1) 5560(2) 560
(1) better than (2) for reductionin tenderness pain and speed ofpain reduction
Predel 2012 (1) Diclofenac gel (VoltarenEmulgel 232) 2 x 5 cm daily n= 80(2)Diclofenac gel (Voltaren Emulgel232 ) 3 x 5 cm daily n = 80(3) Placebo gel n = 82
ge 50 red in PI on movement at5 days(1) 5780(2) 5980(3) 1782
PGE efficacy (5 point) ldquogoodrdquo orldquovery goodrdquo at 8 days(1) 6880(2) 7380(3) 2482
Predel 2013a (1) Diclofenac 4 spray gel 4 or5 sprays 3 x daily (96-120 mg di-clofenac sodium) n = 118(2) Placebo spray gel n = 114
None or slight PI on movement at7-8 days(1) 111118(2) 93114
None or slight PI on movement at3-4 days(1) 76118(2) 58114
Predel 2013b (1) Diclofenac gel (VoltarenEmulgel) 4 x 2 g daily n = 36(2) Placebo gel 4 x 2 g daily n =36
PGE (5 point) ldquogoodrdquo or ldquoexcel-lentrdquo(1) 3636(2) 736
ge 50 reduction in PI on move-ment after 48 hours(1) 3436(2) 336
Ramesh 1983 (1) Ibuprofen cream 5 3 or 4 x5-10 cm daily n = 40(2) Placebo cream n = 40
Pain on movement (4 point)ldquononerdquo or ldquoslightrdquo at 7 days(1) 2340(2) 2340
Physician global assessment at 10days ldquogoodrdquo(1) 2940(2) 1640
Rowbotham 2003 (1) Diclofenac epolamine plaster(Flector Tissuegel) 2 x daily n =191(2) Placebo plaster n = 181
Pain intensity le 210 for 2 daysor 4 consecutive evaluations by 7days(1) 75191(2) 48181
Mean pain on rest significantlybetter with (1) than (2) after 7 days
127Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Russell 1991 (1) Piroxicam gel 05 4 x 5 mgdaily n = 100(2) Placebo gel n = 100
PGE (4 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 8 days(1) 79100(2) 45100
Statistically greater reduction inmean pain on movement at 8 dayswith (1) than (2)
Saillant 1998 (1) DHEP plaster (Flector Tis-sugel 1) 1 x daily n = 70(2) Placebo plaster 1 x daily n =70
PGE (4 point) ldquoexcellentrdquo at 7 days(1) 4370(2) 2570
ge 30 decrease in PI at 7 days(1) 6470(2) 5070
Sanguinetti 1989 (1) Felbinac gel 3 3 x daily n =42(2) Placebo gel n = 40
PGE ldquogoodrdquo or ldquovery goodrdquo at 7days(1) 3442(2) 1140
(1) better than (2) by 2 days
Sinniger 1981 (1) Fentiazac cream 5 2 or 3 xdaily n = 10(2) Placebo cream n = 10
Complete pain relief within 10days(1) 710(2) 110
Improvement in active pain onmovement at 5 days(1) 67(2) 32
Spacca 2005 (1) DHEP lecithin gel (Effigel) 3x 5 g daily n = 79(2) Placebo gel n = 76
No dichotomous data Mean pain scores improved morerapidly in (1) than (2) - statisticallysignificant at 3 and 6 days
Sugioka 1984 (1) Piroxicam gel 05 3 or 4 x 1g daily n = 183(2) Indomethacin gel 1 3 or 4 x1 g daily n = 183
PGE (5 point) ldquobetterrdquo or ldquomuchbetterrdquo at 14 days(1) 85175(2) 55165
Pain on movement ldquoreducedrdquo orldquodisappearedrdquo at 7 days(1) 77175(2) 63165
Thorling 1990 (1) Naproxen gel 10 2-6 x dailyn = 60(2) Placebo gel n = 60
PGE (5 point) ldquogoodrdquo or ldquoverygoodrdquo at 7 days(1) 3860(2) 2760
Participants using naproxen im-proved more rapidly and had sig-nificantly lower severity scores byday 3
Tonutti 1994 (1) Ketoprofen gel 5 3 x 2-3 gdaily n = 15(2) Etofenamate gel 5 3 x 2-3g n = 15
PGE (4 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 7 days(1) 1015(2) 1115
Significant reductions in pain onmovement by 7 days in bothgroups
Vecchiet 1991 (1) Meclofenamic acid gel 5 2 x10 cm daily (2 g) n = 30(2) Placebo n = 30
PGE (4 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 10 days(1) 3030(2) 1930
(1) significantly better than (2)for mean improvement in sponta-neous pain movement pain func-tional restriction
Whitefield 2002 (1) Ibuprofen gel 5 + placebotablet 3 x daily n = 50(2) Ibuprofen 400 mg tablet +placebo gel 3 x daily n = 50
Participant satisfied at 7 days(1) 3050(2) 3650
ldquoCompletely betterrdquo at 14 days(1) 2450(2) 3050
128Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
DHEP diclofenac epolamine HCl hydrochloride n number PGE participant global evaluation PI pain intensity SD standarddeviation VAS visual analogue scale
Appendix 5 Summary of outcomes adverse events and withdrawals
Study ID Treatment Local AEs Systemic AEs Serious AEs Withdrawals
Airaksinen 1993 (1) Ketoprofen gel 2x 5 g (125 mg) dailyn = 29(2) Placebo gel n =27
(1) 529(2) 427
(1) 129 (nausea af-ter paracetamol)(2) 027
None AE noneOther nonereported
kermark 1990 (1)Indomethacin spray1 (Elmetacin) 3-5 x 05-15 mL dailyn = 23(2) Indomethacincapsules 3 x 25 mgdaily n = 23(3) Placebo sprayand capsules n = 24
(1) 422(2) 022(3) 024
No usable data -reported for eventsnot participants
None reported AE (1) 1 (2) 1 (3)0Lost to follow-up(1) 1 (2) 2 (3) 3
Aoki 1984 (1) Piroxicam gel5 3 or 4 x 1 gdaily n = 84(2) Indomethacingel 1 3 or 4 x 1 gdaily n = 84(3) Placebo gel n =84
(1) 179(2) 270(3) 274
None None reported AE none23 excluded for pro-tocol violations (1)7 (2) 7 (3) 926 withdrew for rea-sons unrelated totreatment (1) 5 (2)13 (3) 8
Auclair 1989 (1) Niflumic acidgel 25 3 x 5 gdaily n = 117(2) Placebo gel n =110
All AEs(1) 5123(2) 6116Most commonly cu-taneous eruptions
No usable data None reported AE (1) 1123 (2) 011626 excluded fromefficacy analysis fornot meeting entrycriteria and protocolviolations
Billigmann 1996 (1) Ibuprofen mi-crogel 5 3 x 200mg daily n = 80(2) Placebo gel n =
(1) 1180(2) 480
None reported None reported AE (1) 280 (aller-gic rash dermatitis)No reason given (1)380 (2) 580
129Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
80 Symptom-free (1)180 (2) 180
Campbell 1994 (1) Ibuprofen cream5 (Proflex) 4 x 4rdquodaily n = 26(2) Placebo cream n= 25
No data (1) 126 (headache)(2) 025
No data AE noneExclusions 493 presented late 2missing forms 1 ap-peared twice 43 didnot return diaries
Chatterjee 1977 (1) BenzydamineHCl cream 3 3 xdaily n = 25(2) Placebo cream n= 25
None None None AE none1 participant lostto follow-up (groupnot reported)
Costantino 2011 (1) DHEP-hep-arin plaster (Flector-parin) x 1 daily n =142(2) DHEP plaster(Flector) x 1 daily n= 146(3) Placebo plastern = 142
Possibly or probablydrug-related(1) 2142(2) 1145(3) 6142
All mild and goneby 14 days ex-cept 1 (moderate) inplacebo group
Not reportedNo systemic GIevents
None AE noneOther(1) 0142(2) 2146 (proto-col deviation lost tofollow-up)(3) 2142 (lost tofollow-up)
Coudreuse 2010 (1) DHEP-hep-arin plaster (Flector-parin) n = 120(2) Placebo n = 120
(1) 1117(2) 7116
All AEs mild ormoderate resolvedspontaneously
(1) 1117(2) 0116
None AE(1) 1117 (increasedoedema)(2) 0116Other(1) 6117 (recovery3 no reason 3)(2) 3116 (recovery1 no reason 2)
Curioni 1985 (1) Ibuproxam n =20(2) Ketoprofen n =20(3) Etofenamate n =20
None None None AE noneOther none
Diebshlag 1990 (1) Ketorolac gel2 3 x 3 g daily n =13(2) Etofenamate gel5 3 x 3 g daily n =12
(1) 113(2) 112(3) 012
None None AE none1 ketorolac partic-ipant did not at-tend 15 day follow-up due to car acci-
130Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
(3) Placebo gel n =12
dent
Dreiser 1988 (1) Ibuprofen cream5 3 x 4 cm dailyn = 32 (3 x 10 cmfor large joints)(2) Placebo cream n= 32
No usable data None Not reported AE none4 placebo partici-pants lost to follow-up
Dreiser 1989 (1) Ketoprofen gel25 2 x 5 cm dailyn = 30(2) Placebo gel n =30
(1) 030(2) 230
None None reported AE (2) 230 (intol-erance)LoE (1) 130 (2) 130
Dreiser 1990 (1) Niflumic acidgel 25 3 x 5 gdaily n = 30(2) Placebo gel n =30
(1) 030(2) 330
None None AE (2) 130 (ery-thema)Exclusion 1 from(2) from efficacyanalysis for inade-quate baseline pain
Dreiser 1994 (1)Flurbiprofen plaster2 x 40 mg daily n =65(2) Placebo plastern = 66
(1) 265(2) 066
None None AE 0(1) 165 excludedfrom efficacy analy-sis for protocol vio-lation(2) 266 (1 LoE 1cured)
Fioravanti 1999 (1) DHEP lecithingel 3 x 5 g (= 65 mg)daily n = 50(2) DHEP gel 3 x 5g (= 65 mg) daily n= 50
(1) 050(2) 150
No data None reported AE noneOther none
Fujimaki 1985 (1) Piroxicam gel 05 3 or 4 x 1 g dailyn = 92(2) Indomethacingel 1 3 or 4 x 1 gdaily n = 90(3) Placebo gel n =89
(1) 183(2) 582(3) 282
(1) 083(2) 182 (nauseaand vomiting)(3) 082
None AE (1) 0 (2) 4 (3)0Unknown reasons(1) 2 (2) 1Did not return af-ter 1st visitirregularvisits (1) 6 (2) 6(3) 7
Gallacchi 1990 (1) Diclofenac hy-droxyethylpyrro-lidine gel 1 4 x 2
No AEs None None AE noneOther none
131Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
g daily n = 25 (Flec-tor gel)(2) Di-clofenac sodium 14 x 2 g daily n = 25(Voltaren Emugel)
Gonzaacutelez de Vega2013
(1) Traumeel gel 3 x2 g daily n = 140(2) Traumeel oint-ment n = 143(3) Diclofenac gel1 3 x 2 g daily n =137
No usable data
Infrequent mild tomoderate
No usable data None AE noneOther(1) 11140(2) 12137
Governali 1995 (1) Ketoprofen gel5 3 x 2-3 g dailyn = 15(2) Ketoprofencream 1 3 x 2-3 gdaily n = 15
No side effects None None AE noneOther none
Gualdi 1987 (1) Flunaxaprofengel 2 x 3-5 cm dailyn = 30(2) Ketoprofen gel 2x 3-5 cm daily n =30
(1) 130(2) 330
No data None reported AE noneOther none
Haig 1986 (1) Benzydaminecream 3 6 x dailyn = 21(2) Placebo cream n= 22
No AEs reported None None reported AE none reportedOther no data
Hoffmann 2012 (1) DHEP-hep-arin plaster (Flector-parin) x 1 daily n =121(2) DHEP (Flector)plaster x 1 daily n =115(3) Placebo plastern = 118
Most AEs were mi-nor local reactions(eg pruritus anderythema) in areaof plaster of mildto moderate inten-sity and resolvedwithout interrupt-ing treatment
No treatment-related systemic AEsrecorded
(2) 1 participanthad 3 SAEs nonejudged related tostudy medication
AE noneExclusions(1) 5121(2) 10115(3) 7118Excluded fromper protocol analysisdue to poor compli-ance or personal de-cision
Hofman 2000 (1) Diclofenacsodium gel 1 4 x2 cm daily n = 69(2) Lysine clonixi-
(1) 158(2) 161
None None AE noneLoE (1) 9 (2) 8
132Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
nate gel 5 4 x 2cm (225 mg) dailyn = 73
Hosie 1993 (1) Felbinac foam3 3 x 2 g daily+ placebo tablets 3x 1 daily n = 140(127 analysed for ef-ficacy)(2) Ibupro-fen tablets 3 x 400mg daily + placebofoam 3 x 2 g daily n= 147 (134 analysedfor efficacy)
(1) 1127(2) 3134
GI events (1) 14127 (2) 11134For (1) more mildnone definitely drugrelated for(2) definitely relatedto study drug
None AE noneExclusions (1) 13(2) 13 did not returnfor 7 day follow-up
Jenoure 1997 (1) DHEP plaster(Tissugel) 2 x dailyn = 44(2) Placebo plaster 2x daily n = 41
(1) 144(2) 141
No data None reported AE none reportedOther nonereported
Joussellin 2003 (1) DHEPplaster (Flector Tis-sugel 1) 1 x dailyn = 68(2) Placebo plaster 1x daily n = 66
(1) 168 (pruritus)(2) 366 (pruritus 2burning 1)
All AEs mild ormoderate
(1) 168 (allergic re-action)(2) 066
None reported AE(1) 066(2) 166Other(1) 366(2) 266
Julien 1989 (1) Ketoprofen gel25 2 x 5 cm (= 50mg) daily n = 30(2) Placebo gel n =30
(1) 130(2) 030
Not reported None AE noneOther none
Klainguti 2010 (1) DHEP-hep-arin plaster (Flector-parin) x 1 daily n =62(2) DHEP plaster(Flector) x 1 daily n= 61(3) Placebo n = 59
(1) 062(2) 061(3) 159All AEs mild in na-ture and resolvedspontaneously
(1) 165 (facial in-fection)(2) 161 (abdominalpain)(3) 059
None AE noneOther(1) 365(2) 161(3) 359
Kockelbergh 1985a (1) Ketoprofen gel25 2 x 5 cm (= 15mg) daily n = 38(2) Placebo gel n =
(1) 138(2) 126
Not reported None AE noneOther none
133Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
36
Kuehl 2011 (1) DETP 13plaster 2 x daily n =207(2) Placebo plastern = 211
(1) 16207(2) 12211
(1) 15207(2) 23211
None AE(1) 4207(2) 9211LoE(1) 21207(2) 25211Other(1) 58207(2) 62211
Li 2013 (1) DHEP plaster(Flector) 2 x daily n= 192(2) Placebo plaster 2x daily n = 192
(1) 4192(2) 3192
(1) 10192(2) 4192
None AE(1) 2192(2) 0192Other(1) 5192(2) 1192
Linde 1985 (1) Benzydamine3 cream 3 x dailyn = 50(2) Placebo gel n =50
(1) 440(2) 241
None None AE none(1) 6 (2) 6 excludedfrom 1st assessment(1) 3 (2) 4 ex-cluded from final as-sessment
Machen 2002 (1) Ibuprofen gel5 3 x daily n = 40(2) Placebo gel n =41
(1) 440(2) 241
None None AE none(1) 1 LoE 1 proto-col violation(2) 4 LoE
Mahler 2003 (1) DHEP + lethicingel 3 x 5 g daily n =52(2) DHEP gel 3 x 5g daily n = 48
(1) 152(2) 048
(1) 152(2) 048
None AE none5 lost to follow-up
Maziegraveres 2005b (1) Ketoprofen plas-ter 100 mg x 1 dailyn = 81(2) Placebo plastern = 82
At 21 days(1) 1281(2) 682
(1) 1381(2) 1482
None AE (1) 381(1) 781 (1 LoE 6cured)(2) 782 (5 LoE 2cured)
Maziegraveres 2005a (1) Ketoprofen plas-ter 100 mg x 1 dailyn = 87(2) Placebo plastern = 85
At 21 days(1) 2987(2) 2785
(1) 1187(2) 785
None AE (1) 987 (2) 685(1) 687 (2 LoE 4cured)(2) 585 (4 LoE 1cured)
134Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
McLatchie 1989 (1) Felbinac gel 33 x 3 cm daily n =118(2) Placebo gel n =113
(1) 3118(2) 2113Mild transient localirritation
None reported None AE noneOther none
Morris 1991 (1) Felbinac gel 33 x 1 cm daily n =41(2) Placebo gel n =43
None None None AE none(1) 4 (protocol vio-lations)(2) 1 (lost to follow-up)Exclusions 11 fromefficacy analysis be-cause evaluated by4 different investi-gators
NCT01255423 (1) Diclofenacsodium gel 1 x 4daily n = 104(2) Placebo gel x 4daily n = 100
(1) 1104(2) 3102
Total AE(1) 11104(2) 8102
None None
NCT01272934 (1) Diclofenacsodium gel 1 x 4daily n = 104(2) Placebo gel x 4daily n = 100
(1) 1102(2) 0103
Total AE (excludingSAE)(1) 6102(2) 3103
(1) 0102(2) 1103 (rupturedligaments in wrist)
AE see SAE
NCT01272947 (1) Diclofenacsodium gel 1 x 4daily n = 104(2) Placebo gel x 4daily n = 100
None (1) 2104(2) 2100
None None
Noret 1987 (1) Ketoprofen gel25 2 x 5 cm (75mg) daily n = 48(2) Placebo gel n =45
(1) 151(2) 047
None reported Not reported AE (1) 151 (skinallergy)(1) 1 LoE 1 unre-lated to trial(2) 1 LoE 1 unre-lated to trial
Parrini 1992 (1) Ketopro-fen foam 15 3 x 2g (200 mg) daily n= 83(2) Placebo foam n= 86
None None None AE noneOther none
135Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Picchio 1981 (1) Ibuprofen gel10 3 x daily n =20(2) Ketoprofen gel1 3 x daily n = 20
None None None AE noneOther not reported
Predel 2004 (1) Diclofenacsodium plaster 2 xdaily (140 mgplas-ter) n = 60(2) Placebo plastern = 60
12 participants ex-pe-rienced 16 mild AEswith no differencesbetween groups
None None AE (1) 160Other none
Predel 2012 (1) Diclofenac gel(Voltaren Emulgel232) 2 x 5 cmdaily n = 80(2) Diclofenac gel(Voltaren Emulgel232) 3 x 5 cmdaily n = 80(3) Placebo gel n =82
(1) 080(2) 180(3) 182All mild to moder-ate
(1) 280(2) 180(3) 382
None AE(1) 080(2) 080(3) 182Other(1) 180(2) 280(3) 282(protocol violationslost to follow-up)
Predel 2013a (1) Diclofenac 4spray gel 4 or 5sprays 3 x daily (96-120 mg diclofenacsodium) n = 118(2) Placebo spraygel n = 114
(1) 1118(2) 4114
(1) 5118(2) 4114All AEs mild re-versible
None AE(1) 1118(2) 1114Other(1) 3118(2) 43114
Predel 2013b (1) Diclofenac gel(Voltaren Emulgel)4 x 2 g daily n = 36(2) Placebo gel 4 x 2g daily n = 36
None (1) 036(2) 136
None AE noneOther none
Ramesh 1983 (1) Ibuprofen cream5 3 or 4 x 5-10 cmdaily n = 40(2) Placebo cream n= 40
(1) 140(2) 140
None reported Not reported AE (1) 140 (2) 140Other none
Rowbotham 2003 (1) DHEP plaster(Flector Tissuegel) 2x daily n = 191(2) Placebo plastern = 181
(1) 27191 (pruritis14)(2) 31181 (pruritis21)
(1) 21191(2) 22181
None re-ported (ldquovast major-ity mildrdquo)
AE none(1) 3191 (2) 4181(did not finish trialand complete dailydiaries)
136Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Russell 1991 (1) Piroxicam gel 05 4 x 5 mg daily n= 100(2) Placebo gel n =100
(1) 4102(2) 10102
GI or CNS events(1) 4 (2) 7Any AE (1) 7102(2) 15102
None reported AE (1) 1102 (2) 8102Other (1) 6 LoE 1ldquootherrdquo(2) 42 LoEExclusions7 did not complywith study medica-tion schedule 6 lostto follow-up 1 pro-tocol violation
Saillant 1998 (1) DHEPplaster (Flector Tis-sugel 1) 1 x dailyn = 70(2) Placebo plaster 1x daily n = 70
None None None AE noneOther(1) 570(2) 570
Sanguinetti 1989 (1) Felbinac gel 33 x daily n = 42(2) Placebo gel n =40
(1) 342(2) 140
None None reported AE noneOther nonereported
Sinniger 1981 (1) Fentiazac cream5 2 or 3 x daily n= 10(2) Placebo cream n= 10
ldquoNo untoward sideeffectsrdquo
None None AE noneOther nonereported
Spacca 2005 (1) DHEP lecithingel (Effigel) 3 x 5 gdaily n = 79(2) Placebo gel n =76
ldquoNo signsof cutaneous irrita-tion or sensitisationobservedrdquo
No AEs observed None AE noneOther nonereported
Sugioka 1984 (1) Piroxicam gel 05 3 or 4 x 1 g dailyn = 183(2) Indomethacingel 1 3 or 4 x 1 gdaily n = 183
(1) 5178(2) 26179
None reported None reported AE none reportedExclusions due toprotocol violations(1) 8 (2) 18Withdrawals(1) 11 (2) 12
Thorling 1990 (1) Naproxen gel10 2-6 x daily n =60(2) Placebo gel n =60
(1) 160(2) 060
None None AE none(1) 1 LoE 1 proto-col violation(2) 1 participant re-quest
137Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Tonutti 1994 (1) Ketoprofen gel5 3 x 2-3 g dailyn = 15(2) Etofenamate gel5 3 x 2-3 g n =15
None No AEs attributableto the medication
None AE NoneLoE (1) 1 (2) 2
Vecchiet 1991 (1) Meclofe-namic acid gel 5 2x 10 cm daily (2 g)n = 30(2) Placebo n = 30
Tolerabil-ity excellent or goodin nearly all partici-pants
No data None AE none reported(2) 5 lost to follow-up
Whitefield 2002 (1) Ibuprofen gel5 + placebo tablet3 x daily n = 50(2) Ibuprofen 400mg tablet + placebogel 3 x daily n = 50
No data 6 AEs reportednone judged relatedto study medication
None reported AE noneRecovered (1) 3 (2)2LoE (2) 1Lost to follow-up(1) 1 (2) 1
AE adverse event CNS central nervous system DHEP diclofenac epolamine GI gastrointestinal HCl hydrochloride LoE lackof efficacy n number SAE serious adverse event
Appendix 6 Concentration amount and frequency of dosing
Study Drug Concentration Quantity Frequency Estimated daily dose of topi-cal NSAID
Joussellin 2003 Diclofenac 1 Plaster 1 180 mg epolamine salt 140 mgNa salt
Li 2013 Diclofenac 1 Plaster 2 360 mg epolamine salt 280 mgNa salt
Rowbotham 2003 Diclofenac 1 Plaster 2 360 mg epolamine salt 280 mgNa salt
Saillant 1998 Diclofenac 1 Plaster 1 180 mg epolamine salt 140 mgNa salt
Coudreuse 2010 Diclofenac 1 Plaster 1 180 mg epolamine salt 140 mgNa salt
Klainguti 2010 Diclofenac 1 Plaster 1 180 mg epolamine salt 140 mgNa salt
138Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Predel 2004 Diclofenac - Plaster 2 280 mg Na salt
Predel 2012 Diclofenac 232 5 cm ribbon~ 2 g 2 or 3 92-138 mg (Na equiv) as di-ethylamine salt
Predel 2013b Diclofenac 116 2 g 4 92 mg (Na equiv) as diethy-lamine salt
Predel 2013a Diclofenac 4 4-5 sprays 3 96-120 mg Na salt
Costantino 2011 Diclofenac 1 Plaster 1 180 mg epolamine salt 140 mgNa salt
Fioravanti 1999 Diclofenac - 5 g 3 195 mg epolamine salt
Gallacchi 1990 Diclofenac 1 2 g 4 80 mg
Gonzaacutelez de Vega2013
Diclofenac 1 6 cm 2g 3 60 mg
Hoffmann 2012 Diclofenac 1 Plaster 1 180 mg epolamine salt 140 mgNa salt
Hofman 2000 Diclofenac 1 2 cm 4 probably 30-40 mg Na salt
Jenoure 1997 Diclofenac 1 Plaster 2 360 mg epolamine salt 280 mgNa salt
Mahler 2003 Diclofenac - 5 g 3 195 mg epolamine salt
NCT01255423 Diclofenac 1 - 4 -
NCT01272934 Diclofenac 1 - 4 -
NCT01272947 Diclofenac 1 - 4 -
Spacca 2005 Diclofenac - 5 g 3 195 mg epolamine salt
Campbell 1994 Ibuprofen 5 4 inch ribbon 4 Assume 800 mg
Dreiser 1988 Ibuprofen 5 4 cm ribbon(10 cm for largerjoints)
4 Assume up-800 mg
Ramesh 1983 Ibuprofen 5 5-10 cm ribbon 3-4 Assume 300-800 mg
Billigmann 1996 Ibuprofen 5 10 cm 4 g gel 3 600 mg
139Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Machen 2002 Ibuprofen 5 - 3 -
Picchio 1981 Ibuprofen 10 - 3 -
Whitefield 2002 Ibuprofen 5 - 3 -
Maziegraveres 2005b Ketoprofen - Plaster 1 100 mg
Maziegraveres 2005a Ketoprofen - Plaster 1 100 mg
Airaksinen 1993 Ketoprofen - 5 g 2 125 mg
Dreiser 1989 Ketoprofen 25 5 cm 2 100 mg
Julien 1989 Ketoprofen 25 5 cm 2 100 mg
Noret 1987 Ketoprofen 25 5 cm 75 g 2 375 mg
Curioni 1985 Ketoprofen No details 2 -
Governali 1995 Ketoprofen 5 2-3 g 3 300-450 mg
Governali 1995 Ketoprofen 1 2-3 g 3 60-90 mg
Gualdi 1987 Ketoprofen - 3-5 cm 2 -
Kockelbergh 1985 Ketoprofen 25 5 cm 2 100 mg
Parrini 1992 Ketoprofen 150 2 g 3 600 mg
Picchio 1981 Ketoprofen 1 - 3 -
Tonutti 1994 Ketoprofen 5 2-3 g 3 300-450 mg
kermark 1990 Indomethacin 1 05-15 mL 3-5 12-60 mg
Aoki 1984 Indomethacin 1 1 g 3-4 30-40 mg
Fujimaki 1985 Indomethacin 1 1 g 3-4 30-40 mg
Sugioka 1984 Indomethacin 1 1 g 3-4 30-40 mg
Aoki 1984 Piroxicam 05 1 g 3-4 15-20 mg
Fujimaki 1985 Piroxicam 05 1 g 3-4 15-20 mg
140Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Russell 1991 Piroxicam 05 5 mg 4 20 mg
Sugioka 1984 Piroxicam 05 1 g 3-4 15-20 mg
Chatterjee 1977 Benzydamine 3 - 3 -
Haig 1986 Benzydamine 3 - 6 -
Linde 1985 Benzydamine 3 - 3 -
Auclair 1989 Niflumic acid 25 10 cm 5 g 3 375 mg
Dreiser 1990 Niflumic acid 25 10 cm 5 g 3 375 mg
Curioni 1985 Ibuproxam gel 10 - 2 -
Curioni 1985 Etofenamate No details - 2 -
Diebshlag 1990 Etofenamate 5 3 g 3 450 mg
Tonutti 1994 Etofenamate 5 2-3 g 3 300-450 mg
Diebshlag 1990 Ketorolac 2 3 g 3 360 mg
Dreiser 1994 Flurbiprofen Patch 2 80 mg
Gualdi 1987 Flunoxaprofen - 3-5 cm 2 -
Hofman 2000 Lysine clonixinate 5 2 cm 4 90 mg
Hosie 1993 Felbinac 3 2 g 3 180 mg
McLatchie 1989 Felbinac 3 3 cm 3 -
Morris 1991 Felbinac 3 1 cm 3 -
Sanguinetti 1989 Felbinac 3 - 3 -
Sinniger 1981 Fentiazac 5 Varied according toinvolved areas
2-3 -
Thorling 1990 Naproxen 10 - 2-6 -
Vecchiet 1991 Meclofenamic acid 5 10 cm 4 g 2 400 mg
equiv equivalent Na sodium
141Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
F E E D B A C K
Query on formulations of topical NSAIDs particularly DMSO from Dr Chrubasik 11 April 2012
Summary
Dr Chrubasik highlighted this letter to the Editor postgradmedorgdoi103810pgm2011092482
DMSO [dimethyl sulphoxide] but also other additives eg nonivamide (which is a capsaicinoid added as drug enhancer) may contributeto the overall effect of topical NSAIDs Nonivamide certainly contributes to the analgesic effect and to adverse events (heat sensationburning pruritus etc) This has not been considered in the Cochrane review by Massey T Derry S Moore RA McQuay HJ TopicalNSAIDs for acute pain in adults but Dr Chrubasik believes should be done otherwise the effect size of the NSAID topicals is favoured
Reply
We have been asked by Dr Chrubasik to comment on a letter (Roth 2011) about the formulations of topical NSAIDs particularly howDMSO and other penetration enhancers can affect efficacy estimates or adverse event reporting in osteoarthritis It was suggested thatthe review of Topical NSAIDs for acute pain in adults did not consider this resulting in a bias towards the topical NSAIDThere are a number of points to be made here
1 Penetration enhancers are used in formulations of topical products to encourage local absorption through the skin and produce ahigh local concentration Topical NSAIDs use penetration enhancers and the result is high local concentration in joints for instancebut low systemic concentrations (Moore 2008) That is how they work Formulation is an important part of medicinal chemistry as awhole not just for topical agents
2 In our analysis of topical NSAIDs we were aware that a range of properties are or have been ascribed to the analgesia resultingfrom application of topical agents and which could contribute to overestimation of treatment effect of topical NSAID These includefeelings of heat or cold and even the act of rubbing itself For that reason we have chosen to include only double-blind studies wherethe placebo agent is identical to the active with the exception of course of the NSAID So heat cold rubbing and penetrationenhancers should be identical as best we can judge That leaves only the NSAID itself to provide any additional analgesic effect andit is that which we measure This is analogous for example to use of acupuncture say where the better studies show no differencebetween ldquotruerdquo acupuncture and ldquoshamrdquo acupuncture performed at nonspecific sites but better than non-treatment controls Theargument that we should only use high quality studies to evaluate evidence about pain interventions is well made
3 Overestimation of analgesic effect because of effects of enhancers themselves would be better made in direct comparisons oftopical and oral NSAIDs where local or even systemic effects would not be balanced in the oral study arm However our reviewconcentrated on placebo-controlled studies and had few studies with active controls Moreover the real test would be in chronicrather than acute conditions with long duration (12 week) outcomes using current best evidence rules (Moore 2010) includingimputation (Moore 2012) In their response to Rothrsquos letter the authors of the original review of products available in the USA showrather similar effect sizes of oral diclofenac and topical diclofenac with different penetration enhancers (Barthel 2011) in such studies
4 The Roth letter sought to differentiate between topical diclofenac preparations based on the penetration enhancers used Thatdifferent formulations may have different effect sizes is a fair point to make Two of the studies in our review of topical NSAIDs inacute conditions used diclofenac sodium 1 gel comparing it with either diclofenac epolamine gel (Gallacchi 1990 50 participants)or lysine clonixinate gel (Hofman 2000 142 participants) no difference between formulations was demonstrated It is difficult tomake any judgement for topical NSAIDs in acute conditions due to the relatively small number and particularly the small size ofstudies We did an analysis by drug and this showed that some topical NSAIDs were consistently beneficial irrespective offormulation while others had little or no efficacy This fits in with some theoretical considerations of molecular architecture andtissue penetration (Moore 2008) In chronic pain where there are larger studies and much more data we have considered formulation(Derry in preparation)
References
Roth SH Letter to the editor The importance of differentiating between topical NSAIDs Postgraduate Medicine 2011123251-2[103810pgm2011092482]Moore RA Derry S McQuay HJ Topical agents in the treatment of rheumatic pain Rheumatic Diseases Clinics of North America200834415-32
142Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Moore RA Eccleston C Derry S Wiffen P Bell RF Straube S et al ACTINPAIN Writing Group of the IASP Special Interest Groupon Systematic Reviews in Pain Relief Cochrane Pain Palliative and Supportive Care Systematic Review Group Editors ldquoEvidencerdquo inchronic pain - establishing best practice in the reporting of systematic reviews Pain 2010150386-9Moore RA Straube S Eccleston C Derry S Aldington D Wiffen P et al Estimate at your peril imputation methods for patientwithdrawal can bias efficacy outcomes in chronic pain trials using responder analyses Pain 2012153265-8Barthel HR Axford-Gatley RA Response to Roth Letter to the Editor Postgraduate Medicine 2011123253-4 [103810pgm2011092483]
Contributors
Feedback from Sigrun ChrubasikAuthors involved with responding Andrew Moore Sheena DerryFeedback Editor Kate Seers
Query submitted by Peter C Goslashtzsche 3 November 2015
Summary
Date of Submission 03-Nov-2015Name Peter C GoslashtzscheEmail Address pcgcochranedkAffiliation Nordic Cochrane CentreRole DirectorComment The authors found that the results were missing from 5900 patients Furthermore there was extreme heterogeneity in theirmeta-analyses eg I square was 92 for the diclofenac trials which were the most common ones and there was extreme funnel plotasymmetry with the largest trials showing the smallest effects (the authors didnrsquot show funnel plots but I constructed one for diclofenac)Moreover the trials were industry funded of relatively poor quality and the authors analysed published data not data from clinicalstudy reports and did not try to obtain all the missing trials and data from the manufacturersThe authors concluded that topical NSAIDs are effective in providing pain relief but also cautioned that the large amounts of unpublisheddata ldquocould influence results in updates of this reviewrdquo They certainly could I believe it is plain wrong to perform meta-analyses on theauthorsrsquo data When I most recently reviewed this area for the BMJ in 2010 I concluded that we donrsquot know whether topical NSAIDsare beneficial (1)1 Goslashtzsche PC NSAIDs Clin Evid (Online) 2010 Jun 28I agree with the conflict of interest statement belowI certify that I have no affiliations with or involvement in any organization or entity with a financial interest in the subject matter ofmy feedback
Reply
Response submitted by authors Sheena Derry and Professor Andrew Moore on 4 November 2015Peter has made these same points elsewhere [PubMed Commons PubPeer gt Cochrane Database Syst Rev October 2015 httpspubpeercompublicationsA9E5BEA36549727357F9FD14CC2537]As a preliminary it is important to stress that the number of trials available in pain has increased over time and newer trials are typicallybetter and larger This means we can move from answering the simple question of whether an intervention works to more importantand relevant questions such as how well the intervention works and for drugs particularly to examine effects of dose and formulationFormulation is now recognised to have profound effects For oral analgesics for example fast acting formulations demonstrate up todouble the analgesic effect for a given dose as our recently updated overview points out (Cochrane Database Syst Rev 2015 Sep 289CD008659 see also Pain 2014 Jan155(1)14-21)What was true for oral analgesics in acute pain is now true also for topical NSAIDs in acute pain where manufacturers have beenputting a lot of effort into trying to produce new formulations that work better This updated review sought to examine the influenceof formulation and drug although dose is somewhat more difficult There are good reasons why formulation may be important(Rheumatic Disease Clinics 2008 Vol 34 Issue 2 p 415-432)
143Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Our searches did identify a large number of unpublished trials We deplore this but are powerless to change things For a previousreview of topical NSAIDs we contacted all identified manufacturers and asked for published or unpublished data The yield was small(as others have found) but some unpublished studies were brought into the public domain Waiting for all studies and clinical trialreports of all studies would take forever and could probably never be achieved We believe that many of these unpublished studiesrelate to drugs andor formulations that have never been manufactured commercially While these would be of interest in determiningwhat does and doesnrsquot work and to direct future research they would probably have little clinical relevance because these formulationsare unlikely to come to marketPeterrsquos main issue appears to be heterogeneity Tests for heterogeneity are problematical anyway (Pain 2000 85415-24) and the I squareof 92 that he quotes was for all topical diclofenac formulations combined While we do give an overall summary for diclofenac wedemonstrated in the review that the different diclofenac formulations produced different results from one another using LrsquoAbbeacute plotsand in our detailed analyses showing large variations in efficacy between them In the circumstance a high I square for all combined(clinical heterogeneity) is to be expected but it is not relevant The bulk of the studies on diclofenac were published in the last fiveyears were of decent quality and moderate to large size There were older data for ketoprofen but again these showed major differencesbetween formulations Differences between formulations are highlighted throughout the reviewTrying to determine publication bias using funnel plots or other measures is something of a lost cause (Journal of Clinical Epidemiology200053207-16) It is especially so with small numbers of trials (Journal of Clinical Epidemiology 2000 53 477-484) and makingsense of funnel plots is anything but easy for most people (Journal of Clinical Epidemiology 2005 58 894-901) A useless methodseems an odd choice to make to criticise our reviewThere are very good scientific reasons why drug and formulation may play a big part in the efficacy of topical NSAIDs This is alsothe case for oral analgesics used in acute pain where formulation improvements generating rapid absorption confers greater efficacySimple lumping strategies may have been permissible in past systematic review methodology but a more forensic approach is needednow and in the future This is what we have attempted to do in this latest reviewComparisons with Peterrsquos 2010 review seem inappropriate since that review was based on other reviews that are now out of date
Contributors
Feedback from Peter C GoslashtzscheAuthors involved with responding Andrew Moore Sheena DerryFeedback Editor Kate Seers
W H A T rsquo S N E W
Last assessed as up-to-date 4 February 2015
Date Event Description
4 November 2015 Feedback has been incorporated Feedback has been incorporated See Feedback 2
H I S T O R Y
Protocol first published Issue 4 2008
Review first published Issue 6 2010
144Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Date Event Description
13 October 2015 Amended Small errors found in Summary Table B for total num-ber of participants with piroxicam and indomethacinand in text for NNT with diclofenac (other gel) Nochange to conclusions
6 October 2015 Amended Small error found in Summary Table B percentagesfor NSAID and placebo were the wrong way round forcomparison of all NSAIDs and placebo RR and textwere correct No change to results or conclusions
13 February 2015 New citation required but conclusions have notchanged
Conclusions not changed Results remain essentiallythe same but the focus has changed to examine drugand formulation combination
3 February 2015 New search has been performed Title changed from Topical NSAIDs for acute pain inadults to Topical NSAIDs for acute musculoskeletalpain in adults to increase specificityNew searches run and new studies identified Fourteennew included studies using diclofenac (3489 new par-ticipants a 63 increase) four new excluded studiesFifteen studies awaiting classification (completed noresults available but likely to satisfy inclusion criteria)
23 May 2014 Amended Error in data reported for clinical success in Hosie 1993was brought to our attention and has been corrected
12 June 2012 Feedback has been incorporated We have incorporated feedback received from DrSigrun Chrubasik and the authorrsquos response on DMSOand other additives
C O N T R I B U T I O N S O F A U T H O R S
For the earlier review Tom Massey and SD identified studies and carried out data extraction analysis and drafting RAM and HJMwere involved in planning acted as adjudicators and were involved with writing
For this update SD and RAM carried out searches data extraction and analysis All authors were involved with writing the review
145Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
D E C L A R A T I O N S O F I N T E R E S T
SD has no conflicts relating to this review or any similar product
RAM has no conflicts relating to this review or any similar product
PW has no conflicts relating to this review or any similar product
HG has no conflicts relating to this review or any similar product
MM has no conflicts relating to this review or any similar product
S O U R C E S O F S U P P O R T
Internal sources
bull Oxford Pain Relief Trust UKGeneral institutional support
External sources
bull No sources of support supplied
D I F F E R E N C E S B E T W E E N P R O T O C O L A N D R E V I E W
For this update in 2015 we have changed the title to specify musculoskeletal pain because topical NSAIDs are not normally used totreat visceral pain or headache We felt that the new title better reflected the content of the review We have also changed the focus ofthe review from pooled analysis of all topical NSAIDs and all studies of a particular NSAID to an examination of individual drug andits formulation This makes the review much more relevant We have expanded the rsquoRisk of biasrsquo assessment and added a rsquoSummaryof findingsrsquo table and PRISMA flow chart We have removed a number of sensitivity analyses because they were not appropriate giventhe current information on the impact of formulation on efficacy The sensitivity analyses have been superseded by the rsquoRisk of biasrsquoassessment and taken into account in the rsquoSummary of findingsrsquo tables
An earlier review in 2004 chose to exclude studies using benzydamine on the grounds that it was no longer considered to be an NSAID(Mason 2004a) Although the protocol for this review stated that we would not include benzydamine after further consultation wenow believe that it should be classified as an NSAID albeit with a different mode of action which is not fully understood (Quane1998) Thus we have reinstated studies using topical benzydamine
I N D E X T E R M S
Medical Subject Headings (MeSH)
Acute Disease Administration Topical Anti-Inflammatory Agents Non-Steroidal [lowastadministration amp dosage adverse effects] AthleticInjuries [drug therapy] Pain [lowastdrug therapy] Randomized Controlled Trials as Topic Sprains and Strains [drug therapy]
146Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
MeSH check words
Adult Humans
147Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
[Intervention Review]
Topical NSAIDs for acute musculoskeletal pain in adults
Sheena Derry1 R Andrew Moore1 Helen Gaskell2 Mairead McIntyre3 Philip J Wiffen1
1Pain Research and Nuffield Department of Clinical Neurosciences (Nuffield Division of Anaesthetics) University of Oxford OxfordUK 2Department of Clinical Geratology Oxford University Hospitals NHS Trust John Radcliffe Hospital Oxford UK 3West HoeSurgery Plymouth UK
Contact address Sheena Derry Pain Research and Nuffield Department of Clinical Neurosciences (Nuffield Division of Anaesthetics)University of Oxford Pain Research Unit Churchill Hospital Oxford Oxfordshire OX3 7LE UK sheenaderryndcnoxacuk
Editorial group Cochrane Pain Palliative and Supportive Care GroupPublication status and date Edited (no change to conclusions) comment added to review published in Issue 11 2015Review content assessed as up-to-date 4 February 2015
Citation Derry S Moore RA Gaskell H McIntyre M Wiffen PJ Topical NSAIDs for acute musculoskeletal pain in adults CochraneDatabase of Systematic Reviews 2015 Issue 6 Art No CD007402 DOI 10100214651858CD007402pub3
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
A B S T R A C T
Background
Use of topical NSAIDs to treat acute musculoskeletal conditions has become widely accepted because they can provide pain reliefwithout associated systemic adverse events This review is an update of rsquoTopical NSAIDs for acute pain in adultsrsquo originally publishedin Issue 6 2010
Objectives
To determine the efficacy and safety of topically applied NSAIDs in acute musculoskeletal pain in adults
Search methods
We searched the Cochrane Register of Studies Online MEDLINE and EMBASE to February 2015 We sought unpublished studies byasking personal contacts and searching online clinical trial registers and manufacturers websites For the earlier review we also searchedour own in-house database and contacted manufacturers
Selection criteria
We included randomised double-blind active or placebo (inert carrier)-controlled trials in which treatments were administered toadults with acute pain resulting from strains sprains or sports or overuse-type injuries (twisted ankle for instance) There had to be atleast 10 participants in each treatment arm with application of treatment at least once daily
Data collection and analysis
Two review authors independently assessed studies for inclusion and extracted data We used numbers of participants achieving eachoutcome to calculate the risk ratio and numbers needed to treat for an additional beneficial outcome (NNT) or additional harmfuloutcome (NNH) compared with placebo or other active treatment We reported 95 confidence intervals (CI) We were particularlyinterested to compare different formulations (gel cream plaster) of individual NSAIDs
1Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Main results
For this update we added 14 new included studies (3489 participants) and excluded four studies We also identified 20 additionalreports of completed or ongoing studies that have not been published in full The earlier review included 47 studies
This update included 61 studies Most compared topical NSAIDs in the form of a gel spray or cream with a similar topical placebo5311 participants were treated with a topical NSAID 3470 with placebo and 220 with an oral NSAID This was a 63 increase inthe number of included participants over the previous version of this review We also identified a number of studies in clinical trialregistries with unavailable results amounting to about 5900 participants for efficacy and 5300 for adverse events
Formulations of topical diclofenac ibuprofen ketoprofen piroxicam and indomethacin demonstrated significantly higher rates ofclinical success (more participants with at least 50 pain relief ) than matching topical placebo (moderate or high quality data)Benzydamine did not Three drug and formulation combinations had NNTs for clinical success below 4 For diclofenac the Emulgelregformulation had the lowest NNT of 18 (95 CI 15 to 21) in two studies using at least 50 pain intensity reduction as the outcomeDiclofenac plasters other than Flectorreg also had a low NNT of 32 (26 to 42) based on good or excellent responses in some studiesKetoprofen gel had an NNT of 25 (20 to 34) from five studies in the 1980s some with less well defined outcomes Ibuprofen gelhad an NNT of 39 (27 to 67) from two studies with outcomes of marked improvement or complete remission All other drug andformulation combinations had NNT values above 4 indicating lesser efficacy
There were insufficient data to compare reliably individual topical NSAIDs with each other or the same oral NSAID
Local skin reactions were generally mild and transient and did not differ from placebo (high quality data) There were very few systemicadverse events (high quality data) or withdrawals due to adverse events (low quality data)
Authorsrsquo conclusions
Topical NSAIDs provided good levels of pain relief in acute conditions such as sprains strains and overuse injuries probably similar tothat provided by oral NSAIDs Gel formulations of diclofenac (as Emugelreg) ibuprofen and ketoprofen and some diclofenac patchesprovided the best effects Adverse events were usually minimal
Since the last version of this review the new included studies have provided additional information In particular information ontopical diclofenac is greatly expanded The present review supports the previous review in concluding that topical NSAIDs are effectivein providing pain relief and goes further to demonstrate that certain formulations mainly gel formulations of diclofenac ibuprofenand ketoprofen provide the best results Large amounts of unpublished data have been identified and this could influence results inupdates of this review
P L A I N L A N G U A G E S U M M A R Y
Topical non-steroidal anti-inflammatory drugs for acute musculoskeletal pain in adults
Acute musculoskeletal pain describes conditions like a sprained ankle or a muscle pull These usually get better over two or three weekswithout treatment but can be very painful while they last
Topical non-steroidal anti-inflammatory drugs (NSAIDs) are applied to unbroken skin where it hurts as gels creams sprays or plastersTopical NSAIDs penetrate the skin enter tissues or joints and reduce processes causing pain in the tissue Drug levels in the bloodwith topical NSAIDs are very much lower than with the same drug taken by mouth This minimises the risk of harmful effects
We searched medical databases for clinical trials comparing topical NSAIDs with placebo (creams or gels that do not contain a medicine)or other medicines in adults aged 16 years or older with musculoskeletal pain (typically sports injuries) The evidence is current toFebruary 2015
This review is an update of rsquoTopical NSAIDs for acute pain in adultsrsquo originally published in Issue 6 2010 We identified 14 new studiesto add to the 47 studies included in the earlier review We also identified 14 studies in a clinical trial registry that are completed andthree short reports from meetings for which we could not find full details (about 4500 participants) Three more studies are ongoing(almost 900 participants)
The 61 included studies involving 8386 participants were generally of high-quality They tested a number of different topical drugsmostly against a topical placebo (carrier without the NSAID) with application at least once a day We were interested in participants
2Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
having good pain reduction (by about half ) around seven days after treatment started At later times most people are expected to getbetter even without treatment
We looked at particular formulations of individual drugs Gel formulations of diclofenac and ketoprofen were among the most effectivealong with ibuprofen gel and diclofenac plaster For diclofenac and ketoprofen gels 7 or 8 people out of 10 with a painful strain sprainor muscle pull had much reduced pain after seven days compared with only 2 or 3 out of 10 with placebo (high quality data) OtherNSAIDs and formulations were better than placebo but not by as much Because both topical NSAIDs and topical placebo are rubbedinto the skin in these studies we know that any effect is not just from rubbing
About 1 in 20 people experienced a mild and short-lived side effect like redness at the application site This was the same for topicalNSAID and topical placebo (high quality data) Side effects like a stomach upset or feeling sick were uncommon with no differencebetween topical NSAID and topical placebo (high quality data) There were no serious side effects
3Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
S U M M A R Y O F F I N D I N G S F O R T H E M A I N C O M P A R I S O N [Explanation]
Topical NSAIDs compared with topical placebo for acute musculoskeletal pain in adults
Patient or population adults with strains sprains or muscle pullSettings communityIntervention topical NSAID (topical diclofenac ibuprofen and ketoprofen gels only shown here for efficacy)Comparison topical placebo
Outcomes Probable outcome withintervention
Probable outcome withcomparator
RR NNT NNTp or NNH(95 CI)
No of studies partici-pants
Quality of the evidence(GRADE)
Comments
Topical diclofenac gel
(as Emulgel)
Clinical success (eg50 reduction in pain)
780 in 1000 200 in 1000 RR34 (27 to 55)NNT18 (15 to 21)
2 studies314 participants
High Consistent results in 2moderately sized recentstudies of high quality
Topical ibuprofen gel
Clinical success (eg
50 reduction in pain)
420 in 1000 160 in 1000 RR27 (17 to 42)NNT39 (27 to 67)
2 studies241 participants
Moderate Modest effect size andnumbers of participants
Topical ketoprofen gel
Clinical success (eg
50 reduction in pain)
720 in 1000 330 in 1000 RR22 (17 to 28)NNT25 (20 to 34)
5 studies348 participants
Moderate Modest effect size andnumbers of participantsbut studies small withnone recent
All topical NSAIDs
Local adverse events
46 in 1000 50 in 1000 RR10 (080 to 12)NNH not calculated
42 studies6125 participants
High Large number of studiesand participants with con-sistent results
All topical NSAIDs
Systemic adverse
events
32 in 1000 35 in 1000 RR10 (07 to 13)NNH not calculated
38 studies5372 participants
High Large number of studiesand participants with con-sistent results
All topical NSAIDs
Withdrawals - adverse
events
11 in 1000 11 in 1000 RR10 (07 to 17)NNH not calculated
42 studies5790 participants
High Large number of studiesand participants with con-sistent results4
To
pica
lN
SA
IDs
for
acu
tem
uscu
losk
ele
talp
ain
inad
ults
(Revie
w)
Co
pyrig
ht
copy2015
Th
eC
och
ran
eC
olla
bo
ratio
nP
ub
lished
by
Joh
nW
iley
ampS
on
sL
td
Serious adverse events 1 in total 0 in total Not calculated All data Low Small numbers of events
GRADE Working Group grades of evidenceHigh quality Further research is very unlikely to change our confidence in the estimate of effectModerate quality Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimateLow quality Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimateVery low quality We are very uncertain about the estimate
CI confidence interval RR risk ratio NNT number needed to treat for an additional beneficial outcome NNTp number needed to treatto prevent an event happening NNH number needed to treat for an additional harmful outcome
xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
5To
pica
lN
SA
IDs
for
acu
tem
uscu
losk
ele
talp
ain
inad
ults
(Revie
w)
Co
pyrig
ht
copy2015
Th
eC
och
ran
eC
olla
bo
ratio
nP
ub
lished
by
Joh
nW
iley
ampS
on
sL
td
B A C K G R O U N D
This review is an update of a review originally published in Issue 62010 on rsquoTopical NSAIDs for acute pain in adultsrsquo (Massey 2010)We have changed the title to specify musculoskeletal pain becausetopical NSAIDs are not normally used to treat visceral pain orheadache We felt that the new title better reflected the content ofthe review
The use of topical NSAIDs for pain relief has been a controversialsubject in analgesic practice In some parts of the world (includingmuch of Western Europe) they have been available for many yearsare widely available without prescription widely advertised usedextensively and evidence for their use is considered adequate Inother parts of the world they were regarded as little more thanplacebo with any apparent effect attributed to the process of rub-bing at the site of the affected area In some places (for examplethe US) their use was almost unknown until the mid-2010s InEngland 52 million prescriptions for topical NSAIDs were dis-pensed in 2013 (PACT 2014) mainly for formulations of ibupro-fen (245 million) piroxicam (118 million) and diclofenac (127million)
There is good evidence for the efficacy of topical NSAIDs in acuteand chronic musculoskeletal pain (Mason 2004a Mason 2004bMoore 1998a) In the US the Food and Drug Administration li-censed topical nonsteroidal products in 2007 and in England theNational Institute for Health and Care Excellence (NICE) recom-mended topical therapies as first line treatment in its guidelinesfor osteoarthritis in 2008 (NICE 2008) Earlier reviews of topicalanalgesics covered studies investigating the underlying science toexplain biological plausibility in addition to clinical trials (Anon2005 Moore 2008a)
This review is one of a series on topical analgesics including topicalcapsaicin at low and high doses (Derry 2012a Derry 2013) andtopical NSAIDs in chronic pain conditions (Derry 2012b) andsalicylate-containing rubefacients (Derry 2014)
Description of the condition
Acute pain is usually defined as pain of less than three monthsrsquo du-ration It is often associated with injury including trauma surgerymusculoskeletal injuries such as strains sprains and over-use in-juries or soft tissue injuries such as muscle soreness or cramps
Description of the intervention
Clinicians prescribe NSAIDs on a routine basis for a range of mildto moderate pain NSAIDs are the most commonly prescribedanalgesic medications worldwide and their efficacy for treatingacute pain has been well demonstrated (Moore 2003) They re-versibly inhibit cyclooxygenase (prostaglandin endoperoxide syn-thase) the enzyme mediating production of prostaglandins and
thromboxane A2 (FitzGerald 2001) Prostaglandins mediate a va-riety of physiological functions such as maintenance of the gastricmucosal barrier regulation of renal blood flow and regulation ofendothelial tone They also play an important role in inflamma-tory and nociceptive processes However relatively little is knownabout the mechanism of action of this class of compounds asidefrom their ability to inhibit cyclooxygenase-dependent prostanoidformation (Hawkey 1999)NSAIDs taken orally or intravenously are transported to all partsof the body in the blood and relatively high blood concentrationsare needed to achieve effective tissue concentrations at the site ofthe pain and inflammation These high concentrations throughoutthe body can give rise to a number of adverse events that canbe unpleasant (for example dyspepsia) or potentially serious (forexample gastrointestinal bleeding)A topical medication is a one applied to body surfaces such asthe skin or mucous membranes to treat ailments A large rangeof types of topical formulation may be used including but notlimited to creams foams gels lotions ointments and plastersThe exact formulation of a topical medication is often determinedby the speed of drug absorption required The need may be forslow absorption into the circulation to maintain low drug concen-trations and perhaps avoiding extensive first pass metabolism inthe liver plasters containing drug reservoirs may be used for thisas with transdermal opioids or contraceptive steroids For rapidabsorption the formulation is enhanced by substances to improveor assist skin penetration perhaps only to generate high concen-trations in tissues rather than in the blood gel formulations areuseful for this purpose which is why they are sometimes used fortopical NSAIDs
Topical NSAIDs
Topical NSAIDs are formulated for direct application to thepainful site and to produce a local pain-relieving effect whileavoiding body-wide distribution of the drug at physiologically ac-tive levels (McPherson 2013) This method of application (dos-ing) necessarily limits their use to more superficial painful condi-tions such as sprains strains and muscle or tendon soreness Theywould not for example be indicated for deep visceral pain orheadaches They are also not appropriate for use on broken skinso would not be used on open wounds (accidental or surgical)
How the intervention might work
For a topical formulation to be effective it must first penetratethe skin Only when the drug has entered the lower layers ofthe skin can it be absorbed by the blood and transported to thesite of action or penetrate deeper into areas where inflammationoccurs Individual drugs have different degrees of penetration Abalance between lipid and aqueous solubility is needed to optimisepenetration and use of prodrug esters has been suggested as a way
6Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
of enhancing permeability Formulation is also crucial to goodskin penetration Experiments with artificial membranes or humanepidermis suggest that creams are generally less effective than gelsor sprays but newer formulations such as microemulsions mayhave greater potentialOnce the drug has reached the site of action it must be presentat a sufficiently high concentration to inhibit cyclooxygenase en-zymes thereby reducing prostaglandin synthesis This in turn re-duces inflammation and relieves pain It is probable that in acuteconditions topical NSAIDs exert their action primarily by lo-cal reduction of symptoms independent of any systemic uptakeand delivery Tissue levels of NSAIDs applied topically certainlyreach levels high enough to inhibit cyclooxygenase-2 (Anon 2005Haroutiunian 2010 Moore 2008a) However plasma concentra-tions found after topical administration are only a fraction (usu-ally much less than 5) of the levels found in plasma followingoral administration Topical application can potentially limit sys-temic adverse events by increasing local effects and minimisingsystemic concentrations of the drug We know that the incidenceof upper gastrointestinal bleeding is low with chronic use of topi-cal NSAIDs (Evans 1995) although it has been reported partic-ularly in people with risk factors (Zimmerman 1995) We haveno certain knowledge of lower effects on cardiovascular events orrenal failure both of which have been associated with oral NSAIDuse
Why it is important to do this review
Since the last review in 2010 a number of new studies have beenpublished nearly all of which investigated various formulations ofdiclofenac These new studies are generally of higher quality thanmany of the earlier ones in this review and have the potential toinfluence the strength of its conclusions substantially Moreoverthe additional information allows for analysis based not only on aparticular drug but also on the formulation of that drug This canprovide better insight into whether formulation affects efficacy oftopical NSAIDs in acute musculoskeletal painAn updated review of evidence for topical NSAIDs was needed toinform choices made by consumers prescribers and commission-ers (purchasers of healthcare)
O B J E C T I V E S
To determine the efficacy and safety of topically applied NSAIDsin acute musculoskeletal pain in adults
M E T H O D S
Criteria for considering studies for this review
Types of studies
Randomised controlled double-blind studies comparing topicalNSAIDs with placebo (inert carrier) or other active treatment foracute pain with at least 10 participants per treatment arm andoutcomes close to seven days (minimum three days) We excludedstudies published only as short abstracts (which report insufficientdata to assess methods) or studying experimentally induced pain(which does not correlate well with clinical pain) Because a cross-over design is not appropriate for self limiting conditions such assprains strains and contusions we only considered parallel-groupdesigns
Types of participants
Adults (aged 16 years or more) with acute musculoskeletal pain ofat least moderate intensity resulting mainly from strains sprainsor sports injuries Typically for sports injuries the injury wouldhave occurred within 24 or 48 hours
Types of interventions
Included studies had at least one treatment arm using a topicalNSAID and a comparator arm using placebo (inert carrier withoutNSAID or other active treatment) The topical NSAID had to beapplied at least once daily We did not include salicylates in thisreview as they are no longer classified as topical NSAIDs and arecovered in a separate review (Derry 2014)
Types of outcome measures
We sought information on participant characteristics includingage sex and condition treated
Primary outcomes
The primary outcome was rsquoclinical successrsquo defined as at least a50 reduction in pain or equivalent measure such as a rsquovery goodrsquoor rsquoexcellentrsquo global assessment of treatment or rsquononersquo or rsquoslightrsquopain on rest or movement measured on a categorical scale (Moore1998a) We used the following hierarchy of outcomes to extractdata for the primary outcome
bull Participant reported reduction in pain of at least 50bull Participant reported global assessment of treatmentbull Pain on movementbull Pain at rest or spontaneous painbull Undefined rsquoimprovementrsquo
We used only participant reported outcomes of efficacy and notphysician or investigator reported outcomes
7Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Secondary outcomes
bull Numbers of participants with adverse events local andsystemic
bull Numbers of withdrawals all cause lack of efficacy andadverse events
We anticipated that outcomes would be reported after differentdurations of treatment and extracted data reported as close toseven days as possible with a minimum of three days We also ex-tracted data for outcomes reported after longer durations of treat-ment We anticipated that reporting of adverse events would varybetween studies with regard to the terminology used method ofascertainment and categories reported (for example occurring inat least 5 of participants or where there is a statistically signifi-cant difference between treatment groups) We took care to iden-tify these details where relevant
Search methods for identification of studies
Electronic searches
We searched the following databases without language restrictionbull CENTRAL (The Cochrane Library) Issue 4 2009 for the
original review and the Cochrane Register of Studies Online(CRSO) to 3 February 2015 for this update
bull MEDLINE (via Ovid) from inception to December 2009for the original review and from 2008 to 3 February 2015 forthis update
bull EMBASE (via Ovid) from inception to December 2009for the original review and from 2008 to 3 February 2015 forthis update
bull Oxford Pain Relief Database for the original review (Jadad1996a) This resource is no longer being updated
See Appendix 1 for the CENTRAL search strategy Appendix 2 forthe MEDLINE search strategy and Appendix 3 for the EMBASEsearch strategy
Searching other resources
We searched the reference lists of review articles and includedstudies We have previously asked manufacturers for details ofunpublished studies but did not make new requestsWe searched two clinical trial registries (clinicaltrialsgov (clinicaltrialsgov) and the World Health Organization In-ternational Clinical Trials Registry Platform (appswhointtrialsearch)) and asked personal contacts about ongoing and un-published studies
Data collection and analysis
Selection of studies
We screened the titles and abstracts of studies identified by thesearches to eliminate those that clearly did not satisfy the inclu-sion criteria and obtained full reports of the remaining studies todetermine inclusion in the review
Data extraction and management
Review authors were not blinded to the authorsrsquo names and in-stitutions journal of publication or study results at any stage ofthe review Two review authors independently selected the stud-ies for inclusion assessed methodological quality and risk of biasand extracted data We resolved disagreements and uncertaintiesthrough discussionWe abstracted information on participants interventions and out-comes from the original reports into a standard data extractionform One review author entered data suitable for meta-analysisinto Review Manager 5 (RevMan 2014) and another review au-thor checked it
Assessment of risk of bias in included studies
We used the Oxford Quality Score as the basis for inclusion lim-iting inclusion to studies that were randomised and double-blindas a minimum (Jadad 1996b)Two review authors independently assessed risk of bias for eachstudy using the criteria outlined in the Cochrane Handbook forSystematic Reviews of Interventions with any disagreements resolvedby discussion (Higgins 2011) We assessed the following for eachstudy
bull Random sequence generation (checking for possibleselection bias) We assessed the method used to generate theallocation sequence as low risk of bias (any truly randomprocess for example random number table computer randomnumber generator) unclear risk of bias (method used to generatesequence was not clearly stated) We excluded studies using anon-random process that were therefore at high risk of bias (forexample odd or even date of birth hospital or clinic recordnumber)
bull Allocation concealment (checking for possible selectionbias) The method used to conceal allocation to interventionsbefore assignment determines whether intervention allocationcould have been foreseen in advance of or during recruitmentor changed after assignment We assessed the methods as lowrisk of bias (for example telephone or central randomisationconsecutively numbered sealed opaque envelopes) unclear riskof bias (method was not clearly stated) We excluded studies thatdid not conceal allocation and were therefore at high risk of bias(for example open list)
8Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
bull Blinding of outcome assessment (checking for possibledetection bias) We assessed the methods used to blind studyparticipants and outcome assessors from knowledge of whichintervention a participant received We assessed the methods aslow risk of bias (study stated that it was blinded and describedthe method used to achieve blinding for example identicaltubes containing gel or identical plasters matched in appearanceand smell) unclear risk of bias (study stated that it was blindedbut did not provide an adequate description of how blinding wasachieved) We excluded studies that were not double-blind andwere therefore at high risk of bias
bull Incomplete outcome data (checking for possible attritionbias due to the amount nature and handling of incompleteoutcome data) We assessed the methods used to deal withincomplete data as low risk of bias (less than 10 ofparticipants did not complete the study or used rsquobaselineobservation carried forwardrsquo analysis or both) unclear risk ofbias (used rsquolast observation carried forwardrsquo analysis) or high riskof bias (used rsquocompleterrsquo analysis)
bull Size (checking for possible biases confounded by small size)Small studies have been shown to overestimate treatment effectsprobably due to methodological weaknesses (Dechartres 2013Nuumlesch 2010) We assessed studies as at low risk of bias if theyhad at least 200 participants at unclear risk if they had 50 to 200participants and at high risk if they had fewer than 50participants
Measures of treatment effect
We used risk ratio (RR) to establish statistical difference and num-bers needed to treat for an additional beneficial outcome (NNT)with 95 confidence intervals (CI) We pooled percentages as ab-solute measures of benefit or harmWe used the following terms to describe adverse outcomes in termsof harm or prevention of harm
bull When significantly fewer adverse outcomes occurred withtreatment than with control (placebo or active) we used the termthenumber needed to treat to prevent one event (NNTp)
bull When significantly more adverse outcomes occurred withtreatment compared with control (placebo or active) we usedthe term the number needed to treat for an additional harmfuloutcome or cause one event (NNH)
Unit of analysis issues
Randomisation was to the individual participant
Dealing with missing data
Wherever possible we used intention-to-treat (ITT) analysis wherethe ITT population consists of participants who were randomisedapplied at least one dose of the assigned study medication andprovided at least one post-baseline assessment We assigned miss-ing participants zero improvement
We also looked for information about methods of imputation formissing data
Assessment of heterogeneity
We examined heterogeneity visually using LrsquoAbbeacute plots (LrsquoAbbeacute1987) a visual method for assessing differences in results of indi-vidual studies and with the I2 statistic
Assessment of reporting biases
The aim of this review was to use dichotomous outcomes of knownutility and of value to patients (Moore 2013) The review did notdepend on what the authors of the original studies chose to reportor not Studies that did not report dichotomous results but onlyaverage pain data did not contribute to analysesWe assessed publication bias using a method designed to detect theamount of unpublished data with a null effect required to makeany result clinically irrelevant (usually taken to mean an NNT of10 or higher Moore 2008b)
Data synthesis
We pooled data only for comparisons and outcomes where therewere at least two studies and 200 participants (Moore 1998b)When two active treatment arms were compared with a placeboarm we took care to avoid double counting of participants in theplacebo arm if both active groups contributed to an analysis wesplit the placebo group between themWe calculated RRs with 95 CIs using the fixed-effect model(Morris 1995) A statistically significant benefit of topical NSAIDover control was assumed when the lower limit of the 95 CI ofthe RR was greater than one A statistically significant benefit ofcontrol over active treatment was assumed when the upper limitof the 95 CI was less than one We calculated NNTs with 95CIs using the pooled number of events by the method of Cookand Sackett (Cook 1995)Statistically significant differences between NNTs for differenttopical NSAIDs were tested using the z test (Tramer 1997) wherethere were sufficient data to do so and where the studies weresufficiently similar in types of participant outcome and durationto make such comparisons sensible
Subgroup analysis and investigation of heterogeneity
We carried out separate analyses for individual NSAIDs andwhere the data permitted for different formulations of individualNSAIDs
Sensitivity analysis
The earlier review included sensitivity analyses for various factorsthat are now covered by the assessment of risk of bias
9Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
R E S U L T S
Description of studies
Results of the search
New searches for this update identified 20 publications that wereexamined in further detail to determine inclusion status We alsoidentified 20 additional studies in clinical trial registries See Figure1
10Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Figure 1 Study flow diagram
11Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
We identified 14 new studies (11 publications three unpub-lished reports) satisfying our inclusion criteria (Costantino2011 Coudreuse 2010 Gonzaacutelez de Vega 2013 Hofman2000 Klainguti 2010 Kuehl 2011 Li 2013 NCT01255423NCT01272934 NCT01272947 Predel 2012 Predel 2013aPredel 2013b Saillant 1998) one post hoc analysis of a study thatwas included in the earlier review (Mueller 2010 in Predel 2004)and an additional publication and a pooled analysis that includednew data for another study from the earlier review (Lionberger2011 pooled analysis in Joussellin 2003)
Included studies
All except one of the new studies compared diclofenac withplacebo A number of different formulations were used includ-ing diclofenac epolamine (DHEP) with or without heparin (Flec-toparin Tissugel or Flector EP Tissugel) applied as a plaster (orpatch) diclofenac diethylamine (DDEA) applied as a gel and di-clofenac with lethicin applied as a spray gel The remaining studycompared diclofenac gel with traumeel a ldquofixed combination ofplant and mineral extractsrdquo applied as a gel or an ointmentThere were 47 studies in the original review 14 new studies wereincluded making a total of 61 studies in this updated reviewAll used a parallel group design Forty-four compared a topicalNSAID with placebo 13 a topical NSAID with an active com-parator (a different topical NSAID an oral NSAID the same top-ical NSAID in a different formulation or a compound of plantand mineral extracts) and four had both placebo and active com-parators In total 5311 participants were treated with a topicalNSAID 3470 with placebo and 220 with an oral NSAID Top-ical NSAIDs used were benzydamine diclofenac etofenamatefelbinac fentiazac flunoxaprophen flurbiprofen ibuprofen in-domethacin ketoprofen ketorolac lysine clonixinate meclofe-namic acid naproxen niflumic acid and piroxicam They wereapplied as creams gels sprays foams or plasters (patches) Topicalplacebos were the inert carriers without the active NSAID OralNSAIDs used were ibuprofen (as tablets) and indomethacin (ascapsules)Most studies enrolled participants who had sprains strains andcontusions usually as a result of sports injuries and treatmentwas started within a few hours or days Other studies enrolledparticipants with overuse-type injuries such as tendinitis and acutelow back pain where pain had been present for days or weeks butless than three monthsParticipants were treated for at least five days and up to threeweeks with most studies lasting seven to 14 days Participantswere usually assessed in clinic at intervals during treatment andsometimes also at home using daily patient diaries We used out-comes closest to seven days because many of these injuries are self
limiting with differences between active treatment and placebobeing diminished or lost after longer intervalsMost studies reported dichotomous outcomes suitable for a re-sponder analysis although group mean change (for pain or phys-ical function for example) was usually the primary outcomesHowever the definition of response varied both in the parametermeasured (for example pain pain on movement patient globalevaluation of treatment) and in the scale used to measure it (forexample a 3- 4- or 5-point scale for patient global evaluation)Details of included studies are in the Characteristics of includedstudies tableWe identified 14 completed but apparently unpublished studies ina clinical trial registry for which no results have been posted (4403participants NCT00351104 NCT00352625 NCT00426985NCT00640705 NCT00640939 NCT00680472NCT00680784 NCT00765700 NCT00869063NCT00869180 NCT00931866 NCT01874626NCT01957215 NCT02324270) We have placed these underCharacteristics of studies awaiting classification We also identi-fied three conference abstracts that relate to completed studies thatdo not appear to have been published but that may satisfy ourinclusion criteria One is likely to be the same study as one of theincluded studies identified in a clinical trial registry (Pallay 2013in NCT01272947) one relates to another study identified in theclinical trial registry that is awaiting classification (Ekman 2010in NCT00765700) while we could find no published or unpub-lished reports of the other Sarzi-Puttini 2014)We also identified three ongoing studies with an estimated en-rolment of 880 participants (NCT01945034 NCT02100670NCT02290821) Details are in the Characteristics of ongoingstudies table
Excluded studies
For the original review 25 studies were excluded after readingthe full paper For this update we excluded four new studies (Cesarone 2008 Coulibaly 2009 Kuwabara 2013 Vinciguerra2008) Details are in the Characteristics of excluded studies table
Risk of bias in included studies
All studies were randomised and double-blind One study scored25 (Sinniger 1981) 23 scored 35 23 scored 45 and 14 scored55 for methodological quality using the Oxford Quality Scale Abreakdown of the scores for individual studies is reported in theCharacteristics of included studies tableComments on potential biases in individual studies are reported inthe rsquoRisk of biasrsquo section of the Characteristics of included studiestable The findings are displayed in Figure 2 and Figure 3
12Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Figure 2 Risk of bias summary review authorsrsquo judgements about each risk of bias item for each includedstudy
13Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Figure 3 Methodological quality graph review authorsrsquo judgements about each methodological quality
item presented as percentages across all included studies
Allocation
All the studies were randomised but only 17 adequately describedthe method used to generate the random sequence Thirteen stud-ies adequately described the method used to conceal allocation ofthe sequence No studies were at high risk of bias for this item
Blinding
All studies were double-blind and 38 adequately described themethod used to maintain the blinding No studies were at highrisk of bias for this item
Incomplete outcome data
Thirty-six studies included all participants in the primary analysisor provided sufficient data to allow missing participants to be in-cluded as non-responders and were judged at low risk of bias Wejudged four studies to be at high risk of attrition bias (Campbell1994 Kuehl 2011 Maziegraveres 2005a Russell 1991) Three unpub-lished studies contributed only to adverse event analyses and ac-counted for all participants for these outcomes (NCT01255423NCT01272934 NCT01272947)
Other potential sources of bias
We judged one study that included more than 200 participantsin each treatment arm to be at low risk of bias from size butthis study had a very high attrition rate (see rsquoIncomplete outcome
data (attrition bias)rsquo) and did not report all the efficacy outcomesmeasured (Kuehl 2011) We judged 27 studies to be at high riskbecause they included fewer than 50 participants per treatmentarm
Effects of interventions
See Summary of findings for the main comparison TopicalNSAIDs compared with topical placebo for acute musculoskeletalpain in adultsThree studies did not contribute data suitable for analysis of at leastone outcome (Gonzaacutelez de Vega 2013 Gualdi 1987 Hoffmann2012)
1 Topical NSAID versus placebo
Details of efficacy outcomes in individual studies are in Appendix4 and of adverse events and withdrawals in Appendix 5 Appendix6 has details of the concentration of topical products the amountapplied the frequency of application and an estimation of thedaily dose of topical NSAID applied Not all studies providedsufficient information to allow calculation of daily dose appliedFor example for topical diclofenac the estimated doses appliedvaried between about 60 and 280 mg for topical ketoprofen 100to about 450 mg for topical ibuprofen 300 to 800 mg
Participants with clinical success
14Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Topical diclofenac versus placebo
Ten studies contributed to this analysis (Coudreuse 2010Joussellin 2003 Klainguti 2010 Li 2013 Predel 2004 Predel2012 Predel 2013a Predel 2013b Rowbotham 2003) of whichone (Predel 2012) had two active treatment arms A total of 1074participants were treated with topical diclofenac and 976 withplacebo (Analysis 11 Figure 4)
15Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Figure 4 Forest plot of comparison 2 Individual NSAID versus placebo outcome 21 Clinical success
16Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
bull The proportion of participants experiencing successfultreatment with topical diclofenac was 74 (8001074 range39 to 100)
bull The proportion of participants experiencing successfultreatment with placebo was 47 (461976 range 8 to 82)
bull The RR for treatment compared with placebo was 16(95 CI 15 to 17)
bull The NNT for successful treatment was 37 (32 to 43) Forevery four participants treated with topical diclofenac one wouldexperience successful treatment who would not have done sowith placebo
Effect of formulation
The effects of formulation are shown in Analysis 21 and Figure5
17Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Figure 5 LrsquoAbbeacute plot of clinical success in studies of topical diclofenac versus topical placebo The size ofthe symbol is proportional to the size of the study (inset scale) Dark blue Emulgel light blue spraygel red
Flector pink other patch or plaster
bull Four studies used a Flectorreg plaster (1030 participantsJoussellin 2003 Li 2013 Rowbotham 2003 Saillant 1998) TheRR for treatment compared with placebo was 15 (14 to 17)and the NNT was 47 (37 to 65)
bull Three studies used other makes of plaster (474 participantsCoudreuse 2010 Klainguti 2010 Predel 2004) The RR fortreatment compared with placebo was 16 (14 to 18) and theNNT was 32 (26 to 42)
bull Two studies used Voltaren Emulgel (314 participants Predel2012 Predel 2013b) The RR for treatment compared withplacebo was 38 (27 to 55) and the NNT was 18 (15 to 21)
bull One study used a spray gel (232 participants Predel2013a) The RR for treatment compared with placebo was 12(105 to 13) and the NNT was 80 (48 to 24)
Diclofenac as the gel formulation Emulgel was statistically more ef-ficacious than the plaster formulation as Flector plaster (z = 6360P value lt 000001)
Topical ibuprofen versus placebo
Five studies contributed to this analysis (Billigmann 1996Campbell 1994 Dreiser 1988 Machen 2002 Ramesh 1983) Atotal of 218 participants were treated with topical ibuprofen and218 with placebo (Analysis 11)
bull The proportion of participants experiencing successfultreatment with topical ibuprofen was 55 (120218 range 31to 81)
18Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
bull The proportion of participants experiencing successfultreatment with placebo was 33 (73218 range 13 to 76)
bull The RR of treatment compared with placebo was 16 (13to 20)
bull The NNT for successful treatment was 46 (33 to 80) Forevery five participants treated with topical ibuprofen one wouldexperience successful treatment who would not have done sowith placebo
Effect of formulation
The effects of formulation are shown in Analysis 31bull Three studies used cream formulations (195 participants
Campbell 1994 Dreiser 1988 Ramesh 1983) Although this isjust below our threshold for pooled analysis we have includedthis analysis for completeness and the results should beinterpreted with caution The RR for treatment compared withplacebo was 13 (103 to 16) and the NNT was 64 (34 to 41)
bull Two studies used gel formulations (241 participantsBilligmann 1996 Machen 2002) The RR for treatmentcompared with placebo was 27 (17 to 42) and the NNT was39 (27 to 67)
There was no statistically significant difference between the geland cream formulations (z = 1160 P value = 0246)
Topical ketoprofen versus placebo
Seven studies contributed to this analysis (Airaksinen 1993Dreiser 1989 Julien 1989 Kockelbergh 1985 Maziegraveres 2005bMaziegraveres 2005a Noret 1987) A total of 346 participants weretreated with topical ketoprofen and 337 with placebo (Analysis11)
bull The proportion of participants experiencing successfultreatment with topical ketoprofen was 73 (251346 range57 to 89)
bull The proportion of participants experiencing successfultreatment with placebo was 47 (157337 range 17 to 73)
bull The RR of treatment compared with placebo was 16 (14to 18)
bull The NNT for successful treatment was 39 (30 to 53) Forevery four participants treated with topical ketoprofen onewould experience successful treatment who would not have doneso with placebo
Effect of formulation
The effects of formulation are shown in Analysis 41 and Figure6
19Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Figure 6 LrsquoAbbeacute plot of clinical success in studies of topical ketoprofen versus topical placebo The size ofthe symbol is proportional to the size of the study (inset scale) Light blue ketoprofen gel pink ketoprofen
plaster
Two studies used a plaster formulation (335 participants Maziegraveres2005b Maziegraveres 2005a) The RR for treatment compared withplacebo was 12 (104 to 14) and the NNT was 82 (45 to 47)Five studies used gel formulations (348 participants Airaksinen1993 Dreiser 1989 Julien 1989 Kockelbergh 1985 Noret 1987)The RR for treatment compared with placebo was 22 (17 to 28)and the NNT was 25 (20 to 34)Ketoprofen as a gel formulation was statistically more efficaciousthan a plaster formulation (z = 3860 P value = 000014)
Topical piroxicam versus placebo
Three studies contributed to this analysis (Aoki 1984 Fujimaki1985 Russell 1991) A total of 255 participants were treated withtopical piroxicam and 249 with placebo (Analysis 11)
bull The proportion of participants experiencing successfultreatment with topical piroxicam was 68 (179255 range 53to 79)
bull The proportion of participants experiencing successfultreatment with placebo was 47 (118249 range 45 to 49)
bull The RR of treatment compared with placebo was 15 (13to 17)
bull The NNT for successful treatment was 44 (32 to 69) Forevery four participants treated with topical piroxicam one wouldexperience successful treatment who would not have done sowith placebo
Topical indomethacin versus placebo
20Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Three studies contributed to this analysis ( kermark 1990 Aoki1984 Fujimaki 1985) A total of 168 participants were treatedwith topical indomethacin and 173 with placebo (Analysis 11)
bull The proportion of participants experiencing successfultreatment with topical indomethacin was 58 (97168 range54 to 64)
bull The proportion of participants experiencing successfultreatment with placebo was 46 (79173 range 25 to 49)
bull The RR of treatment compared with placebo was 13 (103to 16)
bull The NNT for successful treatment was 83 (44 to 65) Forevery eight participants treated with topical indomethacin onewould experience successful treatment who would not have doneso with placebo
Topical benzydamine versus placebo
Three studies contributed to this analysis (Chatterjee 1977 Haig1986 Linde 1985) A total of 96 participants were treated withtopical benzydamine and 97 with placebo (Analysis 11) Al-though this is just below our threshold for pooled analysis we haveincluded this analysis for completeness and the results should beinterpreted with caution
bull The proportion of participants experiencing successfultreatment with topical benzydamine was 77 (7496 range70 to 86)
bull The proportion of participants experiencing successfultreatment with placebo was 67 (6597 range 48 to 80)
bull The RR of treatment compared with placebo was 12 (096to 14) There was no statistically significant difference betweentreatments (Figure 4)
Results for participants with clinical success with individual topi-cal NSAIDs where there were adequate data for analysis are sum-marised below in rsquoSummary of results Arsquo and Analysis 11 Analysis21 Analysis 31 and Analysis 41
Summary of results A Participants with clinical success
Comparison Studies Participants NSAID()
Placebo () Relative benefit (95CI)
NNT (95 CI)
Diclofenac -Flector plaster
4 1030 63 41 15 (14 to 17) 47 (37 to 65)
Diclofenac -other plaster
3 474 88 57 16 (14 to 18) 32 (26 to 42)
Diclofenac -Emulgel
2 314 78 20 38 (27 to 55) 18 (15 to 21)
Diclofenac -other gel
1 232 94 82 12 (11 to 13) 80 (48 to 24)
Ibuprofen -cream
3 195 71 56 13 (103 to 16) 64 (34 to 41)
Ibuprofen - gel 2 241 42 16 27 (17 to 42) 39 (27 to 67)
Ketoprofen -plaster
2 335 73 60 12 (104 to 14) 82 (45 to 47)
Ketoprofen - gel 5 348 72 33 22 (17 to 28) 25 (20 to 34)
Piroxicam 3 504 70 47 15 (13 to 17) 44 (32 to 69)
Indomethacin 3 341 58 46 13 (103 to 16) 83 (44 to 65)
21Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Benzydamine 3 193 77 67 12 (096 to 14) not calculated
Results for these two comparisons are derived from very smallamounts of data and are provided here for completeness Theyshould be interpreted with caution
Local adverse events
Local adverse events were irritation of the area to which the topicalNSAID was applied including redness or erythema and itch orpruritus Where reported these were usually described as mild andtransient
All topical NSAIDs versus placebo
Forty-two studies contributed to this analysis of which three com-pared two different drugs with placebo (Aoki 1984 Diebshlag1990 Fujimaki 1985) Three studies had two treatment armscomparing different formulations or application regimens for di-clofenac with placebo which have been combined for this analysis
(Costantino 2011 Klainguti 2010 Predel 2012) In total 3619participants were treated with topical NSAIDs and 3121 withplacebo (Analysis 51)
bull The proportion of participants experiencing a local adverseevent with a topical NSAID was 43 (1553619 range 0 to33)
bull The proportion of participants experiencing a local adverseevent with placebo was 46 (1453121 range 0 to 32)
bull The RR of topical NSAID compared with placebo was 098(080 to 12)
bull There was no significant difference between treatmentgroups so the NNH was not calculated
Individual topical NSAIDs versus placebo
Results for local adverse events with individual topical NSAIDswhere there were adequate data for analysis are in Summary ofresults B and Analysis 12
Summary of results B Participants with local adverse events
Comparison Studies Participants NSAID()
Placebo()
RR(95 CI)
NNH(95 CI)
All NSAIDs 42 6740 43 46 098 (080 to 12) Not calculated
Diclofenac 15 3271 31 43 078 (056 to 11) Not calculated
Ketoprofen 8 852 11 95 12 (083 to 17) Not calculated
Piroxicam 3 522 23 54 042 (017 to 11) Not calculated
Felbinac 3 397 30 15 19 (049 to 75) Not calculated
Indomethacin 3 354 63 22 27 (091 to 77) Not calculated
Ibuprofen 3 321 10 43 23 (098 to 54) Not calculated
Systemic adverse events
All topical NSAIDs versus placebo
Thirty-six studies contributed data on systemic adverse eventsof which three compared two different drugs with placebo (Aoki1984 Diebshlag 1990 Fujimaki 1985) Two studies had two treat-
22Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
ment arms comparing different formulations or application regi-mens for diclofenac with placebo which have been combined forthis analysis (Klainguti 2010 Predel 2012) In total 2956 partic-ipants were treated with a topical NSAID and 2620 with placebo(Analysis 52)
bull Twenty-three studies reported no systemic adverse events inany arm of the study
bull The proportion of participants experiencing a systemicadverse event with a topical NSAID was 31 (922956)
bull The proportion of participants experiencing a systemicadverse event with placebo was 35 (912620)
bull The RR of topical NSAID compared with placebo was 096(073 to 13)
bull There was no significant difference between treatmentgroups so the NNH was not calculated
A further six studies did not report the occurrence or oth-erwise of systemic adverse events (Billigmann 1996 Julien1989 Kockelbergh 1985 Noret 1987 Ramesh 1983 Vecchiet1991) while two studies did not report numbers of participantswith systemic adverse events ( kermark 1990 Auclair 1989)Costantino 2011 reported that there were no systemic gastroin-testinal adverse events Two studies reported only on total adverseevents without distinguishing between local and systemic events(NCT01255423 NCT01272934)
Serious adverse events
Two studies reported serious adverse events In Hoffmann 2012one participant experienced three serious adverse events none ofwhich was judged to be related to the study medication (diclofenacplaster) In NCT01272934 one participant using diclofenac gelruptured the ligaments of the wrist There was no statement aboutlikely relationship to the study medication but this seems unlikely
Adverse event withdrawals
Forty-two studies reported data relating to adverse event with-drawals of which three compared two different drugs with placebo(Aoki 1984 Diebshlag 1990 Fujimaki 1985) Two studies hadtwo treatment arms comparing different formulations or applica-tion regimens for diclofenac with placebo which have been com-bined for this analysis (Klainguti 2010 Predel 2012) In total3365 participants received a topical NSAID and 3040 placebo(Analysis 53)
bull Forty-four comparisons reported no adverse eventwithdrawals
bull The proportion of participants withdrawing from the studydue to an adverse event after treatment with a topical NSAIDwas 098 (333365)
bull The proportion of participants withdrawing from the studydue to an adverse event after treatment with placebo was 099(303040)
bull The RR of topical NSAID compared to placebo was 10(064 to 16)
bull There was no significant difference between treatmentgroups so the NNH was not calculated
Four studies did not specifically mention adverse event with-drawals (Haig 1986 Hoffmann 2012 Klainguti 2010 Vecchiet1991) while one reported that one participant withdrew with mildpruritus but did not state the treatment arm (Joussellin 2003)Ten studies specifically reported withdrawals due to lack of efficacy(Dreiser 1989 Dreiser 1994 Kuehl 2011 Machen 2002 Maziegraveres2005b Maziegraveres 2005a Noret 1987 Predel 2013a Russell 1991Thorling 1990) (Appendix 5) Numbers of participants withdraw-ing were generally low with rates of 6 or less except in Kuehl2011 where the rate was 10 with active treatment (diclofenacplaster) and 12 with placebo We did not carry out any analysisbecause the outcome was inconsistently reported
2 Topical NSAID versus active comparator
Details of efficacy outcomes in individual studies are in Appendix4 and of adverse events and withdrawals in Appendix 5
Participants with clinical success
Topical NSAID versus oral NSAID
bull kermark 1990 compared indomethacin spray withindomethacin capsules with response rates of 55 (1222) withspray and 23 (522) with capsules
bull Hosie 1993 compared felbinac foam with ibuprofen tabletswith response rates of 64 (81127) with felbinac foam and72 (96133) with ibuprofen tablets
bull Whitefield 2002 compared ibuprofen gel with ibuprofentablets with response rates of 60 (3050) with gel and 54(3650) with tablets
There were insufficient data for meta-analysis for any one of thesecomparisons felbinac is not known to be better than placebo
Topical NSAID versus different formulation of the same
topical NSAID
bull Fioravanti 1999 compared DHEP (diclofenac) gelformulated with and without lecithin with response rates of70 (3550) in both treatment arms
bull Mahler 2003 compared DHEP (diclofenac) gel formulatedwith and without lecithin with response rates of 89 (8292)with lecithin and 70 (6288) without lecithin
bull Gallacchi 1990 compared topical diclofenac formulated asFlectorreg gel and Emugelreg with response rates of 76 (1925)in both treatment arms
23Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
bull Governali 1995 compared topical ketoprofen cream withgel with response rates of 93 (1415) with cream and 27 (415) with gel
There were insufficient data for analysis
Topical NSAID versus different topical NSAID
Eight studies compared one topical NSAID versus at least oneother topical NSAID piroxicam versus indomethacin (Aoki1984 Fujimaki 1985 Sugioka 1984) ibuprofen versus ketopro-fen (Curioni 1985 Picchio 1981) ketoprofen versus etofena-mate (Curioni 1985 Tonutti 1994) ibuprofen versus etofenamate(Curioni 1985) ketorolac versus etofenamate (Diebshlag 1990)and diclofenac versus lysine clonixinate (Hofman 2000)There were sufficient data to compare only piroxicam with in-domethacin (Aoki 1984 Fujimaki 1985 Sugioka 1984 Analysis61)
bull The proportion of participants experiencing clinical successwith topical piroxicam was 56 (185330 range 49 to 78)
bull The proportion of participants experiencing clinical successwith topical indomethacin was 45 (140311 range 33 to64)
bull The RR of piroxicam compared with indomethacin was 12(11 to 14)
bull The NNT for successful treatment was 91 (53 to 30) Forevery nine participants treated with topical piroxicam onewould experience a clinical success who would not haveexperienced one with topical indomethacin
Topical NSAID versus different topical intervention
One study compared diclofenac gel with a herbal product calledTraumeel gel under double-blind conditions with response ratesfor being pain-free at seven days of 8137 (58) with diclofenacgel and 7140 (50) with Traumeel gel (Gonzaacutelez de Vega 2013)
Local adverse events
Topical NSAID versus oral NSAID
Two studies comparing a topical NSAID with an oral NSAIDprovided data on local adverse events ( kermark 1990 Hosie1993) There were five events with topical NSAID and three withoral NSAID which were too few for analysis
Topical NSAID versus different topical NSAID
All nine studies comparing one topical NSAID with at least oneother reported on local adverse events with a total of 48 events in1005 participants (48) There were sufficient data to compare
only piroxicam with indomethacin (Aoki 1984 Fujimaki 1985Sugioka 1984 Analysis 62)
bull The proportion of participants experiencing local adverseevents with topical piroxicam was 21 (7340 range 12 to28)
bull The proportion of participants experiencing local adverseevents with topical indomethacin was 10 (33331 range 29to 15)
bull The RR of piroxicam compared with indomethacin was021 (009 to 047)
bull The NNT to prevent a local adverse event was 13 (87 to23) For every thirteen participants treated with topicalpiroxicam one would not experience a local adverse event whowould have experienced one with topical indomethacin
Topical NSAID versus different topical intervention
Gonzaacutelez de Vega 2013 reported that adverse events were infre-quent and mild to moderate in intensity but did not distinguishbetween local and systemic events Numbers of participants expe-riencing any adverse event were 8147 with diclofenac gel and 14148 with Traumeel gel
Systemic adverse events
kermark 1990 reported numbers of events rather than numbersof participants with events while Tonutti 1994 and Whitefield2002 reported no adverse events attributable to the study medi-cation and Fioravanti 1999 Gallacchi 1990 Gualdi 1987 andSugioka 1984 did not mention systemic adverse events Gonzaacutelezde Vega 2013 did not distinguish between local and systemicevents In the remaining studies a total of 16 events were reportedin topical NSAID treatment arms (797 participants 2) and 11with ibuprofen tablets (134 participants 8)
Serious adverse events
No serious adverse events were reported in any treatment arm
Withdrawals
The only withdrawals reported due to adverse events were in stud-ies with placebo treatment arms ( kermark 1990 Fujimaki 1985)and have been reviewedThree studies reported withdrawals due to lack of efficacy (Gonzaacutelez de Vega 2013 Hofman 2000 Tonutti 1994) (Appendix5) There were insufficient data for analysisSome studies reported exclusions from analysis (efficacy or safetyor both) following randomisation mainly due to protocol vio-lations or loss to follow-up (Appendix 5) There is no reason tobelieve these exclusions would introduce systematic bias and thenumbers involved were not likely to influence results
24Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
D I S C U S S I O N
This updated review of topical NSAIDs for acute musculoskele-tal pain in adults differs from previous reviews Previously dataallowed only for comparison of individual topical NSAIDs withplacebo irrespective of formulation With a substantial amount ofnew data for diclofenac it is possible to distinguish effects of for-mulation for individual NSAIDs Because formulation chemistrycan substantially affect the rate and total amount of drug accessingsubcutaneous injured tissues the effect of formulation may be asimportant as the individual NSAID used Drug and formulationshould thus be considered together when assessing efficacy andthis is now possible
We have also included an assessment of the daily dose of NSAIDapplied to the skin This involved having information on the con-centration of NSAID in the preparation the amount used andthe frequency of use Not all studies reported all three but an esti-mation of topical doses applied was possible It varied by factors ofthree or four for each topical NSAID However for topical formu-lations the key issue is less the dose applied but the amount thatpenetrates locally (producing analgesic effect) and the amount en-tering the systemic circulation (producing potential harms) Bothwill depend on the exact formulation of the topical agent andwhether there is occlusion (Moore 2008a)
Summary of main results
This review included 61 studies comparing a topical NSAID withplacebo another topical NSAID or an oral NSAID In total5311 participants were treated with a topical NSAID 3470 withplacebo and 220 with an oral NSAID There were 63 moreparticipants than in the previous version of this review Conditionstreated were sprains strains and contusions mainly resulting fromsports injuries and overuse injuries such as tendinitisFormulations of topical diclofenac ibuprofen ketoprofen piroxi-cam and indomethacin demonstrated significantly higher rates ofclinical success than matching topical placebo lacking the NSAIDbenzydamine did not Three drug and formulation combinationshad NNTs for clinical success below 4 For diclofenac Emulgelreghad the lowest NNT of 18 (15 to 21) in two studies using atleast 50 pain intensity reduction as the outcome (high qualityevidence) Diclofenac plasters other than Flectorreg also had a lowNNT of 32 (26 to 42) based on good or excellent responses inrelatively recent studies (high quality evidence) Ketoprofen gelhad an NNT of 25 (20 to 34) from five studies in the 1980ssome with less well defined outcomes (moderate quality evidence)Ibuprofen gel had an NNT of 39 (27 to 67) from two stud-ies with outcomes of marked improvement or complete remission(moderate quality evidence) All other drug and formulation com-binations had NNT values above 4 indicating lesser efficacyThese results are better than alternative topical products that mightbe used for acute musculoskeletal pain There is no evidence tosupport the use of topical salicylate rubefacients (Derry 2014)
Treatment with a topical NSAID was not associated with anincrease in local adverse events (skin reactions) compared withplacebo (inert carrier) or in withdrawals due to adverse events(high quality evidence) The inert carrier was sometimes associ-ated with mild skin irritation but this rarely led to cessation oftreatment and quickly resolved Systemic adverse events were un-common and did not differ between topical NSAID and placebo(high quality evidence) Two participants experienced serious ad-verse events with diclofenac plaster and diclofenac gel but it isunlikely that these were related to the study medicationsThere were insufficient data directly comparing a topical NSAIDwith the same oral NSAID to draw conclusions about efficacyBased on very limited data for oral NSAIDs there were fewersystemic adverse events with topical than oral treatment Therewere sufficient data only for topical piroxicam compared withtopical indomethacin to compare one topical agent with anotherThese limited data suggested that piroxicam was more effectivethan indomethacin and was less likely to cause local adverse eventsIt is worth noting here that topical indomethacin did not givesignificantly better pain relief than placebo in two of the threestudies in this analysis
Overall completeness and applicability ofevidence
There is a tension between pooling studies to produce analyseswith larger numbers and the subsequent large increases in clinicaland statistical heterogeneity on the one hand and using the ap-proach of clinical homogeneity with subsequent smaller numbersof participants on the other hand Previous reviews have takenthe former approach that is useful in demonstrating that topicalNSAIDs rsquoworkrsquo by being significantly better than placebo Becauseof the substantial increase in the amount of data available in thisreview we have chosen to seek greater clinical homogeneity thisproduces results that are more relevant to patient and prescriberchoiceThere were too few studies comparing one topical NSAID versusanother or versus the same oral NSAID to allow meaningful directcomparisons between individual drugs or routes of administrationThe conditions treated in these studies are representative of thoselikely to be suitable for acute treatment with topical NSAIDs Themean age of participants in individual studies ranged from 25 to57 years and the nature of recruitment in many studies meant thatparticipants were actively engaged in sporting activities Neverthe-less older people in their 60s to 80s were also included in somestudies and the low levels of predominantly mild adverse eventsmeans that this route of administration of NSAIDs is suitable forall age groups able to manage the application processInformation from other sources mainly randomised studies last-ing 12 weeks or more in older populations with arthritis tendto confirm this A systematic review of topical NSAIDs in olderadults was difficult to interpret but suggested that the range of
25Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
withdrawal rates in these studies was similar with topical and oralNSAIDs (Makris 2010) It also claimed potentiation of warfarineffects but that was with topical salicylate not an NSAID Incontrast a pooled safety analysis of topical diclofenac in peopleaged 75 years or older reported minimal changes with a mean re-duction in haemoglobin of less than 1 gL with topical diclofenacand a mean systolic blood pressure reduction of almost 4 mm Hg(Roth 2012) Two large randomised 12-week studies comparingtopical with oral diclofenac in arthritis reported lower rates of gas-trointestinal adverse events with topical than oral especially severeevents but larger reductions in haemoglobin with oral diclofenac(Simon 2009 Tugwell 2004)The available evidence was limited by numbers to comment on rarebut potentially serious adverse events One example is the potentialfor photo-sensitivity reactions with topical ketoprofen Currentadvice from the Medicines and Healthcare products RegulatoryAgency in the UK is to avoid direct sunlight ultraviolet (UV) rayssunlamps and sunbeds while using topical ketoprofen and to seea healthcare professional or go to hospital if they experience a skinreaction to sunlight sunlamps or sunbeds (MHRA 2009)
Quality of the evidence
All included studies were both randomised and double-blind nonewas considered at high risk of methodological bias Many werecarried out in the 1980s and 1990s when methodological rigorand detailed reporting were not given such high priority and stud-ies did not always report details of the randomisation treatmentallocation and blinding processes More recent studies often didreport methodological details and tended to be larger (see Figure4 for a comparison of quality of reporting for different dates andNSAIDs) Our primary outcome of clinical success was not alwayswell-defined and was measured using different scales but againmore recent studies tended to report outcomes betterThe studies were conducted in different conditions with some-what different outcome definitions and duration and with differ-ent topical NSAIDs and formulations Moreover the small size ofmany of the studies is likely to result in considerable chance varia-tion (Counsell 1994 Moore 1998b) These factors would accountfor the high I2 values seen in several analyses Despite these sourcesof potential clinical heterogeneity most studies showed benefit oftopical NSAID over placeboThe design of studies to be able to demonstrate analgesic sensi-tivity is important in self limiting conditions such as strains andsprains Too long a duration and the condition results in sponta-neous resolution of painful symptoms while too short a durationmay be inadequate to show any effect The decision by trialists toconcentrate on outcomes closest to seven days of treatment appearsto be prudent and has been adopted in this and previous reviewsThere are potential differences in response to treatment betweenstrains and sprains and overuse-type injuries such as tendinitis
and future reviews may examine this At the present time thereare too few existing trials to explore any differences adequatelyBaseline pain may be a cause for concern Seven studies didnot report baseline pain levels (Billigmann 1996 Curioni 1985Haig 1986 NCT01255423 NCT01272934 NCT01272947Sinniger 1981) and a further 11 reported either mean levels ofless than moderate pain or a significant proportion of individualswith less than moderate pain ( kermark 1990 Aoki 1984 Auclair1989 Diebshlag 1990 Fujimaki 1985 Jenoure 1997 Linde 1985Picchio 1981 Ramesh 1983 Sugioka 1984 Whitefield 2002)using recognised scales Insufficient pain at baseline compromisesthe ability of a study to demonstrate any improvement All thenewly added studies reported baseline pain to be of at least mod-erate intensity
Potential biases in the review process
There has been greater interest in topical NSAIDs in recent yearsmainly because lower systemic drug levels reduce the risk of trou-blesome and severe adverse events particularly in the gastroin-testinal tract and renal and cardiovascular systems Most of theattention has been in chronic conditions such as osteoarthritiswith few studies in acute painful conditions Low levels of seriousadverse events with topical NSAIDs has been noted previously(Evans 1995) and the near absence of serious adverse events inthis review is unlikely to be due to any biases in the review processOne potential bias is that clinical trials for topical NSAIDs may nothave been published One previous review did find previously un-published trials (Moore 1998a) but a subsequent attempt that in-cluded extensive contacts with pharmaceutical companies revealedno additional data (Mason 2004a) While some old unpublishedstudies of topical NSAIDs in acute painful conditions may existthey constitute an unknown number of studies and participantswhose results are unknown and are likely to remain unknownFurthermore their relevance to current clinical practice may belimited as better formulations are developed New systems of trialregistration mean that we know what recent studies have been doneor are ongoing the number of studies and participants is knowneven if their results remain unknown We identified in Clinicaltri-alsgov three unpublished studies (612 participants) with adverseevent data but no dichotomous efficacy data 14 completed un-published studies (4403 participants) with no results posted andthree ongoing studies (880 participants)For the main topical NSAIDs of interest and where most informa-tion exists about 4200 participants in this review provided data onefficacy for diclofenac ibuprofen ketoprofen indomethacin andbenzydamine compared with placebo For efficacy there are un-known results from almost 5900 participants in studies known tohave been done but essentially unpublished Almost 6500 partic-ipants in this review provided information on local adverse eventsfor topical NSAIDs compared with placebo For local adverseevents the unknown results from known studies represents almost
26Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
5300 participants It is clear that identified unpublished but un-available study data amounts to a further potentially large increasein knowledge over and above the 60 increase in numbers ofparticipants already included in this updated reviewBased on efficacy data on known and available study results un-published trials showing no difference between any topical NSAIDand topical placebo and involving 5500 participants would haveto exist in order for the NNT to be as high as 9 at which pointthe effectiveness of topical NSAIDs would become clinically irrel-evant (Moore 2006) This amount of unpublished negative data isobviously available and while a negative result in all the identifiedstudies is unlikely knowledge would be greatly served by havingthese unpublished trial results availableWe have not yet attempted to obtain results from these clinicaltrials from the trial sponsors because this takes a considerableamount of time and may not be successful Moreover the studieswere spilt between nine different sponsors Cerimon Pharmaceu-ticals (five studies) Novartis (four studies) Endo Pharmaceuticals(three studies) GlaxoSmithKline (two studies) Hisamatsu Phar-maceutical (two studies) and one each from Actavis Pfizer Im-primis Pharmaceutical and Strategic Science amp Technologies
Agreements and disagreements with otherstudies or reviews
A review published in 2004 included some of the studies in thisreview and reported an NNT for all topical NSAIDs combinedof 38 (34 to 44) for clinical success equivalent to half pain reliefat seven days (Mason 2004a) That review found no differencebetween topical NSAID and placebo for local adverse events as didthis review In turn the Mason review was in broad agreement withthe original systematic review on topical NSAIDs (Moore 1998a)To our knowledge no previous review assembled sufficient trialdata to analyse results by both drug and formulation as was donehere
A U T H O R S rsquo C O N C L U S I O N S
Implications for practice
For people with acute musculoskeletal pain
Topical NSAIDs can provide good levels of pain relief in acuteconditions such as sprains strains and overuse injuries proba-bly similar to that provided by oral NSAIDs Gel formulations ofdiclofenac (as Emulgelreg) ibuprofen and ketoprofen and somediclofenac patches provide the best effects Adverse events are usu-ally minimal with topical NSAIDs
For clinicians
Topical diclofenac ibuprofen or ketoprofen gels provide goodpain relief for painful acute musculoskeletal conditions and arebetter tolerated than oral formulations These drugs and formu-lations are more likely to be cost effective than alternative topicalpreparations such as topical rubefacients
For policy makers
Topical NSAIDs are not associated with an increased incidence oflocal skin reactions compared with the inert carrier and while thecarrier may cause mild transient irritation it is rarely troublesomeTopical NSAIDs do not cause systemic (mainly gastrointestinal)problems commonly seen with oral NSAIDs making them partic-ularly useful for individuals unable to tolerate oral administrationor for whom it is contraindicated
For funders
Topical diclofenac ibuprofen or ketoprofen gels should be consid-ered for initial treatment of acute musculoskeletal painful condi-tions where there are no contraindications such as damaged skinThese drugs and formulations are more likely to be cost effectivethan alternative topical preparations such as topical rubefacientsBecause formulations of topical NSAIDs are likely to change overtime the relevant trials performed and reported in or before the1990s must be limited and may be questionable Funders mightwish to consider asking pharmaceutical companies without recenttrial evidence for their products to produce it
Implications for research
General
The general thrust of these findings is that gel formulations of top-ical diclofenac ibuprofen and ketoprofen work best but for somedrugs (ketoprofen for instance) studies were pre-1990 Thesestudies may not be relevant to products available now Becauseformulation can have a significant effect on efficacy formulationchanges should be accompanied by relevant randomised trials
Design
The design of the trials is generally good and the sports injurymodel appears to be reliable and reproducible Modern studieshave ensured that participants entering the trials have at least mod-erate pain and this helps sensitivity to detect an analgesic responseMajor changes to the design of these trials would not appear to beneeded
27Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Measurement (endpoints)
A major issue is not in the measurement of pain as most studiesespecially modern studies have used standard pain intensity andpain relief scales However reporting of average pain changes isinadequate and the use of responder analyses (at least 50 painintensity reduction or people experiencing mild or no pain) ispreferred
Comparison between active treatments
Indirect comparisons with placebo are probably as informative asuse of an active comparator
A C K N O W L E D G E M E N T S
The Oxford Pain Relief Trust the NHS Cochrane CollaborationProgramme Grant Scheme and the NIHR Biomedical ResearchCentre Programme provided support for the earlier review
The National Institute for Health Research (NIHR) is the largestsingle funder of the Cochrane Pain Palliative and SupportiveCare Review Group Disclaimer the views and opinions expressedherein are those of the authors and do not necessarily reflect thoseof the NIHR National Health Service (NHS) or the Departmentof Health
R E F E R E N C E S
References to studies included in this review
Airaksinen 1993 published data onlyAiraksinen O Venaumllaumlinen J Pietilaumlinen T Ketoprofen 25gel versus placebo gel in the treatment of acute soft tissueinjuries International Journal of Clinical PharmacologyTherapy and Toxicology 199331(11)561ndash3
kermark 1990 published data only
kermark C Forsskaringhl B Topical indomethacin in overuseinjuries in athletes A randomized double blind studycomparing Elmetacinreg with oral indomethacin andplacebo International Journal of Sports Medicine 199011(5)393ndash6
Aoki 1984 published data only
Aoki T Numajiri M Yamamoto M A well controlledcomparative study of piroxicam gel indomethacin geland placebo gel in the treatment of trauma JapanesePharmacology and Therapeutics 198412(12)101ndash17
Auclair 1989 published data onlyAuclair J Georges M Grapton X Gryp L DrsquoHooghe MMeisser RG et al A double-blind controlled multi-centerstudy of percutaneous niflumic acid gel and placebo in thetreatment of Achilles heel tendinitis Current TherapeuticResearch 198946(4)782ndash8
Billigmann 1996 published data onlyBilligmann PW Treatment of ankle distortion withibuprofen gel [Therapie von Sprunggelenksndashdistosionenmit ibuprofenndashmikrogel] Therapiewoche 199646(21)1187ndash92
Campbell 1994 published data onlyCampbell J Dunn T Evaluation of topical ibuprofen creamin the treatment of acute ankle sprains Journal of Accidentand Emergency Medicine 199411(3)178ndash82
Chatterjee 1977 published data onlyChatterjee DS A double-blind clinical study withbenzydamine 3 cream on soft tissue injuries in anoccupational health centre Journal of International MedicalResearch 19775(6)450ndash8
Costantino 2011 published data only
Costantino C Kwarecki J Samokhin AV Mautone GRovati S Diclofenac epolamine plus heparin plasterversus diclofenac epolamine plaster in mild to moderateankle sprain a randomized double-blind parallel-groupplacebo-controlled multicentre phase III trial ClinicalDrug Investigation 201131(1)15ndash26 [DOI 10216511585890-000000000-00000]
Coudreuse 2010 published data onlyCoudreuse JM de Vathaire F Effect of a plaster containingDHEP and heparin in acute ankle sprains with oedemaa randomized double-blind placebo-controlled clinicalstudy Current Medical Research and Opinion 201026(9)2221ndash8 [DOI 101185030079952010508020]
Curioni 1985 published data only
Curioni BG Di Domenica F Daolio P Spignoli GEvaluation of ibuproxam gel in traumatology [Valutazionedellrsquoefficacia terapeutica di ibudros gel in traumatologia]Clinica Europea 198524(3)456ndash60
Diebshlag 1990 published data only
Diebschlag W Nocker W Bullingham R A double-blindstudy of the efficacy of topical ketorolac tromethamine gelin the treatment of ankle sprain in comparison to placeboand etofenamate Journal of Clinical Pharmacology 199030
(1)82ndash9
Dreiser 1988 published data onlyDreiser RL Clinical trial of efficacy and tolerability oftopical ibuprofen in the treatment of tendinitis Le JournalInternational De Meacutedecine 198811915ndash31
Dreiser 1989 published data only
Dreiser RL Clinical trial - Fatsum gel FG-6 Supplied byMenarini 1989
Dreiser 1990 published data onlyDreiser RE Charlot J Lopez A Ditisheim A Clinicalevaluation of niflumic acid gel in the treatment ofuncomplicated ankle sprains Current Medical Research andOpinion 199012(2)93ndash9
28Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Dreiser 1994 published data only
Dreiser RL Roche R de Sahb R Thomas F Leutenegger EFlurbiprofen local action transcutaneous (LAT tm) clinicalevaluation in the treatment of acute ankle sprains EuropeanJournal of Rheumatology and Inflammation 199414(4)9ndash13
Fioravanti 1999 published data onlyFioravanti A Cicero MR Nerucci F Manopulo RMarcolongo R Double-blind controlled clinical study of theefficacy and tolerability of diclofenac-N-(2-hydroxyethyl)-pyrrolidine lecithin gel compared with diclofenac-N-(2-hydroxyethyl)-pyrrolidine gel in patients with periand extraarticular inflammatory diseases Drugs underExperimental and Clinical Research 199925(5)235ndash40
Fujimaki 1985 published data only
Fujimaki E et al Clinical evaluation of piroxicam gelversus indomethacin gel and placebo in the treatment ofmuscle pain a double-blind multicenter study JapanesePharmacology and Therapeutics 198512(12)119ndash37
Gallacchi 1990 published data only
Gallacchi G Mautone G Lauldi P Topical treatmentwith diclofenac hydroxyethylpyrrilidine (Flector gel 1)Clinical Trials Journal 199027(1)58ndash64
Gonzaacutelez de Vega 2013 published data only
Gonzaacutelez de Vega C Speed C Wolfarth B Gonzaacutelez JTraumeel vs diclofenac for reducing pain and improvingankle mobility after acute ankle sprain a multicentrerandomised blinded controlled and non-inferiority trialInternational Journal of Clinical Practice 201367(10)979ndash89 [DOI 101111ijcp12219]
Governali 1995 published data only
Governali E Casalini D A controlled clinical studyon ketoprofen gel 5 versus ketoprofen ointment 1in patients with post-traumatic lesions [Ricerda clinicacontrollata tra ketoprofene gel 5 e ketoprofene crema1 in pazienti con postumi di lesioni traumatiche]Riabilitazione 199528(1)61ndash9
Gualdi 1987 published data only
Gualdi A Bonollo L Martini A Forgione A Non-steroidal anti-inflammatory drugs for topical therapy intraumatology a double-blind study with flunoxaprofen andketoprofen [Antinflammatori no steroidi per uso topicoin traumatologia studio clinico con flunoxaprofene echetoprofene] Riforma Medica 1987102(10)401ndash4
Haig 1986 published data only
Haig G Portable thermogram technique for topicallyapplied benzydamine cream in acute soft-tissue injuriesInternation Journal of Tissue Reactions 19868(2)145ndash7
Hoffmann 2012 published data only
Hoffmann P Kopa ka P Gugliotta B Rovati S Efficacyand tolerability of DHEP-heparin plaster in reducingpain in mild-to-moderate muscle contusions a double-blind randomized trial Current Medical Researchand Opinion 201228(8)1313ndash21 [DOI 101185030079952012709182]
Hofman 2000 published data only
Hofman J Nasswetter G Cayetti LM Lysine clonixinate gelin soft tissue injuries Controlled randomized prospectivedouble-blind clinical trial with diclofenac [Clonixinatode lisina gel en lesiones de tejidos blandos ensayo clinicoprospectivo doble ciego al azar controlado con diclofenac]Prensa Medica Argentina 200087(5)513ndash20
Hosie 1993 published data only
Hosie GAC The topical NSAID felbinac versus oralibuprofen a comparison of efficacy in the treatment ofacute lower back injury British Journal of Clinical Research199345ndash17
Jenoure 1997 published data only
Jenoure PJ Rostan A Gremion G Meier JL Grossen RBielinki R et al Multicentre double-blind controlledclinical study on the efficacy of diclofenac epolamineTissugel plaster in patients with epicondylitis [Studiomulticentrico controllato in doppio cieco su diclofenacTissugel plaster in pazienti con epicondilite] MedicinaDello Sport 197750(3)285ndash92
Joussellin 2003 published data onlylowast Joussellin E Flector Tissugel for the treatment ofpainful ankle sprains [Flector Tissugel dans le traitementdes entorses douloureuses de la cheville] Journal deTraumatologie du Sport 2003201S5ndash9Lionberger DR Joussellin E Lanzarotti A Yanchick JMagelli M Diclofenac epolamine topical patch relieves painassociated with ankle sprain Journal of Pain Research 2011447ndash53 [DOI 102147JPRS15380]Lionberger DR Joussellin E Yanchick J Magelli MLanzarotti A Pooled analysis of clinical trial data evaluatingthe safety and effectiveness of diclofenac epolamine topicalpatch 13 for the treatment of acute ankle sprain OpenAccess Journal of Sports Medicine 2011275ndash84 [DOI102147OAJSMS17048]
Julien 1989 published data onlyJulien D Clinical trial - Fastum gel FG-8 Supplied byMenarini 1989
Klainguti 2010 published data onlyKlainguti A Forgacs A Berkes I Castellacci E A plastercontaining DHEP and heparin for mild to moderatecontusions and sprains with haematoma a double-blind randomized study Current Medical Researchand Opinion 201026(9)2243ndash51 [DOI 101185030079952010508022]
Kockelbergh 1985 published data onlyKockelbergh M Verspeelt P Caloine R Dermaux F Localanti inflammatory treatment with a ketoprofen gel currentclinical findings Journal Belge de Medecine Physique et deRehabilitation 19858(4)205ndash13
Kuehl 2011 published data only
Kuehl K Carr W Yanchick J Magelli M Rovati SAnalgesic efficacy and safety of the diclofenac epolaminetopical patch 13 (DETP) in minor soft tissue injuryInternational Journal of Sports Medicine 201132(8)635ndash43[DOI 101055s-0031-1275359]
29Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Li 2013 published data only
Li C Frangione V Rovati S Zheng Q Diclofenacepolamine medicated plaster in the treatment of minor softtissue injuries a multicenter randomized controlled trialCurrent Medical Research and Opinion 201329(9)1137ndash46[DOI 101185030079952013816669]
Linde 1985 published data only
Linde F Hvass I Juumlrgensen U Madsen F Treatment ofsprained ankles with 5 benzydamine creme A double-blind study [Ankelforstuvninger behandlet med benzydamin5 creme] Ugeskrift for Laeger 1985148(1)12ndash3
Machen 2002 published data onlyMachen J Whitefield M Efficacy of a proprietary ibuprofengel in soft tissue injuries a randomised double blindplacebo controlled study International Journal of ClinicalPharmacology 200256(2)102ndash6
Mahler 2003 published data only
Mahler P Mahler F Duruz H Ramazzina M Liguori VMautone G Double-blind randomized controlled studyon the efficacy and safety of a novel diclofenac epolaminegel formulated with lecithin for the treatment of sprainsstrains and contusions Drugs under Experimental andClinical Research 200329(1)45ndash52
Maziegraveres 2005a published data onlyMaziegraveres B Rouanet S Velicy J Scarsi C Reiner V Topicalketoprofen patch (100 mg) for the treatment of anklesprain a randomized double-blind placebo-controlledstudy American Journal of Sports Medicine 200533(4)515ndash23 [DOI 1011770363546504268135]
Maziegraveres 2005b published data onlyMaziegraveres B Rouanet S Guillon Y Scarsi C Reiner VTopical ketoprofen patch in the treatment of tendinitisa randomized double blind placebo controlled studyJournal of Rheumatology 200532(8)1563ndash70
McLatchie 1989 published data onlyMcLatchie GR McDonald M Lawrence GF RogmansD Lisai P Hibberd M Soft tissue trauma a randomisedcontrolled trial of the topical application of felbinac a newNSAID British Journal of Clinical Practice 198943(8)277ndash80
Morris 1991 published data only
Morris WD Scott HV Peters WA Ketelbey JW Felbinactopical gel for acute soft tissue sports injuries New ZealandJournal of Sports Medicine 19911945ndash7
NCT01255423 unpublished data only
Novartis (Sponsors) A randomized double-blind multi-center placebo-controlled parallel group study to evaluatethe efficacy and safety of diclofenac sodium topical gel(DSG) 1 applied 4 times daily in subjects with acuteankle sprain clinicaltrialsgovct2showNCT01255423(accessed 3 February 2015) 2012 [CTG NCT01255423]
NCT01272934 unpublished data only
Novartis (Sponsors) A randomized double-blind multi-center placebo-controlled parallel group study to evaluatethe efficacy and safety of diclofenac sodium topical gel
(DSG) 1 applied four times daily in subjects withacute ankle sprain clinicaltrialsgovct2showresultsNCT01272934 (accessed 3 February 2015) 2013 [CTGNCT01272934]
NCT01272947 unpublished data onlylowast Novartis (Sponsors) A randomized double-blindmulti-center placebo-controlled parallel group study toevaluate the efficacy and safety of diclofenac sodium topicalgel (DSG) 1 applied four times daily in subjects withacute blunt soft tissue injuriescontusions of the limbsclinicaltrialsgovct2showresultsNCT01272947 (accessed3 February 2015) 2012 [CTG NCT01272947]Pallay R Pabst H Giannetti B Burnett I Monnet JDiclofenac sodium topical gel (DSG) 1 for acute softtissue injuriescontusions of the limbs 2013 AnnualAssembly of the American Academy of Physical Medicineand Rehabilitation National Harbor MD United States2013 Vol 5 (9 Suppl 1)S204
Noret 1987 published data only
Noret A Roty V Allington N Hauters P Zuinen C PoelsR Ketoprofen gel as topical treatment for sport injuriesActa Therapeutica 198713367ndash78
Parrini 1992 published data onlyParrini M Cabitza P Arrigo A Vanasia M Efficacy andtolerability of ketoprofen lysine salt foam for topical use inthe treatment of traumatic pathologies of the locomotorapparatus [Efficacia e tollerabilita del ketoprofene sale dilisina schiuma per uso topico nel trattamento di alcunepatologie traumatiche dellrsquoapparato locomotore] La ClinicaTerapeutica 1992141(9)199ndash204
Picchio 1981 published data only
Picchio AA Volta S Longoni A Controlled clinical trial ofibuprofen for topical use in sport injuries [Studio clinicocontrollato sullrsquoimpiego dellrsquoibuprofen per uso topico intraumatologica sportiva] Medicina dello Sport 198134403ndash6
Predel 2004 published data only
Mueller EA Kirch W Reiter S Extent and time courseof pain intensity upon treatment with a topical diclofenacsodium patch versus placebo in acute traumatic injury basedon a validated end point post hoc analysis of a randomizedplacebo-controlled trial Expert Opinion on Pharmacotherapy201011(4)493ndash8 [DOI 10151714656560903535898]lowast Predel HG Koll R Pabst H Dieter R Gallacchi GGiannetti B et al Diclofenac patch for topical treatmentof acute impact injuries a randomised double blindplacebo controlled multicentre study British Journal ofSports Medicine 200438(3)318ndash23 [DOI 101136bjsm2003005017]
Predel 2012 published data onlyPredel HG Hamelsky S Gold M Giannetti B Efficacy andsafety of diclofenac diethylamine 232 gel in acute anklesprain Medicine and Science in Sports and Exercise 201244
(9)1629ndash36 [DOI 101249MSS0b013e318257ed41]
30Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Predel 2013a published data only
Predel HG Giannetti B Seigfried B Novellini R MenkeG A randomized double-blind placebo-controlledmulticentre study to evaluate the efficacy and safetyof diclofenac 4 spray gel in the treatment of acuteuncomplicated ankle sprain Journal of InternationalMedical Research 201341(4)1187ndash202 [DOI 1011770300060513487639]
Predel 2013b published data onlyPredel HG Giannetti B Pabst H Schaefer A Hug AMBurnett I Efficacy and safety of diclofenac diethylamine116 gel in acute neck pain a randomized double-blindplacebo-controlled study BMC Musculoskeletal Disorders201314250 [DOI 1011861471-2474-14-250]
Ramesh 1983 published data only
Ramesh N Steuber U Ibuprofen in a cream vehicle foraccidental and sport injuries [Dolgit creme bei unfallndash undsportverletzungen] Therapiewoche 1983334563ndash70
Rowbotham 2003 published data only
Rowbotham M Galer B Block J Backonja M FlectorTissugel efficacy and safety in minor sport injuries Avs controlled clinical trial Journal of Sports Traumatology200320IS15ndashIS20
Russell 1991 published data only
Russell AL Piroxicam 05 topical gel compared to placeboin the treatment of acute soft tissue injuries a double-blind study comparing efficacy and safety Clinical andInvestigative Medicine 199114(1)35ndash43
Saillant 1998 published data only
Lionberger DR Joussellin E Yanchick J Magelli MLanzarotti A Pooled analysis of clinical trial data evaluatingthe safety and effectiveness of diclofenac epolamine topicalpatch 13 for the treatment of acute ankle sprain OpenAccess Journal of Sports Medicine 2011275ndash84 [DOI102147OAJSMS17048]lowast Saillant G Study comparing the efficacy and toleranceof Flector Tissugelreg to that of a placebo in the treatmentof benign ankle sprains [Eacutetude comparant lrsquoefficaciteacute et latoleacuterance de FlectorTissugelreg agrave celles drsquoun placebo dans letraitement des entorses beacutenignes de la cheville] Medicinedu Sport 1998721ndash5
Sanguinetti 1989 published data onlySanguinetti C Treatment of soft tissue injury with BPAAgel Results of an Italian multicenter study vs placebo[Trattemento con BPAA gel dei traumi dei tessuti moli]Clinica Terapeutica 1989130(5)255ndash258
Sinniger 1981 published data onlySinniger M Blanchard P Controlled clinical trial withFentiazac cream in sport microtraumatology Journal ofInternational Medical Research 19819(4)300ndash2
Spacca 2005 published data onlySpacca G Cacchio A Forgcs A Monteforte PRovetta G Analgesic efficacy of a lecithin-vehiculateddiclofenac epolamine gel in shoulder periarthritis andlateral epicondylitis a placebo-controlled multicenter
randomized double-blind clinical trial Drugs underExperimental and Clinical Research 200531(4)147ndash54
Sugioka 1984 published data only
Sugioka Y Multicenter clinical evaluation of piroxicam gelvs indomethacin gel in the treatment of non-traumaticdiseases of tendon or muscle Japanese Pharmacology andTherapeutics 198412139ndash53
Thorling 1990 published data only
Thorling J Linden B Berg R Sandahl A A double blindcomparison of naproxen gel and placebo in the treatment ofsoft tissue injuries Current Medical Research and Opinion199012242ndash8
Tonutti 1994 published data only
Tonutti A The use of ketoprofen gel 5 (orudis gel) intraumatology controlled double-blind study vs etofenamate[Utilizzazione del ketoprofene gel 5 (orudis gel) nellapratica traumatologica studio in doppio cieco controllatoverso etofenamato] Ortopedia e Traumatologia Oggi 199414(3)119ndash25
Vecchiet 1991 published data only
Vecchiet L Colozzi A Effects of meclofenamic acid in thetreatment of lesions deriving from minor traumatologyClinical Journal of Pain 19917 Suppl 1S54ndash9
Whitefield 2002 published data only
Whitefield M OrsquoKane CJA Anderson S Comparativeefficacy of a proprietary topical ibuprofen gel and oralibuprofen in acute soft tissue injuries a randomised doubleblind study Journal of Clinical Pharmacy and Therapeutics200227(6)409ndash17
References to studies excluded from this review
Ambrus 1987 published data only
Ambrus P Boumlhmer D Mobilat ointment in acute sprains[Mobilat Salbe bei akuten Distorsionen] Fortschritte derMedizin 1987105(13)259ndash62
Anon 1993 unpublished data only
Anon Comparative clinical efficacy of Oruvail piroxicamand diclofenac gels in soft tissue injury Unpublished
Ascherl 1982 published data onlyAscherl R Schlemmer H Blumel G Lechner F Theeffectiveness of etofenamate in minor sports injuries ofthe knee and ankle joint A double blind study [DieWirksamkeit der Etofenamat in kleineren Sportverletzungenam Knie und Sprunggelenk Eine DoppelblindndashStudie]Fortschritte der Medizin 1982100(37)1729ndash34
Bagliani 1976 published data only
Bagliani A Montalbetti L Topical treatment ofthrombophlebitis with feprazone and benzydamineControlled clinical study [Il trattamento topico ditromboflebite con feprazone e benzidamina Studio clinicocontrollato] Minerva Medica 197667(14)880ndash4
Baracchi 1982 published data only
Baracchi G Messina Denaro S Piscini S Experience ofthe topical use of isobutylfenylproprionic acid (Ibuprofen)in traumatic inflammation A double-blind comparison
31Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
with placebo [Esperienza sullrsquoimpiego topico dellrsquoacidoisobutilndashfenilndashpropionico (ibuprofen) nella flogositraumatica Confronto a doppia cecita con placebo]Gazzetta Medica Italiana 1982141(12)691ndash4
Boumlhmer 1995 published data onlyBoumlhmer D Ambrus P Treatment of muscular injurieswith diclofenac-diethylammonium emugel [Behandlungvon muskelverletzungen mit diclofenacndashdiethylammoniumemugel] Sportverletzung Sportschaden 1995994ndash5
Burnham 1998 published data only
Burnham R Gregg R Healy P Steadward R Theeffectiveness of topical diclofenac for lateral epicondylitisClinical Journal of Sport Medicine 19988(2)78ndash81
Cesarone 2008 published data onlyCesarone MR Belcaro G Pellegrini L Ledda A VinciguerraG Ricci A et al Treatment of ankle sprain in patientswith vascular diseases of the lower limbs MinervaCardioangiologica 200856 (5 Suppl)39ndash46
Coulibaly 2009 published data only
Coulibaly SK Courau S Staiger C The sprained anklecomparative study of a comfreyroot extract ointment versusa diclofenac gel [Entorses de la cheville eacutetude comparativeentre un extrait agrave base de racines de consoude et lediclofeacutenac] Phytotheacuterpie 20097147ndash9 [DOI 101007s10298-009-0383-x]
Diebschlag 1985 published data only
Diebschlag W Benzydamine cream in post-traumaticoedema International Journal of Tissue Reactions 19857(3)219ndash23
Diebschlag 1986 published data onlyDiebschlag W Diclofenac in blunt traumatic ankle jointswelling Volumetric monitoring in a placebo controlleddouble blind trial [Diclofenac in stumpfen traumatischenSprunggelenk Schwellung Volumetrische Uumlberwachungin einer Placebondashkontrollierten DoppelblindndashStudie]Fortschritte der Medizin 1986104(21)437ndash40
Diebschlag 1992 published data onlyDiebschlag W Nocker W Lehmacher W Treatment of acutesprains of the ankle joint A comparison of the effectivenessand tolerance of two gel preparations containingindomethacin [Die Behandlung der akuten Verstauchungendes Sprunggelenks Ein Vergleich der Wirksamkeitund Vertraumlglichkeit von zwei GelndashZubereitungen mitIndometacin] Fortschritte der Medizin 1992110(6)64ndash72
Fantato 1971 published data only
Fantato S De Gregorio M Clinical evaluation of topicalbenzydamine in traumatology Arzneimittel-Forschung197121(10)1530ndash5
Galer 2000 published data onlyGaler BS Rowbotham M Perander J Devers A FriedmanE Topical diclofenac patch relieves minor sports injury painresults of a multicenter controlled clinical trial Journal ofPain and Symptom Management 200019(4)287ndash94
Hallmeier 1986 published data only
Hallmeier B Michelbach B Etofenamate under tapebandages - a controlled study [Etofenamat untertapendashverbaumlnden] Medizinische Welt 198637(43)1344ndash8
Hallmeier 1988 published data onlyHallmeier B Efficacy and tolerance of etofenamate anddiclofenac in acute sports injuries [Wirksamkeit undVertraumlglichkeit von Etofenamat und Diclofenac bei akutenSportverletzungen] Rheuma 19888183ndash6
Kaneko 1999 published data only
Kaneko M Shimojo H Saito H Onuma Y YamashitaK Clinical evaluation of felbinac patch (SELSPOT) onpost-traumatic disease clinical comparative study versuscommercially available patch Japanese Pharmacology andTherapeutics 19992775ndash85
Kockelbergh 1985b published data only
Kockelbergh M Verspeelt P Caloine R Dermaux F Localanti-inflammatory treatment with a ketoprofen gel currentclinical findings [Traitement antindashinflammatoire local parun gel de keacutetoprofegravene donneacutees cliniques reacutecentes] JournalBelge de Medecine Physique et de Rehabilitation 19858(4)205-13 (study 2)
Kuwabara 2013 published data only
Kuwabara Y Hamamoto H Hikake S Miwa Y Arandomized multi-center double-blind placebo-controlledphase IIIII trial to evaluate the eftopical patch in thetreatment of pain Journal of Pain 2013 Vol 14 (4 Suppl1)S73
Lee 1991 published data only
Lee EH Lee PY Ngai AT Chiu EH Treatment of acute softtissue trauma with a topical non-steroidal anti-inflammatorydrug (biphenylacetic acid 3 gel) Singapore MedicalJournal 199132(4)238ndash41
Link 1996 published data onlyLink R Balint G Pavlik G Otto J Krause W Topicaltreatment of soft tissue rheumatism and athletic injuriesEffectiveness and tolerance of a new ketoprofen gel[Topische Behandlung von WeichteilndashRheumatismus undSportverletzungen Wirksamkeit und Vertraumlglichkeit einesneuen Ketoprofen Gel] Fortschritte der Medizin 1996114(25)311ndash4
May 2007 published data onlyMay JJ Lovell G Hopkins WG Effectiveness of1 diclofenac gel in the treatment of wrist extensortenosynovitis in long distance kayakers Journal of Scienceand Medicine in Sport 200710(1)59ndash65
Oakland 1993 published data only
Oakland C Rapier C A comparison of the efficacy of thetopical NSAID felbinac and ultrasound in the treatmentof acute ankle injuries British Journal of Clinical Research1993489ndash96
Odaglia 1987 unpublished data only
Odaglia G Sereni G Sports minor traumatology results ofa double-blind controlled clinical study ketoprofen (fastumgel 25) versus placebo Menarini unpublished data
32Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Picardi 1993 published data only
Picardi E De Iasio R Efficacy of percutaneous anti-inflammatory drugs (NSAIDs) in swimmers and waterpoloplayers [Efficacia dei farmaci antinfiammatori (FANS) pervia percutanea in atleti di nuoto e pallanuoto] ClinicaTerapeutica 1993143(6)507ndash9
Taboada 1992 published data onlyTaboada A Controlled trial of piroxicam gel associated withultrasound in acute disturbances of the locomotive system[Experienca controlada con gel de piroxicam asociado aultrasonidos en afecciones agudas del aparato locomotor]Prensa Medica Argentina 199279(10)630ndash2
Vanderstraeten 1990 published data only
Vanderstraeten G Schuermans P Study on the effect ofetofenamate 10 cream in comparison with an oral NSAIDin strains and sprains due to sports injuries Acta BelgicaMedica Physica 199013(3)139ndash41
Vinciguerra 2008 published data only
Vinciguerra G Belcaro G Cesarone MR Errichi BM DiRenzo A Errichi S et al Management of uncomplicatedankle sprains with topical or oral ketoprofen treatmentA registry study Minerva Cardioangiologica 200856 (5
Suppl)47ndash53
Von Klug 1977 published data onlyVon Klug H Experience with a locally applied anti-inflammatory drug [Erfahrungen mit einem localanwendbaren antirheumatikum] Arzneimittel-ForschungDrug Research 1977271350ndash4
References to studies awaiting assessment
NCT00351104 published data onlyEndo Pharmaceuticals (Sponsors) A randomized double-blind placebo-controlled parallel group phase III study ofthe efficacy tolerability and safety of ketoprofen topicalpatch 20 (KTP) in the treatment of pain associated withgrade 1 or grade 2 ankle sprain or strain clinicaltrialsgovct2showNCT00351104 (accessed 3 February 2015) 2008[CTG NCT00351104]
NCT00352625 unpublished data only
Endo Pharmaceuticals (Sponsors) A randomized double-blind placebo-controlled parallel group phase III study ofthe efficacy tolerability and safety of ketoprofen topicalpatch 20 (KTP) in the treatment of pain associatedwith tendonitis or bursitis of the shoulder elbow or kneeclinicaltrialsgovct2showNCT00352625 (accessed 3February 2015) 2008 [CTG NCT00352625]
NCT00426985 unpublished data only
Endo Pharmaceuticals (Sponsors) A randomized double-blind placebo-controlled parallel group phase iii study ofthe efficacy tolerability and safety of ketoprofen topicalpatch 20 (KTP) in the treatment of pain associatedwith tendonitis or bursitis of the shoulder elbow or kneeclinicaltrialsgovct2showNCT00426985 (accessed 3February 2015) 2008 [CTG NCT00426985]
NCT00640705 unpublished data only
Cerimon Pharmaceuticals (Sponsors) A randomizeddouble-blind placebo-controlled study of the efficacyand safety of a diclofenac sodium patch for the topicaltreatment of pain due to mild to moderate ankle sprainclinicaltrialsgovct2showNCT00640705 (accessed 3February 2015) 2008 [CTG NCT00640705]
NCT00640939 unpublished data onlyCerimon Pharmaceuticals (Sponsors) A randomizeddouble-blind placebo-controlled study of the efficacy andsafety of a diclofenac sodium patch for the topical treatmentof pain due to mild to moderate tendonitis or bursitisclinicaltrialsgovct2showNCT00640939 (accessed 3February 2015) 2008 [CTG NCT00640939]
NCT00680472 unpublished data onlyHisamitsu Pharmaceutical Co Inc (Sponsors) Arandomized multicenter double-blind placebo-controlledtwo-week study to assess the efficacy and safety of HKT-500 in subjects with acute shoulder pain clinicaltrialsgovct2showNCT00680472 (accessed 3 February 2015) 2015[CTG NCT00680472]
NCT00680784 unpublished data only
Hisamitsu Pharmaceutical Co Inc (Sponsors) Arandomized multicenter double-blind placebo-controlledstudy to evaluate the efficacy and safety of HKT-500 inthe treatment of pain associated with Grade I or Grade IIankle sprain clinicaltrialsgovct2showNCT00680784(accessed 3 February 2015) 2015 [CTG NCT00680784]
NCT00765700 unpublished data onlyEkman E Skrepnik S Jones M Lawson K Schupp JEfficacy and safety of ketoprofen 10 cream in acute softtissue injuries (phase 3 study tdlp-110-001) Proceedings ofthe 13th World Congress on Pain 2010 Aug 29 - Sept 2Montreal CanadaImprimis Pharmaceuticals Inc (Sponsors) A randomizedmulticenter double-blind placebo-controlled parallel-group phase 3 study to assess the efficacy and safetyof Ketotransdeltrade (Ketoprofen topical cream 10) inthe treatment of pain associated with mild to moderateacute soft tissue injury clinicaltrialsgovct2showNCT00765700 (accessed 3 February 2015) 2013 [CTGNCT00765700]
NCT00869063 unpublished data onlyCerimon Pharmaceuticals (Sponsors) A randomizeddouble-blind placebo-controlled study of the efficacy andsafety of a diclofenac sodium patch for the topical treatmentof acute pain due to mild to moderate wrist sprain strainor contusion clinicaltrialsgovct2showNCT00869063(accessed 3 February 2015) 2010 [CTG NCT00869063]
NCT00869180 unpublished data onlyCerimon Pharmaceuticals (Sponsors) A randomizeddouble-blind placebo-controlled study of the efficacy andsafety of a diclofenac sodium patch for the topical treatmentof acute pain due to mild to moderate ankle sprainclinicaltrialsgovct2showNCT00869180 (accessed 3February 2015) 2010 [CTG NCT00869180]
33Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT00931866 unpublished data only
Cerimon Pharmaceuticals (Sponsors) A randomizeddouble-blind placebo-controlled study of the efficacy andsafety of a diclofenac sodium patch for the topical treatmentof acute pain due to mild to moderate soft tissue injuriesclinicaltrialsgovct2showNCT00931866 (accessed 3February 2015) 2010 [CTG NCT00931866]
NCT01874626 unpublished data only
Strategic Science amp Technologies LLC (Sponsors) Arandomized double-blind placebo-controlled multi-dosepivotal study to determine the efficacy and safety of SST0225 a topical ibuprofen cream in the treatment of painassociated with acute ankle sprain clinicaltrialsgovct2showNCT01874626 (accessed 3 February 2015) 2014[CTG NCT01874626]
NCT01957215 unpublished data only
GlaxoSmithKline (Sponsors) A clinical study to assess theefficacy of pain relief of topical indomethacin patch overplacebo in ankle sprain patients clinicaltrialsgovct2showNCT01957215 (accessed 3 February 2015) 2015 [CTGNCT01957215]
NCT02324270 unpublished data only
Actavis Inc (Sponsors) Randomized double-blindmultiple-center placebo-controlled study comparing thesafety and efficacy of generic diclofenac epolamine toFlectorreg patch in the treatment of acute pain due to minorankle sprain clinicaltrialsgovct2showNCT02324270(accessed 3 February 2015) 2014 [CTG NCT02324270]
Sarzi-Puttini 2014 published data only
Sarzi-Puttini P Atzeni F Damiani C Casale R BarbagalloM Cazzola M A double-blind randomised parallelgroup active controlled multicentre study to assess thetherapeutic non-inferiority of SKP-021 a 03 ketoprofenpatch versus diclofenac sodium patch in patients withacute inflammatory musculoskeletal injuries Proceeding ofthe Annual European Congress of Rheumatology of theEuropean League Against Rheumatism EULAR 2014 June11-14 Paris France 201473
References to ongoing studies
NCT01945034 unpublished data onlyPfizer (Sponsors) Placebo-controlled double-blindevaluation of the efficacy and safety of ibuprofen 5 topicalgel for the treatment of ankle sprain clinicaltrialsgovct2showNCT01945034 (accessed 3 February 2015) 2014[CTG NCT01945034]
NCT02100670 unpublished data onlyGlaxoSmithKline (Sponsors) A clinical study to assessthe efficacy and onset of pain relief of topical MFC51123diclofenac-menthol gel versus controls in ankle sprainclinicaltrialsgovct2showNCT02100670 (accessed 3February 2015) 2014 [ClinicalTiralsgov NCT02100670]
NCT02290821 unpublished data only
Novartis (Sponsors) A randomized double-blind placebo-controlled parallel group study to evaluate the efficacy andsafety of diclofenac sodium topical gel (DSG) 1 applied
four times daily in subjects with acute blunt soft tissueinjuriescontusions of the limbs clinicaltrialsgovct2showNCT02290821 (accessed 3 February 2015) 2015 [CTGNCT02290821]
Additional references
Anon 2005
Anon Topical analgesics a review of reviews and a bit ofperspective wwwjr2oxacukBandolierExtraforbandoTopextra3pdf 2005
Cook 1995
Cook RJ Sackett DL The number needed to treat aclinically useful measure of treatment effect BMJ 1995310
(6977)452ndash4
Counsell 1994Counsell CE Clarke MJ Slattery J Sandercock PA Themiracle of DICE therapy for acute stroke fact or fictionalproduct of subgroup analysis BMJ 1994309(6970)1677ndash81
Dechartres 2013Dechartres A Trinquart L Boutron I Ravaud P Influenceof trial sample size on treatment effect estimates meta-epidemiological study BMJ 2013346f2304 [DOI101136bmjf2304]
Derry 2012a
Derry S Moore RA Topical capsaicin (low concentration)for chronic neuropathic pain in adults Cochrane Databaseof Systematic Reviews 2012 Issue 9 [DOI 10100214651858CD010111]
Derry 2012bDerry S Moore RA Rabbie R Topical NSAIDs forchronic musculoskeletal pain in adults Cochrane Databaseof Systematic Reviews 2012 Issue 9 [DOI 10100214651858CD007400pub2]
Derry 2013Derry S Sven-Rice A Cole P Tan T Moore RA Topicalcapsaicin (high concentration) for chronic neuropathic painin adults Cochrane Database of Systematic Reviews 2013Issue 2 [DOI 10100214651858CD007393pub3]
Derry 2014
Derry S Matthews PR Wiffen PJ Moore RASalicylate-containing rubefacients for acute and chronicmusculoskeletal pain in adults Cochrane Database ofSystematic Reviews 2014 Issue 11 [DOI 10100214651858CD007403pub3]
Evans 1995Evans JM McMahon AD McGilchrist MM White GMurray FE McDevitt DG et al Topical non-steroidal anti-inflammatory drugs and admission to hospital for uppergastrointestinal bleeding and perforation a record linkagecase-control study BMJ 1995311(6996)22ndash6
FitzGerald 2001
FitzGerald GA Patrono C The coxibs selective inhibitorsof cyclooxygenase-2 New England Journal of Medicine2001345(6)433ndash42
34Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Haroutiunian 2010
Haroutiunian S Drennan DA Lipman AG Topical NSAIDtherapy for musculoskeletal pain Pain Medicine 201011
(4)535ndash49 [DOI 101111j1526-4637201000809x]
Hawkey 1999
Hawkey CJ Cox-2 inhibitors Lancet 1999353(9149)307ndash14
Higgins 2011
Higgins JPT Altman DG Sterne JAC Chapter 8 Assessingrisk of bias in included studies In Higgins JPT GreenS (editors) Cochrane Handbook for Systematic Reviewsof Interventions Version 510 [updated March 2011]The Cochrane Collaboration 2011 Available fromwwwcochrane-handbookorg
Jadad 1996a
Jadad AR Carroll D Moore A McQuay H Developing adatabase of published reports of randomised clinical trials inpain research Pain 199666(2-3)239ndash46
Jadad 1996b
Jadad AR Moore RA Carroll D Jenkinson C ReynoldsDJ Gavaghan DJ et al Assessing the quality of reports ofrandomized clinical trials is blinding necessary ControlledClinical Trials 199617(1)1ndash12
LrsquoAbbeacute 1987
LrsquoAbbeacute KA Detsky AS OrsquoRourke K Meta-analysis inclinical research Annals of Internal Medicine 1987107(2)224ndash33
Makris 2010Makris UE Kohler MJ Fraenkel L Adverse effects oftopical nonsteroidal antiinflammatory drugs in older adultswith osteoarthritis a systematic literature review Journalof Rheumatology 201037(6)1236ndash43 [DOI 103899jrheum090935]
Mason 2004a
Mason L Moore RA Edwards JE Derry S McQuay HJTopical NSAIDs for acute pain a meta-analysis BMCFamily Practice 2004510
Mason 2004bMason L Moore RA Edwards JE Derry S McQuayHJ Topical NSAIDs for chronic musculoskeletal painsystematic review and meta-analysis BMC MusculoskeletalDisorders 2004528
McPherson 2013McPherson ML Cimino NM Topical NSAIDformulations Pain Medicine 201314 Suppl 1S35ndash9[DOI 101111pme12288]
MHRA 2009
Medicines and Healthcare Products Regulatory AgencyTopical ketoprofen reminder on risk of photo-sensitivityreactions Drug Safety Update 2009 June Vol 2 issue 116
Moore 1998aMoore RA Tramegraver MR Carroll D Wiffen PJ McQuayHJ Quantitative systematic review of topically applied non-
steroidal anti-inflammatory drugs BMJ 1998316(7128)333ndash8
Moore 1998bMoore RA Gavaghan D Tramegraver MR Collins SL McQuayHJ Size is everything - large amounts of information areneeded to overcome random effects in estimating directionand magnitude of treatment effects Pain 199878(3)209ndash16
Moore 2003
Moore RA Edwards J Barden J McQuay HJ BandolierrsquosLittle Book of Pain Oxford Oxford University Press 2003[ISBN 0ndash19ndash263247ndash7]
Moore 2006
Moore RA Barden J Derry S McQuay HJ Managingpotential publication bias In McQuay HJ Kalso EMoore RA editor(s) Systematic Reviews in Pain ResearchMethodology Refined Seattle IASP Press 200615ndash23[ISBN 978ndash0ndash931092ndash69ndash5]
Moore 2008aMoore RA Derry S McQuay HJ Topical agents in thetreatment of rheumatic pain Rheumatic Diseases Clinics ofNorth America 200834(2)415ndash32
Moore 2008bMoore RA Barden J Derry S McQuay HJ Managingpotential publication bias In McQuay HJ Kalso EMoore RA editor(s) Systematic Reviews in Pain ResearchMethodology Refined Seattle IASP Press 200815ndash24[ISBN 978ndash0ndash931092ndash69ndash5]
Moore 2013
Moore RA Straube S Aldington D Pain measures andcut-offs - rsquono worse than mild painrsquo as a simple universaloutcome Anaesthesia 201368(4)400ndash12 [DOI 101111anae12148]
Morris 1995
Morris JA Gardner MJ Calculating confidence intervalsfor relative risk odds ratios and standardised ratios andrates In Gardner MJ Altman DG editor(s) Statistics withConfidence - Confidence Intervals and Statistical GuidelinesLondon British Medical Journal 199550ndash63
NICE 2008National Institute for Health and Care ExcellenceOsteoarthritis The care and management of osteoarthritisin adults 2008 wwwniceorgukCG059fullguideline
Nuumlesch 2010Nuumlesch E Trelle S Reichenbach S Rutjes AW TschannenB Altman DG et al Small study effects in meta-analysesof osteoarthritis trials meta-epidemiological study BMJ2010341c3515 [DOI 101136bmjc3515]
PACT 2014
Prescribing and Primary Care team Health and SocialCare Information Centre Prescrption Cost AnalysisEngland 2013 Health and Social Care Information CentreThe NHS Information Centre 2014298ndash301 [ISBN978ndash1ndash78386ndash089ndash0]
35Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Quane 1998
Quane PA Graham GG Zeigler JB Pharmacology ofbenzydamine Inflammopharmacology 19986(2)95ndash107
RevMan 2014The Nordic Cochrane Centre The Cochrane CollaborationReview Manager (RevMan) 53 Copenhagen The NordicCochrane Centre The Cochrane Collaboration 2014
Roth 2012Roth SH Fuller P Pooled safety analysis of diclofenacsodium topical solution 15 (ww) in the treatment ofosteoarthritis in patients aged 75 years or older ClinicalInterventions in Aging 20127127ndash37 [DOI 102147CIAS30884]
Simon 2009
Simon LS Grierson LM Naseer Z Bookman AA ZevShainhouse J Efficacy and safety of topical diclofenaccontaining dimethyl sulfoxide (DMSO) compared withthose of topical placebo DMSO vehicle and oral diclofenacfor knee osteoarthritis Pain 2009143(3)238ndash45 [DOI101016jpain200903008]
Tramer 1997
Tramer MR Reynolds DJ Moore RA McQuay HJ Impact
of covert duplicate publication on meta-analysis a casestudy BMJ 1997315(7109)635ndash40
Tugwell 2004Tugwell PS Wells GA Shainhouse JZ Equivalence studyof a topical diclofenac solution (pennsaid) compared withoral diclofenac in symptomatic treatment of osteoarthritisof the knee a randomized controlled trial Journal ofRheumatology 200431(10)2002ndash12
Zimmerman 1995
Zimmerman J Siguencia J Tsvang E Upper gastrointestinalhemorrhage associated with cutaneous application ofdiclofenac gel American Journal of Gastroenterology 199590(11)2032ndash4 [PUBMED 7485017]
References to other published versions of this review
Massey 2010
Massey T Derry S Moore RA McQuay HJ TopicalNSAIDs for acute pain in adults Cochrane Databaseof Systematic Reviews 2010 Issue 6 [DOI 10100214651858CD007402pub2]
lowast Indicates the major publication for the study
36Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
C H A R A C T E R I S T I C S O F S T U D I E S
Characteristics of included studies [ordered by study ID]
Airaksinen 1993
Methods R DB PC parallel groupsGel applied to the painful area twice daily for 7 daysAssessment at baseline 3 7 days
Participants Minor soft tissue injuries (lt 7 days)N = 56M 45 F 11Age not reportedMean baseline pain at rest 25-26 mm
Interventions Ketoprofen gel 2 x 5 g (125 mg) daily n = 29Placebo gel n = 27Rescue medication paracetamol 500 mgNo other treatment allowed
Outcomes PGE 5-point scale but reported as ldquoimprovedrdquo or ldquosame or worserdquo (responder = ldquoim-provedrdquo)Improvement in pain with movement 100 mm VAS reported as group meanAdverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Unclear risk No mention of early withdrawals ormethod of imputation
Size High risk lt 50 participants per treatment group
37Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Aoki 1984
Methods R DB PC AC parallel groupsGel applied to affected area 3 or 4 times daily with no occlusion for 7 daysAssessment at baseline 3 7 days
Participants Acute orthopaedic trauma (contusion distortion fracture lt 7 days)N = 252 (203 analysed for efficacy)M 98 F 105Age range 8 to 86 years 13 younger than 20 yearsBaseline pain mild in 35Exclusions 23 protocol violations 26 reasons ldquonot relatedrdquo to drug Equally distributedbetween groups
Interventions Piroxicam gel 05 1 g 3 to 4 x daily n = 84Indomethacin gel 1 1 g 3 to 4 x daily n = 84Placebo gel n = 84No other medication or initiation of physical therapy allowed
Outcomes PGE 5-point scale (responder = ldquobetterrdquo and ldquomuch betterrdquo)Adverse eventsWithdrawals and exclusions
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Low risk ldquokey code sealed until end of studyrdquo
Blinding (performance bias and detectionbias)All outcomes
Low risk Gels in ldquoidentical tubesrdquo
Incomplete outcome data (attrition bias)All outcomes
Unclear risk gt 10 withdrawals ldquounrelated to treat-mentrdquo and for ldquoprotocol violationsrdquo Nofurther details but no significant differ-ences between groups
Size Unclear risk 50 to 200 participants per treatment group
38Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Auclair 1989
Methods R DB PC parallel groupsGel massaged into skin over affected heel 3 times daily after cleaning with soap and waterfor up to 21 daysAssessment at baseline 7 21 days
Participants Acute Achilles heel tendinitis (not associated with continuous pain at rest or gt 1 monthhistory)N = 243 (227 analysed for efficacy)MF not reportedMean age 29 yearsBaseline pain ~ 10 had lt 26 mm on palpation of tendon ~ 30 had mild or no painon dorsiflexion of footExclusions failure to meet inclusion criteria major protocol violations failure to takestudy medication for full duration
Interventions Niflumic acid gel 25 3 x 5 g daily n = 117Placebo gel n = 110No other analgesics and anti-inflammatories physiotherapy or supportive measures al-lowed
Outcomes PGE 5-point scale (responder = ldquogoodrdquo or ldquovery goodrdquo)Pain improved or disappeared on dorsiflexionAdverse eventsWithdrawals and exclusions
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Unclear risk 10 excluded for ldquofailing to meet entry cri-teria and protocol violationsrdquo No furtherdetails
Size Unclear risk 50 to 200 participants per treatment group
39Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Billigmann 1996
Methods R DB PC parallel groupsGel applied 3 times daily with rubbingAssessed at baseline 3 5 7 days
Participants Distortion of ankle jointN = 160M and FAge 18+ yearsBaseline pain not reported
Interventions Ibuprofen microgel 5 3 x 10 cm (= 200 mg) daily n = 80Placebo gel n = 80
Outcomes Pain with movement VAS (responder = decreased by 20)Complete remissionAdverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Unclear risk Insufficient data to assess
Size Unclear risk 50 to 200 participants per treatment group
Campbell 1994
Methods R DB PC parallel groupsCream applied 4 times daily for 7 days (up to 14 days optional)Self assessed using daily diary for 7 days and up to 14 days
Participants Acute ankle sprain (lt 24 hours no fracture)N = 100 (51 analysed)M 33 F 18Mean age 29 years
40Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Campbell 1994 (Continued)
Baseline pain at rest gt 35 mm on walking 80 mmExclusions did not return diaries protocol exclusions (25 ibuprofen 24 placebo)
Interventions Ibuprofen cream 5 (Proflex) 4 x 4rdquo (10 cm) daily n = 26Placebo cream n = 25Advised to use rest and regular icing for 48 hours then walking and exerciseRescue medication paracetamol
Outcomes Improvement in walking ability 4-point scale (responder = ldquoimprovementrdquo)Pain on walking 100 mm VAS (mean data)Withdrawals and exclusions
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Low risk Randomisation carried out by sponsorTubes dispensed by hospital pharmacy whoheld the codes
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical creamrdquo
Incomplete outcome data (attrition bias)All outcomes
High risk gt 40 lost to follow-up Approximatelyequal between groups
Size High risk lt 50 participants per treatment arm as anal-ysed
Chatterjee 1977
Methods R DB PC parallel groupsCream applied to site of injury 3 times daily for 6 daysAssessment at baseline 2 6 days
Participants Soft tissue injuries (recent)N = 51MF not reportedAge not reportedBaseline pain on passive movement moderate or severe in all but 3 participants
41Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Chatterjee 1977 (Continued)
Interventions Benzydamine HCl cream 3 3 x daily n = 25Placebo cream n = 25(5 active 6 placebo participants also received ultrasound)No other topical agent allowed
Outcomes Pain on passive movement 4-point scale (responder = ldquoabsentrdquo or ldquoslightrdquo)Tenderness with pressure 4-point scale (responder = ldquoabsentrdquo or ldquoslightrdquo)Adverse eventsWithdrawals and exclusions
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk ldquopredetermined randomised schedulerdquo
Allocation concealment (selection bias) Low risk Sealed copy of schedule held by investiga-tor and duplicate copy kept by clinical trialco-ordinator Looked at only in event ofadverse reaction (not necessary)
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoindistinguishable in appearance andconsistencyrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk 2 lost to follow-up no withdrawals dueto adverse events Responder analysis
Size Unclear risk lt 50 participants per treatment arm
Costantino 2011
Methods R DB PC multicentre parallel groupPlaster applied daily for 7 daysPI assessment daily overall treatment efficacy and tolerability assessed at 3 and 7 days
Participants Grade I or II ankle sprain (lt 48 hours) with lateral external ligament involvementPI on movement ge 50100 oedema ge 20 mm difference between anklesN = 430M 249 F 175 (for analysis)Mean age 35 yearsBaseline PI on movement 72100 (SD 12)
42Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Costantino 2011 (Continued)
Interventions DHEPhep n = 142DHEP n = 146Placebo n = 142Rescue medication paracetamol to maximum 3 g dailyNo other treatments allowed
Outcomes Mean change in PI on movement from baseline to 3 daysMean reduction in oedema at 3 daysTolerability at 3 daysRescue medicationAdverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk Computer-generated
Allocation concealment (selection bias) Unclear risk ldquosealed envelopesrdquo
Blinding (performance bias and detectionbias)All outcomes
Low risk Shape colour size and application methodidentical for all plasters
Incomplete outcome data (attrition bias)All outcomes
Unclear risk Very few withdrawals but used LOCF
Size Unclear risk 50 to 200 participants per treatment arm
Coudreuse 2010
Methods R DB PC multicentre parallel groupPlaster applied daily for 7 daysPI assessed twice daily over 3 days then day 7 Overall treatment efficacy and tolerabilityassessed on days 3 7
Participants Ankle sprain (lt 48 hours) with lateral external ligament involvementPI on movement ge 50100 oedema ge 20 mm difference between anklesN = 240 (233 for analysis)M 148 F 86 6 unknown
Interventions DHEPhep n = 120Placebo n = 120Rescue medication paracetamol to maximum 4 g daily
43Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Coudreuse 2010 (Continued)
No other treatments allowed
Outcomes Global efficacy at 7 days 4-point scale (responder = ldquoexcellentrdquo or ldquogoodrdquo)Mean change in PI on movement at 6 hours and 7 daysMean change in oedema at 3 and 7 daysTolerabilityAdverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk Computer-generated
Allocation concealment (selection bias) Unclear risk ldquosealed envelopesrdquo
Blinding (performance bias and detectionbias)All outcomes
Low risk Appearance and odour identical for all plas-ters
Incomplete outcome data (attrition bias)All outcomes
Low risk ITT analysis lt 5 excluded for missingdata equal between groups
Size Unclear risk 50 to 200 participants per treatment arm
Curioni 1985
Methods R DB PC AC parallel groupsGel rubbed into affected area until absorbed twice daily for 10 daysAssessed at baseline and daily to 10 days
Participants Acute soft tissue injuriesN = 60M 33 F 27Median age 33 yearsBaseline pain not given
Interventions Ibuproxam gel 10 n = 20Ketoprofen gel n = 20Etofenamate gel n = 20
Outcomes PGE 4-point scale (ldquogoodrdquo or ldquoexcellentrdquo)Resolution of symptomsAdverse events
44Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Curioni 1985 (Continued)
Withdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk Medication supplied in identical tubes
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
Diebshlag 1990
Methods R DB PC AC parallel groupsGel applied 3 times daily without occlusion for 14 daysAssessment at baseline 2 3 4 8 15 days
Participants Ankle sprain (lt 24 hours)N = 37M 24 F 13Mean age 28 yearsBaseline pain slight to moderate
Interventions Ketorolac gel 2 3 x 3 g daily n = 13Etofenamate gel 5 3 x 3 g daily n = 12Placebo gel n = 12Rescue medication paracetamolNo other analgesic or anti-inflammatory medication ice packs or physiotherapy allowed
Outcomes Reduction in PI 100 mm VAS and 4-point scale (responder = ldquoimprovedrdquo)Adverse eventsWithdrawals and exclusions
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
45Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Diebshlag 1990 (Continued)
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Low risk ldquoMedication assignment supplied in asealed enveloperdquo Opened only if seriousparticipant event necessitation treatmentdisclosure occurred (not necessary)
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical appearancerdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
Dreiser 1988
Methods R DB PC parallel groupsCream applied 3 times dailyAssessment at baseline 7 days
Participants Acute tendinitis (lt 1 month)N = 64M 35 F 25Mean age 36 yearsBaseline spontaneous pain ge 60 mm
Interventions Ibuprofen cream 5 3 x 4 cm daily n = 32 (3 x 10 cm for large joints)Placebo cream n = 32No other topical systemic or physical treatment allowed
Outcomes PGE scale not reported (responder = ldquoimprovementrdquo or ldquocomplete relief rdquo)Improvement in pain VAS (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35Oxford Validity Score 1016
Risk of bias
Bias Authorsrsquo judgement Support for judgement
46Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Dreiser 1988 (Continued)
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Low risk Missing participants added back usingBOCF
Size High risk lt 50 participants per treatment arm
Dreiser 1989
Methods R DB PC parallel groupsGel applied twice daily to affected area with light massage then covered with standardcompressAssessed at baseline 3 7 days
Participants Uncomplicated recent ankle sprainN = 60M 36 F 24Mean age 33 yearsMean baseline pain 54 mm
Interventions Ketoprofen gel 25 2 x 5 cm daily n = 30Placebo gel n = 30No concomitant therapy other than simple oral analgesia allowed
Outcomes PGE 3-point scale (responder = ldquobetterrdquo)Improvement in pain VAS (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk ldquodrawing lotsrdquo
Allocation concealment (selection bias) Unclear risk Not described
47Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Dreiser 1989 (Continued)
Blinding (performance bias and detectionbias)All outcomes
Low risk Treatments ldquoidentical in every way exceptthat placebo did not contain active princi-plerdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participant per treatment arm
Dreiser 1990
Methods R DB PC parallel groupsGel lightly massaged into skin over affected area 3 times daily then covered with standardcompressAssessed at baseline 3 7 days
Participants Uncomplicated ankle sprain (lt 4 days)N = 60 (59 analysed)M 29 F 29 (not stated for 1 participant)Mean age 33 yearsBaseline pain ge moderately severeExclusions 1 participant had only moderate pain at baseline
Interventions Niflumic acid gel 25 3 x 5 g daily n = 30Placebo gel n = 30Concomitant treatment with systemic NSAIDs local therapies or physiotherapy werenot allowed
Outcomes PGE 4-point scale (responder = ldquocuredrdquo or ldquoimprovedrdquo)Improvement in pain VAS (mean data)Adverse eventsWithdrawals and exclusions
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
48Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Dreiser 1990 (Continued)
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
Dreiser 1994
Methods R DB PC parallel groupsPatch applied twice dailyAssessed at baseline 3 7 days
Participants Traumatic ankle sprain (lt 2 days)N = 131M 84 F 47Mean age 34 yearsBaseline pain ge 50 mm
Interventions Flurbiprofen patch 2 x 40 mg daily n = 65Placebo patch n = 66Rescue medication paracetamol Ice or light restraint allowedExclusions 1 from flurbiprofen group for protocol violation
Outcomes PGE 4-point scale (responder = ldquogoodrdquo or ldquovery goodrdquo)Improvement in pain VAS (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk Placebo patch was ldquonon-medicated (butotherwise identical)rdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
49Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Fioravanti 1999
Methods R DB AC parallel groupsGel lightly massaged into skin 3 times daily and kept dry for 6 to 8 hoursAssessed at baseline 3 10 days
Participants Peri and extra-articular inflammatory diseasesN = 100M 32 F 68Mean age 49 yearsBaseline spontaneous pain ge 40 mm
Interventions DHEP lecithin gel 3 x 5 g (= 65 mg) daily n = 50DHEP gel 3 x 5 g (= 65 mg) daily n = 50
Outcomes PGE 4-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo)Pain on movement meanAdverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size Unclear risk 50 participants per treatment arm
Fujimaki 1985
Methods R DB PC AC parallel groupsGel applied to affected area 3 or 4 times daily with no occlusion for up to 14 daysAssessed at baseline 7 14 days
Participants Muscle pain or inflammation in neck shoulder back chest and upper and lower ex-tremities or a combinationN = 271 (247 analysed)M 97 F 149
50Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Fujimaki 1985 (Continued)
Age lt 20 to 89 yearsBaseline pain mostly mild to moderateExclusions 24 due to protocol violations loss to follow-up
Interventions Piroxicam gel 05 1 g 3 to 4 x daily n = 92Indomethacin gel 1 1 g 3 to 4 x daily n = 90Placebo gel n = 89No concomitant oral or topical analgesic or anti-inflammatory medication allowed Nophysical therapy initiated after start of study
Outcomes PGE 5-point scale (responder = ldquobetterrdquo or ldquomuch betterrdquo)Physician rated improvement 5-point scale (responder = ldquomarked improvementrdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total =45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Low risk Cartons numbered randomly and numbersheld in a key code until study completion
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical tubesrdquo packed in numbered car-ton Gel bases slightly different in appear-ance so dispensing physician did not haveaccess to them
Incomplete outcome data (attrition bias)All outcomes
Unclear risk ~ 8 excluded from analysis for unknownreasons or lost to or inadequate follow-upApproximately equal between groups
Size Unclear risk 50 to 200 participants per treatment arm
Gallacchi 1990
Methods R DB AC parallel groupsGel applied to affected area 4 times daily with light massage for 14 daysAssessment at baseline 7 14 days
Participants Painful inflammatory conditionsN = 50M 20 F 30Mean age 50 years
51Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Gallacchi 1990 (Continued)
Baseline pain ge moderate severity
Interventions Diclofenac gel 1 2 g 4 x daily n = 25 (Flector)Diclofenac sodium 1 2 g 4 x daily n = 25 (Voltaren Emugel)No other medication that could interfere with test drugs allowed
Outcomes PGE 5-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo)Improvement in pain on pressure 4-point scale (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
Gonzaacutelez de Vega 2013
Methods R DB (for gels) SB (for ointment) AC multicentre parallel groups2 g (~ 6 cm of gel) applied over injured area x 3 daily for 14 daysAssessments at 0 4 7 14 days of treatment and day 42 follow-up
Participants Grade I or II ankle sprain (lt 24 hours) with lateral external ligament involvement PIon weight bearing ge 30100N = 449M = 308 F = 112 (for analysis)Mean age 28 years
Interventions Traumeel gel 3 x 2 g daily n = 140Traumeel ointment n = 143 (not analysed in this review)Diclofenac gel 1 3 x 2 g daily n = 137Rescue medication paracetamol to maximum 2 g daily
52Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Gonzaacutelez de Vega 2013 (Continued)
Outcomes Pain-free on day 7Global efficacy 5-point scale (responder = ldquogoodrdquo and ldquovery goodrdquo) on day 14Normal function on day 14 yes or no (responder = ldquoyesrdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 (gels) W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk Computer-generated
Allocation concealment (selection bias) Low risk Central allocation kits assigned in orderreceived and used envelopes (no furtherdetails)
Blinding (performance bias and detectionbias)All outcomes
Low risk For gel comparison ldquoidentical containersrdquo
Incomplete outcome data (attrition bias)All outcomes
Unclear risk Used LOCF Most withdrawals due to earlyrecovery (within 14 days) approximatelyequal between groups
Size Unclear risk 50 to 200 participants per treatment arm
Governali 1995
Methods R DB AC parallel groupsGel or cream applied 3 times daily for up to 14 daysAssessed at baseline 7 14 days
Participants Soft tissue injuries + 2 fracturesN = 30M = 21 F = 9Median age 38 yearsMean baseline pain on movement moderate to severe (28 scale 0 to 4)
Interventions Ketoprofen gel 5 3 x 2 to 3 g daily n = 15Ketoprofen cream 1 3 x 2 to 3 g daily n = 15
Outcomes PGE 5-point scale (responder = ldquogoodrdquo and ldquoexcellentrdquo)Adverse eventsWithdrawals
53Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Governali 1995 (Continued)
Notes Oxford Quality Score R1 DB1 W1 Total =35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Treatments were given in identical tubesand measurements made by blinded ob-servers but one was a cream and the othera gel
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
Gualdi 1987
Methods R DB AC parallel groupsGel applied twice daily for 10 daysAssessed at baseline 4 7 10 days
Participants Soft tissue injuriesN = 60M = 37 F = 23Mean age 32 years (range 13 to 78)Mean baseline pain on movement moderate to severe (22 scale 0 to 3)
Interventions Flunoxaprofen gel 2 x 3 to 5 cm daily n = 30Ketoprofen gel 2 x 3 to 5 cm daily n = 30
Outcomes Improvement in pain on pressure (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
54Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Gualdi 1987 (Continued)
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Unclear risk Insufficient data
Size High risk lt 50 participants per treatment arm
Haig 1986
Methods R DB PC parallel groupsCream applied lightly to affected area 6 times daily for 6 daysAssessed at baseline 2 4 6 days
Participants Soft tissue injuries (lt 24 hours)N = 43MF not reportedAge not reportedBaseline pain not reported
Interventions Benzydamine cream 3 6 x daily n = 21Placebo cream n = 22
Outcomes Pain on movement 4-point scale (responder = ldquoimprovedrdquo)Adverse events
Notes Oxford Quality Score R1 DB2 W0 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not reported
Allocation concealment (selection bias) Unclear risk Not reported
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquomatching placebordquo
55Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Haig 1986 (Continued)
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants apparently included inanalysis
Size High risk lt 50 participants per treatment arm
Hoffmann 2012
Methods R DB PC multicentre parallel groupPlaster applied x 1 daily (minimum of 20 consecutive hours of application per day) for14 daysPI assessed daily overall efficacy assessed at 7 and 14 days
Participants Unilateral mild to moderate muscle contusion of upper or lower limb (lt 72 hours) PIge 50100 superficial haematoma le 10 x 14 cm at injured siteN = 354M 126 F 228Mean age 39 years
Interventions DHEPhep plaster x 1 daily n = 121DHEP plaster x 1 daily n = 115Placebo plaster n = 118Rescue medication paracetamol 500 mg (no limit reported)
Outcomes PGE 5-point scale (responder = excellent or good) at 14 daysMean reduction in PI at 3 and 8 daysRescue medicationAdverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Method not reported
Allocation concealment (selection bias) Unclear risk Method not reported ldquoMedication packedand labelled in order to render all partici-pants and personnel fully blinded to treat-ment administeredrdquo
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoIndistinguishable regarding appearanceshape colour size and odourrdquo
56Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Hoffmann 2012 (Continued)
Incomplete outcome data (attrition bias)All outcomes
Unclear risk No responder analysis reported and impu-tation method not mentioned
Size Unclear risk 50 to 200 participants per treatment arm
Hofman 2000
Methods R DB AC parallel groupsGel applied to affected region 4 times daily with gentle massageAssessed at baseline 8 days in clinic and daily diary
Participants Soft tissue articular pain (le 15 days)N = 142M 19 F 123Mean age 57 yearsMean baseline PI moderate to severe
Interventions Diclofenac sodium gel 1 4 x 2 cm daily n = 69Lysine clonixinate gel 5 4 x 2 cm daily n = 73(2 cm = 225 mg)No other analgesic local treatment (including immobilisation bandaging) or acupunc-tureRescue mediation allowed after 2 applications if needed
Outcomes PGE 3-point scale (ldquogoodrdquo)PI participant diary (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk Diclofenac gel repackaged to maintain DBwith lysine clonixinate gel Minor differ-ences between gels only apparent when di-rectly compared
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
57Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Hofman 2000 (Continued)
Size Unclear risk 50 to 200 participants per treatment arm
Hosie 1993
Methods R DB AC parallel groupsFoam (approximately the size of a golf ball) applied and 1 tablet taken 3 times daily for7 days and up to 14 daysAssessed at baseline 7 14 (if necessary) days
Participants Acute lower back injury (lt 1 month)N = 287 (261 analysed for efficacy)M 151 F 136Mean age 37 years (range 18 to 63)Most participants had moderate to severe pain on movement one had noneExclusions 25 lost to follow up one assessed at 14 days but not 7 days
Interventions Felbinac foam 3 3 x 2 g daily + placebo tablets 3 x 1 daily n = 140 (127 analysed forefficacy)Ibuprofen tablets 3 x 400 mg daily + placebo foam 3 x 2 g daily n = 147 (134 analysedfor efficacy but one had no pain at baseline)No other oral injectable or topical analgesic or anti-inflammatory medication Ongoingphysiotherapy to continue without change
Outcomes Pain on movement 5-point scale (responder = ldquononerdquo or ldquomildrdquo)Spontaneous pain 5-point scale (responder = ldquononerdquo or ldquomildrdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquodouble dummyrdquo
Incomplete outcome data (attrition bias)All outcomes
Unclear risk Imputation method not mentioned
Size Unclear risk 50 to 200 participants per treatment arm
58Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Jenoure 1997
Methods R DB PC parallel groupsPlaster applied to skin over affected area twice daily and kept in place with an elasticbandageAssessed at baseline 7 14 days and after further 14 days without treatment
Participants Humero-radial epicondyl pain (tendinopathic) - nearly all tennis elbowN = 85M 54 F 31Mean age 45 yearsBaseline pain ldquomildrdquo in ~ 10 of placebo group and 29 of active group
Interventions DHEP plaster (Tissugel) x 2 daily n = 44Placebo plaster x 2 daily n = 41
Outcomes Pain on pressure 5-point scale (responder = ldquononerdquo or ldquomildrdquo)Spontaneous pain 5-point scale (responder = ldquono painrdquo)Adverse events
Notes Oxford Quality Score R1 DB2 W0 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not reported
Allocation concealment (selection bias) Unclear risk Not reported
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical characteristicsrdquo
Incomplete outcome data (attrition bias)All outcomes
Unclear risk No useable data
Size High risk lt 50 participants per treatment arm
Joussellin 2003
Methods R DB PC parallel groupsPlaster applied to skin over affected area once dailyAssessed at baseline 1 2 3 7 days
Participants Ankle sprain (lt 48 hours)N = 134M 72 F 62Age range 18 to 65 years
59Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Joussellin 2003 (Continued)
Baseline spontaneous pain ge 50 mm
Interventions DHEP plaster (Flector Tissugel 1) x 1 daily n = 68Placebo plaster x 1 daily n = 66Rescue medication paracetamol
Outcomes PGE 4-point scale (responder = ldquoexcellentrdquo)Pain on movement VAS (mean)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoIdentical in size appearance and usedsame formula as active patch without ac-tive ingredientrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in responder anal-ysis
Size Unclear risk 50 to 200 participants per treatment arm
Julien 1989
Methods R DB PC parallel groupsGel applied to affected area twice daily with light massageAssessed at baseline 3 7 days in clinic and daily diary
Participants TendinitisN = 60M 29 F 31Mean age 41 yearsBaseline pain gt 50 mm
Interventions Ketoprofen gel 25 2 x 5 cm (= 50 mg) daily n = 30Placebo gel 2 x 5 cm daily n = 30No concomitant therapy other than simple analgesia
60Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Julien 1989 (Continued)
Outcomes PGE 4-point scale (responder = ldquoimprovedrdquo or ldquorecoveredrdquo)Pain on movement 4-point scale (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB1 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk Random numbers table
Allocation concealment (selection bias) Unclear risk Randomisation code supplied by Menarinilaboratories remote from allocation
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in responder anal-ysis
Size High risk lt 50 participants per treatment arm
Klainguti 2010
Methods R DB PC multicentre parallel groupPlaster applied x 1 daily (minimum 20 consecutive hours of application daily) for 10daysPI assessed daily Overall treatment efficacy and tolerability assessed at days 3 and 7
Participants Unilateral mild to moderate muscle contusion or strain of upper or lower limb (lt 72hours) superficial haematoma le 140 cm2 PI ge 40100N = 185M 90 F 95Mean age 39 years
Interventions DHEPhep plaster x 1 daily n = 65DHEP plaster x 1 daily n = 61Placebo plaster x 1 daily n = 59
Outcomes Overall treatment efficacy at 3 days 5-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo)Resolution of haematoma at 10 days yes or noRescue medicationAdverse eventsWithdrawals
61Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Klainguti 2010 (Continued)
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk Computer-generated
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoIdentical in size appearance and odourrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk Withdrawals lt 5 All participants in-cluded in analysis
Size Unclear risk 50 to 200 participants per treatment arm
Kockelbergh 1985
Methods R DB PC parallel groupsGel applied twice dailyAssessed at baseline 3 7 days
Participants Acute soft tissue trauma (lt 24 hours)N = 74M 60 F 14Mean age 27 yearsBaseline pain gt 65 mm
Interventions Ketoprofen gel 25 2 x 5 cm (= 15 mg) daily n = 38Placebo gel 2 x 5 cm daily n = 36No concomitant treatmentRescue medication glafenine
Outcomes PGE 3-point scale (responder = ldquogoodrdquo)Spontaneous pain 100 mm VAS (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
62Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Kockelbergh 1985 (Continued)
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in responder anal-ysis
Size High risk lt 50 participants per treatment arm
Kuehl 2011
Methods R DB PC multicentre parallel groupsPlaster applied x 2 daily for 14 days or until resolutionPI assessed daily Overall treatment efficacy assessed at 14 days
Participants Mild or moderate sprain strain or contusion (lt 7 days) PI ge 510N = 418M 206 F 212Mean age 39 years
Interventions DHEP patch x 2 daily n = 207Placebo patch x 2 daily n = 211Rescue medication no analgesic allowed (or icewrapping) use = discontinuation
Outcomes Resolution of injury (VAS le 210)Overall tolerability 5-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk Computer-generated
Allocation concealment (selection bias) Unclear risk Not described
63Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Kuehl 2011 (Continued)
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical in appearance with same contentexcept diclofenacrdquo
Incomplete outcome data (attrition bias)All outcomes
High risk Used LOCF high withdrawal rate (43)Did not report all efficacy outcomes mea-sured
Size Low risk gt 200 participants per treatment arm
Li 2013
Methods R DB PC multicentre parallel groupsPlaster applied x 2 daily for 7 daysPI assessed daily overall treatment efficacy and tolerability assessed at 7 days
Participants Mild or moderate ankle or knee sprain muscle strain or contusion (lt 72 hours) PI ge
50100N = 384M 144 F 240Mean age 42 years
Interventions DHEP plaster x 2 daily n = 192Placebo plaster x 2 daily n = 192Rescue medication paracetamol to maximum 2 g daily
Outcomes ge 50 reduction in PI at 7 daysOverall treatment efficacy 5-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo)Overall tolerability 5-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo)Rescue medicationAdverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk Computer-generated
Allocation concealment (selection bias) Unclear risk Method not fully described ldquosealed en-velopesrdquo ldquoconcealed from investigators andparticipantsrdquo
64Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Li 2013 (Continued)
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical in texture size color and odorrdquo
Incomplete outcome data (attrition bias)All outcomes
Unclear risk Withdrawals 2 Methods states LOCFbut appears to describe BOCF for with-drawals
Size Unclear risk 50 to 200 participants per treatment arm
Linde 1985
Methods R DB PC parallel groupsCream applied x 3 daily for 5 days with elastic support for the first 3 daysAssessed at baseline 4 8 days
Participants Sprained ankle (lt 24 hours)N = 100M 58 F 42Mean age 28 yearsBaseline pain all participants had ldquowalking painrdquo
Interventions Benzydamine 3 cream x 3 daily n = 50Placebo gel x 3 daily n = 50
Outcomes Pain on movement responder = ldquofree of walking painrdquoAdverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35Paper describes a benzydamine cream and a placebo gel
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described Paper describes a benzy-damine cream and a placebo gel
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in responder anal-ysis
65Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Linde 1985 (Continued)
Size Unclear risk 50 participants per treatment arm
Machen 2002
Methods R DB PC parallel groupsGel gently (ldquominimal rubrdquo not vigorously) massaged into skin over affected site untilabsorbed 3 times daily until symptoms disappeared or for maximum of 7 daysAssessment at baseline and once daily using diary cards to 7 days
Participants Soft tissue injury (lt 2 weeks and untreated)N = 85 (81 analysed)M 42 F 39Mean age 41 yearsBaseline pain gt 50 mm4 placebo participants lost to follow-up
Interventions Ibuprofen gel 5 x 3 daily n = 40Placebo gel x 3 daily n = 41Initiation of other medication or physiotherapy not allowed during study
Outcomes PGE 5-point scale (responder = ldquomarked improvementrdquo or ldquocomplete clearancerdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk Gels had similar physical characteristicsand were supplied in identical tubes
Incomplete outcome data (attrition bias)All outcomes
Low risk lt 5 missing participants Remaining par-ticipants included in responder analysis
Size High risk lt 50 participants per treatment arm
66Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Mahler 2003
Methods R DB AC parallel groupsGel applied with gentle massage to affected area 3 times daily without occlusion for 10daysAssessed at baseline 3 10 days in clinic and daily patient diary
Participants First-degree ankle or knee sprains first-degree muscle strains and mild-to-moderatecontusionsn = 100M 69 F 31Mean age 32 yearsMean baseline pain with activity ge 65 mm
Interventions DHEP lethicin gel 3 x 5 g (= 65 mg) daily n = 52DHEP gel 3 x 5 g (= 65 mg) daily n = 48All participants treated with ice at site of inflammation for first 48 hours but no immo-bilisation allowedRescue medication paracetamol 500 mg if strictly necessary
Outcomes PGE 4-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo)Pain on movement 100-mm VAS (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 5
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk ldquoComputer-generated randomization listrdquo
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk Pharmaceutically inert colouring agentsadded to reference formulation so that gelswere indistinguishable
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in responder anal-ysis
Size High risk lt 50 participants in each treatment arm
67Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Maziegraveres 2005a
Methods R DB PC parallel groupsNew patch applied directly to skin over painful area each morningAssessed at baseline 3 7 14 days
Participants Symptomatic tendonitis in upper or lower limbs not requiring surgery (le 15 days)N = 172M 72 F 100Mean age 46 yearsBaseline pain with activity ge 40 mm
Interventions Ketoprofen patch 100 mg x 1 daily n = 87Placebo patch x 1 daily n = 85No analgesic or steroid by any route or other topical medication or physical therapyallowedRescue medication permitted but not within 12 hours of assessment
Outcomes PGE 4-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk ldquocomputer generated global randomizationcoderdquo
Allocation concealment (selection bias) Low risk ldquoThe randomization list and code en-velopes were prepared by the company ap-pointed for clinical supplies packaging Therandom code was disclosed only after studycompletion and database closurerdquo
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquothe same indistinguishable patch with noingredientrdquo
Incomplete outcome data (attrition bias)All outcomes
High risk LOCF withdrawals 12 All participantsincluded in analysis
Size Unclear risk 50 to 200 participants per treatment arm
68Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Maziegraveres 2005b
Methods R DB PC parallel groupsNew patch applied directly to skin over painful area each morningAssessed at baseline 3 7 14 days
Participants Painful benign ankle sprain (le 48 hours)N = 163M 83 F 80Mean age 37 yearsBaseline spontaneous pain ge 50 mm
Interventions Ketoprofen patch 100 mg x 1 daily n = 81Placebo patch x 1 daily n = 82No analgesic or steroid by any route or other topical medication or physical therapyallowedRescue medication permitted but not within 12 hours of assessment
Outcomes PGE 4-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk ldquocomputer-generated global randomiza-tion coderdquo
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk The same transdermal patch with no activeingredient
Incomplete outcome data (attrition bias)All outcomes
Unclear risk LOCF withdrawals 5 to 6 All partic-ipants included in analysis
Size Unclear risk 50 to 200 participants per treatment arm
McLatchie 1989
Methods R DB PC parallel groupsGel applied to injured site 3 times daily for 7 daysAssessment at baseline 4 7 days at clinic daily patient diary
Participants Acute soft tissue injury (lt 48 hours)N = 231
69Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
McLatchie 1989 (Continued)
M 143 F 88Mean age 33 yearsBaseline pain moderate to severe
Interventions Felbinac gel 3 3 x 3 cm daily n = 118Placebo gel 3 x 3 cm daily n = 113Rescue medication paracetamol
Outcomes Patient diary mean changeAdverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquotubes identical in all aspectsrdquo
Incomplete outcome data (attrition bias)All outcomes
Unclear risk No usable efficacy data
Size Unclear risk 50 to 200 participants per treatment arm
Morris 1991
Methods R DB PC parallel groupsGel applied to site of injury 3 times daily for 7 daysAssessed at baseline 7 days at clinic and daily patient diary
Participants Acute soft tissue injury (lt 3 days)N = 100 (84 analysed for efficacy)M 70 F 14Mean age 25 yearsBaseline pain moderate to severeExclusions 1 participant in placebo group lost to follow-up 15 protocol violations
Interventions Felbinac gel 3 3 x 1 cm daily n = 41Placebo gel n = 43Ice joint immobilisation bandaging and compression allowed
70Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Morris 1991 (Continued)
No concomitant oral NSAID occlusive dressing physiotherapy or liniments allowedRescue medication paracetamol
Outcomes PGE 5-point scale (responder = ldquogoodrdquo and ldquovery goodrdquo)Change in PI patient diary 10 cm VAS (mean data)Adverse eventsWithdrawals and exclusions
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Low risk ldquoRandomisation was undertaken at theproduction facility and a sealed copy of thelist supplied to the investigator for refer-ence only in defined circumstancesrdquo
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical tubes and outer boxesrdquo ldquoplacebowas a similarly constituted gelrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All relevant participants included in analy-sis
Size High risk lt 50 participants per treatment arm
NCT01255423
Methods R DB PC parallel group studyGel applied 4 x daily
Participants Acute lateral ankle sprain Grade I to II le 12 hoursN = 206Mean (plusmn SD) age 31 years (plusmn 13)M 87 F 119Baseline PI not reported
Interventions Diclofenac sodium gel 1 x 4 daily n = 104Placebo gel n = 102
Outcomes Mean PI on movementAdverse events
71Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT01255423 (Continued)
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Method of blinding not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants accounted for
Size Unclear risk 50 to 200 participants per treatment arm
NCT01272934
Methods R DB PC parallel group studyGel applied 4 x daily
Participants Acute lateral ankle sprain Grade I to II le 12 hoursN = 205M 101 F 104Mean (plusmn SD) age 32 years (plusmn 11)
Baseline PI not reported
Interventions Diclofenac sodium gel 1 x 4 daily n = 102Placebo gel x 4 daily n = 103
Outcomes Mean PI on movementAdverse events
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
72Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT01272934 (Continued)
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Method of blinding not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants accounted for
Size Unclear risk 50 to 200 participants per treatment arm
NCT01272947
Methods R DB PC parallel groupsGel applied 4 x daily
Participants Acute blunt soft tissue injuriescontusions of the limbs lt 3 hoursN = 204M 101 F 103Mean (plusmn SD) age 30 years (plusmn 11)
Baseline PI not reported
Interventions Diclofenac sodium gel 1 x 4 daily n = 104Placebo gel x 4 daily n = 100
Outcomes Mean PI on movementAdverse events
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Method of blinding not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants accounted for
73Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT01272947 (Continued)
Size Unclear risk 50 to 200 participants per treatment arm
Noret 1987
Methods R DB PC parallel groupsGel applied twice daily for 7 daysAssessment at baseline 3 8 days
Participants Minor sports injuries (lt 24 hours)N = 98 (93 analysed)M 71 F 27Mean age 29 yearsBaseline pain gt 60 mm
Interventions Ketoprofen gel 25 2 x 5 cm daily (= 15 mg) n = 48Placebo gel n = 45No other treatment given
Outcomes PGE 4-point scale (responder = ldquogoodrdquo and ldquoexcellentrdquo)Spontaneous pain 100 mm VAS (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Low risk ldquoallocated according to a randomization listand a corresponding code in a sealed enve-loperdquo
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
74Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Parrini 1992
Methods R DB PC parallel groupsFoam (the size of a walnut or a 1-second spray) applied with massage 3 x daily for 7days
Participants Articular trauma strains distortionsN = 169M 94 F 75Mean age 37 yearsMean baseline pain on movement 31 (scale 1 to 4)
Interventions Ketoprofen foam 15 3 x 2 g (= 600 mg) daily n = 83Placebo foam 3 x 2 g n = 86
Outcomes Pain on movement 4-point scale (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB1 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk ldquopatients were randomised according to themethod of random numbersrdquo [translated]
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Unclear risk No usable efficacy data
Size Unclear risk 50 to 200 participants per treatment arm
Picchio 1981
Methods R DB AC parallel groupsCream applied with slight massage until completely absorbed 3 times daily for up to 16daysAssessed at baseline 4 8 12 16 days
Participants Acute sports injuriesN = 40M 24 F 16Mean age 22 years (range 12 to 46)Most participants had mild to moderate baseline pain (12 and 9 with slight pain on
75Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Picchio 1981 (Continued)
movement)
Interventions Ibuprofen gel 10 3 x daily n = 20Ketoprofen gel 1 3 x daily n = 20
Outcomes Pain on movement (responder = ldquononerdquo)Adverse events
Notes Oxford Quality Score R1 DB2 W0 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk ldquotubes were identical in appearancerdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
Predel 2004
Methods R DB PC parallel groupsNew patch applied to injured area twice daily for 7 days Contact of patch with humidityor water to be avoidedAssessment at baseline 3 7 days
Participants Traumatic blunt soft tissue injuries (lt 3 hours no treatment)N = 120M 73 F 47Mean age 32 yearsBaseline pain gt 60 mm
Interventions Diclofenac sodium patch 2 x daily (140 mg per patch) n = 60Placebo patch 2 x daily n = 60NSAIDs analgesics psychotropic agents other topical preparations and bandages notallowed
Outcomes PGE 4-point scale (responder = rdquogoodldquo and rdquoexcellentldquo)Pain on movement 10 cm VAS (mean data)
76Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Predel 2004 (Continued)
Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk rdquocomputer generated block randomisationlist
Allocation concealment (selection bias) Low risk An independent statistician produced ran-domisation list and an independent con-tract research organisation packaged med-ication according to list Nobody else hadaccess to the randomisation list until thedatabase was closed
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoThe placebo patch was visually indistin-guishable from the active patchrdquo To avoidunblinding due to different smell anystudy nurse involved with medication wasnot involved in outcome assessment
Incomplete outcome data (attrition bias)All outcomes
Low risk Withdrawals lt 1 All participants in-cluded in analysis
Size Unclear risk 50 to 200 participants per treatment arm
Predel 2012
Methods R DB PC parallel groupsTreatment for 7 daysAssessed at baseline 1 3 5 and 8 days
Participants Grade I or II ankle sprain with lateral external ligament involvement lt 12 hoursPI on movement ge 50100N = 242M 152 F 90Mean age 32 years
Mean baseline pain on movement = 75100
Interventions Diclofenac gel (Voltaren Emulgel 232) 2 x 5 cm + 1 x 5 cm placebo gel daily n = 80Diclofenac gel (Voltaren Emulgel 232) 3 x 5 cm daily n = 80Placebo gel 3 x 5 cm daily n = 82Rescue medication paracetamol (to maximum 2 g daily)
77Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Predel 2012 (Continued)
No ice or bandages after randomisation
Outcomes ge 50 reduction in PI on movement at 5 daysPGE 5-point scale (responder = ldquogoodrdquo or rsquovery goodldquo)Patient satisfaction 5-point scale (responder = rdquogoodldquo or rsquovery goodrdquo or ldquoexcellentrdquo)Rescue medicationAdverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical in composition appearance tex-ture and smellrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk LOCF but withdrawals lt 3 in all treat-ment arms
Size Unclear risk 50 to 200 participants per treatment arm
Predel 2013a
Methods R DB PC parallel groupsTreatment for 14 daysAssessed at baseline 3 7 10 and 14 days (plusmn 1 day)
Participants Uncomplicated one-sided ankle sprain with swelling ge 12 mm (2 to 18 hours)PI on movement tenderness joint mobility (scale 0 to 3) summed as ge 5 and le 7N = 232M 126 F 106Mean age 29 years
Interventions Diclofenac 4 spray gel 3 x 4 to 5 sprays daily (96 to 120 mg diclofenac sodium) n =118Placebo spray gel 3 x 4 to 5 sprays daily n = 114Rescue medication paracetamol 500 mg (maximum 10 tablets per week)No ice or bandaging allowed
78Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Predel 2013a (Continued)
Outcomes None or slight PI on movement at 3 and 7 daysPGE 4-point scale (responder = ldquovery goodrdquo at 14 days)gt 50 reduction in swelling at 10 daysAdverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk ldquoplacebo (vehicle only no active ingredi-ent)rdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk LOCF for full or early cure Withdrawalslt 2
Size Unclear risk 50 to 200 participants per treatment arm
Predel 2013b
Methods R DB PC parallel groups2 g gel applied with fingertips over affected area and massaged in for 1 minute Treatmentfor 5 daysAssessed at baseline 2 3 5 days
Participants Uncomplicated neck pain originating from cervical joints and accompanying soft tissues(ge 12 hours but lt 3 months) PI ge 50100N = 72M 33 F 39Mean age 34 years
Interventions Diclofenac gel (Voltaren Emulgel) 4 x 2 g daily n = 36Placebo gel 4 x 2 g daily n = 36Rescue medication paracetamol up to 2 g dailyNo concomitant therapies allowed
Outcomes ge 50 decrease in PI on movement after 48 hoursPGE 5-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo at 5 daysAdverse eventsWithdrawals
79Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Predel 2013b (Continued)
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk Sponsor ldquoproduced randomisation listrdquo
Allocation concealment (selection bias) Low risk Automated remote system
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical in packaging labelling scheduleof administration appearance and odourrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk No missing data
Size High risk lt 50 participants per treatment arm
Ramesh 1983
Methods R DB PC parallel groupsCream applied to painful area and rubbed into skin over a large area for up to 10 daysAssessment at baseline 3 7 10 days
Participants Strains sprains contusions compressionsN = 80M 42 F 38Age 11 to 81 yearsBaseline pain 5 ibuprofen 2 placebo participants had no or slight pain
Interventions Ibuprofen cream 5 3 to 4 x 5 to 10 cm daily n = 40Placebo cream 3 to 4 x 5 to 10 cm daily n = 40Adjuvant therapy was not administered
Outcomes Pain on movement 4-point scale (responder = ldquononerdquo or ldquoslightrdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
80Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Ramesh 1983 (Continued)
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Low risk Randomisation key in sealed envelopeavailable for emergencies but opened onlyafter completion
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical appearance and odourrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
Rowbotham 2003
Methods R DB PC parallel groupsNew patches applied to the affected painful area for 12 consecutive hours twice dailyfor up to 14 daysAssessed at baseline 14 days in clinic and daily patient diary
Participants Minor sports injuries (sprains sprains contusions lt 72 hours)N = 372M 253 F 119Mean age 33 yearsBaseline pain at rest ge 510
Interventions DHEP patch (Flector Tissuegel) 2 x daily (equivalent to diclofenac sodium 140 mg perpatch) n = 191Placebo patch 2 x daily n = 181
Outcomes PGE 5-point scale (responder = ldquogoodrdquo and ldquoexcellentrdquo)Pain resolved lt moderate for 2 daysSpontaneous pain 10 cm VAS (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
81Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Rowbotham 2003 (Continued)
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk ldquoSystegraveme identiquerdquo without diclofenac
Incomplete outcome data (attrition bias)All outcomes
Unclear risk No information about imputation All par-ticipants included in analysis
Size Unclear risk 50 to 200 participants per treatment arm
Russell 1991
Methods R DB PC parallel groupsAffected area washed with soap and water and dried then gel applied and carefullyrubbed into skin 4 times daily for at least 7 daysAssessed at baseline 4 8 15 (if necessary) days at clinic and daily patient diary
Participants Acute soft tissue injuries (recent not recurrent)N = 214 (200 analysed)M 95 F 105Mean age 40 yearsBaseline pain gt 65 mm
Interventions Piroxicam gel 05 4 x 5 mg daily n = 100Placebo gel n = 100No other NSAIDs or analgesic drugs including liniments containing salicylates allowedAncillary therapy at the discretion of the investigator
Outcomes PGE 4-point scale (responder = ldquogoodrdquo and ldquoexcellentrdquo)Spontaneous pain mean reductionAdverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk ldquocomputer generated randomization coderdquo
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical base formulationrdquo
82Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Russell 1991 (Continued)
Incomplete outcome data (attrition bias)All outcomes
High risk High withdrawal rate (8 with piroxicam50 with placebo) No information aboutimputation
Size Unclear risk 50 to 100 participants per treatment group
Saillant 1998
Methods R DB PC parallel groupsTreatment for 7 daysAssessed at baseline 1 2 3 and 7 days
Participants Ankle sprain (lt 48 hours)N = 140M 72 F 62Age 18 to 65 yearsBaseline spontaneous pain ge 50 mm
Interventions (1) DHEP plaster (Flector Tissugel 1) 1 x daily n = 70(2) Placebo plaster 1 x daily n = 70Rescue medication paracetamol (to maximum 3 g daily)Ice allowed
Outcomes ge 30 decrease in PI at 7 daysPGE 4-point scale (responder = ldquoexcellentrdquo)No or low pain on passive stretchSingle foot leaning OK without painWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk Identical in size appearance and used sameformula as active patch without active in-gredient
Incomplete outcome data (attrition bias)All outcomes
Unclear risk Used LOCF for primary outcome With-drawals lt 7 and equal between groupsAll participants included in analysis
83Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Saillant 1998 (Continued)
Size Unclear risk 50 to 200 participants per treatment arm
Sanguinetti 1989
Methods R DB PC parallel groupsGel applied 3 times daily for 7 consecutive daysAssessment at baseline 7 days
Participants Soft tissue trauma (lt 48 hours)N = 82M = 47 F = 35Mean age 34 yearsBaseline pain moderate to severe
Interventions Felbinac gel 3 3 x daily n = 42Placebo gel 3 x daily n = 40No other NSAID steroid other topical application allowedRescue medication paracetamol felbinac is an active metabolite of the NSAID fenbufen
Outcomes PGE scale not reported (responder = ldquogoodrdquo and ldquovery goodrdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoindistinguishable in appearance colour orodourrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
84Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Sinniger 1981
Methods R DB PC parallel groupsCream applied 2 to 3 times daily with gentle massage or if massage not possible (toopainful) with protective dressingAssessment at baseline 5 10 days
Participants Minor soft tissue injuriesN = 20M 11 F 9Mean age 40 yearsBaseline pain not reported
Interventions Fentiazac cream 5 2 to 3 x daily n = 10Placebo cream 2 to 3 x daily n = 10All participants told to restNo other local and systemic treatments allowedRescue medication analgesic if actually needed
Outcomes Pain relief scale not reported (responder = total pain relief ) improvement in pain on movement pain scale not reported (mean data)Adverse events
Notes Oxford Quality Score R1 DB1 W0 Total = 25
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
85Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Spacca 2005
Methods R DB PC parallel groupsGel applied 3 times daily with gentle massage until complete absorption for up to 10daysAssessment at baseline 10 days in clinic and daily patient diary
Participants Shoulder periarthritis or lateral epicondylitis (lt 5 days)N = 155M 74 F 81Mean age 51 yearsBaseline pain with activity gt 70 mm
Interventions DHEP lecithin gel (Effigel) 3 x 5 g daily n = 79Placebo gel 3 x 5 g daily n = 76Rescue medication (paracetamol) allowed if pain unbearableNo other analgesic or anti-inflammatory drug allowed
Outcomes Improvement in pain 100 mm VAS (mean data)Adverse events
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Unclear risk No usable efficacy data
Size Unclear risk 50 to 200 participants per treatment arm
Sugioka 1984
Methods R DB AC parallel groupsGel applied to affected area 3 to 4 times daily without occlusion for 14 daysAssessed at baseline 7 14 days
Participants Non-traumatic diseases of muscle or tendonN = 366 (340 analysed for efficacy)M 115 F 202 (completers)
86Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Sugioka 1984 (Continued)
Age range 12 to 84 years (most 30 to 70)Baseline pain on movement ldquononerdquo or ldquomildrdquo in about 13 of participantsExclusions for protocol violations 8 piroxicam 18 indomethacin
Interventions Piroxicam gel 05 3 to 4 x 1 g daily n = 183Indomethacin gel 1 3 to 4 x 1 g daily n = 183No concomitant anti-inflammatory or analgesic drug including steroids or initiationof physical therapy allowed
Outcomes PGE 5-point scale (responder = ldquobetterrdquo or ldquomuch betterrdquo)Pain on movement 4-point scale (responder = ldquoreducedrdquo or ldquodisappearedrdquo)
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Low risk Key code sealed and retained until end ofstudy
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoboth packages were of the same appear-ance and indistinguishablerdquo and investiga-tors did not see contents
Incomplete outcome data (attrition bias)All outcomes
Low risk All relevant participants included in analy-sis
Size Unclear risk 50 to 200 participants per treatment arm
Thorling 1990
Methods R DB PC parallel groupsParticipants given specific instructions on how to apply gel (not reported) to affectedarea 2 to 6 times daily as requiredAssessment at baseline 3 7 days in clinic
Participants Soft tissue injuries (lt 48 hours)N = 120M 85 F 35Mean age 27 yearsBaseline pain moderate to severe
87Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Thorling 1990 (Continued)
Interventions Naproxen gel 10 2 to 6 x daily n = 60Placebo gel 2 to 6 x daily n = 60Rescue medication paracetamol 500 mg
Outcomes PGE 5-point scale (responder = ldquogoodrdquo and ldquovery goodrdquo)Pain on passive movement 4-point scale (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquosupplied in unmarked tubesrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size Unclear risk 50 to 200 participants per treatment arm
Tonutti 1994
Methods R DB AC parallel groupsGel applied 3 times daily for 2 to 3 weeksAssessed at baseline and intervals of 7 days
Participants Muscle or joint traumaN = 30M 20 F 10Mean age 34 years1 participant had injury of mild severity Mean baseline pain on active movement 28(scale 0 to 4)
Interventions Ketoprofen gel 5 3 x 2 to 3 g daily n = 15Etofenamate gel 5 3 x 2 to 3 g n = 15No concomitant treatment with NSAID aspirin steroid or physical therapy
88Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Tonutti 1994 (Continued)
Outcomes PGE 4-point scale (responder = ldquogoodrdquo and ldquoexcellentrdquo)Pain on movement 5-point scale (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not reported
Allocation concealment (selection bias) Unclear risk Not reported
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquothe two drugs were packed in indistin-guishable tubesrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
Vecchiet 1991
Methods R DB PC parallel groupsGel applied to the skin on and around painful area and gently rubbed in until absorbedtwice daily for up to 10 daysAssessed at baseline 5 10 days
Participants Soft tissue trauma (minor sports injuries)N = 60M 60Mean age 25 yearsMean baseline pain on active movement moderate
Interventions Meclofenamic acid gel 5 2 x 10 cm daily (= 4 g) n = 30Placebo gel 2 x 10 cm daily n = 30Both groups treated with ice rest and bandage for first 48 hours before starting testtreatmentRescue medication paracetamol
Outcomes PGE 4-point scale (responder = ldquogoodrdquo and ldquoexcellentrdquo)Pain on movement 4-point scale (mean data)Withdrawals
89Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Vecchiet 1991 (Continued)
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
Whitefield 2002
Methods R DB (double dummy) AC parallel groupsGel applied to affected site with gentle massage and 1 tablet taken 3 times daily for atleast 7 daysAssessed at baseline 7 14 (if necessary) days in clinic and daily patient diary
Participants Soft tissue injuries (lt 24 hours)N = 100M 95 F 5Mean age 26 years (range 18 to 50)Mean baseline pain on movement 22 cm
Interventions Ibuprofen gel 5 + placebo tablet 3 x daily n = 50Ibuprofen 400 mg tablet + placebo gel 3 x daily n = 50No other medication or physical therapy was prescribed and no other analgesics wereallowed
Outcomes PGE 3-point scale (responder = ldquoexcellentrdquo)Change in condition of injury site 5-point scale (responder = ldquocompletely betterrdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
90Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Whitefield 2002 (Continued)
Random sequence generation (selectionbias)
Unclear risk Not reported
Allocation concealment (selection bias) Unclear risk Not reported
Blinding (performance bias and detectionbias)All outcomes
Low risk Placebo tablets Identical in appearance toactive tablets Active and placebo gels hadsimilar physical characteristics and weresupplied in identical tubes
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size Unclear risk 50 participants per treatment arm
kermark 1990
Methods R DB (double dummy) PC AC parallel groupsSpray applied to affected area and capsules taken 3 times daily for 2 weeksAssessment at baseline 3 or 4 7 and 14 days
Participants Superficial overuse sports injuries (symptom onset 74 weeks)N = 70M 44 F 18 (completers)Mean age 30 yearsBaseline pain on palpation mostly slight to moderate
Interventions Elmetacin spray (indomethacin 1) 3 to 5 x 05 to 15 ml daily + placebo capsules n= 23Indomethacin capsules 3 x 25 mg daily + placebo spray n = 23Placebo spray and capsules n = 24Rescue medication paracetamol
Outcomes No pain on palpation (= responder)Participant improvement 100 mm VAS (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk ldquorandom number coderdquo
91Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
kermark 1990 (Continued)
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical in appearancerdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk Responder analysis lt 5 withdrawals
Size High risk lt 50 participants per treatment group
AC active control BOCF baseline observation carried forward DB double-blind DHEP diclofenac hydroxyethylpyrrolidine ordiclofenac epolamine F female HCl hydrochloride hep heparin LOCF last observation carried forward M male N numberof participants in study n number of participants in treatment arm PC placebo control PGE Participant Global Evaluation PIpain intensity R randomised VAS visual analogue scale W withdrawals
Characteristics of excluded studies [ordered by study ID]
Study Reason for exclusion
Ambrus 1987 No usable dichotomous data
Anon 1993 Not double-blind
Ascherl 1982 No usable dichotomous data
Bagliani 1976 Not an RCT
Baracchi 1982 No usable data
Burnham 1998 lt10 participantstreatment arm in first period of crossover study
Boumlhmer 1995 Active control invalid
Cesarone 2008 Not an RCT
Coulibaly 2009 Not double-blind
Diebschlag 1985 No usable dichotomous data
Diebschlag 1986 Inappropriate randomisation
Diebschlag 1992 No usable dichotomous data
92Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Fantato 1971 No usable dichotomous data
Galer 2000 No usable data
Hallmeier 1986 Not double-blind
Hallmeier 1988 Not double-blind
Kaneko 1999 Inappropriate randomisation - quasi-randomised
Kockelbergh 1985b Treatment not applied daily
Kuwabara 2013 Used NSAID-lidocaine combination (conference abstract)
Lee 1991 Not an RCT
Link 1996 No usable dichotomous data
May 2007 No usable dichotomous data
Oakland 1993 Inappropriate comparator
Odaglia 1987 Not an RCT
Picardi 1993 Not an RCT
Taboada 1992 Dose and duration of treatment unclear
Vanderstraeten 1990 Not double-blind
Vinciguerra 2008 Not an RCT
Von Klug 1977 Chronic and acute outcomes combined
RCT randomised controlled trial
Characteristics of studies awaiting assessment [ordered by study ID]
NCT00351104
Methods R DB PC parallel groups
Participants Ankle sprain or strain Grade I or II Age ge 18 yearsN = 220
93Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT00351104 (Continued)
Interventions Ketoprofen patch 20 x 1 dailyPlacebo patch x 1 dailyTreatment for 21 days
Outcomes Assessments days 3 7 21PI (0-10) at rest and during activitiesPGEFunctionRescue medication
Notes Sponsor Endo PharmaceuticalsEstimated completion February 2007 No results
NCT00352625
Methods R DB PC parallel groups
Participants Shoulder elbow or knee tendonitis or bursitis Age ge 18 yearsN = 330
Interventions Ketoprofen patch 20 x 1 dailyPlacebo patch x 1 dailyTreatment for 21 days
Outcomes Assessments days 3 7 21PI (0-10) at rest and during activitiesPGEFunctionRescue medication
Notes Sponsor Endo PharmaceuticalsEstimated completion February 2007 No results
NCT00426985
Methods R DB PC parallel groups
Participants Shoulder elbow or knee tendonitis or bursitis Age ge 18 years
Interventions Ketoprofen patch 20 x 1 dailyPlacebo patchTreatment for 21 days
Outcomes PI during activityPI at restPGERescue medication
94Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT00426985 (Continued)
Safety
Notes Study terminated May 2008 but ldquosufficient number of subjects accrued to conduct analysisrdquo No results
NCT00640705
Methods R DB PC parallel groups
Participants Ankle sprain or strain Grade I or II le 48 hours PI 510 to 910 Age 18 to 75 yearsN = 170
Interventions Diclofenac sodium patch (15 mg) x 1 dailyPlaceboTreatment for 7 days
Outcomes EfficacySafety
Notes Study terminated July 2008 (Sponsor decision - no further details) No results posted
NCT00640939
Methods R DB PC parallel groups
Participants Shoulder elbow or knee tendonitis or bursitis Age 18 to 75 yearsBaseline pain mild to moderate but states 510 to 910N = 308
Interventions Diclofenac sodium patch 1 (15 mg) x 1 dailyPlacebo patch x 1 dailyDuration of treatment not reported probably 14 days
Outcomes EfficacySafety
Notes Sponsor Cerimon PharmaceuticalsEstimated completion April 2008 No results
NCT00680472
Methods R DB PC parallel groups
Participants Acute unilateral shoulder pain requiring treatment for ge 2 weeks Age ge 18 yearsN = 368
95Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT00680472 (Continued)
Interventions Ketoprofen patch (HKT-500) x 1 dailyPlacebo x 1 dailyTreatment for 14 days
Outcomes PainSafety
Notes Sponsor Hisamitsu Pharmaceutical Co IncEstimated completion October 2008 No results
NCT00680784
Methods R DB PC parallel groups
Participants Acute benign ankle sprain Grade I-II le 48 hours Age ge 18 yearsN = 260
Interventions Ketoprofen patch (HKT-500) x 1 dailyPlacebo patch x 1 dailyTreatment for 14 days
Outcomes PainSafety
Notes Sponsor Hisamitsu Pharmaceutical Co IncEstimated completion November 2008 No results
NCT00765700
Methods R DB PC parallel groups
Participants Acute sprain or strain of upper and lower extremities mild to moderate injuries le 72 hours Age 18 to 70 yearsN = 364
Interventions Ketoprofen cream 10 1 g x 3 dailyPlacebo cream x 3 dailyTreatment for 7 days
Outcomes PainSafety
Notes Sponsor Imprimis Pharmaceuticals IncEstimated completion September 2009 No results
96Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT00869063
Methods R DB PC parallel groups
Participants Wrist sprain strain or contusion (mild to moderate) ldquorecentrdquo Age 17 to 75 yearsN = 214
Interventions Diclofenac sodium patch 1 x 1 dailyPlacebo patch x 1 dailyTreatment for 7 days
Outcomes Change in PI during activity at 3 and 7 daysSafety
Notes Sponsor Cerimon PharmaceuticalsEstimated completion September 2009 No results
NCT00869180
Methods R DB PC parallel groups
Participants Unilateral ankle sprain (mild or moderate) ldquorecentrdquo Age 17 to 75 yearsN = 219
Interventions Diclofenac sodium patch x 1 dailyPlacebo patch x 1 dailyTreatment for 7 days
Outcomes Change in PI during activity at 3 and 7 days
Notes Sponsor Cerimon PharmaceuticalsEstimated completion August 2009 No results
NCT00931866
Methods R DB PC parallel groups
Participants Unilateral soft tissue injury between mid-bicep to wrist or mid-thigh to ankle (mild to moderate) ldquorecentrdquo Age 18to 75 yearsN = 407
Interventions Diclofenac sodium patch x 1 dailyPlacebo patch x 1 dailyTreatment for 14 days
Outcomes Change in PI during activity at 7 and 14 daysPIPGERescue medicationSafety
97Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT00931866 (Continued)
Notes Sponsor Cerimon PharmaceuticalsEstimated completion December 2009 No results
NCT01874626
Methods R DB PC parallel groups
Participants Acute ankle sprain Grade I or II le 24 hours Age ge 16 years Baseline PI ge 510N = 305
Interventions Ibuprofen cream 200 mg in 27 g cream x 4 dailyPlacebo cream x 4 dailyTreatment for 7 days
Outcomes PI on movement daily to 7 daysPGE 7 daysSystemic and local adverse events
Notes Estimated completion January 2014 No results
NCT01957215
Methods R DB PC parallel groups
Participants Acute lateral ankle sprain Grade I-II le 24 hours PI ge 510 Age 18 to 65 yearsN = 270
Interventions Indomethacin patch x 2 dailyPlacebo patch x 2 dailyDuration of treatment not reported probably 7 days
Outcomes PI on movement dailyPR dailyPGERescue medication
Notes Estimated completion September 2014 No results
NCT02324270
Methods R DB PC and AC parallel groups
Participants Uncomplicated acute ankle sprain Grade I or II lt 48 hours PI gt 50100 Age 18 to 65 yearsN = 658
98Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT02324270 (Continued)
Interventions Diclofenac epolamine (Flector) 13 patchGeneric diclofenac epolamine 13 patchPlacebo patchDuration of treatment not reported
Outcomes Bioequivalence studyChange from baseline in PI (VAS) at 3 daysApplication site reactions
Notes Estimated completion December 2014 No results
Sarzi-Puttini 2014
Methods R DB AC parallel group non-inferiority studyDuration 7 days
Participants Acute musculoskeletal injury (mainly muscular joint tendon)N = 697M 271 F 426Mean age 52 years
Interventions Ketoprofen (SKP-021) patch ketoprofen 30 mgDiclofenac (Voltadola) patch diclofenac sodium 140 mgPatches applied twice daily for 7 days
Outcomes ge 50 reduction in PI from baseline to end of study (100 mm VAS)Patient overall ratingClinical symptoms (4-point scale)Time to responseAdverse events skin reactionsSerious adverse events
Notes ldquoThe analysis of the data of this trial showed that the two formulations were equally effective and well tolerated inthe treatment of acute musculoskeletal injuriesrdquo
AC active-controlled DB double-blind F female M male N number of participants in study PC placebo-controlled PGE PatientGlobal Evaluation of treatment PI pain intensity R randomised VAS visual analogue scale
99Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Characteristics of ongoing studies [ordered by study ID]
NCT01945034
Trial name or title Placebo-controlled double-blind evaluation of the efficacy and safety of ibuprofen 5 topical gel for thetreatment of ankle sprain
Methods R DB PC parallel groups
Participants Ankle sprain Grade I or II age ge 12 yearsEstimated enrolment 280
Interventions Ibuprofen gel 5 x 2 dailyIbuprofen gel 5 x 3 dailyPlacebo gelTreatment for 7 days with additional 3 days as needed
Outcomes PI on weight bearing and rest at 7 daysPGE at 10 daysRescue medicationSafety
Starting date November 2013
Contact information Pfizer
Notes Estimated completion February 2015Includes participants aged ge 12 years
NCT02100670
Trial name or title A clinical study to assess the efficacy and onset of pain relief of topical MFC51123 diclofenac-menthol gelversus controls in ankle sprain
Methods R DB PC and AC parallel groups
Participants Acute lateral ankle sprain Grade I or II le 24 hours PI ge 510 Age 16 to 65 yearsEstimated enrolment 400
Interventions Diclofenac sodium 1 + methanol 3 gelDiclofenac sodium 1 + methanol 009 gelMethanol 3 gelPlacebo + 009 methanol gelTreatment for 10 days with 4 g gel x 4 daily
Outcomes PRPID on movement (0 to 10 days)PGETime to complete recoveryAdverse events
100Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT02100670 (Continued)
Starting date November 2013
Contact information GSKClinicalSupportHDgskcom
Notes Estimated completion November 2014
NCT02290821
Trial name or title A randomised double-blind placebo-controlled parallel group study to evaluate the efficacy and safety ofdiclofenac sodium topical gel (DSG) 1 applied four times daily in subjects with acute blunt soft tissueinjuriescontusions of the limbs
Methods R DB PC parallel groups
Participants Acute blunt soft tissue injuries or contusions of the limbs le 6 hours (ldquofreshrdquo) Age ge 16 yearsEstimated enrolment 200
Interventions Diclofenac sodium gel 1 x 4 dailyPlacebo gelDuration of treatment not reported
Outcomes PainSafety
Starting date December 2014
Contact information Novartis
Notes Estimated completion August 2015
AC active control DB double-blind PC placebo-controlled PGE Patient Global Evaluation of treatment PI pain intensity PIDpain intensity difference PR pain relief R randomised
101Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
D A T A A N D A N A L Y S E S
Comparison 1 Individual NSAID versus placebo
Outcome or subgroup titleNo ofstudies
No ofparticipants Statistical method Effect size
1 Clinical success 29 Risk Ratio (M-H Fixed 95 CI) Subtotals only11 Diclofenac 10 2050 Risk Ratio (M-H Fixed 95 CI) 160 [149 172]12 Ibuprofen 5 436 Risk Ratio (M-H Fixed 95 CI) 164 [133 201]13 Ketoprofen 7 683 Risk Ratio (M-H Fixed 95 CI) 156 [137 177]14 Piroxicam 3 504 Risk Ratio (M-H Fixed 95 CI) 148 [127 173]15 Indomethacin 3 341 Risk Ratio (M-H Fixed 95 CI) 126 [103 155]16 Benzydamine 3 193 Risk Ratio (M-H Fixed 95 CI) 115 [096 138]
2 Local adverse events 33 Risk Ratio (M-H Fixed 95 CI) Subtotals only21 Diclofenac 15 3271 Risk Ratio (M-H Fixed 95 CI) 078 [056 110]22 Ibuprofen 3 321 Risk Ratio (M-H Fixed 95 CI) 230 [098 543]23 Ketoprofen 8 852 Risk Ratio (M-H Fixed 95 CI) 119 [083 170]24 Piroxicam 3 522 Risk Ratio (M-H Fixed 95 CI) 042 [017 108]25 Felbinac 3 397 Risk Ratio (M-H Fixed 95 CI) 191 [049 750]26 Indomethacin 3 354 Risk Ratio (M-H Fixed 95 CI) 265 [091 773]
Comparison 2 Diclofenac versus placebo (effect of formulation)
Outcome or subgroup titleNo ofstudies
No ofparticipants Statistical method Effect size
1 Clinical success 10 2050 Risk Ratio (M-H Fixed 95 CI) 160 [149 172]11 Plaster - Flector 4 1030 Risk Ratio (M-H Fixed 95 CI) 153 [136 171]12 Plaster - other 3 474 Risk Ratio (M-H Fixed 95 CI) 155 [137 175]13 Gel - Emulgel 2 314 Risk Ratio (M-H Fixed 95 CI) 384 [268 550]14 Gel - other 1 232 Risk Ratio (M-H Fixed 95 CI) 115 [105 127]
Comparison 3 Ibuprofen versus placebo (effect of formulation)
Outcome or subgroup titleNo ofstudies
No ofparticipants Statistical method Effect size
1 Clinical success 5 436 Risk Ratio (M-H Fixed 95 CI) 164 [133 201]11 Cream 3 195 Risk Ratio (M-H Fixed 95 CI) 128 [103 159]12 Gel 2 241 Risk Ratio (M-H Fixed 95 CI) 266 [169 421]
102Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Comparison 4 Ketoprofen versus placebo
Outcome or subgroup titleNo ofstudies
No ofparticipants Statistical method Effect size
1 Clinical success 7 683 Risk Ratio (M-H Fixed 95 CI) 156 [137 177]11 Plaster 2 335 Risk Ratio (M-H Fixed 95 CI) 121 [104 140]12 Gel 5 348 Risk Ratio (M-H Fixed 95 CI) 219 [174 275]
Comparison 5 All topical NSAIDs versus placebo
Outcome or subgroup titleNo ofstudies
No ofparticipants Statistical method Effect size
1 Local adverse events 42 6740 Risk Ratio (M-H Fixed 95 CI) 098 [080 121]2 Systemic adverse events 36 5576 Risk Ratio (M-H Fixed 95 CI) 096 [073 126]3 Adverse event withdrawals 42 6405 Risk Ratio (M-H Fixed 95 CI) 101 [064 159]
Comparison 6 Topical NSAID versus active comparator
Outcome or subgroup titleNo ofstudies
No ofparticipants Statistical method Effect size
1 Clinical success - topicalpiroxicam vs topicalindomethacin
3 641 Risk Ratio (M-H Fixed 95 CI) 124 [107 144]
2 Local adverse events -topical piroxicam vs topicalindomethacin
3 671 Risk Ratio (M-H Fixed 95 CI) 021 [009 047]
103Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Analysis 11 Comparison 1 Individual NSAID versus placebo Outcome 1 Clinical success
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 1 Individual NSAID versus placebo
Outcome 1 Clinical success
Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
1 Diclofenac
Coudreuse 2010 99117 82116 175 120 [ 104 138 ]
Joussellin 2003 3668 2466 52 146 [ 099 215 ]
Klainguti 2010 5662 4659 100 116 [ 099 136 ]
Li 2013 173192 114192 242 152 [ 134 172 ]
Predel 2004 5560 560 11 1100 [ 474 2555 ]
Predel 2012 (1) 5980 941 25 336 [ 186 607 ]
Predel 2012 (2) 5780 841 22 365 [ 193 690 ]
Predel 2013a 111118 93114 201 115 [ 105 127 ]
Predel 2013b 3636 736 16 487 [ 257 923 ]
Rowbotham 2003 75191 48181 104 148 [ 110 200 ]
Saillant 1998 4370 2570 53 172 [ 119 248 ]
Subtotal (95 CI) 1074 976 1000 160 [ 149 172 ]Total events 800 (NSAID) 461 (Placebo)
Heterogeneity Chi2 = 12176 df = 10 (Plt000001) I2 =92
Test for overall effect Z = 1278 (P lt 000001)
2 Ibuprofen
Billigmann 1996 2580 1080 136 250 [ 129 486 ]
Campbell 1994 2126 1925 264 106 [ 080 142 ]
Dreiser 1988 2632 1232 164 217 [ 134 349 ]
Machen 2002 2540 941 121 285 [ 152 532 ]
Ramesh 1983 2340 2340 314 100 [ 069 146 ]
Subtotal (95 CI) 218 218 1000 164 [ 133 201 ]Total events 120 (NSAID) 73 (Placebo)
Heterogeneity Chi2 = 2102 df = 4 (P = 000031) I2 =81
Test for overall effect Z = 465 (P lt 000001)
3 Ketoprofen
Airaksinen 1993 2429 1427 91 160 [ 107 238 ]
Dreiser 1989 1830 530 32 360 [ 154 844 ]
02 05 1 2 5
Favours placebo Favours NSAID
(Continued )
104Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
( Continued)Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Julien 1989 1830 630 38 300 [ 138 650 ]
Kockelbergh 1985 3038 2236 142 129 [ 095 176 ]
Mazi res 2005a 7281 6082 376 121 [ 104 141 ]
Mazi res 2005b 5087 4185 262 119 [ 090 158 ]
Noret 1987 3951 947 59 399 [ 218 733 ]
Subtotal (95 CI) 346 337 1000 156 [ 137 177 ]Total events 251 (NSAID) 157 (Placebo)
Heterogeneity Chi2 = 3104 df = 6 (P = 000002) I2 =81
Test for overall effect Z = 683 (P lt 000001)
4 Piroxicam
Aoki 1984 5672 3367 286 158 [ 120 207 ]
Fujimaki 1985 4483 4082 337 109 [ 080 147 ]
Russell 1991 79100 45100 377 176 [ 138 223 ]
Subtotal (95 CI) 255 249 1000 148 [ 127 173 ]Total events 179 (NSAID) 118 (Placebo)
Heterogeneity Chi2 = 624 df = 2 (P = 004) I2 =68
Test for overall effect Z = 499 (P lt 000001)
5 Indomethacin
Aoki 1984 4164 3367 413 130 [ 096 176 ]
Fujimaki 1985 4482 4082 513 110 [ 082 148 ]
kermark 1990 1222 624 74 218 [ 099 481 ]
Subtotal (95 CI) 168 173 1000 126 [ 103 155 ]Total events 97 (NSAID) 79 (Placebo)
Heterogeneity Chi2 = 269 df = 2 (P = 026) I2 =26
Test for overall effect Z = 221 (P = 0027)
6 Benzydamine
Chatterjee 1977 2125 1225 185 175 [ 112 272 ]
Haig 1986 1821 1322 196 145 [ 098 214 ]
Linde 1985 3550 4050 618 088 [ 070 110 ]
Subtotal (95 CI) 96 97 1000 115 [ 096 138 ]Total events 74 (NSAID) 65 (Placebo)
Heterogeneity Chi2 = 1033 df = 2 (P = 001) I2 =81
Test for overall effect Z = 152 (P = 013)
02 05 1 2 5
Favours placebo Favours NSAID
(1) Three times daily application
(2) Twice daily application
105Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Analysis 12 Comparison 1 Individual NSAID versus placebo Outcome 2 Local adverse events
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 1 Individual NSAID versus placebo
Outcome 2 Local adverse events
Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
1 Diclofenac
Coudreuse 2010 1117 7116 102 014 [ 002 113 ]
Jenoure 1997 144 141 15 093 [ 006 1442 ]
Joussellin 2003 168 366 44 032 [ 003 303 ]
Klainguti 2010 0126 159 30 016 [ 001 381 ]
Kuehl 2011 16207 12211 173 136 [ 066 280 ]
Li 2013 4192 3192 44 133 [ 030 588 ]
NCT01255423 1104 3102 44 033 [ 003 309 ]
NCT01272934 1102 0103 07 303 [ 012 7350 ]
NCT01272947 0104 0100 Not estimable
Predel 2012 (1) 1160 182 19 051 [ 003 809 ]
Predel 2013a 1120 4116 59 024 [ 003 213 ]
Predel 2013b 036 036 Not estimable
Rowbotham 2003 27191 31181 463 083 [ 051 133 ]
Saillant 1998 070 070 Not estimable
Spacca 2005 079 076 Not estimable
Subtotal (95 CI) 1720 1551 1000 078 [ 056 110 ]Total events 54 (NSAID) 66 (Placebo)
Heterogeneity Chi2 = 944 df = 10 (P = 049) I2 =00
Test for overall effect Z = 142 (P = 016)
2 Ibuprofen
Billigmann 1996 1180 480 573 275 [ 091 827 ]
Machen 2002 440 241 283 205 [ 040 1057 ]
Ramesh 1983 140 140 143 100 [ 006 1544 ]
Subtotal (95 CI) 160 161 1000 230 [ 098 543 ]Total events 16 (NSAID) 7 (Placebo)
Heterogeneity Chi2 = 048 df = 2 (P = 079) I2 =00
Test for overall effect Z = 190 (P = 0057)
3 Ketoprofen
002 01 1 10 50
Favours NSAID Favours placebo
(Continued )
106Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
( Continued)
Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Airaksinen 1993 529 427 99 116 [ 035 389 ]
Dreiser 1989 030 230 60 020 [ 001 400 ]
Julien 1989 130 030 12 300 [ 013 7083 ]
Kockelbergh 1985 138 136 24 095 [ 006 1459 ]
Mazi res 2005a 1281 682 142 202 [ 080 513 ]
Mazi res 2005b 2987 2785 651 105 [ 068 161 ]
Noret 1987 151 047 12 277 [ 012 6636 ]
Parrini 1992 083 086 Not estimable
Subtotal (95 CI) 429 423 1000 119 [ 083 170 ]Total events 49 (NSAID) 40 (Placebo)
Heterogeneity Chi2 = 357 df = 6 (P = 073) I2 =00
Test for overall effect Z = 095 (P = 034)
4 Piroxicam
Aoki 1984 179 274 147 047 [ 004 506 ]
Fujimaki 1985 183 282 143 049 [ 005 534 ]
Russell 1991 4102 10102 710 040 [ 013 123 ]
Subtotal (95 CI) 264 258 1000 042 [ 017 108 ]Total events 6 (NSAID) 14 (Placebo)
Heterogeneity Chi2 = 003 df = 2 (P = 098) I2 =00
Test for overall effect Z = 180 (P = 0072)
5 Felbinac
McLatchie 1989 3118 2113 666 144 [ 024 844 ]
Morris 1991 041 043 Not estimable
Sanguinetti 1989 342 140 334 286 [ 031 2634 ]
Subtotal (95 CI) 201 196 1000 191 [ 049 750 ]Total events 6 (NSAID) 3 (Placebo)
Heterogeneity Chi2 = 023 df = 1 (P = 063) I2 =00
Test for overall effect Z = 093 (P = 035)
6 Indomethacin
Aoki 1984 270 274 440 106 [ 015 730 ]
Fujimaki 1985 582 282 452 250 [ 050 1252 ]
kermark 1990 422 024 108 978 [ 056 17191 ]
Subtotal (95 CI) 174 180 1000 265 [ 091 773 ]Total events 11 (NSAID) 4 (Placebo)
Heterogeneity Chi2 = 167 df = 2 (P = 043) I2 =00
Test for overall effect Z = 179 (P = 0073)
002 01 1 10 50
Favours NSAID Favours placebo
(1) DHEP + heparin and DHEP combined
107Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Analysis 21 Comparison 2 Diclofenac versus placebo (effect of formulation) Outcome 1 Clinical success
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 2 Diclofenac versus placebo (effect of formulation)
Outcome 1 Clinical success
Study or subgroup Experimental Control Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
1 Plaster - Flector
Joussellin 2003 3668 2466 52 146 [ 099 215 ]
Li 2013 173192 114192 242 152 [ 134 172 ]
Rowbotham 2003 75191 48181 104 148 [ 110 200 ]
Saillant 1998 4370 2570 53 172 [ 119 248 ]
Subtotal (95 CI) 521 509 451 153 [ 136 171 ]Total events 327 (Experimental) 211 (Control)
Heterogeneity Chi2 = 051 df = 3 (P = 092) I2 =00
Test for overall effect Z = 720 (P lt 000001)
2 Plaster - other
Coudreuse 2010 99117 82116 175 120 [ 104 138 ]
Klainguti 2010 5662 4659 100 116 [ 099 136 ]
Predel 2004 5560 560 11 1100 [ 474 2555 ]
Subtotal (95 CI) 239 235 285 155 [ 137 175 ]Total events 210 (Experimental) 133 (Control)
Heterogeneity Chi2 = 4658 df = 2 (Plt000001) I2 =96
Test for overall effect Z = 699 (P lt 000001)
3 Gel - Emulgel
Predel 2012 (1) 116160 1782 48 350 [ 227 540 ]
Predel 2013b 3636 736 16 487 [ 257 923 ]
Subtotal (95 CI) 196 118 64 384 [ 268 550 ]Total events 152 (Experimental) 24 (Control)
Heterogeneity Chi2 = 070 df = 1 (P = 040) I2 =00
Test for overall effect Z = 734 (P lt 000001)
4 Gel - other
Predel 2013a 111118 93114 201 115 [ 105 127 ]
Subtotal (95 CI) 118 114 201 115 [ 105 127 ]Total events 111 (Experimental) 93 (Control)
02 05 1 2 5
Favours placebo Favours diclofenac
(Continued )
108Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
( Continued)Study or subgroup Experimental Control Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Heterogeneity not applicable
Test for overall effect Z = 284 (P = 00045)
Total (95 CI) 1074 976 1000 160 [ 149 172 ]Total events 800 (Experimental) 461 (Control)
Heterogeneity Chi2 = 12176 df = 9 (Plt000001) I2 =93
Test for overall effect Z = 1278 (P lt 000001)
Test for subgroup differences Chi2 = 5003 df = 3 (P = 000) I2 =94
02 05 1 2 5
Favours placebo Favours diclofenac
(1) twice daily and three times daily applications combined
Analysis 31 Comparison 3 Ibuprofen versus placebo (effect of formulation) Outcome 1 Clinical success
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 3 Ibuprofen versus placebo (effect of formulation)
Outcome 1 Clinical success
Study or subgroup Ibuprofen Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
1 Cream
Campbell 1994 2126 1925 264 106 [ 080 142 ]
Dreiser 1988 2632 1232 164 217 [ 134 349 ]
Ramesh 1983 2340 2340 314 100 [ 069 146 ]
Subtotal (95 CI) 98 97 742 128 [ 103 159 ]Total events 70 (Ibuprofen) 54 (Placebo)
Heterogeneity Chi2 = 791 df = 2 (P = 002) I2 =75
Test for overall effect Z = 222 (P = 0026)
2 Gel
Billigmann 1996 2580 1080 136 250 [ 129 486 ]
Machen 2002 2540 941 121 285 [ 152 532 ]
Subtotal (95 CI) 120 121 258 266 [ 169 421 ]Total events 50 (Ibuprofen) 19 (Placebo)
001 01 1 10 100
Favours placebo Favours ibuprofen
(Continued )
109Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
( Continued)Study or subgroup Ibuprofen Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Heterogeneity Chi2 = 008 df = 1 (P = 078) I2 =00
Test for overall effect Z = 420 (P = 0000026)
Total (95 CI) 218 218 1000 164 [ 133 201 ]Total events 120 (Ibuprofen) 73 (Placebo)
Heterogeneity Chi2 = 2102 df = 4 (P = 000031) I2 =81
Test for overall effect Z = 465 (P lt 000001)
Test for subgroup differences Chi2 = 805 df = 1 (P = 000) I2 =88
001 01 1 10 100
Favours placebo Favours ibuprofen
Analysis 41 Comparison 4 Ketoprofen versus placebo Outcome 1 Clinical success
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 4 Ketoprofen versus placebo
Outcome 1 Clinical success
Study or subgroup Diclofenac Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
1 Plaster
Mazi res 2005a 7281 6082 376 121 [ 104 141 ]
Mazi res 2005b 5087 4185 262 119 [ 090 158 ]
Subtotal (95 CI) 168 167 638 121 [ 104 140 ]Total events 122 (Diclofenac) 101 (Placebo)
Heterogeneity Chi2 = 002 df = 1 (P = 090) I2 =00
Test for overall effect Z = 249 (P = 0013)
2 Gel
Airaksinen 1993 2429 1427 91 160 [ 107 238 ]
Dreiser 1989 1830 530 32 360 [ 154 844 ]
Julien 1989 1830 630 38 300 [ 138 650 ]
Kockelbergh 1985 3038 2236 142 129 [ 095 176 ]
Noret 1987 3951 947 59 399 [ 218 733 ]
001 01 1 10 100
Favours placebo Favours diclofenac
(Continued )
110Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
( Continued)Study or subgroup Diclofenac Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Subtotal (95 CI) 178 170 362 219 [ 174 275 ]Total events 129 (Diclofenac) 56 (Placebo)
Heterogeneity Chi2 = 1937 df = 4 (P = 000067) I2 =79
Test for overall effect Z = 674 (P lt 000001)
Total (95 CI) 346 337 1000 156 [ 137 177 ]Total events 251 (Diclofenac) 157 (Placebo)
Heterogeneity Chi2 = 3104 df = 6 (P = 000002) I2 =81
Test for overall effect Z = 683 (P lt 000001)
Test for subgroup differences Chi2 = 1859 df = 1 (P = 000) I2 =95
001 01 1 10 100
Favours placebo Favours diclofenac
Analysis 51 Comparison 5 All topical NSAIDs versus placebo Outcome 1 Local adverse events
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 5 All topical NSAIDs versus placebo
Outcome 1 Local adverse events
Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Airaksinen 1993 529 427 26 116 [ 035 389 ]
Aoki 1984 (1) 270 137 08 106 [ 010 1128 ]
Aoki 1984 (2) 179 137 09 047 [ 003 728 ]
Auclair 1989 5123 6116 39 079 [ 025 251 ]
Billigmann 1996 1180 480 26 275 [ 091 827 ]
Costantino 2011 (3) 3287 6142 51 025 [ 006 097 ]
Coudreuse 2010 1117 7116 45 014 [ 002 113 ]
Diebshlag 1990 (4) 113 06 04 150 [ 007 3229 ]
Diebshlag 1990 (5) 112 06 04 162 [ 008 3466 ]
Dreiser 1989 030 230 16 020 [ 001 400 ]
002 01 1 10 50
Favours NSAID Favours placebo
(Continued )
111Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
( Continued)Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Dreiser 1990 030 330 22 014 [ 001 265 ]
Dreiser 1994 265 066 03 508 [ 025 10373 ]
Fujimaki 1985 (6) 183 141 09 049 [ 003 770 ]
Fujimaki 1985 (7) 582 141 09 250 [ 030 2070 ]
Haig 1986 021 022 Not estimable
Jenoure 1997 144 141 07 093 [ 006 1442 ]
Joussellin 2003 168 366 19 032 [ 003 303 ]
Julien 1989 130 030 03 300 [ 013 7083 ]
Klainguti 2010 (8) 0126 159 13 016 [ 001 381 ]
Kockelbergh 1985 138 136 07 095 [ 006 1459 ]
Kuehl 2011 16207 12211 76 136 [ 066 280 ]
Li 2013 4192 3192 19 133 [ 030 588 ]
Linde 1985 050 050 Not estimable
Machen 2002 440 241 13 205 [ 040 1057 ]
Mazi res 2005a 1281 682 38 202 [ 080 513 ]
Mazi res 2005b 2987 2785 174 105 [ 068 161 ]
McLatchie 1989 3118 2113 13 144 [ 024 844 ]
Morris 1991 041 043 Not estimable
NCT01255423 1104 3102 19 033 [ 003 309 ]
NCT01272934 1102 0103 03 303 [ 012 7350 ]
NCT01272947 0104 0100 Not estimable
Noret 1987 151 047 03 277 [ 012 6636 ]
Parrini 1992 083 086 Not estimable
Predel 2012 (9) 1160 182 08 051 [ 003 809 ]
Predel 2013a 1120 4116 26 024 [ 003 213 ]
Predel 2013b 036 036 Not estimable
Ramesh 1983 140 140 06 100 [ 006 1544 ]
Rowbotham 2003 27191 31181 203 083 [ 051 133 ]
Russell 1991 4102 10102 64 040 [ 013 123 ]
Saillant 1998 070 070 Not estimable
Sanguinetti 1989 342 140 07 286 [ 031 2634 ]
002 01 1 10 50
Favours NSAID Favours placebo
(Continued )
112Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
( Continued)Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Sinniger 1981 010 010 Not estimable
Spacca 2005 079 076 Not estimable
Thorling 1990 160 060 03 300 [ 012 7220 ]
kermark 1990 422 024 03 978 [ 056 17191 ]
Total (95 CI) 3619 3121 1000 098 [ 080 121 ]Total events 155 (NSAID) 145 (Placebo)
Heterogeneity Chi2 = 3354 df = 35 (P = 054) I2 =00
Test for overall effect Z = 017 (P = 087)
Test for subgroup differences Not applicable
002 01 1 10 50
Favours NSAID Favours placebo
(1) indomethacin
(2) piroxicam
(3) DHEP + heparin and DHEP combined
(4) ketorolac
(5) etofenamate
(6) piroxicam
(7) indomethacin
(8) DHEP + heparin and DHEP combined
(9) Twice and three times daily application combined
113Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Analysis 52 Comparison 5 All topical NSAIDs versus placebo Outcome 2 Systemic adverse events
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 5 All topical NSAIDs versus placebo
Outcome 2 Systemic adverse events
Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Airaksinen 1993 129 027 05 280 [ 012 6593 ]
Aoki 1984 (1) 079 037 Not estimable
Aoki 1984 (2) 070 037 Not estimable
Campbell 1994 126 025 05 289 [ 012 6775 ]
Chatterjee 1977 025 025 Not estimable
Coudreuse 2010 1117 0116 05 297 [ 012 7227 ]
Diebshlag 1990 (3) 013 06 Not estimable
Diebshlag 1990 (4) 012 06 Not estimable
Dreiser 1988 032 032 Not estimable
Dreiser 1989 030 030 Not estimable
Dreiser 1990 030 030 Not estimable
Dreiser 1994 065 066 Not estimable
Fujimaki 1985 (5) 083 041 Not estimable
Fujimaki 1985 (6) 182 041 07 152 [ 006 3647 ]
Haig 1986 021 022 Not estimable
Hoffmann 2012 0115 0118 Not estimable
Joussellin 2003 168 066 05 291 [ 012 7025 ]
Klainguti 2010 (7) 2126 059 07 236 [ 012 4844 ]
Kuehl 2011 15207 23211 236 066 [ 036 124 ]
Li 2013 10192 4192 41 250 [ 080 783 ]
Linde 1985 050 050 Not estimable
Machen 2002 040 041 Not estimable
Mazi res 2005a 1187 785 73 154 [ 062 377 ]
Mazi res 2005b 1381 1482 144 094 [ 047 187 ]
McLatchie 1989 0118 0113 Not estimable
Morris 1991 041 043 Not estimable
001 01 1 10 100
Favours topical NSAID Favours placebo
(Continued )
114Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
( Continued)Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
NCT01272947 2104 2100 21 096 [ 014 670 ]
Parrini 1992 083 086 Not estimable
Predel 2004 060 060 Not estimable
Predel 2012 (8) 3160 382 41 051 [ 011 248 ]
Predel 2013a 6120 8116 84 073 [ 026 203 ]
Predel 2013b 036 136 16 033 [ 001 792 ]
Rowbotham 2003 21191 22181 234 090 [ 052 159 ]
Russell 1991 4102 7102 73 057 [ 017 189 ]
Saillant 1998 070 070 Not estimable
Sanguinetti 1989 042 040 Not estimable
Sinniger 1981 010 010 Not estimable
Spacca 2005 079 076 Not estimable
Thorling 1990 060 060 Not estimable
Total (95 CI) 2956 2620 1000 096 [ 073 126 ]Total events 92 (NSAID) 91 (Placebo)
Heterogeneity Chi2 = 946 df = 15 (P = 085) I2 =00
Test for overall effect Z = 030 (P = 076)
Test for subgroup differences Not applicable
001 01 1 10 100
Favours topical NSAID Favours placebo
(1) piroxicam
(2) indomethacin
(3) ketorolac
(4) etofenamate
(5) piroxicam
(6) indomethacin
(7) DHEP + heparin and DHEP combined
(8) Twice and three times daily application combined
115Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Analysis 53 Comparison 5 All topical NSAIDs versus placebo Outcome 3 Adverse event withdrawals
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 5 All topical NSAIDs versus placebo
Outcome 3 Adverse event withdrawals
Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Airaksinen 1993 029 027 Not estimable
Aoki 1984 072 034 Not estimable
Aoki 1984 064 033 Not estimable
Auclair 1989 1123 0116 14 283 [ 012 6879 ]
Billigmann 1996 280 080 14 500 [ 024 10253 ]
Campbell 1994 026 025 Not estimable
Chatterjee 1977 025 025 Not estimable
Costantino 2011 0188 0142 Not estimable
Coudreuse 2010 0117 0116 Not estimable
Diebshlag 1990 013 06 Not estimable
Diebshlag 1990 012 06 Not estimable
Dreiser 1988 032 032 Not estimable
Dreiser 1989 030 230 69 020 [ 001 400 ]
Dreiser 1990 030 130 42 033 [ 001 787 ]
Dreiser 1994 065 066 Not estimable
Fujimaki 1985 083 041 Not estimable
Fujimaki 1985 482 041 18 455 [ 025 8261 ]
Julien 1989 030 030 Not estimable
Kockelbergh 1985 038 036 Not estimable
Kuehl 2011 4207 9211 248 045 [ 014 145 ]
Li 2013 2192 0192 14 500 [ 024 10347 ]
Linde 1985 050 050 Not estimable
Machen 2002 040 041 Not estimable
Mazi res 2005a 987 685 169 147 [ 055 394 ]
Mazi res 2005b 381 082 14 709 [ 037 13503 ]
McLatchie 1989 0118 0113 Not estimable
001 01 1 10 100
Favours topical NSAID Favours placebo
(Continued )
116Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
( Continued)Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Morris 1991 041 043 Not estimable
NCT01255423 0104 0102 Not estimable
NCT01272934 0102 1103 41 034 [ 001 817 ]
NCT01272947 0104 0100 Not estimable
Noret 1987 148 045 14 282 [ 012 6740 ]
Parrini 1992 060 060 Not estimable
Predel 2004 160 060 14 300 [ 012 7220 ]
Predel 2012 2160 182 37 103 [ 009 1114 ]
Predel 2013a 1120 1116 28 097 [ 006 1527 ]
Predel 2013b 036 036 Not estimable
Ramesh 1983 140 140 28 100 [ 006 1544 ]
Rowbotham 2003 0191 0181 Not estimable
Russell 1991 1102 8102 222 013 [ 002 098 ]
Saillant 1998 070 070 Not estimable
Sanguinetti 1989 042 040 Not estimable
Sinniger 1981 010 010 Not estimable
Spacca 2005 079 076 Not estimable
Thorling 1990 060 060 Not estimable
kermark 1990 122 024 13 326 [ 014 7610 ]
Total (95 CI) 3365 3040 1000 101 [ 064 159 ]Total events 33 (NSAID) 30 (Placebo)
Heterogeneity Chi2 = 1502 df = 16 (P = 052) I2 =00
Test for overall effect Z = 004 (P = 097)
Test for subgroup differences Not applicable
001 01 1 10 100
Favours topical NSAID Favours placebo
117Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Analysis 61 Comparison 6 Topical NSAID versus active comparator Outcome 1 Clinical success - topical
piroxicam vs topical indomethacin
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 6 Topical NSAID versus active comparator
Outcome 1 Clinical success - topical piroxicam vs topical indomethacin
Study or subgroup Piroxicam Indomethacin Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Aoki 1984 5672 4164 301 121 [ 097 151 ]
Fujimaki 1985 4483 4482 307 099 [ 074 131 ]
Sugioka 1984 85175 55165 392 146 [ 112 190 ]
Total (95 CI) 330 311 1000 124 [ 107 144 ]Total events 185 (Piroxicam) 140 (Indomethacin)
Heterogeneity Chi2 = 390 df = 2 (P = 014) I2 =49
Test for overall effect Z = 278 (P = 00055)
Test for subgroup differences Not applicable
01 02 05 1 2 5 10
Favours indomethacin Favours piroxicam
118Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Analysis 62 Comparison 6 Topical NSAID versus active comparator Outcome 2 Local adverse events -
topical piroxicam vs topical indomethacin
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 6 Topical NSAID versus active comparator
Outcome 2 Local adverse events - topical piroxicam vs topical indomethacin
Study or subgroup Piroxicam Indomethacin Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Aoki 1984 179 270 64 044 [ 004 478 ]
Fujimaki 1985 183 582 152 020 [ 002 165 ]
Sugioka 1984 5178 26179 784 019 [ 008 049 ]
Total (95 CI) 340 331 1000 021 [ 009 047 ]Total events 7 (Piroxicam) 33 (Indomethacin)
Heterogeneity Chi2 = 041 df = 2 (P = 081) I2 =00
Test for overall effect Z = 382 (P = 000013)
Test for subgroup differences Not applicable
002 01 1 10 50
Favours piroxicam Favours indomethacin
A P P E N D I C E S
Appendix 1 CENTRAL search strategy (via CRSO) for 2015 update
1 MESH DESCRIPTOR Anti-Inflammatory Agents Non-Steroidal EXPLODE ALL TREES (13419)2 (bufexamac OR bufexine OR calmaderm OR ekzemase OR dicoflenac OR solaraze OR pennsaid OR voltarol OR emugel OR
voltarene OR voltarol OR optha OR voltaren OR etofenamate OR afrolate OR algesalona OR bayro OR deiron OR etofen ORflexium OR flogoprofen OR rheuma-gel OR rheumon OR traumalix OR traumon OR zenavan OR felbinac OR dolinac OR flexfreeOR napageln OR target OR traxam OR fentiazac OR domureuma OR fentiazaco OR norvedan OR riscalon OR fepradinol ORdalgen OR flexidol OR cocresol OR rangozona OR reuflodol OR pinazone OR zepelin OR flufenamic OR dignodolin OR rheumaOR lindofluid OR sastridex OR lunoxaprofen OR priaxim OR flubiprofen OR fenomel OR ocufen OR ocuflur OR ldquoTrans ActLATrdquo OR tulip OR ibuprofen OR cuprofen OR ldquodeep relief rdquo OR fenbid OR ibu-cream OR ibugel OR ibuleve OR ibumousse ORibuspray OR ldquonurofen gelrdquo OR proflex OR motrin OR advil OR radian OR ralgex OR ibutop OR indomethacin OR indocin ORindospray OR isonixin OR nixyn OR ketoprofen OR tiloket OR oruvail OR powergel OR solpaflex OR ketorolac OR acular ORtrometamol OR meclofenamic OR naproxen OR naprosyn OR niflumic OR actol OR flunir OR niflactol topico OR niflugel ORnifluril OR oxyphenbutazone OR californit OR diflamil OR otone OR tanderil OR piketoprofen OR calmatel OR triparsean ORpiroxicam OR feldene OR pranoprofen OR oftalar OR pranox OR suxibuzone OR danilon OR flamilon OR ufenamate OR fenazolOR flector OR benzydamine) TIABKY (25220)
3 1 OR 2 (32484)4 MESH DESCRIPTOR Administration Topical EXPLODE ALL TREES (2169)
119Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
5 (topical OR cutaneous OR dermal OR transcutaneous OR transdermal OR percutaneous OR skin OR massage ORembrocation OR gel OR ointment OR aerosol OR cream OR cregraveme OR lotion OR mouse OR foam OR liniment OR spray OR rubOR balm OR salve OR emulsion OR oil OR patch OR plaster)TIABKY (67940)
6 4 OR 5 (70486)7 MESH DESCRIPTOR Athletic Injuries EXPLODE ALL TREES (411)8 (strain OR sprain OR contusion OR distortion OR compression OR ldquosports injurrdquo OR ldquosoft tissue injurrdquo OR tendnitis OR
ldquomuscle painrdquo OR periarthritis OR epicondylitis OR tenosynovitis)TIABKY (9158)9 7 OR 8 (9448)
10 MESH DESCRIPTOR pain EXPLODE ALL TREES (29943)11 (pain OR analgesi)TIABKY (74815)12 10 OR 11 (80041)13 3 AND 6 AND 9 AND 12 (110)
Appendix 2 MEDLINE search strategy via Ovid (for 2015 update)
1 exp Anti-inflammatory Agents non-steroidal (162888)2 (bufexamac OR bufexine OR calmaderm OR ekzemase OR diclofenac OR solaraze OR pennsaid OR voltarol OR emugel OR
voltarene OR voltarol OR optha OR voltaren OR etofenamate OR afrolate OR algesalona OR bayro OR deiron OR etofen ORflexium OR flogoprofen OR rheuma-gel OR rheumon OR traumalix OR traumon OR zenavan OR felbinac OR dolinac OR flexfreeOR napageln OR target OR traxam OR fentiazac OR domureuma OR fentiazaco OR norvedan OR riscalon OR fepradinol ORdalgen OR flexidol OR cocresol OR rangozona OR reuflodol OR pinazone OR zepelin OR flufenamic OR dignodolin OR rheumaOR lindofluid OR sastridex OR lunoxaprofen OR priaxim OR flubiprofen OR fenomel OR ocufen OR ocuflur OR ldquoTrans ActLATrdquo OR tulip OR ibuprofen OR cuprofen OR ldquodeep relief rdquo OR fenbid OR ibu-cream OR ibugel OR ibuleve OR ibumousse ORibuspray OR ldquonurofen gelrdquo OR proflex OR motrin OR advil OR radian OR ralgex OR ibutop OR indomethacin OR indocin ORindospray OR isonixin OR nixyn OR ketoprofen OR tiloket OR oruvail OR powergel OR solpaflex OR ketorolac OR acular ORtrometamol OR meclofenamic OR naproxen OR naprosyn OR niflumic OR actol OR flunir OR niflactol topico OR niflugel ORnifluril OR oxyphenbutazone OR californit OR diflamil OR otone OR tanderil OR piketoprofen OR calmatel OR triparsean ORpiroxicam OR feldene OR pranoprofen OR oftalar OR pranox OR suxibuzone OR danilon OR flamilon OR ufenamate OR fenazolOR flector OR benzydamine)mp (558284)
3 1 OR 2 (664691)4 exp Administration Topical (69697)5 (topical OR cutaneous OR dermal OR transcutaneous OR transdermal OR percutaneous OR skin OR massage OR
embrocation OR gel OR ointment OR aerosol OR cream OR cregraveme OR lotion OR mouse OR foam OR liniment OR spray OR rubOR balm OR salve OR emulsion OR oil OR patch OR plaster)mp (1982395)
6 4 OR 5 (1997720)7 exp Athletic Injuries (21464)8 (strain OR sprain OR contusion OR distortion OR compression OR ldquosports injurrdquo OR ldquosoft tissue injurrdquo OR tendnitis OR
ldquomuscle painrdquo OR periarthritis OR epicondylitis OR tenosynovitis)mp (420133)9 7 OR 8 (439228)
10 Pain (113906)11 (pain OR analgesi)mp (585249)12 10 or 11 (585249)13 randomized controlled trialpt (401171)14 randomizedab (296222)15 placeboab (155341)16 drug therapyfs (1789858)17 randomlyab (207517)18 trialab (308477)19 groupsab (1318386)20 OR13-1921 3 AND 6 AND 9 AND 12 AND 20 (139)22 limit 21 to yr=ldquo2008-Currentrdquo (56)
120Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Appendix 3 EMBASE search strategy via Ovid (for 2015 update)
1 exp Anti-inflammatory Agents non-steroidal (452266)2 (bufexamac OR bufexine OR calmaderm OR ekzemase OR dicoflenac OR solaraze OR pennsaid OR voltarol OR emugel OR
voltarene OR voltarol OR optha OR voltaren OR etofenamate OR afrolate OR algesalona OR bayro OR deiron OR etofen ORflexium OR flogoprofen OR rheuma-gel OR rheumon OR traumalix OR traumon OR zenavan OR felbinac OR dolinac OR flexfreeOR napageln OR target OR traxam OR fentiazac OR domureuma OR fentiazaco OR norvedan OR riscalon OR fepradinol ORdalgen OR flexidol OR cocresol OR rangozona OR reuflodol OR pinazone OR zepelin OR flufenamic OR dignodolin OR rheumaOR lindofluid OR sastridex OR lunoxaprofen OR priaxim OR flubiprofen OR fenomel OR ocufen OR ocuflur OR ldquoTrans ActLATrdquo OR tulip OR ibuprofen OR cuprofen OR ldquodeep relief rdquo OR fenbid OR ibu-cream OR ibugel OR ibuleve OR ibumousse ORibuspray OR ldquonurofen gelrdquo OR proflex OR motrin OR advil OR radian OR ralgex OR ibutop OR indomethacin OR indocin ORindospray OR isonixin OR nixyn OR ketoprofen OR tiloket OR oruvail OR powergel OR solpaflex OR ketorolac OR acular ORtrometamol OR meclofenamic OR naproxen OR naprosyn OR niflumic OR actol OR flunir OR niflactol topico OR niflugel ORnifluril OR oxyphenbutazone OR californit OR diflamil OR otone OR tanderil OR piketoprofen OR calmatel OR triparsean ORpiroxicam OR feldene OR pranoprofen OR oftalar OR pranox OR suxibuzone OR danilon OR flamilon OR ufenamate OR fenazolOR flector OR benzydamine)mp (790435)
3 1 OR 2 (1108683)4 exp Administration Topical (68490)5 (topical OR cutaneous OR dermal OR transcutaneous OR transdermal OR percutaneous OR skin OR massage OR
embrocation OR gel OR ointment OR aerosol OR cream OR cregraveme OR lotion OR mouse OR foam OR liniment OR spray OR rubOR balm OR salve OR emulsion OR oil OR patch OR plaster)mp (3208844)
6 4 OR 5 (3208844)7 exp Athletic Injuries (24675)8 (strain OR sprain OR contusion OR distortion OR compression OR ldquosports injurrdquo OR ldquosoft tissue injurrdquo OR tendnitis OR
ldquomuscle painrdquo OR periarthritis OR epicondylitis OR tenosynovitis)mp (804322)9 7 OR 8 (825164)
10 Pain (216406)11 (pain OR analgesi)mp (971593)12 10 OR 11 (971593)13 clinical trialssh (841252)14 controlled clinical trialssh (389335)15 randomized controlled trialsh (357169)16 double-blind proceduresh (118945)17 (clin adj25 trial)ab (331002)18 ((doubl or trebl or tripl) adj25 (blind or mask))ab (141726)19 placeboab (203390)20 randomab (906176)21 OR13-20 (1731380)22 3 AND 6 AND 9 AND 10 AND 19 (439)23 limit 22 to yr=ldquo2008-Currentrdquo (204)
Appendix 4 Summary of outcomes efficacy
Study ID Treatment Clinical response Other response
Airaksinen 1993 (1) Ketoprofen gel 2 x 5 g (125mg) daily n = 29(2) Placebo gel n = 27
PGE rdquoimprovedldquo at 7 days(1) 2429(2) 1427
No additional data
121Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
kermark 1990 (1) Indomethacin spray 1 (El-metacin) 3-5 x 05-15 mL dailyn = 23(2) Indomethacin capsules 3 x 25mg daily n = 23(3) Placebo spray and capsules n= 24
No pain on palpation at 7 days(1) 1222(2) 522(3) 624
Patient assessment of improve-ment at 7 days (scale 0 to 100)(1) 57(2) 49(3) 30
Aoki 1984 (1) Piroxicam gel 5 3 or 4 x 1 gdaily n = 84(2) Indomethacin gel 1 3 or 4x 1 g daily n = 84(3) Placebo gel n = 84
PGE (5 point) rdquobetter or muchbetterldquo at 7 days(1) 5672(2) 4164(3) 3367
Pain on movement rdquoreducedldquo orrdquodisappearedldquo at 7 days(1) 4861(2) 3860(3) 3563
Auclair 1989 (1) Niflumic acid gel 25 3 x 5g daily n = 117(2) Placebo gel n = 110
PGE (5 point) rdquogood or verygoodldquo at 7 days(1) 69117(2) 54110
Pain on palpation rdquoimprovedldquo at7 days(1) 69117(2) 53110
Billigmann 1996 (1) Ibuprofen microgel 5 3 x200 mg daily n = 80(2) Placebo gel n = 80
Complete remission(1) 2580(2) 1080
Improvement in pain with move-ment of 20 at 7 days(1) 6580(2) 5580
Campbell 1994 (1) Ibuprofen cream 5 (Proflex)4 x 4rdquo daily n = 26(2) Placebo cream n = 25
Improvement in walking ability (4point) at 7 days(1) 2126(2) 1925
No additional data
Chatterjee 1977 (1) Benzydamine HCl cream 33 x daily n = 25(2) Placebo cream n = 25
Pain on movement ldquoabsentslightrdquoat 6 days(1) 2125(2) 1225
Tenderness with pressure ldquoabsentslightrdquo at 6 days(1) 2125(2) 1225
Costantino 2011 (1) DHEP-heparin plaster (Flec-torparin) x 1 daily n = 142(2) DHEP plaster (Flector) x 1daily n = 146(3) Placebo plaster n = 142
No responder analysis for efficacyOverall treatment efficacy not re-ported
Mean reduction from baseline forPI on movement at 3 days (fromgraph)(1) 24 mm(2) 19 mm(3) 14 mm
Coudreuse 2010 (1) DHEP-heparin plaster (Flec-torparin) n = 120(2) Placebo n = 120
PGE ldquogoodrdquo or ldquoexcellentrdquo at 7days(1) 99117(2) 82116These are results for physician-re-ported judgement - not partici-pant
Mean reduction from baseline forPI on movement over 6 hours(1) about 30(2) about 20Overall greater in DHEP-heparingroup than placebo over 7 days
122Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Curioni 1985 (1) Ibuproxam n = 20(2) Ketoprofen n = 20(3) Etofenamate n = 20
Resolution of symptoms by 7 days(1) 1520(2) 1320(3) 1320
PGE ldquogoodrdquo or ldquoexcellentrdquo at 10days(1) 1920(2) not reported(3) 1620
Diebshlag 1990 (1) Ketorolac gel 2 3 x 3 g dailyn = 13(2) Etofenamate gel 5 3 x 3 gdaily n = 12(3) Placebo gel n = 12
Improvement in pain at 7 days(1) 1213(2) 1012(3) 912
No additional data
Dreiser 1988 (1) Ibuprofen cream 5 3 x 4 cmdaily n = 32 (3 x 10 cm for largejoints)(2) Placebo cream n = 32
PGE ldquoimprovementrdquo or ldquocom-plete relief rdquo at 7 days(1) 2632(2) 1232
(1) significantly better than (2)for mean improvement in sponta-neous pain movement pain restpain tenderness to pressure (VAS)
Dreiser 1989 (1) Ketoprofen gel 25 2 x 5 cmdaily n = 30(2) Placebo gel n = 30
PGE (3 point) ldquobetterrdquo at 7 days(1) 1830(2) 530
(1) significantly better than (2) formean improvement in pain (restand movement) (VAS)
Dreiser 1990 (1) Niflumic acid gel 25 3 x 5g daily n = 30(2) Placebo gel n = 30
PGE (4 point) ldquocuredrdquo or ldquoim-provedrdquo at 7 days(1) 2330(2) 1030
(1) significantly better than (2) formean improvement in pain (VAS)
Dreiser 1994 (1) Flurbiprofen plaster 2 x 40 mgdaily n = 65(2) Placebo plaster n = 66
PGE (4 point) ldquogoodrdquo or ldquoverygoodrdquo at 7 days(1) 4865(2) 4166
(1) significantly better than (2)for mean improvement in sponta-neous pain but not pain on move-ment or palpation (VAS)
Fioravanti 1999 (1) DHEP lecithin gel 3 x 5 g (=65 mg) daily n = 50(2) DHEP gel 3 x 5 g (= 65 mg)daily n = 50
PGE (4 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 10 days(1) 3550(2) 3550
(1) significantly better than (2)for mean improvement in spon-taneous pain at 7 days but notfor pain on movement at 10 days(VAS)
Fujimaki 1985 (1) Piroxicam gel 05 3 or 4 x 1g daily n = 92(2) Indomethacin gel 1 3 or 4 x1 g daily n = 90(3) Placebo gel n = 89
PGE (5 point) ldquobetterrdquo or ldquomuchbetterrdquo at end of treatment at 14days(1) 4483(2) 4482(3) 4082
No additional data
Gallacchi 1990 (1) Diclofenac hy-droxyethylpyrrolidine gel 1 4 x2 g daily n = 25 (Flector gel)(2) Diclofenac sodium 1 4 x 2 g
PGE (5 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 14 days(1) 1925(2) 1925
No significant difference betweengroups for pain on applied pres-sure at 7 and 14 days
123Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
daily n = 25 (Voltaren Emugel)
Gonzaacutelez de Vega 2013 (1) Traumeel gel 3 x 2 g daily n =140(2) Traumeel ointment n = 143(3) Diclofenac gel 1 3 x 2 g dailyn = 137
Pain-free at 7 days(1) 7140(3) 8137PGE (5 point) ldquogoodrdquo or ldquoverygoodrdquo(1) 128140(3) 127137
Normal function (5 point) scoreof 0 or 1 at 14 days(1) 133140(3) 131137
Governali 1995 (1) Ketoprofen gel 5 3 x 2-3 gdaily n = 15(2) Ketoprofen cream 1 3 x 2-3 g daily n = 15
PGE (5 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 7 days(1) 1415(2) 415
No additional data
Gualdi 1987 (1) Flunaxaprofen gel 2 x 3-5 cmdaily n = 30(2) Ketoprofen gel 2 x 3-5 cmdaily n = 30
No dichotomous data No significant difference betweengroups for pain on movement at 7days
Haig 1986 (1) Benzydamine cream 3 6 xdaily n = 21(2) Placebo cream n = 22
Pain on movement ldquoimprovedrdquo by6 days(1) 1821(2) 1322
No additional data
Hoffmann 2012 (1) DHEP-heparin plaster (Flec-torparin) x 1 daily n = 121(2) DHEP (Flector) plaster x 1daily n = 115(3) Placebo plaster n = 118
PGE (5 point) ldquogoodrdquo or ldquoexcel-lentrdquo over 80 in all groups
Mean reduction from baseline inPI (100 mm VAS)Day 3(1) 191 mm(2) 114 mm(3) 52 mmDay 8(1) 443 mm(2) 372 mm(3) 296 mm
Hofman 2000 (1) Diclofenac sodium gel 1 4x 2 cm daily n = 69(2) Lysine clonixinate gel 5 4 x2 cm (225 mg) daily n = 73
PGE (3 point) at 8 days ldquogoodrdquo(1) 3869(2) 3673
No significant difference betweentreatments for any pain outcomes
Hosie 1993 (1) Felbinac foam 3 3 x 2 g daily+ placebo tablets 3 x 1 daily n =140 (127 analysed for efficacy)(2) Ibuprofen tablets 3 x 400 mgdaily + placebo foam 3 x 2 g dailyn = 147 (134 analysed for efficacy)
Pain on movement ldquononerdquo orldquomildrdquo at 7 days(1) 81127(2) 96133
Spontaneous pain ldquononerdquo orldquomildrdquo at 7 days(1) 99127(2) 108134
124Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Jenoure 1997 (1) DHEP plaster (Tissugel) 2 xdaily n = 44(2) Placebo plaster 2 x daily n =41
Baseline pain in two groups notbalanced and data in table and fig-ure do not agree so efficacy out-comes not used
No additional data
Joussellin 2003 (1) DHEP plaster (Flector Tis-sugel 1) 1 x daily n = 68(2) Placebo plaster 1 x daily n =66
PGE (4 point) ldquoexcellentrdquo at 7 days(1) 3668(2) 2466ge 30 reduction in PI at 7 days(1) 2568(2) 1166
(1) significantly better than (2) formean pain on movement at 6 days
Julien 1989 (1) Ketoprofen gel 25 2 x 5 cm(= 50 mg) daily n = 30(2) Placebo gel n = 30
PGE (4 point) ldquorecoveredrdquo at 7days(1) 1830(2) 630
PGE (4 point) ldquorecoveredrdquo or ldquoim-provedrdquo at 7 days(1) 2530(2) 1330
Klainguti 2010 (1) DHEP-heparin plaster (Flec-torparin) x 1 daily n = 62(2) DHEP plaster (Flector) x 1daily n = 61(3) Placebo n = 59
Overall treatment efficacy (partic-ipant and investigator) (5 point)ldquogoodrdquo or ldquoexcellentrdquo at 3 days(1) 5662(2) not reported(3) 4659
Mean reduction in PI at day 3(from graph)(1) 33 mm(2) Not reported(3) 24 mm
Kockelbergh 1985 (1) Ketoprofen gel 25 2 x 5 cm(= 15 mg) daily n = 38(2) Placebo gel n = 36
PGE (3 point) ldquogoodrdquo at 7 days(1) 3038(2) 2236
(1) and (2) slightly better than (3)for mean spontaneous pain at 7days
Kuehl 2011 (1) DETP 13 plaster 2 x dailyn = 207(2) Placebo plaster n = 211
No dichotomous data Percentage reduction from base-line at last application(1) 73(2) 62
Li 2013 (1) DHEP plaster (Flector) 2 xdaily n = 192(2) Placebo plaster 2 x daily n =192
ge 50 pain reduction at day 7(posthoc analysis)(1) 173192(2) 114192
Overall treatment efficacy (5point) ldquogoodrdquo or ldquoexcellentrdquo(1) 161192(2) 81192Participant and investigator ratingMean plusmn SD reduction in PI onmovement at 7 days(1) 538 plusmn 170(2) 370 plusmn 183
Linde 1985 (1) Benzydamine 3 cream 3 xdaily n = 50(2) Placebo gel n = 50
No pain on movement (walking)at 8 days(1) 3550(2) 4050
No additional data
125Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Machen 2002 (1) Ibuprofen gel 5 3 x daily n= 40(2) Placebo gel n = 41
PGE (5 point) ldquomarked improve-mentrdquo or ldquocomplete clearancerdquo at7 days(1) 2540(2) 941
Clinically meaningful (ge 30 mm)pain relief at day 7(1) 3040(2) 1641
Mahler 2003 (1) DHEP + lethicin gel 3 x 5 gdaily n = 52(2) DHEP gel 3 x 5 g daily n = 48
PGE (4 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 10 days(1) 4952(2) 3948
Mean reduction in pain on move-ment at 3 and 10 days significantlygreater with (1) than (2)
Maziegraveres 2005b (1) Ketoprofen plaster 100 mg x1 daily n = 81(2) Placebo plaster n = 82
PGE (4 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 14 days(1) 7281(2) 6082
All mean efficacy measures im-proved more for (1) than (2) mostwere statistically significant
Maziegraveres 2005a (1) Ketoprofen plaster 100 mg x1 daily n = 87(2) Placebo plaster n = 85
PGE (4 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 14 days(1) 5087(2) 4185
All mean efficacy measures im-proved more for (1) than (2) mostwere statistically significant
McLatchie 1989 (1) Felbinac gel 3 3 x 3 cm dailyn = 118(2) Placebo gel n = 113
No dichotomous data Patient daily self-assessment formean pain on rest movement atnight interference with normaland leisure activities show betterefficacy for (1) than (2) from day2 (VAS)
Morris 1991 (1) Felbinac gel 3 3 x 1 cm dailyn = 41(2) Placebo gel n = 43
PGE (5 point) ldquogoodrdquo or ldquoverygoodrdquo at 7 days(1) 2341(2) 2743
(1) better than (2) for meanimprovement in symptoms andsporting function at 7 days
NCT01255423 (1) Diclofenac sodium gel 1 x 4daily n = 104(2) Placebo gel x 4 daily n = 100
No dichotomous data VAS (mean plusmn SD) at 72 hours(baseline PI not reported)(1) 374 plusmn 252(2) 388 plusmn 241
NCT01272934 (1) Diclofenac sodium gel 1 x 4daily n = 104(2) Placebo gel x 4 daily n = 100
No dichotomous data VAS (mean plusmn SD) at 72 hours(baseline PI not reported)(1) 256 plusmn 159(2) 612 plusmn 166
NCT01272947 (1) Diclofenac sodium gel 1 x 4daily n = 104(2) Placebo gel x 4 daily n = 100
No dichotomous data VAS (mean plusmn SD) at 24 hours(baseline PI not reported)(1) 331 plusmn 214(2) 654 plusmn 169
126Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Noret 1987 (1) Ketoprofen gel 25 2 x 5 cm(75 mg) daily n = 48(2) Placebo gel n = 45
PGE (4 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 8 days(1) 3948(2) 945
Decrease in mean spontaneouspain significantly greater in (1)than (2) by 3 days
Parrini 1992 (1) Ketoprofen foam 15 3 x 2 g(200 mg) daily n = 83(2) Placebo foam n = 86
No dichotomous data Mean pain on movement andpressure significantly decreased by7 days in (1) compared with (2)
Picchio 1981 (1) Ibuprofen gel 10 3 x daily n = 20(2) Ketoprofen gel 1 3 x daily n= 20
No pain on movement at 8 days(1) 320(2) 020
Spontaneous pain ldquononerdquo at 8days(1) 620(2) 020
Predel 2004 (1) Diclofenac sodium plaster 2 xdaily (140 mgplaster) n = 60(2) Placebo plaster n = 60
PGE (4 point) ldquogoodrdquo ldquoexcellentrdquoat 7 days(1) 5560(2) 560
(1) better than (2) for reductionin tenderness pain and speed ofpain reduction
Predel 2012 (1) Diclofenac gel (VoltarenEmulgel 232) 2 x 5 cm daily n= 80(2)Diclofenac gel (Voltaren Emulgel232 ) 3 x 5 cm daily n = 80(3) Placebo gel n = 82
ge 50 red in PI on movement at5 days(1) 5780(2) 5980(3) 1782
PGE efficacy (5 point) ldquogoodrdquo orldquovery goodrdquo at 8 days(1) 6880(2) 7380(3) 2482
Predel 2013a (1) Diclofenac 4 spray gel 4 or5 sprays 3 x daily (96-120 mg di-clofenac sodium) n = 118(2) Placebo spray gel n = 114
None or slight PI on movement at7-8 days(1) 111118(2) 93114
None or slight PI on movement at3-4 days(1) 76118(2) 58114
Predel 2013b (1) Diclofenac gel (VoltarenEmulgel) 4 x 2 g daily n = 36(2) Placebo gel 4 x 2 g daily n =36
PGE (5 point) ldquogoodrdquo or ldquoexcel-lentrdquo(1) 3636(2) 736
ge 50 reduction in PI on move-ment after 48 hours(1) 3436(2) 336
Ramesh 1983 (1) Ibuprofen cream 5 3 or 4 x5-10 cm daily n = 40(2) Placebo cream n = 40
Pain on movement (4 point)ldquononerdquo or ldquoslightrdquo at 7 days(1) 2340(2) 2340
Physician global assessment at 10days ldquogoodrdquo(1) 2940(2) 1640
Rowbotham 2003 (1) Diclofenac epolamine plaster(Flector Tissuegel) 2 x daily n =191(2) Placebo plaster n = 181
Pain intensity le 210 for 2 daysor 4 consecutive evaluations by 7days(1) 75191(2) 48181
Mean pain on rest significantlybetter with (1) than (2) after 7 days
127Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Russell 1991 (1) Piroxicam gel 05 4 x 5 mgdaily n = 100(2) Placebo gel n = 100
PGE (4 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 8 days(1) 79100(2) 45100
Statistically greater reduction inmean pain on movement at 8 dayswith (1) than (2)
Saillant 1998 (1) DHEP plaster (Flector Tis-sugel 1) 1 x daily n = 70(2) Placebo plaster 1 x daily n =70
PGE (4 point) ldquoexcellentrdquo at 7 days(1) 4370(2) 2570
ge 30 decrease in PI at 7 days(1) 6470(2) 5070
Sanguinetti 1989 (1) Felbinac gel 3 3 x daily n =42(2) Placebo gel n = 40
PGE ldquogoodrdquo or ldquovery goodrdquo at 7days(1) 3442(2) 1140
(1) better than (2) by 2 days
Sinniger 1981 (1) Fentiazac cream 5 2 or 3 xdaily n = 10(2) Placebo cream n = 10
Complete pain relief within 10days(1) 710(2) 110
Improvement in active pain onmovement at 5 days(1) 67(2) 32
Spacca 2005 (1) DHEP lecithin gel (Effigel) 3x 5 g daily n = 79(2) Placebo gel n = 76
No dichotomous data Mean pain scores improved morerapidly in (1) than (2) - statisticallysignificant at 3 and 6 days
Sugioka 1984 (1) Piroxicam gel 05 3 or 4 x 1g daily n = 183(2) Indomethacin gel 1 3 or 4 x1 g daily n = 183
PGE (5 point) ldquobetterrdquo or ldquomuchbetterrdquo at 14 days(1) 85175(2) 55165
Pain on movement ldquoreducedrdquo orldquodisappearedrdquo at 7 days(1) 77175(2) 63165
Thorling 1990 (1) Naproxen gel 10 2-6 x dailyn = 60(2) Placebo gel n = 60
PGE (5 point) ldquogoodrdquo or ldquoverygoodrdquo at 7 days(1) 3860(2) 2760
Participants using naproxen im-proved more rapidly and had sig-nificantly lower severity scores byday 3
Tonutti 1994 (1) Ketoprofen gel 5 3 x 2-3 gdaily n = 15(2) Etofenamate gel 5 3 x 2-3g n = 15
PGE (4 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 7 days(1) 1015(2) 1115
Significant reductions in pain onmovement by 7 days in bothgroups
Vecchiet 1991 (1) Meclofenamic acid gel 5 2 x10 cm daily (2 g) n = 30(2) Placebo n = 30
PGE (4 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 10 days(1) 3030(2) 1930
(1) significantly better than (2)for mean improvement in sponta-neous pain movement pain func-tional restriction
Whitefield 2002 (1) Ibuprofen gel 5 + placebotablet 3 x daily n = 50(2) Ibuprofen 400 mg tablet +placebo gel 3 x daily n = 50
Participant satisfied at 7 days(1) 3050(2) 3650
ldquoCompletely betterrdquo at 14 days(1) 2450(2) 3050
128Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
DHEP diclofenac epolamine HCl hydrochloride n number PGE participant global evaluation PI pain intensity SD standarddeviation VAS visual analogue scale
Appendix 5 Summary of outcomes adverse events and withdrawals
Study ID Treatment Local AEs Systemic AEs Serious AEs Withdrawals
Airaksinen 1993 (1) Ketoprofen gel 2x 5 g (125 mg) dailyn = 29(2) Placebo gel n =27
(1) 529(2) 427
(1) 129 (nausea af-ter paracetamol)(2) 027
None AE noneOther nonereported
kermark 1990 (1)Indomethacin spray1 (Elmetacin) 3-5 x 05-15 mL dailyn = 23(2) Indomethacincapsules 3 x 25 mgdaily n = 23(3) Placebo sprayand capsules n = 24
(1) 422(2) 022(3) 024
No usable data -reported for eventsnot participants
None reported AE (1) 1 (2) 1 (3)0Lost to follow-up(1) 1 (2) 2 (3) 3
Aoki 1984 (1) Piroxicam gel5 3 or 4 x 1 gdaily n = 84(2) Indomethacingel 1 3 or 4 x 1 gdaily n = 84(3) Placebo gel n =84
(1) 179(2) 270(3) 274
None None reported AE none23 excluded for pro-tocol violations (1)7 (2) 7 (3) 926 withdrew for rea-sons unrelated totreatment (1) 5 (2)13 (3) 8
Auclair 1989 (1) Niflumic acidgel 25 3 x 5 gdaily n = 117(2) Placebo gel n =110
All AEs(1) 5123(2) 6116Most commonly cu-taneous eruptions
No usable data None reported AE (1) 1123 (2) 011626 excluded fromefficacy analysis fornot meeting entrycriteria and protocolviolations
Billigmann 1996 (1) Ibuprofen mi-crogel 5 3 x 200mg daily n = 80(2) Placebo gel n =
(1) 1180(2) 480
None reported None reported AE (1) 280 (aller-gic rash dermatitis)No reason given (1)380 (2) 580
129Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
80 Symptom-free (1)180 (2) 180
Campbell 1994 (1) Ibuprofen cream5 (Proflex) 4 x 4rdquodaily n = 26(2) Placebo cream n= 25
No data (1) 126 (headache)(2) 025
No data AE noneExclusions 493 presented late 2missing forms 1 ap-peared twice 43 didnot return diaries
Chatterjee 1977 (1) BenzydamineHCl cream 3 3 xdaily n = 25(2) Placebo cream n= 25
None None None AE none1 participant lostto follow-up (groupnot reported)
Costantino 2011 (1) DHEP-hep-arin plaster (Flector-parin) x 1 daily n =142(2) DHEP plaster(Flector) x 1 daily n= 146(3) Placebo plastern = 142
Possibly or probablydrug-related(1) 2142(2) 1145(3) 6142
All mild and goneby 14 days ex-cept 1 (moderate) inplacebo group
Not reportedNo systemic GIevents
None AE noneOther(1) 0142(2) 2146 (proto-col deviation lost tofollow-up)(3) 2142 (lost tofollow-up)
Coudreuse 2010 (1) DHEP-hep-arin plaster (Flector-parin) n = 120(2) Placebo n = 120
(1) 1117(2) 7116
All AEs mild ormoderate resolvedspontaneously
(1) 1117(2) 0116
None AE(1) 1117 (increasedoedema)(2) 0116Other(1) 6117 (recovery3 no reason 3)(2) 3116 (recovery1 no reason 2)
Curioni 1985 (1) Ibuproxam n =20(2) Ketoprofen n =20(3) Etofenamate n =20
None None None AE noneOther none
Diebshlag 1990 (1) Ketorolac gel2 3 x 3 g daily n =13(2) Etofenamate gel5 3 x 3 g daily n =12
(1) 113(2) 112(3) 012
None None AE none1 ketorolac partic-ipant did not at-tend 15 day follow-up due to car acci-
130Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
(3) Placebo gel n =12
dent
Dreiser 1988 (1) Ibuprofen cream5 3 x 4 cm dailyn = 32 (3 x 10 cmfor large joints)(2) Placebo cream n= 32
No usable data None Not reported AE none4 placebo partici-pants lost to follow-up
Dreiser 1989 (1) Ketoprofen gel25 2 x 5 cm dailyn = 30(2) Placebo gel n =30
(1) 030(2) 230
None None reported AE (2) 230 (intol-erance)LoE (1) 130 (2) 130
Dreiser 1990 (1) Niflumic acidgel 25 3 x 5 gdaily n = 30(2) Placebo gel n =30
(1) 030(2) 330
None None AE (2) 130 (ery-thema)Exclusion 1 from(2) from efficacyanalysis for inade-quate baseline pain
Dreiser 1994 (1)Flurbiprofen plaster2 x 40 mg daily n =65(2) Placebo plastern = 66
(1) 265(2) 066
None None AE 0(1) 165 excludedfrom efficacy analy-sis for protocol vio-lation(2) 266 (1 LoE 1cured)
Fioravanti 1999 (1) DHEP lecithingel 3 x 5 g (= 65 mg)daily n = 50(2) DHEP gel 3 x 5g (= 65 mg) daily n= 50
(1) 050(2) 150
No data None reported AE noneOther none
Fujimaki 1985 (1) Piroxicam gel 05 3 or 4 x 1 g dailyn = 92(2) Indomethacingel 1 3 or 4 x 1 gdaily n = 90(3) Placebo gel n =89
(1) 183(2) 582(3) 282
(1) 083(2) 182 (nauseaand vomiting)(3) 082
None AE (1) 0 (2) 4 (3)0Unknown reasons(1) 2 (2) 1Did not return af-ter 1st visitirregularvisits (1) 6 (2) 6(3) 7
Gallacchi 1990 (1) Diclofenac hy-droxyethylpyrro-lidine gel 1 4 x 2
No AEs None None AE noneOther none
131Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
g daily n = 25 (Flec-tor gel)(2) Di-clofenac sodium 14 x 2 g daily n = 25(Voltaren Emugel)
Gonzaacutelez de Vega2013
(1) Traumeel gel 3 x2 g daily n = 140(2) Traumeel oint-ment n = 143(3) Diclofenac gel1 3 x 2 g daily n =137
No usable data
Infrequent mild tomoderate
No usable data None AE noneOther(1) 11140(2) 12137
Governali 1995 (1) Ketoprofen gel5 3 x 2-3 g dailyn = 15(2) Ketoprofencream 1 3 x 2-3 gdaily n = 15
No side effects None None AE noneOther none
Gualdi 1987 (1) Flunaxaprofengel 2 x 3-5 cm dailyn = 30(2) Ketoprofen gel 2x 3-5 cm daily n =30
(1) 130(2) 330
No data None reported AE noneOther none
Haig 1986 (1) Benzydaminecream 3 6 x dailyn = 21(2) Placebo cream n= 22
No AEs reported None None reported AE none reportedOther no data
Hoffmann 2012 (1) DHEP-hep-arin plaster (Flector-parin) x 1 daily n =121(2) DHEP (Flector)plaster x 1 daily n =115(3) Placebo plastern = 118
Most AEs were mi-nor local reactions(eg pruritus anderythema) in areaof plaster of mildto moderate inten-sity and resolvedwithout interrupt-ing treatment
No treatment-related systemic AEsrecorded
(2) 1 participanthad 3 SAEs nonejudged related tostudy medication
AE noneExclusions(1) 5121(2) 10115(3) 7118Excluded fromper protocol analysisdue to poor compli-ance or personal de-cision
Hofman 2000 (1) Diclofenacsodium gel 1 4 x2 cm daily n = 69(2) Lysine clonixi-
(1) 158(2) 161
None None AE noneLoE (1) 9 (2) 8
132Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
nate gel 5 4 x 2cm (225 mg) dailyn = 73
Hosie 1993 (1) Felbinac foam3 3 x 2 g daily+ placebo tablets 3x 1 daily n = 140(127 analysed for ef-ficacy)(2) Ibupro-fen tablets 3 x 400mg daily + placebofoam 3 x 2 g daily n= 147 (134 analysedfor efficacy)
(1) 1127(2) 3134
GI events (1) 14127 (2) 11134For (1) more mildnone definitely drugrelated for(2) definitely relatedto study drug
None AE noneExclusions (1) 13(2) 13 did not returnfor 7 day follow-up
Jenoure 1997 (1) DHEP plaster(Tissugel) 2 x dailyn = 44(2) Placebo plaster 2x daily n = 41
(1) 144(2) 141
No data None reported AE none reportedOther nonereported
Joussellin 2003 (1) DHEPplaster (Flector Tis-sugel 1) 1 x dailyn = 68(2) Placebo plaster 1x daily n = 66
(1) 168 (pruritus)(2) 366 (pruritus 2burning 1)
All AEs mild ormoderate
(1) 168 (allergic re-action)(2) 066
None reported AE(1) 066(2) 166Other(1) 366(2) 266
Julien 1989 (1) Ketoprofen gel25 2 x 5 cm (= 50mg) daily n = 30(2) Placebo gel n =30
(1) 130(2) 030
Not reported None AE noneOther none
Klainguti 2010 (1) DHEP-hep-arin plaster (Flector-parin) x 1 daily n =62(2) DHEP plaster(Flector) x 1 daily n= 61(3) Placebo n = 59
(1) 062(2) 061(3) 159All AEs mild in na-ture and resolvedspontaneously
(1) 165 (facial in-fection)(2) 161 (abdominalpain)(3) 059
None AE noneOther(1) 365(2) 161(3) 359
Kockelbergh 1985a (1) Ketoprofen gel25 2 x 5 cm (= 15mg) daily n = 38(2) Placebo gel n =
(1) 138(2) 126
Not reported None AE noneOther none
133Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
36
Kuehl 2011 (1) DETP 13plaster 2 x daily n =207(2) Placebo plastern = 211
(1) 16207(2) 12211
(1) 15207(2) 23211
None AE(1) 4207(2) 9211LoE(1) 21207(2) 25211Other(1) 58207(2) 62211
Li 2013 (1) DHEP plaster(Flector) 2 x daily n= 192(2) Placebo plaster 2x daily n = 192
(1) 4192(2) 3192
(1) 10192(2) 4192
None AE(1) 2192(2) 0192Other(1) 5192(2) 1192
Linde 1985 (1) Benzydamine3 cream 3 x dailyn = 50(2) Placebo gel n =50
(1) 440(2) 241
None None AE none(1) 6 (2) 6 excludedfrom 1st assessment(1) 3 (2) 4 ex-cluded from final as-sessment
Machen 2002 (1) Ibuprofen gel5 3 x daily n = 40(2) Placebo gel n =41
(1) 440(2) 241
None None AE none(1) 1 LoE 1 proto-col violation(2) 4 LoE
Mahler 2003 (1) DHEP + lethicingel 3 x 5 g daily n =52(2) DHEP gel 3 x 5g daily n = 48
(1) 152(2) 048
(1) 152(2) 048
None AE none5 lost to follow-up
Maziegraveres 2005b (1) Ketoprofen plas-ter 100 mg x 1 dailyn = 81(2) Placebo plastern = 82
At 21 days(1) 1281(2) 682
(1) 1381(2) 1482
None AE (1) 381(1) 781 (1 LoE 6cured)(2) 782 (5 LoE 2cured)
Maziegraveres 2005a (1) Ketoprofen plas-ter 100 mg x 1 dailyn = 87(2) Placebo plastern = 85
At 21 days(1) 2987(2) 2785
(1) 1187(2) 785
None AE (1) 987 (2) 685(1) 687 (2 LoE 4cured)(2) 585 (4 LoE 1cured)
134Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
McLatchie 1989 (1) Felbinac gel 33 x 3 cm daily n =118(2) Placebo gel n =113
(1) 3118(2) 2113Mild transient localirritation
None reported None AE noneOther none
Morris 1991 (1) Felbinac gel 33 x 1 cm daily n =41(2) Placebo gel n =43
None None None AE none(1) 4 (protocol vio-lations)(2) 1 (lost to follow-up)Exclusions 11 fromefficacy analysis be-cause evaluated by4 different investi-gators
NCT01255423 (1) Diclofenacsodium gel 1 x 4daily n = 104(2) Placebo gel x 4daily n = 100
(1) 1104(2) 3102
Total AE(1) 11104(2) 8102
None None
NCT01272934 (1) Diclofenacsodium gel 1 x 4daily n = 104(2) Placebo gel x 4daily n = 100
(1) 1102(2) 0103
Total AE (excludingSAE)(1) 6102(2) 3103
(1) 0102(2) 1103 (rupturedligaments in wrist)
AE see SAE
NCT01272947 (1) Diclofenacsodium gel 1 x 4daily n = 104(2) Placebo gel x 4daily n = 100
None (1) 2104(2) 2100
None None
Noret 1987 (1) Ketoprofen gel25 2 x 5 cm (75mg) daily n = 48(2) Placebo gel n =45
(1) 151(2) 047
None reported Not reported AE (1) 151 (skinallergy)(1) 1 LoE 1 unre-lated to trial(2) 1 LoE 1 unre-lated to trial
Parrini 1992 (1) Ketopro-fen foam 15 3 x 2g (200 mg) daily n= 83(2) Placebo foam n= 86
None None None AE noneOther none
135Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Picchio 1981 (1) Ibuprofen gel10 3 x daily n =20(2) Ketoprofen gel1 3 x daily n = 20
None None None AE noneOther not reported
Predel 2004 (1) Diclofenacsodium plaster 2 xdaily (140 mgplas-ter) n = 60(2) Placebo plastern = 60
12 participants ex-pe-rienced 16 mild AEswith no differencesbetween groups
None None AE (1) 160Other none
Predel 2012 (1) Diclofenac gel(Voltaren Emulgel232) 2 x 5 cmdaily n = 80(2) Diclofenac gel(Voltaren Emulgel232) 3 x 5 cmdaily n = 80(3) Placebo gel n =82
(1) 080(2) 180(3) 182All mild to moder-ate
(1) 280(2) 180(3) 382
None AE(1) 080(2) 080(3) 182Other(1) 180(2) 280(3) 282(protocol violationslost to follow-up)
Predel 2013a (1) Diclofenac 4spray gel 4 or 5sprays 3 x daily (96-120 mg diclofenacsodium) n = 118(2) Placebo spraygel n = 114
(1) 1118(2) 4114
(1) 5118(2) 4114All AEs mild re-versible
None AE(1) 1118(2) 1114Other(1) 3118(2) 43114
Predel 2013b (1) Diclofenac gel(Voltaren Emulgel)4 x 2 g daily n = 36(2) Placebo gel 4 x 2g daily n = 36
None (1) 036(2) 136
None AE noneOther none
Ramesh 1983 (1) Ibuprofen cream5 3 or 4 x 5-10 cmdaily n = 40(2) Placebo cream n= 40
(1) 140(2) 140
None reported Not reported AE (1) 140 (2) 140Other none
Rowbotham 2003 (1) DHEP plaster(Flector Tissuegel) 2x daily n = 191(2) Placebo plastern = 181
(1) 27191 (pruritis14)(2) 31181 (pruritis21)
(1) 21191(2) 22181
None re-ported (ldquovast major-ity mildrdquo)
AE none(1) 3191 (2) 4181(did not finish trialand complete dailydiaries)
136Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Russell 1991 (1) Piroxicam gel 05 4 x 5 mg daily n= 100(2) Placebo gel n =100
(1) 4102(2) 10102
GI or CNS events(1) 4 (2) 7Any AE (1) 7102(2) 15102
None reported AE (1) 1102 (2) 8102Other (1) 6 LoE 1ldquootherrdquo(2) 42 LoEExclusions7 did not complywith study medica-tion schedule 6 lostto follow-up 1 pro-tocol violation
Saillant 1998 (1) DHEPplaster (Flector Tis-sugel 1) 1 x dailyn = 70(2) Placebo plaster 1x daily n = 70
None None None AE noneOther(1) 570(2) 570
Sanguinetti 1989 (1) Felbinac gel 33 x daily n = 42(2) Placebo gel n =40
(1) 342(2) 140
None None reported AE noneOther nonereported
Sinniger 1981 (1) Fentiazac cream5 2 or 3 x daily n= 10(2) Placebo cream n= 10
ldquoNo untoward sideeffectsrdquo
None None AE noneOther nonereported
Spacca 2005 (1) DHEP lecithingel (Effigel) 3 x 5 gdaily n = 79(2) Placebo gel n =76
ldquoNo signsof cutaneous irrita-tion or sensitisationobservedrdquo
No AEs observed None AE noneOther nonereported
Sugioka 1984 (1) Piroxicam gel 05 3 or 4 x 1 g dailyn = 183(2) Indomethacingel 1 3 or 4 x 1 gdaily n = 183
(1) 5178(2) 26179
None reported None reported AE none reportedExclusions due toprotocol violations(1) 8 (2) 18Withdrawals(1) 11 (2) 12
Thorling 1990 (1) Naproxen gel10 2-6 x daily n =60(2) Placebo gel n =60
(1) 160(2) 060
None None AE none(1) 1 LoE 1 proto-col violation(2) 1 participant re-quest
137Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Tonutti 1994 (1) Ketoprofen gel5 3 x 2-3 g dailyn = 15(2) Etofenamate gel5 3 x 2-3 g n =15
None No AEs attributableto the medication
None AE NoneLoE (1) 1 (2) 2
Vecchiet 1991 (1) Meclofe-namic acid gel 5 2x 10 cm daily (2 g)n = 30(2) Placebo n = 30
Tolerabil-ity excellent or goodin nearly all partici-pants
No data None AE none reported(2) 5 lost to follow-up
Whitefield 2002 (1) Ibuprofen gel5 + placebo tablet3 x daily n = 50(2) Ibuprofen 400mg tablet + placebogel 3 x daily n = 50
No data 6 AEs reportednone judged relatedto study medication
None reported AE noneRecovered (1) 3 (2)2LoE (2) 1Lost to follow-up(1) 1 (2) 1
AE adverse event CNS central nervous system DHEP diclofenac epolamine GI gastrointestinal HCl hydrochloride LoE lackof efficacy n number SAE serious adverse event
Appendix 6 Concentration amount and frequency of dosing
Study Drug Concentration Quantity Frequency Estimated daily dose of topi-cal NSAID
Joussellin 2003 Diclofenac 1 Plaster 1 180 mg epolamine salt 140 mgNa salt
Li 2013 Diclofenac 1 Plaster 2 360 mg epolamine salt 280 mgNa salt
Rowbotham 2003 Diclofenac 1 Plaster 2 360 mg epolamine salt 280 mgNa salt
Saillant 1998 Diclofenac 1 Plaster 1 180 mg epolamine salt 140 mgNa salt
Coudreuse 2010 Diclofenac 1 Plaster 1 180 mg epolamine salt 140 mgNa salt
Klainguti 2010 Diclofenac 1 Plaster 1 180 mg epolamine salt 140 mgNa salt
138Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Predel 2004 Diclofenac - Plaster 2 280 mg Na salt
Predel 2012 Diclofenac 232 5 cm ribbon~ 2 g 2 or 3 92-138 mg (Na equiv) as di-ethylamine salt
Predel 2013b Diclofenac 116 2 g 4 92 mg (Na equiv) as diethy-lamine salt
Predel 2013a Diclofenac 4 4-5 sprays 3 96-120 mg Na salt
Costantino 2011 Diclofenac 1 Plaster 1 180 mg epolamine salt 140 mgNa salt
Fioravanti 1999 Diclofenac - 5 g 3 195 mg epolamine salt
Gallacchi 1990 Diclofenac 1 2 g 4 80 mg
Gonzaacutelez de Vega2013
Diclofenac 1 6 cm 2g 3 60 mg
Hoffmann 2012 Diclofenac 1 Plaster 1 180 mg epolamine salt 140 mgNa salt
Hofman 2000 Diclofenac 1 2 cm 4 probably 30-40 mg Na salt
Jenoure 1997 Diclofenac 1 Plaster 2 360 mg epolamine salt 280 mgNa salt
Mahler 2003 Diclofenac - 5 g 3 195 mg epolamine salt
NCT01255423 Diclofenac 1 - 4 -
NCT01272934 Diclofenac 1 - 4 -
NCT01272947 Diclofenac 1 - 4 -
Spacca 2005 Diclofenac - 5 g 3 195 mg epolamine salt
Campbell 1994 Ibuprofen 5 4 inch ribbon 4 Assume 800 mg
Dreiser 1988 Ibuprofen 5 4 cm ribbon(10 cm for largerjoints)
4 Assume up-800 mg
Ramesh 1983 Ibuprofen 5 5-10 cm ribbon 3-4 Assume 300-800 mg
Billigmann 1996 Ibuprofen 5 10 cm 4 g gel 3 600 mg
139Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Machen 2002 Ibuprofen 5 - 3 -
Picchio 1981 Ibuprofen 10 - 3 -
Whitefield 2002 Ibuprofen 5 - 3 -
Maziegraveres 2005b Ketoprofen - Plaster 1 100 mg
Maziegraveres 2005a Ketoprofen - Plaster 1 100 mg
Airaksinen 1993 Ketoprofen - 5 g 2 125 mg
Dreiser 1989 Ketoprofen 25 5 cm 2 100 mg
Julien 1989 Ketoprofen 25 5 cm 2 100 mg
Noret 1987 Ketoprofen 25 5 cm 75 g 2 375 mg
Curioni 1985 Ketoprofen No details 2 -
Governali 1995 Ketoprofen 5 2-3 g 3 300-450 mg
Governali 1995 Ketoprofen 1 2-3 g 3 60-90 mg
Gualdi 1987 Ketoprofen - 3-5 cm 2 -
Kockelbergh 1985 Ketoprofen 25 5 cm 2 100 mg
Parrini 1992 Ketoprofen 150 2 g 3 600 mg
Picchio 1981 Ketoprofen 1 - 3 -
Tonutti 1994 Ketoprofen 5 2-3 g 3 300-450 mg
kermark 1990 Indomethacin 1 05-15 mL 3-5 12-60 mg
Aoki 1984 Indomethacin 1 1 g 3-4 30-40 mg
Fujimaki 1985 Indomethacin 1 1 g 3-4 30-40 mg
Sugioka 1984 Indomethacin 1 1 g 3-4 30-40 mg
Aoki 1984 Piroxicam 05 1 g 3-4 15-20 mg
Fujimaki 1985 Piroxicam 05 1 g 3-4 15-20 mg
140Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Russell 1991 Piroxicam 05 5 mg 4 20 mg
Sugioka 1984 Piroxicam 05 1 g 3-4 15-20 mg
Chatterjee 1977 Benzydamine 3 - 3 -
Haig 1986 Benzydamine 3 - 6 -
Linde 1985 Benzydamine 3 - 3 -
Auclair 1989 Niflumic acid 25 10 cm 5 g 3 375 mg
Dreiser 1990 Niflumic acid 25 10 cm 5 g 3 375 mg
Curioni 1985 Ibuproxam gel 10 - 2 -
Curioni 1985 Etofenamate No details - 2 -
Diebshlag 1990 Etofenamate 5 3 g 3 450 mg
Tonutti 1994 Etofenamate 5 2-3 g 3 300-450 mg
Diebshlag 1990 Ketorolac 2 3 g 3 360 mg
Dreiser 1994 Flurbiprofen Patch 2 80 mg
Gualdi 1987 Flunoxaprofen - 3-5 cm 2 -
Hofman 2000 Lysine clonixinate 5 2 cm 4 90 mg
Hosie 1993 Felbinac 3 2 g 3 180 mg
McLatchie 1989 Felbinac 3 3 cm 3 -
Morris 1991 Felbinac 3 1 cm 3 -
Sanguinetti 1989 Felbinac 3 - 3 -
Sinniger 1981 Fentiazac 5 Varied according toinvolved areas
2-3 -
Thorling 1990 Naproxen 10 - 2-6 -
Vecchiet 1991 Meclofenamic acid 5 10 cm 4 g 2 400 mg
equiv equivalent Na sodium
141Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
F E E D B A C K
Query on formulations of topical NSAIDs particularly DMSO from Dr Chrubasik 11 April 2012
Summary
Dr Chrubasik highlighted this letter to the Editor postgradmedorgdoi103810pgm2011092482
DMSO [dimethyl sulphoxide] but also other additives eg nonivamide (which is a capsaicinoid added as drug enhancer) may contributeto the overall effect of topical NSAIDs Nonivamide certainly contributes to the analgesic effect and to adverse events (heat sensationburning pruritus etc) This has not been considered in the Cochrane review by Massey T Derry S Moore RA McQuay HJ TopicalNSAIDs for acute pain in adults but Dr Chrubasik believes should be done otherwise the effect size of the NSAID topicals is favoured
Reply
We have been asked by Dr Chrubasik to comment on a letter (Roth 2011) about the formulations of topical NSAIDs particularly howDMSO and other penetration enhancers can affect efficacy estimates or adverse event reporting in osteoarthritis It was suggested thatthe review of Topical NSAIDs for acute pain in adults did not consider this resulting in a bias towards the topical NSAIDThere are a number of points to be made here
1 Penetration enhancers are used in formulations of topical products to encourage local absorption through the skin and produce ahigh local concentration Topical NSAIDs use penetration enhancers and the result is high local concentration in joints for instancebut low systemic concentrations (Moore 2008) That is how they work Formulation is an important part of medicinal chemistry as awhole not just for topical agents
2 In our analysis of topical NSAIDs we were aware that a range of properties are or have been ascribed to the analgesia resultingfrom application of topical agents and which could contribute to overestimation of treatment effect of topical NSAID These includefeelings of heat or cold and even the act of rubbing itself For that reason we have chosen to include only double-blind studies wherethe placebo agent is identical to the active with the exception of course of the NSAID So heat cold rubbing and penetrationenhancers should be identical as best we can judge That leaves only the NSAID itself to provide any additional analgesic effect andit is that which we measure This is analogous for example to use of acupuncture say where the better studies show no differencebetween ldquotruerdquo acupuncture and ldquoshamrdquo acupuncture performed at nonspecific sites but better than non-treatment controls Theargument that we should only use high quality studies to evaluate evidence about pain interventions is well made
3 Overestimation of analgesic effect because of effects of enhancers themselves would be better made in direct comparisons oftopical and oral NSAIDs where local or even systemic effects would not be balanced in the oral study arm However our reviewconcentrated on placebo-controlled studies and had few studies with active controls Moreover the real test would be in chronicrather than acute conditions with long duration (12 week) outcomes using current best evidence rules (Moore 2010) includingimputation (Moore 2012) In their response to Rothrsquos letter the authors of the original review of products available in the USA showrather similar effect sizes of oral diclofenac and topical diclofenac with different penetration enhancers (Barthel 2011) in such studies
4 The Roth letter sought to differentiate between topical diclofenac preparations based on the penetration enhancers used Thatdifferent formulations may have different effect sizes is a fair point to make Two of the studies in our review of topical NSAIDs inacute conditions used diclofenac sodium 1 gel comparing it with either diclofenac epolamine gel (Gallacchi 1990 50 participants)or lysine clonixinate gel (Hofman 2000 142 participants) no difference between formulations was demonstrated It is difficult tomake any judgement for topical NSAIDs in acute conditions due to the relatively small number and particularly the small size ofstudies We did an analysis by drug and this showed that some topical NSAIDs were consistently beneficial irrespective offormulation while others had little or no efficacy This fits in with some theoretical considerations of molecular architecture andtissue penetration (Moore 2008) In chronic pain where there are larger studies and much more data we have considered formulation(Derry in preparation)
References
Roth SH Letter to the editor The importance of differentiating between topical NSAIDs Postgraduate Medicine 2011123251-2[103810pgm2011092482]Moore RA Derry S McQuay HJ Topical agents in the treatment of rheumatic pain Rheumatic Diseases Clinics of North America200834415-32
142Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Moore RA Eccleston C Derry S Wiffen P Bell RF Straube S et al ACTINPAIN Writing Group of the IASP Special Interest Groupon Systematic Reviews in Pain Relief Cochrane Pain Palliative and Supportive Care Systematic Review Group Editors ldquoEvidencerdquo inchronic pain - establishing best practice in the reporting of systematic reviews Pain 2010150386-9Moore RA Straube S Eccleston C Derry S Aldington D Wiffen P et al Estimate at your peril imputation methods for patientwithdrawal can bias efficacy outcomes in chronic pain trials using responder analyses Pain 2012153265-8Barthel HR Axford-Gatley RA Response to Roth Letter to the Editor Postgraduate Medicine 2011123253-4 [103810pgm2011092483]
Contributors
Feedback from Sigrun ChrubasikAuthors involved with responding Andrew Moore Sheena DerryFeedback Editor Kate Seers
Query submitted by Peter C Goslashtzsche 3 November 2015
Summary
Date of Submission 03-Nov-2015Name Peter C GoslashtzscheEmail Address pcgcochranedkAffiliation Nordic Cochrane CentreRole DirectorComment The authors found that the results were missing from 5900 patients Furthermore there was extreme heterogeneity in theirmeta-analyses eg I square was 92 for the diclofenac trials which were the most common ones and there was extreme funnel plotasymmetry with the largest trials showing the smallest effects (the authors didnrsquot show funnel plots but I constructed one for diclofenac)Moreover the trials were industry funded of relatively poor quality and the authors analysed published data not data from clinicalstudy reports and did not try to obtain all the missing trials and data from the manufacturersThe authors concluded that topical NSAIDs are effective in providing pain relief but also cautioned that the large amounts of unpublisheddata ldquocould influence results in updates of this reviewrdquo They certainly could I believe it is plain wrong to perform meta-analyses on theauthorsrsquo data When I most recently reviewed this area for the BMJ in 2010 I concluded that we donrsquot know whether topical NSAIDsare beneficial (1)1 Goslashtzsche PC NSAIDs Clin Evid (Online) 2010 Jun 28I agree with the conflict of interest statement belowI certify that I have no affiliations with or involvement in any organization or entity with a financial interest in the subject matter ofmy feedback
Reply
Response submitted by authors Sheena Derry and Professor Andrew Moore on 4 November 2015Peter has made these same points elsewhere [PubMed Commons PubPeer gt Cochrane Database Syst Rev October 2015 httpspubpeercompublicationsA9E5BEA36549727357F9FD14CC2537]As a preliminary it is important to stress that the number of trials available in pain has increased over time and newer trials are typicallybetter and larger This means we can move from answering the simple question of whether an intervention works to more importantand relevant questions such as how well the intervention works and for drugs particularly to examine effects of dose and formulationFormulation is now recognised to have profound effects For oral analgesics for example fast acting formulations demonstrate up todouble the analgesic effect for a given dose as our recently updated overview points out (Cochrane Database Syst Rev 2015 Sep 289CD008659 see also Pain 2014 Jan155(1)14-21)What was true for oral analgesics in acute pain is now true also for topical NSAIDs in acute pain where manufacturers have beenputting a lot of effort into trying to produce new formulations that work better This updated review sought to examine the influenceof formulation and drug although dose is somewhat more difficult There are good reasons why formulation may be important(Rheumatic Disease Clinics 2008 Vol 34 Issue 2 p 415-432)
143Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Our searches did identify a large number of unpublished trials We deplore this but are powerless to change things For a previousreview of topical NSAIDs we contacted all identified manufacturers and asked for published or unpublished data The yield was small(as others have found) but some unpublished studies were brought into the public domain Waiting for all studies and clinical trialreports of all studies would take forever and could probably never be achieved We believe that many of these unpublished studiesrelate to drugs andor formulations that have never been manufactured commercially While these would be of interest in determiningwhat does and doesnrsquot work and to direct future research they would probably have little clinical relevance because these formulationsare unlikely to come to marketPeterrsquos main issue appears to be heterogeneity Tests for heterogeneity are problematical anyway (Pain 2000 85415-24) and the I squareof 92 that he quotes was for all topical diclofenac formulations combined While we do give an overall summary for diclofenac wedemonstrated in the review that the different diclofenac formulations produced different results from one another using LrsquoAbbeacute plotsand in our detailed analyses showing large variations in efficacy between them In the circumstance a high I square for all combined(clinical heterogeneity) is to be expected but it is not relevant The bulk of the studies on diclofenac were published in the last fiveyears were of decent quality and moderate to large size There were older data for ketoprofen but again these showed major differencesbetween formulations Differences between formulations are highlighted throughout the reviewTrying to determine publication bias using funnel plots or other measures is something of a lost cause (Journal of Clinical Epidemiology200053207-16) It is especially so with small numbers of trials (Journal of Clinical Epidemiology 2000 53 477-484) and makingsense of funnel plots is anything but easy for most people (Journal of Clinical Epidemiology 2005 58 894-901) A useless methodseems an odd choice to make to criticise our reviewThere are very good scientific reasons why drug and formulation may play a big part in the efficacy of topical NSAIDs This is alsothe case for oral analgesics used in acute pain where formulation improvements generating rapid absorption confers greater efficacySimple lumping strategies may have been permissible in past systematic review methodology but a more forensic approach is needednow and in the future This is what we have attempted to do in this latest reviewComparisons with Peterrsquos 2010 review seem inappropriate since that review was based on other reviews that are now out of date
Contributors
Feedback from Peter C GoslashtzscheAuthors involved with responding Andrew Moore Sheena DerryFeedback Editor Kate Seers
W H A T rsquo S N E W
Last assessed as up-to-date 4 February 2015
Date Event Description
4 November 2015 Feedback has been incorporated Feedback has been incorporated See Feedback 2
H I S T O R Y
Protocol first published Issue 4 2008
Review first published Issue 6 2010
144Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Date Event Description
13 October 2015 Amended Small errors found in Summary Table B for total num-ber of participants with piroxicam and indomethacinand in text for NNT with diclofenac (other gel) Nochange to conclusions
6 October 2015 Amended Small error found in Summary Table B percentagesfor NSAID and placebo were the wrong way round forcomparison of all NSAIDs and placebo RR and textwere correct No change to results or conclusions
13 February 2015 New citation required but conclusions have notchanged
Conclusions not changed Results remain essentiallythe same but the focus has changed to examine drugand formulation combination
3 February 2015 New search has been performed Title changed from Topical NSAIDs for acute pain inadults to Topical NSAIDs for acute musculoskeletalpain in adults to increase specificityNew searches run and new studies identified Fourteennew included studies using diclofenac (3489 new par-ticipants a 63 increase) four new excluded studiesFifteen studies awaiting classification (completed noresults available but likely to satisfy inclusion criteria)
23 May 2014 Amended Error in data reported for clinical success in Hosie 1993was brought to our attention and has been corrected
12 June 2012 Feedback has been incorporated We have incorporated feedback received from DrSigrun Chrubasik and the authorrsquos response on DMSOand other additives
C O N T R I B U T I O N S O F A U T H O R S
For the earlier review Tom Massey and SD identified studies and carried out data extraction analysis and drafting RAM and HJMwere involved in planning acted as adjudicators and were involved with writing
For this update SD and RAM carried out searches data extraction and analysis All authors were involved with writing the review
145Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
D E C L A R A T I O N S O F I N T E R E S T
SD has no conflicts relating to this review or any similar product
RAM has no conflicts relating to this review or any similar product
PW has no conflicts relating to this review or any similar product
HG has no conflicts relating to this review or any similar product
MM has no conflicts relating to this review or any similar product
S O U R C E S O F S U P P O R T
Internal sources
bull Oxford Pain Relief Trust UKGeneral institutional support
External sources
bull No sources of support supplied
D I F F E R E N C E S B E T W E E N P R O T O C O L A N D R E V I E W
For this update in 2015 we have changed the title to specify musculoskeletal pain because topical NSAIDs are not normally used totreat visceral pain or headache We felt that the new title better reflected the content of the review We have also changed the focus ofthe review from pooled analysis of all topical NSAIDs and all studies of a particular NSAID to an examination of individual drug andits formulation This makes the review much more relevant We have expanded the rsquoRisk of biasrsquo assessment and added a rsquoSummaryof findingsrsquo table and PRISMA flow chart We have removed a number of sensitivity analyses because they were not appropriate giventhe current information on the impact of formulation on efficacy The sensitivity analyses have been superseded by the rsquoRisk of biasrsquoassessment and taken into account in the rsquoSummary of findingsrsquo tables
An earlier review in 2004 chose to exclude studies using benzydamine on the grounds that it was no longer considered to be an NSAID(Mason 2004a) Although the protocol for this review stated that we would not include benzydamine after further consultation wenow believe that it should be classified as an NSAID albeit with a different mode of action which is not fully understood (Quane1998) Thus we have reinstated studies using topical benzydamine
I N D E X T E R M S
Medical Subject Headings (MeSH)
Acute Disease Administration Topical Anti-Inflammatory Agents Non-Steroidal [lowastadministration amp dosage adverse effects] AthleticInjuries [drug therapy] Pain [lowastdrug therapy] Randomized Controlled Trials as Topic Sprains and Strains [drug therapy]
146Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
MeSH check words
Adult Humans
147Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Main results
For this update we added 14 new included studies (3489 participants) and excluded four studies We also identified 20 additionalreports of completed or ongoing studies that have not been published in full The earlier review included 47 studies
This update included 61 studies Most compared topical NSAIDs in the form of a gel spray or cream with a similar topical placebo5311 participants were treated with a topical NSAID 3470 with placebo and 220 with an oral NSAID This was a 63 increase inthe number of included participants over the previous version of this review We also identified a number of studies in clinical trialregistries with unavailable results amounting to about 5900 participants for efficacy and 5300 for adverse events
Formulations of topical diclofenac ibuprofen ketoprofen piroxicam and indomethacin demonstrated significantly higher rates ofclinical success (more participants with at least 50 pain relief ) than matching topical placebo (moderate or high quality data)Benzydamine did not Three drug and formulation combinations had NNTs for clinical success below 4 For diclofenac the Emulgelregformulation had the lowest NNT of 18 (95 CI 15 to 21) in two studies using at least 50 pain intensity reduction as the outcomeDiclofenac plasters other than Flectorreg also had a low NNT of 32 (26 to 42) based on good or excellent responses in some studiesKetoprofen gel had an NNT of 25 (20 to 34) from five studies in the 1980s some with less well defined outcomes Ibuprofen gelhad an NNT of 39 (27 to 67) from two studies with outcomes of marked improvement or complete remission All other drug andformulation combinations had NNT values above 4 indicating lesser efficacy
There were insufficient data to compare reliably individual topical NSAIDs with each other or the same oral NSAID
Local skin reactions were generally mild and transient and did not differ from placebo (high quality data) There were very few systemicadverse events (high quality data) or withdrawals due to adverse events (low quality data)
Authorsrsquo conclusions
Topical NSAIDs provided good levels of pain relief in acute conditions such as sprains strains and overuse injuries probably similar tothat provided by oral NSAIDs Gel formulations of diclofenac (as Emugelreg) ibuprofen and ketoprofen and some diclofenac patchesprovided the best effects Adverse events were usually minimal
Since the last version of this review the new included studies have provided additional information In particular information ontopical diclofenac is greatly expanded The present review supports the previous review in concluding that topical NSAIDs are effectivein providing pain relief and goes further to demonstrate that certain formulations mainly gel formulations of diclofenac ibuprofenand ketoprofen provide the best results Large amounts of unpublished data have been identified and this could influence results inupdates of this review
P L A I N L A N G U A G E S U M M A R Y
Topical non-steroidal anti-inflammatory drugs for acute musculoskeletal pain in adults
Acute musculoskeletal pain describes conditions like a sprained ankle or a muscle pull These usually get better over two or three weekswithout treatment but can be very painful while they last
Topical non-steroidal anti-inflammatory drugs (NSAIDs) are applied to unbroken skin where it hurts as gels creams sprays or plastersTopical NSAIDs penetrate the skin enter tissues or joints and reduce processes causing pain in the tissue Drug levels in the bloodwith topical NSAIDs are very much lower than with the same drug taken by mouth This minimises the risk of harmful effects
We searched medical databases for clinical trials comparing topical NSAIDs with placebo (creams or gels that do not contain a medicine)or other medicines in adults aged 16 years or older with musculoskeletal pain (typically sports injuries) The evidence is current toFebruary 2015
This review is an update of rsquoTopical NSAIDs for acute pain in adultsrsquo originally published in Issue 6 2010 We identified 14 new studiesto add to the 47 studies included in the earlier review We also identified 14 studies in a clinical trial registry that are completed andthree short reports from meetings for which we could not find full details (about 4500 participants) Three more studies are ongoing(almost 900 participants)
The 61 included studies involving 8386 participants were generally of high-quality They tested a number of different topical drugsmostly against a topical placebo (carrier without the NSAID) with application at least once a day We were interested in participants
2Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
having good pain reduction (by about half ) around seven days after treatment started At later times most people are expected to getbetter even without treatment
We looked at particular formulations of individual drugs Gel formulations of diclofenac and ketoprofen were among the most effectivealong with ibuprofen gel and diclofenac plaster For diclofenac and ketoprofen gels 7 or 8 people out of 10 with a painful strain sprainor muscle pull had much reduced pain after seven days compared with only 2 or 3 out of 10 with placebo (high quality data) OtherNSAIDs and formulations were better than placebo but not by as much Because both topical NSAIDs and topical placebo are rubbedinto the skin in these studies we know that any effect is not just from rubbing
About 1 in 20 people experienced a mild and short-lived side effect like redness at the application site This was the same for topicalNSAID and topical placebo (high quality data) Side effects like a stomach upset or feeling sick were uncommon with no differencebetween topical NSAID and topical placebo (high quality data) There were no serious side effects
3Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
S U M M A R Y O F F I N D I N G S F O R T H E M A I N C O M P A R I S O N [Explanation]
Topical NSAIDs compared with topical placebo for acute musculoskeletal pain in adults
Patient or population adults with strains sprains or muscle pullSettings communityIntervention topical NSAID (topical diclofenac ibuprofen and ketoprofen gels only shown here for efficacy)Comparison topical placebo
Outcomes Probable outcome withintervention
Probable outcome withcomparator
RR NNT NNTp or NNH(95 CI)
No of studies partici-pants
Quality of the evidence(GRADE)
Comments
Topical diclofenac gel
(as Emulgel)
Clinical success (eg50 reduction in pain)
780 in 1000 200 in 1000 RR34 (27 to 55)NNT18 (15 to 21)
2 studies314 participants
High Consistent results in 2moderately sized recentstudies of high quality
Topical ibuprofen gel
Clinical success (eg
50 reduction in pain)
420 in 1000 160 in 1000 RR27 (17 to 42)NNT39 (27 to 67)
2 studies241 participants
Moderate Modest effect size andnumbers of participants
Topical ketoprofen gel
Clinical success (eg
50 reduction in pain)
720 in 1000 330 in 1000 RR22 (17 to 28)NNT25 (20 to 34)
5 studies348 participants
Moderate Modest effect size andnumbers of participantsbut studies small withnone recent
All topical NSAIDs
Local adverse events
46 in 1000 50 in 1000 RR10 (080 to 12)NNH not calculated
42 studies6125 participants
High Large number of studiesand participants with con-sistent results
All topical NSAIDs
Systemic adverse
events
32 in 1000 35 in 1000 RR10 (07 to 13)NNH not calculated
38 studies5372 participants
High Large number of studiesand participants with con-sistent results
All topical NSAIDs
Withdrawals - adverse
events
11 in 1000 11 in 1000 RR10 (07 to 17)NNH not calculated
42 studies5790 participants
High Large number of studiesand participants with con-sistent results4
To
pica
lN
SA
IDs
for
acu
tem
uscu
losk
ele
talp
ain
inad
ults
(Revie
w)
Co
pyrig
ht
copy2015
Th
eC
och
ran
eC
olla
bo
ratio
nP
ub
lished
by
Joh
nW
iley
ampS
on
sL
td
Serious adverse events 1 in total 0 in total Not calculated All data Low Small numbers of events
GRADE Working Group grades of evidenceHigh quality Further research is very unlikely to change our confidence in the estimate of effectModerate quality Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimateLow quality Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimateVery low quality We are very uncertain about the estimate
CI confidence interval RR risk ratio NNT number needed to treat for an additional beneficial outcome NNTp number needed to treatto prevent an event happening NNH number needed to treat for an additional harmful outcome
xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
5To
pica
lN
SA
IDs
for
acu
tem
uscu
losk
ele
talp
ain
inad
ults
(Revie
w)
Co
pyrig
ht
copy2015
Th
eC
och
ran
eC
olla
bo
ratio
nP
ub
lished
by
Joh
nW
iley
ampS
on
sL
td
B A C K G R O U N D
This review is an update of a review originally published in Issue 62010 on rsquoTopical NSAIDs for acute pain in adultsrsquo (Massey 2010)We have changed the title to specify musculoskeletal pain becausetopical NSAIDs are not normally used to treat visceral pain orheadache We felt that the new title better reflected the content ofthe review
The use of topical NSAIDs for pain relief has been a controversialsubject in analgesic practice In some parts of the world (includingmuch of Western Europe) they have been available for many yearsare widely available without prescription widely advertised usedextensively and evidence for their use is considered adequate Inother parts of the world they were regarded as little more thanplacebo with any apparent effect attributed to the process of rub-bing at the site of the affected area In some places (for examplethe US) their use was almost unknown until the mid-2010s InEngland 52 million prescriptions for topical NSAIDs were dis-pensed in 2013 (PACT 2014) mainly for formulations of ibupro-fen (245 million) piroxicam (118 million) and diclofenac (127million)
There is good evidence for the efficacy of topical NSAIDs in acuteand chronic musculoskeletal pain (Mason 2004a Mason 2004bMoore 1998a) In the US the Food and Drug Administration li-censed topical nonsteroidal products in 2007 and in England theNational Institute for Health and Care Excellence (NICE) recom-mended topical therapies as first line treatment in its guidelinesfor osteoarthritis in 2008 (NICE 2008) Earlier reviews of topicalanalgesics covered studies investigating the underlying science toexplain biological plausibility in addition to clinical trials (Anon2005 Moore 2008a)
This review is one of a series on topical analgesics including topicalcapsaicin at low and high doses (Derry 2012a Derry 2013) andtopical NSAIDs in chronic pain conditions (Derry 2012b) andsalicylate-containing rubefacients (Derry 2014)
Description of the condition
Acute pain is usually defined as pain of less than three monthsrsquo du-ration It is often associated with injury including trauma surgerymusculoskeletal injuries such as strains sprains and over-use in-juries or soft tissue injuries such as muscle soreness or cramps
Description of the intervention
Clinicians prescribe NSAIDs on a routine basis for a range of mildto moderate pain NSAIDs are the most commonly prescribedanalgesic medications worldwide and their efficacy for treatingacute pain has been well demonstrated (Moore 2003) They re-versibly inhibit cyclooxygenase (prostaglandin endoperoxide syn-thase) the enzyme mediating production of prostaglandins and
thromboxane A2 (FitzGerald 2001) Prostaglandins mediate a va-riety of physiological functions such as maintenance of the gastricmucosal barrier regulation of renal blood flow and regulation ofendothelial tone They also play an important role in inflamma-tory and nociceptive processes However relatively little is knownabout the mechanism of action of this class of compounds asidefrom their ability to inhibit cyclooxygenase-dependent prostanoidformation (Hawkey 1999)NSAIDs taken orally or intravenously are transported to all partsof the body in the blood and relatively high blood concentrationsare needed to achieve effective tissue concentrations at the site ofthe pain and inflammation These high concentrations throughoutthe body can give rise to a number of adverse events that canbe unpleasant (for example dyspepsia) or potentially serious (forexample gastrointestinal bleeding)A topical medication is a one applied to body surfaces such asthe skin or mucous membranes to treat ailments A large rangeof types of topical formulation may be used including but notlimited to creams foams gels lotions ointments and plastersThe exact formulation of a topical medication is often determinedby the speed of drug absorption required The need may be forslow absorption into the circulation to maintain low drug concen-trations and perhaps avoiding extensive first pass metabolism inthe liver plasters containing drug reservoirs may be used for thisas with transdermal opioids or contraceptive steroids For rapidabsorption the formulation is enhanced by substances to improveor assist skin penetration perhaps only to generate high concen-trations in tissues rather than in the blood gel formulations areuseful for this purpose which is why they are sometimes used fortopical NSAIDs
Topical NSAIDs
Topical NSAIDs are formulated for direct application to thepainful site and to produce a local pain-relieving effect whileavoiding body-wide distribution of the drug at physiologically ac-tive levels (McPherson 2013) This method of application (dos-ing) necessarily limits their use to more superficial painful condi-tions such as sprains strains and muscle or tendon soreness Theywould not for example be indicated for deep visceral pain orheadaches They are also not appropriate for use on broken skinso would not be used on open wounds (accidental or surgical)
How the intervention might work
For a topical formulation to be effective it must first penetratethe skin Only when the drug has entered the lower layers ofthe skin can it be absorbed by the blood and transported to thesite of action or penetrate deeper into areas where inflammationoccurs Individual drugs have different degrees of penetration Abalance between lipid and aqueous solubility is needed to optimisepenetration and use of prodrug esters has been suggested as a way
6Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
of enhancing permeability Formulation is also crucial to goodskin penetration Experiments with artificial membranes or humanepidermis suggest that creams are generally less effective than gelsor sprays but newer formulations such as microemulsions mayhave greater potentialOnce the drug has reached the site of action it must be presentat a sufficiently high concentration to inhibit cyclooxygenase en-zymes thereby reducing prostaglandin synthesis This in turn re-duces inflammation and relieves pain It is probable that in acuteconditions topical NSAIDs exert their action primarily by lo-cal reduction of symptoms independent of any systemic uptakeand delivery Tissue levels of NSAIDs applied topically certainlyreach levels high enough to inhibit cyclooxygenase-2 (Anon 2005Haroutiunian 2010 Moore 2008a) However plasma concentra-tions found after topical administration are only a fraction (usu-ally much less than 5) of the levels found in plasma followingoral administration Topical application can potentially limit sys-temic adverse events by increasing local effects and minimisingsystemic concentrations of the drug We know that the incidenceof upper gastrointestinal bleeding is low with chronic use of topi-cal NSAIDs (Evans 1995) although it has been reported partic-ularly in people with risk factors (Zimmerman 1995) We haveno certain knowledge of lower effects on cardiovascular events orrenal failure both of which have been associated with oral NSAIDuse
Why it is important to do this review
Since the last review in 2010 a number of new studies have beenpublished nearly all of which investigated various formulations ofdiclofenac These new studies are generally of higher quality thanmany of the earlier ones in this review and have the potential toinfluence the strength of its conclusions substantially Moreoverthe additional information allows for analysis based not only on aparticular drug but also on the formulation of that drug This canprovide better insight into whether formulation affects efficacy oftopical NSAIDs in acute musculoskeletal painAn updated review of evidence for topical NSAIDs was needed toinform choices made by consumers prescribers and commission-ers (purchasers of healthcare)
O B J E C T I V E S
To determine the efficacy and safety of topically applied NSAIDsin acute musculoskeletal pain in adults
M E T H O D S
Criteria for considering studies for this review
Types of studies
Randomised controlled double-blind studies comparing topicalNSAIDs with placebo (inert carrier) or other active treatment foracute pain with at least 10 participants per treatment arm andoutcomes close to seven days (minimum three days) We excludedstudies published only as short abstracts (which report insufficientdata to assess methods) or studying experimentally induced pain(which does not correlate well with clinical pain) Because a cross-over design is not appropriate for self limiting conditions such assprains strains and contusions we only considered parallel-groupdesigns
Types of participants
Adults (aged 16 years or more) with acute musculoskeletal pain ofat least moderate intensity resulting mainly from strains sprainsor sports injuries Typically for sports injuries the injury wouldhave occurred within 24 or 48 hours
Types of interventions
Included studies had at least one treatment arm using a topicalNSAID and a comparator arm using placebo (inert carrier withoutNSAID or other active treatment) The topical NSAID had to beapplied at least once daily We did not include salicylates in thisreview as they are no longer classified as topical NSAIDs and arecovered in a separate review (Derry 2014)
Types of outcome measures
We sought information on participant characteristics includingage sex and condition treated
Primary outcomes
The primary outcome was rsquoclinical successrsquo defined as at least a50 reduction in pain or equivalent measure such as a rsquovery goodrsquoor rsquoexcellentrsquo global assessment of treatment or rsquononersquo or rsquoslightrsquopain on rest or movement measured on a categorical scale (Moore1998a) We used the following hierarchy of outcomes to extractdata for the primary outcome
bull Participant reported reduction in pain of at least 50bull Participant reported global assessment of treatmentbull Pain on movementbull Pain at rest or spontaneous painbull Undefined rsquoimprovementrsquo
We used only participant reported outcomes of efficacy and notphysician or investigator reported outcomes
7Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Secondary outcomes
bull Numbers of participants with adverse events local andsystemic
bull Numbers of withdrawals all cause lack of efficacy andadverse events
We anticipated that outcomes would be reported after differentdurations of treatment and extracted data reported as close toseven days as possible with a minimum of three days We also ex-tracted data for outcomes reported after longer durations of treat-ment We anticipated that reporting of adverse events would varybetween studies with regard to the terminology used method ofascertainment and categories reported (for example occurring inat least 5 of participants or where there is a statistically signifi-cant difference between treatment groups) We took care to iden-tify these details where relevant
Search methods for identification of studies
Electronic searches
We searched the following databases without language restrictionbull CENTRAL (The Cochrane Library) Issue 4 2009 for the
original review and the Cochrane Register of Studies Online(CRSO) to 3 February 2015 for this update
bull MEDLINE (via Ovid) from inception to December 2009for the original review and from 2008 to 3 February 2015 forthis update
bull EMBASE (via Ovid) from inception to December 2009for the original review and from 2008 to 3 February 2015 forthis update
bull Oxford Pain Relief Database for the original review (Jadad1996a) This resource is no longer being updated
See Appendix 1 for the CENTRAL search strategy Appendix 2 forthe MEDLINE search strategy and Appendix 3 for the EMBASEsearch strategy
Searching other resources
We searched the reference lists of review articles and includedstudies We have previously asked manufacturers for details ofunpublished studies but did not make new requestsWe searched two clinical trial registries (clinicaltrialsgov (clinicaltrialsgov) and the World Health Organization In-ternational Clinical Trials Registry Platform (appswhointtrialsearch)) and asked personal contacts about ongoing and un-published studies
Data collection and analysis
Selection of studies
We screened the titles and abstracts of studies identified by thesearches to eliminate those that clearly did not satisfy the inclu-sion criteria and obtained full reports of the remaining studies todetermine inclusion in the review
Data extraction and management
Review authors were not blinded to the authorsrsquo names and in-stitutions journal of publication or study results at any stage ofthe review Two review authors independently selected the stud-ies for inclusion assessed methodological quality and risk of biasand extracted data We resolved disagreements and uncertaintiesthrough discussionWe abstracted information on participants interventions and out-comes from the original reports into a standard data extractionform One review author entered data suitable for meta-analysisinto Review Manager 5 (RevMan 2014) and another review au-thor checked it
Assessment of risk of bias in included studies
We used the Oxford Quality Score as the basis for inclusion lim-iting inclusion to studies that were randomised and double-blindas a minimum (Jadad 1996b)Two review authors independently assessed risk of bias for eachstudy using the criteria outlined in the Cochrane Handbook forSystematic Reviews of Interventions with any disagreements resolvedby discussion (Higgins 2011) We assessed the following for eachstudy
bull Random sequence generation (checking for possibleselection bias) We assessed the method used to generate theallocation sequence as low risk of bias (any truly randomprocess for example random number table computer randomnumber generator) unclear risk of bias (method used to generatesequence was not clearly stated) We excluded studies using anon-random process that were therefore at high risk of bias (forexample odd or even date of birth hospital or clinic recordnumber)
bull Allocation concealment (checking for possible selectionbias) The method used to conceal allocation to interventionsbefore assignment determines whether intervention allocationcould have been foreseen in advance of or during recruitmentor changed after assignment We assessed the methods as lowrisk of bias (for example telephone or central randomisationconsecutively numbered sealed opaque envelopes) unclear riskof bias (method was not clearly stated) We excluded studies thatdid not conceal allocation and were therefore at high risk of bias(for example open list)
8Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
bull Blinding of outcome assessment (checking for possibledetection bias) We assessed the methods used to blind studyparticipants and outcome assessors from knowledge of whichintervention a participant received We assessed the methods aslow risk of bias (study stated that it was blinded and describedthe method used to achieve blinding for example identicaltubes containing gel or identical plasters matched in appearanceand smell) unclear risk of bias (study stated that it was blindedbut did not provide an adequate description of how blinding wasachieved) We excluded studies that were not double-blind andwere therefore at high risk of bias
bull Incomplete outcome data (checking for possible attritionbias due to the amount nature and handling of incompleteoutcome data) We assessed the methods used to deal withincomplete data as low risk of bias (less than 10 ofparticipants did not complete the study or used rsquobaselineobservation carried forwardrsquo analysis or both) unclear risk ofbias (used rsquolast observation carried forwardrsquo analysis) or high riskof bias (used rsquocompleterrsquo analysis)
bull Size (checking for possible biases confounded by small size)Small studies have been shown to overestimate treatment effectsprobably due to methodological weaknesses (Dechartres 2013Nuumlesch 2010) We assessed studies as at low risk of bias if theyhad at least 200 participants at unclear risk if they had 50 to 200participants and at high risk if they had fewer than 50participants
Measures of treatment effect
We used risk ratio (RR) to establish statistical difference and num-bers needed to treat for an additional beneficial outcome (NNT)with 95 confidence intervals (CI) We pooled percentages as ab-solute measures of benefit or harmWe used the following terms to describe adverse outcomes in termsof harm or prevention of harm
bull When significantly fewer adverse outcomes occurred withtreatment than with control (placebo or active) we used the termthenumber needed to treat to prevent one event (NNTp)
bull When significantly more adverse outcomes occurred withtreatment compared with control (placebo or active) we usedthe term the number needed to treat for an additional harmfuloutcome or cause one event (NNH)
Unit of analysis issues
Randomisation was to the individual participant
Dealing with missing data
Wherever possible we used intention-to-treat (ITT) analysis wherethe ITT population consists of participants who were randomisedapplied at least one dose of the assigned study medication andprovided at least one post-baseline assessment We assigned miss-ing participants zero improvement
We also looked for information about methods of imputation formissing data
Assessment of heterogeneity
We examined heterogeneity visually using LrsquoAbbeacute plots (LrsquoAbbeacute1987) a visual method for assessing differences in results of indi-vidual studies and with the I2 statistic
Assessment of reporting biases
The aim of this review was to use dichotomous outcomes of knownutility and of value to patients (Moore 2013) The review did notdepend on what the authors of the original studies chose to reportor not Studies that did not report dichotomous results but onlyaverage pain data did not contribute to analysesWe assessed publication bias using a method designed to detect theamount of unpublished data with a null effect required to makeany result clinically irrelevant (usually taken to mean an NNT of10 or higher Moore 2008b)
Data synthesis
We pooled data only for comparisons and outcomes where therewere at least two studies and 200 participants (Moore 1998b)When two active treatment arms were compared with a placeboarm we took care to avoid double counting of participants in theplacebo arm if both active groups contributed to an analysis wesplit the placebo group between themWe calculated RRs with 95 CIs using the fixed-effect model(Morris 1995) A statistically significant benefit of topical NSAIDover control was assumed when the lower limit of the 95 CI ofthe RR was greater than one A statistically significant benefit ofcontrol over active treatment was assumed when the upper limitof the 95 CI was less than one We calculated NNTs with 95CIs using the pooled number of events by the method of Cookand Sackett (Cook 1995)Statistically significant differences between NNTs for differenttopical NSAIDs were tested using the z test (Tramer 1997) wherethere were sufficient data to do so and where the studies weresufficiently similar in types of participant outcome and durationto make such comparisons sensible
Subgroup analysis and investigation of heterogeneity
We carried out separate analyses for individual NSAIDs andwhere the data permitted for different formulations of individualNSAIDs
Sensitivity analysis
The earlier review included sensitivity analyses for various factorsthat are now covered by the assessment of risk of bias
9Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
R E S U L T S
Description of studies
Results of the search
New searches for this update identified 20 publications that wereexamined in further detail to determine inclusion status We alsoidentified 20 additional studies in clinical trial registries See Figure1
10Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Figure 1 Study flow diagram
11Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
We identified 14 new studies (11 publications three unpub-lished reports) satisfying our inclusion criteria (Costantino2011 Coudreuse 2010 Gonzaacutelez de Vega 2013 Hofman2000 Klainguti 2010 Kuehl 2011 Li 2013 NCT01255423NCT01272934 NCT01272947 Predel 2012 Predel 2013aPredel 2013b Saillant 1998) one post hoc analysis of a study thatwas included in the earlier review (Mueller 2010 in Predel 2004)and an additional publication and a pooled analysis that includednew data for another study from the earlier review (Lionberger2011 pooled analysis in Joussellin 2003)
Included studies
All except one of the new studies compared diclofenac withplacebo A number of different formulations were used includ-ing diclofenac epolamine (DHEP) with or without heparin (Flec-toparin Tissugel or Flector EP Tissugel) applied as a plaster (orpatch) diclofenac diethylamine (DDEA) applied as a gel and di-clofenac with lethicin applied as a spray gel The remaining studycompared diclofenac gel with traumeel a ldquofixed combination ofplant and mineral extractsrdquo applied as a gel or an ointmentThere were 47 studies in the original review 14 new studies wereincluded making a total of 61 studies in this updated reviewAll used a parallel group design Forty-four compared a topicalNSAID with placebo 13 a topical NSAID with an active com-parator (a different topical NSAID an oral NSAID the same top-ical NSAID in a different formulation or a compound of plantand mineral extracts) and four had both placebo and active com-parators In total 5311 participants were treated with a topicalNSAID 3470 with placebo and 220 with an oral NSAID Top-ical NSAIDs used were benzydamine diclofenac etofenamatefelbinac fentiazac flunoxaprophen flurbiprofen ibuprofen in-domethacin ketoprofen ketorolac lysine clonixinate meclofe-namic acid naproxen niflumic acid and piroxicam They wereapplied as creams gels sprays foams or plasters (patches) Topicalplacebos were the inert carriers without the active NSAID OralNSAIDs used were ibuprofen (as tablets) and indomethacin (ascapsules)Most studies enrolled participants who had sprains strains andcontusions usually as a result of sports injuries and treatmentwas started within a few hours or days Other studies enrolledparticipants with overuse-type injuries such as tendinitis and acutelow back pain where pain had been present for days or weeks butless than three monthsParticipants were treated for at least five days and up to threeweeks with most studies lasting seven to 14 days Participantswere usually assessed in clinic at intervals during treatment andsometimes also at home using daily patient diaries We used out-comes closest to seven days because many of these injuries are self
limiting with differences between active treatment and placebobeing diminished or lost after longer intervalsMost studies reported dichotomous outcomes suitable for a re-sponder analysis although group mean change (for pain or phys-ical function for example) was usually the primary outcomesHowever the definition of response varied both in the parametermeasured (for example pain pain on movement patient globalevaluation of treatment) and in the scale used to measure it (forexample a 3- 4- or 5-point scale for patient global evaluation)Details of included studies are in the Characteristics of includedstudies tableWe identified 14 completed but apparently unpublished studies ina clinical trial registry for which no results have been posted (4403participants NCT00351104 NCT00352625 NCT00426985NCT00640705 NCT00640939 NCT00680472NCT00680784 NCT00765700 NCT00869063NCT00869180 NCT00931866 NCT01874626NCT01957215 NCT02324270) We have placed these underCharacteristics of studies awaiting classification We also identi-fied three conference abstracts that relate to completed studies thatdo not appear to have been published but that may satisfy ourinclusion criteria One is likely to be the same study as one of theincluded studies identified in a clinical trial registry (Pallay 2013in NCT01272947) one relates to another study identified in theclinical trial registry that is awaiting classification (Ekman 2010in NCT00765700) while we could find no published or unpub-lished reports of the other Sarzi-Puttini 2014)We also identified three ongoing studies with an estimated en-rolment of 880 participants (NCT01945034 NCT02100670NCT02290821) Details are in the Characteristics of ongoingstudies table
Excluded studies
For the original review 25 studies were excluded after readingthe full paper For this update we excluded four new studies (Cesarone 2008 Coulibaly 2009 Kuwabara 2013 Vinciguerra2008) Details are in the Characteristics of excluded studies table
Risk of bias in included studies
All studies were randomised and double-blind One study scored25 (Sinniger 1981) 23 scored 35 23 scored 45 and 14 scored55 for methodological quality using the Oxford Quality Scale Abreakdown of the scores for individual studies is reported in theCharacteristics of included studies tableComments on potential biases in individual studies are reported inthe rsquoRisk of biasrsquo section of the Characteristics of included studiestable The findings are displayed in Figure 2 and Figure 3
12Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Figure 2 Risk of bias summary review authorsrsquo judgements about each risk of bias item for each includedstudy
13Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Figure 3 Methodological quality graph review authorsrsquo judgements about each methodological quality
item presented as percentages across all included studies
Allocation
All the studies were randomised but only 17 adequately describedthe method used to generate the random sequence Thirteen stud-ies adequately described the method used to conceal allocation ofthe sequence No studies were at high risk of bias for this item
Blinding
All studies were double-blind and 38 adequately described themethod used to maintain the blinding No studies were at highrisk of bias for this item
Incomplete outcome data
Thirty-six studies included all participants in the primary analysisor provided sufficient data to allow missing participants to be in-cluded as non-responders and were judged at low risk of bias Wejudged four studies to be at high risk of attrition bias (Campbell1994 Kuehl 2011 Maziegraveres 2005a Russell 1991) Three unpub-lished studies contributed only to adverse event analyses and ac-counted for all participants for these outcomes (NCT01255423NCT01272934 NCT01272947)
Other potential sources of bias
We judged one study that included more than 200 participantsin each treatment arm to be at low risk of bias from size butthis study had a very high attrition rate (see rsquoIncomplete outcome
data (attrition bias)rsquo) and did not report all the efficacy outcomesmeasured (Kuehl 2011) We judged 27 studies to be at high riskbecause they included fewer than 50 participants per treatmentarm
Effects of interventions
See Summary of findings for the main comparison TopicalNSAIDs compared with topical placebo for acute musculoskeletalpain in adultsThree studies did not contribute data suitable for analysis of at leastone outcome (Gonzaacutelez de Vega 2013 Gualdi 1987 Hoffmann2012)
1 Topical NSAID versus placebo
Details of efficacy outcomes in individual studies are in Appendix4 and of adverse events and withdrawals in Appendix 5 Appendix6 has details of the concentration of topical products the amountapplied the frequency of application and an estimation of thedaily dose of topical NSAID applied Not all studies providedsufficient information to allow calculation of daily dose appliedFor example for topical diclofenac the estimated doses appliedvaried between about 60 and 280 mg for topical ketoprofen 100to about 450 mg for topical ibuprofen 300 to 800 mg
Participants with clinical success
14Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Topical diclofenac versus placebo
Ten studies contributed to this analysis (Coudreuse 2010Joussellin 2003 Klainguti 2010 Li 2013 Predel 2004 Predel2012 Predel 2013a Predel 2013b Rowbotham 2003) of whichone (Predel 2012) had two active treatment arms A total of 1074participants were treated with topical diclofenac and 976 withplacebo (Analysis 11 Figure 4)
15Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Figure 4 Forest plot of comparison 2 Individual NSAID versus placebo outcome 21 Clinical success
16Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
bull The proportion of participants experiencing successfultreatment with topical diclofenac was 74 (8001074 range39 to 100)
bull The proportion of participants experiencing successfultreatment with placebo was 47 (461976 range 8 to 82)
bull The RR for treatment compared with placebo was 16(95 CI 15 to 17)
bull The NNT for successful treatment was 37 (32 to 43) Forevery four participants treated with topical diclofenac one wouldexperience successful treatment who would not have done sowith placebo
Effect of formulation
The effects of formulation are shown in Analysis 21 and Figure5
17Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Figure 5 LrsquoAbbeacute plot of clinical success in studies of topical diclofenac versus topical placebo The size ofthe symbol is proportional to the size of the study (inset scale) Dark blue Emulgel light blue spraygel red
Flector pink other patch or plaster
bull Four studies used a Flectorreg plaster (1030 participantsJoussellin 2003 Li 2013 Rowbotham 2003 Saillant 1998) TheRR for treatment compared with placebo was 15 (14 to 17)and the NNT was 47 (37 to 65)
bull Three studies used other makes of plaster (474 participantsCoudreuse 2010 Klainguti 2010 Predel 2004) The RR fortreatment compared with placebo was 16 (14 to 18) and theNNT was 32 (26 to 42)
bull Two studies used Voltaren Emulgel (314 participants Predel2012 Predel 2013b) The RR for treatment compared withplacebo was 38 (27 to 55) and the NNT was 18 (15 to 21)
bull One study used a spray gel (232 participants Predel2013a) The RR for treatment compared with placebo was 12(105 to 13) and the NNT was 80 (48 to 24)
Diclofenac as the gel formulation Emulgel was statistically more ef-ficacious than the plaster formulation as Flector plaster (z = 6360P value lt 000001)
Topical ibuprofen versus placebo
Five studies contributed to this analysis (Billigmann 1996Campbell 1994 Dreiser 1988 Machen 2002 Ramesh 1983) Atotal of 218 participants were treated with topical ibuprofen and218 with placebo (Analysis 11)
bull The proportion of participants experiencing successfultreatment with topical ibuprofen was 55 (120218 range 31to 81)
18Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
bull The proportion of participants experiencing successfultreatment with placebo was 33 (73218 range 13 to 76)
bull The RR of treatment compared with placebo was 16 (13to 20)
bull The NNT for successful treatment was 46 (33 to 80) Forevery five participants treated with topical ibuprofen one wouldexperience successful treatment who would not have done sowith placebo
Effect of formulation
The effects of formulation are shown in Analysis 31bull Three studies used cream formulations (195 participants
Campbell 1994 Dreiser 1988 Ramesh 1983) Although this isjust below our threshold for pooled analysis we have includedthis analysis for completeness and the results should beinterpreted with caution The RR for treatment compared withplacebo was 13 (103 to 16) and the NNT was 64 (34 to 41)
bull Two studies used gel formulations (241 participantsBilligmann 1996 Machen 2002) The RR for treatmentcompared with placebo was 27 (17 to 42) and the NNT was39 (27 to 67)
There was no statistically significant difference between the geland cream formulations (z = 1160 P value = 0246)
Topical ketoprofen versus placebo
Seven studies contributed to this analysis (Airaksinen 1993Dreiser 1989 Julien 1989 Kockelbergh 1985 Maziegraveres 2005bMaziegraveres 2005a Noret 1987) A total of 346 participants weretreated with topical ketoprofen and 337 with placebo (Analysis11)
bull The proportion of participants experiencing successfultreatment with topical ketoprofen was 73 (251346 range57 to 89)
bull The proportion of participants experiencing successfultreatment with placebo was 47 (157337 range 17 to 73)
bull The RR of treatment compared with placebo was 16 (14to 18)
bull The NNT for successful treatment was 39 (30 to 53) Forevery four participants treated with topical ketoprofen onewould experience successful treatment who would not have doneso with placebo
Effect of formulation
The effects of formulation are shown in Analysis 41 and Figure6
19Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Figure 6 LrsquoAbbeacute plot of clinical success in studies of topical ketoprofen versus topical placebo The size ofthe symbol is proportional to the size of the study (inset scale) Light blue ketoprofen gel pink ketoprofen
plaster
Two studies used a plaster formulation (335 participants Maziegraveres2005b Maziegraveres 2005a) The RR for treatment compared withplacebo was 12 (104 to 14) and the NNT was 82 (45 to 47)Five studies used gel formulations (348 participants Airaksinen1993 Dreiser 1989 Julien 1989 Kockelbergh 1985 Noret 1987)The RR for treatment compared with placebo was 22 (17 to 28)and the NNT was 25 (20 to 34)Ketoprofen as a gel formulation was statistically more efficaciousthan a plaster formulation (z = 3860 P value = 000014)
Topical piroxicam versus placebo
Three studies contributed to this analysis (Aoki 1984 Fujimaki1985 Russell 1991) A total of 255 participants were treated withtopical piroxicam and 249 with placebo (Analysis 11)
bull The proportion of participants experiencing successfultreatment with topical piroxicam was 68 (179255 range 53to 79)
bull The proportion of participants experiencing successfultreatment with placebo was 47 (118249 range 45 to 49)
bull The RR of treatment compared with placebo was 15 (13to 17)
bull The NNT for successful treatment was 44 (32 to 69) Forevery four participants treated with topical piroxicam one wouldexperience successful treatment who would not have done sowith placebo
Topical indomethacin versus placebo
20Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Three studies contributed to this analysis ( kermark 1990 Aoki1984 Fujimaki 1985) A total of 168 participants were treatedwith topical indomethacin and 173 with placebo (Analysis 11)
bull The proportion of participants experiencing successfultreatment with topical indomethacin was 58 (97168 range54 to 64)
bull The proportion of participants experiencing successfultreatment with placebo was 46 (79173 range 25 to 49)
bull The RR of treatment compared with placebo was 13 (103to 16)
bull The NNT for successful treatment was 83 (44 to 65) Forevery eight participants treated with topical indomethacin onewould experience successful treatment who would not have doneso with placebo
Topical benzydamine versus placebo
Three studies contributed to this analysis (Chatterjee 1977 Haig1986 Linde 1985) A total of 96 participants were treated withtopical benzydamine and 97 with placebo (Analysis 11) Al-though this is just below our threshold for pooled analysis we haveincluded this analysis for completeness and the results should beinterpreted with caution
bull The proportion of participants experiencing successfultreatment with topical benzydamine was 77 (7496 range70 to 86)
bull The proportion of participants experiencing successfultreatment with placebo was 67 (6597 range 48 to 80)
bull The RR of treatment compared with placebo was 12 (096to 14) There was no statistically significant difference betweentreatments (Figure 4)
Results for participants with clinical success with individual topi-cal NSAIDs where there were adequate data for analysis are sum-marised below in rsquoSummary of results Arsquo and Analysis 11 Analysis21 Analysis 31 and Analysis 41
Summary of results A Participants with clinical success
Comparison Studies Participants NSAID()
Placebo () Relative benefit (95CI)
NNT (95 CI)
Diclofenac -Flector plaster
4 1030 63 41 15 (14 to 17) 47 (37 to 65)
Diclofenac -other plaster
3 474 88 57 16 (14 to 18) 32 (26 to 42)
Diclofenac -Emulgel
2 314 78 20 38 (27 to 55) 18 (15 to 21)
Diclofenac -other gel
1 232 94 82 12 (11 to 13) 80 (48 to 24)
Ibuprofen -cream
3 195 71 56 13 (103 to 16) 64 (34 to 41)
Ibuprofen - gel 2 241 42 16 27 (17 to 42) 39 (27 to 67)
Ketoprofen -plaster
2 335 73 60 12 (104 to 14) 82 (45 to 47)
Ketoprofen - gel 5 348 72 33 22 (17 to 28) 25 (20 to 34)
Piroxicam 3 504 70 47 15 (13 to 17) 44 (32 to 69)
Indomethacin 3 341 58 46 13 (103 to 16) 83 (44 to 65)
21Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Benzydamine 3 193 77 67 12 (096 to 14) not calculated
Results for these two comparisons are derived from very smallamounts of data and are provided here for completeness Theyshould be interpreted with caution
Local adverse events
Local adverse events were irritation of the area to which the topicalNSAID was applied including redness or erythema and itch orpruritus Where reported these were usually described as mild andtransient
All topical NSAIDs versus placebo
Forty-two studies contributed to this analysis of which three com-pared two different drugs with placebo (Aoki 1984 Diebshlag1990 Fujimaki 1985) Three studies had two treatment armscomparing different formulations or application regimens for di-clofenac with placebo which have been combined for this analysis
(Costantino 2011 Klainguti 2010 Predel 2012) In total 3619participants were treated with topical NSAIDs and 3121 withplacebo (Analysis 51)
bull The proportion of participants experiencing a local adverseevent with a topical NSAID was 43 (1553619 range 0 to33)
bull The proportion of participants experiencing a local adverseevent with placebo was 46 (1453121 range 0 to 32)
bull The RR of topical NSAID compared with placebo was 098(080 to 12)
bull There was no significant difference between treatmentgroups so the NNH was not calculated
Individual topical NSAIDs versus placebo
Results for local adverse events with individual topical NSAIDswhere there were adequate data for analysis are in Summary ofresults B and Analysis 12
Summary of results B Participants with local adverse events
Comparison Studies Participants NSAID()
Placebo()
RR(95 CI)
NNH(95 CI)
All NSAIDs 42 6740 43 46 098 (080 to 12) Not calculated
Diclofenac 15 3271 31 43 078 (056 to 11) Not calculated
Ketoprofen 8 852 11 95 12 (083 to 17) Not calculated
Piroxicam 3 522 23 54 042 (017 to 11) Not calculated
Felbinac 3 397 30 15 19 (049 to 75) Not calculated
Indomethacin 3 354 63 22 27 (091 to 77) Not calculated
Ibuprofen 3 321 10 43 23 (098 to 54) Not calculated
Systemic adverse events
All topical NSAIDs versus placebo
Thirty-six studies contributed data on systemic adverse eventsof which three compared two different drugs with placebo (Aoki1984 Diebshlag 1990 Fujimaki 1985) Two studies had two treat-
22Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
ment arms comparing different formulations or application regi-mens for diclofenac with placebo which have been combined forthis analysis (Klainguti 2010 Predel 2012) In total 2956 partic-ipants were treated with a topical NSAID and 2620 with placebo(Analysis 52)
bull Twenty-three studies reported no systemic adverse events inany arm of the study
bull The proportion of participants experiencing a systemicadverse event with a topical NSAID was 31 (922956)
bull The proportion of participants experiencing a systemicadverse event with placebo was 35 (912620)
bull The RR of topical NSAID compared with placebo was 096(073 to 13)
bull There was no significant difference between treatmentgroups so the NNH was not calculated
A further six studies did not report the occurrence or oth-erwise of systemic adverse events (Billigmann 1996 Julien1989 Kockelbergh 1985 Noret 1987 Ramesh 1983 Vecchiet1991) while two studies did not report numbers of participantswith systemic adverse events ( kermark 1990 Auclair 1989)Costantino 2011 reported that there were no systemic gastroin-testinal adverse events Two studies reported only on total adverseevents without distinguishing between local and systemic events(NCT01255423 NCT01272934)
Serious adverse events
Two studies reported serious adverse events In Hoffmann 2012one participant experienced three serious adverse events none ofwhich was judged to be related to the study medication (diclofenacplaster) In NCT01272934 one participant using diclofenac gelruptured the ligaments of the wrist There was no statement aboutlikely relationship to the study medication but this seems unlikely
Adverse event withdrawals
Forty-two studies reported data relating to adverse event with-drawals of which three compared two different drugs with placebo(Aoki 1984 Diebshlag 1990 Fujimaki 1985) Two studies hadtwo treatment arms comparing different formulations or applica-tion regimens for diclofenac with placebo which have been com-bined for this analysis (Klainguti 2010 Predel 2012) In total3365 participants received a topical NSAID and 3040 placebo(Analysis 53)
bull Forty-four comparisons reported no adverse eventwithdrawals
bull The proportion of participants withdrawing from the studydue to an adverse event after treatment with a topical NSAIDwas 098 (333365)
bull The proportion of participants withdrawing from the studydue to an adverse event after treatment with placebo was 099(303040)
bull The RR of topical NSAID compared to placebo was 10(064 to 16)
bull There was no significant difference between treatmentgroups so the NNH was not calculated
Four studies did not specifically mention adverse event with-drawals (Haig 1986 Hoffmann 2012 Klainguti 2010 Vecchiet1991) while one reported that one participant withdrew with mildpruritus but did not state the treatment arm (Joussellin 2003)Ten studies specifically reported withdrawals due to lack of efficacy(Dreiser 1989 Dreiser 1994 Kuehl 2011 Machen 2002 Maziegraveres2005b Maziegraveres 2005a Noret 1987 Predel 2013a Russell 1991Thorling 1990) (Appendix 5) Numbers of participants withdraw-ing were generally low with rates of 6 or less except in Kuehl2011 where the rate was 10 with active treatment (diclofenacplaster) and 12 with placebo We did not carry out any analysisbecause the outcome was inconsistently reported
2 Topical NSAID versus active comparator
Details of efficacy outcomes in individual studies are in Appendix4 and of adverse events and withdrawals in Appendix 5
Participants with clinical success
Topical NSAID versus oral NSAID
bull kermark 1990 compared indomethacin spray withindomethacin capsules with response rates of 55 (1222) withspray and 23 (522) with capsules
bull Hosie 1993 compared felbinac foam with ibuprofen tabletswith response rates of 64 (81127) with felbinac foam and72 (96133) with ibuprofen tablets
bull Whitefield 2002 compared ibuprofen gel with ibuprofentablets with response rates of 60 (3050) with gel and 54(3650) with tablets
There were insufficient data for meta-analysis for any one of thesecomparisons felbinac is not known to be better than placebo
Topical NSAID versus different formulation of the same
topical NSAID
bull Fioravanti 1999 compared DHEP (diclofenac) gelformulated with and without lecithin with response rates of70 (3550) in both treatment arms
bull Mahler 2003 compared DHEP (diclofenac) gel formulatedwith and without lecithin with response rates of 89 (8292)with lecithin and 70 (6288) without lecithin
bull Gallacchi 1990 compared topical diclofenac formulated asFlectorreg gel and Emugelreg with response rates of 76 (1925)in both treatment arms
23Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
bull Governali 1995 compared topical ketoprofen cream withgel with response rates of 93 (1415) with cream and 27 (415) with gel
There were insufficient data for analysis
Topical NSAID versus different topical NSAID
Eight studies compared one topical NSAID versus at least oneother topical NSAID piroxicam versus indomethacin (Aoki1984 Fujimaki 1985 Sugioka 1984) ibuprofen versus ketopro-fen (Curioni 1985 Picchio 1981) ketoprofen versus etofena-mate (Curioni 1985 Tonutti 1994) ibuprofen versus etofenamate(Curioni 1985) ketorolac versus etofenamate (Diebshlag 1990)and diclofenac versus lysine clonixinate (Hofman 2000)There were sufficient data to compare only piroxicam with in-domethacin (Aoki 1984 Fujimaki 1985 Sugioka 1984 Analysis61)
bull The proportion of participants experiencing clinical successwith topical piroxicam was 56 (185330 range 49 to 78)
bull The proportion of participants experiencing clinical successwith topical indomethacin was 45 (140311 range 33 to64)
bull The RR of piroxicam compared with indomethacin was 12(11 to 14)
bull The NNT for successful treatment was 91 (53 to 30) Forevery nine participants treated with topical piroxicam onewould experience a clinical success who would not haveexperienced one with topical indomethacin
Topical NSAID versus different topical intervention
One study compared diclofenac gel with a herbal product calledTraumeel gel under double-blind conditions with response ratesfor being pain-free at seven days of 8137 (58) with diclofenacgel and 7140 (50) with Traumeel gel (Gonzaacutelez de Vega 2013)
Local adverse events
Topical NSAID versus oral NSAID
Two studies comparing a topical NSAID with an oral NSAIDprovided data on local adverse events ( kermark 1990 Hosie1993) There were five events with topical NSAID and three withoral NSAID which were too few for analysis
Topical NSAID versus different topical NSAID
All nine studies comparing one topical NSAID with at least oneother reported on local adverse events with a total of 48 events in1005 participants (48) There were sufficient data to compare
only piroxicam with indomethacin (Aoki 1984 Fujimaki 1985Sugioka 1984 Analysis 62)
bull The proportion of participants experiencing local adverseevents with topical piroxicam was 21 (7340 range 12 to28)
bull The proportion of participants experiencing local adverseevents with topical indomethacin was 10 (33331 range 29to 15)
bull The RR of piroxicam compared with indomethacin was021 (009 to 047)
bull The NNT to prevent a local adverse event was 13 (87 to23) For every thirteen participants treated with topicalpiroxicam one would not experience a local adverse event whowould have experienced one with topical indomethacin
Topical NSAID versus different topical intervention
Gonzaacutelez de Vega 2013 reported that adverse events were infre-quent and mild to moderate in intensity but did not distinguishbetween local and systemic events Numbers of participants expe-riencing any adverse event were 8147 with diclofenac gel and 14148 with Traumeel gel
Systemic adverse events
kermark 1990 reported numbers of events rather than numbersof participants with events while Tonutti 1994 and Whitefield2002 reported no adverse events attributable to the study medi-cation and Fioravanti 1999 Gallacchi 1990 Gualdi 1987 andSugioka 1984 did not mention systemic adverse events Gonzaacutelezde Vega 2013 did not distinguish between local and systemicevents In the remaining studies a total of 16 events were reportedin topical NSAID treatment arms (797 participants 2) and 11with ibuprofen tablets (134 participants 8)
Serious adverse events
No serious adverse events were reported in any treatment arm
Withdrawals
The only withdrawals reported due to adverse events were in stud-ies with placebo treatment arms ( kermark 1990 Fujimaki 1985)and have been reviewedThree studies reported withdrawals due to lack of efficacy (Gonzaacutelez de Vega 2013 Hofman 2000 Tonutti 1994) (Appendix5) There were insufficient data for analysisSome studies reported exclusions from analysis (efficacy or safetyor both) following randomisation mainly due to protocol vio-lations or loss to follow-up (Appendix 5) There is no reason tobelieve these exclusions would introduce systematic bias and thenumbers involved were not likely to influence results
24Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
D I S C U S S I O N
This updated review of topical NSAIDs for acute musculoskele-tal pain in adults differs from previous reviews Previously dataallowed only for comparison of individual topical NSAIDs withplacebo irrespective of formulation With a substantial amount ofnew data for diclofenac it is possible to distinguish effects of for-mulation for individual NSAIDs Because formulation chemistrycan substantially affect the rate and total amount of drug accessingsubcutaneous injured tissues the effect of formulation may be asimportant as the individual NSAID used Drug and formulationshould thus be considered together when assessing efficacy andthis is now possible
We have also included an assessment of the daily dose of NSAIDapplied to the skin This involved having information on the con-centration of NSAID in the preparation the amount used andthe frequency of use Not all studies reported all three but an esti-mation of topical doses applied was possible It varied by factors ofthree or four for each topical NSAID However for topical formu-lations the key issue is less the dose applied but the amount thatpenetrates locally (producing analgesic effect) and the amount en-tering the systemic circulation (producing potential harms) Bothwill depend on the exact formulation of the topical agent andwhether there is occlusion (Moore 2008a)
Summary of main results
This review included 61 studies comparing a topical NSAID withplacebo another topical NSAID or an oral NSAID In total5311 participants were treated with a topical NSAID 3470 withplacebo and 220 with an oral NSAID There were 63 moreparticipants than in the previous version of this review Conditionstreated were sprains strains and contusions mainly resulting fromsports injuries and overuse injuries such as tendinitisFormulations of topical diclofenac ibuprofen ketoprofen piroxi-cam and indomethacin demonstrated significantly higher rates ofclinical success than matching topical placebo lacking the NSAIDbenzydamine did not Three drug and formulation combinationshad NNTs for clinical success below 4 For diclofenac Emulgelreghad the lowest NNT of 18 (15 to 21) in two studies using atleast 50 pain intensity reduction as the outcome (high qualityevidence) Diclofenac plasters other than Flectorreg also had a lowNNT of 32 (26 to 42) based on good or excellent responses inrelatively recent studies (high quality evidence) Ketoprofen gelhad an NNT of 25 (20 to 34) from five studies in the 1980ssome with less well defined outcomes (moderate quality evidence)Ibuprofen gel had an NNT of 39 (27 to 67) from two stud-ies with outcomes of marked improvement or complete remission(moderate quality evidence) All other drug and formulation com-binations had NNT values above 4 indicating lesser efficacyThese results are better than alternative topical products that mightbe used for acute musculoskeletal pain There is no evidence tosupport the use of topical salicylate rubefacients (Derry 2014)
Treatment with a topical NSAID was not associated with anincrease in local adverse events (skin reactions) compared withplacebo (inert carrier) or in withdrawals due to adverse events(high quality evidence) The inert carrier was sometimes associ-ated with mild skin irritation but this rarely led to cessation oftreatment and quickly resolved Systemic adverse events were un-common and did not differ between topical NSAID and placebo(high quality evidence) Two participants experienced serious ad-verse events with diclofenac plaster and diclofenac gel but it isunlikely that these were related to the study medicationsThere were insufficient data directly comparing a topical NSAIDwith the same oral NSAID to draw conclusions about efficacyBased on very limited data for oral NSAIDs there were fewersystemic adverse events with topical than oral treatment Therewere sufficient data only for topical piroxicam compared withtopical indomethacin to compare one topical agent with anotherThese limited data suggested that piroxicam was more effectivethan indomethacin and was less likely to cause local adverse eventsIt is worth noting here that topical indomethacin did not givesignificantly better pain relief than placebo in two of the threestudies in this analysis
Overall completeness and applicability ofevidence
There is a tension between pooling studies to produce analyseswith larger numbers and the subsequent large increases in clinicaland statistical heterogeneity on the one hand and using the ap-proach of clinical homogeneity with subsequent smaller numbersof participants on the other hand Previous reviews have takenthe former approach that is useful in demonstrating that topicalNSAIDs rsquoworkrsquo by being significantly better than placebo Becauseof the substantial increase in the amount of data available in thisreview we have chosen to seek greater clinical homogeneity thisproduces results that are more relevant to patient and prescriberchoiceThere were too few studies comparing one topical NSAID versusanother or versus the same oral NSAID to allow meaningful directcomparisons between individual drugs or routes of administrationThe conditions treated in these studies are representative of thoselikely to be suitable for acute treatment with topical NSAIDs Themean age of participants in individual studies ranged from 25 to57 years and the nature of recruitment in many studies meant thatparticipants were actively engaged in sporting activities Neverthe-less older people in their 60s to 80s were also included in somestudies and the low levels of predominantly mild adverse eventsmeans that this route of administration of NSAIDs is suitable forall age groups able to manage the application processInformation from other sources mainly randomised studies last-ing 12 weeks or more in older populations with arthritis tendto confirm this A systematic review of topical NSAIDs in olderadults was difficult to interpret but suggested that the range of
25Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
withdrawal rates in these studies was similar with topical and oralNSAIDs (Makris 2010) It also claimed potentiation of warfarineffects but that was with topical salicylate not an NSAID Incontrast a pooled safety analysis of topical diclofenac in peopleaged 75 years or older reported minimal changes with a mean re-duction in haemoglobin of less than 1 gL with topical diclofenacand a mean systolic blood pressure reduction of almost 4 mm Hg(Roth 2012) Two large randomised 12-week studies comparingtopical with oral diclofenac in arthritis reported lower rates of gas-trointestinal adverse events with topical than oral especially severeevents but larger reductions in haemoglobin with oral diclofenac(Simon 2009 Tugwell 2004)The available evidence was limited by numbers to comment on rarebut potentially serious adverse events One example is the potentialfor photo-sensitivity reactions with topical ketoprofen Currentadvice from the Medicines and Healthcare products RegulatoryAgency in the UK is to avoid direct sunlight ultraviolet (UV) rayssunlamps and sunbeds while using topical ketoprofen and to seea healthcare professional or go to hospital if they experience a skinreaction to sunlight sunlamps or sunbeds (MHRA 2009)
Quality of the evidence
All included studies were both randomised and double-blind nonewas considered at high risk of methodological bias Many werecarried out in the 1980s and 1990s when methodological rigorand detailed reporting were not given such high priority and stud-ies did not always report details of the randomisation treatmentallocation and blinding processes More recent studies often didreport methodological details and tended to be larger (see Figure4 for a comparison of quality of reporting for different dates andNSAIDs) Our primary outcome of clinical success was not alwayswell-defined and was measured using different scales but againmore recent studies tended to report outcomes betterThe studies were conducted in different conditions with some-what different outcome definitions and duration and with differ-ent topical NSAIDs and formulations Moreover the small size ofmany of the studies is likely to result in considerable chance varia-tion (Counsell 1994 Moore 1998b) These factors would accountfor the high I2 values seen in several analyses Despite these sourcesof potential clinical heterogeneity most studies showed benefit oftopical NSAID over placeboThe design of studies to be able to demonstrate analgesic sensi-tivity is important in self limiting conditions such as strains andsprains Too long a duration and the condition results in sponta-neous resolution of painful symptoms while too short a durationmay be inadequate to show any effect The decision by trialists toconcentrate on outcomes closest to seven days of treatment appearsto be prudent and has been adopted in this and previous reviewsThere are potential differences in response to treatment betweenstrains and sprains and overuse-type injuries such as tendinitis
and future reviews may examine this At the present time thereare too few existing trials to explore any differences adequatelyBaseline pain may be a cause for concern Seven studies didnot report baseline pain levels (Billigmann 1996 Curioni 1985Haig 1986 NCT01255423 NCT01272934 NCT01272947Sinniger 1981) and a further 11 reported either mean levels ofless than moderate pain or a significant proportion of individualswith less than moderate pain ( kermark 1990 Aoki 1984 Auclair1989 Diebshlag 1990 Fujimaki 1985 Jenoure 1997 Linde 1985Picchio 1981 Ramesh 1983 Sugioka 1984 Whitefield 2002)using recognised scales Insufficient pain at baseline compromisesthe ability of a study to demonstrate any improvement All thenewly added studies reported baseline pain to be of at least mod-erate intensity
Potential biases in the review process
There has been greater interest in topical NSAIDs in recent yearsmainly because lower systemic drug levels reduce the risk of trou-blesome and severe adverse events particularly in the gastroin-testinal tract and renal and cardiovascular systems Most of theattention has been in chronic conditions such as osteoarthritiswith few studies in acute painful conditions Low levels of seriousadverse events with topical NSAIDs has been noted previously(Evans 1995) and the near absence of serious adverse events inthis review is unlikely to be due to any biases in the review processOne potential bias is that clinical trials for topical NSAIDs may nothave been published One previous review did find previously un-published trials (Moore 1998a) but a subsequent attempt that in-cluded extensive contacts with pharmaceutical companies revealedno additional data (Mason 2004a) While some old unpublishedstudies of topical NSAIDs in acute painful conditions may existthey constitute an unknown number of studies and participantswhose results are unknown and are likely to remain unknownFurthermore their relevance to current clinical practice may belimited as better formulations are developed New systems of trialregistration mean that we know what recent studies have been doneor are ongoing the number of studies and participants is knowneven if their results remain unknown We identified in Clinicaltri-alsgov three unpublished studies (612 participants) with adverseevent data but no dichotomous efficacy data 14 completed un-published studies (4403 participants) with no results posted andthree ongoing studies (880 participants)For the main topical NSAIDs of interest and where most informa-tion exists about 4200 participants in this review provided data onefficacy for diclofenac ibuprofen ketoprofen indomethacin andbenzydamine compared with placebo For efficacy there are un-known results from almost 5900 participants in studies known tohave been done but essentially unpublished Almost 6500 partic-ipants in this review provided information on local adverse eventsfor topical NSAIDs compared with placebo For local adverseevents the unknown results from known studies represents almost
26Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
5300 participants It is clear that identified unpublished but un-available study data amounts to a further potentially large increasein knowledge over and above the 60 increase in numbers ofparticipants already included in this updated reviewBased on efficacy data on known and available study results un-published trials showing no difference between any topical NSAIDand topical placebo and involving 5500 participants would haveto exist in order for the NNT to be as high as 9 at which pointthe effectiveness of topical NSAIDs would become clinically irrel-evant (Moore 2006) This amount of unpublished negative data isobviously available and while a negative result in all the identifiedstudies is unlikely knowledge would be greatly served by havingthese unpublished trial results availableWe have not yet attempted to obtain results from these clinicaltrials from the trial sponsors because this takes a considerableamount of time and may not be successful Moreover the studieswere spilt between nine different sponsors Cerimon Pharmaceu-ticals (five studies) Novartis (four studies) Endo Pharmaceuticals(three studies) GlaxoSmithKline (two studies) Hisamatsu Phar-maceutical (two studies) and one each from Actavis Pfizer Im-primis Pharmaceutical and Strategic Science amp Technologies
Agreements and disagreements with otherstudies or reviews
A review published in 2004 included some of the studies in thisreview and reported an NNT for all topical NSAIDs combinedof 38 (34 to 44) for clinical success equivalent to half pain reliefat seven days (Mason 2004a) That review found no differencebetween topical NSAID and placebo for local adverse events as didthis review In turn the Mason review was in broad agreement withthe original systematic review on topical NSAIDs (Moore 1998a)To our knowledge no previous review assembled sufficient trialdata to analyse results by both drug and formulation as was donehere
A U T H O R S rsquo C O N C L U S I O N S
Implications for practice
For people with acute musculoskeletal pain
Topical NSAIDs can provide good levels of pain relief in acuteconditions such as sprains strains and overuse injuries proba-bly similar to that provided by oral NSAIDs Gel formulations ofdiclofenac (as Emulgelreg) ibuprofen and ketoprofen and somediclofenac patches provide the best effects Adverse events are usu-ally minimal with topical NSAIDs
For clinicians
Topical diclofenac ibuprofen or ketoprofen gels provide goodpain relief for painful acute musculoskeletal conditions and arebetter tolerated than oral formulations These drugs and formu-lations are more likely to be cost effective than alternative topicalpreparations such as topical rubefacients
For policy makers
Topical NSAIDs are not associated with an increased incidence oflocal skin reactions compared with the inert carrier and while thecarrier may cause mild transient irritation it is rarely troublesomeTopical NSAIDs do not cause systemic (mainly gastrointestinal)problems commonly seen with oral NSAIDs making them partic-ularly useful for individuals unable to tolerate oral administrationor for whom it is contraindicated
For funders
Topical diclofenac ibuprofen or ketoprofen gels should be consid-ered for initial treatment of acute musculoskeletal painful condi-tions where there are no contraindications such as damaged skinThese drugs and formulations are more likely to be cost effectivethan alternative topical preparations such as topical rubefacientsBecause formulations of topical NSAIDs are likely to change overtime the relevant trials performed and reported in or before the1990s must be limited and may be questionable Funders mightwish to consider asking pharmaceutical companies without recenttrial evidence for their products to produce it
Implications for research
General
The general thrust of these findings is that gel formulations of top-ical diclofenac ibuprofen and ketoprofen work best but for somedrugs (ketoprofen for instance) studies were pre-1990 Thesestudies may not be relevant to products available now Becauseformulation can have a significant effect on efficacy formulationchanges should be accompanied by relevant randomised trials
Design
The design of the trials is generally good and the sports injurymodel appears to be reliable and reproducible Modern studieshave ensured that participants entering the trials have at least mod-erate pain and this helps sensitivity to detect an analgesic responseMajor changes to the design of these trials would not appear to beneeded
27Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Measurement (endpoints)
A major issue is not in the measurement of pain as most studiesespecially modern studies have used standard pain intensity andpain relief scales However reporting of average pain changes isinadequate and the use of responder analyses (at least 50 painintensity reduction or people experiencing mild or no pain) ispreferred
Comparison between active treatments
Indirect comparisons with placebo are probably as informative asuse of an active comparator
A C K N O W L E D G E M E N T S
The Oxford Pain Relief Trust the NHS Cochrane CollaborationProgramme Grant Scheme and the NIHR Biomedical ResearchCentre Programme provided support for the earlier review
The National Institute for Health Research (NIHR) is the largestsingle funder of the Cochrane Pain Palliative and SupportiveCare Review Group Disclaimer the views and opinions expressedherein are those of the authors and do not necessarily reflect thoseof the NIHR National Health Service (NHS) or the Departmentof Health
R E F E R E N C E S
References to studies included in this review
Airaksinen 1993 published data onlyAiraksinen O Venaumllaumlinen J Pietilaumlinen T Ketoprofen 25gel versus placebo gel in the treatment of acute soft tissueinjuries International Journal of Clinical PharmacologyTherapy and Toxicology 199331(11)561ndash3
kermark 1990 published data only
kermark C Forsskaringhl B Topical indomethacin in overuseinjuries in athletes A randomized double blind studycomparing Elmetacinreg with oral indomethacin andplacebo International Journal of Sports Medicine 199011(5)393ndash6
Aoki 1984 published data only
Aoki T Numajiri M Yamamoto M A well controlledcomparative study of piroxicam gel indomethacin geland placebo gel in the treatment of trauma JapanesePharmacology and Therapeutics 198412(12)101ndash17
Auclair 1989 published data onlyAuclair J Georges M Grapton X Gryp L DrsquoHooghe MMeisser RG et al A double-blind controlled multi-centerstudy of percutaneous niflumic acid gel and placebo in thetreatment of Achilles heel tendinitis Current TherapeuticResearch 198946(4)782ndash8
Billigmann 1996 published data onlyBilligmann PW Treatment of ankle distortion withibuprofen gel [Therapie von Sprunggelenksndashdistosionenmit ibuprofenndashmikrogel] Therapiewoche 199646(21)1187ndash92
Campbell 1994 published data onlyCampbell J Dunn T Evaluation of topical ibuprofen creamin the treatment of acute ankle sprains Journal of Accidentand Emergency Medicine 199411(3)178ndash82
Chatterjee 1977 published data onlyChatterjee DS A double-blind clinical study withbenzydamine 3 cream on soft tissue injuries in anoccupational health centre Journal of International MedicalResearch 19775(6)450ndash8
Costantino 2011 published data only
Costantino C Kwarecki J Samokhin AV Mautone GRovati S Diclofenac epolamine plus heparin plasterversus diclofenac epolamine plaster in mild to moderateankle sprain a randomized double-blind parallel-groupplacebo-controlled multicentre phase III trial ClinicalDrug Investigation 201131(1)15ndash26 [DOI 10216511585890-000000000-00000]
Coudreuse 2010 published data onlyCoudreuse JM de Vathaire F Effect of a plaster containingDHEP and heparin in acute ankle sprains with oedemaa randomized double-blind placebo-controlled clinicalstudy Current Medical Research and Opinion 201026(9)2221ndash8 [DOI 101185030079952010508020]
Curioni 1985 published data only
Curioni BG Di Domenica F Daolio P Spignoli GEvaluation of ibuproxam gel in traumatology [Valutazionedellrsquoefficacia terapeutica di ibudros gel in traumatologia]Clinica Europea 198524(3)456ndash60
Diebshlag 1990 published data only
Diebschlag W Nocker W Bullingham R A double-blindstudy of the efficacy of topical ketorolac tromethamine gelin the treatment of ankle sprain in comparison to placeboand etofenamate Journal of Clinical Pharmacology 199030
(1)82ndash9
Dreiser 1988 published data onlyDreiser RL Clinical trial of efficacy and tolerability oftopical ibuprofen in the treatment of tendinitis Le JournalInternational De Meacutedecine 198811915ndash31
Dreiser 1989 published data only
Dreiser RL Clinical trial - Fatsum gel FG-6 Supplied byMenarini 1989
Dreiser 1990 published data onlyDreiser RE Charlot J Lopez A Ditisheim A Clinicalevaluation of niflumic acid gel in the treatment ofuncomplicated ankle sprains Current Medical Research andOpinion 199012(2)93ndash9
28Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Dreiser 1994 published data only
Dreiser RL Roche R de Sahb R Thomas F Leutenegger EFlurbiprofen local action transcutaneous (LAT tm) clinicalevaluation in the treatment of acute ankle sprains EuropeanJournal of Rheumatology and Inflammation 199414(4)9ndash13
Fioravanti 1999 published data onlyFioravanti A Cicero MR Nerucci F Manopulo RMarcolongo R Double-blind controlled clinical study of theefficacy and tolerability of diclofenac-N-(2-hydroxyethyl)-pyrrolidine lecithin gel compared with diclofenac-N-(2-hydroxyethyl)-pyrrolidine gel in patients with periand extraarticular inflammatory diseases Drugs underExperimental and Clinical Research 199925(5)235ndash40
Fujimaki 1985 published data only
Fujimaki E et al Clinical evaluation of piroxicam gelversus indomethacin gel and placebo in the treatment ofmuscle pain a double-blind multicenter study JapanesePharmacology and Therapeutics 198512(12)119ndash37
Gallacchi 1990 published data only
Gallacchi G Mautone G Lauldi P Topical treatmentwith diclofenac hydroxyethylpyrrilidine (Flector gel 1)Clinical Trials Journal 199027(1)58ndash64
Gonzaacutelez de Vega 2013 published data only
Gonzaacutelez de Vega C Speed C Wolfarth B Gonzaacutelez JTraumeel vs diclofenac for reducing pain and improvingankle mobility after acute ankle sprain a multicentrerandomised blinded controlled and non-inferiority trialInternational Journal of Clinical Practice 201367(10)979ndash89 [DOI 101111ijcp12219]
Governali 1995 published data only
Governali E Casalini D A controlled clinical studyon ketoprofen gel 5 versus ketoprofen ointment 1in patients with post-traumatic lesions [Ricerda clinicacontrollata tra ketoprofene gel 5 e ketoprofene crema1 in pazienti con postumi di lesioni traumatiche]Riabilitazione 199528(1)61ndash9
Gualdi 1987 published data only
Gualdi A Bonollo L Martini A Forgione A Non-steroidal anti-inflammatory drugs for topical therapy intraumatology a double-blind study with flunoxaprofen andketoprofen [Antinflammatori no steroidi per uso topicoin traumatologia studio clinico con flunoxaprofene echetoprofene] Riforma Medica 1987102(10)401ndash4
Haig 1986 published data only
Haig G Portable thermogram technique for topicallyapplied benzydamine cream in acute soft-tissue injuriesInternation Journal of Tissue Reactions 19868(2)145ndash7
Hoffmann 2012 published data only
Hoffmann P Kopa ka P Gugliotta B Rovati S Efficacyand tolerability of DHEP-heparin plaster in reducingpain in mild-to-moderate muscle contusions a double-blind randomized trial Current Medical Researchand Opinion 201228(8)1313ndash21 [DOI 101185030079952012709182]
Hofman 2000 published data only
Hofman J Nasswetter G Cayetti LM Lysine clonixinate gelin soft tissue injuries Controlled randomized prospectivedouble-blind clinical trial with diclofenac [Clonixinatode lisina gel en lesiones de tejidos blandos ensayo clinicoprospectivo doble ciego al azar controlado con diclofenac]Prensa Medica Argentina 200087(5)513ndash20
Hosie 1993 published data only
Hosie GAC The topical NSAID felbinac versus oralibuprofen a comparison of efficacy in the treatment ofacute lower back injury British Journal of Clinical Research199345ndash17
Jenoure 1997 published data only
Jenoure PJ Rostan A Gremion G Meier JL Grossen RBielinki R et al Multicentre double-blind controlledclinical study on the efficacy of diclofenac epolamineTissugel plaster in patients with epicondylitis [Studiomulticentrico controllato in doppio cieco su diclofenacTissugel plaster in pazienti con epicondilite] MedicinaDello Sport 197750(3)285ndash92
Joussellin 2003 published data onlylowast Joussellin E Flector Tissugel for the treatment ofpainful ankle sprains [Flector Tissugel dans le traitementdes entorses douloureuses de la cheville] Journal deTraumatologie du Sport 2003201S5ndash9Lionberger DR Joussellin E Lanzarotti A Yanchick JMagelli M Diclofenac epolamine topical patch relieves painassociated with ankle sprain Journal of Pain Research 2011447ndash53 [DOI 102147JPRS15380]Lionberger DR Joussellin E Yanchick J Magelli MLanzarotti A Pooled analysis of clinical trial data evaluatingthe safety and effectiveness of diclofenac epolamine topicalpatch 13 for the treatment of acute ankle sprain OpenAccess Journal of Sports Medicine 2011275ndash84 [DOI102147OAJSMS17048]
Julien 1989 published data onlyJulien D Clinical trial - Fastum gel FG-8 Supplied byMenarini 1989
Klainguti 2010 published data onlyKlainguti A Forgacs A Berkes I Castellacci E A plastercontaining DHEP and heparin for mild to moderatecontusions and sprains with haematoma a double-blind randomized study Current Medical Researchand Opinion 201026(9)2243ndash51 [DOI 101185030079952010508022]
Kockelbergh 1985 published data onlyKockelbergh M Verspeelt P Caloine R Dermaux F Localanti inflammatory treatment with a ketoprofen gel currentclinical findings Journal Belge de Medecine Physique et deRehabilitation 19858(4)205ndash13
Kuehl 2011 published data only
Kuehl K Carr W Yanchick J Magelli M Rovati SAnalgesic efficacy and safety of the diclofenac epolaminetopical patch 13 (DETP) in minor soft tissue injuryInternational Journal of Sports Medicine 201132(8)635ndash43[DOI 101055s-0031-1275359]
29Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Li 2013 published data only
Li C Frangione V Rovati S Zheng Q Diclofenacepolamine medicated plaster in the treatment of minor softtissue injuries a multicenter randomized controlled trialCurrent Medical Research and Opinion 201329(9)1137ndash46[DOI 101185030079952013816669]
Linde 1985 published data only
Linde F Hvass I Juumlrgensen U Madsen F Treatment ofsprained ankles with 5 benzydamine creme A double-blind study [Ankelforstuvninger behandlet med benzydamin5 creme] Ugeskrift for Laeger 1985148(1)12ndash3
Machen 2002 published data onlyMachen J Whitefield M Efficacy of a proprietary ibuprofengel in soft tissue injuries a randomised double blindplacebo controlled study International Journal of ClinicalPharmacology 200256(2)102ndash6
Mahler 2003 published data only
Mahler P Mahler F Duruz H Ramazzina M Liguori VMautone G Double-blind randomized controlled studyon the efficacy and safety of a novel diclofenac epolaminegel formulated with lecithin for the treatment of sprainsstrains and contusions Drugs under Experimental andClinical Research 200329(1)45ndash52
Maziegraveres 2005a published data onlyMaziegraveres B Rouanet S Velicy J Scarsi C Reiner V Topicalketoprofen patch (100 mg) for the treatment of anklesprain a randomized double-blind placebo-controlledstudy American Journal of Sports Medicine 200533(4)515ndash23 [DOI 1011770363546504268135]
Maziegraveres 2005b published data onlyMaziegraveres B Rouanet S Guillon Y Scarsi C Reiner VTopical ketoprofen patch in the treatment of tendinitisa randomized double blind placebo controlled studyJournal of Rheumatology 200532(8)1563ndash70
McLatchie 1989 published data onlyMcLatchie GR McDonald M Lawrence GF RogmansD Lisai P Hibberd M Soft tissue trauma a randomisedcontrolled trial of the topical application of felbinac a newNSAID British Journal of Clinical Practice 198943(8)277ndash80
Morris 1991 published data only
Morris WD Scott HV Peters WA Ketelbey JW Felbinactopical gel for acute soft tissue sports injuries New ZealandJournal of Sports Medicine 19911945ndash7
NCT01255423 unpublished data only
Novartis (Sponsors) A randomized double-blind multi-center placebo-controlled parallel group study to evaluatethe efficacy and safety of diclofenac sodium topical gel(DSG) 1 applied 4 times daily in subjects with acuteankle sprain clinicaltrialsgovct2showNCT01255423(accessed 3 February 2015) 2012 [CTG NCT01255423]
NCT01272934 unpublished data only
Novartis (Sponsors) A randomized double-blind multi-center placebo-controlled parallel group study to evaluatethe efficacy and safety of diclofenac sodium topical gel
(DSG) 1 applied four times daily in subjects withacute ankle sprain clinicaltrialsgovct2showresultsNCT01272934 (accessed 3 February 2015) 2013 [CTGNCT01272934]
NCT01272947 unpublished data onlylowast Novartis (Sponsors) A randomized double-blindmulti-center placebo-controlled parallel group study toevaluate the efficacy and safety of diclofenac sodium topicalgel (DSG) 1 applied four times daily in subjects withacute blunt soft tissue injuriescontusions of the limbsclinicaltrialsgovct2showresultsNCT01272947 (accessed3 February 2015) 2012 [CTG NCT01272947]Pallay R Pabst H Giannetti B Burnett I Monnet JDiclofenac sodium topical gel (DSG) 1 for acute softtissue injuriescontusions of the limbs 2013 AnnualAssembly of the American Academy of Physical Medicineand Rehabilitation National Harbor MD United States2013 Vol 5 (9 Suppl 1)S204
Noret 1987 published data only
Noret A Roty V Allington N Hauters P Zuinen C PoelsR Ketoprofen gel as topical treatment for sport injuriesActa Therapeutica 198713367ndash78
Parrini 1992 published data onlyParrini M Cabitza P Arrigo A Vanasia M Efficacy andtolerability of ketoprofen lysine salt foam for topical use inthe treatment of traumatic pathologies of the locomotorapparatus [Efficacia e tollerabilita del ketoprofene sale dilisina schiuma per uso topico nel trattamento di alcunepatologie traumatiche dellrsquoapparato locomotore] La ClinicaTerapeutica 1992141(9)199ndash204
Picchio 1981 published data only
Picchio AA Volta S Longoni A Controlled clinical trial ofibuprofen for topical use in sport injuries [Studio clinicocontrollato sullrsquoimpiego dellrsquoibuprofen per uso topico intraumatologica sportiva] Medicina dello Sport 198134403ndash6
Predel 2004 published data only
Mueller EA Kirch W Reiter S Extent and time courseof pain intensity upon treatment with a topical diclofenacsodium patch versus placebo in acute traumatic injury basedon a validated end point post hoc analysis of a randomizedplacebo-controlled trial Expert Opinion on Pharmacotherapy201011(4)493ndash8 [DOI 10151714656560903535898]lowast Predel HG Koll R Pabst H Dieter R Gallacchi GGiannetti B et al Diclofenac patch for topical treatmentof acute impact injuries a randomised double blindplacebo controlled multicentre study British Journal ofSports Medicine 200438(3)318ndash23 [DOI 101136bjsm2003005017]
Predel 2012 published data onlyPredel HG Hamelsky S Gold M Giannetti B Efficacy andsafety of diclofenac diethylamine 232 gel in acute anklesprain Medicine and Science in Sports and Exercise 201244
(9)1629ndash36 [DOI 101249MSS0b013e318257ed41]
30Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Predel 2013a published data only
Predel HG Giannetti B Seigfried B Novellini R MenkeG A randomized double-blind placebo-controlledmulticentre study to evaluate the efficacy and safetyof diclofenac 4 spray gel in the treatment of acuteuncomplicated ankle sprain Journal of InternationalMedical Research 201341(4)1187ndash202 [DOI 1011770300060513487639]
Predel 2013b published data onlyPredel HG Giannetti B Pabst H Schaefer A Hug AMBurnett I Efficacy and safety of diclofenac diethylamine116 gel in acute neck pain a randomized double-blindplacebo-controlled study BMC Musculoskeletal Disorders201314250 [DOI 1011861471-2474-14-250]
Ramesh 1983 published data only
Ramesh N Steuber U Ibuprofen in a cream vehicle foraccidental and sport injuries [Dolgit creme bei unfallndash undsportverletzungen] Therapiewoche 1983334563ndash70
Rowbotham 2003 published data only
Rowbotham M Galer B Block J Backonja M FlectorTissugel efficacy and safety in minor sport injuries Avs controlled clinical trial Journal of Sports Traumatology200320IS15ndashIS20
Russell 1991 published data only
Russell AL Piroxicam 05 topical gel compared to placeboin the treatment of acute soft tissue injuries a double-blind study comparing efficacy and safety Clinical andInvestigative Medicine 199114(1)35ndash43
Saillant 1998 published data only
Lionberger DR Joussellin E Yanchick J Magelli MLanzarotti A Pooled analysis of clinical trial data evaluatingthe safety and effectiveness of diclofenac epolamine topicalpatch 13 for the treatment of acute ankle sprain OpenAccess Journal of Sports Medicine 2011275ndash84 [DOI102147OAJSMS17048]lowast Saillant G Study comparing the efficacy and toleranceof Flector Tissugelreg to that of a placebo in the treatmentof benign ankle sprains [Eacutetude comparant lrsquoefficaciteacute et latoleacuterance de FlectorTissugelreg agrave celles drsquoun placebo dans letraitement des entorses beacutenignes de la cheville] Medicinedu Sport 1998721ndash5
Sanguinetti 1989 published data onlySanguinetti C Treatment of soft tissue injury with BPAAgel Results of an Italian multicenter study vs placebo[Trattemento con BPAA gel dei traumi dei tessuti moli]Clinica Terapeutica 1989130(5)255ndash258
Sinniger 1981 published data onlySinniger M Blanchard P Controlled clinical trial withFentiazac cream in sport microtraumatology Journal ofInternational Medical Research 19819(4)300ndash2
Spacca 2005 published data onlySpacca G Cacchio A Forgcs A Monteforte PRovetta G Analgesic efficacy of a lecithin-vehiculateddiclofenac epolamine gel in shoulder periarthritis andlateral epicondylitis a placebo-controlled multicenter
randomized double-blind clinical trial Drugs underExperimental and Clinical Research 200531(4)147ndash54
Sugioka 1984 published data only
Sugioka Y Multicenter clinical evaluation of piroxicam gelvs indomethacin gel in the treatment of non-traumaticdiseases of tendon or muscle Japanese Pharmacology andTherapeutics 198412139ndash53
Thorling 1990 published data only
Thorling J Linden B Berg R Sandahl A A double blindcomparison of naproxen gel and placebo in the treatment ofsoft tissue injuries Current Medical Research and Opinion199012242ndash8
Tonutti 1994 published data only
Tonutti A The use of ketoprofen gel 5 (orudis gel) intraumatology controlled double-blind study vs etofenamate[Utilizzazione del ketoprofene gel 5 (orudis gel) nellapratica traumatologica studio in doppio cieco controllatoverso etofenamato] Ortopedia e Traumatologia Oggi 199414(3)119ndash25
Vecchiet 1991 published data only
Vecchiet L Colozzi A Effects of meclofenamic acid in thetreatment of lesions deriving from minor traumatologyClinical Journal of Pain 19917 Suppl 1S54ndash9
Whitefield 2002 published data only
Whitefield M OrsquoKane CJA Anderson S Comparativeefficacy of a proprietary topical ibuprofen gel and oralibuprofen in acute soft tissue injuries a randomised doubleblind study Journal of Clinical Pharmacy and Therapeutics200227(6)409ndash17
References to studies excluded from this review
Ambrus 1987 published data only
Ambrus P Boumlhmer D Mobilat ointment in acute sprains[Mobilat Salbe bei akuten Distorsionen] Fortschritte derMedizin 1987105(13)259ndash62
Anon 1993 unpublished data only
Anon Comparative clinical efficacy of Oruvail piroxicamand diclofenac gels in soft tissue injury Unpublished
Ascherl 1982 published data onlyAscherl R Schlemmer H Blumel G Lechner F Theeffectiveness of etofenamate in minor sports injuries ofthe knee and ankle joint A double blind study [DieWirksamkeit der Etofenamat in kleineren Sportverletzungenam Knie und Sprunggelenk Eine DoppelblindndashStudie]Fortschritte der Medizin 1982100(37)1729ndash34
Bagliani 1976 published data only
Bagliani A Montalbetti L Topical treatment ofthrombophlebitis with feprazone and benzydamineControlled clinical study [Il trattamento topico ditromboflebite con feprazone e benzidamina Studio clinicocontrollato] Minerva Medica 197667(14)880ndash4
Baracchi 1982 published data only
Baracchi G Messina Denaro S Piscini S Experience ofthe topical use of isobutylfenylproprionic acid (Ibuprofen)in traumatic inflammation A double-blind comparison
31Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
with placebo [Esperienza sullrsquoimpiego topico dellrsquoacidoisobutilndashfenilndashpropionico (ibuprofen) nella flogositraumatica Confronto a doppia cecita con placebo]Gazzetta Medica Italiana 1982141(12)691ndash4
Boumlhmer 1995 published data onlyBoumlhmer D Ambrus P Treatment of muscular injurieswith diclofenac-diethylammonium emugel [Behandlungvon muskelverletzungen mit diclofenacndashdiethylammoniumemugel] Sportverletzung Sportschaden 1995994ndash5
Burnham 1998 published data only
Burnham R Gregg R Healy P Steadward R Theeffectiveness of topical diclofenac for lateral epicondylitisClinical Journal of Sport Medicine 19988(2)78ndash81
Cesarone 2008 published data onlyCesarone MR Belcaro G Pellegrini L Ledda A VinciguerraG Ricci A et al Treatment of ankle sprain in patientswith vascular diseases of the lower limbs MinervaCardioangiologica 200856 (5 Suppl)39ndash46
Coulibaly 2009 published data only
Coulibaly SK Courau S Staiger C The sprained anklecomparative study of a comfreyroot extract ointment versusa diclofenac gel [Entorses de la cheville eacutetude comparativeentre un extrait agrave base de racines de consoude et lediclofeacutenac] Phytotheacuterpie 20097147ndash9 [DOI 101007s10298-009-0383-x]
Diebschlag 1985 published data only
Diebschlag W Benzydamine cream in post-traumaticoedema International Journal of Tissue Reactions 19857(3)219ndash23
Diebschlag 1986 published data onlyDiebschlag W Diclofenac in blunt traumatic ankle jointswelling Volumetric monitoring in a placebo controlleddouble blind trial [Diclofenac in stumpfen traumatischenSprunggelenk Schwellung Volumetrische Uumlberwachungin einer Placebondashkontrollierten DoppelblindndashStudie]Fortschritte der Medizin 1986104(21)437ndash40
Diebschlag 1992 published data onlyDiebschlag W Nocker W Lehmacher W Treatment of acutesprains of the ankle joint A comparison of the effectivenessand tolerance of two gel preparations containingindomethacin [Die Behandlung der akuten Verstauchungendes Sprunggelenks Ein Vergleich der Wirksamkeitund Vertraumlglichkeit von zwei GelndashZubereitungen mitIndometacin] Fortschritte der Medizin 1992110(6)64ndash72
Fantato 1971 published data only
Fantato S De Gregorio M Clinical evaluation of topicalbenzydamine in traumatology Arzneimittel-Forschung197121(10)1530ndash5
Galer 2000 published data onlyGaler BS Rowbotham M Perander J Devers A FriedmanE Topical diclofenac patch relieves minor sports injury painresults of a multicenter controlled clinical trial Journal ofPain and Symptom Management 200019(4)287ndash94
Hallmeier 1986 published data only
Hallmeier B Michelbach B Etofenamate under tapebandages - a controlled study [Etofenamat untertapendashverbaumlnden] Medizinische Welt 198637(43)1344ndash8
Hallmeier 1988 published data onlyHallmeier B Efficacy and tolerance of etofenamate anddiclofenac in acute sports injuries [Wirksamkeit undVertraumlglichkeit von Etofenamat und Diclofenac bei akutenSportverletzungen] Rheuma 19888183ndash6
Kaneko 1999 published data only
Kaneko M Shimojo H Saito H Onuma Y YamashitaK Clinical evaluation of felbinac patch (SELSPOT) onpost-traumatic disease clinical comparative study versuscommercially available patch Japanese Pharmacology andTherapeutics 19992775ndash85
Kockelbergh 1985b published data only
Kockelbergh M Verspeelt P Caloine R Dermaux F Localanti-inflammatory treatment with a ketoprofen gel currentclinical findings [Traitement antindashinflammatoire local parun gel de keacutetoprofegravene donneacutees cliniques reacutecentes] JournalBelge de Medecine Physique et de Rehabilitation 19858(4)205-13 (study 2)
Kuwabara 2013 published data only
Kuwabara Y Hamamoto H Hikake S Miwa Y Arandomized multi-center double-blind placebo-controlledphase IIIII trial to evaluate the eftopical patch in thetreatment of pain Journal of Pain 2013 Vol 14 (4 Suppl1)S73
Lee 1991 published data only
Lee EH Lee PY Ngai AT Chiu EH Treatment of acute softtissue trauma with a topical non-steroidal anti-inflammatorydrug (biphenylacetic acid 3 gel) Singapore MedicalJournal 199132(4)238ndash41
Link 1996 published data onlyLink R Balint G Pavlik G Otto J Krause W Topicaltreatment of soft tissue rheumatism and athletic injuriesEffectiveness and tolerance of a new ketoprofen gel[Topische Behandlung von WeichteilndashRheumatismus undSportverletzungen Wirksamkeit und Vertraumlglichkeit einesneuen Ketoprofen Gel] Fortschritte der Medizin 1996114(25)311ndash4
May 2007 published data onlyMay JJ Lovell G Hopkins WG Effectiveness of1 diclofenac gel in the treatment of wrist extensortenosynovitis in long distance kayakers Journal of Scienceand Medicine in Sport 200710(1)59ndash65
Oakland 1993 published data only
Oakland C Rapier C A comparison of the efficacy of thetopical NSAID felbinac and ultrasound in the treatmentof acute ankle injuries British Journal of Clinical Research1993489ndash96
Odaglia 1987 unpublished data only
Odaglia G Sereni G Sports minor traumatology results ofa double-blind controlled clinical study ketoprofen (fastumgel 25) versus placebo Menarini unpublished data
32Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Picardi 1993 published data only
Picardi E De Iasio R Efficacy of percutaneous anti-inflammatory drugs (NSAIDs) in swimmers and waterpoloplayers [Efficacia dei farmaci antinfiammatori (FANS) pervia percutanea in atleti di nuoto e pallanuoto] ClinicaTerapeutica 1993143(6)507ndash9
Taboada 1992 published data onlyTaboada A Controlled trial of piroxicam gel associated withultrasound in acute disturbances of the locomotive system[Experienca controlada con gel de piroxicam asociado aultrasonidos en afecciones agudas del aparato locomotor]Prensa Medica Argentina 199279(10)630ndash2
Vanderstraeten 1990 published data only
Vanderstraeten G Schuermans P Study on the effect ofetofenamate 10 cream in comparison with an oral NSAIDin strains and sprains due to sports injuries Acta BelgicaMedica Physica 199013(3)139ndash41
Vinciguerra 2008 published data only
Vinciguerra G Belcaro G Cesarone MR Errichi BM DiRenzo A Errichi S et al Management of uncomplicatedankle sprains with topical or oral ketoprofen treatmentA registry study Minerva Cardioangiologica 200856 (5
Suppl)47ndash53
Von Klug 1977 published data onlyVon Klug H Experience with a locally applied anti-inflammatory drug [Erfahrungen mit einem localanwendbaren antirheumatikum] Arzneimittel-ForschungDrug Research 1977271350ndash4
References to studies awaiting assessment
NCT00351104 published data onlyEndo Pharmaceuticals (Sponsors) A randomized double-blind placebo-controlled parallel group phase III study ofthe efficacy tolerability and safety of ketoprofen topicalpatch 20 (KTP) in the treatment of pain associated withgrade 1 or grade 2 ankle sprain or strain clinicaltrialsgovct2showNCT00351104 (accessed 3 February 2015) 2008[CTG NCT00351104]
NCT00352625 unpublished data only
Endo Pharmaceuticals (Sponsors) A randomized double-blind placebo-controlled parallel group phase III study ofthe efficacy tolerability and safety of ketoprofen topicalpatch 20 (KTP) in the treatment of pain associatedwith tendonitis or bursitis of the shoulder elbow or kneeclinicaltrialsgovct2showNCT00352625 (accessed 3February 2015) 2008 [CTG NCT00352625]
NCT00426985 unpublished data only
Endo Pharmaceuticals (Sponsors) A randomized double-blind placebo-controlled parallel group phase iii study ofthe efficacy tolerability and safety of ketoprofen topicalpatch 20 (KTP) in the treatment of pain associatedwith tendonitis or bursitis of the shoulder elbow or kneeclinicaltrialsgovct2showNCT00426985 (accessed 3February 2015) 2008 [CTG NCT00426985]
NCT00640705 unpublished data only
Cerimon Pharmaceuticals (Sponsors) A randomizeddouble-blind placebo-controlled study of the efficacyand safety of a diclofenac sodium patch for the topicaltreatment of pain due to mild to moderate ankle sprainclinicaltrialsgovct2showNCT00640705 (accessed 3February 2015) 2008 [CTG NCT00640705]
NCT00640939 unpublished data onlyCerimon Pharmaceuticals (Sponsors) A randomizeddouble-blind placebo-controlled study of the efficacy andsafety of a diclofenac sodium patch for the topical treatmentof pain due to mild to moderate tendonitis or bursitisclinicaltrialsgovct2showNCT00640939 (accessed 3February 2015) 2008 [CTG NCT00640939]
NCT00680472 unpublished data onlyHisamitsu Pharmaceutical Co Inc (Sponsors) Arandomized multicenter double-blind placebo-controlledtwo-week study to assess the efficacy and safety of HKT-500 in subjects with acute shoulder pain clinicaltrialsgovct2showNCT00680472 (accessed 3 February 2015) 2015[CTG NCT00680472]
NCT00680784 unpublished data only
Hisamitsu Pharmaceutical Co Inc (Sponsors) Arandomized multicenter double-blind placebo-controlledstudy to evaluate the efficacy and safety of HKT-500 inthe treatment of pain associated with Grade I or Grade IIankle sprain clinicaltrialsgovct2showNCT00680784(accessed 3 February 2015) 2015 [CTG NCT00680784]
NCT00765700 unpublished data onlyEkman E Skrepnik S Jones M Lawson K Schupp JEfficacy and safety of ketoprofen 10 cream in acute softtissue injuries (phase 3 study tdlp-110-001) Proceedings ofthe 13th World Congress on Pain 2010 Aug 29 - Sept 2Montreal CanadaImprimis Pharmaceuticals Inc (Sponsors) A randomizedmulticenter double-blind placebo-controlled parallel-group phase 3 study to assess the efficacy and safetyof Ketotransdeltrade (Ketoprofen topical cream 10) inthe treatment of pain associated with mild to moderateacute soft tissue injury clinicaltrialsgovct2showNCT00765700 (accessed 3 February 2015) 2013 [CTGNCT00765700]
NCT00869063 unpublished data onlyCerimon Pharmaceuticals (Sponsors) A randomizeddouble-blind placebo-controlled study of the efficacy andsafety of a diclofenac sodium patch for the topical treatmentof acute pain due to mild to moderate wrist sprain strainor contusion clinicaltrialsgovct2showNCT00869063(accessed 3 February 2015) 2010 [CTG NCT00869063]
NCT00869180 unpublished data onlyCerimon Pharmaceuticals (Sponsors) A randomizeddouble-blind placebo-controlled study of the efficacy andsafety of a diclofenac sodium patch for the topical treatmentof acute pain due to mild to moderate ankle sprainclinicaltrialsgovct2showNCT00869180 (accessed 3February 2015) 2010 [CTG NCT00869180]
33Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT00931866 unpublished data only
Cerimon Pharmaceuticals (Sponsors) A randomizeddouble-blind placebo-controlled study of the efficacy andsafety of a diclofenac sodium patch for the topical treatmentof acute pain due to mild to moderate soft tissue injuriesclinicaltrialsgovct2showNCT00931866 (accessed 3February 2015) 2010 [CTG NCT00931866]
NCT01874626 unpublished data only
Strategic Science amp Technologies LLC (Sponsors) Arandomized double-blind placebo-controlled multi-dosepivotal study to determine the efficacy and safety of SST0225 a topical ibuprofen cream in the treatment of painassociated with acute ankle sprain clinicaltrialsgovct2showNCT01874626 (accessed 3 February 2015) 2014[CTG NCT01874626]
NCT01957215 unpublished data only
GlaxoSmithKline (Sponsors) A clinical study to assess theefficacy of pain relief of topical indomethacin patch overplacebo in ankle sprain patients clinicaltrialsgovct2showNCT01957215 (accessed 3 February 2015) 2015 [CTGNCT01957215]
NCT02324270 unpublished data only
Actavis Inc (Sponsors) Randomized double-blindmultiple-center placebo-controlled study comparing thesafety and efficacy of generic diclofenac epolamine toFlectorreg patch in the treatment of acute pain due to minorankle sprain clinicaltrialsgovct2showNCT02324270(accessed 3 February 2015) 2014 [CTG NCT02324270]
Sarzi-Puttini 2014 published data only
Sarzi-Puttini P Atzeni F Damiani C Casale R BarbagalloM Cazzola M A double-blind randomised parallelgroup active controlled multicentre study to assess thetherapeutic non-inferiority of SKP-021 a 03 ketoprofenpatch versus diclofenac sodium patch in patients withacute inflammatory musculoskeletal injuries Proceeding ofthe Annual European Congress of Rheumatology of theEuropean League Against Rheumatism EULAR 2014 June11-14 Paris France 201473
References to ongoing studies
NCT01945034 unpublished data onlyPfizer (Sponsors) Placebo-controlled double-blindevaluation of the efficacy and safety of ibuprofen 5 topicalgel for the treatment of ankle sprain clinicaltrialsgovct2showNCT01945034 (accessed 3 February 2015) 2014[CTG NCT01945034]
NCT02100670 unpublished data onlyGlaxoSmithKline (Sponsors) A clinical study to assessthe efficacy and onset of pain relief of topical MFC51123diclofenac-menthol gel versus controls in ankle sprainclinicaltrialsgovct2showNCT02100670 (accessed 3February 2015) 2014 [ClinicalTiralsgov NCT02100670]
NCT02290821 unpublished data only
Novartis (Sponsors) A randomized double-blind placebo-controlled parallel group study to evaluate the efficacy andsafety of diclofenac sodium topical gel (DSG) 1 applied
four times daily in subjects with acute blunt soft tissueinjuriescontusions of the limbs clinicaltrialsgovct2showNCT02290821 (accessed 3 February 2015) 2015 [CTGNCT02290821]
Additional references
Anon 2005
Anon Topical analgesics a review of reviews and a bit ofperspective wwwjr2oxacukBandolierExtraforbandoTopextra3pdf 2005
Cook 1995
Cook RJ Sackett DL The number needed to treat aclinically useful measure of treatment effect BMJ 1995310
(6977)452ndash4
Counsell 1994Counsell CE Clarke MJ Slattery J Sandercock PA Themiracle of DICE therapy for acute stroke fact or fictionalproduct of subgroup analysis BMJ 1994309(6970)1677ndash81
Dechartres 2013Dechartres A Trinquart L Boutron I Ravaud P Influenceof trial sample size on treatment effect estimates meta-epidemiological study BMJ 2013346f2304 [DOI101136bmjf2304]
Derry 2012a
Derry S Moore RA Topical capsaicin (low concentration)for chronic neuropathic pain in adults Cochrane Databaseof Systematic Reviews 2012 Issue 9 [DOI 10100214651858CD010111]
Derry 2012bDerry S Moore RA Rabbie R Topical NSAIDs forchronic musculoskeletal pain in adults Cochrane Databaseof Systematic Reviews 2012 Issue 9 [DOI 10100214651858CD007400pub2]
Derry 2013Derry S Sven-Rice A Cole P Tan T Moore RA Topicalcapsaicin (high concentration) for chronic neuropathic painin adults Cochrane Database of Systematic Reviews 2013Issue 2 [DOI 10100214651858CD007393pub3]
Derry 2014
Derry S Matthews PR Wiffen PJ Moore RASalicylate-containing rubefacients for acute and chronicmusculoskeletal pain in adults Cochrane Database ofSystematic Reviews 2014 Issue 11 [DOI 10100214651858CD007403pub3]
Evans 1995Evans JM McMahon AD McGilchrist MM White GMurray FE McDevitt DG et al Topical non-steroidal anti-inflammatory drugs and admission to hospital for uppergastrointestinal bleeding and perforation a record linkagecase-control study BMJ 1995311(6996)22ndash6
FitzGerald 2001
FitzGerald GA Patrono C The coxibs selective inhibitorsof cyclooxygenase-2 New England Journal of Medicine2001345(6)433ndash42
34Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Haroutiunian 2010
Haroutiunian S Drennan DA Lipman AG Topical NSAIDtherapy for musculoskeletal pain Pain Medicine 201011
(4)535ndash49 [DOI 101111j1526-4637201000809x]
Hawkey 1999
Hawkey CJ Cox-2 inhibitors Lancet 1999353(9149)307ndash14
Higgins 2011
Higgins JPT Altman DG Sterne JAC Chapter 8 Assessingrisk of bias in included studies In Higgins JPT GreenS (editors) Cochrane Handbook for Systematic Reviewsof Interventions Version 510 [updated March 2011]The Cochrane Collaboration 2011 Available fromwwwcochrane-handbookorg
Jadad 1996a
Jadad AR Carroll D Moore A McQuay H Developing adatabase of published reports of randomised clinical trials inpain research Pain 199666(2-3)239ndash46
Jadad 1996b
Jadad AR Moore RA Carroll D Jenkinson C ReynoldsDJ Gavaghan DJ et al Assessing the quality of reports ofrandomized clinical trials is blinding necessary ControlledClinical Trials 199617(1)1ndash12
LrsquoAbbeacute 1987
LrsquoAbbeacute KA Detsky AS OrsquoRourke K Meta-analysis inclinical research Annals of Internal Medicine 1987107(2)224ndash33
Makris 2010Makris UE Kohler MJ Fraenkel L Adverse effects oftopical nonsteroidal antiinflammatory drugs in older adultswith osteoarthritis a systematic literature review Journalof Rheumatology 201037(6)1236ndash43 [DOI 103899jrheum090935]
Mason 2004a
Mason L Moore RA Edwards JE Derry S McQuay HJTopical NSAIDs for acute pain a meta-analysis BMCFamily Practice 2004510
Mason 2004bMason L Moore RA Edwards JE Derry S McQuayHJ Topical NSAIDs for chronic musculoskeletal painsystematic review and meta-analysis BMC MusculoskeletalDisorders 2004528
McPherson 2013McPherson ML Cimino NM Topical NSAIDformulations Pain Medicine 201314 Suppl 1S35ndash9[DOI 101111pme12288]
MHRA 2009
Medicines and Healthcare Products Regulatory AgencyTopical ketoprofen reminder on risk of photo-sensitivityreactions Drug Safety Update 2009 June Vol 2 issue 116
Moore 1998aMoore RA Tramegraver MR Carroll D Wiffen PJ McQuayHJ Quantitative systematic review of topically applied non-
steroidal anti-inflammatory drugs BMJ 1998316(7128)333ndash8
Moore 1998bMoore RA Gavaghan D Tramegraver MR Collins SL McQuayHJ Size is everything - large amounts of information areneeded to overcome random effects in estimating directionand magnitude of treatment effects Pain 199878(3)209ndash16
Moore 2003
Moore RA Edwards J Barden J McQuay HJ BandolierrsquosLittle Book of Pain Oxford Oxford University Press 2003[ISBN 0ndash19ndash263247ndash7]
Moore 2006
Moore RA Barden J Derry S McQuay HJ Managingpotential publication bias In McQuay HJ Kalso EMoore RA editor(s) Systematic Reviews in Pain ResearchMethodology Refined Seattle IASP Press 200615ndash23[ISBN 978ndash0ndash931092ndash69ndash5]
Moore 2008aMoore RA Derry S McQuay HJ Topical agents in thetreatment of rheumatic pain Rheumatic Diseases Clinics ofNorth America 200834(2)415ndash32
Moore 2008bMoore RA Barden J Derry S McQuay HJ Managingpotential publication bias In McQuay HJ Kalso EMoore RA editor(s) Systematic Reviews in Pain ResearchMethodology Refined Seattle IASP Press 200815ndash24[ISBN 978ndash0ndash931092ndash69ndash5]
Moore 2013
Moore RA Straube S Aldington D Pain measures andcut-offs - rsquono worse than mild painrsquo as a simple universaloutcome Anaesthesia 201368(4)400ndash12 [DOI 101111anae12148]
Morris 1995
Morris JA Gardner MJ Calculating confidence intervalsfor relative risk odds ratios and standardised ratios andrates In Gardner MJ Altman DG editor(s) Statistics withConfidence - Confidence Intervals and Statistical GuidelinesLondon British Medical Journal 199550ndash63
NICE 2008National Institute for Health and Care ExcellenceOsteoarthritis The care and management of osteoarthritisin adults 2008 wwwniceorgukCG059fullguideline
Nuumlesch 2010Nuumlesch E Trelle S Reichenbach S Rutjes AW TschannenB Altman DG et al Small study effects in meta-analysesof osteoarthritis trials meta-epidemiological study BMJ2010341c3515 [DOI 101136bmjc3515]
PACT 2014
Prescribing and Primary Care team Health and SocialCare Information Centre Prescrption Cost AnalysisEngland 2013 Health and Social Care Information CentreThe NHS Information Centre 2014298ndash301 [ISBN978ndash1ndash78386ndash089ndash0]
35Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Quane 1998
Quane PA Graham GG Zeigler JB Pharmacology ofbenzydamine Inflammopharmacology 19986(2)95ndash107
RevMan 2014The Nordic Cochrane Centre The Cochrane CollaborationReview Manager (RevMan) 53 Copenhagen The NordicCochrane Centre The Cochrane Collaboration 2014
Roth 2012Roth SH Fuller P Pooled safety analysis of diclofenacsodium topical solution 15 (ww) in the treatment ofosteoarthritis in patients aged 75 years or older ClinicalInterventions in Aging 20127127ndash37 [DOI 102147CIAS30884]
Simon 2009
Simon LS Grierson LM Naseer Z Bookman AA ZevShainhouse J Efficacy and safety of topical diclofenaccontaining dimethyl sulfoxide (DMSO) compared withthose of topical placebo DMSO vehicle and oral diclofenacfor knee osteoarthritis Pain 2009143(3)238ndash45 [DOI101016jpain200903008]
Tramer 1997
Tramer MR Reynolds DJ Moore RA McQuay HJ Impact
of covert duplicate publication on meta-analysis a casestudy BMJ 1997315(7109)635ndash40
Tugwell 2004Tugwell PS Wells GA Shainhouse JZ Equivalence studyof a topical diclofenac solution (pennsaid) compared withoral diclofenac in symptomatic treatment of osteoarthritisof the knee a randomized controlled trial Journal ofRheumatology 200431(10)2002ndash12
Zimmerman 1995
Zimmerman J Siguencia J Tsvang E Upper gastrointestinalhemorrhage associated with cutaneous application ofdiclofenac gel American Journal of Gastroenterology 199590(11)2032ndash4 [PUBMED 7485017]
References to other published versions of this review
Massey 2010
Massey T Derry S Moore RA McQuay HJ TopicalNSAIDs for acute pain in adults Cochrane Databaseof Systematic Reviews 2010 Issue 6 [DOI 10100214651858CD007402pub2]
lowast Indicates the major publication for the study
36Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
C H A R A C T E R I S T I C S O F S T U D I E S
Characteristics of included studies [ordered by study ID]
Airaksinen 1993
Methods R DB PC parallel groupsGel applied to the painful area twice daily for 7 daysAssessment at baseline 3 7 days
Participants Minor soft tissue injuries (lt 7 days)N = 56M 45 F 11Age not reportedMean baseline pain at rest 25-26 mm
Interventions Ketoprofen gel 2 x 5 g (125 mg) daily n = 29Placebo gel n = 27Rescue medication paracetamol 500 mgNo other treatment allowed
Outcomes PGE 5-point scale but reported as ldquoimprovedrdquo or ldquosame or worserdquo (responder = ldquoim-provedrdquo)Improvement in pain with movement 100 mm VAS reported as group meanAdverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Unclear risk No mention of early withdrawals ormethod of imputation
Size High risk lt 50 participants per treatment group
37Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Aoki 1984
Methods R DB PC AC parallel groupsGel applied to affected area 3 or 4 times daily with no occlusion for 7 daysAssessment at baseline 3 7 days
Participants Acute orthopaedic trauma (contusion distortion fracture lt 7 days)N = 252 (203 analysed for efficacy)M 98 F 105Age range 8 to 86 years 13 younger than 20 yearsBaseline pain mild in 35Exclusions 23 protocol violations 26 reasons ldquonot relatedrdquo to drug Equally distributedbetween groups
Interventions Piroxicam gel 05 1 g 3 to 4 x daily n = 84Indomethacin gel 1 1 g 3 to 4 x daily n = 84Placebo gel n = 84No other medication or initiation of physical therapy allowed
Outcomes PGE 5-point scale (responder = ldquobetterrdquo and ldquomuch betterrdquo)Adverse eventsWithdrawals and exclusions
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Low risk ldquokey code sealed until end of studyrdquo
Blinding (performance bias and detectionbias)All outcomes
Low risk Gels in ldquoidentical tubesrdquo
Incomplete outcome data (attrition bias)All outcomes
Unclear risk gt 10 withdrawals ldquounrelated to treat-mentrdquo and for ldquoprotocol violationsrdquo Nofurther details but no significant differ-ences between groups
Size Unclear risk 50 to 200 participants per treatment group
38Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Auclair 1989
Methods R DB PC parallel groupsGel massaged into skin over affected heel 3 times daily after cleaning with soap and waterfor up to 21 daysAssessment at baseline 7 21 days
Participants Acute Achilles heel tendinitis (not associated with continuous pain at rest or gt 1 monthhistory)N = 243 (227 analysed for efficacy)MF not reportedMean age 29 yearsBaseline pain ~ 10 had lt 26 mm on palpation of tendon ~ 30 had mild or no painon dorsiflexion of footExclusions failure to meet inclusion criteria major protocol violations failure to takestudy medication for full duration
Interventions Niflumic acid gel 25 3 x 5 g daily n = 117Placebo gel n = 110No other analgesics and anti-inflammatories physiotherapy or supportive measures al-lowed
Outcomes PGE 5-point scale (responder = ldquogoodrdquo or ldquovery goodrdquo)Pain improved or disappeared on dorsiflexionAdverse eventsWithdrawals and exclusions
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Unclear risk 10 excluded for ldquofailing to meet entry cri-teria and protocol violationsrdquo No furtherdetails
Size Unclear risk 50 to 200 participants per treatment group
39Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Billigmann 1996
Methods R DB PC parallel groupsGel applied 3 times daily with rubbingAssessed at baseline 3 5 7 days
Participants Distortion of ankle jointN = 160M and FAge 18+ yearsBaseline pain not reported
Interventions Ibuprofen microgel 5 3 x 10 cm (= 200 mg) daily n = 80Placebo gel n = 80
Outcomes Pain with movement VAS (responder = decreased by 20)Complete remissionAdverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Unclear risk Insufficient data to assess
Size Unclear risk 50 to 200 participants per treatment group
Campbell 1994
Methods R DB PC parallel groupsCream applied 4 times daily for 7 days (up to 14 days optional)Self assessed using daily diary for 7 days and up to 14 days
Participants Acute ankle sprain (lt 24 hours no fracture)N = 100 (51 analysed)M 33 F 18Mean age 29 years
40Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Campbell 1994 (Continued)
Baseline pain at rest gt 35 mm on walking 80 mmExclusions did not return diaries protocol exclusions (25 ibuprofen 24 placebo)
Interventions Ibuprofen cream 5 (Proflex) 4 x 4rdquo (10 cm) daily n = 26Placebo cream n = 25Advised to use rest and regular icing for 48 hours then walking and exerciseRescue medication paracetamol
Outcomes Improvement in walking ability 4-point scale (responder = ldquoimprovementrdquo)Pain on walking 100 mm VAS (mean data)Withdrawals and exclusions
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Low risk Randomisation carried out by sponsorTubes dispensed by hospital pharmacy whoheld the codes
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical creamrdquo
Incomplete outcome data (attrition bias)All outcomes
High risk gt 40 lost to follow-up Approximatelyequal between groups
Size High risk lt 50 participants per treatment arm as anal-ysed
Chatterjee 1977
Methods R DB PC parallel groupsCream applied to site of injury 3 times daily for 6 daysAssessment at baseline 2 6 days
Participants Soft tissue injuries (recent)N = 51MF not reportedAge not reportedBaseline pain on passive movement moderate or severe in all but 3 participants
41Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Chatterjee 1977 (Continued)
Interventions Benzydamine HCl cream 3 3 x daily n = 25Placebo cream n = 25(5 active 6 placebo participants also received ultrasound)No other topical agent allowed
Outcomes Pain on passive movement 4-point scale (responder = ldquoabsentrdquo or ldquoslightrdquo)Tenderness with pressure 4-point scale (responder = ldquoabsentrdquo or ldquoslightrdquo)Adverse eventsWithdrawals and exclusions
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk ldquopredetermined randomised schedulerdquo
Allocation concealment (selection bias) Low risk Sealed copy of schedule held by investiga-tor and duplicate copy kept by clinical trialco-ordinator Looked at only in event ofadverse reaction (not necessary)
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoindistinguishable in appearance andconsistencyrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk 2 lost to follow-up no withdrawals dueto adverse events Responder analysis
Size Unclear risk lt 50 participants per treatment arm
Costantino 2011
Methods R DB PC multicentre parallel groupPlaster applied daily for 7 daysPI assessment daily overall treatment efficacy and tolerability assessed at 3 and 7 days
Participants Grade I or II ankle sprain (lt 48 hours) with lateral external ligament involvementPI on movement ge 50100 oedema ge 20 mm difference between anklesN = 430M 249 F 175 (for analysis)Mean age 35 yearsBaseline PI on movement 72100 (SD 12)
42Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Costantino 2011 (Continued)
Interventions DHEPhep n = 142DHEP n = 146Placebo n = 142Rescue medication paracetamol to maximum 3 g dailyNo other treatments allowed
Outcomes Mean change in PI on movement from baseline to 3 daysMean reduction in oedema at 3 daysTolerability at 3 daysRescue medicationAdverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk Computer-generated
Allocation concealment (selection bias) Unclear risk ldquosealed envelopesrdquo
Blinding (performance bias and detectionbias)All outcomes
Low risk Shape colour size and application methodidentical for all plasters
Incomplete outcome data (attrition bias)All outcomes
Unclear risk Very few withdrawals but used LOCF
Size Unclear risk 50 to 200 participants per treatment arm
Coudreuse 2010
Methods R DB PC multicentre parallel groupPlaster applied daily for 7 daysPI assessed twice daily over 3 days then day 7 Overall treatment efficacy and tolerabilityassessed on days 3 7
Participants Ankle sprain (lt 48 hours) with lateral external ligament involvementPI on movement ge 50100 oedema ge 20 mm difference between anklesN = 240 (233 for analysis)M 148 F 86 6 unknown
Interventions DHEPhep n = 120Placebo n = 120Rescue medication paracetamol to maximum 4 g daily
43Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Coudreuse 2010 (Continued)
No other treatments allowed
Outcomes Global efficacy at 7 days 4-point scale (responder = ldquoexcellentrdquo or ldquogoodrdquo)Mean change in PI on movement at 6 hours and 7 daysMean change in oedema at 3 and 7 daysTolerabilityAdverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk Computer-generated
Allocation concealment (selection bias) Unclear risk ldquosealed envelopesrdquo
Blinding (performance bias and detectionbias)All outcomes
Low risk Appearance and odour identical for all plas-ters
Incomplete outcome data (attrition bias)All outcomes
Low risk ITT analysis lt 5 excluded for missingdata equal between groups
Size Unclear risk 50 to 200 participants per treatment arm
Curioni 1985
Methods R DB PC AC parallel groupsGel rubbed into affected area until absorbed twice daily for 10 daysAssessed at baseline and daily to 10 days
Participants Acute soft tissue injuriesN = 60M 33 F 27Median age 33 yearsBaseline pain not given
Interventions Ibuproxam gel 10 n = 20Ketoprofen gel n = 20Etofenamate gel n = 20
Outcomes PGE 4-point scale (ldquogoodrdquo or ldquoexcellentrdquo)Resolution of symptomsAdverse events
44Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Curioni 1985 (Continued)
Withdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk Medication supplied in identical tubes
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
Diebshlag 1990
Methods R DB PC AC parallel groupsGel applied 3 times daily without occlusion for 14 daysAssessment at baseline 2 3 4 8 15 days
Participants Ankle sprain (lt 24 hours)N = 37M 24 F 13Mean age 28 yearsBaseline pain slight to moderate
Interventions Ketorolac gel 2 3 x 3 g daily n = 13Etofenamate gel 5 3 x 3 g daily n = 12Placebo gel n = 12Rescue medication paracetamolNo other analgesic or anti-inflammatory medication ice packs or physiotherapy allowed
Outcomes Reduction in PI 100 mm VAS and 4-point scale (responder = ldquoimprovedrdquo)Adverse eventsWithdrawals and exclusions
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
45Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Diebshlag 1990 (Continued)
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Low risk ldquoMedication assignment supplied in asealed enveloperdquo Opened only if seriousparticipant event necessitation treatmentdisclosure occurred (not necessary)
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical appearancerdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
Dreiser 1988
Methods R DB PC parallel groupsCream applied 3 times dailyAssessment at baseline 7 days
Participants Acute tendinitis (lt 1 month)N = 64M 35 F 25Mean age 36 yearsBaseline spontaneous pain ge 60 mm
Interventions Ibuprofen cream 5 3 x 4 cm daily n = 32 (3 x 10 cm for large joints)Placebo cream n = 32No other topical systemic or physical treatment allowed
Outcomes PGE scale not reported (responder = ldquoimprovementrdquo or ldquocomplete relief rdquo)Improvement in pain VAS (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35Oxford Validity Score 1016
Risk of bias
Bias Authorsrsquo judgement Support for judgement
46Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Dreiser 1988 (Continued)
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Low risk Missing participants added back usingBOCF
Size High risk lt 50 participants per treatment arm
Dreiser 1989
Methods R DB PC parallel groupsGel applied twice daily to affected area with light massage then covered with standardcompressAssessed at baseline 3 7 days
Participants Uncomplicated recent ankle sprainN = 60M 36 F 24Mean age 33 yearsMean baseline pain 54 mm
Interventions Ketoprofen gel 25 2 x 5 cm daily n = 30Placebo gel n = 30No concomitant therapy other than simple oral analgesia allowed
Outcomes PGE 3-point scale (responder = ldquobetterrdquo)Improvement in pain VAS (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk ldquodrawing lotsrdquo
Allocation concealment (selection bias) Unclear risk Not described
47Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Dreiser 1989 (Continued)
Blinding (performance bias and detectionbias)All outcomes
Low risk Treatments ldquoidentical in every way exceptthat placebo did not contain active princi-plerdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participant per treatment arm
Dreiser 1990
Methods R DB PC parallel groupsGel lightly massaged into skin over affected area 3 times daily then covered with standardcompressAssessed at baseline 3 7 days
Participants Uncomplicated ankle sprain (lt 4 days)N = 60 (59 analysed)M 29 F 29 (not stated for 1 participant)Mean age 33 yearsBaseline pain ge moderately severeExclusions 1 participant had only moderate pain at baseline
Interventions Niflumic acid gel 25 3 x 5 g daily n = 30Placebo gel n = 30Concomitant treatment with systemic NSAIDs local therapies or physiotherapy werenot allowed
Outcomes PGE 4-point scale (responder = ldquocuredrdquo or ldquoimprovedrdquo)Improvement in pain VAS (mean data)Adverse eventsWithdrawals and exclusions
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
48Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Dreiser 1990 (Continued)
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
Dreiser 1994
Methods R DB PC parallel groupsPatch applied twice dailyAssessed at baseline 3 7 days
Participants Traumatic ankle sprain (lt 2 days)N = 131M 84 F 47Mean age 34 yearsBaseline pain ge 50 mm
Interventions Flurbiprofen patch 2 x 40 mg daily n = 65Placebo patch n = 66Rescue medication paracetamol Ice or light restraint allowedExclusions 1 from flurbiprofen group for protocol violation
Outcomes PGE 4-point scale (responder = ldquogoodrdquo or ldquovery goodrdquo)Improvement in pain VAS (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk Placebo patch was ldquonon-medicated (butotherwise identical)rdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
49Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Fioravanti 1999
Methods R DB AC parallel groupsGel lightly massaged into skin 3 times daily and kept dry for 6 to 8 hoursAssessed at baseline 3 10 days
Participants Peri and extra-articular inflammatory diseasesN = 100M 32 F 68Mean age 49 yearsBaseline spontaneous pain ge 40 mm
Interventions DHEP lecithin gel 3 x 5 g (= 65 mg) daily n = 50DHEP gel 3 x 5 g (= 65 mg) daily n = 50
Outcomes PGE 4-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo)Pain on movement meanAdverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size Unclear risk 50 participants per treatment arm
Fujimaki 1985
Methods R DB PC AC parallel groupsGel applied to affected area 3 or 4 times daily with no occlusion for up to 14 daysAssessed at baseline 7 14 days
Participants Muscle pain or inflammation in neck shoulder back chest and upper and lower ex-tremities or a combinationN = 271 (247 analysed)M 97 F 149
50Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Fujimaki 1985 (Continued)
Age lt 20 to 89 yearsBaseline pain mostly mild to moderateExclusions 24 due to protocol violations loss to follow-up
Interventions Piroxicam gel 05 1 g 3 to 4 x daily n = 92Indomethacin gel 1 1 g 3 to 4 x daily n = 90Placebo gel n = 89No concomitant oral or topical analgesic or anti-inflammatory medication allowed Nophysical therapy initiated after start of study
Outcomes PGE 5-point scale (responder = ldquobetterrdquo or ldquomuch betterrdquo)Physician rated improvement 5-point scale (responder = ldquomarked improvementrdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total =45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Low risk Cartons numbered randomly and numbersheld in a key code until study completion
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical tubesrdquo packed in numbered car-ton Gel bases slightly different in appear-ance so dispensing physician did not haveaccess to them
Incomplete outcome data (attrition bias)All outcomes
Unclear risk ~ 8 excluded from analysis for unknownreasons or lost to or inadequate follow-upApproximately equal between groups
Size Unclear risk 50 to 200 participants per treatment arm
Gallacchi 1990
Methods R DB AC parallel groupsGel applied to affected area 4 times daily with light massage for 14 daysAssessment at baseline 7 14 days
Participants Painful inflammatory conditionsN = 50M 20 F 30Mean age 50 years
51Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Gallacchi 1990 (Continued)
Baseline pain ge moderate severity
Interventions Diclofenac gel 1 2 g 4 x daily n = 25 (Flector)Diclofenac sodium 1 2 g 4 x daily n = 25 (Voltaren Emugel)No other medication that could interfere with test drugs allowed
Outcomes PGE 5-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo)Improvement in pain on pressure 4-point scale (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
Gonzaacutelez de Vega 2013
Methods R DB (for gels) SB (for ointment) AC multicentre parallel groups2 g (~ 6 cm of gel) applied over injured area x 3 daily for 14 daysAssessments at 0 4 7 14 days of treatment and day 42 follow-up
Participants Grade I or II ankle sprain (lt 24 hours) with lateral external ligament involvement PIon weight bearing ge 30100N = 449M = 308 F = 112 (for analysis)Mean age 28 years
Interventions Traumeel gel 3 x 2 g daily n = 140Traumeel ointment n = 143 (not analysed in this review)Diclofenac gel 1 3 x 2 g daily n = 137Rescue medication paracetamol to maximum 2 g daily
52Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Gonzaacutelez de Vega 2013 (Continued)
Outcomes Pain-free on day 7Global efficacy 5-point scale (responder = ldquogoodrdquo and ldquovery goodrdquo) on day 14Normal function on day 14 yes or no (responder = ldquoyesrdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 (gels) W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk Computer-generated
Allocation concealment (selection bias) Low risk Central allocation kits assigned in orderreceived and used envelopes (no furtherdetails)
Blinding (performance bias and detectionbias)All outcomes
Low risk For gel comparison ldquoidentical containersrdquo
Incomplete outcome data (attrition bias)All outcomes
Unclear risk Used LOCF Most withdrawals due to earlyrecovery (within 14 days) approximatelyequal between groups
Size Unclear risk 50 to 200 participants per treatment arm
Governali 1995
Methods R DB AC parallel groupsGel or cream applied 3 times daily for up to 14 daysAssessed at baseline 7 14 days
Participants Soft tissue injuries + 2 fracturesN = 30M = 21 F = 9Median age 38 yearsMean baseline pain on movement moderate to severe (28 scale 0 to 4)
Interventions Ketoprofen gel 5 3 x 2 to 3 g daily n = 15Ketoprofen cream 1 3 x 2 to 3 g daily n = 15
Outcomes PGE 5-point scale (responder = ldquogoodrdquo and ldquoexcellentrdquo)Adverse eventsWithdrawals
53Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Governali 1995 (Continued)
Notes Oxford Quality Score R1 DB1 W1 Total =35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Treatments were given in identical tubesand measurements made by blinded ob-servers but one was a cream and the othera gel
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
Gualdi 1987
Methods R DB AC parallel groupsGel applied twice daily for 10 daysAssessed at baseline 4 7 10 days
Participants Soft tissue injuriesN = 60M = 37 F = 23Mean age 32 years (range 13 to 78)Mean baseline pain on movement moderate to severe (22 scale 0 to 3)
Interventions Flunoxaprofen gel 2 x 3 to 5 cm daily n = 30Ketoprofen gel 2 x 3 to 5 cm daily n = 30
Outcomes Improvement in pain on pressure (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
54Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Gualdi 1987 (Continued)
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Unclear risk Insufficient data
Size High risk lt 50 participants per treatment arm
Haig 1986
Methods R DB PC parallel groupsCream applied lightly to affected area 6 times daily for 6 daysAssessed at baseline 2 4 6 days
Participants Soft tissue injuries (lt 24 hours)N = 43MF not reportedAge not reportedBaseline pain not reported
Interventions Benzydamine cream 3 6 x daily n = 21Placebo cream n = 22
Outcomes Pain on movement 4-point scale (responder = ldquoimprovedrdquo)Adverse events
Notes Oxford Quality Score R1 DB2 W0 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not reported
Allocation concealment (selection bias) Unclear risk Not reported
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquomatching placebordquo
55Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Haig 1986 (Continued)
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants apparently included inanalysis
Size High risk lt 50 participants per treatment arm
Hoffmann 2012
Methods R DB PC multicentre parallel groupPlaster applied x 1 daily (minimum of 20 consecutive hours of application per day) for14 daysPI assessed daily overall efficacy assessed at 7 and 14 days
Participants Unilateral mild to moderate muscle contusion of upper or lower limb (lt 72 hours) PIge 50100 superficial haematoma le 10 x 14 cm at injured siteN = 354M 126 F 228Mean age 39 years
Interventions DHEPhep plaster x 1 daily n = 121DHEP plaster x 1 daily n = 115Placebo plaster n = 118Rescue medication paracetamol 500 mg (no limit reported)
Outcomes PGE 5-point scale (responder = excellent or good) at 14 daysMean reduction in PI at 3 and 8 daysRescue medicationAdverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Method not reported
Allocation concealment (selection bias) Unclear risk Method not reported ldquoMedication packedand labelled in order to render all partici-pants and personnel fully blinded to treat-ment administeredrdquo
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoIndistinguishable regarding appearanceshape colour size and odourrdquo
56Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Hoffmann 2012 (Continued)
Incomplete outcome data (attrition bias)All outcomes
Unclear risk No responder analysis reported and impu-tation method not mentioned
Size Unclear risk 50 to 200 participants per treatment arm
Hofman 2000
Methods R DB AC parallel groupsGel applied to affected region 4 times daily with gentle massageAssessed at baseline 8 days in clinic and daily diary
Participants Soft tissue articular pain (le 15 days)N = 142M 19 F 123Mean age 57 yearsMean baseline PI moderate to severe
Interventions Diclofenac sodium gel 1 4 x 2 cm daily n = 69Lysine clonixinate gel 5 4 x 2 cm daily n = 73(2 cm = 225 mg)No other analgesic local treatment (including immobilisation bandaging) or acupunc-tureRescue mediation allowed after 2 applications if needed
Outcomes PGE 3-point scale (ldquogoodrdquo)PI participant diary (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk Diclofenac gel repackaged to maintain DBwith lysine clonixinate gel Minor differ-ences between gels only apparent when di-rectly compared
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
57Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Hofman 2000 (Continued)
Size Unclear risk 50 to 200 participants per treatment arm
Hosie 1993
Methods R DB AC parallel groupsFoam (approximately the size of a golf ball) applied and 1 tablet taken 3 times daily for7 days and up to 14 daysAssessed at baseline 7 14 (if necessary) days
Participants Acute lower back injury (lt 1 month)N = 287 (261 analysed for efficacy)M 151 F 136Mean age 37 years (range 18 to 63)Most participants had moderate to severe pain on movement one had noneExclusions 25 lost to follow up one assessed at 14 days but not 7 days
Interventions Felbinac foam 3 3 x 2 g daily + placebo tablets 3 x 1 daily n = 140 (127 analysed forefficacy)Ibuprofen tablets 3 x 400 mg daily + placebo foam 3 x 2 g daily n = 147 (134 analysedfor efficacy but one had no pain at baseline)No other oral injectable or topical analgesic or anti-inflammatory medication Ongoingphysiotherapy to continue without change
Outcomes Pain on movement 5-point scale (responder = ldquononerdquo or ldquomildrdquo)Spontaneous pain 5-point scale (responder = ldquononerdquo or ldquomildrdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquodouble dummyrdquo
Incomplete outcome data (attrition bias)All outcomes
Unclear risk Imputation method not mentioned
Size Unclear risk 50 to 200 participants per treatment arm
58Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Jenoure 1997
Methods R DB PC parallel groupsPlaster applied to skin over affected area twice daily and kept in place with an elasticbandageAssessed at baseline 7 14 days and after further 14 days without treatment
Participants Humero-radial epicondyl pain (tendinopathic) - nearly all tennis elbowN = 85M 54 F 31Mean age 45 yearsBaseline pain ldquomildrdquo in ~ 10 of placebo group and 29 of active group
Interventions DHEP plaster (Tissugel) x 2 daily n = 44Placebo plaster x 2 daily n = 41
Outcomes Pain on pressure 5-point scale (responder = ldquononerdquo or ldquomildrdquo)Spontaneous pain 5-point scale (responder = ldquono painrdquo)Adverse events
Notes Oxford Quality Score R1 DB2 W0 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not reported
Allocation concealment (selection bias) Unclear risk Not reported
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical characteristicsrdquo
Incomplete outcome data (attrition bias)All outcomes
Unclear risk No useable data
Size High risk lt 50 participants per treatment arm
Joussellin 2003
Methods R DB PC parallel groupsPlaster applied to skin over affected area once dailyAssessed at baseline 1 2 3 7 days
Participants Ankle sprain (lt 48 hours)N = 134M 72 F 62Age range 18 to 65 years
59Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Joussellin 2003 (Continued)
Baseline spontaneous pain ge 50 mm
Interventions DHEP plaster (Flector Tissugel 1) x 1 daily n = 68Placebo plaster x 1 daily n = 66Rescue medication paracetamol
Outcomes PGE 4-point scale (responder = ldquoexcellentrdquo)Pain on movement VAS (mean)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoIdentical in size appearance and usedsame formula as active patch without ac-tive ingredientrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in responder anal-ysis
Size Unclear risk 50 to 200 participants per treatment arm
Julien 1989
Methods R DB PC parallel groupsGel applied to affected area twice daily with light massageAssessed at baseline 3 7 days in clinic and daily diary
Participants TendinitisN = 60M 29 F 31Mean age 41 yearsBaseline pain gt 50 mm
Interventions Ketoprofen gel 25 2 x 5 cm (= 50 mg) daily n = 30Placebo gel 2 x 5 cm daily n = 30No concomitant therapy other than simple analgesia
60Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Julien 1989 (Continued)
Outcomes PGE 4-point scale (responder = ldquoimprovedrdquo or ldquorecoveredrdquo)Pain on movement 4-point scale (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB1 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk Random numbers table
Allocation concealment (selection bias) Unclear risk Randomisation code supplied by Menarinilaboratories remote from allocation
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in responder anal-ysis
Size High risk lt 50 participants per treatment arm
Klainguti 2010
Methods R DB PC multicentre parallel groupPlaster applied x 1 daily (minimum 20 consecutive hours of application daily) for 10daysPI assessed daily Overall treatment efficacy and tolerability assessed at days 3 and 7
Participants Unilateral mild to moderate muscle contusion or strain of upper or lower limb (lt 72hours) superficial haematoma le 140 cm2 PI ge 40100N = 185M 90 F 95Mean age 39 years
Interventions DHEPhep plaster x 1 daily n = 65DHEP plaster x 1 daily n = 61Placebo plaster x 1 daily n = 59
Outcomes Overall treatment efficacy at 3 days 5-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo)Resolution of haematoma at 10 days yes or noRescue medicationAdverse eventsWithdrawals
61Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Klainguti 2010 (Continued)
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk Computer-generated
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoIdentical in size appearance and odourrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk Withdrawals lt 5 All participants in-cluded in analysis
Size Unclear risk 50 to 200 participants per treatment arm
Kockelbergh 1985
Methods R DB PC parallel groupsGel applied twice dailyAssessed at baseline 3 7 days
Participants Acute soft tissue trauma (lt 24 hours)N = 74M 60 F 14Mean age 27 yearsBaseline pain gt 65 mm
Interventions Ketoprofen gel 25 2 x 5 cm (= 15 mg) daily n = 38Placebo gel 2 x 5 cm daily n = 36No concomitant treatmentRescue medication glafenine
Outcomes PGE 3-point scale (responder = ldquogoodrdquo)Spontaneous pain 100 mm VAS (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
62Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Kockelbergh 1985 (Continued)
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in responder anal-ysis
Size High risk lt 50 participants per treatment arm
Kuehl 2011
Methods R DB PC multicentre parallel groupsPlaster applied x 2 daily for 14 days or until resolutionPI assessed daily Overall treatment efficacy assessed at 14 days
Participants Mild or moderate sprain strain or contusion (lt 7 days) PI ge 510N = 418M 206 F 212Mean age 39 years
Interventions DHEP patch x 2 daily n = 207Placebo patch x 2 daily n = 211Rescue medication no analgesic allowed (or icewrapping) use = discontinuation
Outcomes Resolution of injury (VAS le 210)Overall tolerability 5-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk Computer-generated
Allocation concealment (selection bias) Unclear risk Not described
63Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Kuehl 2011 (Continued)
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical in appearance with same contentexcept diclofenacrdquo
Incomplete outcome data (attrition bias)All outcomes
High risk Used LOCF high withdrawal rate (43)Did not report all efficacy outcomes mea-sured
Size Low risk gt 200 participants per treatment arm
Li 2013
Methods R DB PC multicentre parallel groupsPlaster applied x 2 daily for 7 daysPI assessed daily overall treatment efficacy and tolerability assessed at 7 days
Participants Mild or moderate ankle or knee sprain muscle strain or contusion (lt 72 hours) PI ge
50100N = 384M 144 F 240Mean age 42 years
Interventions DHEP plaster x 2 daily n = 192Placebo plaster x 2 daily n = 192Rescue medication paracetamol to maximum 2 g daily
Outcomes ge 50 reduction in PI at 7 daysOverall treatment efficacy 5-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo)Overall tolerability 5-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo)Rescue medicationAdverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk Computer-generated
Allocation concealment (selection bias) Unclear risk Method not fully described ldquosealed en-velopesrdquo ldquoconcealed from investigators andparticipantsrdquo
64Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Li 2013 (Continued)
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical in texture size color and odorrdquo
Incomplete outcome data (attrition bias)All outcomes
Unclear risk Withdrawals 2 Methods states LOCFbut appears to describe BOCF for with-drawals
Size Unclear risk 50 to 200 participants per treatment arm
Linde 1985
Methods R DB PC parallel groupsCream applied x 3 daily for 5 days with elastic support for the first 3 daysAssessed at baseline 4 8 days
Participants Sprained ankle (lt 24 hours)N = 100M 58 F 42Mean age 28 yearsBaseline pain all participants had ldquowalking painrdquo
Interventions Benzydamine 3 cream x 3 daily n = 50Placebo gel x 3 daily n = 50
Outcomes Pain on movement responder = ldquofree of walking painrdquoAdverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35Paper describes a benzydamine cream and a placebo gel
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described Paper describes a benzy-damine cream and a placebo gel
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in responder anal-ysis
65Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Linde 1985 (Continued)
Size Unclear risk 50 participants per treatment arm
Machen 2002
Methods R DB PC parallel groupsGel gently (ldquominimal rubrdquo not vigorously) massaged into skin over affected site untilabsorbed 3 times daily until symptoms disappeared or for maximum of 7 daysAssessment at baseline and once daily using diary cards to 7 days
Participants Soft tissue injury (lt 2 weeks and untreated)N = 85 (81 analysed)M 42 F 39Mean age 41 yearsBaseline pain gt 50 mm4 placebo participants lost to follow-up
Interventions Ibuprofen gel 5 x 3 daily n = 40Placebo gel x 3 daily n = 41Initiation of other medication or physiotherapy not allowed during study
Outcomes PGE 5-point scale (responder = ldquomarked improvementrdquo or ldquocomplete clearancerdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk Gels had similar physical characteristicsand were supplied in identical tubes
Incomplete outcome data (attrition bias)All outcomes
Low risk lt 5 missing participants Remaining par-ticipants included in responder analysis
Size High risk lt 50 participants per treatment arm
66Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Mahler 2003
Methods R DB AC parallel groupsGel applied with gentle massage to affected area 3 times daily without occlusion for 10daysAssessed at baseline 3 10 days in clinic and daily patient diary
Participants First-degree ankle or knee sprains first-degree muscle strains and mild-to-moderatecontusionsn = 100M 69 F 31Mean age 32 yearsMean baseline pain with activity ge 65 mm
Interventions DHEP lethicin gel 3 x 5 g (= 65 mg) daily n = 52DHEP gel 3 x 5 g (= 65 mg) daily n = 48All participants treated with ice at site of inflammation for first 48 hours but no immo-bilisation allowedRescue medication paracetamol 500 mg if strictly necessary
Outcomes PGE 4-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo)Pain on movement 100-mm VAS (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 5
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk ldquoComputer-generated randomization listrdquo
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk Pharmaceutically inert colouring agentsadded to reference formulation so that gelswere indistinguishable
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in responder anal-ysis
Size High risk lt 50 participants in each treatment arm
67Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Maziegraveres 2005a
Methods R DB PC parallel groupsNew patch applied directly to skin over painful area each morningAssessed at baseline 3 7 14 days
Participants Symptomatic tendonitis in upper or lower limbs not requiring surgery (le 15 days)N = 172M 72 F 100Mean age 46 yearsBaseline pain with activity ge 40 mm
Interventions Ketoprofen patch 100 mg x 1 daily n = 87Placebo patch x 1 daily n = 85No analgesic or steroid by any route or other topical medication or physical therapyallowedRescue medication permitted but not within 12 hours of assessment
Outcomes PGE 4-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk ldquocomputer generated global randomizationcoderdquo
Allocation concealment (selection bias) Low risk ldquoThe randomization list and code en-velopes were prepared by the company ap-pointed for clinical supplies packaging Therandom code was disclosed only after studycompletion and database closurerdquo
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquothe same indistinguishable patch with noingredientrdquo
Incomplete outcome data (attrition bias)All outcomes
High risk LOCF withdrawals 12 All participantsincluded in analysis
Size Unclear risk 50 to 200 participants per treatment arm
68Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Maziegraveres 2005b
Methods R DB PC parallel groupsNew patch applied directly to skin over painful area each morningAssessed at baseline 3 7 14 days
Participants Painful benign ankle sprain (le 48 hours)N = 163M 83 F 80Mean age 37 yearsBaseline spontaneous pain ge 50 mm
Interventions Ketoprofen patch 100 mg x 1 daily n = 81Placebo patch x 1 daily n = 82No analgesic or steroid by any route or other topical medication or physical therapyallowedRescue medication permitted but not within 12 hours of assessment
Outcomes PGE 4-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk ldquocomputer-generated global randomiza-tion coderdquo
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk The same transdermal patch with no activeingredient
Incomplete outcome data (attrition bias)All outcomes
Unclear risk LOCF withdrawals 5 to 6 All partic-ipants included in analysis
Size Unclear risk 50 to 200 participants per treatment arm
McLatchie 1989
Methods R DB PC parallel groupsGel applied to injured site 3 times daily for 7 daysAssessment at baseline 4 7 days at clinic daily patient diary
Participants Acute soft tissue injury (lt 48 hours)N = 231
69Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
McLatchie 1989 (Continued)
M 143 F 88Mean age 33 yearsBaseline pain moderate to severe
Interventions Felbinac gel 3 3 x 3 cm daily n = 118Placebo gel 3 x 3 cm daily n = 113Rescue medication paracetamol
Outcomes Patient diary mean changeAdverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquotubes identical in all aspectsrdquo
Incomplete outcome data (attrition bias)All outcomes
Unclear risk No usable efficacy data
Size Unclear risk 50 to 200 participants per treatment arm
Morris 1991
Methods R DB PC parallel groupsGel applied to site of injury 3 times daily for 7 daysAssessed at baseline 7 days at clinic and daily patient diary
Participants Acute soft tissue injury (lt 3 days)N = 100 (84 analysed for efficacy)M 70 F 14Mean age 25 yearsBaseline pain moderate to severeExclusions 1 participant in placebo group lost to follow-up 15 protocol violations
Interventions Felbinac gel 3 3 x 1 cm daily n = 41Placebo gel n = 43Ice joint immobilisation bandaging and compression allowed
70Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Morris 1991 (Continued)
No concomitant oral NSAID occlusive dressing physiotherapy or liniments allowedRescue medication paracetamol
Outcomes PGE 5-point scale (responder = ldquogoodrdquo and ldquovery goodrdquo)Change in PI patient diary 10 cm VAS (mean data)Adverse eventsWithdrawals and exclusions
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Low risk ldquoRandomisation was undertaken at theproduction facility and a sealed copy of thelist supplied to the investigator for refer-ence only in defined circumstancesrdquo
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical tubes and outer boxesrdquo ldquoplacebowas a similarly constituted gelrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All relevant participants included in analy-sis
Size High risk lt 50 participants per treatment arm
NCT01255423
Methods R DB PC parallel group studyGel applied 4 x daily
Participants Acute lateral ankle sprain Grade I to II le 12 hoursN = 206Mean (plusmn SD) age 31 years (plusmn 13)M 87 F 119Baseline PI not reported
Interventions Diclofenac sodium gel 1 x 4 daily n = 104Placebo gel n = 102
Outcomes Mean PI on movementAdverse events
71Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT01255423 (Continued)
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Method of blinding not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants accounted for
Size Unclear risk 50 to 200 participants per treatment arm
NCT01272934
Methods R DB PC parallel group studyGel applied 4 x daily
Participants Acute lateral ankle sprain Grade I to II le 12 hoursN = 205M 101 F 104Mean (plusmn SD) age 32 years (plusmn 11)
Baseline PI not reported
Interventions Diclofenac sodium gel 1 x 4 daily n = 102Placebo gel x 4 daily n = 103
Outcomes Mean PI on movementAdverse events
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
72Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT01272934 (Continued)
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Method of blinding not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants accounted for
Size Unclear risk 50 to 200 participants per treatment arm
NCT01272947
Methods R DB PC parallel groupsGel applied 4 x daily
Participants Acute blunt soft tissue injuriescontusions of the limbs lt 3 hoursN = 204M 101 F 103Mean (plusmn SD) age 30 years (plusmn 11)
Baseline PI not reported
Interventions Diclofenac sodium gel 1 x 4 daily n = 104Placebo gel x 4 daily n = 100
Outcomes Mean PI on movementAdverse events
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Method of blinding not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants accounted for
73Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT01272947 (Continued)
Size Unclear risk 50 to 200 participants per treatment arm
Noret 1987
Methods R DB PC parallel groupsGel applied twice daily for 7 daysAssessment at baseline 3 8 days
Participants Minor sports injuries (lt 24 hours)N = 98 (93 analysed)M 71 F 27Mean age 29 yearsBaseline pain gt 60 mm
Interventions Ketoprofen gel 25 2 x 5 cm daily (= 15 mg) n = 48Placebo gel n = 45No other treatment given
Outcomes PGE 4-point scale (responder = ldquogoodrdquo and ldquoexcellentrdquo)Spontaneous pain 100 mm VAS (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Low risk ldquoallocated according to a randomization listand a corresponding code in a sealed enve-loperdquo
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
74Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Parrini 1992
Methods R DB PC parallel groupsFoam (the size of a walnut or a 1-second spray) applied with massage 3 x daily for 7days
Participants Articular trauma strains distortionsN = 169M 94 F 75Mean age 37 yearsMean baseline pain on movement 31 (scale 1 to 4)
Interventions Ketoprofen foam 15 3 x 2 g (= 600 mg) daily n = 83Placebo foam 3 x 2 g n = 86
Outcomes Pain on movement 4-point scale (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB1 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk ldquopatients were randomised according to themethod of random numbersrdquo [translated]
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Unclear risk No usable efficacy data
Size Unclear risk 50 to 200 participants per treatment arm
Picchio 1981
Methods R DB AC parallel groupsCream applied with slight massage until completely absorbed 3 times daily for up to 16daysAssessed at baseline 4 8 12 16 days
Participants Acute sports injuriesN = 40M 24 F 16Mean age 22 years (range 12 to 46)Most participants had mild to moderate baseline pain (12 and 9 with slight pain on
75Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Picchio 1981 (Continued)
movement)
Interventions Ibuprofen gel 10 3 x daily n = 20Ketoprofen gel 1 3 x daily n = 20
Outcomes Pain on movement (responder = ldquononerdquo)Adverse events
Notes Oxford Quality Score R1 DB2 W0 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk ldquotubes were identical in appearancerdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
Predel 2004
Methods R DB PC parallel groupsNew patch applied to injured area twice daily for 7 days Contact of patch with humidityor water to be avoidedAssessment at baseline 3 7 days
Participants Traumatic blunt soft tissue injuries (lt 3 hours no treatment)N = 120M 73 F 47Mean age 32 yearsBaseline pain gt 60 mm
Interventions Diclofenac sodium patch 2 x daily (140 mg per patch) n = 60Placebo patch 2 x daily n = 60NSAIDs analgesics psychotropic agents other topical preparations and bandages notallowed
Outcomes PGE 4-point scale (responder = rdquogoodldquo and rdquoexcellentldquo)Pain on movement 10 cm VAS (mean data)
76Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Predel 2004 (Continued)
Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk rdquocomputer generated block randomisationlist
Allocation concealment (selection bias) Low risk An independent statistician produced ran-domisation list and an independent con-tract research organisation packaged med-ication according to list Nobody else hadaccess to the randomisation list until thedatabase was closed
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoThe placebo patch was visually indistin-guishable from the active patchrdquo To avoidunblinding due to different smell anystudy nurse involved with medication wasnot involved in outcome assessment
Incomplete outcome data (attrition bias)All outcomes
Low risk Withdrawals lt 1 All participants in-cluded in analysis
Size Unclear risk 50 to 200 participants per treatment arm
Predel 2012
Methods R DB PC parallel groupsTreatment for 7 daysAssessed at baseline 1 3 5 and 8 days
Participants Grade I or II ankle sprain with lateral external ligament involvement lt 12 hoursPI on movement ge 50100N = 242M 152 F 90Mean age 32 years
Mean baseline pain on movement = 75100
Interventions Diclofenac gel (Voltaren Emulgel 232) 2 x 5 cm + 1 x 5 cm placebo gel daily n = 80Diclofenac gel (Voltaren Emulgel 232) 3 x 5 cm daily n = 80Placebo gel 3 x 5 cm daily n = 82Rescue medication paracetamol (to maximum 2 g daily)
77Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Predel 2012 (Continued)
No ice or bandages after randomisation
Outcomes ge 50 reduction in PI on movement at 5 daysPGE 5-point scale (responder = ldquogoodrdquo or rsquovery goodldquo)Patient satisfaction 5-point scale (responder = rdquogoodldquo or rsquovery goodrdquo or ldquoexcellentrdquo)Rescue medicationAdverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical in composition appearance tex-ture and smellrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk LOCF but withdrawals lt 3 in all treat-ment arms
Size Unclear risk 50 to 200 participants per treatment arm
Predel 2013a
Methods R DB PC parallel groupsTreatment for 14 daysAssessed at baseline 3 7 10 and 14 days (plusmn 1 day)
Participants Uncomplicated one-sided ankle sprain with swelling ge 12 mm (2 to 18 hours)PI on movement tenderness joint mobility (scale 0 to 3) summed as ge 5 and le 7N = 232M 126 F 106Mean age 29 years
Interventions Diclofenac 4 spray gel 3 x 4 to 5 sprays daily (96 to 120 mg diclofenac sodium) n =118Placebo spray gel 3 x 4 to 5 sprays daily n = 114Rescue medication paracetamol 500 mg (maximum 10 tablets per week)No ice or bandaging allowed
78Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Predel 2013a (Continued)
Outcomes None or slight PI on movement at 3 and 7 daysPGE 4-point scale (responder = ldquovery goodrdquo at 14 days)gt 50 reduction in swelling at 10 daysAdverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk ldquoplacebo (vehicle only no active ingredi-ent)rdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk LOCF for full or early cure Withdrawalslt 2
Size Unclear risk 50 to 200 participants per treatment arm
Predel 2013b
Methods R DB PC parallel groups2 g gel applied with fingertips over affected area and massaged in for 1 minute Treatmentfor 5 daysAssessed at baseline 2 3 5 days
Participants Uncomplicated neck pain originating from cervical joints and accompanying soft tissues(ge 12 hours but lt 3 months) PI ge 50100N = 72M 33 F 39Mean age 34 years
Interventions Diclofenac gel (Voltaren Emulgel) 4 x 2 g daily n = 36Placebo gel 4 x 2 g daily n = 36Rescue medication paracetamol up to 2 g dailyNo concomitant therapies allowed
Outcomes ge 50 decrease in PI on movement after 48 hoursPGE 5-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo at 5 daysAdverse eventsWithdrawals
79Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Predel 2013b (Continued)
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk Sponsor ldquoproduced randomisation listrdquo
Allocation concealment (selection bias) Low risk Automated remote system
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical in packaging labelling scheduleof administration appearance and odourrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk No missing data
Size High risk lt 50 participants per treatment arm
Ramesh 1983
Methods R DB PC parallel groupsCream applied to painful area and rubbed into skin over a large area for up to 10 daysAssessment at baseline 3 7 10 days
Participants Strains sprains contusions compressionsN = 80M 42 F 38Age 11 to 81 yearsBaseline pain 5 ibuprofen 2 placebo participants had no or slight pain
Interventions Ibuprofen cream 5 3 to 4 x 5 to 10 cm daily n = 40Placebo cream 3 to 4 x 5 to 10 cm daily n = 40Adjuvant therapy was not administered
Outcomes Pain on movement 4-point scale (responder = ldquononerdquo or ldquoslightrdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
80Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Ramesh 1983 (Continued)
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Low risk Randomisation key in sealed envelopeavailable for emergencies but opened onlyafter completion
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical appearance and odourrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
Rowbotham 2003
Methods R DB PC parallel groupsNew patches applied to the affected painful area for 12 consecutive hours twice dailyfor up to 14 daysAssessed at baseline 14 days in clinic and daily patient diary
Participants Minor sports injuries (sprains sprains contusions lt 72 hours)N = 372M 253 F 119Mean age 33 yearsBaseline pain at rest ge 510
Interventions DHEP patch (Flector Tissuegel) 2 x daily (equivalent to diclofenac sodium 140 mg perpatch) n = 191Placebo patch 2 x daily n = 181
Outcomes PGE 5-point scale (responder = ldquogoodrdquo and ldquoexcellentrdquo)Pain resolved lt moderate for 2 daysSpontaneous pain 10 cm VAS (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
81Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Rowbotham 2003 (Continued)
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk ldquoSystegraveme identiquerdquo without diclofenac
Incomplete outcome data (attrition bias)All outcomes
Unclear risk No information about imputation All par-ticipants included in analysis
Size Unclear risk 50 to 200 participants per treatment arm
Russell 1991
Methods R DB PC parallel groupsAffected area washed with soap and water and dried then gel applied and carefullyrubbed into skin 4 times daily for at least 7 daysAssessed at baseline 4 8 15 (if necessary) days at clinic and daily patient diary
Participants Acute soft tissue injuries (recent not recurrent)N = 214 (200 analysed)M 95 F 105Mean age 40 yearsBaseline pain gt 65 mm
Interventions Piroxicam gel 05 4 x 5 mg daily n = 100Placebo gel n = 100No other NSAIDs or analgesic drugs including liniments containing salicylates allowedAncillary therapy at the discretion of the investigator
Outcomes PGE 4-point scale (responder = ldquogoodrdquo and ldquoexcellentrdquo)Spontaneous pain mean reductionAdverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk ldquocomputer generated randomization coderdquo
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical base formulationrdquo
82Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Russell 1991 (Continued)
Incomplete outcome data (attrition bias)All outcomes
High risk High withdrawal rate (8 with piroxicam50 with placebo) No information aboutimputation
Size Unclear risk 50 to 100 participants per treatment group
Saillant 1998
Methods R DB PC parallel groupsTreatment for 7 daysAssessed at baseline 1 2 3 and 7 days
Participants Ankle sprain (lt 48 hours)N = 140M 72 F 62Age 18 to 65 yearsBaseline spontaneous pain ge 50 mm
Interventions (1) DHEP plaster (Flector Tissugel 1) 1 x daily n = 70(2) Placebo plaster 1 x daily n = 70Rescue medication paracetamol (to maximum 3 g daily)Ice allowed
Outcomes ge 30 decrease in PI at 7 daysPGE 4-point scale (responder = ldquoexcellentrdquo)No or low pain on passive stretchSingle foot leaning OK without painWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk Identical in size appearance and used sameformula as active patch without active in-gredient
Incomplete outcome data (attrition bias)All outcomes
Unclear risk Used LOCF for primary outcome With-drawals lt 7 and equal between groupsAll participants included in analysis
83Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Saillant 1998 (Continued)
Size Unclear risk 50 to 200 participants per treatment arm
Sanguinetti 1989
Methods R DB PC parallel groupsGel applied 3 times daily for 7 consecutive daysAssessment at baseline 7 days
Participants Soft tissue trauma (lt 48 hours)N = 82M = 47 F = 35Mean age 34 yearsBaseline pain moderate to severe
Interventions Felbinac gel 3 3 x daily n = 42Placebo gel 3 x daily n = 40No other NSAID steroid other topical application allowedRescue medication paracetamol felbinac is an active metabolite of the NSAID fenbufen
Outcomes PGE scale not reported (responder = ldquogoodrdquo and ldquovery goodrdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoindistinguishable in appearance colour orodourrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
84Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Sinniger 1981
Methods R DB PC parallel groupsCream applied 2 to 3 times daily with gentle massage or if massage not possible (toopainful) with protective dressingAssessment at baseline 5 10 days
Participants Minor soft tissue injuriesN = 20M 11 F 9Mean age 40 yearsBaseline pain not reported
Interventions Fentiazac cream 5 2 to 3 x daily n = 10Placebo cream 2 to 3 x daily n = 10All participants told to restNo other local and systemic treatments allowedRescue medication analgesic if actually needed
Outcomes Pain relief scale not reported (responder = total pain relief ) improvement in pain on movement pain scale not reported (mean data)Adverse events
Notes Oxford Quality Score R1 DB1 W0 Total = 25
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
85Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Spacca 2005
Methods R DB PC parallel groupsGel applied 3 times daily with gentle massage until complete absorption for up to 10daysAssessment at baseline 10 days in clinic and daily patient diary
Participants Shoulder periarthritis or lateral epicondylitis (lt 5 days)N = 155M 74 F 81Mean age 51 yearsBaseline pain with activity gt 70 mm
Interventions DHEP lecithin gel (Effigel) 3 x 5 g daily n = 79Placebo gel 3 x 5 g daily n = 76Rescue medication (paracetamol) allowed if pain unbearableNo other analgesic or anti-inflammatory drug allowed
Outcomes Improvement in pain 100 mm VAS (mean data)Adverse events
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Unclear risk No usable efficacy data
Size Unclear risk 50 to 200 participants per treatment arm
Sugioka 1984
Methods R DB AC parallel groupsGel applied to affected area 3 to 4 times daily without occlusion for 14 daysAssessed at baseline 7 14 days
Participants Non-traumatic diseases of muscle or tendonN = 366 (340 analysed for efficacy)M 115 F 202 (completers)
86Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Sugioka 1984 (Continued)
Age range 12 to 84 years (most 30 to 70)Baseline pain on movement ldquononerdquo or ldquomildrdquo in about 13 of participantsExclusions for protocol violations 8 piroxicam 18 indomethacin
Interventions Piroxicam gel 05 3 to 4 x 1 g daily n = 183Indomethacin gel 1 3 to 4 x 1 g daily n = 183No concomitant anti-inflammatory or analgesic drug including steroids or initiationof physical therapy allowed
Outcomes PGE 5-point scale (responder = ldquobetterrdquo or ldquomuch betterrdquo)Pain on movement 4-point scale (responder = ldquoreducedrdquo or ldquodisappearedrdquo)
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Low risk Key code sealed and retained until end ofstudy
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoboth packages were of the same appear-ance and indistinguishablerdquo and investiga-tors did not see contents
Incomplete outcome data (attrition bias)All outcomes
Low risk All relevant participants included in analy-sis
Size Unclear risk 50 to 200 participants per treatment arm
Thorling 1990
Methods R DB PC parallel groupsParticipants given specific instructions on how to apply gel (not reported) to affectedarea 2 to 6 times daily as requiredAssessment at baseline 3 7 days in clinic
Participants Soft tissue injuries (lt 48 hours)N = 120M 85 F 35Mean age 27 yearsBaseline pain moderate to severe
87Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Thorling 1990 (Continued)
Interventions Naproxen gel 10 2 to 6 x daily n = 60Placebo gel 2 to 6 x daily n = 60Rescue medication paracetamol 500 mg
Outcomes PGE 5-point scale (responder = ldquogoodrdquo and ldquovery goodrdquo)Pain on passive movement 4-point scale (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquosupplied in unmarked tubesrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size Unclear risk 50 to 200 participants per treatment arm
Tonutti 1994
Methods R DB AC parallel groupsGel applied 3 times daily for 2 to 3 weeksAssessed at baseline and intervals of 7 days
Participants Muscle or joint traumaN = 30M 20 F 10Mean age 34 years1 participant had injury of mild severity Mean baseline pain on active movement 28(scale 0 to 4)
Interventions Ketoprofen gel 5 3 x 2 to 3 g daily n = 15Etofenamate gel 5 3 x 2 to 3 g n = 15No concomitant treatment with NSAID aspirin steroid or physical therapy
88Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Tonutti 1994 (Continued)
Outcomes PGE 4-point scale (responder = ldquogoodrdquo and ldquoexcellentrdquo)Pain on movement 5-point scale (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not reported
Allocation concealment (selection bias) Unclear risk Not reported
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquothe two drugs were packed in indistin-guishable tubesrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
Vecchiet 1991
Methods R DB PC parallel groupsGel applied to the skin on and around painful area and gently rubbed in until absorbedtwice daily for up to 10 daysAssessed at baseline 5 10 days
Participants Soft tissue trauma (minor sports injuries)N = 60M 60Mean age 25 yearsMean baseline pain on active movement moderate
Interventions Meclofenamic acid gel 5 2 x 10 cm daily (= 4 g) n = 30Placebo gel 2 x 10 cm daily n = 30Both groups treated with ice rest and bandage for first 48 hours before starting testtreatmentRescue medication paracetamol
Outcomes PGE 4-point scale (responder = ldquogoodrdquo and ldquoexcellentrdquo)Pain on movement 4-point scale (mean data)Withdrawals
89Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Vecchiet 1991 (Continued)
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
Whitefield 2002
Methods R DB (double dummy) AC parallel groupsGel applied to affected site with gentle massage and 1 tablet taken 3 times daily for atleast 7 daysAssessed at baseline 7 14 (if necessary) days in clinic and daily patient diary
Participants Soft tissue injuries (lt 24 hours)N = 100M 95 F 5Mean age 26 years (range 18 to 50)Mean baseline pain on movement 22 cm
Interventions Ibuprofen gel 5 + placebo tablet 3 x daily n = 50Ibuprofen 400 mg tablet + placebo gel 3 x daily n = 50No other medication or physical therapy was prescribed and no other analgesics wereallowed
Outcomes PGE 3-point scale (responder = ldquoexcellentrdquo)Change in condition of injury site 5-point scale (responder = ldquocompletely betterrdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
90Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Whitefield 2002 (Continued)
Random sequence generation (selectionbias)
Unclear risk Not reported
Allocation concealment (selection bias) Unclear risk Not reported
Blinding (performance bias and detectionbias)All outcomes
Low risk Placebo tablets Identical in appearance toactive tablets Active and placebo gels hadsimilar physical characteristics and weresupplied in identical tubes
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size Unclear risk 50 participants per treatment arm
kermark 1990
Methods R DB (double dummy) PC AC parallel groupsSpray applied to affected area and capsules taken 3 times daily for 2 weeksAssessment at baseline 3 or 4 7 and 14 days
Participants Superficial overuse sports injuries (symptom onset 74 weeks)N = 70M 44 F 18 (completers)Mean age 30 yearsBaseline pain on palpation mostly slight to moderate
Interventions Elmetacin spray (indomethacin 1) 3 to 5 x 05 to 15 ml daily + placebo capsules n= 23Indomethacin capsules 3 x 25 mg daily + placebo spray n = 23Placebo spray and capsules n = 24Rescue medication paracetamol
Outcomes No pain on palpation (= responder)Participant improvement 100 mm VAS (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk ldquorandom number coderdquo
91Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
kermark 1990 (Continued)
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical in appearancerdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk Responder analysis lt 5 withdrawals
Size High risk lt 50 participants per treatment group
AC active control BOCF baseline observation carried forward DB double-blind DHEP diclofenac hydroxyethylpyrrolidine ordiclofenac epolamine F female HCl hydrochloride hep heparin LOCF last observation carried forward M male N numberof participants in study n number of participants in treatment arm PC placebo control PGE Participant Global Evaluation PIpain intensity R randomised VAS visual analogue scale W withdrawals
Characteristics of excluded studies [ordered by study ID]
Study Reason for exclusion
Ambrus 1987 No usable dichotomous data
Anon 1993 Not double-blind
Ascherl 1982 No usable dichotomous data
Bagliani 1976 Not an RCT
Baracchi 1982 No usable data
Burnham 1998 lt10 participantstreatment arm in first period of crossover study
Boumlhmer 1995 Active control invalid
Cesarone 2008 Not an RCT
Coulibaly 2009 Not double-blind
Diebschlag 1985 No usable dichotomous data
Diebschlag 1986 Inappropriate randomisation
Diebschlag 1992 No usable dichotomous data
92Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Fantato 1971 No usable dichotomous data
Galer 2000 No usable data
Hallmeier 1986 Not double-blind
Hallmeier 1988 Not double-blind
Kaneko 1999 Inappropriate randomisation - quasi-randomised
Kockelbergh 1985b Treatment not applied daily
Kuwabara 2013 Used NSAID-lidocaine combination (conference abstract)
Lee 1991 Not an RCT
Link 1996 No usable dichotomous data
May 2007 No usable dichotomous data
Oakland 1993 Inappropriate comparator
Odaglia 1987 Not an RCT
Picardi 1993 Not an RCT
Taboada 1992 Dose and duration of treatment unclear
Vanderstraeten 1990 Not double-blind
Vinciguerra 2008 Not an RCT
Von Klug 1977 Chronic and acute outcomes combined
RCT randomised controlled trial
Characteristics of studies awaiting assessment [ordered by study ID]
NCT00351104
Methods R DB PC parallel groups
Participants Ankle sprain or strain Grade I or II Age ge 18 yearsN = 220
93Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT00351104 (Continued)
Interventions Ketoprofen patch 20 x 1 dailyPlacebo patch x 1 dailyTreatment for 21 days
Outcomes Assessments days 3 7 21PI (0-10) at rest and during activitiesPGEFunctionRescue medication
Notes Sponsor Endo PharmaceuticalsEstimated completion February 2007 No results
NCT00352625
Methods R DB PC parallel groups
Participants Shoulder elbow or knee tendonitis or bursitis Age ge 18 yearsN = 330
Interventions Ketoprofen patch 20 x 1 dailyPlacebo patch x 1 dailyTreatment for 21 days
Outcomes Assessments days 3 7 21PI (0-10) at rest and during activitiesPGEFunctionRescue medication
Notes Sponsor Endo PharmaceuticalsEstimated completion February 2007 No results
NCT00426985
Methods R DB PC parallel groups
Participants Shoulder elbow or knee tendonitis or bursitis Age ge 18 years
Interventions Ketoprofen patch 20 x 1 dailyPlacebo patchTreatment for 21 days
Outcomes PI during activityPI at restPGERescue medication
94Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT00426985 (Continued)
Safety
Notes Study terminated May 2008 but ldquosufficient number of subjects accrued to conduct analysisrdquo No results
NCT00640705
Methods R DB PC parallel groups
Participants Ankle sprain or strain Grade I or II le 48 hours PI 510 to 910 Age 18 to 75 yearsN = 170
Interventions Diclofenac sodium patch (15 mg) x 1 dailyPlaceboTreatment for 7 days
Outcomes EfficacySafety
Notes Study terminated July 2008 (Sponsor decision - no further details) No results posted
NCT00640939
Methods R DB PC parallel groups
Participants Shoulder elbow or knee tendonitis or bursitis Age 18 to 75 yearsBaseline pain mild to moderate but states 510 to 910N = 308
Interventions Diclofenac sodium patch 1 (15 mg) x 1 dailyPlacebo patch x 1 dailyDuration of treatment not reported probably 14 days
Outcomes EfficacySafety
Notes Sponsor Cerimon PharmaceuticalsEstimated completion April 2008 No results
NCT00680472
Methods R DB PC parallel groups
Participants Acute unilateral shoulder pain requiring treatment for ge 2 weeks Age ge 18 yearsN = 368
95Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT00680472 (Continued)
Interventions Ketoprofen patch (HKT-500) x 1 dailyPlacebo x 1 dailyTreatment for 14 days
Outcomes PainSafety
Notes Sponsor Hisamitsu Pharmaceutical Co IncEstimated completion October 2008 No results
NCT00680784
Methods R DB PC parallel groups
Participants Acute benign ankle sprain Grade I-II le 48 hours Age ge 18 yearsN = 260
Interventions Ketoprofen patch (HKT-500) x 1 dailyPlacebo patch x 1 dailyTreatment for 14 days
Outcomes PainSafety
Notes Sponsor Hisamitsu Pharmaceutical Co IncEstimated completion November 2008 No results
NCT00765700
Methods R DB PC parallel groups
Participants Acute sprain or strain of upper and lower extremities mild to moderate injuries le 72 hours Age 18 to 70 yearsN = 364
Interventions Ketoprofen cream 10 1 g x 3 dailyPlacebo cream x 3 dailyTreatment for 7 days
Outcomes PainSafety
Notes Sponsor Imprimis Pharmaceuticals IncEstimated completion September 2009 No results
96Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT00869063
Methods R DB PC parallel groups
Participants Wrist sprain strain or contusion (mild to moderate) ldquorecentrdquo Age 17 to 75 yearsN = 214
Interventions Diclofenac sodium patch 1 x 1 dailyPlacebo patch x 1 dailyTreatment for 7 days
Outcomes Change in PI during activity at 3 and 7 daysSafety
Notes Sponsor Cerimon PharmaceuticalsEstimated completion September 2009 No results
NCT00869180
Methods R DB PC parallel groups
Participants Unilateral ankle sprain (mild or moderate) ldquorecentrdquo Age 17 to 75 yearsN = 219
Interventions Diclofenac sodium patch x 1 dailyPlacebo patch x 1 dailyTreatment for 7 days
Outcomes Change in PI during activity at 3 and 7 days
Notes Sponsor Cerimon PharmaceuticalsEstimated completion August 2009 No results
NCT00931866
Methods R DB PC parallel groups
Participants Unilateral soft tissue injury between mid-bicep to wrist or mid-thigh to ankle (mild to moderate) ldquorecentrdquo Age 18to 75 yearsN = 407
Interventions Diclofenac sodium patch x 1 dailyPlacebo patch x 1 dailyTreatment for 14 days
Outcomes Change in PI during activity at 7 and 14 daysPIPGERescue medicationSafety
97Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT00931866 (Continued)
Notes Sponsor Cerimon PharmaceuticalsEstimated completion December 2009 No results
NCT01874626
Methods R DB PC parallel groups
Participants Acute ankle sprain Grade I or II le 24 hours Age ge 16 years Baseline PI ge 510N = 305
Interventions Ibuprofen cream 200 mg in 27 g cream x 4 dailyPlacebo cream x 4 dailyTreatment for 7 days
Outcomes PI on movement daily to 7 daysPGE 7 daysSystemic and local adverse events
Notes Estimated completion January 2014 No results
NCT01957215
Methods R DB PC parallel groups
Participants Acute lateral ankle sprain Grade I-II le 24 hours PI ge 510 Age 18 to 65 yearsN = 270
Interventions Indomethacin patch x 2 dailyPlacebo patch x 2 dailyDuration of treatment not reported probably 7 days
Outcomes PI on movement dailyPR dailyPGERescue medication
Notes Estimated completion September 2014 No results
NCT02324270
Methods R DB PC and AC parallel groups
Participants Uncomplicated acute ankle sprain Grade I or II lt 48 hours PI gt 50100 Age 18 to 65 yearsN = 658
98Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT02324270 (Continued)
Interventions Diclofenac epolamine (Flector) 13 patchGeneric diclofenac epolamine 13 patchPlacebo patchDuration of treatment not reported
Outcomes Bioequivalence studyChange from baseline in PI (VAS) at 3 daysApplication site reactions
Notes Estimated completion December 2014 No results
Sarzi-Puttini 2014
Methods R DB AC parallel group non-inferiority studyDuration 7 days
Participants Acute musculoskeletal injury (mainly muscular joint tendon)N = 697M 271 F 426Mean age 52 years
Interventions Ketoprofen (SKP-021) patch ketoprofen 30 mgDiclofenac (Voltadola) patch diclofenac sodium 140 mgPatches applied twice daily for 7 days
Outcomes ge 50 reduction in PI from baseline to end of study (100 mm VAS)Patient overall ratingClinical symptoms (4-point scale)Time to responseAdverse events skin reactionsSerious adverse events
Notes ldquoThe analysis of the data of this trial showed that the two formulations were equally effective and well tolerated inthe treatment of acute musculoskeletal injuriesrdquo
AC active-controlled DB double-blind F female M male N number of participants in study PC placebo-controlled PGE PatientGlobal Evaluation of treatment PI pain intensity R randomised VAS visual analogue scale
99Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Characteristics of ongoing studies [ordered by study ID]
NCT01945034
Trial name or title Placebo-controlled double-blind evaluation of the efficacy and safety of ibuprofen 5 topical gel for thetreatment of ankle sprain
Methods R DB PC parallel groups
Participants Ankle sprain Grade I or II age ge 12 yearsEstimated enrolment 280
Interventions Ibuprofen gel 5 x 2 dailyIbuprofen gel 5 x 3 dailyPlacebo gelTreatment for 7 days with additional 3 days as needed
Outcomes PI on weight bearing and rest at 7 daysPGE at 10 daysRescue medicationSafety
Starting date November 2013
Contact information Pfizer
Notes Estimated completion February 2015Includes participants aged ge 12 years
NCT02100670
Trial name or title A clinical study to assess the efficacy and onset of pain relief of topical MFC51123 diclofenac-menthol gelversus controls in ankle sprain
Methods R DB PC and AC parallel groups
Participants Acute lateral ankle sprain Grade I or II le 24 hours PI ge 510 Age 16 to 65 yearsEstimated enrolment 400
Interventions Diclofenac sodium 1 + methanol 3 gelDiclofenac sodium 1 + methanol 009 gelMethanol 3 gelPlacebo + 009 methanol gelTreatment for 10 days with 4 g gel x 4 daily
Outcomes PRPID on movement (0 to 10 days)PGETime to complete recoveryAdverse events
100Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT02100670 (Continued)
Starting date November 2013
Contact information GSKClinicalSupportHDgskcom
Notes Estimated completion November 2014
NCT02290821
Trial name or title A randomised double-blind placebo-controlled parallel group study to evaluate the efficacy and safety ofdiclofenac sodium topical gel (DSG) 1 applied four times daily in subjects with acute blunt soft tissueinjuriescontusions of the limbs
Methods R DB PC parallel groups
Participants Acute blunt soft tissue injuries or contusions of the limbs le 6 hours (ldquofreshrdquo) Age ge 16 yearsEstimated enrolment 200
Interventions Diclofenac sodium gel 1 x 4 dailyPlacebo gelDuration of treatment not reported
Outcomes PainSafety
Starting date December 2014
Contact information Novartis
Notes Estimated completion August 2015
AC active control DB double-blind PC placebo-controlled PGE Patient Global Evaluation of treatment PI pain intensity PIDpain intensity difference PR pain relief R randomised
101Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
D A T A A N D A N A L Y S E S
Comparison 1 Individual NSAID versus placebo
Outcome or subgroup titleNo ofstudies
No ofparticipants Statistical method Effect size
1 Clinical success 29 Risk Ratio (M-H Fixed 95 CI) Subtotals only11 Diclofenac 10 2050 Risk Ratio (M-H Fixed 95 CI) 160 [149 172]12 Ibuprofen 5 436 Risk Ratio (M-H Fixed 95 CI) 164 [133 201]13 Ketoprofen 7 683 Risk Ratio (M-H Fixed 95 CI) 156 [137 177]14 Piroxicam 3 504 Risk Ratio (M-H Fixed 95 CI) 148 [127 173]15 Indomethacin 3 341 Risk Ratio (M-H Fixed 95 CI) 126 [103 155]16 Benzydamine 3 193 Risk Ratio (M-H Fixed 95 CI) 115 [096 138]
2 Local adverse events 33 Risk Ratio (M-H Fixed 95 CI) Subtotals only21 Diclofenac 15 3271 Risk Ratio (M-H Fixed 95 CI) 078 [056 110]22 Ibuprofen 3 321 Risk Ratio (M-H Fixed 95 CI) 230 [098 543]23 Ketoprofen 8 852 Risk Ratio (M-H Fixed 95 CI) 119 [083 170]24 Piroxicam 3 522 Risk Ratio (M-H Fixed 95 CI) 042 [017 108]25 Felbinac 3 397 Risk Ratio (M-H Fixed 95 CI) 191 [049 750]26 Indomethacin 3 354 Risk Ratio (M-H Fixed 95 CI) 265 [091 773]
Comparison 2 Diclofenac versus placebo (effect of formulation)
Outcome or subgroup titleNo ofstudies
No ofparticipants Statistical method Effect size
1 Clinical success 10 2050 Risk Ratio (M-H Fixed 95 CI) 160 [149 172]11 Plaster - Flector 4 1030 Risk Ratio (M-H Fixed 95 CI) 153 [136 171]12 Plaster - other 3 474 Risk Ratio (M-H Fixed 95 CI) 155 [137 175]13 Gel - Emulgel 2 314 Risk Ratio (M-H Fixed 95 CI) 384 [268 550]14 Gel - other 1 232 Risk Ratio (M-H Fixed 95 CI) 115 [105 127]
Comparison 3 Ibuprofen versus placebo (effect of formulation)
Outcome or subgroup titleNo ofstudies
No ofparticipants Statistical method Effect size
1 Clinical success 5 436 Risk Ratio (M-H Fixed 95 CI) 164 [133 201]11 Cream 3 195 Risk Ratio (M-H Fixed 95 CI) 128 [103 159]12 Gel 2 241 Risk Ratio (M-H Fixed 95 CI) 266 [169 421]
102Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Comparison 4 Ketoprofen versus placebo
Outcome or subgroup titleNo ofstudies
No ofparticipants Statistical method Effect size
1 Clinical success 7 683 Risk Ratio (M-H Fixed 95 CI) 156 [137 177]11 Plaster 2 335 Risk Ratio (M-H Fixed 95 CI) 121 [104 140]12 Gel 5 348 Risk Ratio (M-H Fixed 95 CI) 219 [174 275]
Comparison 5 All topical NSAIDs versus placebo
Outcome or subgroup titleNo ofstudies
No ofparticipants Statistical method Effect size
1 Local adverse events 42 6740 Risk Ratio (M-H Fixed 95 CI) 098 [080 121]2 Systemic adverse events 36 5576 Risk Ratio (M-H Fixed 95 CI) 096 [073 126]3 Adverse event withdrawals 42 6405 Risk Ratio (M-H Fixed 95 CI) 101 [064 159]
Comparison 6 Topical NSAID versus active comparator
Outcome or subgroup titleNo ofstudies
No ofparticipants Statistical method Effect size
1 Clinical success - topicalpiroxicam vs topicalindomethacin
3 641 Risk Ratio (M-H Fixed 95 CI) 124 [107 144]
2 Local adverse events -topical piroxicam vs topicalindomethacin
3 671 Risk Ratio (M-H Fixed 95 CI) 021 [009 047]
103Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Analysis 11 Comparison 1 Individual NSAID versus placebo Outcome 1 Clinical success
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 1 Individual NSAID versus placebo
Outcome 1 Clinical success
Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
1 Diclofenac
Coudreuse 2010 99117 82116 175 120 [ 104 138 ]
Joussellin 2003 3668 2466 52 146 [ 099 215 ]
Klainguti 2010 5662 4659 100 116 [ 099 136 ]
Li 2013 173192 114192 242 152 [ 134 172 ]
Predel 2004 5560 560 11 1100 [ 474 2555 ]
Predel 2012 (1) 5980 941 25 336 [ 186 607 ]
Predel 2012 (2) 5780 841 22 365 [ 193 690 ]
Predel 2013a 111118 93114 201 115 [ 105 127 ]
Predel 2013b 3636 736 16 487 [ 257 923 ]
Rowbotham 2003 75191 48181 104 148 [ 110 200 ]
Saillant 1998 4370 2570 53 172 [ 119 248 ]
Subtotal (95 CI) 1074 976 1000 160 [ 149 172 ]Total events 800 (NSAID) 461 (Placebo)
Heterogeneity Chi2 = 12176 df = 10 (Plt000001) I2 =92
Test for overall effect Z = 1278 (P lt 000001)
2 Ibuprofen
Billigmann 1996 2580 1080 136 250 [ 129 486 ]
Campbell 1994 2126 1925 264 106 [ 080 142 ]
Dreiser 1988 2632 1232 164 217 [ 134 349 ]
Machen 2002 2540 941 121 285 [ 152 532 ]
Ramesh 1983 2340 2340 314 100 [ 069 146 ]
Subtotal (95 CI) 218 218 1000 164 [ 133 201 ]Total events 120 (NSAID) 73 (Placebo)
Heterogeneity Chi2 = 2102 df = 4 (P = 000031) I2 =81
Test for overall effect Z = 465 (P lt 000001)
3 Ketoprofen
Airaksinen 1993 2429 1427 91 160 [ 107 238 ]
Dreiser 1989 1830 530 32 360 [ 154 844 ]
02 05 1 2 5
Favours placebo Favours NSAID
(Continued )
104Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
( Continued)Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Julien 1989 1830 630 38 300 [ 138 650 ]
Kockelbergh 1985 3038 2236 142 129 [ 095 176 ]
Mazi res 2005a 7281 6082 376 121 [ 104 141 ]
Mazi res 2005b 5087 4185 262 119 [ 090 158 ]
Noret 1987 3951 947 59 399 [ 218 733 ]
Subtotal (95 CI) 346 337 1000 156 [ 137 177 ]Total events 251 (NSAID) 157 (Placebo)
Heterogeneity Chi2 = 3104 df = 6 (P = 000002) I2 =81
Test for overall effect Z = 683 (P lt 000001)
4 Piroxicam
Aoki 1984 5672 3367 286 158 [ 120 207 ]
Fujimaki 1985 4483 4082 337 109 [ 080 147 ]
Russell 1991 79100 45100 377 176 [ 138 223 ]
Subtotal (95 CI) 255 249 1000 148 [ 127 173 ]Total events 179 (NSAID) 118 (Placebo)
Heterogeneity Chi2 = 624 df = 2 (P = 004) I2 =68
Test for overall effect Z = 499 (P lt 000001)
5 Indomethacin
Aoki 1984 4164 3367 413 130 [ 096 176 ]
Fujimaki 1985 4482 4082 513 110 [ 082 148 ]
kermark 1990 1222 624 74 218 [ 099 481 ]
Subtotal (95 CI) 168 173 1000 126 [ 103 155 ]Total events 97 (NSAID) 79 (Placebo)
Heterogeneity Chi2 = 269 df = 2 (P = 026) I2 =26
Test for overall effect Z = 221 (P = 0027)
6 Benzydamine
Chatterjee 1977 2125 1225 185 175 [ 112 272 ]
Haig 1986 1821 1322 196 145 [ 098 214 ]
Linde 1985 3550 4050 618 088 [ 070 110 ]
Subtotal (95 CI) 96 97 1000 115 [ 096 138 ]Total events 74 (NSAID) 65 (Placebo)
Heterogeneity Chi2 = 1033 df = 2 (P = 001) I2 =81
Test for overall effect Z = 152 (P = 013)
02 05 1 2 5
Favours placebo Favours NSAID
(1) Three times daily application
(2) Twice daily application
105Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Analysis 12 Comparison 1 Individual NSAID versus placebo Outcome 2 Local adverse events
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 1 Individual NSAID versus placebo
Outcome 2 Local adverse events
Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
1 Diclofenac
Coudreuse 2010 1117 7116 102 014 [ 002 113 ]
Jenoure 1997 144 141 15 093 [ 006 1442 ]
Joussellin 2003 168 366 44 032 [ 003 303 ]
Klainguti 2010 0126 159 30 016 [ 001 381 ]
Kuehl 2011 16207 12211 173 136 [ 066 280 ]
Li 2013 4192 3192 44 133 [ 030 588 ]
NCT01255423 1104 3102 44 033 [ 003 309 ]
NCT01272934 1102 0103 07 303 [ 012 7350 ]
NCT01272947 0104 0100 Not estimable
Predel 2012 (1) 1160 182 19 051 [ 003 809 ]
Predel 2013a 1120 4116 59 024 [ 003 213 ]
Predel 2013b 036 036 Not estimable
Rowbotham 2003 27191 31181 463 083 [ 051 133 ]
Saillant 1998 070 070 Not estimable
Spacca 2005 079 076 Not estimable
Subtotal (95 CI) 1720 1551 1000 078 [ 056 110 ]Total events 54 (NSAID) 66 (Placebo)
Heterogeneity Chi2 = 944 df = 10 (P = 049) I2 =00
Test for overall effect Z = 142 (P = 016)
2 Ibuprofen
Billigmann 1996 1180 480 573 275 [ 091 827 ]
Machen 2002 440 241 283 205 [ 040 1057 ]
Ramesh 1983 140 140 143 100 [ 006 1544 ]
Subtotal (95 CI) 160 161 1000 230 [ 098 543 ]Total events 16 (NSAID) 7 (Placebo)
Heterogeneity Chi2 = 048 df = 2 (P = 079) I2 =00
Test for overall effect Z = 190 (P = 0057)
3 Ketoprofen
002 01 1 10 50
Favours NSAID Favours placebo
(Continued )
106Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
( Continued)
Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Airaksinen 1993 529 427 99 116 [ 035 389 ]
Dreiser 1989 030 230 60 020 [ 001 400 ]
Julien 1989 130 030 12 300 [ 013 7083 ]
Kockelbergh 1985 138 136 24 095 [ 006 1459 ]
Mazi res 2005a 1281 682 142 202 [ 080 513 ]
Mazi res 2005b 2987 2785 651 105 [ 068 161 ]
Noret 1987 151 047 12 277 [ 012 6636 ]
Parrini 1992 083 086 Not estimable
Subtotal (95 CI) 429 423 1000 119 [ 083 170 ]Total events 49 (NSAID) 40 (Placebo)
Heterogeneity Chi2 = 357 df = 6 (P = 073) I2 =00
Test for overall effect Z = 095 (P = 034)
4 Piroxicam
Aoki 1984 179 274 147 047 [ 004 506 ]
Fujimaki 1985 183 282 143 049 [ 005 534 ]
Russell 1991 4102 10102 710 040 [ 013 123 ]
Subtotal (95 CI) 264 258 1000 042 [ 017 108 ]Total events 6 (NSAID) 14 (Placebo)
Heterogeneity Chi2 = 003 df = 2 (P = 098) I2 =00
Test for overall effect Z = 180 (P = 0072)
5 Felbinac
McLatchie 1989 3118 2113 666 144 [ 024 844 ]
Morris 1991 041 043 Not estimable
Sanguinetti 1989 342 140 334 286 [ 031 2634 ]
Subtotal (95 CI) 201 196 1000 191 [ 049 750 ]Total events 6 (NSAID) 3 (Placebo)
Heterogeneity Chi2 = 023 df = 1 (P = 063) I2 =00
Test for overall effect Z = 093 (P = 035)
6 Indomethacin
Aoki 1984 270 274 440 106 [ 015 730 ]
Fujimaki 1985 582 282 452 250 [ 050 1252 ]
kermark 1990 422 024 108 978 [ 056 17191 ]
Subtotal (95 CI) 174 180 1000 265 [ 091 773 ]Total events 11 (NSAID) 4 (Placebo)
Heterogeneity Chi2 = 167 df = 2 (P = 043) I2 =00
Test for overall effect Z = 179 (P = 0073)
002 01 1 10 50
Favours NSAID Favours placebo
(1) DHEP + heparin and DHEP combined
107Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Analysis 21 Comparison 2 Diclofenac versus placebo (effect of formulation) Outcome 1 Clinical success
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 2 Diclofenac versus placebo (effect of formulation)
Outcome 1 Clinical success
Study or subgroup Experimental Control Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
1 Plaster - Flector
Joussellin 2003 3668 2466 52 146 [ 099 215 ]
Li 2013 173192 114192 242 152 [ 134 172 ]
Rowbotham 2003 75191 48181 104 148 [ 110 200 ]
Saillant 1998 4370 2570 53 172 [ 119 248 ]
Subtotal (95 CI) 521 509 451 153 [ 136 171 ]Total events 327 (Experimental) 211 (Control)
Heterogeneity Chi2 = 051 df = 3 (P = 092) I2 =00
Test for overall effect Z = 720 (P lt 000001)
2 Plaster - other
Coudreuse 2010 99117 82116 175 120 [ 104 138 ]
Klainguti 2010 5662 4659 100 116 [ 099 136 ]
Predel 2004 5560 560 11 1100 [ 474 2555 ]
Subtotal (95 CI) 239 235 285 155 [ 137 175 ]Total events 210 (Experimental) 133 (Control)
Heterogeneity Chi2 = 4658 df = 2 (Plt000001) I2 =96
Test for overall effect Z = 699 (P lt 000001)
3 Gel - Emulgel
Predel 2012 (1) 116160 1782 48 350 [ 227 540 ]
Predel 2013b 3636 736 16 487 [ 257 923 ]
Subtotal (95 CI) 196 118 64 384 [ 268 550 ]Total events 152 (Experimental) 24 (Control)
Heterogeneity Chi2 = 070 df = 1 (P = 040) I2 =00
Test for overall effect Z = 734 (P lt 000001)
4 Gel - other
Predel 2013a 111118 93114 201 115 [ 105 127 ]
Subtotal (95 CI) 118 114 201 115 [ 105 127 ]Total events 111 (Experimental) 93 (Control)
02 05 1 2 5
Favours placebo Favours diclofenac
(Continued )
108Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
( Continued)Study or subgroup Experimental Control Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Heterogeneity not applicable
Test for overall effect Z = 284 (P = 00045)
Total (95 CI) 1074 976 1000 160 [ 149 172 ]Total events 800 (Experimental) 461 (Control)
Heterogeneity Chi2 = 12176 df = 9 (Plt000001) I2 =93
Test for overall effect Z = 1278 (P lt 000001)
Test for subgroup differences Chi2 = 5003 df = 3 (P = 000) I2 =94
02 05 1 2 5
Favours placebo Favours diclofenac
(1) twice daily and three times daily applications combined
Analysis 31 Comparison 3 Ibuprofen versus placebo (effect of formulation) Outcome 1 Clinical success
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 3 Ibuprofen versus placebo (effect of formulation)
Outcome 1 Clinical success
Study or subgroup Ibuprofen Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
1 Cream
Campbell 1994 2126 1925 264 106 [ 080 142 ]
Dreiser 1988 2632 1232 164 217 [ 134 349 ]
Ramesh 1983 2340 2340 314 100 [ 069 146 ]
Subtotal (95 CI) 98 97 742 128 [ 103 159 ]Total events 70 (Ibuprofen) 54 (Placebo)
Heterogeneity Chi2 = 791 df = 2 (P = 002) I2 =75
Test for overall effect Z = 222 (P = 0026)
2 Gel
Billigmann 1996 2580 1080 136 250 [ 129 486 ]
Machen 2002 2540 941 121 285 [ 152 532 ]
Subtotal (95 CI) 120 121 258 266 [ 169 421 ]Total events 50 (Ibuprofen) 19 (Placebo)
001 01 1 10 100
Favours placebo Favours ibuprofen
(Continued )
109Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
( Continued)Study or subgroup Ibuprofen Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Heterogeneity Chi2 = 008 df = 1 (P = 078) I2 =00
Test for overall effect Z = 420 (P = 0000026)
Total (95 CI) 218 218 1000 164 [ 133 201 ]Total events 120 (Ibuprofen) 73 (Placebo)
Heterogeneity Chi2 = 2102 df = 4 (P = 000031) I2 =81
Test for overall effect Z = 465 (P lt 000001)
Test for subgroup differences Chi2 = 805 df = 1 (P = 000) I2 =88
001 01 1 10 100
Favours placebo Favours ibuprofen
Analysis 41 Comparison 4 Ketoprofen versus placebo Outcome 1 Clinical success
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 4 Ketoprofen versus placebo
Outcome 1 Clinical success
Study or subgroup Diclofenac Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
1 Plaster
Mazi res 2005a 7281 6082 376 121 [ 104 141 ]
Mazi res 2005b 5087 4185 262 119 [ 090 158 ]
Subtotal (95 CI) 168 167 638 121 [ 104 140 ]Total events 122 (Diclofenac) 101 (Placebo)
Heterogeneity Chi2 = 002 df = 1 (P = 090) I2 =00
Test for overall effect Z = 249 (P = 0013)
2 Gel
Airaksinen 1993 2429 1427 91 160 [ 107 238 ]
Dreiser 1989 1830 530 32 360 [ 154 844 ]
Julien 1989 1830 630 38 300 [ 138 650 ]
Kockelbergh 1985 3038 2236 142 129 [ 095 176 ]
Noret 1987 3951 947 59 399 [ 218 733 ]
001 01 1 10 100
Favours placebo Favours diclofenac
(Continued )
110Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
( Continued)Study or subgroup Diclofenac Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Subtotal (95 CI) 178 170 362 219 [ 174 275 ]Total events 129 (Diclofenac) 56 (Placebo)
Heterogeneity Chi2 = 1937 df = 4 (P = 000067) I2 =79
Test for overall effect Z = 674 (P lt 000001)
Total (95 CI) 346 337 1000 156 [ 137 177 ]Total events 251 (Diclofenac) 157 (Placebo)
Heterogeneity Chi2 = 3104 df = 6 (P = 000002) I2 =81
Test for overall effect Z = 683 (P lt 000001)
Test for subgroup differences Chi2 = 1859 df = 1 (P = 000) I2 =95
001 01 1 10 100
Favours placebo Favours diclofenac
Analysis 51 Comparison 5 All topical NSAIDs versus placebo Outcome 1 Local adverse events
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 5 All topical NSAIDs versus placebo
Outcome 1 Local adverse events
Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Airaksinen 1993 529 427 26 116 [ 035 389 ]
Aoki 1984 (1) 270 137 08 106 [ 010 1128 ]
Aoki 1984 (2) 179 137 09 047 [ 003 728 ]
Auclair 1989 5123 6116 39 079 [ 025 251 ]
Billigmann 1996 1180 480 26 275 [ 091 827 ]
Costantino 2011 (3) 3287 6142 51 025 [ 006 097 ]
Coudreuse 2010 1117 7116 45 014 [ 002 113 ]
Diebshlag 1990 (4) 113 06 04 150 [ 007 3229 ]
Diebshlag 1990 (5) 112 06 04 162 [ 008 3466 ]
Dreiser 1989 030 230 16 020 [ 001 400 ]
002 01 1 10 50
Favours NSAID Favours placebo
(Continued )
111Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
( Continued)Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Dreiser 1990 030 330 22 014 [ 001 265 ]
Dreiser 1994 265 066 03 508 [ 025 10373 ]
Fujimaki 1985 (6) 183 141 09 049 [ 003 770 ]
Fujimaki 1985 (7) 582 141 09 250 [ 030 2070 ]
Haig 1986 021 022 Not estimable
Jenoure 1997 144 141 07 093 [ 006 1442 ]
Joussellin 2003 168 366 19 032 [ 003 303 ]
Julien 1989 130 030 03 300 [ 013 7083 ]
Klainguti 2010 (8) 0126 159 13 016 [ 001 381 ]
Kockelbergh 1985 138 136 07 095 [ 006 1459 ]
Kuehl 2011 16207 12211 76 136 [ 066 280 ]
Li 2013 4192 3192 19 133 [ 030 588 ]
Linde 1985 050 050 Not estimable
Machen 2002 440 241 13 205 [ 040 1057 ]
Mazi res 2005a 1281 682 38 202 [ 080 513 ]
Mazi res 2005b 2987 2785 174 105 [ 068 161 ]
McLatchie 1989 3118 2113 13 144 [ 024 844 ]
Morris 1991 041 043 Not estimable
NCT01255423 1104 3102 19 033 [ 003 309 ]
NCT01272934 1102 0103 03 303 [ 012 7350 ]
NCT01272947 0104 0100 Not estimable
Noret 1987 151 047 03 277 [ 012 6636 ]
Parrini 1992 083 086 Not estimable
Predel 2012 (9) 1160 182 08 051 [ 003 809 ]
Predel 2013a 1120 4116 26 024 [ 003 213 ]
Predel 2013b 036 036 Not estimable
Ramesh 1983 140 140 06 100 [ 006 1544 ]
Rowbotham 2003 27191 31181 203 083 [ 051 133 ]
Russell 1991 4102 10102 64 040 [ 013 123 ]
Saillant 1998 070 070 Not estimable
Sanguinetti 1989 342 140 07 286 [ 031 2634 ]
002 01 1 10 50
Favours NSAID Favours placebo
(Continued )
112Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
( Continued)Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Sinniger 1981 010 010 Not estimable
Spacca 2005 079 076 Not estimable
Thorling 1990 160 060 03 300 [ 012 7220 ]
kermark 1990 422 024 03 978 [ 056 17191 ]
Total (95 CI) 3619 3121 1000 098 [ 080 121 ]Total events 155 (NSAID) 145 (Placebo)
Heterogeneity Chi2 = 3354 df = 35 (P = 054) I2 =00
Test for overall effect Z = 017 (P = 087)
Test for subgroup differences Not applicable
002 01 1 10 50
Favours NSAID Favours placebo
(1) indomethacin
(2) piroxicam
(3) DHEP + heparin and DHEP combined
(4) ketorolac
(5) etofenamate
(6) piroxicam
(7) indomethacin
(8) DHEP + heparin and DHEP combined
(9) Twice and three times daily application combined
113Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Analysis 52 Comparison 5 All topical NSAIDs versus placebo Outcome 2 Systemic adverse events
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 5 All topical NSAIDs versus placebo
Outcome 2 Systemic adverse events
Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Airaksinen 1993 129 027 05 280 [ 012 6593 ]
Aoki 1984 (1) 079 037 Not estimable
Aoki 1984 (2) 070 037 Not estimable
Campbell 1994 126 025 05 289 [ 012 6775 ]
Chatterjee 1977 025 025 Not estimable
Coudreuse 2010 1117 0116 05 297 [ 012 7227 ]
Diebshlag 1990 (3) 013 06 Not estimable
Diebshlag 1990 (4) 012 06 Not estimable
Dreiser 1988 032 032 Not estimable
Dreiser 1989 030 030 Not estimable
Dreiser 1990 030 030 Not estimable
Dreiser 1994 065 066 Not estimable
Fujimaki 1985 (5) 083 041 Not estimable
Fujimaki 1985 (6) 182 041 07 152 [ 006 3647 ]
Haig 1986 021 022 Not estimable
Hoffmann 2012 0115 0118 Not estimable
Joussellin 2003 168 066 05 291 [ 012 7025 ]
Klainguti 2010 (7) 2126 059 07 236 [ 012 4844 ]
Kuehl 2011 15207 23211 236 066 [ 036 124 ]
Li 2013 10192 4192 41 250 [ 080 783 ]
Linde 1985 050 050 Not estimable
Machen 2002 040 041 Not estimable
Mazi res 2005a 1187 785 73 154 [ 062 377 ]
Mazi res 2005b 1381 1482 144 094 [ 047 187 ]
McLatchie 1989 0118 0113 Not estimable
Morris 1991 041 043 Not estimable
001 01 1 10 100
Favours topical NSAID Favours placebo
(Continued )
114Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
( Continued)Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
NCT01272947 2104 2100 21 096 [ 014 670 ]
Parrini 1992 083 086 Not estimable
Predel 2004 060 060 Not estimable
Predel 2012 (8) 3160 382 41 051 [ 011 248 ]
Predel 2013a 6120 8116 84 073 [ 026 203 ]
Predel 2013b 036 136 16 033 [ 001 792 ]
Rowbotham 2003 21191 22181 234 090 [ 052 159 ]
Russell 1991 4102 7102 73 057 [ 017 189 ]
Saillant 1998 070 070 Not estimable
Sanguinetti 1989 042 040 Not estimable
Sinniger 1981 010 010 Not estimable
Spacca 2005 079 076 Not estimable
Thorling 1990 060 060 Not estimable
Total (95 CI) 2956 2620 1000 096 [ 073 126 ]Total events 92 (NSAID) 91 (Placebo)
Heterogeneity Chi2 = 946 df = 15 (P = 085) I2 =00
Test for overall effect Z = 030 (P = 076)
Test for subgroup differences Not applicable
001 01 1 10 100
Favours topical NSAID Favours placebo
(1) piroxicam
(2) indomethacin
(3) ketorolac
(4) etofenamate
(5) piroxicam
(6) indomethacin
(7) DHEP + heparin and DHEP combined
(8) Twice and three times daily application combined
115Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Analysis 53 Comparison 5 All topical NSAIDs versus placebo Outcome 3 Adverse event withdrawals
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 5 All topical NSAIDs versus placebo
Outcome 3 Adverse event withdrawals
Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Airaksinen 1993 029 027 Not estimable
Aoki 1984 072 034 Not estimable
Aoki 1984 064 033 Not estimable
Auclair 1989 1123 0116 14 283 [ 012 6879 ]
Billigmann 1996 280 080 14 500 [ 024 10253 ]
Campbell 1994 026 025 Not estimable
Chatterjee 1977 025 025 Not estimable
Costantino 2011 0188 0142 Not estimable
Coudreuse 2010 0117 0116 Not estimable
Diebshlag 1990 013 06 Not estimable
Diebshlag 1990 012 06 Not estimable
Dreiser 1988 032 032 Not estimable
Dreiser 1989 030 230 69 020 [ 001 400 ]
Dreiser 1990 030 130 42 033 [ 001 787 ]
Dreiser 1994 065 066 Not estimable
Fujimaki 1985 083 041 Not estimable
Fujimaki 1985 482 041 18 455 [ 025 8261 ]
Julien 1989 030 030 Not estimable
Kockelbergh 1985 038 036 Not estimable
Kuehl 2011 4207 9211 248 045 [ 014 145 ]
Li 2013 2192 0192 14 500 [ 024 10347 ]
Linde 1985 050 050 Not estimable
Machen 2002 040 041 Not estimable
Mazi res 2005a 987 685 169 147 [ 055 394 ]
Mazi res 2005b 381 082 14 709 [ 037 13503 ]
McLatchie 1989 0118 0113 Not estimable
001 01 1 10 100
Favours topical NSAID Favours placebo
(Continued )
116Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
( Continued)Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Morris 1991 041 043 Not estimable
NCT01255423 0104 0102 Not estimable
NCT01272934 0102 1103 41 034 [ 001 817 ]
NCT01272947 0104 0100 Not estimable
Noret 1987 148 045 14 282 [ 012 6740 ]
Parrini 1992 060 060 Not estimable
Predel 2004 160 060 14 300 [ 012 7220 ]
Predel 2012 2160 182 37 103 [ 009 1114 ]
Predel 2013a 1120 1116 28 097 [ 006 1527 ]
Predel 2013b 036 036 Not estimable
Ramesh 1983 140 140 28 100 [ 006 1544 ]
Rowbotham 2003 0191 0181 Not estimable
Russell 1991 1102 8102 222 013 [ 002 098 ]
Saillant 1998 070 070 Not estimable
Sanguinetti 1989 042 040 Not estimable
Sinniger 1981 010 010 Not estimable
Spacca 2005 079 076 Not estimable
Thorling 1990 060 060 Not estimable
kermark 1990 122 024 13 326 [ 014 7610 ]
Total (95 CI) 3365 3040 1000 101 [ 064 159 ]Total events 33 (NSAID) 30 (Placebo)
Heterogeneity Chi2 = 1502 df = 16 (P = 052) I2 =00
Test for overall effect Z = 004 (P = 097)
Test for subgroup differences Not applicable
001 01 1 10 100
Favours topical NSAID Favours placebo
117Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Analysis 61 Comparison 6 Topical NSAID versus active comparator Outcome 1 Clinical success - topical
piroxicam vs topical indomethacin
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 6 Topical NSAID versus active comparator
Outcome 1 Clinical success - topical piroxicam vs topical indomethacin
Study or subgroup Piroxicam Indomethacin Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Aoki 1984 5672 4164 301 121 [ 097 151 ]
Fujimaki 1985 4483 4482 307 099 [ 074 131 ]
Sugioka 1984 85175 55165 392 146 [ 112 190 ]
Total (95 CI) 330 311 1000 124 [ 107 144 ]Total events 185 (Piroxicam) 140 (Indomethacin)
Heterogeneity Chi2 = 390 df = 2 (P = 014) I2 =49
Test for overall effect Z = 278 (P = 00055)
Test for subgroup differences Not applicable
01 02 05 1 2 5 10
Favours indomethacin Favours piroxicam
118Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Analysis 62 Comparison 6 Topical NSAID versus active comparator Outcome 2 Local adverse events -
topical piroxicam vs topical indomethacin
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 6 Topical NSAID versus active comparator
Outcome 2 Local adverse events - topical piroxicam vs topical indomethacin
Study or subgroup Piroxicam Indomethacin Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Aoki 1984 179 270 64 044 [ 004 478 ]
Fujimaki 1985 183 582 152 020 [ 002 165 ]
Sugioka 1984 5178 26179 784 019 [ 008 049 ]
Total (95 CI) 340 331 1000 021 [ 009 047 ]Total events 7 (Piroxicam) 33 (Indomethacin)
Heterogeneity Chi2 = 041 df = 2 (P = 081) I2 =00
Test for overall effect Z = 382 (P = 000013)
Test for subgroup differences Not applicable
002 01 1 10 50
Favours piroxicam Favours indomethacin
A P P E N D I C E S
Appendix 1 CENTRAL search strategy (via CRSO) for 2015 update
1 MESH DESCRIPTOR Anti-Inflammatory Agents Non-Steroidal EXPLODE ALL TREES (13419)2 (bufexamac OR bufexine OR calmaderm OR ekzemase OR dicoflenac OR solaraze OR pennsaid OR voltarol OR emugel OR
voltarene OR voltarol OR optha OR voltaren OR etofenamate OR afrolate OR algesalona OR bayro OR deiron OR etofen ORflexium OR flogoprofen OR rheuma-gel OR rheumon OR traumalix OR traumon OR zenavan OR felbinac OR dolinac OR flexfreeOR napageln OR target OR traxam OR fentiazac OR domureuma OR fentiazaco OR norvedan OR riscalon OR fepradinol ORdalgen OR flexidol OR cocresol OR rangozona OR reuflodol OR pinazone OR zepelin OR flufenamic OR dignodolin OR rheumaOR lindofluid OR sastridex OR lunoxaprofen OR priaxim OR flubiprofen OR fenomel OR ocufen OR ocuflur OR ldquoTrans ActLATrdquo OR tulip OR ibuprofen OR cuprofen OR ldquodeep relief rdquo OR fenbid OR ibu-cream OR ibugel OR ibuleve OR ibumousse ORibuspray OR ldquonurofen gelrdquo OR proflex OR motrin OR advil OR radian OR ralgex OR ibutop OR indomethacin OR indocin ORindospray OR isonixin OR nixyn OR ketoprofen OR tiloket OR oruvail OR powergel OR solpaflex OR ketorolac OR acular ORtrometamol OR meclofenamic OR naproxen OR naprosyn OR niflumic OR actol OR flunir OR niflactol topico OR niflugel ORnifluril OR oxyphenbutazone OR californit OR diflamil OR otone OR tanderil OR piketoprofen OR calmatel OR triparsean ORpiroxicam OR feldene OR pranoprofen OR oftalar OR pranox OR suxibuzone OR danilon OR flamilon OR ufenamate OR fenazolOR flector OR benzydamine) TIABKY (25220)
3 1 OR 2 (32484)4 MESH DESCRIPTOR Administration Topical EXPLODE ALL TREES (2169)
119Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
5 (topical OR cutaneous OR dermal OR transcutaneous OR transdermal OR percutaneous OR skin OR massage ORembrocation OR gel OR ointment OR aerosol OR cream OR cregraveme OR lotion OR mouse OR foam OR liniment OR spray OR rubOR balm OR salve OR emulsion OR oil OR patch OR plaster)TIABKY (67940)
6 4 OR 5 (70486)7 MESH DESCRIPTOR Athletic Injuries EXPLODE ALL TREES (411)8 (strain OR sprain OR contusion OR distortion OR compression OR ldquosports injurrdquo OR ldquosoft tissue injurrdquo OR tendnitis OR
ldquomuscle painrdquo OR periarthritis OR epicondylitis OR tenosynovitis)TIABKY (9158)9 7 OR 8 (9448)
10 MESH DESCRIPTOR pain EXPLODE ALL TREES (29943)11 (pain OR analgesi)TIABKY (74815)12 10 OR 11 (80041)13 3 AND 6 AND 9 AND 12 (110)
Appendix 2 MEDLINE search strategy via Ovid (for 2015 update)
1 exp Anti-inflammatory Agents non-steroidal (162888)2 (bufexamac OR bufexine OR calmaderm OR ekzemase OR diclofenac OR solaraze OR pennsaid OR voltarol OR emugel OR
voltarene OR voltarol OR optha OR voltaren OR etofenamate OR afrolate OR algesalona OR bayro OR deiron OR etofen ORflexium OR flogoprofen OR rheuma-gel OR rheumon OR traumalix OR traumon OR zenavan OR felbinac OR dolinac OR flexfreeOR napageln OR target OR traxam OR fentiazac OR domureuma OR fentiazaco OR norvedan OR riscalon OR fepradinol ORdalgen OR flexidol OR cocresol OR rangozona OR reuflodol OR pinazone OR zepelin OR flufenamic OR dignodolin OR rheumaOR lindofluid OR sastridex OR lunoxaprofen OR priaxim OR flubiprofen OR fenomel OR ocufen OR ocuflur OR ldquoTrans ActLATrdquo OR tulip OR ibuprofen OR cuprofen OR ldquodeep relief rdquo OR fenbid OR ibu-cream OR ibugel OR ibuleve OR ibumousse ORibuspray OR ldquonurofen gelrdquo OR proflex OR motrin OR advil OR radian OR ralgex OR ibutop OR indomethacin OR indocin ORindospray OR isonixin OR nixyn OR ketoprofen OR tiloket OR oruvail OR powergel OR solpaflex OR ketorolac OR acular ORtrometamol OR meclofenamic OR naproxen OR naprosyn OR niflumic OR actol OR flunir OR niflactol topico OR niflugel ORnifluril OR oxyphenbutazone OR californit OR diflamil OR otone OR tanderil OR piketoprofen OR calmatel OR triparsean ORpiroxicam OR feldene OR pranoprofen OR oftalar OR pranox OR suxibuzone OR danilon OR flamilon OR ufenamate OR fenazolOR flector OR benzydamine)mp (558284)
3 1 OR 2 (664691)4 exp Administration Topical (69697)5 (topical OR cutaneous OR dermal OR transcutaneous OR transdermal OR percutaneous OR skin OR massage OR
embrocation OR gel OR ointment OR aerosol OR cream OR cregraveme OR lotion OR mouse OR foam OR liniment OR spray OR rubOR balm OR salve OR emulsion OR oil OR patch OR plaster)mp (1982395)
6 4 OR 5 (1997720)7 exp Athletic Injuries (21464)8 (strain OR sprain OR contusion OR distortion OR compression OR ldquosports injurrdquo OR ldquosoft tissue injurrdquo OR tendnitis OR
ldquomuscle painrdquo OR periarthritis OR epicondylitis OR tenosynovitis)mp (420133)9 7 OR 8 (439228)
10 Pain (113906)11 (pain OR analgesi)mp (585249)12 10 or 11 (585249)13 randomized controlled trialpt (401171)14 randomizedab (296222)15 placeboab (155341)16 drug therapyfs (1789858)17 randomlyab (207517)18 trialab (308477)19 groupsab (1318386)20 OR13-1921 3 AND 6 AND 9 AND 12 AND 20 (139)22 limit 21 to yr=ldquo2008-Currentrdquo (56)
120Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Appendix 3 EMBASE search strategy via Ovid (for 2015 update)
1 exp Anti-inflammatory Agents non-steroidal (452266)2 (bufexamac OR bufexine OR calmaderm OR ekzemase OR dicoflenac OR solaraze OR pennsaid OR voltarol OR emugel OR
voltarene OR voltarol OR optha OR voltaren OR etofenamate OR afrolate OR algesalona OR bayro OR deiron OR etofen ORflexium OR flogoprofen OR rheuma-gel OR rheumon OR traumalix OR traumon OR zenavan OR felbinac OR dolinac OR flexfreeOR napageln OR target OR traxam OR fentiazac OR domureuma OR fentiazaco OR norvedan OR riscalon OR fepradinol ORdalgen OR flexidol OR cocresol OR rangozona OR reuflodol OR pinazone OR zepelin OR flufenamic OR dignodolin OR rheumaOR lindofluid OR sastridex OR lunoxaprofen OR priaxim OR flubiprofen OR fenomel OR ocufen OR ocuflur OR ldquoTrans ActLATrdquo OR tulip OR ibuprofen OR cuprofen OR ldquodeep relief rdquo OR fenbid OR ibu-cream OR ibugel OR ibuleve OR ibumousse ORibuspray OR ldquonurofen gelrdquo OR proflex OR motrin OR advil OR radian OR ralgex OR ibutop OR indomethacin OR indocin ORindospray OR isonixin OR nixyn OR ketoprofen OR tiloket OR oruvail OR powergel OR solpaflex OR ketorolac OR acular ORtrometamol OR meclofenamic OR naproxen OR naprosyn OR niflumic OR actol OR flunir OR niflactol topico OR niflugel ORnifluril OR oxyphenbutazone OR californit OR diflamil OR otone OR tanderil OR piketoprofen OR calmatel OR triparsean ORpiroxicam OR feldene OR pranoprofen OR oftalar OR pranox OR suxibuzone OR danilon OR flamilon OR ufenamate OR fenazolOR flector OR benzydamine)mp (790435)
3 1 OR 2 (1108683)4 exp Administration Topical (68490)5 (topical OR cutaneous OR dermal OR transcutaneous OR transdermal OR percutaneous OR skin OR massage OR
embrocation OR gel OR ointment OR aerosol OR cream OR cregraveme OR lotion OR mouse OR foam OR liniment OR spray OR rubOR balm OR salve OR emulsion OR oil OR patch OR plaster)mp (3208844)
6 4 OR 5 (3208844)7 exp Athletic Injuries (24675)8 (strain OR sprain OR contusion OR distortion OR compression OR ldquosports injurrdquo OR ldquosoft tissue injurrdquo OR tendnitis OR
ldquomuscle painrdquo OR periarthritis OR epicondylitis OR tenosynovitis)mp (804322)9 7 OR 8 (825164)
10 Pain (216406)11 (pain OR analgesi)mp (971593)12 10 OR 11 (971593)13 clinical trialssh (841252)14 controlled clinical trialssh (389335)15 randomized controlled trialsh (357169)16 double-blind proceduresh (118945)17 (clin adj25 trial)ab (331002)18 ((doubl or trebl or tripl) adj25 (blind or mask))ab (141726)19 placeboab (203390)20 randomab (906176)21 OR13-20 (1731380)22 3 AND 6 AND 9 AND 10 AND 19 (439)23 limit 22 to yr=ldquo2008-Currentrdquo (204)
Appendix 4 Summary of outcomes efficacy
Study ID Treatment Clinical response Other response
Airaksinen 1993 (1) Ketoprofen gel 2 x 5 g (125mg) daily n = 29(2) Placebo gel n = 27
PGE rdquoimprovedldquo at 7 days(1) 2429(2) 1427
No additional data
121Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
kermark 1990 (1) Indomethacin spray 1 (El-metacin) 3-5 x 05-15 mL dailyn = 23(2) Indomethacin capsules 3 x 25mg daily n = 23(3) Placebo spray and capsules n= 24
No pain on palpation at 7 days(1) 1222(2) 522(3) 624
Patient assessment of improve-ment at 7 days (scale 0 to 100)(1) 57(2) 49(3) 30
Aoki 1984 (1) Piroxicam gel 5 3 or 4 x 1 gdaily n = 84(2) Indomethacin gel 1 3 or 4x 1 g daily n = 84(3) Placebo gel n = 84
PGE (5 point) rdquobetter or muchbetterldquo at 7 days(1) 5672(2) 4164(3) 3367
Pain on movement rdquoreducedldquo orrdquodisappearedldquo at 7 days(1) 4861(2) 3860(3) 3563
Auclair 1989 (1) Niflumic acid gel 25 3 x 5g daily n = 117(2) Placebo gel n = 110
PGE (5 point) rdquogood or verygoodldquo at 7 days(1) 69117(2) 54110
Pain on palpation rdquoimprovedldquo at7 days(1) 69117(2) 53110
Billigmann 1996 (1) Ibuprofen microgel 5 3 x200 mg daily n = 80(2) Placebo gel n = 80
Complete remission(1) 2580(2) 1080
Improvement in pain with move-ment of 20 at 7 days(1) 6580(2) 5580
Campbell 1994 (1) Ibuprofen cream 5 (Proflex)4 x 4rdquo daily n = 26(2) Placebo cream n = 25
Improvement in walking ability (4point) at 7 days(1) 2126(2) 1925
No additional data
Chatterjee 1977 (1) Benzydamine HCl cream 33 x daily n = 25(2) Placebo cream n = 25
Pain on movement ldquoabsentslightrdquoat 6 days(1) 2125(2) 1225
Tenderness with pressure ldquoabsentslightrdquo at 6 days(1) 2125(2) 1225
Costantino 2011 (1) DHEP-heparin plaster (Flec-torparin) x 1 daily n = 142(2) DHEP plaster (Flector) x 1daily n = 146(3) Placebo plaster n = 142
No responder analysis for efficacyOverall treatment efficacy not re-ported
Mean reduction from baseline forPI on movement at 3 days (fromgraph)(1) 24 mm(2) 19 mm(3) 14 mm
Coudreuse 2010 (1) DHEP-heparin plaster (Flec-torparin) n = 120(2) Placebo n = 120
PGE ldquogoodrdquo or ldquoexcellentrdquo at 7days(1) 99117(2) 82116These are results for physician-re-ported judgement - not partici-pant
Mean reduction from baseline forPI on movement over 6 hours(1) about 30(2) about 20Overall greater in DHEP-heparingroup than placebo over 7 days
122Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Curioni 1985 (1) Ibuproxam n = 20(2) Ketoprofen n = 20(3) Etofenamate n = 20
Resolution of symptoms by 7 days(1) 1520(2) 1320(3) 1320
PGE ldquogoodrdquo or ldquoexcellentrdquo at 10days(1) 1920(2) not reported(3) 1620
Diebshlag 1990 (1) Ketorolac gel 2 3 x 3 g dailyn = 13(2) Etofenamate gel 5 3 x 3 gdaily n = 12(3) Placebo gel n = 12
Improvement in pain at 7 days(1) 1213(2) 1012(3) 912
No additional data
Dreiser 1988 (1) Ibuprofen cream 5 3 x 4 cmdaily n = 32 (3 x 10 cm for largejoints)(2) Placebo cream n = 32
PGE ldquoimprovementrdquo or ldquocom-plete relief rdquo at 7 days(1) 2632(2) 1232
(1) significantly better than (2)for mean improvement in sponta-neous pain movement pain restpain tenderness to pressure (VAS)
Dreiser 1989 (1) Ketoprofen gel 25 2 x 5 cmdaily n = 30(2) Placebo gel n = 30
PGE (3 point) ldquobetterrdquo at 7 days(1) 1830(2) 530
(1) significantly better than (2) formean improvement in pain (restand movement) (VAS)
Dreiser 1990 (1) Niflumic acid gel 25 3 x 5g daily n = 30(2) Placebo gel n = 30
PGE (4 point) ldquocuredrdquo or ldquoim-provedrdquo at 7 days(1) 2330(2) 1030
(1) significantly better than (2) formean improvement in pain (VAS)
Dreiser 1994 (1) Flurbiprofen plaster 2 x 40 mgdaily n = 65(2) Placebo plaster n = 66
PGE (4 point) ldquogoodrdquo or ldquoverygoodrdquo at 7 days(1) 4865(2) 4166
(1) significantly better than (2)for mean improvement in sponta-neous pain but not pain on move-ment or palpation (VAS)
Fioravanti 1999 (1) DHEP lecithin gel 3 x 5 g (=65 mg) daily n = 50(2) DHEP gel 3 x 5 g (= 65 mg)daily n = 50
PGE (4 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 10 days(1) 3550(2) 3550
(1) significantly better than (2)for mean improvement in spon-taneous pain at 7 days but notfor pain on movement at 10 days(VAS)
Fujimaki 1985 (1) Piroxicam gel 05 3 or 4 x 1g daily n = 92(2) Indomethacin gel 1 3 or 4 x1 g daily n = 90(3) Placebo gel n = 89
PGE (5 point) ldquobetterrdquo or ldquomuchbetterrdquo at end of treatment at 14days(1) 4483(2) 4482(3) 4082
No additional data
Gallacchi 1990 (1) Diclofenac hy-droxyethylpyrrolidine gel 1 4 x2 g daily n = 25 (Flector gel)(2) Diclofenac sodium 1 4 x 2 g
PGE (5 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 14 days(1) 1925(2) 1925
No significant difference betweengroups for pain on applied pres-sure at 7 and 14 days
123Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
daily n = 25 (Voltaren Emugel)
Gonzaacutelez de Vega 2013 (1) Traumeel gel 3 x 2 g daily n =140(2) Traumeel ointment n = 143(3) Diclofenac gel 1 3 x 2 g dailyn = 137
Pain-free at 7 days(1) 7140(3) 8137PGE (5 point) ldquogoodrdquo or ldquoverygoodrdquo(1) 128140(3) 127137
Normal function (5 point) scoreof 0 or 1 at 14 days(1) 133140(3) 131137
Governali 1995 (1) Ketoprofen gel 5 3 x 2-3 gdaily n = 15(2) Ketoprofen cream 1 3 x 2-3 g daily n = 15
PGE (5 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 7 days(1) 1415(2) 415
No additional data
Gualdi 1987 (1) Flunaxaprofen gel 2 x 3-5 cmdaily n = 30(2) Ketoprofen gel 2 x 3-5 cmdaily n = 30
No dichotomous data No significant difference betweengroups for pain on movement at 7days
Haig 1986 (1) Benzydamine cream 3 6 xdaily n = 21(2) Placebo cream n = 22
Pain on movement ldquoimprovedrdquo by6 days(1) 1821(2) 1322
No additional data
Hoffmann 2012 (1) DHEP-heparin plaster (Flec-torparin) x 1 daily n = 121(2) DHEP (Flector) plaster x 1daily n = 115(3) Placebo plaster n = 118
PGE (5 point) ldquogoodrdquo or ldquoexcel-lentrdquo over 80 in all groups
Mean reduction from baseline inPI (100 mm VAS)Day 3(1) 191 mm(2) 114 mm(3) 52 mmDay 8(1) 443 mm(2) 372 mm(3) 296 mm
Hofman 2000 (1) Diclofenac sodium gel 1 4x 2 cm daily n = 69(2) Lysine clonixinate gel 5 4 x2 cm (225 mg) daily n = 73
PGE (3 point) at 8 days ldquogoodrdquo(1) 3869(2) 3673
No significant difference betweentreatments for any pain outcomes
Hosie 1993 (1) Felbinac foam 3 3 x 2 g daily+ placebo tablets 3 x 1 daily n =140 (127 analysed for efficacy)(2) Ibuprofen tablets 3 x 400 mgdaily + placebo foam 3 x 2 g dailyn = 147 (134 analysed for efficacy)
Pain on movement ldquononerdquo orldquomildrdquo at 7 days(1) 81127(2) 96133
Spontaneous pain ldquononerdquo orldquomildrdquo at 7 days(1) 99127(2) 108134
124Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Jenoure 1997 (1) DHEP plaster (Tissugel) 2 xdaily n = 44(2) Placebo plaster 2 x daily n =41
Baseline pain in two groups notbalanced and data in table and fig-ure do not agree so efficacy out-comes not used
No additional data
Joussellin 2003 (1) DHEP plaster (Flector Tis-sugel 1) 1 x daily n = 68(2) Placebo plaster 1 x daily n =66
PGE (4 point) ldquoexcellentrdquo at 7 days(1) 3668(2) 2466ge 30 reduction in PI at 7 days(1) 2568(2) 1166
(1) significantly better than (2) formean pain on movement at 6 days
Julien 1989 (1) Ketoprofen gel 25 2 x 5 cm(= 50 mg) daily n = 30(2) Placebo gel n = 30
PGE (4 point) ldquorecoveredrdquo at 7days(1) 1830(2) 630
PGE (4 point) ldquorecoveredrdquo or ldquoim-provedrdquo at 7 days(1) 2530(2) 1330
Klainguti 2010 (1) DHEP-heparin plaster (Flec-torparin) x 1 daily n = 62(2) DHEP plaster (Flector) x 1daily n = 61(3) Placebo n = 59
Overall treatment efficacy (partic-ipant and investigator) (5 point)ldquogoodrdquo or ldquoexcellentrdquo at 3 days(1) 5662(2) not reported(3) 4659
Mean reduction in PI at day 3(from graph)(1) 33 mm(2) Not reported(3) 24 mm
Kockelbergh 1985 (1) Ketoprofen gel 25 2 x 5 cm(= 15 mg) daily n = 38(2) Placebo gel n = 36
PGE (3 point) ldquogoodrdquo at 7 days(1) 3038(2) 2236
(1) and (2) slightly better than (3)for mean spontaneous pain at 7days
Kuehl 2011 (1) DETP 13 plaster 2 x dailyn = 207(2) Placebo plaster n = 211
No dichotomous data Percentage reduction from base-line at last application(1) 73(2) 62
Li 2013 (1) DHEP plaster (Flector) 2 xdaily n = 192(2) Placebo plaster 2 x daily n =192
ge 50 pain reduction at day 7(posthoc analysis)(1) 173192(2) 114192
Overall treatment efficacy (5point) ldquogoodrdquo or ldquoexcellentrdquo(1) 161192(2) 81192Participant and investigator ratingMean plusmn SD reduction in PI onmovement at 7 days(1) 538 plusmn 170(2) 370 plusmn 183
Linde 1985 (1) Benzydamine 3 cream 3 xdaily n = 50(2) Placebo gel n = 50
No pain on movement (walking)at 8 days(1) 3550(2) 4050
No additional data
125Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Machen 2002 (1) Ibuprofen gel 5 3 x daily n= 40(2) Placebo gel n = 41
PGE (5 point) ldquomarked improve-mentrdquo or ldquocomplete clearancerdquo at7 days(1) 2540(2) 941
Clinically meaningful (ge 30 mm)pain relief at day 7(1) 3040(2) 1641
Mahler 2003 (1) DHEP + lethicin gel 3 x 5 gdaily n = 52(2) DHEP gel 3 x 5 g daily n = 48
PGE (4 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 10 days(1) 4952(2) 3948
Mean reduction in pain on move-ment at 3 and 10 days significantlygreater with (1) than (2)
Maziegraveres 2005b (1) Ketoprofen plaster 100 mg x1 daily n = 81(2) Placebo plaster n = 82
PGE (4 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 14 days(1) 7281(2) 6082
All mean efficacy measures im-proved more for (1) than (2) mostwere statistically significant
Maziegraveres 2005a (1) Ketoprofen plaster 100 mg x1 daily n = 87(2) Placebo plaster n = 85
PGE (4 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 14 days(1) 5087(2) 4185
All mean efficacy measures im-proved more for (1) than (2) mostwere statistically significant
McLatchie 1989 (1) Felbinac gel 3 3 x 3 cm dailyn = 118(2) Placebo gel n = 113
No dichotomous data Patient daily self-assessment formean pain on rest movement atnight interference with normaland leisure activities show betterefficacy for (1) than (2) from day2 (VAS)
Morris 1991 (1) Felbinac gel 3 3 x 1 cm dailyn = 41(2) Placebo gel n = 43
PGE (5 point) ldquogoodrdquo or ldquoverygoodrdquo at 7 days(1) 2341(2) 2743
(1) better than (2) for meanimprovement in symptoms andsporting function at 7 days
NCT01255423 (1) Diclofenac sodium gel 1 x 4daily n = 104(2) Placebo gel x 4 daily n = 100
No dichotomous data VAS (mean plusmn SD) at 72 hours(baseline PI not reported)(1) 374 plusmn 252(2) 388 plusmn 241
NCT01272934 (1) Diclofenac sodium gel 1 x 4daily n = 104(2) Placebo gel x 4 daily n = 100
No dichotomous data VAS (mean plusmn SD) at 72 hours(baseline PI not reported)(1) 256 plusmn 159(2) 612 plusmn 166
NCT01272947 (1) Diclofenac sodium gel 1 x 4daily n = 104(2) Placebo gel x 4 daily n = 100
No dichotomous data VAS (mean plusmn SD) at 24 hours(baseline PI not reported)(1) 331 plusmn 214(2) 654 plusmn 169
126Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Noret 1987 (1) Ketoprofen gel 25 2 x 5 cm(75 mg) daily n = 48(2) Placebo gel n = 45
PGE (4 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 8 days(1) 3948(2) 945
Decrease in mean spontaneouspain significantly greater in (1)than (2) by 3 days
Parrini 1992 (1) Ketoprofen foam 15 3 x 2 g(200 mg) daily n = 83(2) Placebo foam n = 86
No dichotomous data Mean pain on movement andpressure significantly decreased by7 days in (1) compared with (2)
Picchio 1981 (1) Ibuprofen gel 10 3 x daily n = 20(2) Ketoprofen gel 1 3 x daily n= 20
No pain on movement at 8 days(1) 320(2) 020
Spontaneous pain ldquononerdquo at 8days(1) 620(2) 020
Predel 2004 (1) Diclofenac sodium plaster 2 xdaily (140 mgplaster) n = 60(2) Placebo plaster n = 60
PGE (4 point) ldquogoodrdquo ldquoexcellentrdquoat 7 days(1) 5560(2) 560
(1) better than (2) for reductionin tenderness pain and speed ofpain reduction
Predel 2012 (1) Diclofenac gel (VoltarenEmulgel 232) 2 x 5 cm daily n= 80(2)Diclofenac gel (Voltaren Emulgel232 ) 3 x 5 cm daily n = 80(3) Placebo gel n = 82
ge 50 red in PI on movement at5 days(1) 5780(2) 5980(3) 1782
PGE efficacy (5 point) ldquogoodrdquo orldquovery goodrdquo at 8 days(1) 6880(2) 7380(3) 2482
Predel 2013a (1) Diclofenac 4 spray gel 4 or5 sprays 3 x daily (96-120 mg di-clofenac sodium) n = 118(2) Placebo spray gel n = 114
None or slight PI on movement at7-8 days(1) 111118(2) 93114
None or slight PI on movement at3-4 days(1) 76118(2) 58114
Predel 2013b (1) Diclofenac gel (VoltarenEmulgel) 4 x 2 g daily n = 36(2) Placebo gel 4 x 2 g daily n =36
PGE (5 point) ldquogoodrdquo or ldquoexcel-lentrdquo(1) 3636(2) 736
ge 50 reduction in PI on move-ment after 48 hours(1) 3436(2) 336
Ramesh 1983 (1) Ibuprofen cream 5 3 or 4 x5-10 cm daily n = 40(2) Placebo cream n = 40
Pain on movement (4 point)ldquononerdquo or ldquoslightrdquo at 7 days(1) 2340(2) 2340
Physician global assessment at 10days ldquogoodrdquo(1) 2940(2) 1640
Rowbotham 2003 (1) Diclofenac epolamine plaster(Flector Tissuegel) 2 x daily n =191(2) Placebo plaster n = 181
Pain intensity le 210 for 2 daysor 4 consecutive evaluations by 7days(1) 75191(2) 48181
Mean pain on rest significantlybetter with (1) than (2) after 7 days
127Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Russell 1991 (1) Piroxicam gel 05 4 x 5 mgdaily n = 100(2) Placebo gel n = 100
PGE (4 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 8 days(1) 79100(2) 45100
Statistically greater reduction inmean pain on movement at 8 dayswith (1) than (2)
Saillant 1998 (1) DHEP plaster (Flector Tis-sugel 1) 1 x daily n = 70(2) Placebo plaster 1 x daily n =70
PGE (4 point) ldquoexcellentrdquo at 7 days(1) 4370(2) 2570
ge 30 decrease in PI at 7 days(1) 6470(2) 5070
Sanguinetti 1989 (1) Felbinac gel 3 3 x daily n =42(2) Placebo gel n = 40
PGE ldquogoodrdquo or ldquovery goodrdquo at 7days(1) 3442(2) 1140
(1) better than (2) by 2 days
Sinniger 1981 (1) Fentiazac cream 5 2 or 3 xdaily n = 10(2) Placebo cream n = 10
Complete pain relief within 10days(1) 710(2) 110
Improvement in active pain onmovement at 5 days(1) 67(2) 32
Spacca 2005 (1) DHEP lecithin gel (Effigel) 3x 5 g daily n = 79(2) Placebo gel n = 76
No dichotomous data Mean pain scores improved morerapidly in (1) than (2) - statisticallysignificant at 3 and 6 days
Sugioka 1984 (1) Piroxicam gel 05 3 or 4 x 1g daily n = 183(2) Indomethacin gel 1 3 or 4 x1 g daily n = 183
PGE (5 point) ldquobetterrdquo or ldquomuchbetterrdquo at 14 days(1) 85175(2) 55165
Pain on movement ldquoreducedrdquo orldquodisappearedrdquo at 7 days(1) 77175(2) 63165
Thorling 1990 (1) Naproxen gel 10 2-6 x dailyn = 60(2) Placebo gel n = 60
PGE (5 point) ldquogoodrdquo or ldquoverygoodrdquo at 7 days(1) 3860(2) 2760
Participants using naproxen im-proved more rapidly and had sig-nificantly lower severity scores byday 3
Tonutti 1994 (1) Ketoprofen gel 5 3 x 2-3 gdaily n = 15(2) Etofenamate gel 5 3 x 2-3g n = 15
PGE (4 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 7 days(1) 1015(2) 1115
Significant reductions in pain onmovement by 7 days in bothgroups
Vecchiet 1991 (1) Meclofenamic acid gel 5 2 x10 cm daily (2 g) n = 30(2) Placebo n = 30
PGE (4 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 10 days(1) 3030(2) 1930
(1) significantly better than (2)for mean improvement in sponta-neous pain movement pain func-tional restriction
Whitefield 2002 (1) Ibuprofen gel 5 + placebotablet 3 x daily n = 50(2) Ibuprofen 400 mg tablet +placebo gel 3 x daily n = 50
Participant satisfied at 7 days(1) 3050(2) 3650
ldquoCompletely betterrdquo at 14 days(1) 2450(2) 3050
128Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
DHEP diclofenac epolamine HCl hydrochloride n number PGE participant global evaluation PI pain intensity SD standarddeviation VAS visual analogue scale
Appendix 5 Summary of outcomes adverse events and withdrawals
Study ID Treatment Local AEs Systemic AEs Serious AEs Withdrawals
Airaksinen 1993 (1) Ketoprofen gel 2x 5 g (125 mg) dailyn = 29(2) Placebo gel n =27
(1) 529(2) 427
(1) 129 (nausea af-ter paracetamol)(2) 027
None AE noneOther nonereported
kermark 1990 (1)Indomethacin spray1 (Elmetacin) 3-5 x 05-15 mL dailyn = 23(2) Indomethacincapsules 3 x 25 mgdaily n = 23(3) Placebo sprayand capsules n = 24
(1) 422(2) 022(3) 024
No usable data -reported for eventsnot participants
None reported AE (1) 1 (2) 1 (3)0Lost to follow-up(1) 1 (2) 2 (3) 3
Aoki 1984 (1) Piroxicam gel5 3 or 4 x 1 gdaily n = 84(2) Indomethacingel 1 3 or 4 x 1 gdaily n = 84(3) Placebo gel n =84
(1) 179(2) 270(3) 274
None None reported AE none23 excluded for pro-tocol violations (1)7 (2) 7 (3) 926 withdrew for rea-sons unrelated totreatment (1) 5 (2)13 (3) 8
Auclair 1989 (1) Niflumic acidgel 25 3 x 5 gdaily n = 117(2) Placebo gel n =110
All AEs(1) 5123(2) 6116Most commonly cu-taneous eruptions
No usable data None reported AE (1) 1123 (2) 011626 excluded fromefficacy analysis fornot meeting entrycriteria and protocolviolations
Billigmann 1996 (1) Ibuprofen mi-crogel 5 3 x 200mg daily n = 80(2) Placebo gel n =
(1) 1180(2) 480
None reported None reported AE (1) 280 (aller-gic rash dermatitis)No reason given (1)380 (2) 580
129Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
80 Symptom-free (1)180 (2) 180
Campbell 1994 (1) Ibuprofen cream5 (Proflex) 4 x 4rdquodaily n = 26(2) Placebo cream n= 25
No data (1) 126 (headache)(2) 025
No data AE noneExclusions 493 presented late 2missing forms 1 ap-peared twice 43 didnot return diaries
Chatterjee 1977 (1) BenzydamineHCl cream 3 3 xdaily n = 25(2) Placebo cream n= 25
None None None AE none1 participant lostto follow-up (groupnot reported)
Costantino 2011 (1) DHEP-hep-arin plaster (Flector-parin) x 1 daily n =142(2) DHEP plaster(Flector) x 1 daily n= 146(3) Placebo plastern = 142
Possibly or probablydrug-related(1) 2142(2) 1145(3) 6142
All mild and goneby 14 days ex-cept 1 (moderate) inplacebo group
Not reportedNo systemic GIevents
None AE noneOther(1) 0142(2) 2146 (proto-col deviation lost tofollow-up)(3) 2142 (lost tofollow-up)
Coudreuse 2010 (1) DHEP-hep-arin plaster (Flector-parin) n = 120(2) Placebo n = 120
(1) 1117(2) 7116
All AEs mild ormoderate resolvedspontaneously
(1) 1117(2) 0116
None AE(1) 1117 (increasedoedema)(2) 0116Other(1) 6117 (recovery3 no reason 3)(2) 3116 (recovery1 no reason 2)
Curioni 1985 (1) Ibuproxam n =20(2) Ketoprofen n =20(3) Etofenamate n =20
None None None AE noneOther none
Diebshlag 1990 (1) Ketorolac gel2 3 x 3 g daily n =13(2) Etofenamate gel5 3 x 3 g daily n =12
(1) 113(2) 112(3) 012
None None AE none1 ketorolac partic-ipant did not at-tend 15 day follow-up due to car acci-
130Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
(3) Placebo gel n =12
dent
Dreiser 1988 (1) Ibuprofen cream5 3 x 4 cm dailyn = 32 (3 x 10 cmfor large joints)(2) Placebo cream n= 32
No usable data None Not reported AE none4 placebo partici-pants lost to follow-up
Dreiser 1989 (1) Ketoprofen gel25 2 x 5 cm dailyn = 30(2) Placebo gel n =30
(1) 030(2) 230
None None reported AE (2) 230 (intol-erance)LoE (1) 130 (2) 130
Dreiser 1990 (1) Niflumic acidgel 25 3 x 5 gdaily n = 30(2) Placebo gel n =30
(1) 030(2) 330
None None AE (2) 130 (ery-thema)Exclusion 1 from(2) from efficacyanalysis for inade-quate baseline pain
Dreiser 1994 (1)Flurbiprofen plaster2 x 40 mg daily n =65(2) Placebo plastern = 66
(1) 265(2) 066
None None AE 0(1) 165 excludedfrom efficacy analy-sis for protocol vio-lation(2) 266 (1 LoE 1cured)
Fioravanti 1999 (1) DHEP lecithingel 3 x 5 g (= 65 mg)daily n = 50(2) DHEP gel 3 x 5g (= 65 mg) daily n= 50
(1) 050(2) 150
No data None reported AE noneOther none
Fujimaki 1985 (1) Piroxicam gel 05 3 or 4 x 1 g dailyn = 92(2) Indomethacingel 1 3 or 4 x 1 gdaily n = 90(3) Placebo gel n =89
(1) 183(2) 582(3) 282
(1) 083(2) 182 (nauseaand vomiting)(3) 082
None AE (1) 0 (2) 4 (3)0Unknown reasons(1) 2 (2) 1Did not return af-ter 1st visitirregularvisits (1) 6 (2) 6(3) 7
Gallacchi 1990 (1) Diclofenac hy-droxyethylpyrro-lidine gel 1 4 x 2
No AEs None None AE noneOther none
131Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
g daily n = 25 (Flec-tor gel)(2) Di-clofenac sodium 14 x 2 g daily n = 25(Voltaren Emugel)
Gonzaacutelez de Vega2013
(1) Traumeel gel 3 x2 g daily n = 140(2) Traumeel oint-ment n = 143(3) Diclofenac gel1 3 x 2 g daily n =137
No usable data
Infrequent mild tomoderate
No usable data None AE noneOther(1) 11140(2) 12137
Governali 1995 (1) Ketoprofen gel5 3 x 2-3 g dailyn = 15(2) Ketoprofencream 1 3 x 2-3 gdaily n = 15
No side effects None None AE noneOther none
Gualdi 1987 (1) Flunaxaprofengel 2 x 3-5 cm dailyn = 30(2) Ketoprofen gel 2x 3-5 cm daily n =30
(1) 130(2) 330
No data None reported AE noneOther none
Haig 1986 (1) Benzydaminecream 3 6 x dailyn = 21(2) Placebo cream n= 22
No AEs reported None None reported AE none reportedOther no data
Hoffmann 2012 (1) DHEP-hep-arin plaster (Flector-parin) x 1 daily n =121(2) DHEP (Flector)plaster x 1 daily n =115(3) Placebo plastern = 118
Most AEs were mi-nor local reactions(eg pruritus anderythema) in areaof plaster of mildto moderate inten-sity and resolvedwithout interrupt-ing treatment
No treatment-related systemic AEsrecorded
(2) 1 participanthad 3 SAEs nonejudged related tostudy medication
AE noneExclusions(1) 5121(2) 10115(3) 7118Excluded fromper protocol analysisdue to poor compli-ance or personal de-cision
Hofman 2000 (1) Diclofenacsodium gel 1 4 x2 cm daily n = 69(2) Lysine clonixi-
(1) 158(2) 161
None None AE noneLoE (1) 9 (2) 8
132Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
nate gel 5 4 x 2cm (225 mg) dailyn = 73
Hosie 1993 (1) Felbinac foam3 3 x 2 g daily+ placebo tablets 3x 1 daily n = 140(127 analysed for ef-ficacy)(2) Ibupro-fen tablets 3 x 400mg daily + placebofoam 3 x 2 g daily n= 147 (134 analysedfor efficacy)
(1) 1127(2) 3134
GI events (1) 14127 (2) 11134For (1) more mildnone definitely drugrelated for(2) definitely relatedto study drug
None AE noneExclusions (1) 13(2) 13 did not returnfor 7 day follow-up
Jenoure 1997 (1) DHEP plaster(Tissugel) 2 x dailyn = 44(2) Placebo plaster 2x daily n = 41
(1) 144(2) 141
No data None reported AE none reportedOther nonereported
Joussellin 2003 (1) DHEPplaster (Flector Tis-sugel 1) 1 x dailyn = 68(2) Placebo plaster 1x daily n = 66
(1) 168 (pruritus)(2) 366 (pruritus 2burning 1)
All AEs mild ormoderate
(1) 168 (allergic re-action)(2) 066
None reported AE(1) 066(2) 166Other(1) 366(2) 266
Julien 1989 (1) Ketoprofen gel25 2 x 5 cm (= 50mg) daily n = 30(2) Placebo gel n =30
(1) 130(2) 030
Not reported None AE noneOther none
Klainguti 2010 (1) DHEP-hep-arin plaster (Flector-parin) x 1 daily n =62(2) DHEP plaster(Flector) x 1 daily n= 61(3) Placebo n = 59
(1) 062(2) 061(3) 159All AEs mild in na-ture and resolvedspontaneously
(1) 165 (facial in-fection)(2) 161 (abdominalpain)(3) 059
None AE noneOther(1) 365(2) 161(3) 359
Kockelbergh 1985a (1) Ketoprofen gel25 2 x 5 cm (= 15mg) daily n = 38(2) Placebo gel n =
(1) 138(2) 126
Not reported None AE noneOther none
133Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
36
Kuehl 2011 (1) DETP 13plaster 2 x daily n =207(2) Placebo plastern = 211
(1) 16207(2) 12211
(1) 15207(2) 23211
None AE(1) 4207(2) 9211LoE(1) 21207(2) 25211Other(1) 58207(2) 62211
Li 2013 (1) DHEP plaster(Flector) 2 x daily n= 192(2) Placebo plaster 2x daily n = 192
(1) 4192(2) 3192
(1) 10192(2) 4192
None AE(1) 2192(2) 0192Other(1) 5192(2) 1192
Linde 1985 (1) Benzydamine3 cream 3 x dailyn = 50(2) Placebo gel n =50
(1) 440(2) 241
None None AE none(1) 6 (2) 6 excludedfrom 1st assessment(1) 3 (2) 4 ex-cluded from final as-sessment
Machen 2002 (1) Ibuprofen gel5 3 x daily n = 40(2) Placebo gel n =41
(1) 440(2) 241
None None AE none(1) 1 LoE 1 proto-col violation(2) 4 LoE
Mahler 2003 (1) DHEP + lethicingel 3 x 5 g daily n =52(2) DHEP gel 3 x 5g daily n = 48
(1) 152(2) 048
(1) 152(2) 048
None AE none5 lost to follow-up
Maziegraveres 2005b (1) Ketoprofen plas-ter 100 mg x 1 dailyn = 81(2) Placebo plastern = 82
At 21 days(1) 1281(2) 682
(1) 1381(2) 1482
None AE (1) 381(1) 781 (1 LoE 6cured)(2) 782 (5 LoE 2cured)
Maziegraveres 2005a (1) Ketoprofen plas-ter 100 mg x 1 dailyn = 87(2) Placebo plastern = 85
At 21 days(1) 2987(2) 2785
(1) 1187(2) 785
None AE (1) 987 (2) 685(1) 687 (2 LoE 4cured)(2) 585 (4 LoE 1cured)
134Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
McLatchie 1989 (1) Felbinac gel 33 x 3 cm daily n =118(2) Placebo gel n =113
(1) 3118(2) 2113Mild transient localirritation
None reported None AE noneOther none
Morris 1991 (1) Felbinac gel 33 x 1 cm daily n =41(2) Placebo gel n =43
None None None AE none(1) 4 (protocol vio-lations)(2) 1 (lost to follow-up)Exclusions 11 fromefficacy analysis be-cause evaluated by4 different investi-gators
NCT01255423 (1) Diclofenacsodium gel 1 x 4daily n = 104(2) Placebo gel x 4daily n = 100
(1) 1104(2) 3102
Total AE(1) 11104(2) 8102
None None
NCT01272934 (1) Diclofenacsodium gel 1 x 4daily n = 104(2) Placebo gel x 4daily n = 100
(1) 1102(2) 0103
Total AE (excludingSAE)(1) 6102(2) 3103
(1) 0102(2) 1103 (rupturedligaments in wrist)
AE see SAE
NCT01272947 (1) Diclofenacsodium gel 1 x 4daily n = 104(2) Placebo gel x 4daily n = 100
None (1) 2104(2) 2100
None None
Noret 1987 (1) Ketoprofen gel25 2 x 5 cm (75mg) daily n = 48(2) Placebo gel n =45
(1) 151(2) 047
None reported Not reported AE (1) 151 (skinallergy)(1) 1 LoE 1 unre-lated to trial(2) 1 LoE 1 unre-lated to trial
Parrini 1992 (1) Ketopro-fen foam 15 3 x 2g (200 mg) daily n= 83(2) Placebo foam n= 86
None None None AE noneOther none
135Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Picchio 1981 (1) Ibuprofen gel10 3 x daily n =20(2) Ketoprofen gel1 3 x daily n = 20
None None None AE noneOther not reported
Predel 2004 (1) Diclofenacsodium plaster 2 xdaily (140 mgplas-ter) n = 60(2) Placebo plastern = 60
12 participants ex-pe-rienced 16 mild AEswith no differencesbetween groups
None None AE (1) 160Other none
Predel 2012 (1) Diclofenac gel(Voltaren Emulgel232) 2 x 5 cmdaily n = 80(2) Diclofenac gel(Voltaren Emulgel232) 3 x 5 cmdaily n = 80(3) Placebo gel n =82
(1) 080(2) 180(3) 182All mild to moder-ate
(1) 280(2) 180(3) 382
None AE(1) 080(2) 080(3) 182Other(1) 180(2) 280(3) 282(protocol violationslost to follow-up)
Predel 2013a (1) Diclofenac 4spray gel 4 or 5sprays 3 x daily (96-120 mg diclofenacsodium) n = 118(2) Placebo spraygel n = 114
(1) 1118(2) 4114
(1) 5118(2) 4114All AEs mild re-versible
None AE(1) 1118(2) 1114Other(1) 3118(2) 43114
Predel 2013b (1) Diclofenac gel(Voltaren Emulgel)4 x 2 g daily n = 36(2) Placebo gel 4 x 2g daily n = 36
None (1) 036(2) 136
None AE noneOther none
Ramesh 1983 (1) Ibuprofen cream5 3 or 4 x 5-10 cmdaily n = 40(2) Placebo cream n= 40
(1) 140(2) 140
None reported Not reported AE (1) 140 (2) 140Other none
Rowbotham 2003 (1) DHEP plaster(Flector Tissuegel) 2x daily n = 191(2) Placebo plastern = 181
(1) 27191 (pruritis14)(2) 31181 (pruritis21)
(1) 21191(2) 22181
None re-ported (ldquovast major-ity mildrdquo)
AE none(1) 3191 (2) 4181(did not finish trialand complete dailydiaries)
136Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Russell 1991 (1) Piroxicam gel 05 4 x 5 mg daily n= 100(2) Placebo gel n =100
(1) 4102(2) 10102
GI or CNS events(1) 4 (2) 7Any AE (1) 7102(2) 15102
None reported AE (1) 1102 (2) 8102Other (1) 6 LoE 1ldquootherrdquo(2) 42 LoEExclusions7 did not complywith study medica-tion schedule 6 lostto follow-up 1 pro-tocol violation
Saillant 1998 (1) DHEPplaster (Flector Tis-sugel 1) 1 x dailyn = 70(2) Placebo plaster 1x daily n = 70
None None None AE noneOther(1) 570(2) 570
Sanguinetti 1989 (1) Felbinac gel 33 x daily n = 42(2) Placebo gel n =40
(1) 342(2) 140
None None reported AE noneOther nonereported
Sinniger 1981 (1) Fentiazac cream5 2 or 3 x daily n= 10(2) Placebo cream n= 10
ldquoNo untoward sideeffectsrdquo
None None AE noneOther nonereported
Spacca 2005 (1) DHEP lecithingel (Effigel) 3 x 5 gdaily n = 79(2) Placebo gel n =76
ldquoNo signsof cutaneous irrita-tion or sensitisationobservedrdquo
No AEs observed None AE noneOther nonereported
Sugioka 1984 (1) Piroxicam gel 05 3 or 4 x 1 g dailyn = 183(2) Indomethacingel 1 3 or 4 x 1 gdaily n = 183
(1) 5178(2) 26179
None reported None reported AE none reportedExclusions due toprotocol violations(1) 8 (2) 18Withdrawals(1) 11 (2) 12
Thorling 1990 (1) Naproxen gel10 2-6 x daily n =60(2) Placebo gel n =60
(1) 160(2) 060
None None AE none(1) 1 LoE 1 proto-col violation(2) 1 participant re-quest
137Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Tonutti 1994 (1) Ketoprofen gel5 3 x 2-3 g dailyn = 15(2) Etofenamate gel5 3 x 2-3 g n =15
None No AEs attributableto the medication
None AE NoneLoE (1) 1 (2) 2
Vecchiet 1991 (1) Meclofe-namic acid gel 5 2x 10 cm daily (2 g)n = 30(2) Placebo n = 30
Tolerabil-ity excellent or goodin nearly all partici-pants
No data None AE none reported(2) 5 lost to follow-up
Whitefield 2002 (1) Ibuprofen gel5 + placebo tablet3 x daily n = 50(2) Ibuprofen 400mg tablet + placebogel 3 x daily n = 50
No data 6 AEs reportednone judged relatedto study medication
None reported AE noneRecovered (1) 3 (2)2LoE (2) 1Lost to follow-up(1) 1 (2) 1
AE adverse event CNS central nervous system DHEP diclofenac epolamine GI gastrointestinal HCl hydrochloride LoE lackof efficacy n number SAE serious adverse event
Appendix 6 Concentration amount and frequency of dosing
Study Drug Concentration Quantity Frequency Estimated daily dose of topi-cal NSAID
Joussellin 2003 Diclofenac 1 Plaster 1 180 mg epolamine salt 140 mgNa salt
Li 2013 Diclofenac 1 Plaster 2 360 mg epolamine salt 280 mgNa salt
Rowbotham 2003 Diclofenac 1 Plaster 2 360 mg epolamine salt 280 mgNa salt
Saillant 1998 Diclofenac 1 Plaster 1 180 mg epolamine salt 140 mgNa salt
Coudreuse 2010 Diclofenac 1 Plaster 1 180 mg epolamine salt 140 mgNa salt
Klainguti 2010 Diclofenac 1 Plaster 1 180 mg epolamine salt 140 mgNa salt
138Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Predel 2004 Diclofenac - Plaster 2 280 mg Na salt
Predel 2012 Diclofenac 232 5 cm ribbon~ 2 g 2 or 3 92-138 mg (Na equiv) as di-ethylamine salt
Predel 2013b Diclofenac 116 2 g 4 92 mg (Na equiv) as diethy-lamine salt
Predel 2013a Diclofenac 4 4-5 sprays 3 96-120 mg Na salt
Costantino 2011 Diclofenac 1 Plaster 1 180 mg epolamine salt 140 mgNa salt
Fioravanti 1999 Diclofenac - 5 g 3 195 mg epolamine salt
Gallacchi 1990 Diclofenac 1 2 g 4 80 mg
Gonzaacutelez de Vega2013
Diclofenac 1 6 cm 2g 3 60 mg
Hoffmann 2012 Diclofenac 1 Plaster 1 180 mg epolamine salt 140 mgNa salt
Hofman 2000 Diclofenac 1 2 cm 4 probably 30-40 mg Na salt
Jenoure 1997 Diclofenac 1 Plaster 2 360 mg epolamine salt 280 mgNa salt
Mahler 2003 Diclofenac - 5 g 3 195 mg epolamine salt
NCT01255423 Diclofenac 1 - 4 -
NCT01272934 Diclofenac 1 - 4 -
NCT01272947 Diclofenac 1 - 4 -
Spacca 2005 Diclofenac - 5 g 3 195 mg epolamine salt
Campbell 1994 Ibuprofen 5 4 inch ribbon 4 Assume 800 mg
Dreiser 1988 Ibuprofen 5 4 cm ribbon(10 cm for largerjoints)
4 Assume up-800 mg
Ramesh 1983 Ibuprofen 5 5-10 cm ribbon 3-4 Assume 300-800 mg
Billigmann 1996 Ibuprofen 5 10 cm 4 g gel 3 600 mg
139Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Machen 2002 Ibuprofen 5 - 3 -
Picchio 1981 Ibuprofen 10 - 3 -
Whitefield 2002 Ibuprofen 5 - 3 -
Maziegraveres 2005b Ketoprofen - Plaster 1 100 mg
Maziegraveres 2005a Ketoprofen - Plaster 1 100 mg
Airaksinen 1993 Ketoprofen - 5 g 2 125 mg
Dreiser 1989 Ketoprofen 25 5 cm 2 100 mg
Julien 1989 Ketoprofen 25 5 cm 2 100 mg
Noret 1987 Ketoprofen 25 5 cm 75 g 2 375 mg
Curioni 1985 Ketoprofen No details 2 -
Governali 1995 Ketoprofen 5 2-3 g 3 300-450 mg
Governali 1995 Ketoprofen 1 2-3 g 3 60-90 mg
Gualdi 1987 Ketoprofen - 3-5 cm 2 -
Kockelbergh 1985 Ketoprofen 25 5 cm 2 100 mg
Parrini 1992 Ketoprofen 150 2 g 3 600 mg
Picchio 1981 Ketoprofen 1 - 3 -
Tonutti 1994 Ketoprofen 5 2-3 g 3 300-450 mg
kermark 1990 Indomethacin 1 05-15 mL 3-5 12-60 mg
Aoki 1984 Indomethacin 1 1 g 3-4 30-40 mg
Fujimaki 1985 Indomethacin 1 1 g 3-4 30-40 mg
Sugioka 1984 Indomethacin 1 1 g 3-4 30-40 mg
Aoki 1984 Piroxicam 05 1 g 3-4 15-20 mg
Fujimaki 1985 Piroxicam 05 1 g 3-4 15-20 mg
140Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Russell 1991 Piroxicam 05 5 mg 4 20 mg
Sugioka 1984 Piroxicam 05 1 g 3-4 15-20 mg
Chatterjee 1977 Benzydamine 3 - 3 -
Haig 1986 Benzydamine 3 - 6 -
Linde 1985 Benzydamine 3 - 3 -
Auclair 1989 Niflumic acid 25 10 cm 5 g 3 375 mg
Dreiser 1990 Niflumic acid 25 10 cm 5 g 3 375 mg
Curioni 1985 Ibuproxam gel 10 - 2 -
Curioni 1985 Etofenamate No details - 2 -
Diebshlag 1990 Etofenamate 5 3 g 3 450 mg
Tonutti 1994 Etofenamate 5 2-3 g 3 300-450 mg
Diebshlag 1990 Ketorolac 2 3 g 3 360 mg
Dreiser 1994 Flurbiprofen Patch 2 80 mg
Gualdi 1987 Flunoxaprofen - 3-5 cm 2 -
Hofman 2000 Lysine clonixinate 5 2 cm 4 90 mg
Hosie 1993 Felbinac 3 2 g 3 180 mg
McLatchie 1989 Felbinac 3 3 cm 3 -
Morris 1991 Felbinac 3 1 cm 3 -
Sanguinetti 1989 Felbinac 3 - 3 -
Sinniger 1981 Fentiazac 5 Varied according toinvolved areas
2-3 -
Thorling 1990 Naproxen 10 - 2-6 -
Vecchiet 1991 Meclofenamic acid 5 10 cm 4 g 2 400 mg
equiv equivalent Na sodium
141Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
F E E D B A C K
Query on formulations of topical NSAIDs particularly DMSO from Dr Chrubasik 11 April 2012
Summary
Dr Chrubasik highlighted this letter to the Editor postgradmedorgdoi103810pgm2011092482
DMSO [dimethyl sulphoxide] but also other additives eg nonivamide (which is a capsaicinoid added as drug enhancer) may contributeto the overall effect of topical NSAIDs Nonivamide certainly contributes to the analgesic effect and to adverse events (heat sensationburning pruritus etc) This has not been considered in the Cochrane review by Massey T Derry S Moore RA McQuay HJ TopicalNSAIDs for acute pain in adults but Dr Chrubasik believes should be done otherwise the effect size of the NSAID topicals is favoured
Reply
We have been asked by Dr Chrubasik to comment on a letter (Roth 2011) about the formulations of topical NSAIDs particularly howDMSO and other penetration enhancers can affect efficacy estimates or adverse event reporting in osteoarthritis It was suggested thatthe review of Topical NSAIDs for acute pain in adults did not consider this resulting in a bias towards the topical NSAIDThere are a number of points to be made here
1 Penetration enhancers are used in formulations of topical products to encourage local absorption through the skin and produce ahigh local concentration Topical NSAIDs use penetration enhancers and the result is high local concentration in joints for instancebut low systemic concentrations (Moore 2008) That is how they work Formulation is an important part of medicinal chemistry as awhole not just for topical agents
2 In our analysis of topical NSAIDs we were aware that a range of properties are or have been ascribed to the analgesia resultingfrom application of topical agents and which could contribute to overestimation of treatment effect of topical NSAID These includefeelings of heat or cold and even the act of rubbing itself For that reason we have chosen to include only double-blind studies wherethe placebo agent is identical to the active with the exception of course of the NSAID So heat cold rubbing and penetrationenhancers should be identical as best we can judge That leaves only the NSAID itself to provide any additional analgesic effect andit is that which we measure This is analogous for example to use of acupuncture say where the better studies show no differencebetween ldquotruerdquo acupuncture and ldquoshamrdquo acupuncture performed at nonspecific sites but better than non-treatment controls Theargument that we should only use high quality studies to evaluate evidence about pain interventions is well made
3 Overestimation of analgesic effect because of effects of enhancers themselves would be better made in direct comparisons oftopical and oral NSAIDs where local or even systemic effects would not be balanced in the oral study arm However our reviewconcentrated on placebo-controlled studies and had few studies with active controls Moreover the real test would be in chronicrather than acute conditions with long duration (12 week) outcomes using current best evidence rules (Moore 2010) includingimputation (Moore 2012) In their response to Rothrsquos letter the authors of the original review of products available in the USA showrather similar effect sizes of oral diclofenac and topical diclofenac with different penetration enhancers (Barthel 2011) in such studies
4 The Roth letter sought to differentiate between topical diclofenac preparations based on the penetration enhancers used Thatdifferent formulations may have different effect sizes is a fair point to make Two of the studies in our review of topical NSAIDs inacute conditions used diclofenac sodium 1 gel comparing it with either diclofenac epolamine gel (Gallacchi 1990 50 participants)or lysine clonixinate gel (Hofman 2000 142 participants) no difference between formulations was demonstrated It is difficult tomake any judgement for topical NSAIDs in acute conditions due to the relatively small number and particularly the small size ofstudies We did an analysis by drug and this showed that some topical NSAIDs were consistently beneficial irrespective offormulation while others had little or no efficacy This fits in with some theoretical considerations of molecular architecture andtissue penetration (Moore 2008) In chronic pain where there are larger studies and much more data we have considered formulation(Derry in preparation)
References
Roth SH Letter to the editor The importance of differentiating between topical NSAIDs Postgraduate Medicine 2011123251-2[103810pgm2011092482]Moore RA Derry S McQuay HJ Topical agents in the treatment of rheumatic pain Rheumatic Diseases Clinics of North America200834415-32
142Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Moore RA Eccleston C Derry S Wiffen P Bell RF Straube S et al ACTINPAIN Writing Group of the IASP Special Interest Groupon Systematic Reviews in Pain Relief Cochrane Pain Palliative and Supportive Care Systematic Review Group Editors ldquoEvidencerdquo inchronic pain - establishing best practice in the reporting of systematic reviews Pain 2010150386-9Moore RA Straube S Eccleston C Derry S Aldington D Wiffen P et al Estimate at your peril imputation methods for patientwithdrawal can bias efficacy outcomes in chronic pain trials using responder analyses Pain 2012153265-8Barthel HR Axford-Gatley RA Response to Roth Letter to the Editor Postgraduate Medicine 2011123253-4 [103810pgm2011092483]
Contributors
Feedback from Sigrun ChrubasikAuthors involved with responding Andrew Moore Sheena DerryFeedback Editor Kate Seers
Query submitted by Peter C Goslashtzsche 3 November 2015
Summary
Date of Submission 03-Nov-2015Name Peter C GoslashtzscheEmail Address pcgcochranedkAffiliation Nordic Cochrane CentreRole DirectorComment The authors found that the results were missing from 5900 patients Furthermore there was extreme heterogeneity in theirmeta-analyses eg I square was 92 for the diclofenac trials which were the most common ones and there was extreme funnel plotasymmetry with the largest trials showing the smallest effects (the authors didnrsquot show funnel plots but I constructed one for diclofenac)Moreover the trials were industry funded of relatively poor quality and the authors analysed published data not data from clinicalstudy reports and did not try to obtain all the missing trials and data from the manufacturersThe authors concluded that topical NSAIDs are effective in providing pain relief but also cautioned that the large amounts of unpublisheddata ldquocould influence results in updates of this reviewrdquo They certainly could I believe it is plain wrong to perform meta-analyses on theauthorsrsquo data When I most recently reviewed this area for the BMJ in 2010 I concluded that we donrsquot know whether topical NSAIDsare beneficial (1)1 Goslashtzsche PC NSAIDs Clin Evid (Online) 2010 Jun 28I agree with the conflict of interest statement belowI certify that I have no affiliations with or involvement in any organization or entity with a financial interest in the subject matter ofmy feedback
Reply
Response submitted by authors Sheena Derry and Professor Andrew Moore on 4 November 2015Peter has made these same points elsewhere [PubMed Commons PubPeer gt Cochrane Database Syst Rev October 2015 httpspubpeercompublicationsA9E5BEA36549727357F9FD14CC2537]As a preliminary it is important to stress that the number of trials available in pain has increased over time and newer trials are typicallybetter and larger This means we can move from answering the simple question of whether an intervention works to more importantand relevant questions such as how well the intervention works and for drugs particularly to examine effects of dose and formulationFormulation is now recognised to have profound effects For oral analgesics for example fast acting formulations demonstrate up todouble the analgesic effect for a given dose as our recently updated overview points out (Cochrane Database Syst Rev 2015 Sep 289CD008659 see also Pain 2014 Jan155(1)14-21)What was true for oral analgesics in acute pain is now true also for topical NSAIDs in acute pain where manufacturers have beenputting a lot of effort into trying to produce new formulations that work better This updated review sought to examine the influenceof formulation and drug although dose is somewhat more difficult There are good reasons why formulation may be important(Rheumatic Disease Clinics 2008 Vol 34 Issue 2 p 415-432)
143Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Our searches did identify a large number of unpublished trials We deplore this but are powerless to change things For a previousreview of topical NSAIDs we contacted all identified manufacturers and asked for published or unpublished data The yield was small(as others have found) but some unpublished studies were brought into the public domain Waiting for all studies and clinical trialreports of all studies would take forever and could probably never be achieved We believe that many of these unpublished studiesrelate to drugs andor formulations that have never been manufactured commercially While these would be of interest in determiningwhat does and doesnrsquot work and to direct future research they would probably have little clinical relevance because these formulationsare unlikely to come to marketPeterrsquos main issue appears to be heterogeneity Tests for heterogeneity are problematical anyway (Pain 2000 85415-24) and the I squareof 92 that he quotes was for all topical diclofenac formulations combined While we do give an overall summary for diclofenac wedemonstrated in the review that the different diclofenac formulations produced different results from one another using LrsquoAbbeacute plotsand in our detailed analyses showing large variations in efficacy between them In the circumstance a high I square for all combined(clinical heterogeneity) is to be expected but it is not relevant The bulk of the studies on diclofenac were published in the last fiveyears were of decent quality and moderate to large size There were older data for ketoprofen but again these showed major differencesbetween formulations Differences between formulations are highlighted throughout the reviewTrying to determine publication bias using funnel plots or other measures is something of a lost cause (Journal of Clinical Epidemiology200053207-16) It is especially so with small numbers of trials (Journal of Clinical Epidemiology 2000 53 477-484) and makingsense of funnel plots is anything but easy for most people (Journal of Clinical Epidemiology 2005 58 894-901) A useless methodseems an odd choice to make to criticise our reviewThere are very good scientific reasons why drug and formulation may play a big part in the efficacy of topical NSAIDs This is alsothe case for oral analgesics used in acute pain where formulation improvements generating rapid absorption confers greater efficacySimple lumping strategies may have been permissible in past systematic review methodology but a more forensic approach is needednow and in the future This is what we have attempted to do in this latest reviewComparisons with Peterrsquos 2010 review seem inappropriate since that review was based on other reviews that are now out of date
Contributors
Feedback from Peter C GoslashtzscheAuthors involved with responding Andrew Moore Sheena DerryFeedback Editor Kate Seers
W H A T rsquo S N E W
Last assessed as up-to-date 4 February 2015
Date Event Description
4 November 2015 Feedback has been incorporated Feedback has been incorporated See Feedback 2
H I S T O R Y
Protocol first published Issue 4 2008
Review first published Issue 6 2010
144Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Date Event Description
13 October 2015 Amended Small errors found in Summary Table B for total num-ber of participants with piroxicam and indomethacinand in text for NNT with diclofenac (other gel) Nochange to conclusions
6 October 2015 Amended Small error found in Summary Table B percentagesfor NSAID and placebo were the wrong way round forcomparison of all NSAIDs and placebo RR and textwere correct No change to results or conclusions
13 February 2015 New citation required but conclusions have notchanged
Conclusions not changed Results remain essentiallythe same but the focus has changed to examine drugand formulation combination
3 February 2015 New search has been performed Title changed from Topical NSAIDs for acute pain inadults to Topical NSAIDs for acute musculoskeletalpain in adults to increase specificityNew searches run and new studies identified Fourteennew included studies using diclofenac (3489 new par-ticipants a 63 increase) four new excluded studiesFifteen studies awaiting classification (completed noresults available but likely to satisfy inclusion criteria)
23 May 2014 Amended Error in data reported for clinical success in Hosie 1993was brought to our attention and has been corrected
12 June 2012 Feedback has been incorporated We have incorporated feedback received from DrSigrun Chrubasik and the authorrsquos response on DMSOand other additives
C O N T R I B U T I O N S O F A U T H O R S
For the earlier review Tom Massey and SD identified studies and carried out data extraction analysis and drafting RAM and HJMwere involved in planning acted as adjudicators and were involved with writing
For this update SD and RAM carried out searches data extraction and analysis All authors were involved with writing the review
145Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
D E C L A R A T I O N S O F I N T E R E S T
SD has no conflicts relating to this review or any similar product
RAM has no conflicts relating to this review or any similar product
PW has no conflicts relating to this review or any similar product
HG has no conflicts relating to this review or any similar product
MM has no conflicts relating to this review or any similar product
S O U R C E S O F S U P P O R T
Internal sources
bull Oxford Pain Relief Trust UKGeneral institutional support
External sources
bull No sources of support supplied
D I F F E R E N C E S B E T W E E N P R O T O C O L A N D R E V I E W
For this update in 2015 we have changed the title to specify musculoskeletal pain because topical NSAIDs are not normally used totreat visceral pain or headache We felt that the new title better reflected the content of the review We have also changed the focus ofthe review from pooled analysis of all topical NSAIDs and all studies of a particular NSAID to an examination of individual drug andits formulation This makes the review much more relevant We have expanded the rsquoRisk of biasrsquo assessment and added a rsquoSummaryof findingsrsquo table and PRISMA flow chart We have removed a number of sensitivity analyses because they were not appropriate giventhe current information on the impact of formulation on efficacy The sensitivity analyses have been superseded by the rsquoRisk of biasrsquoassessment and taken into account in the rsquoSummary of findingsrsquo tables
An earlier review in 2004 chose to exclude studies using benzydamine on the grounds that it was no longer considered to be an NSAID(Mason 2004a) Although the protocol for this review stated that we would not include benzydamine after further consultation wenow believe that it should be classified as an NSAID albeit with a different mode of action which is not fully understood (Quane1998) Thus we have reinstated studies using topical benzydamine
I N D E X T E R M S
Medical Subject Headings (MeSH)
Acute Disease Administration Topical Anti-Inflammatory Agents Non-Steroidal [lowastadministration amp dosage adverse effects] AthleticInjuries [drug therapy] Pain [lowastdrug therapy] Randomized Controlled Trials as Topic Sprains and Strains [drug therapy]
146Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
MeSH check words
Adult Humans
147Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
having good pain reduction (by about half ) around seven days after treatment started At later times most people are expected to getbetter even without treatment
We looked at particular formulations of individual drugs Gel formulations of diclofenac and ketoprofen were among the most effectivealong with ibuprofen gel and diclofenac plaster For diclofenac and ketoprofen gels 7 or 8 people out of 10 with a painful strain sprainor muscle pull had much reduced pain after seven days compared with only 2 or 3 out of 10 with placebo (high quality data) OtherNSAIDs and formulations were better than placebo but not by as much Because both topical NSAIDs and topical placebo are rubbedinto the skin in these studies we know that any effect is not just from rubbing
About 1 in 20 people experienced a mild and short-lived side effect like redness at the application site This was the same for topicalNSAID and topical placebo (high quality data) Side effects like a stomach upset or feeling sick were uncommon with no differencebetween topical NSAID and topical placebo (high quality data) There were no serious side effects
3Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
S U M M A R Y O F F I N D I N G S F O R T H E M A I N C O M P A R I S O N [Explanation]
Topical NSAIDs compared with topical placebo for acute musculoskeletal pain in adults
Patient or population adults with strains sprains or muscle pullSettings communityIntervention topical NSAID (topical diclofenac ibuprofen and ketoprofen gels only shown here for efficacy)Comparison topical placebo
Outcomes Probable outcome withintervention
Probable outcome withcomparator
RR NNT NNTp or NNH(95 CI)
No of studies partici-pants
Quality of the evidence(GRADE)
Comments
Topical diclofenac gel
(as Emulgel)
Clinical success (eg50 reduction in pain)
780 in 1000 200 in 1000 RR34 (27 to 55)NNT18 (15 to 21)
2 studies314 participants
High Consistent results in 2moderately sized recentstudies of high quality
Topical ibuprofen gel
Clinical success (eg
50 reduction in pain)
420 in 1000 160 in 1000 RR27 (17 to 42)NNT39 (27 to 67)
2 studies241 participants
Moderate Modest effect size andnumbers of participants
Topical ketoprofen gel
Clinical success (eg
50 reduction in pain)
720 in 1000 330 in 1000 RR22 (17 to 28)NNT25 (20 to 34)
5 studies348 participants
Moderate Modest effect size andnumbers of participantsbut studies small withnone recent
All topical NSAIDs
Local adverse events
46 in 1000 50 in 1000 RR10 (080 to 12)NNH not calculated
42 studies6125 participants
High Large number of studiesand participants with con-sistent results
All topical NSAIDs
Systemic adverse
events
32 in 1000 35 in 1000 RR10 (07 to 13)NNH not calculated
38 studies5372 participants
High Large number of studiesand participants with con-sistent results
All topical NSAIDs
Withdrawals - adverse
events
11 in 1000 11 in 1000 RR10 (07 to 17)NNH not calculated
42 studies5790 participants
High Large number of studiesand participants with con-sistent results4
To
pica
lN
SA
IDs
for
acu
tem
uscu
losk
ele
talp
ain
inad
ults
(Revie
w)
Co
pyrig
ht
copy2015
Th
eC
och
ran
eC
olla
bo
ratio
nP
ub
lished
by
Joh
nW
iley
ampS
on
sL
td
Serious adverse events 1 in total 0 in total Not calculated All data Low Small numbers of events
GRADE Working Group grades of evidenceHigh quality Further research is very unlikely to change our confidence in the estimate of effectModerate quality Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimateLow quality Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimateVery low quality We are very uncertain about the estimate
CI confidence interval RR risk ratio NNT number needed to treat for an additional beneficial outcome NNTp number needed to treatto prevent an event happening NNH number needed to treat for an additional harmful outcome
xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
5To
pica
lN
SA
IDs
for
acu
tem
uscu
losk
ele
talp
ain
inad
ults
(Revie
w)
Co
pyrig
ht
copy2015
Th
eC
och
ran
eC
olla
bo
ratio
nP
ub
lished
by
Joh
nW
iley
ampS
on
sL
td
B A C K G R O U N D
This review is an update of a review originally published in Issue 62010 on rsquoTopical NSAIDs for acute pain in adultsrsquo (Massey 2010)We have changed the title to specify musculoskeletal pain becausetopical NSAIDs are not normally used to treat visceral pain orheadache We felt that the new title better reflected the content ofthe review
The use of topical NSAIDs for pain relief has been a controversialsubject in analgesic practice In some parts of the world (includingmuch of Western Europe) they have been available for many yearsare widely available without prescription widely advertised usedextensively and evidence for their use is considered adequate Inother parts of the world they were regarded as little more thanplacebo with any apparent effect attributed to the process of rub-bing at the site of the affected area In some places (for examplethe US) their use was almost unknown until the mid-2010s InEngland 52 million prescriptions for topical NSAIDs were dis-pensed in 2013 (PACT 2014) mainly for formulations of ibupro-fen (245 million) piroxicam (118 million) and diclofenac (127million)
There is good evidence for the efficacy of topical NSAIDs in acuteand chronic musculoskeletal pain (Mason 2004a Mason 2004bMoore 1998a) In the US the Food and Drug Administration li-censed topical nonsteroidal products in 2007 and in England theNational Institute for Health and Care Excellence (NICE) recom-mended topical therapies as first line treatment in its guidelinesfor osteoarthritis in 2008 (NICE 2008) Earlier reviews of topicalanalgesics covered studies investigating the underlying science toexplain biological plausibility in addition to clinical trials (Anon2005 Moore 2008a)
This review is one of a series on topical analgesics including topicalcapsaicin at low and high doses (Derry 2012a Derry 2013) andtopical NSAIDs in chronic pain conditions (Derry 2012b) andsalicylate-containing rubefacients (Derry 2014)
Description of the condition
Acute pain is usually defined as pain of less than three monthsrsquo du-ration It is often associated with injury including trauma surgerymusculoskeletal injuries such as strains sprains and over-use in-juries or soft tissue injuries such as muscle soreness or cramps
Description of the intervention
Clinicians prescribe NSAIDs on a routine basis for a range of mildto moderate pain NSAIDs are the most commonly prescribedanalgesic medications worldwide and their efficacy for treatingacute pain has been well demonstrated (Moore 2003) They re-versibly inhibit cyclooxygenase (prostaglandin endoperoxide syn-thase) the enzyme mediating production of prostaglandins and
thromboxane A2 (FitzGerald 2001) Prostaglandins mediate a va-riety of physiological functions such as maintenance of the gastricmucosal barrier regulation of renal blood flow and regulation ofendothelial tone They also play an important role in inflamma-tory and nociceptive processes However relatively little is knownabout the mechanism of action of this class of compounds asidefrom their ability to inhibit cyclooxygenase-dependent prostanoidformation (Hawkey 1999)NSAIDs taken orally or intravenously are transported to all partsof the body in the blood and relatively high blood concentrationsare needed to achieve effective tissue concentrations at the site ofthe pain and inflammation These high concentrations throughoutthe body can give rise to a number of adverse events that canbe unpleasant (for example dyspepsia) or potentially serious (forexample gastrointestinal bleeding)A topical medication is a one applied to body surfaces such asthe skin or mucous membranes to treat ailments A large rangeof types of topical formulation may be used including but notlimited to creams foams gels lotions ointments and plastersThe exact formulation of a topical medication is often determinedby the speed of drug absorption required The need may be forslow absorption into the circulation to maintain low drug concen-trations and perhaps avoiding extensive first pass metabolism inthe liver plasters containing drug reservoirs may be used for thisas with transdermal opioids or contraceptive steroids For rapidabsorption the formulation is enhanced by substances to improveor assist skin penetration perhaps only to generate high concen-trations in tissues rather than in the blood gel formulations areuseful for this purpose which is why they are sometimes used fortopical NSAIDs
Topical NSAIDs
Topical NSAIDs are formulated for direct application to thepainful site and to produce a local pain-relieving effect whileavoiding body-wide distribution of the drug at physiologically ac-tive levels (McPherson 2013) This method of application (dos-ing) necessarily limits their use to more superficial painful condi-tions such as sprains strains and muscle or tendon soreness Theywould not for example be indicated for deep visceral pain orheadaches They are also not appropriate for use on broken skinso would not be used on open wounds (accidental or surgical)
How the intervention might work
For a topical formulation to be effective it must first penetratethe skin Only when the drug has entered the lower layers ofthe skin can it be absorbed by the blood and transported to thesite of action or penetrate deeper into areas where inflammationoccurs Individual drugs have different degrees of penetration Abalance between lipid and aqueous solubility is needed to optimisepenetration and use of prodrug esters has been suggested as a way
6Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
of enhancing permeability Formulation is also crucial to goodskin penetration Experiments with artificial membranes or humanepidermis suggest that creams are generally less effective than gelsor sprays but newer formulations such as microemulsions mayhave greater potentialOnce the drug has reached the site of action it must be presentat a sufficiently high concentration to inhibit cyclooxygenase en-zymes thereby reducing prostaglandin synthesis This in turn re-duces inflammation and relieves pain It is probable that in acuteconditions topical NSAIDs exert their action primarily by lo-cal reduction of symptoms independent of any systemic uptakeand delivery Tissue levels of NSAIDs applied topically certainlyreach levels high enough to inhibit cyclooxygenase-2 (Anon 2005Haroutiunian 2010 Moore 2008a) However plasma concentra-tions found after topical administration are only a fraction (usu-ally much less than 5) of the levels found in plasma followingoral administration Topical application can potentially limit sys-temic adverse events by increasing local effects and minimisingsystemic concentrations of the drug We know that the incidenceof upper gastrointestinal bleeding is low with chronic use of topi-cal NSAIDs (Evans 1995) although it has been reported partic-ularly in people with risk factors (Zimmerman 1995) We haveno certain knowledge of lower effects on cardiovascular events orrenal failure both of which have been associated with oral NSAIDuse
Why it is important to do this review
Since the last review in 2010 a number of new studies have beenpublished nearly all of which investigated various formulations ofdiclofenac These new studies are generally of higher quality thanmany of the earlier ones in this review and have the potential toinfluence the strength of its conclusions substantially Moreoverthe additional information allows for analysis based not only on aparticular drug but also on the formulation of that drug This canprovide better insight into whether formulation affects efficacy oftopical NSAIDs in acute musculoskeletal painAn updated review of evidence for topical NSAIDs was needed toinform choices made by consumers prescribers and commission-ers (purchasers of healthcare)
O B J E C T I V E S
To determine the efficacy and safety of topically applied NSAIDsin acute musculoskeletal pain in adults
M E T H O D S
Criteria for considering studies for this review
Types of studies
Randomised controlled double-blind studies comparing topicalNSAIDs with placebo (inert carrier) or other active treatment foracute pain with at least 10 participants per treatment arm andoutcomes close to seven days (minimum three days) We excludedstudies published only as short abstracts (which report insufficientdata to assess methods) or studying experimentally induced pain(which does not correlate well with clinical pain) Because a cross-over design is not appropriate for self limiting conditions such assprains strains and contusions we only considered parallel-groupdesigns
Types of participants
Adults (aged 16 years or more) with acute musculoskeletal pain ofat least moderate intensity resulting mainly from strains sprainsor sports injuries Typically for sports injuries the injury wouldhave occurred within 24 or 48 hours
Types of interventions
Included studies had at least one treatment arm using a topicalNSAID and a comparator arm using placebo (inert carrier withoutNSAID or other active treatment) The topical NSAID had to beapplied at least once daily We did not include salicylates in thisreview as they are no longer classified as topical NSAIDs and arecovered in a separate review (Derry 2014)
Types of outcome measures
We sought information on participant characteristics includingage sex and condition treated
Primary outcomes
The primary outcome was rsquoclinical successrsquo defined as at least a50 reduction in pain or equivalent measure such as a rsquovery goodrsquoor rsquoexcellentrsquo global assessment of treatment or rsquononersquo or rsquoslightrsquopain on rest or movement measured on a categorical scale (Moore1998a) We used the following hierarchy of outcomes to extractdata for the primary outcome
bull Participant reported reduction in pain of at least 50bull Participant reported global assessment of treatmentbull Pain on movementbull Pain at rest or spontaneous painbull Undefined rsquoimprovementrsquo
We used only participant reported outcomes of efficacy and notphysician or investigator reported outcomes
7Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Secondary outcomes
bull Numbers of participants with adverse events local andsystemic
bull Numbers of withdrawals all cause lack of efficacy andadverse events
We anticipated that outcomes would be reported after differentdurations of treatment and extracted data reported as close toseven days as possible with a minimum of three days We also ex-tracted data for outcomes reported after longer durations of treat-ment We anticipated that reporting of adverse events would varybetween studies with regard to the terminology used method ofascertainment and categories reported (for example occurring inat least 5 of participants or where there is a statistically signifi-cant difference between treatment groups) We took care to iden-tify these details where relevant
Search methods for identification of studies
Electronic searches
We searched the following databases without language restrictionbull CENTRAL (The Cochrane Library) Issue 4 2009 for the
original review and the Cochrane Register of Studies Online(CRSO) to 3 February 2015 for this update
bull MEDLINE (via Ovid) from inception to December 2009for the original review and from 2008 to 3 February 2015 forthis update
bull EMBASE (via Ovid) from inception to December 2009for the original review and from 2008 to 3 February 2015 forthis update
bull Oxford Pain Relief Database for the original review (Jadad1996a) This resource is no longer being updated
See Appendix 1 for the CENTRAL search strategy Appendix 2 forthe MEDLINE search strategy and Appendix 3 for the EMBASEsearch strategy
Searching other resources
We searched the reference lists of review articles and includedstudies We have previously asked manufacturers for details ofunpublished studies but did not make new requestsWe searched two clinical trial registries (clinicaltrialsgov (clinicaltrialsgov) and the World Health Organization In-ternational Clinical Trials Registry Platform (appswhointtrialsearch)) and asked personal contacts about ongoing and un-published studies
Data collection and analysis
Selection of studies
We screened the titles and abstracts of studies identified by thesearches to eliminate those that clearly did not satisfy the inclu-sion criteria and obtained full reports of the remaining studies todetermine inclusion in the review
Data extraction and management
Review authors were not blinded to the authorsrsquo names and in-stitutions journal of publication or study results at any stage ofthe review Two review authors independently selected the stud-ies for inclusion assessed methodological quality and risk of biasand extracted data We resolved disagreements and uncertaintiesthrough discussionWe abstracted information on participants interventions and out-comes from the original reports into a standard data extractionform One review author entered data suitable for meta-analysisinto Review Manager 5 (RevMan 2014) and another review au-thor checked it
Assessment of risk of bias in included studies
We used the Oxford Quality Score as the basis for inclusion lim-iting inclusion to studies that were randomised and double-blindas a minimum (Jadad 1996b)Two review authors independently assessed risk of bias for eachstudy using the criteria outlined in the Cochrane Handbook forSystematic Reviews of Interventions with any disagreements resolvedby discussion (Higgins 2011) We assessed the following for eachstudy
bull Random sequence generation (checking for possibleselection bias) We assessed the method used to generate theallocation sequence as low risk of bias (any truly randomprocess for example random number table computer randomnumber generator) unclear risk of bias (method used to generatesequence was not clearly stated) We excluded studies using anon-random process that were therefore at high risk of bias (forexample odd or even date of birth hospital or clinic recordnumber)
bull Allocation concealment (checking for possible selectionbias) The method used to conceal allocation to interventionsbefore assignment determines whether intervention allocationcould have been foreseen in advance of or during recruitmentor changed after assignment We assessed the methods as lowrisk of bias (for example telephone or central randomisationconsecutively numbered sealed opaque envelopes) unclear riskof bias (method was not clearly stated) We excluded studies thatdid not conceal allocation and were therefore at high risk of bias(for example open list)
8Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
bull Blinding of outcome assessment (checking for possibledetection bias) We assessed the methods used to blind studyparticipants and outcome assessors from knowledge of whichintervention a participant received We assessed the methods aslow risk of bias (study stated that it was blinded and describedthe method used to achieve blinding for example identicaltubes containing gel or identical plasters matched in appearanceand smell) unclear risk of bias (study stated that it was blindedbut did not provide an adequate description of how blinding wasachieved) We excluded studies that were not double-blind andwere therefore at high risk of bias
bull Incomplete outcome data (checking for possible attritionbias due to the amount nature and handling of incompleteoutcome data) We assessed the methods used to deal withincomplete data as low risk of bias (less than 10 ofparticipants did not complete the study or used rsquobaselineobservation carried forwardrsquo analysis or both) unclear risk ofbias (used rsquolast observation carried forwardrsquo analysis) or high riskof bias (used rsquocompleterrsquo analysis)
bull Size (checking for possible biases confounded by small size)Small studies have been shown to overestimate treatment effectsprobably due to methodological weaknesses (Dechartres 2013Nuumlesch 2010) We assessed studies as at low risk of bias if theyhad at least 200 participants at unclear risk if they had 50 to 200participants and at high risk if they had fewer than 50participants
Measures of treatment effect
We used risk ratio (RR) to establish statistical difference and num-bers needed to treat for an additional beneficial outcome (NNT)with 95 confidence intervals (CI) We pooled percentages as ab-solute measures of benefit or harmWe used the following terms to describe adverse outcomes in termsof harm or prevention of harm
bull When significantly fewer adverse outcomes occurred withtreatment than with control (placebo or active) we used the termthenumber needed to treat to prevent one event (NNTp)
bull When significantly more adverse outcomes occurred withtreatment compared with control (placebo or active) we usedthe term the number needed to treat for an additional harmfuloutcome or cause one event (NNH)
Unit of analysis issues
Randomisation was to the individual participant
Dealing with missing data
Wherever possible we used intention-to-treat (ITT) analysis wherethe ITT population consists of participants who were randomisedapplied at least one dose of the assigned study medication andprovided at least one post-baseline assessment We assigned miss-ing participants zero improvement
We also looked for information about methods of imputation formissing data
Assessment of heterogeneity
We examined heterogeneity visually using LrsquoAbbeacute plots (LrsquoAbbeacute1987) a visual method for assessing differences in results of indi-vidual studies and with the I2 statistic
Assessment of reporting biases
The aim of this review was to use dichotomous outcomes of knownutility and of value to patients (Moore 2013) The review did notdepend on what the authors of the original studies chose to reportor not Studies that did not report dichotomous results but onlyaverage pain data did not contribute to analysesWe assessed publication bias using a method designed to detect theamount of unpublished data with a null effect required to makeany result clinically irrelevant (usually taken to mean an NNT of10 or higher Moore 2008b)
Data synthesis
We pooled data only for comparisons and outcomes where therewere at least two studies and 200 participants (Moore 1998b)When two active treatment arms were compared with a placeboarm we took care to avoid double counting of participants in theplacebo arm if both active groups contributed to an analysis wesplit the placebo group between themWe calculated RRs with 95 CIs using the fixed-effect model(Morris 1995) A statistically significant benefit of topical NSAIDover control was assumed when the lower limit of the 95 CI ofthe RR was greater than one A statistically significant benefit ofcontrol over active treatment was assumed when the upper limitof the 95 CI was less than one We calculated NNTs with 95CIs using the pooled number of events by the method of Cookand Sackett (Cook 1995)Statistically significant differences between NNTs for differenttopical NSAIDs were tested using the z test (Tramer 1997) wherethere were sufficient data to do so and where the studies weresufficiently similar in types of participant outcome and durationto make such comparisons sensible
Subgroup analysis and investigation of heterogeneity
We carried out separate analyses for individual NSAIDs andwhere the data permitted for different formulations of individualNSAIDs
Sensitivity analysis
The earlier review included sensitivity analyses for various factorsthat are now covered by the assessment of risk of bias
9Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
R E S U L T S
Description of studies
Results of the search
New searches for this update identified 20 publications that wereexamined in further detail to determine inclusion status We alsoidentified 20 additional studies in clinical trial registries See Figure1
10Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Figure 1 Study flow diagram
11Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
We identified 14 new studies (11 publications three unpub-lished reports) satisfying our inclusion criteria (Costantino2011 Coudreuse 2010 Gonzaacutelez de Vega 2013 Hofman2000 Klainguti 2010 Kuehl 2011 Li 2013 NCT01255423NCT01272934 NCT01272947 Predel 2012 Predel 2013aPredel 2013b Saillant 1998) one post hoc analysis of a study thatwas included in the earlier review (Mueller 2010 in Predel 2004)and an additional publication and a pooled analysis that includednew data for another study from the earlier review (Lionberger2011 pooled analysis in Joussellin 2003)
Included studies
All except one of the new studies compared diclofenac withplacebo A number of different formulations were used includ-ing diclofenac epolamine (DHEP) with or without heparin (Flec-toparin Tissugel or Flector EP Tissugel) applied as a plaster (orpatch) diclofenac diethylamine (DDEA) applied as a gel and di-clofenac with lethicin applied as a spray gel The remaining studycompared diclofenac gel with traumeel a ldquofixed combination ofplant and mineral extractsrdquo applied as a gel or an ointmentThere were 47 studies in the original review 14 new studies wereincluded making a total of 61 studies in this updated reviewAll used a parallel group design Forty-four compared a topicalNSAID with placebo 13 a topical NSAID with an active com-parator (a different topical NSAID an oral NSAID the same top-ical NSAID in a different formulation or a compound of plantand mineral extracts) and four had both placebo and active com-parators In total 5311 participants were treated with a topicalNSAID 3470 with placebo and 220 with an oral NSAID Top-ical NSAIDs used were benzydamine diclofenac etofenamatefelbinac fentiazac flunoxaprophen flurbiprofen ibuprofen in-domethacin ketoprofen ketorolac lysine clonixinate meclofe-namic acid naproxen niflumic acid and piroxicam They wereapplied as creams gels sprays foams or plasters (patches) Topicalplacebos were the inert carriers without the active NSAID OralNSAIDs used were ibuprofen (as tablets) and indomethacin (ascapsules)Most studies enrolled participants who had sprains strains andcontusions usually as a result of sports injuries and treatmentwas started within a few hours or days Other studies enrolledparticipants with overuse-type injuries such as tendinitis and acutelow back pain where pain had been present for days or weeks butless than three monthsParticipants were treated for at least five days and up to threeweeks with most studies lasting seven to 14 days Participantswere usually assessed in clinic at intervals during treatment andsometimes also at home using daily patient diaries We used out-comes closest to seven days because many of these injuries are self
limiting with differences between active treatment and placebobeing diminished or lost after longer intervalsMost studies reported dichotomous outcomes suitable for a re-sponder analysis although group mean change (for pain or phys-ical function for example) was usually the primary outcomesHowever the definition of response varied both in the parametermeasured (for example pain pain on movement patient globalevaluation of treatment) and in the scale used to measure it (forexample a 3- 4- or 5-point scale for patient global evaluation)Details of included studies are in the Characteristics of includedstudies tableWe identified 14 completed but apparently unpublished studies ina clinical trial registry for which no results have been posted (4403participants NCT00351104 NCT00352625 NCT00426985NCT00640705 NCT00640939 NCT00680472NCT00680784 NCT00765700 NCT00869063NCT00869180 NCT00931866 NCT01874626NCT01957215 NCT02324270) We have placed these underCharacteristics of studies awaiting classification We also identi-fied three conference abstracts that relate to completed studies thatdo not appear to have been published but that may satisfy ourinclusion criteria One is likely to be the same study as one of theincluded studies identified in a clinical trial registry (Pallay 2013in NCT01272947) one relates to another study identified in theclinical trial registry that is awaiting classification (Ekman 2010in NCT00765700) while we could find no published or unpub-lished reports of the other Sarzi-Puttini 2014)We also identified three ongoing studies with an estimated en-rolment of 880 participants (NCT01945034 NCT02100670NCT02290821) Details are in the Characteristics of ongoingstudies table
Excluded studies
For the original review 25 studies were excluded after readingthe full paper For this update we excluded four new studies (Cesarone 2008 Coulibaly 2009 Kuwabara 2013 Vinciguerra2008) Details are in the Characteristics of excluded studies table
Risk of bias in included studies
All studies were randomised and double-blind One study scored25 (Sinniger 1981) 23 scored 35 23 scored 45 and 14 scored55 for methodological quality using the Oxford Quality Scale Abreakdown of the scores for individual studies is reported in theCharacteristics of included studies tableComments on potential biases in individual studies are reported inthe rsquoRisk of biasrsquo section of the Characteristics of included studiestable The findings are displayed in Figure 2 and Figure 3
12Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Figure 2 Risk of bias summary review authorsrsquo judgements about each risk of bias item for each includedstudy
13Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Figure 3 Methodological quality graph review authorsrsquo judgements about each methodological quality
item presented as percentages across all included studies
Allocation
All the studies were randomised but only 17 adequately describedthe method used to generate the random sequence Thirteen stud-ies adequately described the method used to conceal allocation ofthe sequence No studies were at high risk of bias for this item
Blinding
All studies were double-blind and 38 adequately described themethod used to maintain the blinding No studies were at highrisk of bias for this item
Incomplete outcome data
Thirty-six studies included all participants in the primary analysisor provided sufficient data to allow missing participants to be in-cluded as non-responders and were judged at low risk of bias Wejudged four studies to be at high risk of attrition bias (Campbell1994 Kuehl 2011 Maziegraveres 2005a Russell 1991) Three unpub-lished studies contributed only to adverse event analyses and ac-counted for all participants for these outcomes (NCT01255423NCT01272934 NCT01272947)
Other potential sources of bias
We judged one study that included more than 200 participantsin each treatment arm to be at low risk of bias from size butthis study had a very high attrition rate (see rsquoIncomplete outcome
data (attrition bias)rsquo) and did not report all the efficacy outcomesmeasured (Kuehl 2011) We judged 27 studies to be at high riskbecause they included fewer than 50 participants per treatmentarm
Effects of interventions
See Summary of findings for the main comparison TopicalNSAIDs compared with topical placebo for acute musculoskeletalpain in adultsThree studies did not contribute data suitable for analysis of at leastone outcome (Gonzaacutelez de Vega 2013 Gualdi 1987 Hoffmann2012)
1 Topical NSAID versus placebo
Details of efficacy outcomes in individual studies are in Appendix4 and of adverse events and withdrawals in Appendix 5 Appendix6 has details of the concentration of topical products the amountapplied the frequency of application and an estimation of thedaily dose of topical NSAID applied Not all studies providedsufficient information to allow calculation of daily dose appliedFor example for topical diclofenac the estimated doses appliedvaried between about 60 and 280 mg for topical ketoprofen 100to about 450 mg for topical ibuprofen 300 to 800 mg
Participants with clinical success
14Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Topical diclofenac versus placebo
Ten studies contributed to this analysis (Coudreuse 2010Joussellin 2003 Klainguti 2010 Li 2013 Predel 2004 Predel2012 Predel 2013a Predel 2013b Rowbotham 2003) of whichone (Predel 2012) had two active treatment arms A total of 1074participants were treated with topical diclofenac and 976 withplacebo (Analysis 11 Figure 4)
15Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Figure 4 Forest plot of comparison 2 Individual NSAID versus placebo outcome 21 Clinical success
16Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
bull The proportion of participants experiencing successfultreatment with topical diclofenac was 74 (8001074 range39 to 100)
bull The proportion of participants experiencing successfultreatment with placebo was 47 (461976 range 8 to 82)
bull The RR for treatment compared with placebo was 16(95 CI 15 to 17)
bull The NNT for successful treatment was 37 (32 to 43) Forevery four participants treated with topical diclofenac one wouldexperience successful treatment who would not have done sowith placebo
Effect of formulation
The effects of formulation are shown in Analysis 21 and Figure5
17Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Figure 5 LrsquoAbbeacute plot of clinical success in studies of topical diclofenac versus topical placebo The size ofthe symbol is proportional to the size of the study (inset scale) Dark blue Emulgel light blue spraygel red
Flector pink other patch or plaster
bull Four studies used a Flectorreg plaster (1030 participantsJoussellin 2003 Li 2013 Rowbotham 2003 Saillant 1998) TheRR for treatment compared with placebo was 15 (14 to 17)and the NNT was 47 (37 to 65)
bull Three studies used other makes of plaster (474 participantsCoudreuse 2010 Klainguti 2010 Predel 2004) The RR fortreatment compared with placebo was 16 (14 to 18) and theNNT was 32 (26 to 42)
bull Two studies used Voltaren Emulgel (314 participants Predel2012 Predel 2013b) The RR for treatment compared withplacebo was 38 (27 to 55) and the NNT was 18 (15 to 21)
bull One study used a spray gel (232 participants Predel2013a) The RR for treatment compared with placebo was 12(105 to 13) and the NNT was 80 (48 to 24)
Diclofenac as the gel formulation Emulgel was statistically more ef-ficacious than the plaster formulation as Flector plaster (z = 6360P value lt 000001)
Topical ibuprofen versus placebo
Five studies contributed to this analysis (Billigmann 1996Campbell 1994 Dreiser 1988 Machen 2002 Ramesh 1983) Atotal of 218 participants were treated with topical ibuprofen and218 with placebo (Analysis 11)
bull The proportion of participants experiencing successfultreatment with topical ibuprofen was 55 (120218 range 31to 81)
18Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
bull The proportion of participants experiencing successfultreatment with placebo was 33 (73218 range 13 to 76)
bull The RR of treatment compared with placebo was 16 (13to 20)
bull The NNT for successful treatment was 46 (33 to 80) Forevery five participants treated with topical ibuprofen one wouldexperience successful treatment who would not have done sowith placebo
Effect of formulation
The effects of formulation are shown in Analysis 31bull Three studies used cream formulations (195 participants
Campbell 1994 Dreiser 1988 Ramesh 1983) Although this isjust below our threshold for pooled analysis we have includedthis analysis for completeness and the results should beinterpreted with caution The RR for treatment compared withplacebo was 13 (103 to 16) and the NNT was 64 (34 to 41)
bull Two studies used gel formulations (241 participantsBilligmann 1996 Machen 2002) The RR for treatmentcompared with placebo was 27 (17 to 42) and the NNT was39 (27 to 67)
There was no statistically significant difference between the geland cream formulations (z = 1160 P value = 0246)
Topical ketoprofen versus placebo
Seven studies contributed to this analysis (Airaksinen 1993Dreiser 1989 Julien 1989 Kockelbergh 1985 Maziegraveres 2005bMaziegraveres 2005a Noret 1987) A total of 346 participants weretreated with topical ketoprofen and 337 with placebo (Analysis11)
bull The proportion of participants experiencing successfultreatment with topical ketoprofen was 73 (251346 range57 to 89)
bull The proportion of participants experiencing successfultreatment with placebo was 47 (157337 range 17 to 73)
bull The RR of treatment compared with placebo was 16 (14to 18)
bull The NNT for successful treatment was 39 (30 to 53) Forevery four participants treated with topical ketoprofen onewould experience successful treatment who would not have doneso with placebo
Effect of formulation
The effects of formulation are shown in Analysis 41 and Figure6
19Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Figure 6 LrsquoAbbeacute plot of clinical success in studies of topical ketoprofen versus topical placebo The size ofthe symbol is proportional to the size of the study (inset scale) Light blue ketoprofen gel pink ketoprofen
plaster
Two studies used a plaster formulation (335 participants Maziegraveres2005b Maziegraveres 2005a) The RR for treatment compared withplacebo was 12 (104 to 14) and the NNT was 82 (45 to 47)Five studies used gel formulations (348 participants Airaksinen1993 Dreiser 1989 Julien 1989 Kockelbergh 1985 Noret 1987)The RR for treatment compared with placebo was 22 (17 to 28)and the NNT was 25 (20 to 34)Ketoprofen as a gel formulation was statistically more efficaciousthan a plaster formulation (z = 3860 P value = 000014)
Topical piroxicam versus placebo
Three studies contributed to this analysis (Aoki 1984 Fujimaki1985 Russell 1991) A total of 255 participants were treated withtopical piroxicam and 249 with placebo (Analysis 11)
bull The proportion of participants experiencing successfultreatment with topical piroxicam was 68 (179255 range 53to 79)
bull The proportion of participants experiencing successfultreatment with placebo was 47 (118249 range 45 to 49)
bull The RR of treatment compared with placebo was 15 (13to 17)
bull The NNT for successful treatment was 44 (32 to 69) Forevery four participants treated with topical piroxicam one wouldexperience successful treatment who would not have done sowith placebo
Topical indomethacin versus placebo
20Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Three studies contributed to this analysis ( kermark 1990 Aoki1984 Fujimaki 1985) A total of 168 participants were treatedwith topical indomethacin and 173 with placebo (Analysis 11)
bull The proportion of participants experiencing successfultreatment with topical indomethacin was 58 (97168 range54 to 64)
bull The proportion of participants experiencing successfultreatment with placebo was 46 (79173 range 25 to 49)
bull The RR of treatment compared with placebo was 13 (103to 16)
bull The NNT for successful treatment was 83 (44 to 65) Forevery eight participants treated with topical indomethacin onewould experience successful treatment who would not have doneso with placebo
Topical benzydamine versus placebo
Three studies contributed to this analysis (Chatterjee 1977 Haig1986 Linde 1985) A total of 96 participants were treated withtopical benzydamine and 97 with placebo (Analysis 11) Al-though this is just below our threshold for pooled analysis we haveincluded this analysis for completeness and the results should beinterpreted with caution
bull The proportion of participants experiencing successfultreatment with topical benzydamine was 77 (7496 range70 to 86)
bull The proportion of participants experiencing successfultreatment with placebo was 67 (6597 range 48 to 80)
bull The RR of treatment compared with placebo was 12 (096to 14) There was no statistically significant difference betweentreatments (Figure 4)
Results for participants with clinical success with individual topi-cal NSAIDs where there were adequate data for analysis are sum-marised below in rsquoSummary of results Arsquo and Analysis 11 Analysis21 Analysis 31 and Analysis 41
Summary of results A Participants with clinical success
Comparison Studies Participants NSAID()
Placebo () Relative benefit (95CI)
NNT (95 CI)
Diclofenac -Flector plaster
4 1030 63 41 15 (14 to 17) 47 (37 to 65)
Diclofenac -other plaster
3 474 88 57 16 (14 to 18) 32 (26 to 42)
Diclofenac -Emulgel
2 314 78 20 38 (27 to 55) 18 (15 to 21)
Diclofenac -other gel
1 232 94 82 12 (11 to 13) 80 (48 to 24)
Ibuprofen -cream
3 195 71 56 13 (103 to 16) 64 (34 to 41)
Ibuprofen - gel 2 241 42 16 27 (17 to 42) 39 (27 to 67)
Ketoprofen -plaster
2 335 73 60 12 (104 to 14) 82 (45 to 47)
Ketoprofen - gel 5 348 72 33 22 (17 to 28) 25 (20 to 34)
Piroxicam 3 504 70 47 15 (13 to 17) 44 (32 to 69)
Indomethacin 3 341 58 46 13 (103 to 16) 83 (44 to 65)
21Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Benzydamine 3 193 77 67 12 (096 to 14) not calculated
Results for these two comparisons are derived from very smallamounts of data and are provided here for completeness Theyshould be interpreted with caution
Local adverse events
Local adverse events were irritation of the area to which the topicalNSAID was applied including redness or erythema and itch orpruritus Where reported these were usually described as mild andtransient
All topical NSAIDs versus placebo
Forty-two studies contributed to this analysis of which three com-pared two different drugs with placebo (Aoki 1984 Diebshlag1990 Fujimaki 1985) Three studies had two treatment armscomparing different formulations or application regimens for di-clofenac with placebo which have been combined for this analysis
(Costantino 2011 Klainguti 2010 Predel 2012) In total 3619participants were treated with topical NSAIDs and 3121 withplacebo (Analysis 51)
bull The proportion of participants experiencing a local adverseevent with a topical NSAID was 43 (1553619 range 0 to33)
bull The proportion of participants experiencing a local adverseevent with placebo was 46 (1453121 range 0 to 32)
bull The RR of topical NSAID compared with placebo was 098(080 to 12)
bull There was no significant difference between treatmentgroups so the NNH was not calculated
Individual topical NSAIDs versus placebo
Results for local adverse events with individual topical NSAIDswhere there were adequate data for analysis are in Summary ofresults B and Analysis 12
Summary of results B Participants with local adverse events
Comparison Studies Participants NSAID()
Placebo()
RR(95 CI)
NNH(95 CI)
All NSAIDs 42 6740 43 46 098 (080 to 12) Not calculated
Diclofenac 15 3271 31 43 078 (056 to 11) Not calculated
Ketoprofen 8 852 11 95 12 (083 to 17) Not calculated
Piroxicam 3 522 23 54 042 (017 to 11) Not calculated
Felbinac 3 397 30 15 19 (049 to 75) Not calculated
Indomethacin 3 354 63 22 27 (091 to 77) Not calculated
Ibuprofen 3 321 10 43 23 (098 to 54) Not calculated
Systemic adverse events
All topical NSAIDs versus placebo
Thirty-six studies contributed data on systemic adverse eventsof which three compared two different drugs with placebo (Aoki1984 Diebshlag 1990 Fujimaki 1985) Two studies had two treat-
22Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
ment arms comparing different formulations or application regi-mens for diclofenac with placebo which have been combined forthis analysis (Klainguti 2010 Predel 2012) In total 2956 partic-ipants were treated with a topical NSAID and 2620 with placebo(Analysis 52)
bull Twenty-three studies reported no systemic adverse events inany arm of the study
bull The proportion of participants experiencing a systemicadverse event with a topical NSAID was 31 (922956)
bull The proportion of participants experiencing a systemicadverse event with placebo was 35 (912620)
bull The RR of topical NSAID compared with placebo was 096(073 to 13)
bull There was no significant difference between treatmentgroups so the NNH was not calculated
A further six studies did not report the occurrence or oth-erwise of systemic adverse events (Billigmann 1996 Julien1989 Kockelbergh 1985 Noret 1987 Ramesh 1983 Vecchiet1991) while two studies did not report numbers of participantswith systemic adverse events ( kermark 1990 Auclair 1989)Costantino 2011 reported that there were no systemic gastroin-testinal adverse events Two studies reported only on total adverseevents without distinguishing between local and systemic events(NCT01255423 NCT01272934)
Serious adverse events
Two studies reported serious adverse events In Hoffmann 2012one participant experienced three serious adverse events none ofwhich was judged to be related to the study medication (diclofenacplaster) In NCT01272934 one participant using diclofenac gelruptured the ligaments of the wrist There was no statement aboutlikely relationship to the study medication but this seems unlikely
Adverse event withdrawals
Forty-two studies reported data relating to adverse event with-drawals of which three compared two different drugs with placebo(Aoki 1984 Diebshlag 1990 Fujimaki 1985) Two studies hadtwo treatment arms comparing different formulations or applica-tion regimens for diclofenac with placebo which have been com-bined for this analysis (Klainguti 2010 Predel 2012) In total3365 participants received a topical NSAID and 3040 placebo(Analysis 53)
bull Forty-four comparisons reported no adverse eventwithdrawals
bull The proportion of participants withdrawing from the studydue to an adverse event after treatment with a topical NSAIDwas 098 (333365)
bull The proportion of participants withdrawing from the studydue to an adverse event after treatment with placebo was 099(303040)
bull The RR of topical NSAID compared to placebo was 10(064 to 16)
bull There was no significant difference between treatmentgroups so the NNH was not calculated
Four studies did not specifically mention adverse event with-drawals (Haig 1986 Hoffmann 2012 Klainguti 2010 Vecchiet1991) while one reported that one participant withdrew with mildpruritus but did not state the treatment arm (Joussellin 2003)Ten studies specifically reported withdrawals due to lack of efficacy(Dreiser 1989 Dreiser 1994 Kuehl 2011 Machen 2002 Maziegraveres2005b Maziegraveres 2005a Noret 1987 Predel 2013a Russell 1991Thorling 1990) (Appendix 5) Numbers of participants withdraw-ing were generally low with rates of 6 or less except in Kuehl2011 where the rate was 10 with active treatment (diclofenacplaster) and 12 with placebo We did not carry out any analysisbecause the outcome was inconsistently reported
2 Topical NSAID versus active comparator
Details of efficacy outcomes in individual studies are in Appendix4 and of adverse events and withdrawals in Appendix 5
Participants with clinical success
Topical NSAID versus oral NSAID
bull kermark 1990 compared indomethacin spray withindomethacin capsules with response rates of 55 (1222) withspray and 23 (522) with capsules
bull Hosie 1993 compared felbinac foam with ibuprofen tabletswith response rates of 64 (81127) with felbinac foam and72 (96133) with ibuprofen tablets
bull Whitefield 2002 compared ibuprofen gel with ibuprofentablets with response rates of 60 (3050) with gel and 54(3650) with tablets
There were insufficient data for meta-analysis for any one of thesecomparisons felbinac is not known to be better than placebo
Topical NSAID versus different formulation of the same
topical NSAID
bull Fioravanti 1999 compared DHEP (diclofenac) gelformulated with and without lecithin with response rates of70 (3550) in both treatment arms
bull Mahler 2003 compared DHEP (diclofenac) gel formulatedwith and without lecithin with response rates of 89 (8292)with lecithin and 70 (6288) without lecithin
bull Gallacchi 1990 compared topical diclofenac formulated asFlectorreg gel and Emugelreg with response rates of 76 (1925)in both treatment arms
23Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
bull Governali 1995 compared topical ketoprofen cream withgel with response rates of 93 (1415) with cream and 27 (415) with gel
There were insufficient data for analysis
Topical NSAID versus different topical NSAID
Eight studies compared one topical NSAID versus at least oneother topical NSAID piroxicam versus indomethacin (Aoki1984 Fujimaki 1985 Sugioka 1984) ibuprofen versus ketopro-fen (Curioni 1985 Picchio 1981) ketoprofen versus etofena-mate (Curioni 1985 Tonutti 1994) ibuprofen versus etofenamate(Curioni 1985) ketorolac versus etofenamate (Diebshlag 1990)and diclofenac versus lysine clonixinate (Hofman 2000)There were sufficient data to compare only piroxicam with in-domethacin (Aoki 1984 Fujimaki 1985 Sugioka 1984 Analysis61)
bull The proportion of participants experiencing clinical successwith topical piroxicam was 56 (185330 range 49 to 78)
bull The proportion of participants experiencing clinical successwith topical indomethacin was 45 (140311 range 33 to64)
bull The RR of piroxicam compared with indomethacin was 12(11 to 14)
bull The NNT for successful treatment was 91 (53 to 30) Forevery nine participants treated with topical piroxicam onewould experience a clinical success who would not haveexperienced one with topical indomethacin
Topical NSAID versus different topical intervention
One study compared diclofenac gel with a herbal product calledTraumeel gel under double-blind conditions with response ratesfor being pain-free at seven days of 8137 (58) with diclofenacgel and 7140 (50) with Traumeel gel (Gonzaacutelez de Vega 2013)
Local adverse events
Topical NSAID versus oral NSAID
Two studies comparing a topical NSAID with an oral NSAIDprovided data on local adverse events ( kermark 1990 Hosie1993) There were five events with topical NSAID and three withoral NSAID which were too few for analysis
Topical NSAID versus different topical NSAID
All nine studies comparing one topical NSAID with at least oneother reported on local adverse events with a total of 48 events in1005 participants (48) There were sufficient data to compare
only piroxicam with indomethacin (Aoki 1984 Fujimaki 1985Sugioka 1984 Analysis 62)
bull The proportion of participants experiencing local adverseevents with topical piroxicam was 21 (7340 range 12 to28)
bull The proportion of participants experiencing local adverseevents with topical indomethacin was 10 (33331 range 29to 15)
bull The RR of piroxicam compared with indomethacin was021 (009 to 047)
bull The NNT to prevent a local adverse event was 13 (87 to23) For every thirteen participants treated with topicalpiroxicam one would not experience a local adverse event whowould have experienced one with topical indomethacin
Topical NSAID versus different topical intervention
Gonzaacutelez de Vega 2013 reported that adverse events were infre-quent and mild to moderate in intensity but did not distinguishbetween local and systemic events Numbers of participants expe-riencing any adverse event were 8147 with diclofenac gel and 14148 with Traumeel gel
Systemic adverse events
kermark 1990 reported numbers of events rather than numbersof participants with events while Tonutti 1994 and Whitefield2002 reported no adverse events attributable to the study medi-cation and Fioravanti 1999 Gallacchi 1990 Gualdi 1987 andSugioka 1984 did not mention systemic adverse events Gonzaacutelezde Vega 2013 did not distinguish between local and systemicevents In the remaining studies a total of 16 events were reportedin topical NSAID treatment arms (797 participants 2) and 11with ibuprofen tablets (134 participants 8)
Serious adverse events
No serious adverse events were reported in any treatment arm
Withdrawals
The only withdrawals reported due to adverse events were in stud-ies with placebo treatment arms ( kermark 1990 Fujimaki 1985)and have been reviewedThree studies reported withdrawals due to lack of efficacy (Gonzaacutelez de Vega 2013 Hofman 2000 Tonutti 1994) (Appendix5) There were insufficient data for analysisSome studies reported exclusions from analysis (efficacy or safetyor both) following randomisation mainly due to protocol vio-lations or loss to follow-up (Appendix 5) There is no reason tobelieve these exclusions would introduce systematic bias and thenumbers involved were not likely to influence results
24Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
D I S C U S S I O N
This updated review of topical NSAIDs for acute musculoskele-tal pain in adults differs from previous reviews Previously dataallowed only for comparison of individual topical NSAIDs withplacebo irrespective of formulation With a substantial amount ofnew data for diclofenac it is possible to distinguish effects of for-mulation for individual NSAIDs Because formulation chemistrycan substantially affect the rate and total amount of drug accessingsubcutaneous injured tissues the effect of formulation may be asimportant as the individual NSAID used Drug and formulationshould thus be considered together when assessing efficacy andthis is now possible
We have also included an assessment of the daily dose of NSAIDapplied to the skin This involved having information on the con-centration of NSAID in the preparation the amount used andthe frequency of use Not all studies reported all three but an esti-mation of topical doses applied was possible It varied by factors ofthree or four for each topical NSAID However for topical formu-lations the key issue is less the dose applied but the amount thatpenetrates locally (producing analgesic effect) and the amount en-tering the systemic circulation (producing potential harms) Bothwill depend on the exact formulation of the topical agent andwhether there is occlusion (Moore 2008a)
Summary of main results
This review included 61 studies comparing a topical NSAID withplacebo another topical NSAID or an oral NSAID In total5311 participants were treated with a topical NSAID 3470 withplacebo and 220 with an oral NSAID There were 63 moreparticipants than in the previous version of this review Conditionstreated were sprains strains and contusions mainly resulting fromsports injuries and overuse injuries such as tendinitisFormulations of topical diclofenac ibuprofen ketoprofen piroxi-cam and indomethacin demonstrated significantly higher rates ofclinical success than matching topical placebo lacking the NSAIDbenzydamine did not Three drug and formulation combinationshad NNTs for clinical success below 4 For diclofenac Emulgelreghad the lowest NNT of 18 (15 to 21) in two studies using atleast 50 pain intensity reduction as the outcome (high qualityevidence) Diclofenac plasters other than Flectorreg also had a lowNNT of 32 (26 to 42) based on good or excellent responses inrelatively recent studies (high quality evidence) Ketoprofen gelhad an NNT of 25 (20 to 34) from five studies in the 1980ssome with less well defined outcomes (moderate quality evidence)Ibuprofen gel had an NNT of 39 (27 to 67) from two stud-ies with outcomes of marked improvement or complete remission(moderate quality evidence) All other drug and formulation com-binations had NNT values above 4 indicating lesser efficacyThese results are better than alternative topical products that mightbe used for acute musculoskeletal pain There is no evidence tosupport the use of topical salicylate rubefacients (Derry 2014)
Treatment with a topical NSAID was not associated with anincrease in local adverse events (skin reactions) compared withplacebo (inert carrier) or in withdrawals due to adverse events(high quality evidence) The inert carrier was sometimes associ-ated with mild skin irritation but this rarely led to cessation oftreatment and quickly resolved Systemic adverse events were un-common and did not differ between topical NSAID and placebo(high quality evidence) Two participants experienced serious ad-verse events with diclofenac plaster and diclofenac gel but it isunlikely that these were related to the study medicationsThere were insufficient data directly comparing a topical NSAIDwith the same oral NSAID to draw conclusions about efficacyBased on very limited data for oral NSAIDs there were fewersystemic adverse events with topical than oral treatment Therewere sufficient data only for topical piroxicam compared withtopical indomethacin to compare one topical agent with anotherThese limited data suggested that piroxicam was more effectivethan indomethacin and was less likely to cause local adverse eventsIt is worth noting here that topical indomethacin did not givesignificantly better pain relief than placebo in two of the threestudies in this analysis
Overall completeness and applicability ofevidence
There is a tension between pooling studies to produce analyseswith larger numbers and the subsequent large increases in clinicaland statistical heterogeneity on the one hand and using the ap-proach of clinical homogeneity with subsequent smaller numbersof participants on the other hand Previous reviews have takenthe former approach that is useful in demonstrating that topicalNSAIDs rsquoworkrsquo by being significantly better than placebo Becauseof the substantial increase in the amount of data available in thisreview we have chosen to seek greater clinical homogeneity thisproduces results that are more relevant to patient and prescriberchoiceThere were too few studies comparing one topical NSAID versusanother or versus the same oral NSAID to allow meaningful directcomparisons between individual drugs or routes of administrationThe conditions treated in these studies are representative of thoselikely to be suitable for acute treatment with topical NSAIDs Themean age of participants in individual studies ranged from 25 to57 years and the nature of recruitment in many studies meant thatparticipants were actively engaged in sporting activities Neverthe-less older people in their 60s to 80s were also included in somestudies and the low levels of predominantly mild adverse eventsmeans that this route of administration of NSAIDs is suitable forall age groups able to manage the application processInformation from other sources mainly randomised studies last-ing 12 weeks or more in older populations with arthritis tendto confirm this A systematic review of topical NSAIDs in olderadults was difficult to interpret but suggested that the range of
25Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
withdrawal rates in these studies was similar with topical and oralNSAIDs (Makris 2010) It also claimed potentiation of warfarineffects but that was with topical salicylate not an NSAID Incontrast a pooled safety analysis of topical diclofenac in peopleaged 75 years or older reported minimal changes with a mean re-duction in haemoglobin of less than 1 gL with topical diclofenacand a mean systolic blood pressure reduction of almost 4 mm Hg(Roth 2012) Two large randomised 12-week studies comparingtopical with oral diclofenac in arthritis reported lower rates of gas-trointestinal adverse events with topical than oral especially severeevents but larger reductions in haemoglobin with oral diclofenac(Simon 2009 Tugwell 2004)The available evidence was limited by numbers to comment on rarebut potentially serious adverse events One example is the potentialfor photo-sensitivity reactions with topical ketoprofen Currentadvice from the Medicines and Healthcare products RegulatoryAgency in the UK is to avoid direct sunlight ultraviolet (UV) rayssunlamps and sunbeds while using topical ketoprofen and to seea healthcare professional or go to hospital if they experience a skinreaction to sunlight sunlamps or sunbeds (MHRA 2009)
Quality of the evidence
All included studies were both randomised and double-blind nonewas considered at high risk of methodological bias Many werecarried out in the 1980s and 1990s when methodological rigorand detailed reporting were not given such high priority and stud-ies did not always report details of the randomisation treatmentallocation and blinding processes More recent studies often didreport methodological details and tended to be larger (see Figure4 for a comparison of quality of reporting for different dates andNSAIDs) Our primary outcome of clinical success was not alwayswell-defined and was measured using different scales but againmore recent studies tended to report outcomes betterThe studies were conducted in different conditions with some-what different outcome definitions and duration and with differ-ent topical NSAIDs and formulations Moreover the small size ofmany of the studies is likely to result in considerable chance varia-tion (Counsell 1994 Moore 1998b) These factors would accountfor the high I2 values seen in several analyses Despite these sourcesof potential clinical heterogeneity most studies showed benefit oftopical NSAID over placeboThe design of studies to be able to demonstrate analgesic sensi-tivity is important in self limiting conditions such as strains andsprains Too long a duration and the condition results in sponta-neous resolution of painful symptoms while too short a durationmay be inadequate to show any effect The decision by trialists toconcentrate on outcomes closest to seven days of treatment appearsto be prudent and has been adopted in this and previous reviewsThere are potential differences in response to treatment betweenstrains and sprains and overuse-type injuries such as tendinitis
and future reviews may examine this At the present time thereare too few existing trials to explore any differences adequatelyBaseline pain may be a cause for concern Seven studies didnot report baseline pain levels (Billigmann 1996 Curioni 1985Haig 1986 NCT01255423 NCT01272934 NCT01272947Sinniger 1981) and a further 11 reported either mean levels ofless than moderate pain or a significant proportion of individualswith less than moderate pain ( kermark 1990 Aoki 1984 Auclair1989 Diebshlag 1990 Fujimaki 1985 Jenoure 1997 Linde 1985Picchio 1981 Ramesh 1983 Sugioka 1984 Whitefield 2002)using recognised scales Insufficient pain at baseline compromisesthe ability of a study to demonstrate any improvement All thenewly added studies reported baseline pain to be of at least mod-erate intensity
Potential biases in the review process
There has been greater interest in topical NSAIDs in recent yearsmainly because lower systemic drug levels reduce the risk of trou-blesome and severe adverse events particularly in the gastroin-testinal tract and renal and cardiovascular systems Most of theattention has been in chronic conditions such as osteoarthritiswith few studies in acute painful conditions Low levels of seriousadverse events with topical NSAIDs has been noted previously(Evans 1995) and the near absence of serious adverse events inthis review is unlikely to be due to any biases in the review processOne potential bias is that clinical trials for topical NSAIDs may nothave been published One previous review did find previously un-published trials (Moore 1998a) but a subsequent attempt that in-cluded extensive contacts with pharmaceutical companies revealedno additional data (Mason 2004a) While some old unpublishedstudies of topical NSAIDs in acute painful conditions may existthey constitute an unknown number of studies and participantswhose results are unknown and are likely to remain unknownFurthermore their relevance to current clinical practice may belimited as better formulations are developed New systems of trialregistration mean that we know what recent studies have been doneor are ongoing the number of studies and participants is knowneven if their results remain unknown We identified in Clinicaltri-alsgov three unpublished studies (612 participants) with adverseevent data but no dichotomous efficacy data 14 completed un-published studies (4403 participants) with no results posted andthree ongoing studies (880 participants)For the main topical NSAIDs of interest and where most informa-tion exists about 4200 participants in this review provided data onefficacy for diclofenac ibuprofen ketoprofen indomethacin andbenzydamine compared with placebo For efficacy there are un-known results from almost 5900 participants in studies known tohave been done but essentially unpublished Almost 6500 partic-ipants in this review provided information on local adverse eventsfor topical NSAIDs compared with placebo For local adverseevents the unknown results from known studies represents almost
26Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
5300 participants It is clear that identified unpublished but un-available study data amounts to a further potentially large increasein knowledge over and above the 60 increase in numbers ofparticipants already included in this updated reviewBased on efficacy data on known and available study results un-published trials showing no difference between any topical NSAIDand topical placebo and involving 5500 participants would haveto exist in order for the NNT to be as high as 9 at which pointthe effectiveness of topical NSAIDs would become clinically irrel-evant (Moore 2006) This amount of unpublished negative data isobviously available and while a negative result in all the identifiedstudies is unlikely knowledge would be greatly served by havingthese unpublished trial results availableWe have not yet attempted to obtain results from these clinicaltrials from the trial sponsors because this takes a considerableamount of time and may not be successful Moreover the studieswere spilt between nine different sponsors Cerimon Pharmaceu-ticals (five studies) Novartis (four studies) Endo Pharmaceuticals(three studies) GlaxoSmithKline (two studies) Hisamatsu Phar-maceutical (two studies) and one each from Actavis Pfizer Im-primis Pharmaceutical and Strategic Science amp Technologies
Agreements and disagreements with otherstudies or reviews
A review published in 2004 included some of the studies in thisreview and reported an NNT for all topical NSAIDs combinedof 38 (34 to 44) for clinical success equivalent to half pain reliefat seven days (Mason 2004a) That review found no differencebetween topical NSAID and placebo for local adverse events as didthis review In turn the Mason review was in broad agreement withthe original systematic review on topical NSAIDs (Moore 1998a)To our knowledge no previous review assembled sufficient trialdata to analyse results by both drug and formulation as was donehere
A U T H O R S rsquo C O N C L U S I O N S
Implications for practice
For people with acute musculoskeletal pain
Topical NSAIDs can provide good levels of pain relief in acuteconditions such as sprains strains and overuse injuries proba-bly similar to that provided by oral NSAIDs Gel formulations ofdiclofenac (as Emulgelreg) ibuprofen and ketoprofen and somediclofenac patches provide the best effects Adverse events are usu-ally minimal with topical NSAIDs
For clinicians
Topical diclofenac ibuprofen or ketoprofen gels provide goodpain relief for painful acute musculoskeletal conditions and arebetter tolerated than oral formulations These drugs and formu-lations are more likely to be cost effective than alternative topicalpreparations such as topical rubefacients
For policy makers
Topical NSAIDs are not associated with an increased incidence oflocal skin reactions compared with the inert carrier and while thecarrier may cause mild transient irritation it is rarely troublesomeTopical NSAIDs do not cause systemic (mainly gastrointestinal)problems commonly seen with oral NSAIDs making them partic-ularly useful for individuals unable to tolerate oral administrationor for whom it is contraindicated
For funders
Topical diclofenac ibuprofen or ketoprofen gels should be consid-ered for initial treatment of acute musculoskeletal painful condi-tions where there are no contraindications such as damaged skinThese drugs and formulations are more likely to be cost effectivethan alternative topical preparations such as topical rubefacientsBecause formulations of topical NSAIDs are likely to change overtime the relevant trials performed and reported in or before the1990s must be limited and may be questionable Funders mightwish to consider asking pharmaceutical companies without recenttrial evidence for their products to produce it
Implications for research
General
The general thrust of these findings is that gel formulations of top-ical diclofenac ibuprofen and ketoprofen work best but for somedrugs (ketoprofen for instance) studies were pre-1990 Thesestudies may not be relevant to products available now Becauseformulation can have a significant effect on efficacy formulationchanges should be accompanied by relevant randomised trials
Design
The design of the trials is generally good and the sports injurymodel appears to be reliable and reproducible Modern studieshave ensured that participants entering the trials have at least mod-erate pain and this helps sensitivity to detect an analgesic responseMajor changes to the design of these trials would not appear to beneeded
27Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Measurement (endpoints)
A major issue is not in the measurement of pain as most studiesespecially modern studies have used standard pain intensity andpain relief scales However reporting of average pain changes isinadequate and the use of responder analyses (at least 50 painintensity reduction or people experiencing mild or no pain) ispreferred
Comparison between active treatments
Indirect comparisons with placebo are probably as informative asuse of an active comparator
A C K N O W L E D G E M E N T S
The Oxford Pain Relief Trust the NHS Cochrane CollaborationProgramme Grant Scheme and the NIHR Biomedical ResearchCentre Programme provided support for the earlier review
The National Institute for Health Research (NIHR) is the largestsingle funder of the Cochrane Pain Palliative and SupportiveCare Review Group Disclaimer the views and opinions expressedherein are those of the authors and do not necessarily reflect thoseof the NIHR National Health Service (NHS) or the Departmentof Health
R E F E R E N C E S
References to studies included in this review
Airaksinen 1993 published data onlyAiraksinen O Venaumllaumlinen J Pietilaumlinen T Ketoprofen 25gel versus placebo gel in the treatment of acute soft tissueinjuries International Journal of Clinical PharmacologyTherapy and Toxicology 199331(11)561ndash3
kermark 1990 published data only
kermark C Forsskaringhl B Topical indomethacin in overuseinjuries in athletes A randomized double blind studycomparing Elmetacinreg with oral indomethacin andplacebo International Journal of Sports Medicine 199011(5)393ndash6
Aoki 1984 published data only
Aoki T Numajiri M Yamamoto M A well controlledcomparative study of piroxicam gel indomethacin geland placebo gel in the treatment of trauma JapanesePharmacology and Therapeutics 198412(12)101ndash17
Auclair 1989 published data onlyAuclair J Georges M Grapton X Gryp L DrsquoHooghe MMeisser RG et al A double-blind controlled multi-centerstudy of percutaneous niflumic acid gel and placebo in thetreatment of Achilles heel tendinitis Current TherapeuticResearch 198946(4)782ndash8
Billigmann 1996 published data onlyBilligmann PW Treatment of ankle distortion withibuprofen gel [Therapie von Sprunggelenksndashdistosionenmit ibuprofenndashmikrogel] Therapiewoche 199646(21)1187ndash92
Campbell 1994 published data onlyCampbell J Dunn T Evaluation of topical ibuprofen creamin the treatment of acute ankle sprains Journal of Accidentand Emergency Medicine 199411(3)178ndash82
Chatterjee 1977 published data onlyChatterjee DS A double-blind clinical study withbenzydamine 3 cream on soft tissue injuries in anoccupational health centre Journal of International MedicalResearch 19775(6)450ndash8
Costantino 2011 published data only
Costantino C Kwarecki J Samokhin AV Mautone GRovati S Diclofenac epolamine plus heparin plasterversus diclofenac epolamine plaster in mild to moderateankle sprain a randomized double-blind parallel-groupplacebo-controlled multicentre phase III trial ClinicalDrug Investigation 201131(1)15ndash26 [DOI 10216511585890-000000000-00000]
Coudreuse 2010 published data onlyCoudreuse JM de Vathaire F Effect of a plaster containingDHEP and heparin in acute ankle sprains with oedemaa randomized double-blind placebo-controlled clinicalstudy Current Medical Research and Opinion 201026(9)2221ndash8 [DOI 101185030079952010508020]
Curioni 1985 published data only
Curioni BG Di Domenica F Daolio P Spignoli GEvaluation of ibuproxam gel in traumatology [Valutazionedellrsquoefficacia terapeutica di ibudros gel in traumatologia]Clinica Europea 198524(3)456ndash60
Diebshlag 1990 published data only
Diebschlag W Nocker W Bullingham R A double-blindstudy of the efficacy of topical ketorolac tromethamine gelin the treatment of ankle sprain in comparison to placeboand etofenamate Journal of Clinical Pharmacology 199030
(1)82ndash9
Dreiser 1988 published data onlyDreiser RL Clinical trial of efficacy and tolerability oftopical ibuprofen in the treatment of tendinitis Le JournalInternational De Meacutedecine 198811915ndash31
Dreiser 1989 published data only
Dreiser RL Clinical trial - Fatsum gel FG-6 Supplied byMenarini 1989
Dreiser 1990 published data onlyDreiser RE Charlot J Lopez A Ditisheim A Clinicalevaluation of niflumic acid gel in the treatment ofuncomplicated ankle sprains Current Medical Research andOpinion 199012(2)93ndash9
28Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Dreiser 1994 published data only
Dreiser RL Roche R de Sahb R Thomas F Leutenegger EFlurbiprofen local action transcutaneous (LAT tm) clinicalevaluation in the treatment of acute ankle sprains EuropeanJournal of Rheumatology and Inflammation 199414(4)9ndash13
Fioravanti 1999 published data onlyFioravanti A Cicero MR Nerucci F Manopulo RMarcolongo R Double-blind controlled clinical study of theefficacy and tolerability of diclofenac-N-(2-hydroxyethyl)-pyrrolidine lecithin gel compared with diclofenac-N-(2-hydroxyethyl)-pyrrolidine gel in patients with periand extraarticular inflammatory diseases Drugs underExperimental and Clinical Research 199925(5)235ndash40
Fujimaki 1985 published data only
Fujimaki E et al Clinical evaluation of piroxicam gelversus indomethacin gel and placebo in the treatment ofmuscle pain a double-blind multicenter study JapanesePharmacology and Therapeutics 198512(12)119ndash37
Gallacchi 1990 published data only
Gallacchi G Mautone G Lauldi P Topical treatmentwith diclofenac hydroxyethylpyrrilidine (Flector gel 1)Clinical Trials Journal 199027(1)58ndash64
Gonzaacutelez de Vega 2013 published data only
Gonzaacutelez de Vega C Speed C Wolfarth B Gonzaacutelez JTraumeel vs diclofenac for reducing pain and improvingankle mobility after acute ankle sprain a multicentrerandomised blinded controlled and non-inferiority trialInternational Journal of Clinical Practice 201367(10)979ndash89 [DOI 101111ijcp12219]
Governali 1995 published data only
Governali E Casalini D A controlled clinical studyon ketoprofen gel 5 versus ketoprofen ointment 1in patients with post-traumatic lesions [Ricerda clinicacontrollata tra ketoprofene gel 5 e ketoprofene crema1 in pazienti con postumi di lesioni traumatiche]Riabilitazione 199528(1)61ndash9
Gualdi 1987 published data only
Gualdi A Bonollo L Martini A Forgione A Non-steroidal anti-inflammatory drugs for topical therapy intraumatology a double-blind study with flunoxaprofen andketoprofen [Antinflammatori no steroidi per uso topicoin traumatologia studio clinico con flunoxaprofene echetoprofene] Riforma Medica 1987102(10)401ndash4
Haig 1986 published data only
Haig G Portable thermogram technique for topicallyapplied benzydamine cream in acute soft-tissue injuriesInternation Journal of Tissue Reactions 19868(2)145ndash7
Hoffmann 2012 published data only
Hoffmann P Kopa ka P Gugliotta B Rovati S Efficacyand tolerability of DHEP-heparin plaster in reducingpain in mild-to-moderate muscle contusions a double-blind randomized trial Current Medical Researchand Opinion 201228(8)1313ndash21 [DOI 101185030079952012709182]
Hofman 2000 published data only
Hofman J Nasswetter G Cayetti LM Lysine clonixinate gelin soft tissue injuries Controlled randomized prospectivedouble-blind clinical trial with diclofenac [Clonixinatode lisina gel en lesiones de tejidos blandos ensayo clinicoprospectivo doble ciego al azar controlado con diclofenac]Prensa Medica Argentina 200087(5)513ndash20
Hosie 1993 published data only
Hosie GAC The topical NSAID felbinac versus oralibuprofen a comparison of efficacy in the treatment ofacute lower back injury British Journal of Clinical Research199345ndash17
Jenoure 1997 published data only
Jenoure PJ Rostan A Gremion G Meier JL Grossen RBielinki R et al Multicentre double-blind controlledclinical study on the efficacy of diclofenac epolamineTissugel plaster in patients with epicondylitis [Studiomulticentrico controllato in doppio cieco su diclofenacTissugel plaster in pazienti con epicondilite] MedicinaDello Sport 197750(3)285ndash92
Joussellin 2003 published data onlylowast Joussellin E Flector Tissugel for the treatment ofpainful ankle sprains [Flector Tissugel dans le traitementdes entorses douloureuses de la cheville] Journal deTraumatologie du Sport 2003201S5ndash9Lionberger DR Joussellin E Lanzarotti A Yanchick JMagelli M Diclofenac epolamine topical patch relieves painassociated with ankle sprain Journal of Pain Research 2011447ndash53 [DOI 102147JPRS15380]Lionberger DR Joussellin E Yanchick J Magelli MLanzarotti A Pooled analysis of clinical trial data evaluatingthe safety and effectiveness of diclofenac epolamine topicalpatch 13 for the treatment of acute ankle sprain OpenAccess Journal of Sports Medicine 2011275ndash84 [DOI102147OAJSMS17048]
Julien 1989 published data onlyJulien D Clinical trial - Fastum gel FG-8 Supplied byMenarini 1989
Klainguti 2010 published data onlyKlainguti A Forgacs A Berkes I Castellacci E A plastercontaining DHEP and heparin for mild to moderatecontusions and sprains with haematoma a double-blind randomized study Current Medical Researchand Opinion 201026(9)2243ndash51 [DOI 101185030079952010508022]
Kockelbergh 1985 published data onlyKockelbergh M Verspeelt P Caloine R Dermaux F Localanti inflammatory treatment with a ketoprofen gel currentclinical findings Journal Belge de Medecine Physique et deRehabilitation 19858(4)205ndash13
Kuehl 2011 published data only
Kuehl K Carr W Yanchick J Magelli M Rovati SAnalgesic efficacy and safety of the diclofenac epolaminetopical patch 13 (DETP) in minor soft tissue injuryInternational Journal of Sports Medicine 201132(8)635ndash43[DOI 101055s-0031-1275359]
29Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Li 2013 published data only
Li C Frangione V Rovati S Zheng Q Diclofenacepolamine medicated plaster in the treatment of minor softtissue injuries a multicenter randomized controlled trialCurrent Medical Research and Opinion 201329(9)1137ndash46[DOI 101185030079952013816669]
Linde 1985 published data only
Linde F Hvass I Juumlrgensen U Madsen F Treatment ofsprained ankles with 5 benzydamine creme A double-blind study [Ankelforstuvninger behandlet med benzydamin5 creme] Ugeskrift for Laeger 1985148(1)12ndash3
Machen 2002 published data onlyMachen J Whitefield M Efficacy of a proprietary ibuprofengel in soft tissue injuries a randomised double blindplacebo controlled study International Journal of ClinicalPharmacology 200256(2)102ndash6
Mahler 2003 published data only
Mahler P Mahler F Duruz H Ramazzina M Liguori VMautone G Double-blind randomized controlled studyon the efficacy and safety of a novel diclofenac epolaminegel formulated with lecithin for the treatment of sprainsstrains and contusions Drugs under Experimental andClinical Research 200329(1)45ndash52
Maziegraveres 2005a published data onlyMaziegraveres B Rouanet S Velicy J Scarsi C Reiner V Topicalketoprofen patch (100 mg) for the treatment of anklesprain a randomized double-blind placebo-controlledstudy American Journal of Sports Medicine 200533(4)515ndash23 [DOI 1011770363546504268135]
Maziegraveres 2005b published data onlyMaziegraveres B Rouanet S Guillon Y Scarsi C Reiner VTopical ketoprofen patch in the treatment of tendinitisa randomized double blind placebo controlled studyJournal of Rheumatology 200532(8)1563ndash70
McLatchie 1989 published data onlyMcLatchie GR McDonald M Lawrence GF RogmansD Lisai P Hibberd M Soft tissue trauma a randomisedcontrolled trial of the topical application of felbinac a newNSAID British Journal of Clinical Practice 198943(8)277ndash80
Morris 1991 published data only
Morris WD Scott HV Peters WA Ketelbey JW Felbinactopical gel for acute soft tissue sports injuries New ZealandJournal of Sports Medicine 19911945ndash7
NCT01255423 unpublished data only
Novartis (Sponsors) A randomized double-blind multi-center placebo-controlled parallel group study to evaluatethe efficacy and safety of diclofenac sodium topical gel(DSG) 1 applied 4 times daily in subjects with acuteankle sprain clinicaltrialsgovct2showNCT01255423(accessed 3 February 2015) 2012 [CTG NCT01255423]
NCT01272934 unpublished data only
Novartis (Sponsors) A randomized double-blind multi-center placebo-controlled parallel group study to evaluatethe efficacy and safety of diclofenac sodium topical gel
(DSG) 1 applied four times daily in subjects withacute ankle sprain clinicaltrialsgovct2showresultsNCT01272934 (accessed 3 February 2015) 2013 [CTGNCT01272934]
NCT01272947 unpublished data onlylowast Novartis (Sponsors) A randomized double-blindmulti-center placebo-controlled parallel group study toevaluate the efficacy and safety of diclofenac sodium topicalgel (DSG) 1 applied four times daily in subjects withacute blunt soft tissue injuriescontusions of the limbsclinicaltrialsgovct2showresultsNCT01272947 (accessed3 February 2015) 2012 [CTG NCT01272947]Pallay R Pabst H Giannetti B Burnett I Monnet JDiclofenac sodium topical gel (DSG) 1 for acute softtissue injuriescontusions of the limbs 2013 AnnualAssembly of the American Academy of Physical Medicineand Rehabilitation National Harbor MD United States2013 Vol 5 (9 Suppl 1)S204
Noret 1987 published data only
Noret A Roty V Allington N Hauters P Zuinen C PoelsR Ketoprofen gel as topical treatment for sport injuriesActa Therapeutica 198713367ndash78
Parrini 1992 published data onlyParrini M Cabitza P Arrigo A Vanasia M Efficacy andtolerability of ketoprofen lysine salt foam for topical use inthe treatment of traumatic pathologies of the locomotorapparatus [Efficacia e tollerabilita del ketoprofene sale dilisina schiuma per uso topico nel trattamento di alcunepatologie traumatiche dellrsquoapparato locomotore] La ClinicaTerapeutica 1992141(9)199ndash204
Picchio 1981 published data only
Picchio AA Volta S Longoni A Controlled clinical trial ofibuprofen for topical use in sport injuries [Studio clinicocontrollato sullrsquoimpiego dellrsquoibuprofen per uso topico intraumatologica sportiva] Medicina dello Sport 198134403ndash6
Predel 2004 published data only
Mueller EA Kirch W Reiter S Extent and time courseof pain intensity upon treatment with a topical diclofenacsodium patch versus placebo in acute traumatic injury basedon a validated end point post hoc analysis of a randomizedplacebo-controlled trial Expert Opinion on Pharmacotherapy201011(4)493ndash8 [DOI 10151714656560903535898]lowast Predel HG Koll R Pabst H Dieter R Gallacchi GGiannetti B et al Diclofenac patch for topical treatmentof acute impact injuries a randomised double blindplacebo controlled multicentre study British Journal ofSports Medicine 200438(3)318ndash23 [DOI 101136bjsm2003005017]
Predel 2012 published data onlyPredel HG Hamelsky S Gold M Giannetti B Efficacy andsafety of diclofenac diethylamine 232 gel in acute anklesprain Medicine and Science in Sports and Exercise 201244
(9)1629ndash36 [DOI 101249MSS0b013e318257ed41]
30Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Predel 2013a published data only
Predel HG Giannetti B Seigfried B Novellini R MenkeG A randomized double-blind placebo-controlledmulticentre study to evaluate the efficacy and safetyof diclofenac 4 spray gel in the treatment of acuteuncomplicated ankle sprain Journal of InternationalMedical Research 201341(4)1187ndash202 [DOI 1011770300060513487639]
Predel 2013b published data onlyPredel HG Giannetti B Pabst H Schaefer A Hug AMBurnett I Efficacy and safety of diclofenac diethylamine116 gel in acute neck pain a randomized double-blindplacebo-controlled study BMC Musculoskeletal Disorders201314250 [DOI 1011861471-2474-14-250]
Ramesh 1983 published data only
Ramesh N Steuber U Ibuprofen in a cream vehicle foraccidental and sport injuries [Dolgit creme bei unfallndash undsportverletzungen] Therapiewoche 1983334563ndash70
Rowbotham 2003 published data only
Rowbotham M Galer B Block J Backonja M FlectorTissugel efficacy and safety in minor sport injuries Avs controlled clinical trial Journal of Sports Traumatology200320IS15ndashIS20
Russell 1991 published data only
Russell AL Piroxicam 05 topical gel compared to placeboin the treatment of acute soft tissue injuries a double-blind study comparing efficacy and safety Clinical andInvestigative Medicine 199114(1)35ndash43
Saillant 1998 published data only
Lionberger DR Joussellin E Yanchick J Magelli MLanzarotti A Pooled analysis of clinical trial data evaluatingthe safety and effectiveness of diclofenac epolamine topicalpatch 13 for the treatment of acute ankle sprain OpenAccess Journal of Sports Medicine 2011275ndash84 [DOI102147OAJSMS17048]lowast Saillant G Study comparing the efficacy and toleranceof Flector Tissugelreg to that of a placebo in the treatmentof benign ankle sprains [Eacutetude comparant lrsquoefficaciteacute et latoleacuterance de FlectorTissugelreg agrave celles drsquoun placebo dans letraitement des entorses beacutenignes de la cheville] Medicinedu Sport 1998721ndash5
Sanguinetti 1989 published data onlySanguinetti C Treatment of soft tissue injury with BPAAgel Results of an Italian multicenter study vs placebo[Trattemento con BPAA gel dei traumi dei tessuti moli]Clinica Terapeutica 1989130(5)255ndash258
Sinniger 1981 published data onlySinniger M Blanchard P Controlled clinical trial withFentiazac cream in sport microtraumatology Journal ofInternational Medical Research 19819(4)300ndash2
Spacca 2005 published data onlySpacca G Cacchio A Forgcs A Monteforte PRovetta G Analgesic efficacy of a lecithin-vehiculateddiclofenac epolamine gel in shoulder periarthritis andlateral epicondylitis a placebo-controlled multicenter
randomized double-blind clinical trial Drugs underExperimental and Clinical Research 200531(4)147ndash54
Sugioka 1984 published data only
Sugioka Y Multicenter clinical evaluation of piroxicam gelvs indomethacin gel in the treatment of non-traumaticdiseases of tendon or muscle Japanese Pharmacology andTherapeutics 198412139ndash53
Thorling 1990 published data only
Thorling J Linden B Berg R Sandahl A A double blindcomparison of naproxen gel and placebo in the treatment ofsoft tissue injuries Current Medical Research and Opinion199012242ndash8
Tonutti 1994 published data only
Tonutti A The use of ketoprofen gel 5 (orudis gel) intraumatology controlled double-blind study vs etofenamate[Utilizzazione del ketoprofene gel 5 (orudis gel) nellapratica traumatologica studio in doppio cieco controllatoverso etofenamato] Ortopedia e Traumatologia Oggi 199414(3)119ndash25
Vecchiet 1991 published data only
Vecchiet L Colozzi A Effects of meclofenamic acid in thetreatment of lesions deriving from minor traumatologyClinical Journal of Pain 19917 Suppl 1S54ndash9
Whitefield 2002 published data only
Whitefield M OrsquoKane CJA Anderson S Comparativeefficacy of a proprietary topical ibuprofen gel and oralibuprofen in acute soft tissue injuries a randomised doubleblind study Journal of Clinical Pharmacy and Therapeutics200227(6)409ndash17
References to studies excluded from this review
Ambrus 1987 published data only
Ambrus P Boumlhmer D Mobilat ointment in acute sprains[Mobilat Salbe bei akuten Distorsionen] Fortschritte derMedizin 1987105(13)259ndash62
Anon 1993 unpublished data only
Anon Comparative clinical efficacy of Oruvail piroxicamand diclofenac gels in soft tissue injury Unpublished
Ascherl 1982 published data onlyAscherl R Schlemmer H Blumel G Lechner F Theeffectiveness of etofenamate in minor sports injuries ofthe knee and ankle joint A double blind study [DieWirksamkeit der Etofenamat in kleineren Sportverletzungenam Knie und Sprunggelenk Eine DoppelblindndashStudie]Fortschritte der Medizin 1982100(37)1729ndash34
Bagliani 1976 published data only
Bagliani A Montalbetti L Topical treatment ofthrombophlebitis with feprazone and benzydamineControlled clinical study [Il trattamento topico ditromboflebite con feprazone e benzidamina Studio clinicocontrollato] Minerva Medica 197667(14)880ndash4
Baracchi 1982 published data only
Baracchi G Messina Denaro S Piscini S Experience ofthe topical use of isobutylfenylproprionic acid (Ibuprofen)in traumatic inflammation A double-blind comparison
31Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
with placebo [Esperienza sullrsquoimpiego topico dellrsquoacidoisobutilndashfenilndashpropionico (ibuprofen) nella flogositraumatica Confronto a doppia cecita con placebo]Gazzetta Medica Italiana 1982141(12)691ndash4
Boumlhmer 1995 published data onlyBoumlhmer D Ambrus P Treatment of muscular injurieswith diclofenac-diethylammonium emugel [Behandlungvon muskelverletzungen mit diclofenacndashdiethylammoniumemugel] Sportverletzung Sportschaden 1995994ndash5
Burnham 1998 published data only
Burnham R Gregg R Healy P Steadward R Theeffectiveness of topical diclofenac for lateral epicondylitisClinical Journal of Sport Medicine 19988(2)78ndash81
Cesarone 2008 published data onlyCesarone MR Belcaro G Pellegrini L Ledda A VinciguerraG Ricci A et al Treatment of ankle sprain in patientswith vascular diseases of the lower limbs MinervaCardioangiologica 200856 (5 Suppl)39ndash46
Coulibaly 2009 published data only
Coulibaly SK Courau S Staiger C The sprained anklecomparative study of a comfreyroot extract ointment versusa diclofenac gel [Entorses de la cheville eacutetude comparativeentre un extrait agrave base de racines de consoude et lediclofeacutenac] Phytotheacuterpie 20097147ndash9 [DOI 101007s10298-009-0383-x]
Diebschlag 1985 published data only
Diebschlag W Benzydamine cream in post-traumaticoedema International Journal of Tissue Reactions 19857(3)219ndash23
Diebschlag 1986 published data onlyDiebschlag W Diclofenac in blunt traumatic ankle jointswelling Volumetric monitoring in a placebo controlleddouble blind trial [Diclofenac in stumpfen traumatischenSprunggelenk Schwellung Volumetrische Uumlberwachungin einer Placebondashkontrollierten DoppelblindndashStudie]Fortschritte der Medizin 1986104(21)437ndash40
Diebschlag 1992 published data onlyDiebschlag W Nocker W Lehmacher W Treatment of acutesprains of the ankle joint A comparison of the effectivenessand tolerance of two gel preparations containingindomethacin [Die Behandlung der akuten Verstauchungendes Sprunggelenks Ein Vergleich der Wirksamkeitund Vertraumlglichkeit von zwei GelndashZubereitungen mitIndometacin] Fortschritte der Medizin 1992110(6)64ndash72
Fantato 1971 published data only
Fantato S De Gregorio M Clinical evaluation of topicalbenzydamine in traumatology Arzneimittel-Forschung197121(10)1530ndash5
Galer 2000 published data onlyGaler BS Rowbotham M Perander J Devers A FriedmanE Topical diclofenac patch relieves minor sports injury painresults of a multicenter controlled clinical trial Journal ofPain and Symptom Management 200019(4)287ndash94
Hallmeier 1986 published data only
Hallmeier B Michelbach B Etofenamate under tapebandages - a controlled study [Etofenamat untertapendashverbaumlnden] Medizinische Welt 198637(43)1344ndash8
Hallmeier 1988 published data onlyHallmeier B Efficacy and tolerance of etofenamate anddiclofenac in acute sports injuries [Wirksamkeit undVertraumlglichkeit von Etofenamat und Diclofenac bei akutenSportverletzungen] Rheuma 19888183ndash6
Kaneko 1999 published data only
Kaneko M Shimojo H Saito H Onuma Y YamashitaK Clinical evaluation of felbinac patch (SELSPOT) onpost-traumatic disease clinical comparative study versuscommercially available patch Japanese Pharmacology andTherapeutics 19992775ndash85
Kockelbergh 1985b published data only
Kockelbergh M Verspeelt P Caloine R Dermaux F Localanti-inflammatory treatment with a ketoprofen gel currentclinical findings [Traitement antindashinflammatoire local parun gel de keacutetoprofegravene donneacutees cliniques reacutecentes] JournalBelge de Medecine Physique et de Rehabilitation 19858(4)205-13 (study 2)
Kuwabara 2013 published data only
Kuwabara Y Hamamoto H Hikake S Miwa Y Arandomized multi-center double-blind placebo-controlledphase IIIII trial to evaluate the eftopical patch in thetreatment of pain Journal of Pain 2013 Vol 14 (4 Suppl1)S73
Lee 1991 published data only
Lee EH Lee PY Ngai AT Chiu EH Treatment of acute softtissue trauma with a topical non-steroidal anti-inflammatorydrug (biphenylacetic acid 3 gel) Singapore MedicalJournal 199132(4)238ndash41
Link 1996 published data onlyLink R Balint G Pavlik G Otto J Krause W Topicaltreatment of soft tissue rheumatism and athletic injuriesEffectiveness and tolerance of a new ketoprofen gel[Topische Behandlung von WeichteilndashRheumatismus undSportverletzungen Wirksamkeit und Vertraumlglichkeit einesneuen Ketoprofen Gel] Fortschritte der Medizin 1996114(25)311ndash4
May 2007 published data onlyMay JJ Lovell G Hopkins WG Effectiveness of1 diclofenac gel in the treatment of wrist extensortenosynovitis in long distance kayakers Journal of Scienceand Medicine in Sport 200710(1)59ndash65
Oakland 1993 published data only
Oakland C Rapier C A comparison of the efficacy of thetopical NSAID felbinac and ultrasound in the treatmentof acute ankle injuries British Journal of Clinical Research1993489ndash96
Odaglia 1987 unpublished data only
Odaglia G Sereni G Sports minor traumatology results ofa double-blind controlled clinical study ketoprofen (fastumgel 25) versus placebo Menarini unpublished data
32Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Picardi 1993 published data only
Picardi E De Iasio R Efficacy of percutaneous anti-inflammatory drugs (NSAIDs) in swimmers and waterpoloplayers [Efficacia dei farmaci antinfiammatori (FANS) pervia percutanea in atleti di nuoto e pallanuoto] ClinicaTerapeutica 1993143(6)507ndash9
Taboada 1992 published data onlyTaboada A Controlled trial of piroxicam gel associated withultrasound in acute disturbances of the locomotive system[Experienca controlada con gel de piroxicam asociado aultrasonidos en afecciones agudas del aparato locomotor]Prensa Medica Argentina 199279(10)630ndash2
Vanderstraeten 1990 published data only
Vanderstraeten G Schuermans P Study on the effect ofetofenamate 10 cream in comparison with an oral NSAIDin strains and sprains due to sports injuries Acta BelgicaMedica Physica 199013(3)139ndash41
Vinciguerra 2008 published data only
Vinciguerra G Belcaro G Cesarone MR Errichi BM DiRenzo A Errichi S et al Management of uncomplicatedankle sprains with topical or oral ketoprofen treatmentA registry study Minerva Cardioangiologica 200856 (5
Suppl)47ndash53
Von Klug 1977 published data onlyVon Klug H Experience with a locally applied anti-inflammatory drug [Erfahrungen mit einem localanwendbaren antirheumatikum] Arzneimittel-ForschungDrug Research 1977271350ndash4
References to studies awaiting assessment
NCT00351104 published data onlyEndo Pharmaceuticals (Sponsors) A randomized double-blind placebo-controlled parallel group phase III study ofthe efficacy tolerability and safety of ketoprofen topicalpatch 20 (KTP) in the treatment of pain associated withgrade 1 or grade 2 ankle sprain or strain clinicaltrialsgovct2showNCT00351104 (accessed 3 February 2015) 2008[CTG NCT00351104]
NCT00352625 unpublished data only
Endo Pharmaceuticals (Sponsors) A randomized double-blind placebo-controlled parallel group phase III study ofthe efficacy tolerability and safety of ketoprofen topicalpatch 20 (KTP) in the treatment of pain associatedwith tendonitis or bursitis of the shoulder elbow or kneeclinicaltrialsgovct2showNCT00352625 (accessed 3February 2015) 2008 [CTG NCT00352625]
NCT00426985 unpublished data only
Endo Pharmaceuticals (Sponsors) A randomized double-blind placebo-controlled parallel group phase iii study ofthe efficacy tolerability and safety of ketoprofen topicalpatch 20 (KTP) in the treatment of pain associatedwith tendonitis or bursitis of the shoulder elbow or kneeclinicaltrialsgovct2showNCT00426985 (accessed 3February 2015) 2008 [CTG NCT00426985]
NCT00640705 unpublished data only
Cerimon Pharmaceuticals (Sponsors) A randomizeddouble-blind placebo-controlled study of the efficacyand safety of a diclofenac sodium patch for the topicaltreatment of pain due to mild to moderate ankle sprainclinicaltrialsgovct2showNCT00640705 (accessed 3February 2015) 2008 [CTG NCT00640705]
NCT00640939 unpublished data onlyCerimon Pharmaceuticals (Sponsors) A randomizeddouble-blind placebo-controlled study of the efficacy andsafety of a diclofenac sodium patch for the topical treatmentof pain due to mild to moderate tendonitis or bursitisclinicaltrialsgovct2showNCT00640939 (accessed 3February 2015) 2008 [CTG NCT00640939]
NCT00680472 unpublished data onlyHisamitsu Pharmaceutical Co Inc (Sponsors) Arandomized multicenter double-blind placebo-controlledtwo-week study to assess the efficacy and safety of HKT-500 in subjects with acute shoulder pain clinicaltrialsgovct2showNCT00680472 (accessed 3 February 2015) 2015[CTG NCT00680472]
NCT00680784 unpublished data only
Hisamitsu Pharmaceutical Co Inc (Sponsors) Arandomized multicenter double-blind placebo-controlledstudy to evaluate the efficacy and safety of HKT-500 inthe treatment of pain associated with Grade I or Grade IIankle sprain clinicaltrialsgovct2showNCT00680784(accessed 3 February 2015) 2015 [CTG NCT00680784]
NCT00765700 unpublished data onlyEkman E Skrepnik S Jones M Lawson K Schupp JEfficacy and safety of ketoprofen 10 cream in acute softtissue injuries (phase 3 study tdlp-110-001) Proceedings ofthe 13th World Congress on Pain 2010 Aug 29 - Sept 2Montreal CanadaImprimis Pharmaceuticals Inc (Sponsors) A randomizedmulticenter double-blind placebo-controlled parallel-group phase 3 study to assess the efficacy and safetyof Ketotransdeltrade (Ketoprofen topical cream 10) inthe treatment of pain associated with mild to moderateacute soft tissue injury clinicaltrialsgovct2showNCT00765700 (accessed 3 February 2015) 2013 [CTGNCT00765700]
NCT00869063 unpublished data onlyCerimon Pharmaceuticals (Sponsors) A randomizeddouble-blind placebo-controlled study of the efficacy andsafety of a diclofenac sodium patch for the topical treatmentof acute pain due to mild to moderate wrist sprain strainor contusion clinicaltrialsgovct2showNCT00869063(accessed 3 February 2015) 2010 [CTG NCT00869063]
NCT00869180 unpublished data onlyCerimon Pharmaceuticals (Sponsors) A randomizeddouble-blind placebo-controlled study of the efficacy andsafety of a diclofenac sodium patch for the topical treatmentof acute pain due to mild to moderate ankle sprainclinicaltrialsgovct2showNCT00869180 (accessed 3February 2015) 2010 [CTG NCT00869180]
33Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT00931866 unpublished data only
Cerimon Pharmaceuticals (Sponsors) A randomizeddouble-blind placebo-controlled study of the efficacy andsafety of a diclofenac sodium patch for the topical treatmentof acute pain due to mild to moderate soft tissue injuriesclinicaltrialsgovct2showNCT00931866 (accessed 3February 2015) 2010 [CTG NCT00931866]
NCT01874626 unpublished data only
Strategic Science amp Technologies LLC (Sponsors) Arandomized double-blind placebo-controlled multi-dosepivotal study to determine the efficacy and safety of SST0225 a topical ibuprofen cream in the treatment of painassociated with acute ankle sprain clinicaltrialsgovct2showNCT01874626 (accessed 3 February 2015) 2014[CTG NCT01874626]
NCT01957215 unpublished data only
GlaxoSmithKline (Sponsors) A clinical study to assess theefficacy of pain relief of topical indomethacin patch overplacebo in ankle sprain patients clinicaltrialsgovct2showNCT01957215 (accessed 3 February 2015) 2015 [CTGNCT01957215]
NCT02324270 unpublished data only
Actavis Inc (Sponsors) Randomized double-blindmultiple-center placebo-controlled study comparing thesafety and efficacy of generic diclofenac epolamine toFlectorreg patch in the treatment of acute pain due to minorankle sprain clinicaltrialsgovct2showNCT02324270(accessed 3 February 2015) 2014 [CTG NCT02324270]
Sarzi-Puttini 2014 published data only
Sarzi-Puttini P Atzeni F Damiani C Casale R BarbagalloM Cazzola M A double-blind randomised parallelgroup active controlled multicentre study to assess thetherapeutic non-inferiority of SKP-021 a 03 ketoprofenpatch versus diclofenac sodium patch in patients withacute inflammatory musculoskeletal injuries Proceeding ofthe Annual European Congress of Rheumatology of theEuropean League Against Rheumatism EULAR 2014 June11-14 Paris France 201473
References to ongoing studies
NCT01945034 unpublished data onlyPfizer (Sponsors) Placebo-controlled double-blindevaluation of the efficacy and safety of ibuprofen 5 topicalgel for the treatment of ankle sprain clinicaltrialsgovct2showNCT01945034 (accessed 3 February 2015) 2014[CTG NCT01945034]
NCT02100670 unpublished data onlyGlaxoSmithKline (Sponsors) A clinical study to assessthe efficacy and onset of pain relief of topical MFC51123diclofenac-menthol gel versus controls in ankle sprainclinicaltrialsgovct2showNCT02100670 (accessed 3February 2015) 2014 [ClinicalTiralsgov NCT02100670]
NCT02290821 unpublished data only
Novartis (Sponsors) A randomized double-blind placebo-controlled parallel group study to evaluate the efficacy andsafety of diclofenac sodium topical gel (DSG) 1 applied
four times daily in subjects with acute blunt soft tissueinjuriescontusions of the limbs clinicaltrialsgovct2showNCT02290821 (accessed 3 February 2015) 2015 [CTGNCT02290821]
Additional references
Anon 2005
Anon Topical analgesics a review of reviews and a bit ofperspective wwwjr2oxacukBandolierExtraforbandoTopextra3pdf 2005
Cook 1995
Cook RJ Sackett DL The number needed to treat aclinically useful measure of treatment effect BMJ 1995310
(6977)452ndash4
Counsell 1994Counsell CE Clarke MJ Slattery J Sandercock PA Themiracle of DICE therapy for acute stroke fact or fictionalproduct of subgroup analysis BMJ 1994309(6970)1677ndash81
Dechartres 2013Dechartres A Trinquart L Boutron I Ravaud P Influenceof trial sample size on treatment effect estimates meta-epidemiological study BMJ 2013346f2304 [DOI101136bmjf2304]
Derry 2012a
Derry S Moore RA Topical capsaicin (low concentration)for chronic neuropathic pain in adults Cochrane Databaseof Systematic Reviews 2012 Issue 9 [DOI 10100214651858CD010111]
Derry 2012bDerry S Moore RA Rabbie R Topical NSAIDs forchronic musculoskeletal pain in adults Cochrane Databaseof Systematic Reviews 2012 Issue 9 [DOI 10100214651858CD007400pub2]
Derry 2013Derry S Sven-Rice A Cole P Tan T Moore RA Topicalcapsaicin (high concentration) for chronic neuropathic painin adults Cochrane Database of Systematic Reviews 2013Issue 2 [DOI 10100214651858CD007393pub3]
Derry 2014
Derry S Matthews PR Wiffen PJ Moore RASalicylate-containing rubefacients for acute and chronicmusculoskeletal pain in adults Cochrane Database ofSystematic Reviews 2014 Issue 11 [DOI 10100214651858CD007403pub3]
Evans 1995Evans JM McMahon AD McGilchrist MM White GMurray FE McDevitt DG et al Topical non-steroidal anti-inflammatory drugs and admission to hospital for uppergastrointestinal bleeding and perforation a record linkagecase-control study BMJ 1995311(6996)22ndash6
FitzGerald 2001
FitzGerald GA Patrono C The coxibs selective inhibitorsof cyclooxygenase-2 New England Journal of Medicine2001345(6)433ndash42
34Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Haroutiunian 2010
Haroutiunian S Drennan DA Lipman AG Topical NSAIDtherapy for musculoskeletal pain Pain Medicine 201011
(4)535ndash49 [DOI 101111j1526-4637201000809x]
Hawkey 1999
Hawkey CJ Cox-2 inhibitors Lancet 1999353(9149)307ndash14
Higgins 2011
Higgins JPT Altman DG Sterne JAC Chapter 8 Assessingrisk of bias in included studies In Higgins JPT GreenS (editors) Cochrane Handbook for Systematic Reviewsof Interventions Version 510 [updated March 2011]The Cochrane Collaboration 2011 Available fromwwwcochrane-handbookorg
Jadad 1996a
Jadad AR Carroll D Moore A McQuay H Developing adatabase of published reports of randomised clinical trials inpain research Pain 199666(2-3)239ndash46
Jadad 1996b
Jadad AR Moore RA Carroll D Jenkinson C ReynoldsDJ Gavaghan DJ et al Assessing the quality of reports ofrandomized clinical trials is blinding necessary ControlledClinical Trials 199617(1)1ndash12
LrsquoAbbeacute 1987
LrsquoAbbeacute KA Detsky AS OrsquoRourke K Meta-analysis inclinical research Annals of Internal Medicine 1987107(2)224ndash33
Makris 2010Makris UE Kohler MJ Fraenkel L Adverse effects oftopical nonsteroidal antiinflammatory drugs in older adultswith osteoarthritis a systematic literature review Journalof Rheumatology 201037(6)1236ndash43 [DOI 103899jrheum090935]
Mason 2004a
Mason L Moore RA Edwards JE Derry S McQuay HJTopical NSAIDs for acute pain a meta-analysis BMCFamily Practice 2004510
Mason 2004bMason L Moore RA Edwards JE Derry S McQuayHJ Topical NSAIDs for chronic musculoskeletal painsystematic review and meta-analysis BMC MusculoskeletalDisorders 2004528
McPherson 2013McPherson ML Cimino NM Topical NSAIDformulations Pain Medicine 201314 Suppl 1S35ndash9[DOI 101111pme12288]
MHRA 2009
Medicines and Healthcare Products Regulatory AgencyTopical ketoprofen reminder on risk of photo-sensitivityreactions Drug Safety Update 2009 June Vol 2 issue 116
Moore 1998aMoore RA Tramegraver MR Carroll D Wiffen PJ McQuayHJ Quantitative systematic review of topically applied non-
steroidal anti-inflammatory drugs BMJ 1998316(7128)333ndash8
Moore 1998bMoore RA Gavaghan D Tramegraver MR Collins SL McQuayHJ Size is everything - large amounts of information areneeded to overcome random effects in estimating directionand magnitude of treatment effects Pain 199878(3)209ndash16
Moore 2003
Moore RA Edwards J Barden J McQuay HJ BandolierrsquosLittle Book of Pain Oxford Oxford University Press 2003[ISBN 0ndash19ndash263247ndash7]
Moore 2006
Moore RA Barden J Derry S McQuay HJ Managingpotential publication bias In McQuay HJ Kalso EMoore RA editor(s) Systematic Reviews in Pain ResearchMethodology Refined Seattle IASP Press 200615ndash23[ISBN 978ndash0ndash931092ndash69ndash5]
Moore 2008aMoore RA Derry S McQuay HJ Topical agents in thetreatment of rheumatic pain Rheumatic Diseases Clinics ofNorth America 200834(2)415ndash32
Moore 2008bMoore RA Barden J Derry S McQuay HJ Managingpotential publication bias In McQuay HJ Kalso EMoore RA editor(s) Systematic Reviews in Pain ResearchMethodology Refined Seattle IASP Press 200815ndash24[ISBN 978ndash0ndash931092ndash69ndash5]
Moore 2013
Moore RA Straube S Aldington D Pain measures andcut-offs - rsquono worse than mild painrsquo as a simple universaloutcome Anaesthesia 201368(4)400ndash12 [DOI 101111anae12148]
Morris 1995
Morris JA Gardner MJ Calculating confidence intervalsfor relative risk odds ratios and standardised ratios andrates In Gardner MJ Altman DG editor(s) Statistics withConfidence - Confidence Intervals and Statistical GuidelinesLondon British Medical Journal 199550ndash63
NICE 2008National Institute for Health and Care ExcellenceOsteoarthritis The care and management of osteoarthritisin adults 2008 wwwniceorgukCG059fullguideline
Nuumlesch 2010Nuumlesch E Trelle S Reichenbach S Rutjes AW TschannenB Altman DG et al Small study effects in meta-analysesof osteoarthritis trials meta-epidemiological study BMJ2010341c3515 [DOI 101136bmjc3515]
PACT 2014
Prescribing and Primary Care team Health and SocialCare Information Centre Prescrption Cost AnalysisEngland 2013 Health and Social Care Information CentreThe NHS Information Centre 2014298ndash301 [ISBN978ndash1ndash78386ndash089ndash0]
35Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Quane 1998
Quane PA Graham GG Zeigler JB Pharmacology ofbenzydamine Inflammopharmacology 19986(2)95ndash107
RevMan 2014The Nordic Cochrane Centre The Cochrane CollaborationReview Manager (RevMan) 53 Copenhagen The NordicCochrane Centre The Cochrane Collaboration 2014
Roth 2012Roth SH Fuller P Pooled safety analysis of diclofenacsodium topical solution 15 (ww) in the treatment ofosteoarthritis in patients aged 75 years or older ClinicalInterventions in Aging 20127127ndash37 [DOI 102147CIAS30884]
Simon 2009
Simon LS Grierson LM Naseer Z Bookman AA ZevShainhouse J Efficacy and safety of topical diclofenaccontaining dimethyl sulfoxide (DMSO) compared withthose of topical placebo DMSO vehicle and oral diclofenacfor knee osteoarthritis Pain 2009143(3)238ndash45 [DOI101016jpain200903008]
Tramer 1997
Tramer MR Reynolds DJ Moore RA McQuay HJ Impact
of covert duplicate publication on meta-analysis a casestudy BMJ 1997315(7109)635ndash40
Tugwell 2004Tugwell PS Wells GA Shainhouse JZ Equivalence studyof a topical diclofenac solution (pennsaid) compared withoral diclofenac in symptomatic treatment of osteoarthritisof the knee a randomized controlled trial Journal ofRheumatology 200431(10)2002ndash12
Zimmerman 1995
Zimmerman J Siguencia J Tsvang E Upper gastrointestinalhemorrhage associated with cutaneous application ofdiclofenac gel American Journal of Gastroenterology 199590(11)2032ndash4 [PUBMED 7485017]
References to other published versions of this review
Massey 2010
Massey T Derry S Moore RA McQuay HJ TopicalNSAIDs for acute pain in adults Cochrane Databaseof Systematic Reviews 2010 Issue 6 [DOI 10100214651858CD007402pub2]
lowast Indicates the major publication for the study
36Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
C H A R A C T E R I S T I C S O F S T U D I E S
Characteristics of included studies [ordered by study ID]
Airaksinen 1993
Methods R DB PC parallel groupsGel applied to the painful area twice daily for 7 daysAssessment at baseline 3 7 days
Participants Minor soft tissue injuries (lt 7 days)N = 56M 45 F 11Age not reportedMean baseline pain at rest 25-26 mm
Interventions Ketoprofen gel 2 x 5 g (125 mg) daily n = 29Placebo gel n = 27Rescue medication paracetamol 500 mgNo other treatment allowed
Outcomes PGE 5-point scale but reported as ldquoimprovedrdquo or ldquosame or worserdquo (responder = ldquoim-provedrdquo)Improvement in pain with movement 100 mm VAS reported as group meanAdverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Unclear risk No mention of early withdrawals ormethod of imputation
Size High risk lt 50 participants per treatment group
37Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Aoki 1984
Methods R DB PC AC parallel groupsGel applied to affected area 3 or 4 times daily with no occlusion for 7 daysAssessment at baseline 3 7 days
Participants Acute orthopaedic trauma (contusion distortion fracture lt 7 days)N = 252 (203 analysed for efficacy)M 98 F 105Age range 8 to 86 years 13 younger than 20 yearsBaseline pain mild in 35Exclusions 23 protocol violations 26 reasons ldquonot relatedrdquo to drug Equally distributedbetween groups
Interventions Piroxicam gel 05 1 g 3 to 4 x daily n = 84Indomethacin gel 1 1 g 3 to 4 x daily n = 84Placebo gel n = 84No other medication or initiation of physical therapy allowed
Outcomes PGE 5-point scale (responder = ldquobetterrdquo and ldquomuch betterrdquo)Adverse eventsWithdrawals and exclusions
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Low risk ldquokey code sealed until end of studyrdquo
Blinding (performance bias and detectionbias)All outcomes
Low risk Gels in ldquoidentical tubesrdquo
Incomplete outcome data (attrition bias)All outcomes
Unclear risk gt 10 withdrawals ldquounrelated to treat-mentrdquo and for ldquoprotocol violationsrdquo Nofurther details but no significant differ-ences between groups
Size Unclear risk 50 to 200 participants per treatment group
38Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Auclair 1989
Methods R DB PC parallel groupsGel massaged into skin over affected heel 3 times daily after cleaning with soap and waterfor up to 21 daysAssessment at baseline 7 21 days
Participants Acute Achilles heel tendinitis (not associated with continuous pain at rest or gt 1 monthhistory)N = 243 (227 analysed for efficacy)MF not reportedMean age 29 yearsBaseline pain ~ 10 had lt 26 mm on palpation of tendon ~ 30 had mild or no painon dorsiflexion of footExclusions failure to meet inclusion criteria major protocol violations failure to takestudy medication for full duration
Interventions Niflumic acid gel 25 3 x 5 g daily n = 117Placebo gel n = 110No other analgesics and anti-inflammatories physiotherapy or supportive measures al-lowed
Outcomes PGE 5-point scale (responder = ldquogoodrdquo or ldquovery goodrdquo)Pain improved or disappeared on dorsiflexionAdverse eventsWithdrawals and exclusions
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Unclear risk 10 excluded for ldquofailing to meet entry cri-teria and protocol violationsrdquo No furtherdetails
Size Unclear risk 50 to 200 participants per treatment group
39Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Billigmann 1996
Methods R DB PC parallel groupsGel applied 3 times daily with rubbingAssessed at baseline 3 5 7 days
Participants Distortion of ankle jointN = 160M and FAge 18+ yearsBaseline pain not reported
Interventions Ibuprofen microgel 5 3 x 10 cm (= 200 mg) daily n = 80Placebo gel n = 80
Outcomes Pain with movement VAS (responder = decreased by 20)Complete remissionAdverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Unclear risk Insufficient data to assess
Size Unclear risk 50 to 200 participants per treatment group
Campbell 1994
Methods R DB PC parallel groupsCream applied 4 times daily for 7 days (up to 14 days optional)Self assessed using daily diary for 7 days and up to 14 days
Participants Acute ankle sprain (lt 24 hours no fracture)N = 100 (51 analysed)M 33 F 18Mean age 29 years
40Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Campbell 1994 (Continued)
Baseline pain at rest gt 35 mm on walking 80 mmExclusions did not return diaries protocol exclusions (25 ibuprofen 24 placebo)
Interventions Ibuprofen cream 5 (Proflex) 4 x 4rdquo (10 cm) daily n = 26Placebo cream n = 25Advised to use rest and regular icing for 48 hours then walking and exerciseRescue medication paracetamol
Outcomes Improvement in walking ability 4-point scale (responder = ldquoimprovementrdquo)Pain on walking 100 mm VAS (mean data)Withdrawals and exclusions
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Low risk Randomisation carried out by sponsorTubes dispensed by hospital pharmacy whoheld the codes
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical creamrdquo
Incomplete outcome data (attrition bias)All outcomes
High risk gt 40 lost to follow-up Approximatelyequal between groups
Size High risk lt 50 participants per treatment arm as anal-ysed
Chatterjee 1977
Methods R DB PC parallel groupsCream applied to site of injury 3 times daily for 6 daysAssessment at baseline 2 6 days
Participants Soft tissue injuries (recent)N = 51MF not reportedAge not reportedBaseline pain on passive movement moderate or severe in all but 3 participants
41Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Chatterjee 1977 (Continued)
Interventions Benzydamine HCl cream 3 3 x daily n = 25Placebo cream n = 25(5 active 6 placebo participants also received ultrasound)No other topical agent allowed
Outcomes Pain on passive movement 4-point scale (responder = ldquoabsentrdquo or ldquoslightrdquo)Tenderness with pressure 4-point scale (responder = ldquoabsentrdquo or ldquoslightrdquo)Adverse eventsWithdrawals and exclusions
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk ldquopredetermined randomised schedulerdquo
Allocation concealment (selection bias) Low risk Sealed copy of schedule held by investiga-tor and duplicate copy kept by clinical trialco-ordinator Looked at only in event ofadverse reaction (not necessary)
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoindistinguishable in appearance andconsistencyrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk 2 lost to follow-up no withdrawals dueto adverse events Responder analysis
Size Unclear risk lt 50 participants per treatment arm
Costantino 2011
Methods R DB PC multicentre parallel groupPlaster applied daily for 7 daysPI assessment daily overall treatment efficacy and tolerability assessed at 3 and 7 days
Participants Grade I or II ankle sprain (lt 48 hours) with lateral external ligament involvementPI on movement ge 50100 oedema ge 20 mm difference between anklesN = 430M 249 F 175 (for analysis)Mean age 35 yearsBaseline PI on movement 72100 (SD 12)
42Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Costantino 2011 (Continued)
Interventions DHEPhep n = 142DHEP n = 146Placebo n = 142Rescue medication paracetamol to maximum 3 g dailyNo other treatments allowed
Outcomes Mean change in PI on movement from baseline to 3 daysMean reduction in oedema at 3 daysTolerability at 3 daysRescue medicationAdverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk Computer-generated
Allocation concealment (selection bias) Unclear risk ldquosealed envelopesrdquo
Blinding (performance bias and detectionbias)All outcomes
Low risk Shape colour size and application methodidentical for all plasters
Incomplete outcome data (attrition bias)All outcomes
Unclear risk Very few withdrawals but used LOCF
Size Unclear risk 50 to 200 participants per treatment arm
Coudreuse 2010
Methods R DB PC multicentre parallel groupPlaster applied daily for 7 daysPI assessed twice daily over 3 days then day 7 Overall treatment efficacy and tolerabilityassessed on days 3 7
Participants Ankle sprain (lt 48 hours) with lateral external ligament involvementPI on movement ge 50100 oedema ge 20 mm difference between anklesN = 240 (233 for analysis)M 148 F 86 6 unknown
Interventions DHEPhep n = 120Placebo n = 120Rescue medication paracetamol to maximum 4 g daily
43Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Coudreuse 2010 (Continued)
No other treatments allowed
Outcomes Global efficacy at 7 days 4-point scale (responder = ldquoexcellentrdquo or ldquogoodrdquo)Mean change in PI on movement at 6 hours and 7 daysMean change in oedema at 3 and 7 daysTolerabilityAdverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk Computer-generated
Allocation concealment (selection bias) Unclear risk ldquosealed envelopesrdquo
Blinding (performance bias and detectionbias)All outcomes
Low risk Appearance and odour identical for all plas-ters
Incomplete outcome data (attrition bias)All outcomes
Low risk ITT analysis lt 5 excluded for missingdata equal between groups
Size Unclear risk 50 to 200 participants per treatment arm
Curioni 1985
Methods R DB PC AC parallel groupsGel rubbed into affected area until absorbed twice daily for 10 daysAssessed at baseline and daily to 10 days
Participants Acute soft tissue injuriesN = 60M 33 F 27Median age 33 yearsBaseline pain not given
Interventions Ibuproxam gel 10 n = 20Ketoprofen gel n = 20Etofenamate gel n = 20
Outcomes PGE 4-point scale (ldquogoodrdquo or ldquoexcellentrdquo)Resolution of symptomsAdverse events
44Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Curioni 1985 (Continued)
Withdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk Medication supplied in identical tubes
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
Diebshlag 1990
Methods R DB PC AC parallel groupsGel applied 3 times daily without occlusion for 14 daysAssessment at baseline 2 3 4 8 15 days
Participants Ankle sprain (lt 24 hours)N = 37M 24 F 13Mean age 28 yearsBaseline pain slight to moderate
Interventions Ketorolac gel 2 3 x 3 g daily n = 13Etofenamate gel 5 3 x 3 g daily n = 12Placebo gel n = 12Rescue medication paracetamolNo other analgesic or anti-inflammatory medication ice packs or physiotherapy allowed
Outcomes Reduction in PI 100 mm VAS and 4-point scale (responder = ldquoimprovedrdquo)Adverse eventsWithdrawals and exclusions
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
45Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Diebshlag 1990 (Continued)
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Low risk ldquoMedication assignment supplied in asealed enveloperdquo Opened only if seriousparticipant event necessitation treatmentdisclosure occurred (not necessary)
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical appearancerdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
Dreiser 1988
Methods R DB PC parallel groupsCream applied 3 times dailyAssessment at baseline 7 days
Participants Acute tendinitis (lt 1 month)N = 64M 35 F 25Mean age 36 yearsBaseline spontaneous pain ge 60 mm
Interventions Ibuprofen cream 5 3 x 4 cm daily n = 32 (3 x 10 cm for large joints)Placebo cream n = 32No other topical systemic or physical treatment allowed
Outcomes PGE scale not reported (responder = ldquoimprovementrdquo or ldquocomplete relief rdquo)Improvement in pain VAS (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35Oxford Validity Score 1016
Risk of bias
Bias Authorsrsquo judgement Support for judgement
46Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Dreiser 1988 (Continued)
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Low risk Missing participants added back usingBOCF
Size High risk lt 50 participants per treatment arm
Dreiser 1989
Methods R DB PC parallel groupsGel applied twice daily to affected area with light massage then covered with standardcompressAssessed at baseline 3 7 days
Participants Uncomplicated recent ankle sprainN = 60M 36 F 24Mean age 33 yearsMean baseline pain 54 mm
Interventions Ketoprofen gel 25 2 x 5 cm daily n = 30Placebo gel n = 30No concomitant therapy other than simple oral analgesia allowed
Outcomes PGE 3-point scale (responder = ldquobetterrdquo)Improvement in pain VAS (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk ldquodrawing lotsrdquo
Allocation concealment (selection bias) Unclear risk Not described
47Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Dreiser 1989 (Continued)
Blinding (performance bias and detectionbias)All outcomes
Low risk Treatments ldquoidentical in every way exceptthat placebo did not contain active princi-plerdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participant per treatment arm
Dreiser 1990
Methods R DB PC parallel groupsGel lightly massaged into skin over affected area 3 times daily then covered with standardcompressAssessed at baseline 3 7 days
Participants Uncomplicated ankle sprain (lt 4 days)N = 60 (59 analysed)M 29 F 29 (not stated for 1 participant)Mean age 33 yearsBaseline pain ge moderately severeExclusions 1 participant had only moderate pain at baseline
Interventions Niflumic acid gel 25 3 x 5 g daily n = 30Placebo gel n = 30Concomitant treatment with systemic NSAIDs local therapies or physiotherapy werenot allowed
Outcomes PGE 4-point scale (responder = ldquocuredrdquo or ldquoimprovedrdquo)Improvement in pain VAS (mean data)Adverse eventsWithdrawals and exclusions
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
48Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Dreiser 1990 (Continued)
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
Dreiser 1994
Methods R DB PC parallel groupsPatch applied twice dailyAssessed at baseline 3 7 days
Participants Traumatic ankle sprain (lt 2 days)N = 131M 84 F 47Mean age 34 yearsBaseline pain ge 50 mm
Interventions Flurbiprofen patch 2 x 40 mg daily n = 65Placebo patch n = 66Rescue medication paracetamol Ice or light restraint allowedExclusions 1 from flurbiprofen group for protocol violation
Outcomes PGE 4-point scale (responder = ldquogoodrdquo or ldquovery goodrdquo)Improvement in pain VAS (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk Placebo patch was ldquonon-medicated (butotherwise identical)rdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
49Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Fioravanti 1999
Methods R DB AC parallel groupsGel lightly massaged into skin 3 times daily and kept dry for 6 to 8 hoursAssessed at baseline 3 10 days
Participants Peri and extra-articular inflammatory diseasesN = 100M 32 F 68Mean age 49 yearsBaseline spontaneous pain ge 40 mm
Interventions DHEP lecithin gel 3 x 5 g (= 65 mg) daily n = 50DHEP gel 3 x 5 g (= 65 mg) daily n = 50
Outcomes PGE 4-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo)Pain on movement meanAdverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size Unclear risk 50 participants per treatment arm
Fujimaki 1985
Methods R DB PC AC parallel groupsGel applied to affected area 3 or 4 times daily with no occlusion for up to 14 daysAssessed at baseline 7 14 days
Participants Muscle pain or inflammation in neck shoulder back chest and upper and lower ex-tremities or a combinationN = 271 (247 analysed)M 97 F 149
50Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Fujimaki 1985 (Continued)
Age lt 20 to 89 yearsBaseline pain mostly mild to moderateExclusions 24 due to protocol violations loss to follow-up
Interventions Piroxicam gel 05 1 g 3 to 4 x daily n = 92Indomethacin gel 1 1 g 3 to 4 x daily n = 90Placebo gel n = 89No concomitant oral or topical analgesic or anti-inflammatory medication allowed Nophysical therapy initiated after start of study
Outcomes PGE 5-point scale (responder = ldquobetterrdquo or ldquomuch betterrdquo)Physician rated improvement 5-point scale (responder = ldquomarked improvementrdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total =45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Low risk Cartons numbered randomly and numbersheld in a key code until study completion
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical tubesrdquo packed in numbered car-ton Gel bases slightly different in appear-ance so dispensing physician did not haveaccess to them
Incomplete outcome data (attrition bias)All outcomes
Unclear risk ~ 8 excluded from analysis for unknownreasons or lost to or inadequate follow-upApproximately equal between groups
Size Unclear risk 50 to 200 participants per treatment arm
Gallacchi 1990
Methods R DB AC parallel groupsGel applied to affected area 4 times daily with light massage for 14 daysAssessment at baseline 7 14 days
Participants Painful inflammatory conditionsN = 50M 20 F 30Mean age 50 years
51Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Gallacchi 1990 (Continued)
Baseline pain ge moderate severity
Interventions Diclofenac gel 1 2 g 4 x daily n = 25 (Flector)Diclofenac sodium 1 2 g 4 x daily n = 25 (Voltaren Emugel)No other medication that could interfere with test drugs allowed
Outcomes PGE 5-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo)Improvement in pain on pressure 4-point scale (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
Gonzaacutelez de Vega 2013
Methods R DB (for gels) SB (for ointment) AC multicentre parallel groups2 g (~ 6 cm of gel) applied over injured area x 3 daily for 14 daysAssessments at 0 4 7 14 days of treatment and day 42 follow-up
Participants Grade I or II ankle sprain (lt 24 hours) with lateral external ligament involvement PIon weight bearing ge 30100N = 449M = 308 F = 112 (for analysis)Mean age 28 years
Interventions Traumeel gel 3 x 2 g daily n = 140Traumeel ointment n = 143 (not analysed in this review)Diclofenac gel 1 3 x 2 g daily n = 137Rescue medication paracetamol to maximum 2 g daily
52Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Gonzaacutelez de Vega 2013 (Continued)
Outcomes Pain-free on day 7Global efficacy 5-point scale (responder = ldquogoodrdquo and ldquovery goodrdquo) on day 14Normal function on day 14 yes or no (responder = ldquoyesrdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 (gels) W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk Computer-generated
Allocation concealment (selection bias) Low risk Central allocation kits assigned in orderreceived and used envelopes (no furtherdetails)
Blinding (performance bias and detectionbias)All outcomes
Low risk For gel comparison ldquoidentical containersrdquo
Incomplete outcome data (attrition bias)All outcomes
Unclear risk Used LOCF Most withdrawals due to earlyrecovery (within 14 days) approximatelyequal between groups
Size Unclear risk 50 to 200 participants per treatment arm
Governali 1995
Methods R DB AC parallel groupsGel or cream applied 3 times daily for up to 14 daysAssessed at baseline 7 14 days
Participants Soft tissue injuries + 2 fracturesN = 30M = 21 F = 9Median age 38 yearsMean baseline pain on movement moderate to severe (28 scale 0 to 4)
Interventions Ketoprofen gel 5 3 x 2 to 3 g daily n = 15Ketoprofen cream 1 3 x 2 to 3 g daily n = 15
Outcomes PGE 5-point scale (responder = ldquogoodrdquo and ldquoexcellentrdquo)Adverse eventsWithdrawals
53Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Governali 1995 (Continued)
Notes Oxford Quality Score R1 DB1 W1 Total =35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Treatments were given in identical tubesand measurements made by blinded ob-servers but one was a cream and the othera gel
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
Gualdi 1987
Methods R DB AC parallel groupsGel applied twice daily for 10 daysAssessed at baseline 4 7 10 days
Participants Soft tissue injuriesN = 60M = 37 F = 23Mean age 32 years (range 13 to 78)Mean baseline pain on movement moderate to severe (22 scale 0 to 3)
Interventions Flunoxaprofen gel 2 x 3 to 5 cm daily n = 30Ketoprofen gel 2 x 3 to 5 cm daily n = 30
Outcomes Improvement in pain on pressure (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
54Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Gualdi 1987 (Continued)
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Unclear risk Insufficient data
Size High risk lt 50 participants per treatment arm
Haig 1986
Methods R DB PC parallel groupsCream applied lightly to affected area 6 times daily for 6 daysAssessed at baseline 2 4 6 days
Participants Soft tissue injuries (lt 24 hours)N = 43MF not reportedAge not reportedBaseline pain not reported
Interventions Benzydamine cream 3 6 x daily n = 21Placebo cream n = 22
Outcomes Pain on movement 4-point scale (responder = ldquoimprovedrdquo)Adverse events
Notes Oxford Quality Score R1 DB2 W0 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not reported
Allocation concealment (selection bias) Unclear risk Not reported
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquomatching placebordquo
55Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Haig 1986 (Continued)
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants apparently included inanalysis
Size High risk lt 50 participants per treatment arm
Hoffmann 2012
Methods R DB PC multicentre parallel groupPlaster applied x 1 daily (minimum of 20 consecutive hours of application per day) for14 daysPI assessed daily overall efficacy assessed at 7 and 14 days
Participants Unilateral mild to moderate muscle contusion of upper or lower limb (lt 72 hours) PIge 50100 superficial haematoma le 10 x 14 cm at injured siteN = 354M 126 F 228Mean age 39 years
Interventions DHEPhep plaster x 1 daily n = 121DHEP plaster x 1 daily n = 115Placebo plaster n = 118Rescue medication paracetamol 500 mg (no limit reported)
Outcomes PGE 5-point scale (responder = excellent or good) at 14 daysMean reduction in PI at 3 and 8 daysRescue medicationAdverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Method not reported
Allocation concealment (selection bias) Unclear risk Method not reported ldquoMedication packedand labelled in order to render all partici-pants and personnel fully blinded to treat-ment administeredrdquo
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoIndistinguishable regarding appearanceshape colour size and odourrdquo
56Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Hoffmann 2012 (Continued)
Incomplete outcome data (attrition bias)All outcomes
Unclear risk No responder analysis reported and impu-tation method not mentioned
Size Unclear risk 50 to 200 participants per treatment arm
Hofman 2000
Methods R DB AC parallel groupsGel applied to affected region 4 times daily with gentle massageAssessed at baseline 8 days in clinic and daily diary
Participants Soft tissue articular pain (le 15 days)N = 142M 19 F 123Mean age 57 yearsMean baseline PI moderate to severe
Interventions Diclofenac sodium gel 1 4 x 2 cm daily n = 69Lysine clonixinate gel 5 4 x 2 cm daily n = 73(2 cm = 225 mg)No other analgesic local treatment (including immobilisation bandaging) or acupunc-tureRescue mediation allowed after 2 applications if needed
Outcomes PGE 3-point scale (ldquogoodrdquo)PI participant diary (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk Diclofenac gel repackaged to maintain DBwith lysine clonixinate gel Minor differ-ences between gels only apparent when di-rectly compared
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
57Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Hofman 2000 (Continued)
Size Unclear risk 50 to 200 participants per treatment arm
Hosie 1993
Methods R DB AC parallel groupsFoam (approximately the size of a golf ball) applied and 1 tablet taken 3 times daily for7 days and up to 14 daysAssessed at baseline 7 14 (if necessary) days
Participants Acute lower back injury (lt 1 month)N = 287 (261 analysed for efficacy)M 151 F 136Mean age 37 years (range 18 to 63)Most participants had moderate to severe pain on movement one had noneExclusions 25 lost to follow up one assessed at 14 days but not 7 days
Interventions Felbinac foam 3 3 x 2 g daily + placebo tablets 3 x 1 daily n = 140 (127 analysed forefficacy)Ibuprofen tablets 3 x 400 mg daily + placebo foam 3 x 2 g daily n = 147 (134 analysedfor efficacy but one had no pain at baseline)No other oral injectable or topical analgesic or anti-inflammatory medication Ongoingphysiotherapy to continue without change
Outcomes Pain on movement 5-point scale (responder = ldquononerdquo or ldquomildrdquo)Spontaneous pain 5-point scale (responder = ldquononerdquo or ldquomildrdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquodouble dummyrdquo
Incomplete outcome data (attrition bias)All outcomes
Unclear risk Imputation method not mentioned
Size Unclear risk 50 to 200 participants per treatment arm
58Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Jenoure 1997
Methods R DB PC parallel groupsPlaster applied to skin over affected area twice daily and kept in place with an elasticbandageAssessed at baseline 7 14 days and after further 14 days without treatment
Participants Humero-radial epicondyl pain (tendinopathic) - nearly all tennis elbowN = 85M 54 F 31Mean age 45 yearsBaseline pain ldquomildrdquo in ~ 10 of placebo group and 29 of active group
Interventions DHEP plaster (Tissugel) x 2 daily n = 44Placebo plaster x 2 daily n = 41
Outcomes Pain on pressure 5-point scale (responder = ldquononerdquo or ldquomildrdquo)Spontaneous pain 5-point scale (responder = ldquono painrdquo)Adverse events
Notes Oxford Quality Score R1 DB2 W0 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not reported
Allocation concealment (selection bias) Unclear risk Not reported
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical characteristicsrdquo
Incomplete outcome data (attrition bias)All outcomes
Unclear risk No useable data
Size High risk lt 50 participants per treatment arm
Joussellin 2003
Methods R DB PC parallel groupsPlaster applied to skin over affected area once dailyAssessed at baseline 1 2 3 7 days
Participants Ankle sprain (lt 48 hours)N = 134M 72 F 62Age range 18 to 65 years
59Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Joussellin 2003 (Continued)
Baseline spontaneous pain ge 50 mm
Interventions DHEP plaster (Flector Tissugel 1) x 1 daily n = 68Placebo plaster x 1 daily n = 66Rescue medication paracetamol
Outcomes PGE 4-point scale (responder = ldquoexcellentrdquo)Pain on movement VAS (mean)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoIdentical in size appearance and usedsame formula as active patch without ac-tive ingredientrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in responder anal-ysis
Size Unclear risk 50 to 200 participants per treatment arm
Julien 1989
Methods R DB PC parallel groupsGel applied to affected area twice daily with light massageAssessed at baseline 3 7 days in clinic and daily diary
Participants TendinitisN = 60M 29 F 31Mean age 41 yearsBaseline pain gt 50 mm
Interventions Ketoprofen gel 25 2 x 5 cm (= 50 mg) daily n = 30Placebo gel 2 x 5 cm daily n = 30No concomitant therapy other than simple analgesia
60Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Julien 1989 (Continued)
Outcomes PGE 4-point scale (responder = ldquoimprovedrdquo or ldquorecoveredrdquo)Pain on movement 4-point scale (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB1 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk Random numbers table
Allocation concealment (selection bias) Unclear risk Randomisation code supplied by Menarinilaboratories remote from allocation
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in responder anal-ysis
Size High risk lt 50 participants per treatment arm
Klainguti 2010
Methods R DB PC multicentre parallel groupPlaster applied x 1 daily (minimum 20 consecutive hours of application daily) for 10daysPI assessed daily Overall treatment efficacy and tolerability assessed at days 3 and 7
Participants Unilateral mild to moderate muscle contusion or strain of upper or lower limb (lt 72hours) superficial haematoma le 140 cm2 PI ge 40100N = 185M 90 F 95Mean age 39 years
Interventions DHEPhep plaster x 1 daily n = 65DHEP plaster x 1 daily n = 61Placebo plaster x 1 daily n = 59
Outcomes Overall treatment efficacy at 3 days 5-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo)Resolution of haematoma at 10 days yes or noRescue medicationAdverse eventsWithdrawals
61Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Klainguti 2010 (Continued)
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk Computer-generated
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoIdentical in size appearance and odourrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk Withdrawals lt 5 All participants in-cluded in analysis
Size Unclear risk 50 to 200 participants per treatment arm
Kockelbergh 1985
Methods R DB PC parallel groupsGel applied twice dailyAssessed at baseline 3 7 days
Participants Acute soft tissue trauma (lt 24 hours)N = 74M 60 F 14Mean age 27 yearsBaseline pain gt 65 mm
Interventions Ketoprofen gel 25 2 x 5 cm (= 15 mg) daily n = 38Placebo gel 2 x 5 cm daily n = 36No concomitant treatmentRescue medication glafenine
Outcomes PGE 3-point scale (responder = ldquogoodrdquo)Spontaneous pain 100 mm VAS (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
62Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Kockelbergh 1985 (Continued)
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in responder anal-ysis
Size High risk lt 50 participants per treatment arm
Kuehl 2011
Methods R DB PC multicentre parallel groupsPlaster applied x 2 daily for 14 days or until resolutionPI assessed daily Overall treatment efficacy assessed at 14 days
Participants Mild or moderate sprain strain or contusion (lt 7 days) PI ge 510N = 418M 206 F 212Mean age 39 years
Interventions DHEP patch x 2 daily n = 207Placebo patch x 2 daily n = 211Rescue medication no analgesic allowed (or icewrapping) use = discontinuation
Outcomes Resolution of injury (VAS le 210)Overall tolerability 5-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk Computer-generated
Allocation concealment (selection bias) Unclear risk Not described
63Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Kuehl 2011 (Continued)
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical in appearance with same contentexcept diclofenacrdquo
Incomplete outcome data (attrition bias)All outcomes
High risk Used LOCF high withdrawal rate (43)Did not report all efficacy outcomes mea-sured
Size Low risk gt 200 participants per treatment arm
Li 2013
Methods R DB PC multicentre parallel groupsPlaster applied x 2 daily for 7 daysPI assessed daily overall treatment efficacy and tolerability assessed at 7 days
Participants Mild or moderate ankle or knee sprain muscle strain or contusion (lt 72 hours) PI ge
50100N = 384M 144 F 240Mean age 42 years
Interventions DHEP plaster x 2 daily n = 192Placebo plaster x 2 daily n = 192Rescue medication paracetamol to maximum 2 g daily
Outcomes ge 50 reduction in PI at 7 daysOverall treatment efficacy 5-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo)Overall tolerability 5-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo)Rescue medicationAdverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk Computer-generated
Allocation concealment (selection bias) Unclear risk Method not fully described ldquosealed en-velopesrdquo ldquoconcealed from investigators andparticipantsrdquo
64Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Li 2013 (Continued)
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical in texture size color and odorrdquo
Incomplete outcome data (attrition bias)All outcomes
Unclear risk Withdrawals 2 Methods states LOCFbut appears to describe BOCF for with-drawals
Size Unclear risk 50 to 200 participants per treatment arm
Linde 1985
Methods R DB PC parallel groupsCream applied x 3 daily for 5 days with elastic support for the first 3 daysAssessed at baseline 4 8 days
Participants Sprained ankle (lt 24 hours)N = 100M 58 F 42Mean age 28 yearsBaseline pain all participants had ldquowalking painrdquo
Interventions Benzydamine 3 cream x 3 daily n = 50Placebo gel x 3 daily n = 50
Outcomes Pain on movement responder = ldquofree of walking painrdquoAdverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35Paper describes a benzydamine cream and a placebo gel
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described Paper describes a benzy-damine cream and a placebo gel
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in responder anal-ysis
65Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Linde 1985 (Continued)
Size Unclear risk 50 participants per treatment arm
Machen 2002
Methods R DB PC parallel groupsGel gently (ldquominimal rubrdquo not vigorously) massaged into skin over affected site untilabsorbed 3 times daily until symptoms disappeared or for maximum of 7 daysAssessment at baseline and once daily using diary cards to 7 days
Participants Soft tissue injury (lt 2 weeks and untreated)N = 85 (81 analysed)M 42 F 39Mean age 41 yearsBaseline pain gt 50 mm4 placebo participants lost to follow-up
Interventions Ibuprofen gel 5 x 3 daily n = 40Placebo gel x 3 daily n = 41Initiation of other medication or physiotherapy not allowed during study
Outcomes PGE 5-point scale (responder = ldquomarked improvementrdquo or ldquocomplete clearancerdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk Gels had similar physical characteristicsand were supplied in identical tubes
Incomplete outcome data (attrition bias)All outcomes
Low risk lt 5 missing participants Remaining par-ticipants included in responder analysis
Size High risk lt 50 participants per treatment arm
66Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Mahler 2003
Methods R DB AC parallel groupsGel applied with gentle massage to affected area 3 times daily without occlusion for 10daysAssessed at baseline 3 10 days in clinic and daily patient diary
Participants First-degree ankle or knee sprains first-degree muscle strains and mild-to-moderatecontusionsn = 100M 69 F 31Mean age 32 yearsMean baseline pain with activity ge 65 mm
Interventions DHEP lethicin gel 3 x 5 g (= 65 mg) daily n = 52DHEP gel 3 x 5 g (= 65 mg) daily n = 48All participants treated with ice at site of inflammation for first 48 hours but no immo-bilisation allowedRescue medication paracetamol 500 mg if strictly necessary
Outcomes PGE 4-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo)Pain on movement 100-mm VAS (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 5
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk ldquoComputer-generated randomization listrdquo
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk Pharmaceutically inert colouring agentsadded to reference formulation so that gelswere indistinguishable
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in responder anal-ysis
Size High risk lt 50 participants in each treatment arm
67Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Maziegraveres 2005a
Methods R DB PC parallel groupsNew patch applied directly to skin over painful area each morningAssessed at baseline 3 7 14 days
Participants Symptomatic tendonitis in upper or lower limbs not requiring surgery (le 15 days)N = 172M 72 F 100Mean age 46 yearsBaseline pain with activity ge 40 mm
Interventions Ketoprofen patch 100 mg x 1 daily n = 87Placebo patch x 1 daily n = 85No analgesic or steroid by any route or other topical medication or physical therapyallowedRescue medication permitted but not within 12 hours of assessment
Outcomes PGE 4-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk ldquocomputer generated global randomizationcoderdquo
Allocation concealment (selection bias) Low risk ldquoThe randomization list and code en-velopes were prepared by the company ap-pointed for clinical supplies packaging Therandom code was disclosed only after studycompletion and database closurerdquo
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquothe same indistinguishable patch with noingredientrdquo
Incomplete outcome data (attrition bias)All outcomes
High risk LOCF withdrawals 12 All participantsincluded in analysis
Size Unclear risk 50 to 200 participants per treatment arm
68Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Maziegraveres 2005b
Methods R DB PC parallel groupsNew patch applied directly to skin over painful area each morningAssessed at baseline 3 7 14 days
Participants Painful benign ankle sprain (le 48 hours)N = 163M 83 F 80Mean age 37 yearsBaseline spontaneous pain ge 50 mm
Interventions Ketoprofen patch 100 mg x 1 daily n = 81Placebo patch x 1 daily n = 82No analgesic or steroid by any route or other topical medication or physical therapyallowedRescue medication permitted but not within 12 hours of assessment
Outcomes PGE 4-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk ldquocomputer-generated global randomiza-tion coderdquo
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk The same transdermal patch with no activeingredient
Incomplete outcome data (attrition bias)All outcomes
Unclear risk LOCF withdrawals 5 to 6 All partic-ipants included in analysis
Size Unclear risk 50 to 200 participants per treatment arm
McLatchie 1989
Methods R DB PC parallel groupsGel applied to injured site 3 times daily for 7 daysAssessment at baseline 4 7 days at clinic daily patient diary
Participants Acute soft tissue injury (lt 48 hours)N = 231
69Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
McLatchie 1989 (Continued)
M 143 F 88Mean age 33 yearsBaseline pain moderate to severe
Interventions Felbinac gel 3 3 x 3 cm daily n = 118Placebo gel 3 x 3 cm daily n = 113Rescue medication paracetamol
Outcomes Patient diary mean changeAdverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquotubes identical in all aspectsrdquo
Incomplete outcome data (attrition bias)All outcomes
Unclear risk No usable efficacy data
Size Unclear risk 50 to 200 participants per treatment arm
Morris 1991
Methods R DB PC parallel groupsGel applied to site of injury 3 times daily for 7 daysAssessed at baseline 7 days at clinic and daily patient diary
Participants Acute soft tissue injury (lt 3 days)N = 100 (84 analysed for efficacy)M 70 F 14Mean age 25 yearsBaseline pain moderate to severeExclusions 1 participant in placebo group lost to follow-up 15 protocol violations
Interventions Felbinac gel 3 3 x 1 cm daily n = 41Placebo gel n = 43Ice joint immobilisation bandaging and compression allowed
70Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Morris 1991 (Continued)
No concomitant oral NSAID occlusive dressing physiotherapy or liniments allowedRescue medication paracetamol
Outcomes PGE 5-point scale (responder = ldquogoodrdquo and ldquovery goodrdquo)Change in PI patient diary 10 cm VAS (mean data)Adverse eventsWithdrawals and exclusions
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Low risk ldquoRandomisation was undertaken at theproduction facility and a sealed copy of thelist supplied to the investigator for refer-ence only in defined circumstancesrdquo
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical tubes and outer boxesrdquo ldquoplacebowas a similarly constituted gelrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All relevant participants included in analy-sis
Size High risk lt 50 participants per treatment arm
NCT01255423
Methods R DB PC parallel group studyGel applied 4 x daily
Participants Acute lateral ankle sprain Grade I to II le 12 hoursN = 206Mean (plusmn SD) age 31 years (plusmn 13)M 87 F 119Baseline PI not reported
Interventions Diclofenac sodium gel 1 x 4 daily n = 104Placebo gel n = 102
Outcomes Mean PI on movementAdverse events
71Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT01255423 (Continued)
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Method of blinding not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants accounted for
Size Unclear risk 50 to 200 participants per treatment arm
NCT01272934
Methods R DB PC parallel group studyGel applied 4 x daily
Participants Acute lateral ankle sprain Grade I to II le 12 hoursN = 205M 101 F 104Mean (plusmn SD) age 32 years (plusmn 11)
Baseline PI not reported
Interventions Diclofenac sodium gel 1 x 4 daily n = 102Placebo gel x 4 daily n = 103
Outcomes Mean PI on movementAdverse events
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
72Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT01272934 (Continued)
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Method of blinding not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants accounted for
Size Unclear risk 50 to 200 participants per treatment arm
NCT01272947
Methods R DB PC parallel groupsGel applied 4 x daily
Participants Acute blunt soft tissue injuriescontusions of the limbs lt 3 hoursN = 204M 101 F 103Mean (plusmn SD) age 30 years (plusmn 11)
Baseline PI not reported
Interventions Diclofenac sodium gel 1 x 4 daily n = 104Placebo gel x 4 daily n = 100
Outcomes Mean PI on movementAdverse events
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Method of blinding not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants accounted for
73Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT01272947 (Continued)
Size Unclear risk 50 to 200 participants per treatment arm
Noret 1987
Methods R DB PC parallel groupsGel applied twice daily for 7 daysAssessment at baseline 3 8 days
Participants Minor sports injuries (lt 24 hours)N = 98 (93 analysed)M 71 F 27Mean age 29 yearsBaseline pain gt 60 mm
Interventions Ketoprofen gel 25 2 x 5 cm daily (= 15 mg) n = 48Placebo gel n = 45No other treatment given
Outcomes PGE 4-point scale (responder = ldquogoodrdquo and ldquoexcellentrdquo)Spontaneous pain 100 mm VAS (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Low risk ldquoallocated according to a randomization listand a corresponding code in a sealed enve-loperdquo
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
74Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Parrini 1992
Methods R DB PC parallel groupsFoam (the size of a walnut or a 1-second spray) applied with massage 3 x daily for 7days
Participants Articular trauma strains distortionsN = 169M 94 F 75Mean age 37 yearsMean baseline pain on movement 31 (scale 1 to 4)
Interventions Ketoprofen foam 15 3 x 2 g (= 600 mg) daily n = 83Placebo foam 3 x 2 g n = 86
Outcomes Pain on movement 4-point scale (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB1 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk ldquopatients were randomised according to themethod of random numbersrdquo [translated]
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Unclear risk No usable efficacy data
Size Unclear risk 50 to 200 participants per treatment arm
Picchio 1981
Methods R DB AC parallel groupsCream applied with slight massage until completely absorbed 3 times daily for up to 16daysAssessed at baseline 4 8 12 16 days
Participants Acute sports injuriesN = 40M 24 F 16Mean age 22 years (range 12 to 46)Most participants had mild to moderate baseline pain (12 and 9 with slight pain on
75Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Picchio 1981 (Continued)
movement)
Interventions Ibuprofen gel 10 3 x daily n = 20Ketoprofen gel 1 3 x daily n = 20
Outcomes Pain on movement (responder = ldquononerdquo)Adverse events
Notes Oxford Quality Score R1 DB2 W0 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk ldquotubes were identical in appearancerdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
Predel 2004
Methods R DB PC parallel groupsNew patch applied to injured area twice daily for 7 days Contact of patch with humidityor water to be avoidedAssessment at baseline 3 7 days
Participants Traumatic blunt soft tissue injuries (lt 3 hours no treatment)N = 120M 73 F 47Mean age 32 yearsBaseline pain gt 60 mm
Interventions Diclofenac sodium patch 2 x daily (140 mg per patch) n = 60Placebo patch 2 x daily n = 60NSAIDs analgesics psychotropic agents other topical preparations and bandages notallowed
Outcomes PGE 4-point scale (responder = rdquogoodldquo and rdquoexcellentldquo)Pain on movement 10 cm VAS (mean data)
76Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Predel 2004 (Continued)
Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk rdquocomputer generated block randomisationlist
Allocation concealment (selection bias) Low risk An independent statistician produced ran-domisation list and an independent con-tract research organisation packaged med-ication according to list Nobody else hadaccess to the randomisation list until thedatabase was closed
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoThe placebo patch was visually indistin-guishable from the active patchrdquo To avoidunblinding due to different smell anystudy nurse involved with medication wasnot involved in outcome assessment
Incomplete outcome data (attrition bias)All outcomes
Low risk Withdrawals lt 1 All participants in-cluded in analysis
Size Unclear risk 50 to 200 participants per treatment arm
Predel 2012
Methods R DB PC parallel groupsTreatment for 7 daysAssessed at baseline 1 3 5 and 8 days
Participants Grade I or II ankle sprain with lateral external ligament involvement lt 12 hoursPI on movement ge 50100N = 242M 152 F 90Mean age 32 years
Mean baseline pain on movement = 75100
Interventions Diclofenac gel (Voltaren Emulgel 232) 2 x 5 cm + 1 x 5 cm placebo gel daily n = 80Diclofenac gel (Voltaren Emulgel 232) 3 x 5 cm daily n = 80Placebo gel 3 x 5 cm daily n = 82Rescue medication paracetamol (to maximum 2 g daily)
77Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Predel 2012 (Continued)
No ice or bandages after randomisation
Outcomes ge 50 reduction in PI on movement at 5 daysPGE 5-point scale (responder = ldquogoodrdquo or rsquovery goodldquo)Patient satisfaction 5-point scale (responder = rdquogoodldquo or rsquovery goodrdquo or ldquoexcellentrdquo)Rescue medicationAdverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical in composition appearance tex-ture and smellrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk LOCF but withdrawals lt 3 in all treat-ment arms
Size Unclear risk 50 to 200 participants per treatment arm
Predel 2013a
Methods R DB PC parallel groupsTreatment for 14 daysAssessed at baseline 3 7 10 and 14 days (plusmn 1 day)
Participants Uncomplicated one-sided ankle sprain with swelling ge 12 mm (2 to 18 hours)PI on movement tenderness joint mobility (scale 0 to 3) summed as ge 5 and le 7N = 232M 126 F 106Mean age 29 years
Interventions Diclofenac 4 spray gel 3 x 4 to 5 sprays daily (96 to 120 mg diclofenac sodium) n =118Placebo spray gel 3 x 4 to 5 sprays daily n = 114Rescue medication paracetamol 500 mg (maximum 10 tablets per week)No ice or bandaging allowed
78Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Predel 2013a (Continued)
Outcomes None or slight PI on movement at 3 and 7 daysPGE 4-point scale (responder = ldquovery goodrdquo at 14 days)gt 50 reduction in swelling at 10 daysAdverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk ldquoplacebo (vehicle only no active ingredi-ent)rdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk LOCF for full or early cure Withdrawalslt 2
Size Unclear risk 50 to 200 participants per treatment arm
Predel 2013b
Methods R DB PC parallel groups2 g gel applied with fingertips over affected area and massaged in for 1 minute Treatmentfor 5 daysAssessed at baseline 2 3 5 days
Participants Uncomplicated neck pain originating from cervical joints and accompanying soft tissues(ge 12 hours but lt 3 months) PI ge 50100N = 72M 33 F 39Mean age 34 years
Interventions Diclofenac gel (Voltaren Emulgel) 4 x 2 g daily n = 36Placebo gel 4 x 2 g daily n = 36Rescue medication paracetamol up to 2 g dailyNo concomitant therapies allowed
Outcomes ge 50 decrease in PI on movement after 48 hoursPGE 5-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo at 5 daysAdverse eventsWithdrawals
79Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Predel 2013b (Continued)
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk Sponsor ldquoproduced randomisation listrdquo
Allocation concealment (selection bias) Low risk Automated remote system
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical in packaging labelling scheduleof administration appearance and odourrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk No missing data
Size High risk lt 50 participants per treatment arm
Ramesh 1983
Methods R DB PC parallel groupsCream applied to painful area and rubbed into skin over a large area for up to 10 daysAssessment at baseline 3 7 10 days
Participants Strains sprains contusions compressionsN = 80M 42 F 38Age 11 to 81 yearsBaseline pain 5 ibuprofen 2 placebo participants had no or slight pain
Interventions Ibuprofen cream 5 3 to 4 x 5 to 10 cm daily n = 40Placebo cream 3 to 4 x 5 to 10 cm daily n = 40Adjuvant therapy was not administered
Outcomes Pain on movement 4-point scale (responder = ldquononerdquo or ldquoslightrdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
80Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Ramesh 1983 (Continued)
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Low risk Randomisation key in sealed envelopeavailable for emergencies but opened onlyafter completion
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical appearance and odourrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
Rowbotham 2003
Methods R DB PC parallel groupsNew patches applied to the affected painful area for 12 consecutive hours twice dailyfor up to 14 daysAssessed at baseline 14 days in clinic and daily patient diary
Participants Minor sports injuries (sprains sprains contusions lt 72 hours)N = 372M 253 F 119Mean age 33 yearsBaseline pain at rest ge 510
Interventions DHEP patch (Flector Tissuegel) 2 x daily (equivalent to diclofenac sodium 140 mg perpatch) n = 191Placebo patch 2 x daily n = 181
Outcomes PGE 5-point scale (responder = ldquogoodrdquo and ldquoexcellentrdquo)Pain resolved lt moderate for 2 daysSpontaneous pain 10 cm VAS (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
81Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Rowbotham 2003 (Continued)
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk ldquoSystegraveme identiquerdquo without diclofenac
Incomplete outcome data (attrition bias)All outcomes
Unclear risk No information about imputation All par-ticipants included in analysis
Size Unclear risk 50 to 200 participants per treatment arm
Russell 1991
Methods R DB PC parallel groupsAffected area washed with soap and water and dried then gel applied and carefullyrubbed into skin 4 times daily for at least 7 daysAssessed at baseline 4 8 15 (if necessary) days at clinic and daily patient diary
Participants Acute soft tissue injuries (recent not recurrent)N = 214 (200 analysed)M 95 F 105Mean age 40 yearsBaseline pain gt 65 mm
Interventions Piroxicam gel 05 4 x 5 mg daily n = 100Placebo gel n = 100No other NSAIDs or analgesic drugs including liniments containing salicylates allowedAncillary therapy at the discretion of the investigator
Outcomes PGE 4-point scale (responder = ldquogoodrdquo and ldquoexcellentrdquo)Spontaneous pain mean reductionAdverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk ldquocomputer generated randomization coderdquo
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical base formulationrdquo
82Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Russell 1991 (Continued)
Incomplete outcome data (attrition bias)All outcomes
High risk High withdrawal rate (8 with piroxicam50 with placebo) No information aboutimputation
Size Unclear risk 50 to 100 participants per treatment group
Saillant 1998
Methods R DB PC parallel groupsTreatment for 7 daysAssessed at baseline 1 2 3 and 7 days
Participants Ankle sprain (lt 48 hours)N = 140M 72 F 62Age 18 to 65 yearsBaseline spontaneous pain ge 50 mm
Interventions (1) DHEP plaster (Flector Tissugel 1) 1 x daily n = 70(2) Placebo plaster 1 x daily n = 70Rescue medication paracetamol (to maximum 3 g daily)Ice allowed
Outcomes ge 30 decrease in PI at 7 daysPGE 4-point scale (responder = ldquoexcellentrdquo)No or low pain on passive stretchSingle foot leaning OK without painWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk Identical in size appearance and used sameformula as active patch without active in-gredient
Incomplete outcome data (attrition bias)All outcomes
Unclear risk Used LOCF for primary outcome With-drawals lt 7 and equal between groupsAll participants included in analysis
83Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Saillant 1998 (Continued)
Size Unclear risk 50 to 200 participants per treatment arm
Sanguinetti 1989
Methods R DB PC parallel groupsGel applied 3 times daily for 7 consecutive daysAssessment at baseline 7 days
Participants Soft tissue trauma (lt 48 hours)N = 82M = 47 F = 35Mean age 34 yearsBaseline pain moderate to severe
Interventions Felbinac gel 3 3 x daily n = 42Placebo gel 3 x daily n = 40No other NSAID steroid other topical application allowedRescue medication paracetamol felbinac is an active metabolite of the NSAID fenbufen
Outcomes PGE scale not reported (responder = ldquogoodrdquo and ldquovery goodrdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoindistinguishable in appearance colour orodourrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
84Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Sinniger 1981
Methods R DB PC parallel groupsCream applied 2 to 3 times daily with gentle massage or if massage not possible (toopainful) with protective dressingAssessment at baseline 5 10 days
Participants Minor soft tissue injuriesN = 20M 11 F 9Mean age 40 yearsBaseline pain not reported
Interventions Fentiazac cream 5 2 to 3 x daily n = 10Placebo cream 2 to 3 x daily n = 10All participants told to restNo other local and systemic treatments allowedRescue medication analgesic if actually needed
Outcomes Pain relief scale not reported (responder = total pain relief ) improvement in pain on movement pain scale not reported (mean data)Adverse events
Notes Oxford Quality Score R1 DB1 W0 Total = 25
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
85Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Spacca 2005
Methods R DB PC parallel groupsGel applied 3 times daily with gentle massage until complete absorption for up to 10daysAssessment at baseline 10 days in clinic and daily patient diary
Participants Shoulder periarthritis or lateral epicondylitis (lt 5 days)N = 155M 74 F 81Mean age 51 yearsBaseline pain with activity gt 70 mm
Interventions DHEP lecithin gel (Effigel) 3 x 5 g daily n = 79Placebo gel 3 x 5 g daily n = 76Rescue medication (paracetamol) allowed if pain unbearableNo other analgesic or anti-inflammatory drug allowed
Outcomes Improvement in pain 100 mm VAS (mean data)Adverse events
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Unclear risk No usable efficacy data
Size Unclear risk 50 to 200 participants per treatment arm
Sugioka 1984
Methods R DB AC parallel groupsGel applied to affected area 3 to 4 times daily without occlusion for 14 daysAssessed at baseline 7 14 days
Participants Non-traumatic diseases of muscle or tendonN = 366 (340 analysed for efficacy)M 115 F 202 (completers)
86Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Sugioka 1984 (Continued)
Age range 12 to 84 years (most 30 to 70)Baseline pain on movement ldquononerdquo or ldquomildrdquo in about 13 of participantsExclusions for protocol violations 8 piroxicam 18 indomethacin
Interventions Piroxicam gel 05 3 to 4 x 1 g daily n = 183Indomethacin gel 1 3 to 4 x 1 g daily n = 183No concomitant anti-inflammatory or analgesic drug including steroids or initiationof physical therapy allowed
Outcomes PGE 5-point scale (responder = ldquobetterrdquo or ldquomuch betterrdquo)Pain on movement 4-point scale (responder = ldquoreducedrdquo or ldquodisappearedrdquo)
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Low risk Key code sealed and retained until end ofstudy
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoboth packages were of the same appear-ance and indistinguishablerdquo and investiga-tors did not see contents
Incomplete outcome data (attrition bias)All outcomes
Low risk All relevant participants included in analy-sis
Size Unclear risk 50 to 200 participants per treatment arm
Thorling 1990
Methods R DB PC parallel groupsParticipants given specific instructions on how to apply gel (not reported) to affectedarea 2 to 6 times daily as requiredAssessment at baseline 3 7 days in clinic
Participants Soft tissue injuries (lt 48 hours)N = 120M 85 F 35Mean age 27 yearsBaseline pain moderate to severe
87Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Thorling 1990 (Continued)
Interventions Naproxen gel 10 2 to 6 x daily n = 60Placebo gel 2 to 6 x daily n = 60Rescue medication paracetamol 500 mg
Outcomes PGE 5-point scale (responder = ldquogoodrdquo and ldquovery goodrdquo)Pain on passive movement 4-point scale (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquosupplied in unmarked tubesrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size Unclear risk 50 to 200 participants per treatment arm
Tonutti 1994
Methods R DB AC parallel groupsGel applied 3 times daily for 2 to 3 weeksAssessed at baseline and intervals of 7 days
Participants Muscle or joint traumaN = 30M 20 F 10Mean age 34 years1 participant had injury of mild severity Mean baseline pain on active movement 28(scale 0 to 4)
Interventions Ketoprofen gel 5 3 x 2 to 3 g daily n = 15Etofenamate gel 5 3 x 2 to 3 g n = 15No concomitant treatment with NSAID aspirin steroid or physical therapy
88Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Tonutti 1994 (Continued)
Outcomes PGE 4-point scale (responder = ldquogoodrdquo and ldquoexcellentrdquo)Pain on movement 5-point scale (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not reported
Allocation concealment (selection bias) Unclear risk Not reported
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquothe two drugs were packed in indistin-guishable tubesrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
Vecchiet 1991
Methods R DB PC parallel groupsGel applied to the skin on and around painful area and gently rubbed in until absorbedtwice daily for up to 10 daysAssessed at baseline 5 10 days
Participants Soft tissue trauma (minor sports injuries)N = 60M 60Mean age 25 yearsMean baseline pain on active movement moderate
Interventions Meclofenamic acid gel 5 2 x 10 cm daily (= 4 g) n = 30Placebo gel 2 x 10 cm daily n = 30Both groups treated with ice rest and bandage for first 48 hours before starting testtreatmentRescue medication paracetamol
Outcomes PGE 4-point scale (responder = ldquogoodrdquo and ldquoexcellentrdquo)Pain on movement 4-point scale (mean data)Withdrawals
89Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Vecchiet 1991 (Continued)
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
Whitefield 2002
Methods R DB (double dummy) AC parallel groupsGel applied to affected site with gentle massage and 1 tablet taken 3 times daily for atleast 7 daysAssessed at baseline 7 14 (if necessary) days in clinic and daily patient diary
Participants Soft tissue injuries (lt 24 hours)N = 100M 95 F 5Mean age 26 years (range 18 to 50)Mean baseline pain on movement 22 cm
Interventions Ibuprofen gel 5 + placebo tablet 3 x daily n = 50Ibuprofen 400 mg tablet + placebo gel 3 x daily n = 50No other medication or physical therapy was prescribed and no other analgesics wereallowed
Outcomes PGE 3-point scale (responder = ldquoexcellentrdquo)Change in condition of injury site 5-point scale (responder = ldquocompletely betterrdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
90Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Whitefield 2002 (Continued)
Random sequence generation (selectionbias)
Unclear risk Not reported
Allocation concealment (selection bias) Unclear risk Not reported
Blinding (performance bias and detectionbias)All outcomes
Low risk Placebo tablets Identical in appearance toactive tablets Active and placebo gels hadsimilar physical characteristics and weresupplied in identical tubes
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size Unclear risk 50 participants per treatment arm
kermark 1990
Methods R DB (double dummy) PC AC parallel groupsSpray applied to affected area and capsules taken 3 times daily for 2 weeksAssessment at baseline 3 or 4 7 and 14 days
Participants Superficial overuse sports injuries (symptom onset 74 weeks)N = 70M 44 F 18 (completers)Mean age 30 yearsBaseline pain on palpation mostly slight to moderate
Interventions Elmetacin spray (indomethacin 1) 3 to 5 x 05 to 15 ml daily + placebo capsules n= 23Indomethacin capsules 3 x 25 mg daily + placebo spray n = 23Placebo spray and capsules n = 24Rescue medication paracetamol
Outcomes No pain on palpation (= responder)Participant improvement 100 mm VAS (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk ldquorandom number coderdquo
91Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
kermark 1990 (Continued)
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical in appearancerdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk Responder analysis lt 5 withdrawals
Size High risk lt 50 participants per treatment group
AC active control BOCF baseline observation carried forward DB double-blind DHEP diclofenac hydroxyethylpyrrolidine ordiclofenac epolamine F female HCl hydrochloride hep heparin LOCF last observation carried forward M male N numberof participants in study n number of participants in treatment arm PC placebo control PGE Participant Global Evaluation PIpain intensity R randomised VAS visual analogue scale W withdrawals
Characteristics of excluded studies [ordered by study ID]
Study Reason for exclusion
Ambrus 1987 No usable dichotomous data
Anon 1993 Not double-blind
Ascherl 1982 No usable dichotomous data
Bagliani 1976 Not an RCT
Baracchi 1982 No usable data
Burnham 1998 lt10 participantstreatment arm in first period of crossover study
Boumlhmer 1995 Active control invalid
Cesarone 2008 Not an RCT
Coulibaly 2009 Not double-blind
Diebschlag 1985 No usable dichotomous data
Diebschlag 1986 Inappropriate randomisation
Diebschlag 1992 No usable dichotomous data
92Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Fantato 1971 No usable dichotomous data
Galer 2000 No usable data
Hallmeier 1986 Not double-blind
Hallmeier 1988 Not double-blind
Kaneko 1999 Inappropriate randomisation - quasi-randomised
Kockelbergh 1985b Treatment not applied daily
Kuwabara 2013 Used NSAID-lidocaine combination (conference abstract)
Lee 1991 Not an RCT
Link 1996 No usable dichotomous data
May 2007 No usable dichotomous data
Oakland 1993 Inappropriate comparator
Odaglia 1987 Not an RCT
Picardi 1993 Not an RCT
Taboada 1992 Dose and duration of treatment unclear
Vanderstraeten 1990 Not double-blind
Vinciguerra 2008 Not an RCT
Von Klug 1977 Chronic and acute outcomes combined
RCT randomised controlled trial
Characteristics of studies awaiting assessment [ordered by study ID]
NCT00351104
Methods R DB PC parallel groups
Participants Ankle sprain or strain Grade I or II Age ge 18 yearsN = 220
93Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT00351104 (Continued)
Interventions Ketoprofen patch 20 x 1 dailyPlacebo patch x 1 dailyTreatment for 21 days
Outcomes Assessments days 3 7 21PI (0-10) at rest and during activitiesPGEFunctionRescue medication
Notes Sponsor Endo PharmaceuticalsEstimated completion February 2007 No results
NCT00352625
Methods R DB PC parallel groups
Participants Shoulder elbow or knee tendonitis or bursitis Age ge 18 yearsN = 330
Interventions Ketoprofen patch 20 x 1 dailyPlacebo patch x 1 dailyTreatment for 21 days
Outcomes Assessments days 3 7 21PI (0-10) at rest and during activitiesPGEFunctionRescue medication
Notes Sponsor Endo PharmaceuticalsEstimated completion February 2007 No results
NCT00426985
Methods R DB PC parallel groups
Participants Shoulder elbow or knee tendonitis or bursitis Age ge 18 years
Interventions Ketoprofen patch 20 x 1 dailyPlacebo patchTreatment for 21 days
Outcomes PI during activityPI at restPGERescue medication
94Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT00426985 (Continued)
Safety
Notes Study terminated May 2008 but ldquosufficient number of subjects accrued to conduct analysisrdquo No results
NCT00640705
Methods R DB PC parallel groups
Participants Ankle sprain or strain Grade I or II le 48 hours PI 510 to 910 Age 18 to 75 yearsN = 170
Interventions Diclofenac sodium patch (15 mg) x 1 dailyPlaceboTreatment for 7 days
Outcomes EfficacySafety
Notes Study terminated July 2008 (Sponsor decision - no further details) No results posted
NCT00640939
Methods R DB PC parallel groups
Participants Shoulder elbow or knee tendonitis or bursitis Age 18 to 75 yearsBaseline pain mild to moderate but states 510 to 910N = 308
Interventions Diclofenac sodium patch 1 (15 mg) x 1 dailyPlacebo patch x 1 dailyDuration of treatment not reported probably 14 days
Outcomes EfficacySafety
Notes Sponsor Cerimon PharmaceuticalsEstimated completion April 2008 No results
NCT00680472
Methods R DB PC parallel groups
Participants Acute unilateral shoulder pain requiring treatment for ge 2 weeks Age ge 18 yearsN = 368
95Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT00680472 (Continued)
Interventions Ketoprofen patch (HKT-500) x 1 dailyPlacebo x 1 dailyTreatment for 14 days
Outcomes PainSafety
Notes Sponsor Hisamitsu Pharmaceutical Co IncEstimated completion October 2008 No results
NCT00680784
Methods R DB PC parallel groups
Participants Acute benign ankle sprain Grade I-II le 48 hours Age ge 18 yearsN = 260
Interventions Ketoprofen patch (HKT-500) x 1 dailyPlacebo patch x 1 dailyTreatment for 14 days
Outcomes PainSafety
Notes Sponsor Hisamitsu Pharmaceutical Co IncEstimated completion November 2008 No results
NCT00765700
Methods R DB PC parallel groups
Participants Acute sprain or strain of upper and lower extremities mild to moderate injuries le 72 hours Age 18 to 70 yearsN = 364
Interventions Ketoprofen cream 10 1 g x 3 dailyPlacebo cream x 3 dailyTreatment for 7 days
Outcomes PainSafety
Notes Sponsor Imprimis Pharmaceuticals IncEstimated completion September 2009 No results
96Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT00869063
Methods R DB PC parallel groups
Participants Wrist sprain strain or contusion (mild to moderate) ldquorecentrdquo Age 17 to 75 yearsN = 214
Interventions Diclofenac sodium patch 1 x 1 dailyPlacebo patch x 1 dailyTreatment for 7 days
Outcomes Change in PI during activity at 3 and 7 daysSafety
Notes Sponsor Cerimon PharmaceuticalsEstimated completion September 2009 No results
NCT00869180
Methods R DB PC parallel groups
Participants Unilateral ankle sprain (mild or moderate) ldquorecentrdquo Age 17 to 75 yearsN = 219
Interventions Diclofenac sodium patch x 1 dailyPlacebo patch x 1 dailyTreatment for 7 days
Outcomes Change in PI during activity at 3 and 7 days
Notes Sponsor Cerimon PharmaceuticalsEstimated completion August 2009 No results
NCT00931866
Methods R DB PC parallel groups
Participants Unilateral soft tissue injury between mid-bicep to wrist or mid-thigh to ankle (mild to moderate) ldquorecentrdquo Age 18to 75 yearsN = 407
Interventions Diclofenac sodium patch x 1 dailyPlacebo patch x 1 dailyTreatment for 14 days
Outcomes Change in PI during activity at 7 and 14 daysPIPGERescue medicationSafety
97Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT00931866 (Continued)
Notes Sponsor Cerimon PharmaceuticalsEstimated completion December 2009 No results
NCT01874626
Methods R DB PC parallel groups
Participants Acute ankle sprain Grade I or II le 24 hours Age ge 16 years Baseline PI ge 510N = 305
Interventions Ibuprofen cream 200 mg in 27 g cream x 4 dailyPlacebo cream x 4 dailyTreatment for 7 days
Outcomes PI on movement daily to 7 daysPGE 7 daysSystemic and local adverse events
Notes Estimated completion January 2014 No results
NCT01957215
Methods R DB PC parallel groups
Participants Acute lateral ankle sprain Grade I-II le 24 hours PI ge 510 Age 18 to 65 yearsN = 270
Interventions Indomethacin patch x 2 dailyPlacebo patch x 2 dailyDuration of treatment not reported probably 7 days
Outcomes PI on movement dailyPR dailyPGERescue medication
Notes Estimated completion September 2014 No results
NCT02324270
Methods R DB PC and AC parallel groups
Participants Uncomplicated acute ankle sprain Grade I or II lt 48 hours PI gt 50100 Age 18 to 65 yearsN = 658
98Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT02324270 (Continued)
Interventions Diclofenac epolamine (Flector) 13 patchGeneric diclofenac epolamine 13 patchPlacebo patchDuration of treatment not reported
Outcomes Bioequivalence studyChange from baseline in PI (VAS) at 3 daysApplication site reactions
Notes Estimated completion December 2014 No results
Sarzi-Puttini 2014
Methods R DB AC parallel group non-inferiority studyDuration 7 days
Participants Acute musculoskeletal injury (mainly muscular joint tendon)N = 697M 271 F 426Mean age 52 years
Interventions Ketoprofen (SKP-021) patch ketoprofen 30 mgDiclofenac (Voltadola) patch diclofenac sodium 140 mgPatches applied twice daily for 7 days
Outcomes ge 50 reduction in PI from baseline to end of study (100 mm VAS)Patient overall ratingClinical symptoms (4-point scale)Time to responseAdverse events skin reactionsSerious adverse events
Notes ldquoThe analysis of the data of this trial showed that the two formulations were equally effective and well tolerated inthe treatment of acute musculoskeletal injuriesrdquo
AC active-controlled DB double-blind F female M male N number of participants in study PC placebo-controlled PGE PatientGlobal Evaluation of treatment PI pain intensity R randomised VAS visual analogue scale
99Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Characteristics of ongoing studies [ordered by study ID]
NCT01945034
Trial name or title Placebo-controlled double-blind evaluation of the efficacy and safety of ibuprofen 5 topical gel for thetreatment of ankle sprain
Methods R DB PC parallel groups
Participants Ankle sprain Grade I or II age ge 12 yearsEstimated enrolment 280
Interventions Ibuprofen gel 5 x 2 dailyIbuprofen gel 5 x 3 dailyPlacebo gelTreatment for 7 days with additional 3 days as needed
Outcomes PI on weight bearing and rest at 7 daysPGE at 10 daysRescue medicationSafety
Starting date November 2013
Contact information Pfizer
Notes Estimated completion February 2015Includes participants aged ge 12 years
NCT02100670
Trial name or title A clinical study to assess the efficacy and onset of pain relief of topical MFC51123 diclofenac-menthol gelversus controls in ankle sprain
Methods R DB PC and AC parallel groups
Participants Acute lateral ankle sprain Grade I or II le 24 hours PI ge 510 Age 16 to 65 yearsEstimated enrolment 400
Interventions Diclofenac sodium 1 + methanol 3 gelDiclofenac sodium 1 + methanol 009 gelMethanol 3 gelPlacebo + 009 methanol gelTreatment for 10 days with 4 g gel x 4 daily
Outcomes PRPID on movement (0 to 10 days)PGETime to complete recoveryAdverse events
100Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT02100670 (Continued)
Starting date November 2013
Contact information GSKClinicalSupportHDgskcom
Notes Estimated completion November 2014
NCT02290821
Trial name or title A randomised double-blind placebo-controlled parallel group study to evaluate the efficacy and safety ofdiclofenac sodium topical gel (DSG) 1 applied four times daily in subjects with acute blunt soft tissueinjuriescontusions of the limbs
Methods R DB PC parallel groups
Participants Acute blunt soft tissue injuries or contusions of the limbs le 6 hours (ldquofreshrdquo) Age ge 16 yearsEstimated enrolment 200
Interventions Diclofenac sodium gel 1 x 4 dailyPlacebo gelDuration of treatment not reported
Outcomes PainSafety
Starting date December 2014
Contact information Novartis
Notes Estimated completion August 2015
AC active control DB double-blind PC placebo-controlled PGE Patient Global Evaluation of treatment PI pain intensity PIDpain intensity difference PR pain relief R randomised
101Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
D A T A A N D A N A L Y S E S
Comparison 1 Individual NSAID versus placebo
Outcome or subgroup titleNo ofstudies
No ofparticipants Statistical method Effect size
1 Clinical success 29 Risk Ratio (M-H Fixed 95 CI) Subtotals only11 Diclofenac 10 2050 Risk Ratio (M-H Fixed 95 CI) 160 [149 172]12 Ibuprofen 5 436 Risk Ratio (M-H Fixed 95 CI) 164 [133 201]13 Ketoprofen 7 683 Risk Ratio (M-H Fixed 95 CI) 156 [137 177]14 Piroxicam 3 504 Risk Ratio (M-H Fixed 95 CI) 148 [127 173]15 Indomethacin 3 341 Risk Ratio (M-H Fixed 95 CI) 126 [103 155]16 Benzydamine 3 193 Risk Ratio (M-H Fixed 95 CI) 115 [096 138]
2 Local adverse events 33 Risk Ratio (M-H Fixed 95 CI) Subtotals only21 Diclofenac 15 3271 Risk Ratio (M-H Fixed 95 CI) 078 [056 110]22 Ibuprofen 3 321 Risk Ratio (M-H Fixed 95 CI) 230 [098 543]23 Ketoprofen 8 852 Risk Ratio (M-H Fixed 95 CI) 119 [083 170]24 Piroxicam 3 522 Risk Ratio (M-H Fixed 95 CI) 042 [017 108]25 Felbinac 3 397 Risk Ratio (M-H Fixed 95 CI) 191 [049 750]26 Indomethacin 3 354 Risk Ratio (M-H Fixed 95 CI) 265 [091 773]
Comparison 2 Diclofenac versus placebo (effect of formulation)
Outcome or subgroup titleNo ofstudies
No ofparticipants Statistical method Effect size
1 Clinical success 10 2050 Risk Ratio (M-H Fixed 95 CI) 160 [149 172]11 Plaster - Flector 4 1030 Risk Ratio (M-H Fixed 95 CI) 153 [136 171]12 Plaster - other 3 474 Risk Ratio (M-H Fixed 95 CI) 155 [137 175]13 Gel - Emulgel 2 314 Risk Ratio (M-H Fixed 95 CI) 384 [268 550]14 Gel - other 1 232 Risk Ratio (M-H Fixed 95 CI) 115 [105 127]
Comparison 3 Ibuprofen versus placebo (effect of formulation)
Outcome or subgroup titleNo ofstudies
No ofparticipants Statistical method Effect size
1 Clinical success 5 436 Risk Ratio (M-H Fixed 95 CI) 164 [133 201]11 Cream 3 195 Risk Ratio (M-H Fixed 95 CI) 128 [103 159]12 Gel 2 241 Risk Ratio (M-H Fixed 95 CI) 266 [169 421]
102Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Comparison 4 Ketoprofen versus placebo
Outcome or subgroup titleNo ofstudies
No ofparticipants Statistical method Effect size
1 Clinical success 7 683 Risk Ratio (M-H Fixed 95 CI) 156 [137 177]11 Plaster 2 335 Risk Ratio (M-H Fixed 95 CI) 121 [104 140]12 Gel 5 348 Risk Ratio (M-H Fixed 95 CI) 219 [174 275]
Comparison 5 All topical NSAIDs versus placebo
Outcome or subgroup titleNo ofstudies
No ofparticipants Statistical method Effect size
1 Local adverse events 42 6740 Risk Ratio (M-H Fixed 95 CI) 098 [080 121]2 Systemic adverse events 36 5576 Risk Ratio (M-H Fixed 95 CI) 096 [073 126]3 Adverse event withdrawals 42 6405 Risk Ratio (M-H Fixed 95 CI) 101 [064 159]
Comparison 6 Topical NSAID versus active comparator
Outcome or subgroup titleNo ofstudies
No ofparticipants Statistical method Effect size
1 Clinical success - topicalpiroxicam vs topicalindomethacin
3 641 Risk Ratio (M-H Fixed 95 CI) 124 [107 144]
2 Local adverse events -topical piroxicam vs topicalindomethacin
3 671 Risk Ratio (M-H Fixed 95 CI) 021 [009 047]
103Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Analysis 11 Comparison 1 Individual NSAID versus placebo Outcome 1 Clinical success
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 1 Individual NSAID versus placebo
Outcome 1 Clinical success
Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
1 Diclofenac
Coudreuse 2010 99117 82116 175 120 [ 104 138 ]
Joussellin 2003 3668 2466 52 146 [ 099 215 ]
Klainguti 2010 5662 4659 100 116 [ 099 136 ]
Li 2013 173192 114192 242 152 [ 134 172 ]
Predel 2004 5560 560 11 1100 [ 474 2555 ]
Predel 2012 (1) 5980 941 25 336 [ 186 607 ]
Predel 2012 (2) 5780 841 22 365 [ 193 690 ]
Predel 2013a 111118 93114 201 115 [ 105 127 ]
Predel 2013b 3636 736 16 487 [ 257 923 ]
Rowbotham 2003 75191 48181 104 148 [ 110 200 ]
Saillant 1998 4370 2570 53 172 [ 119 248 ]
Subtotal (95 CI) 1074 976 1000 160 [ 149 172 ]Total events 800 (NSAID) 461 (Placebo)
Heterogeneity Chi2 = 12176 df = 10 (Plt000001) I2 =92
Test for overall effect Z = 1278 (P lt 000001)
2 Ibuprofen
Billigmann 1996 2580 1080 136 250 [ 129 486 ]
Campbell 1994 2126 1925 264 106 [ 080 142 ]
Dreiser 1988 2632 1232 164 217 [ 134 349 ]
Machen 2002 2540 941 121 285 [ 152 532 ]
Ramesh 1983 2340 2340 314 100 [ 069 146 ]
Subtotal (95 CI) 218 218 1000 164 [ 133 201 ]Total events 120 (NSAID) 73 (Placebo)
Heterogeneity Chi2 = 2102 df = 4 (P = 000031) I2 =81
Test for overall effect Z = 465 (P lt 000001)
3 Ketoprofen
Airaksinen 1993 2429 1427 91 160 [ 107 238 ]
Dreiser 1989 1830 530 32 360 [ 154 844 ]
02 05 1 2 5
Favours placebo Favours NSAID
(Continued )
104Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
( Continued)Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Julien 1989 1830 630 38 300 [ 138 650 ]
Kockelbergh 1985 3038 2236 142 129 [ 095 176 ]
Mazi res 2005a 7281 6082 376 121 [ 104 141 ]
Mazi res 2005b 5087 4185 262 119 [ 090 158 ]
Noret 1987 3951 947 59 399 [ 218 733 ]
Subtotal (95 CI) 346 337 1000 156 [ 137 177 ]Total events 251 (NSAID) 157 (Placebo)
Heterogeneity Chi2 = 3104 df = 6 (P = 000002) I2 =81
Test for overall effect Z = 683 (P lt 000001)
4 Piroxicam
Aoki 1984 5672 3367 286 158 [ 120 207 ]
Fujimaki 1985 4483 4082 337 109 [ 080 147 ]
Russell 1991 79100 45100 377 176 [ 138 223 ]
Subtotal (95 CI) 255 249 1000 148 [ 127 173 ]Total events 179 (NSAID) 118 (Placebo)
Heterogeneity Chi2 = 624 df = 2 (P = 004) I2 =68
Test for overall effect Z = 499 (P lt 000001)
5 Indomethacin
Aoki 1984 4164 3367 413 130 [ 096 176 ]
Fujimaki 1985 4482 4082 513 110 [ 082 148 ]
kermark 1990 1222 624 74 218 [ 099 481 ]
Subtotal (95 CI) 168 173 1000 126 [ 103 155 ]Total events 97 (NSAID) 79 (Placebo)
Heterogeneity Chi2 = 269 df = 2 (P = 026) I2 =26
Test for overall effect Z = 221 (P = 0027)
6 Benzydamine
Chatterjee 1977 2125 1225 185 175 [ 112 272 ]
Haig 1986 1821 1322 196 145 [ 098 214 ]
Linde 1985 3550 4050 618 088 [ 070 110 ]
Subtotal (95 CI) 96 97 1000 115 [ 096 138 ]Total events 74 (NSAID) 65 (Placebo)
Heterogeneity Chi2 = 1033 df = 2 (P = 001) I2 =81
Test for overall effect Z = 152 (P = 013)
02 05 1 2 5
Favours placebo Favours NSAID
(1) Three times daily application
(2) Twice daily application
105Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Analysis 12 Comparison 1 Individual NSAID versus placebo Outcome 2 Local adverse events
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 1 Individual NSAID versus placebo
Outcome 2 Local adverse events
Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
1 Diclofenac
Coudreuse 2010 1117 7116 102 014 [ 002 113 ]
Jenoure 1997 144 141 15 093 [ 006 1442 ]
Joussellin 2003 168 366 44 032 [ 003 303 ]
Klainguti 2010 0126 159 30 016 [ 001 381 ]
Kuehl 2011 16207 12211 173 136 [ 066 280 ]
Li 2013 4192 3192 44 133 [ 030 588 ]
NCT01255423 1104 3102 44 033 [ 003 309 ]
NCT01272934 1102 0103 07 303 [ 012 7350 ]
NCT01272947 0104 0100 Not estimable
Predel 2012 (1) 1160 182 19 051 [ 003 809 ]
Predel 2013a 1120 4116 59 024 [ 003 213 ]
Predel 2013b 036 036 Not estimable
Rowbotham 2003 27191 31181 463 083 [ 051 133 ]
Saillant 1998 070 070 Not estimable
Spacca 2005 079 076 Not estimable
Subtotal (95 CI) 1720 1551 1000 078 [ 056 110 ]Total events 54 (NSAID) 66 (Placebo)
Heterogeneity Chi2 = 944 df = 10 (P = 049) I2 =00
Test for overall effect Z = 142 (P = 016)
2 Ibuprofen
Billigmann 1996 1180 480 573 275 [ 091 827 ]
Machen 2002 440 241 283 205 [ 040 1057 ]
Ramesh 1983 140 140 143 100 [ 006 1544 ]
Subtotal (95 CI) 160 161 1000 230 [ 098 543 ]Total events 16 (NSAID) 7 (Placebo)
Heterogeneity Chi2 = 048 df = 2 (P = 079) I2 =00
Test for overall effect Z = 190 (P = 0057)
3 Ketoprofen
002 01 1 10 50
Favours NSAID Favours placebo
(Continued )
106Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
( Continued)
Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Airaksinen 1993 529 427 99 116 [ 035 389 ]
Dreiser 1989 030 230 60 020 [ 001 400 ]
Julien 1989 130 030 12 300 [ 013 7083 ]
Kockelbergh 1985 138 136 24 095 [ 006 1459 ]
Mazi res 2005a 1281 682 142 202 [ 080 513 ]
Mazi res 2005b 2987 2785 651 105 [ 068 161 ]
Noret 1987 151 047 12 277 [ 012 6636 ]
Parrini 1992 083 086 Not estimable
Subtotal (95 CI) 429 423 1000 119 [ 083 170 ]Total events 49 (NSAID) 40 (Placebo)
Heterogeneity Chi2 = 357 df = 6 (P = 073) I2 =00
Test for overall effect Z = 095 (P = 034)
4 Piroxicam
Aoki 1984 179 274 147 047 [ 004 506 ]
Fujimaki 1985 183 282 143 049 [ 005 534 ]
Russell 1991 4102 10102 710 040 [ 013 123 ]
Subtotal (95 CI) 264 258 1000 042 [ 017 108 ]Total events 6 (NSAID) 14 (Placebo)
Heterogeneity Chi2 = 003 df = 2 (P = 098) I2 =00
Test for overall effect Z = 180 (P = 0072)
5 Felbinac
McLatchie 1989 3118 2113 666 144 [ 024 844 ]
Morris 1991 041 043 Not estimable
Sanguinetti 1989 342 140 334 286 [ 031 2634 ]
Subtotal (95 CI) 201 196 1000 191 [ 049 750 ]Total events 6 (NSAID) 3 (Placebo)
Heterogeneity Chi2 = 023 df = 1 (P = 063) I2 =00
Test for overall effect Z = 093 (P = 035)
6 Indomethacin
Aoki 1984 270 274 440 106 [ 015 730 ]
Fujimaki 1985 582 282 452 250 [ 050 1252 ]
kermark 1990 422 024 108 978 [ 056 17191 ]
Subtotal (95 CI) 174 180 1000 265 [ 091 773 ]Total events 11 (NSAID) 4 (Placebo)
Heterogeneity Chi2 = 167 df = 2 (P = 043) I2 =00
Test for overall effect Z = 179 (P = 0073)
002 01 1 10 50
Favours NSAID Favours placebo
(1) DHEP + heparin and DHEP combined
107Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Analysis 21 Comparison 2 Diclofenac versus placebo (effect of formulation) Outcome 1 Clinical success
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 2 Diclofenac versus placebo (effect of formulation)
Outcome 1 Clinical success
Study or subgroup Experimental Control Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
1 Plaster - Flector
Joussellin 2003 3668 2466 52 146 [ 099 215 ]
Li 2013 173192 114192 242 152 [ 134 172 ]
Rowbotham 2003 75191 48181 104 148 [ 110 200 ]
Saillant 1998 4370 2570 53 172 [ 119 248 ]
Subtotal (95 CI) 521 509 451 153 [ 136 171 ]Total events 327 (Experimental) 211 (Control)
Heterogeneity Chi2 = 051 df = 3 (P = 092) I2 =00
Test for overall effect Z = 720 (P lt 000001)
2 Plaster - other
Coudreuse 2010 99117 82116 175 120 [ 104 138 ]
Klainguti 2010 5662 4659 100 116 [ 099 136 ]
Predel 2004 5560 560 11 1100 [ 474 2555 ]
Subtotal (95 CI) 239 235 285 155 [ 137 175 ]Total events 210 (Experimental) 133 (Control)
Heterogeneity Chi2 = 4658 df = 2 (Plt000001) I2 =96
Test for overall effect Z = 699 (P lt 000001)
3 Gel - Emulgel
Predel 2012 (1) 116160 1782 48 350 [ 227 540 ]
Predel 2013b 3636 736 16 487 [ 257 923 ]
Subtotal (95 CI) 196 118 64 384 [ 268 550 ]Total events 152 (Experimental) 24 (Control)
Heterogeneity Chi2 = 070 df = 1 (P = 040) I2 =00
Test for overall effect Z = 734 (P lt 000001)
4 Gel - other
Predel 2013a 111118 93114 201 115 [ 105 127 ]
Subtotal (95 CI) 118 114 201 115 [ 105 127 ]Total events 111 (Experimental) 93 (Control)
02 05 1 2 5
Favours placebo Favours diclofenac
(Continued )
108Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
( Continued)Study or subgroup Experimental Control Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Heterogeneity not applicable
Test for overall effect Z = 284 (P = 00045)
Total (95 CI) 1074 976 1000 160 [ 149 172 ]Total events 800 (Experimental) 461 (Control)
Heterogeneity Chi2 = 12176 df = 9 (Plt000001) I2 =93
Test for overall effect Z = 1278 (P lt 000001)
Test for subgroup differences Chi2 = 5003 df = 3 (P = 000) I2 =94
02 05 1 2 5
Favours placebo Favours diclofenac
(1) twice daily and three times daily applications combined
Analysis 31 Comparison 3 Ibuprofen versus placebo (effect of formulation) Outcome 1 Clinical success
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 3 Ibuprofen versus placebo (effect of formulation)
Outcome 1 Clinical success
Study or subgroup Ibuprofen Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
1 Cream
Campbell 1994 2126 1925 264 106 [ 080 142 ]
Dreiser 1988 2632 1232 164 217 [ 134 349 ]
Ramesh 1983 2340 2340 314 100 [ 069 146 ]
Subtotal (95 CI) 98 97 742 128 [ 103 159 ]Total events 70 (Ibuprofen) 54 (Placebo)
Heterogeneity Chi2 = 791 df = 2 (P = 002) I2 =75
Test for overall effect Z = 222 (P = 0026)
2 Gel
Billigmann 1996 2580 1080 136 250 [ 129 486 ]
Machen 2002 2540 941 121 285 [ 152 532 ]
Subtotal (95 CI) 120 121 258 266 [ 169 421 ]Total events 50 (Ibuprofen) 19 (Placebo)
001 01 1 10 100
Favours placebo Favours ibuprofen
(Continued )
109Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
( Continued)Study or subgroup Ibuprofen Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Heterogeneity Chi2 = 008 df = 1 (P = 078) I2 =00
Test for overall effect Z = 420 (P = 0000026)
Total (95 CI) 218 218 1000 164 [ 133 201 ]Total events 120 (Ibuprofen) 73 (Placebo)
Heterogeneity Chi2 = 2102 df = 4 (P = 000031) I2 =81
Test for overall effect Z = 465 (P lt 000001)
Test for subgroup differences Chi2 = 805 df = 1 (P = 000) I2 =88
001 01 1 10 100
Favours placebo Favours ibuprofen
Analysis 41 Comparison 4 Ketoprofen versus placebo Outcome 1 Clinical success
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 4 Ketoprofen versus placebo
Outcome 1 Clinical success
Study or subgroup Diclofenac Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
1 Plaster
Mazi res 2005a 7281 6082 376 121 [ 104 141 ]
Mazi res 2005b 5087 4185 262 119 [ 090 158 ]
Subtotal (95 CI) 168 167 638 121 [ 104 140 ]Total events 122 (Diclofenac) 101 (Placebo)
Heterogeneity Chi2 = 002 df = 1 (P = 090) I2 =00
Test for overall effect Z = 249 (P = 0013)
2 Gel
Airaksinen 1993 2429 1427 91 160 [ 107 238 ]
Dreiser 1989 1830 530 32 360 [ 154 844 ]
Julien 1989 1830 630 38 300 [ 138 650 ]
Kockelbergh 1985 3038 2236 142 129 [ 095 176 ]
Noret 1987 3951 947 59 399 [ 218 733 ]
001 01 1 10 100
Favours placebo Favours diclofenac
(Continued )
110Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
( Continued)Study or subgroup Diclofenac Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Subtotal (95 CI) 178 170 362 219 [ 174 275 ]Total events 129 (Diclofenac) 56 (Placebo)
Heterogeneity Chi2 = 1937 df = 4 (P = 000067) I2 =79
Test for overall effect Z = 674 (P lt 000001)
Total (95 CI) 346 337 1000 156 [ 137 177 ]Total events 251 (Diclofenac) 157 (Placebo)
Heterogeneity Chi2 = 3104 df = 6 (P = 000002) I2 =81
Test for overall effect Z = 683 (P lt 000001)
Test for subgroup differences Chi2 = 1859 df = 1 (P = 000) I2 =95
001 01 1 10 100
Favours placebo Favours diclofenac
Analysis 51 Comparison 5 All topical NSAIDs versus placebo Outcome 1 Local adverse events
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 5 All topical NSAIDs versus placebo
Outcome 1 Local adverse events
Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Airaksinen 1993 529 427 26 116 [ 035 389 ]
Aoki 1984 (1) 270 137 08 106 [ 010 1128 ]
Aoki 1984 (2) 179 137 09 047 [ 003 728 ]
Auclair 1989 5123 6116 39 079 [ 025 251 ]
Billigmann 1996 1180 480 26 275 [ 091 827 ]
Costantino 2011 (3) 3287 6142 51 025 [ 006 097 ]
Coudreuse 2010 1117 7116 45 014 [ 002 113 ]
Diebshlag 1990 (4) 113 06 04 150 [ 007 3229 ]
Diebshlag 1990 (5) 112 06 04 162 [ 008 3466 ]
Dreiser 1989 030 230 16 020 [ 001 400 ]
002 01 1 10 50
Favours NSAID Favours placebo
(Continued )
111Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
( Continued)Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Dreiser 1990 030 330 22 014 [ 001 265 ]
Dreiser 1994 265 066 03 508 [ 025 10373 ]
Fujimaki 1985 (6) 183 141 09 049 [ 003 770 ]
Fujimaki 1985 (7) 582 141 09 250 [ 030 2070 ]
Haig 1986 021 022 Not estimable
Jenoure 1997 144 141 07 093 [ 006 1442 ]
Joussellin 2003 168 366 19 032 [ 003 303 ]
Julien 1989 130 030 03 300 [ 013 7083 ]
Klainguti 2010 (8) 0126 159 13 016 [ 001 381 ]
Kockelbergh 1985 138 136 07 095 [ 006 1459 ]
Kuehl 2011 16207 12211 76 136 [ 066 280 ]
Li 2013 4192 3192 19 133 [ 030 588 ]
Linde 1985 050 050 Not estimable
Machen 2002 440 241 13 205 [ 040 1057 ]
Mazi res 2005a 1281 682 38 202 [ 080 513 ]
Mazi res 2005b 2987 2785 174 105 [ 068 161 ]
McLatchie 1989 3118 2113 13 144 [ 024 844 ]
Morris 1991 041 043 Not estimable
NCT01255423 1104 3102 19 033 [ 003 309 ]
NCT01272934 1102 0103 03 303 [ 012 7350 ]
NCT01272947 0104 0100 Not estimable
Noret 1987 151 047 03 277 [ 012 6636 ]
Parrini 1992 083 086 Not estimable
Predel 2012 (9) 1160 182 08 051 [ 003 809 ]
Predel 2013a 1120 4116 26 024 [ 003 213 ]
Predel 2013b 036 036 Not estimable
Ramesh 1983 140 140 06 100 [ 006 1544 ]
Rowbotham 2003 27191 31181 203 083 [ 051 133 ]
Russell 1991 4102 10102 64 040 [ 013 123 ]
Saillant 1998 070 070 Not estimable
Sanguinetti 1989 342 140 07 286 [ 031 2634 ]
002 01 1 10 50
Favours NSAID Favours placebo
(Continued )
112Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
( Continued)Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Sinniger 1981 010 010 Not estimable
Spacca 2005 079 076 Not estimable
Thorling 1990 160 060 03 300 [ 012 7220 ]
kermark 1990 422 024 03 978 [ 056 17191 ]
Total (95 CI) 3619 3121 1000 098 [ 080 121 ]Total events 155 (NSAID) 145 (Placebo)
Heterogeneity Chi2 = 3354 df = 35 (P = 054) I2 =00
Test for overall effect Z = 017 (P = 087)
Test for subgroup differences Not applicable
002 01 1 10 50
Favours NSAID Favours placebo
(1) indomethacin
(2) piroxicam
(3) DHEP + heparin and DHEP combined
(4) ketorolac
(5) etofenamate
(6) piroxicam
(7) indomethacin
(8) DHEP + heparin and DHEP combined
(9) Twice and three times daily application combined
113Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Analysis 52 Comparison 5 All topical NSAIDs versus placebo Outcome 2 Systemic adverse events
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 5 All topical NSAIDs versus placebo
Outcome 2 Systemic adverse events
Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Airaksinen 1993 129 027 05 280 [ 012 6593 ]
Aoki 1984 (1) 079 037 Not estimable
Aoki 1984 (2) 070 037 Not estimable
Campbell 1994 126 025 05 289 [ 012 6775 ]
Chatterjee 1977 025 025 Not estimable
Coudreuse 2010 1117 0116 05 297 [ 012 7227 ]
Diebshlag 1990 (3) 013 06 Not estimable
Diebshlag 1990 (4) 012 06 Not estimable
Dreiser 1988 032 032 Not estimable
Dreiser 1989 030 030 Not estimable
Dreiser 1990 030 030 Not estimable
Dreiser 1994 065 066 Not estimable
Fujimaki 1985 (5) 083 041 Not estimable
Fujimaki 1985 (6) 182 041 07 152 [ 006 3647 ]
Haig 1986 021 022 Not estimable
Hoffmann 2012 0115 0118 Not estimable
Joussellin 2003 168 066 05 291 [ 012 7025 ]
Klainguti 2010 (7) 2126 059 07 236 [ 012 4844 ]
Kuehl 2011 15207 23211 236 066 [ 036 124 ]
Li 2013 10192 4192 41 250 [ 080 783 ]
Linde 1985 050 050 Not estimable
Machen 2002 040 041 Not estimable
Mazi res 2005a 1187 785 73 154 [ 062 377 ]
Mazi res 2005b 1381 1482 144 094 [ 047 187 ]
McLatchie 1989 0118 0113 Not estimable
Morris 1991 041 043 Not estimable
001 01 1 10 100
Favours topical NSAID Favours placebo
(Continued )
114Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
( Continued)Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
NCT01272947 2104 2100 21 096 [ 014 670 ]
Parrini 1992 083 086 Not estimable
Predel 2004 060 060 Not estimable
Predel 2012 (8) 3160 382 41 051 [ 011 248 ]
Predel 2013a 6120 8116 84 073 [ 026 203 ]
Predel 2013b 036 136 16 033 [ 001 792 ]
Rowbotham 2003 21191 22181 234 090 [ 052 159 ]
Russell 1991 4102 7102 73 057 [ 017 189 ]
Saillant 1998 070 070 Not estimable
Sanguinetti 1989 042 040 Not estimable
Sinniger 1981 010 010 Not estimable
Spacca 2005 079 076 Not estimable
Thorling 1990 060 060 Not estimable
Total (95 CI) 2956 2620 1000 096 [ 073 126 ]Total events 92 (NSAID) 91 (Placebo)
Heterogeneity Chi2 = 946 df = 15 (P = 085) I2 =00
Test for overall effect Z = 030 (P = 076)
Test for subgroup differences Not applicable
001 01 1 10 100
Favours topical NSAID Favours placebo
(1) piroxicam
(2) indomethacin
(3) ketorolac
(4) etofenamate
(5) piroxicam
(6) indomethacin
(7) DHEP + heparin and DHEP combined
(8) Twice and three times daily application combined
115Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Analysis 53 Comparison 5 All topical NSAIDs versus placebo Outcome 3 Adverse event withdrawals
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 5 All topical NSAIDs versus placebo
Outcome 3 Adverse event withdrawals
Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Airaksinen 1993 029 027 Not estimable
Aoki 1984 072 034 Not estimable
Aoki 1984 064 033 Not estimable
Auclair 1989 1123 0116 14 283 [ 012 6879 ]
Billigmann 1996 280 080 14 500 [ 024 10253 ]
Campbell 1994 026 025 Not estimable
Chatterjee 1977 025 025 Not estimable
Costantino 2011 0188 0142 Not estimable
Coudreuse 2010 0117 0116 Not estimable
Diebshlag 1990 013 06 Not estimable
Diebshlag 1990 012 06 Not estimable
Dreiser 1988 032 032 Not estimable
Dreiser 1989 030 230 69 020 [ 001 400 ]
Dreiser 1990 030 130 42 033 [ 001 787 ]
Dreiser 1994 065 066 Not estimable
Fujimaki 1985 083 041 Not estimable
Fujimaki 1985 482 041 18 455 [ 025 8261 ]
Julien 1989 030 030 Not estimable
Kockelbergh 1985 038 036 Not estimable
Kuehl 2011 4207 9211 248 045 [ 014 145 ]
Li 2013 2192 0192 14 500 [ 024 10347 ]
Linde 1985 050 050 Not estimable
Machen 2002 040 041 Not estimable
Mazi res 2005a 987 685 169 147 [ 055 394 ]
Mazi res 2005b 381 082 14 709 [ 037 13503 ]
McLatchie 1989 0118 0113 Not estimable
001 01 1 10 100
Favours topical NSAID Favours placebo
(Continued )
116Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
( Continued)Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Morris 1991 041 043 Not estimable
NCT01255423 0104 0102 Not estimable
NCT01272934 0102 1103 41 034 [ 001 817 ]
NCT01272947 0104 0100 Not estimable
Noret 1987 148 045 14 282 [ 012 6740 ]
Parrini 1992 060 060 Not estimable
Predel 2004 160 060 14 300 [ 012 7220 ]
Predel 2012 2160 182 37 103 [ 009 1114 ]
Predel 2013a 1120 1116 28 097 [ 006 1527 ]
Predel 2013b 036 036 Not estimable
Ramesh 1983 140 140 28 100 [ 006 1544 ]
Rowbotham 2003 0191 0181 Not estimable
Russell 1991 1102 8102 222 013 [ 002 098 ]
Saillant 1998 070 070 Not estimable
Sanguinetti 1989 042 040 Not estimable
Sinniger 1981 010 010 Not estimable
Spacca 2005 079 076 Not estimable
Thorling 1990 060 060 Not estimable
kermark 1990 122 024 13 326 [ 014 7610 ]
Total (95 CI) 3365 3040 1000 101 [ 064 159 ]Total events 33 (NSAID) 30 (Placebo)
Heterogeneity Chi2 = 1502 df = 16 (P = 052) I2 =00
Test for overall effect Z = 004 (P = 097)
Test for subgroup differences Not applicable
001 01 1 10 100
Favours topical NSAID Favours placebo
117Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Analysis 61 Comparison 6 Topical NSAID versus active comparator Outcome 1 Clinical success - topical
piroxicam vs topical indomethacin
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 6 Topical NSAID versus active comparator
Outcome 1 Clinical success - topical piroxicam vs topical indomethacin
Study or subgroup Piroxicam Indomethacin Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Aoki 1984 5672 4164 301 121 [ 097 151 ]
Fujimaki 1985 4483 4482 307 099 [ 074 131 ]
Sugioka 1984 85175 55165 392 146 [ 112 190 ]
Total (95 CI) 330 311 1000 124 [ 107 144 ]Total events 185 (Piroxicam) 140 (Indomethacin)
Heterogeneity Chi2 = 390 df = 2 (P = 014) I2 =49
Test for overall effect Z = 278 (P = 00055)
Test for subgroup differences Not applicable
01 02 05 1 2 5 10
Favours indomethacin Favours piroxicam
118Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Analysis 62 Comparison 6 Topical NSAID versus active comparator Outcome 2 Local adverse events -
topical piroxicam vs topical indomethacin
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 6 Topical NSAID versus active comparator
Outcome 2 Local adverse events - topical piroxicam vs topical indomethacin
Study or subgroup Piroxicam Indomethacin Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Aoki 1984 179 270 64 044 [ 004 478 ]
Fujimaki 1985 183 582 152 020 [ 002 165 ]
Sugioka 1984 5178 26179 784 019 [ 008 049 ]
Total (95 CI) 340 331 1000 021 [ 009 047 ]Total events 7 (Piroxicam) 33 (Indomethacin)
Heterogeneity Chi2 = 041 df = 2 (P = 081) I2 =00
Test for overall effect Z = 382 (P = 000013)
Test for subgroup differences Not applicable
002 01 1 10 50
Favours piroxicam Favours indomethacin
A P P E N D I C E S
Appendix 1 CENTRAL search strategy (via CRSO) for 2015 update
1 MESH DESCRIPTOR Anti-Inflammatory Agents Non-Steroidal EXPLODE ALL TREES (13419)2 (bufexamac OR bufexine OR calmaderm OR ekzemase OR dicoflenac OR solaraze OR pennsaid OR voltarol OR emugel OR
voltarene OR voltarol OR optha OR voltaren OR etofenamate OR afrolate OR algesalona OR bayro OR deiron OR etofen ORflexium OR flogoprofen OR rheuma-gel OR rheumon OR traumalix OR traumon OR zenavan OR felbinac OR dolinac OR flexfreeOR napageln OR target OR traxam OR fentiazac OR domureuma OR fentiazaco OR norvedan OR riscalon OR fepradinol ORdalgen OR flexidol OR cocresol OR rangozona OR reuflodol OR pinazone OR zepelin OR flufenamic OR dignodolin OR rheumaOR lindofluid OR sastridex OR lunoxaprofen OR priaxim OR flubiprofen OR fenomel OR ocufen OR ocuflur OR ldquoTrans ActLATrdquo OR tulip OR ibuprofen OR cuprofen OR ldquodeep relief rdquo OR fenbid OR ibu-cream OR ibugel OR ibuleve OR ibumousse ORibuspray OR ldquonurofen gelrdquo OR proflex OR motrin OR advil OR radian OR ralgex OR ibutop OR indomethacin OR indocin ORindospray OR isonixin OR nixyn OR ketoprofen OR tiloket OR oruvail OR powergel OR solpaflex OR ketorolac OR acular ORtrometamol OR meclofenamic OR naproxen OR naprosyn OR niflumic OR actol OR flunir OR niflactol topico OR niflugel ORnifluril OR oxyphenbutazone OR californit OR diflamil OR otone OR tanderil OR piketoprofen OR calmatel OR triparsean ORpiroxicam OR feldene OR pranoprofen OR oftalar OR pranox OR suxibuzone OR danilon OR flamilon OR ufenamate OR fenazolOR flector OR benzydamine) TIABKY (25220)
3 1 OR 2 (32484)4 MESH DESCRIPTOR Administration Topical EXPLODE ALL TREES (2169)
119Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
5 (topical OR cutaneous OR dermal OR transcutaneous OR transdermal OR percutaneous OR skin OR massage ORembrocation OR gel OR ointment OR aerosol OR cream OR cregraveme OR lotion OR mouse OR foam OR liniment OR spray OR rubOR balm OR salve OR emulsion OR oil OR patch OR plaster)TIABKY (67940)
6 4 OR 5 (70486)7 MESH DESCRIPTOR Athletic Injuries EXPLODE ALL TREES (411)8 (strain OR sprain OR contusion OR distortion OR compression OR ldquosports injurrdquo OR ldquosoft tissue injurrdquo OR tendnitis OR
ldquomuscle painrdquo OR periarthritis OR epicondylitis OR tenosynovitis)TIABKY (9158)9 7 OR 8 (9448)
10 MESH DESCRIPTOR pain EXPLODE ALL TREES (29943)11 (pain OR analgesi)TIABKY (74815)12 10 OR 11 (80041)13 3 AND 6 AND 9 AND 12 (110)
Appendix 2 MEDLINE search strategy via Ovid (for 2015 update)
1 exp Anti-inflammatory Agents non-steroidal (162888)2 (bufexamac OR bufexine OR calmaderm OR ekzemase OR diclofenac OR solaraze OR pennsaid OR voltarol OR emugel OR
voltarene OR voltarol OR optha OR voltaren OR etofenamate OR afrolate OR algesalona OR bayro OR deiron OR etofen ORflexium OR flogoprofen OR rheuma-gel OR rheumon OR traumalix OR traumon OR zenavan OR felbinac OR dolinac OR flexfreeOR napageln OR target OR traxam OR fentiazac OR domureuma OR fentiazaco OR norvedan OR riscalon OR fepradinol ORdalgen OR flexidol OR cocresol OR rangozona OR reuflodol OR pinazone OR zepelin OR flufenamic OR dignodolin OR rheumaOR lindofluid OR sastridex OR lunoxaprofen OR priaxim OR flubiprofen OR fenomel OR ocufen OR ocuflur OR ldquoTrans ActLATrdquo OR tulip OR ibuprofen OR cuprofen OR ldquodeep relief rdquo OR fenbid OR ibu-cream OR ibugel OR ibuleve OR ibumousse ORibuspray OR ldquonurofen gelrdquo OR proflex OR motrin OR advil OR radian OR ralgex OR ibutop OR indomethacin OR indocin ORindospray OR isonixin OR nixyn OR ketoprofen OR tiloket OR oruvail OR powergel OR solpaflex OR ketorolac OR acular ORtrometamol OR meclofenamic OR naproxen OR naprosyn OR niflumic OR actol OR flunir OR niflactol topico OR niflugel ORnifluril OR oxyphenbutazone OR californit OR diflamil OR otone OR tanderil OR piketoprofen OR calmatel OR triparsean ORpiroxicam OR feldene OR pranoprofen OR oftalar OR pranox OR suxibuzone OR danilon OR flamilon OR ufenamate OR fenazolOR flector OR benzydamine)mp (558284)
3 1 OR 2 (664691)4 exp Administration Topical (69697)5 (topical OR cutaneous OR dermal OR transcutaneous OR transdermal OR percutaneous OR skin OR massage OR
embrocation OR gel OR ointment OR aerosol OR cream OR cregraveme OR lotion OR mouse OR foam OR liniment OR spray OR rubOR balm OR salve OR emulsion OR oil OR patch OR plaster)mp (1982395)
6 4 OR 5 (1997720)7 exp Athletic Injuries (21464)8 (strain OR sprain OR contusion OR distortion OR compression OR ldquosports injurrdquo OR ldquosoft tissue injurrdquo OR tendnitis OR
ldquomuscle painrdquo OR periarthritis OR epicondylitis OR tenosynovitis)mp (420133)9 7 OR 8 (439228)
10 Pain (113906)11 (pain OR analgesi)mp (585249)12 10 or 11 (585249)13 randomized controlled trialpt (401171)14 randomizedab (296222)15 placeboab (155341)16 drug therapyfs (1789858)17 randomlyab (207517)18 trialab (308477)19 groupsab (1318386)20 OR13-1921 3 AND 6 AND 9 AND 12 AND 20 (139)22 limit 21 to yr=ldquo2008-Currentrdquo (56)
120Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Appendix 3 EMBASE search strategy via Ovid (for 2015 update)
1 exp Anti-inflammatory Agents non-steroidal (452266)2 (bufexamac OR bufexine OR calmaderm OR ekzemase OR dicoflenac OR solaraze OR pennsaid OR voltarol OR emugel OR
voltarene OR voltarol OR optha OR voltaren OR etofenamate OR afrolate OR algesalona OR bayro OR deiron OR etofen ORflexium OR flogoprofen OR rheuma-gel OR rheumon OR traumalix OR traumon OR zenavan OR felbinac OR dolinac OR flexfreeOR napageln OR target OR traxam OR fentiazac OR domureuma OR fentiazaco OR norvedan OR riscalon OR fepradinol ORdalgen OR flexidol OR cocresol OR rangozona OR reuflodol OR pinazone OR zepelin OR flufenamic OR dignodolin OR rheumaOR lindofluid OR sastridex OR lunoxaprofen OR priaxim OR flubiprofen OR fenomel OR ocufen OR ocuflur OR ldquoTrans ActLATrdquo OR tulip OR ibuprofen OR cuprofen OR ldquodeep relief rdquo OR fenbid OR ibu-cream OR ibugel OR ibuleve OR ibumousse ORibuspray OR ldquonurofen gelrdquo OR proflex OR motrin OR advil OR radian OR ralgex OR ibutop OR indomethacin OR indocin ORindospray OR isonixin OR nixyn OR ketoprofen OR tiloket OR oruvail OR powergel OR solpaflex OR ketorolac OR acular ORtrometamol OR meclofenamic OR naproxen OR naprosyn OR niflumic OR actol OR flunir OR niflactol topico OR niflugel ORnifluril OR oxyphenbutazone OR californit OR diflamil OR otone OR tanderil OR piketoprofen OR calmatel OR triparsean ORpiroxicam OR feldene OR pranoprofen OR oftalar OR pranox OR suxibuzone OR danilon OR flamilon OR ufenamate OR fenazolOR flector OR benzydamine)mp (790435)
3 1 OR 2 (1108683)4 exp Administration Topical (68490)5 (topical OR cutaneous OR dermal OR transcutaneous OR transdermal OR percutaneous OR skin OR massage OR
embrocation OR gel OR ointment OR aerosol OR cream OR cregraveme OR lotion OR mouse OR foam OR liniment OR spray OR rubOR balm OR salve OR emulsion OR oil OR patch OR plaster)mp (3208844)
6 4 OR 5 (3208844)7 exp Athletic Injuries (24675)8 (strain OR sprain OR contusion OR distortion OR compression OR ldquosports injurrdquo OR ldquosoft tissue injurrdquo OR tendnitis OR
ldquomuscle painrdquo OR periarthritis OR epicondylitis OR tenosynovitis)mp (804322)9 7 OR 8 (825164)
10 Pain (216406)11 (pain OR analgesi)mp (971593)12 10 OR 11 (971593)13 clinical trialssh (841252)14 controlled clinical trialssh (389335)15 randomized controlled trialsh (357169)16 double-blind proceduresh (118945)17 (clin adj25 trial)ab (331002)18 ((doubl or trebl or tripl) adj25 (blind or mask))ab (141726)19 placeboab (203390)20 randomab (906176)21 OR13-20 (1731380)22 3 AND 6 AND 9 AND 10 AND 19 (439)23 limit 22 to yr=ldquo2008-Currentrdquo (204)
Appendix 4 Summary of outcomes efficacy
Study ID Treatment Clinical response Other response
Airaksinen 1993 (1) Ketoprofen gel 2 x 5 g (125mg) daily n = 29(2) Placebo gel n = 27
PGE rdquoimprovedldquo at 7 days(1) 2429(2) 1427
No additional data
121Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
kermark 1990 (1) Indomethacin spray 1 (El-metacin) 3-5 x 05-15 mL dailyn = 23(2) Indomethacin capsules 3 x 25mg daily n = 23(3) Placebo spray and capsules n= 24
No pain on palpation at 7 days(1) 1222(2) 522(3) 624
Patient assessment of improve-ment at 7 days (scale 0 to 100)(1) 57(2) 49(3) 30
Aoki 1984 (1) Piroxicam gel 5 3 or 4 x 1 gdaily n = 84(2) Indomethacin gel 1 3 or 4x 1 g daily n = 84(3) Placebo gel n = 84
PGE (5 point) rdquobetter or muchbetterldquo at 7 days(1) 5672(2) 4164(3) 3367
Pain on movement rdquoreducedldquo orrdquodisappearedldquo at 7 days(1) 4861(2) 3860(3) 3563
Auclair 1989 (1) Niflumic acid gel 25 3 x 5g daily n = 117(2) Placebo gel n = 110
PGE (5 point) rdquogood or verygoodldquo at 7 days(1) 69117(2) 54110
Pain on palpation rdquoimprovedldquo at7 days(1) 69117(2) 53110
Billigmann 1996 (1) Ibuprofen microgel 5 3 x200 mg daily n = 80(2) Placebo gel n = 80
Complete remission(1) 2580(2) 1080
Improvement in pain with move-ment of 20 at 7 days(1) 6580(2) 5580
Campbell 1994 (1) Ibuprofen cream 5 (Proflex)4 x 4rdquo daily n = 26(2) Placebo cream n = 25
Improvement in walking ability (4point) at 7 days(1) 2126(2) 1925
No additional data
Chatterjee 1977 (1) Benzydamine HCl cream 33 x daily n = 25(2) Placebo cream n = 25
Pain on movement ldquoabsentslightrdquoat 6 days(1) 2125(2) 1225
Tenderness with pressure ldquoabsentslightrdquo at 6 days(1) 2125(2) 1225
Costantino 2011 (1) DHEP-heparin plaster (Flec-torparin) x 1 daily n = 142(2) DHEP plaster (Flector) x 1daily n = 146(3) Placebo plaster n = 142
No responder analysis for efficacyOverall treatment efficacy not re-ported
Mean reduction from baseline forPI on movement at 3 days (fromgraph)(1) 24 mm(2) 19 mm(3) 14 mm
Coudreuse 2010 (1) DHEP-heparin plaster (Flec-torparin) n = 120(2) Placebo n = 120
PGE ldquogoodrdquo or ldquoexcellentrdquo at 7days(1) 99117(2) 82116These are results for physician-re-ported judgement - not partici-pant
Mean reduction from baseline forPI on movement over 6 hours(1) about 30(2) about 20Overall greater in DHEP-heparingroup than placebo over 7 days
122Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Curioni 1985 (1) Ibuproxam n = 20(2) Ketoprofen n = 20(3) Etofenamate n = 20
Resolution of symptoms by 7 days(1) 1520(2) 1320(3) 1320
PGE ldquogoodrdquo or ldquoexcellentrdquo at 10days(1) 1920(2) not reported(3) 1620
Diebshlag 1990 (1) Ketorolac gel 2 3 x 3 g dailyn = 13(2) Etofenamate gel 5 3 x 3 gdaily n = 12(3) Placebo gel n = 12
Improvement in pain at 7 days(1) 1213(2) 1012(3) 912
No additional data
Dreiser 1988 (1) Ibuprofen cream 5 3 x 4 cmdaily n = 32 (3 x 10 cm for largejoints)(2) Placebo cream n = 32
PGE ldquoimprovementrdquo or ldquocom-plete relief rdquo at 7 days(1) 2632(2) 1232
(1) significantly better than (2)for mean improvement in sponta-neous pain movement pain restpain tenderness to pressure (VAS)
Dreiser 1989 (1) Ketoprofen gel 25 2 x 5 cmdaily n = 30(2) Placebo gel n = 30
PGE (3 point) ldquobetterrdquo at 7 days(1) 1830(2) 530
(1) significantly better than (2) formean improvement in pain (restand movement) (VAS)
Dreiser 1990 (1) Niflumic acid gel 25 3 x 5g daily n = 30(2) Placebo gel n = 30
PGE (4 point) ldquocuredrdquo or ldquoim-provedrdquo at 7 days(1) 2330(2) 1030
(1) significantly better than (2) formean improvement in pain (VAS)
Dreiser 1994 (1) Flurbiprofen plaster 2 x 40 mgdaily n = 65(2) Placebo plaster n = 66
PGE (4 point) ldquogoodrdquo or ldquoverygoodrdquo at 7 days(1) 4865(2) 4166
(1) significantly better than (2)for mean improvement in sponta-neous pain but not pain on move-ment or palpation (VAS)
Fioravanti 1999 (1) DHEP lecithin gel 3 x 5 g (=65 mg) daily n = 50(2) DHEP gel 3 x 5 g (= 65 mg)daily n = 50
PGE (4 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 10 days(1) 3550(2) 3550
(1) significantly better than (2)for mean improvement in spon-taneous pain at 7 days but notfor pain on movement at 10 days(VAS)
Fujimaki 1985 (1) Piroxicam gel 05 3 or 4 x 1g daily n = 92(2) Indomethacin gel 1 3 or 4 x1 g daily n = 90(3) Placebo gel n = 89
PGE (5 point) ldquobetterrdquo or ldquomuchbetterrdquo at end of treatment at 14days(1) 4483(2) 4482(3) 4082
No additional data
Gallacchi 1990 (1) Diclofenac hy-droxyethylpyrrolidine gel 1 4 x2 g daily n = 25 (Flector gel)(2) Diclofenac sodium 1 4 x 2 g
PGE (5 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 14 days(1) 1925(2) 1925
No significant difference betweengroups for pain on applied pres-sure at 7 and 14 days
123Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
daily n = 25 (Voltaren Emugel)
Gonzaacutelez de Vega 2013 (1) Traumeel gel 3 x 2 g daily n =140(2) Traumeel ointment n = 143(3) Diclofenac gel 1 3 x 2 g dailyn = 137
Pain-free at 7 days(1) 7140(3) 8137PGE (5 point) ldquogoodrdquo or ldquoverygoodrdquo(1) 128140(3) 127137
Normal function (5 point) scoreof 0 or 1 at 14 days(1) 133140(3) 131137
Governali 1995 (1) Ketoprofen gel 5 3 x 2-3 gdaily n = 15(2) Ketoprofen cream 1 3 x 2-3 g daily n = 15
PGE (5 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 7 days(1) 1415(2) 415
No additional data
Gualdi 1987 (1) Flunaxaprofen gel 2 x 3-5 cmdaily n = 30(2) Ketoprofen gel 2 x 3-5 cmdaily n = 30
No dichotomous data No significant difference betweengroups for pain on movement at 7days
Haig 1986 (1) Benzydamine cream 3 6 xdaily n = 21(2) Placebo cream n = 22
Pain on movement ldquoimprovedrdquo by6 days(1) 1821(2) 1322
No additional data
Hoffmann 2012 (1) DHEP-heparin plaster (Flec-torparin) x 1 daily n = 121(2) DHEP (Flector) plaster x 1daily n = 115(3) Placebo plaster n = 118
PGE (5 point) ldquogoodrdquo or ldquoexcel-lentrdquo over 80 in all groups
Mean reduction from baseline inPI (100 mm VAS)Day 3(1) 191 mm(2) 114 mm(3) 52 mmDay 8(1) 443 mm(2) 372 mm(3) 296 mm
Hofman 2000 (1) Diclofenac sodium gel 1 4x 2 cm daily n = 69(2) Lysine clonixinate gel 5 4 x2 cm (225 mg) daily n = 73
PGE (3 point) at 8 days ldquogoodrdquo(1) 3869(2) 3673
No significant difference betweentreatments for any pain outcomes
Hosie 1993 (1) Felbinac foam 3 3 x 2 g daily+ placebo tablets 3 x 1 daily n =140 (127 analysed for efficacy)(2) Ibuprofen tablets 3 x 400 mgdaily + placebo foam 3 x 2 g dailyn = 147 (134 analysed for efficacy)
Pain on movement ldquononerdquo orldquomildrdquo at 7 days(1) 81127(2) 96133
Spontaneous pain ldquononerdquo orldquomildrdquo at 7 days(1) 99127(2) 108134
124Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Jenoure 1997 (1) DHEP plaster (Tissugel) 2 xdaily n = 44(2) Placebo plaster 2 x daily n =41
Baseline pain in two groups notbalanced and data in table and fig-ure do not agree so efficacy out-comes not used
No additional data
Joussellin 2003 (1) DHEP plaster (Flector Tis-sugel 1) 1 x daily n = 68(2) Placebo plaster 1 x daily n =66
PGE (4 point) ldquoexcellentrdquo at 7 days(1) 3668(2) 2466ge 30 reduction in PI at 7 days(1) 2568(2) 1166
(1) significantly better than (2) formean pain on movement at 6 days
Julien 1989 (1) Ketoprofen gel 25 2 x 5 cm(= 50 mg) daily n = 30(2) Placebo gel n = 30
PGE (4 point) ldquorecoveredrdquo at 7days(1) 1830(2) 630
PGE (4 point) ldquorecoveredrdquo or ldquoim-provedrdquo at 7 days(1) 2530(2) 1330
Klainguti 2010 (1) DHEP-heparin plaster (Flec-torparin) x 1 daily n = 62(2) DHEP plaster (Flector) x 1daily n = 61(3) Placebo n = 59
Overall treatment efficacy (partic-ipant and investigator) (5 point)ldquogoodrdquo or ldquoexcellentrdquo at 3 days(1) 5662(2) not reported(3) 4659
Mean reduction in PI at day 3(from graph)(1) 33 mm(2) Not reported(3) 24 mm
Kockelbergh 1985 (1) Ketoprofen gel 25 2 x 5 cm(= 15 mg) daily n = 38(2) Placebo gel n = 36
PGE (3 point) ldquogoodrdquo at 7 days(1) 3038(2) 2236
(1) and (2) slightly better than (3)for mean spontaneous pain at 7days
Kuehl 2011 (1) DETP 13 plaster 2 x dailyn = 207(2) Placebo plaster n = 211
No dichotomous data Percentage reduction from base-line at last application(1) 73(2) 62
Li 2013 (1) DHEP plaster (Flector) 2 xdaily n = 192(2) Placebo plaster 2 x daily n =192
ge 50 pain reduction at day 7(posthoc analysis)(1) 173192(2) 114192
Overall treatment efficacy (5point) ldquogoodrdquo or ldquoexcellentrdquo(1) 161192(2) 81192Participant and investigator ratingMean plusmn SD reduction in PI onmovement at 7 days(1) 538 plusmn 170(2) 370 plusmn 183
Linde 1985 (1) Benzydamine 3 cream 3 xdaily n = 50(2) Placebo gel n = 50
No pain on movement (walking)at 8 days(1) 3550(2) 4050
No additional data
125Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Machen 2002 (1) Ibuprofen gel 5 3 x daily n= 40(2) Placebo gel n = 41
PGE (5 point) ldquomarked improve-mentrdquo or ldquocomplete clearancerdquo at7 days(1) 2540(2) 941
Clinically meaningful (ge 30 mm)pain relief at day 7(1) 3040(2) 1641
Mahler 2003 (1) DHEP + lethicin gel 3 x 5 gdaily n = 52(2) DHEP gel 3 x 5 g daily n = 48
PGE (4 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 10 days(1) 4952(2) 3948
Mean reduction in pain on move-ment at 3 and 10 days significantlygreater with (1) than (2)
Maziegraveres 2005b (1) Ketoprofen plaster 100 mg x1 daily n = 81(2) Placebo plaster n = 82
PGE (4 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 14 days(1) 7281(2) 6082
All mean efficacy measures im-proved more for (1) than (2) mostwere statistically significant
Maziegraveres 2005a (1) Ketoprofen plaster 100 mg x1 daily n = 87(2) Placebo plaster n = 85
PGE (4 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 14 days(1) 5087(2) 4185
All mean efficacy measures im-proved more for (1) than (2) mostwere statistically significant
McLatchie 1989 (1) Felbinac gel 3 3 x 3 cm dailyn = 118(2) Placebo gel n = 113
No dichotomous data Patient daily self-assessment formean pain on rest movement atnight interference with normaland leisure activities show betterefficacy for (1) than (2) from day2 (VAS)
Morris 1991 (1) Felbinac gel 3 3 x 1 cm dailyn = 41(2) Placebo gel n = 43
PGE (5 point) ldquogoodrdquo or ldquoverygoodrdquo at 7 days(1) 2341(2) 2743
(1) better than (2) for meanimprovement in symptoms andsporting function at 7 days
NCT01255423 (1) Diclofenac sodium gel 1 x 4daily n = 104(2) Placebo gel x 4 daily n = 100
No dichotomous data VAS (mean plusmn SD) at 72 hours(baseline PI not reported)(1) 374 plusmn 252(2) 388 plusmn 241
NCT01272934 (1) Diclofenac sodium gel 1 x 4daily n = 104(2) Placebo gel x 4 daily n = 100
No dichotomous data VAS (mean plusmn SD) at 72 hours(baseline PI not reported)(1) 256 plusmn 159(2) 612 plusmn 166
NCT01272947 (1) Diclofenac sodium gel 1 x 4daily n = 104(2) Placebo gel x 4 daily n = 100
No dichotomous data VAS (mean plusmn SD) at 24 hours(baseline PI not reported)(1) 331 plusmn 214(2) 654 plusmn 169
126Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Noret 1987 (1) Ketoprofen gel 25 2 x 5 cm(75 mg) daily n = 48(2) Placebo gel n = 45
PGE (4 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 8 days(1) 3948(2) 945
Decrease in mean spontaneouspain significantly greater in (1)than (2) by 3 days
Parrini 1992 (1) Ketoprofen foam 15 3 x 2 g(200 mg) daily n = 83(2) Placebo foam n = 86
No dichotomous data Mean pain on movement andpressure significantly decreased by7 days in (1) compared with (2)
Picchio 1981 (1) Ibuprofen gel 10 3 x daily n = 20(2) Ketoprofen gel 1 3 x daily n= 20
No pain on movement at 8 days(1) 320(2) 020
Spontaneous pain ldquononerdquo at 8days(1) 620(2) 020
Predel 2004 (1) Diclofenac sodium plaster 2 xdaily (140 mgplaster) n = 60(2) Placebo plaster n = 60
PGE (4 point) ldquogoodrdquo ldquoexcellentrdquoat 7 days(1) 5560(2) 560
(1) better than (2) for reductionin tenderness pain and speed ofpain reduction
Predel 2012 (1) Diclofenac gel (VoltarenEmulgel 232) 2 x 5 cm daily n= 80(2)Diclofenac gel (Voltaren Emulgel232 ) 3 x 5 cm daily n = 80(3) Placebo gel n = 82
ge 50 red in PI on movement at5 days(1) 5780(2) 5980(3) 1782
PGE efficacy (5 point) ldquogoodrdquo orldquovery goodrdquo at 8 days(1) 6880(2) 7380(3) 2482
Predel 2013a (1) Diclofenac 4 spray gel 4 or5 sprays 3 x daily (96-120 mg di-clofenac sodium) n = 118(2) Placebo spray gel n = 114
None or slight PI on movement at7-8 days(1) 111118(2) 93114
None or slight PI on movement at3-4 days(1) 76118(2) 58114
Predel 2013b (1) Diclofenac gel (VoltarenEmulgel) 4 x 2 g daily n = 36(2) Placebo gel 4 x 2 g daily n =36
PGE (5 point) ldquogoodrdquo or ldquoexcel-lentrdquo(1) 3636(2) 736
ge 50 reduction in PI on move-ment after 48 hours(1) 3436(2) 336
Ramesh 1983 (1) Ibuprofen cream 5 3 or 4 x5-10 cm daily n = 40(2) Placebo cream n = 40
Pain on movement (4 point)ldquononerdquo or ldquoslightrdquo at 7 days(1) 2340(2) 2340
Physician global assessment at 10days ldquogoodrdquo(1) 2940(2) 1640
Rowbotham 2003 (1) Diclofenac epolamine plaster(Flector Tissuegel) 2 x daily n =191(2) Placebo plaster n = 181
Pain intensity le 210 for 2 daysor 4 consecutive evaluations by 7days(1) 75191(2) 48181
Mean pain on rest significantlybetter with (1) than (2) after 7 days
127Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Russell 1991 (1) Piroxicam gel 05 4 x 5 mgdaily n = 100(2) Placebo gel n = 100
PGE (4 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 8 days(1) 79100(2) 45100
Statistically greater reduction inmean pain on movement at 8 dayswith (1) than (2)
Saillant 1998 (1) DHEP plaster (Flector Tis-sugel 1) 1 x daily n = 70(2) Placebo plaster 1 x daily n =70
PGE (4 point) ldquoexcellentrdquo at 7 days(1) 4370(2) 2570
ge 30 decrease in PI at 7 days(1) 6470(2) 5070
Sanguinetti 1989 (1) Felbinac gel 3 3 x daily n =42(2) Placebo gel n = 40
PGE ldquogoodrdquo or ldquovery goodrdquo at 7days(1) 3442(2) 1140
(1) better than (2) by 2 days
Sinniger 1981 (1) Fentiazac cream 5 2 or 3 xdaily n = 10(2) Placebo cream n = 10
Complete pain relief within 10days(1) 710(2) 110
Improvement in active pain onmovement at 5 days(1) 67(2) 32
Spacca 2005 (1) DHEP lecithin gel (Effigel) 3x 5 g daily n = 79(2) Placebo gel n = 76
No dichotomous data Mean pain scores improved morerapidly in (1) than (2) - statisticallysignificant at 3 and 6 days
Sugioka 1984 (1) Piroxicam gel 05 3 or 4 x 1g daily n = 183(2) Indomethacin gel 1 3 or 4 x1 g daily n = 183
PGE (5 point) ldquobetterrdquo or ldquomuchbetterrdquo at 14 days(1) 85175(2) 55165
Pain on movement ldquoreducedrdquo orldquodisappearedrdquo at 7 days(1) 77175(2) 63165
Thorling 1990 (1) Naproxen gel 10 2-6 x dailyn = 60(2) Placebo gel n = 60
PGE (5 point) ldquogoodrdquo or ldquoverygoodrdquo at 7 days(1) 3860(2) 2760
Participants using naproxen im-proved more rapidly and had sig-nificantly lower severity scores byday 3
Tonutti 1994 (1) Ketoprofen gel 5 3 x 2-3 gdaily n = 15(2) Etofenamate gel 5 3 x 2-3g n = 15
PGE (4 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 7 days(1) 1015(2) 1115
Significant reductions in pain onmovement by 7 days in bothgroups
Vecchiet 1991 (1) Meclofenamic acid gel 5 2 x10 cm daily (2 g) n = 30(2) Placebo n = 30
PGE (4 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 10 days(1) 3030(2) 1930
(1) significantly better than (2)for mean improvement in sponta-neous pain movement pain func-tional restriction
Whitefield 2002 (1) Ibuprofen gel 5 + placebotablet 3 x daily n = 50(2) Ibuprofen 400 mg tablet +placebo gel 3 x daily n = 50
Participant satisfied at 7 days(1) 3050(2) 3650
ldquoCompletely betterrdquo at 14 days(1) 2450(2) 3050
128Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
DHEP diclofenac epolamine HCl hydrochloride n number PGE participant global evaluation PI pain intensity SD standarddeviation VAS visual analogue scale
Appendix 5 Summary of outcomes adverse events and withdrawals
Study ID Treatment Local AEs Systemic AEs Serious AEs Withdrawals
Airaksinen 1993 (1) Ketoprofen gel 2x 5 g (125 mg) dailyn = 29(2) Placebo gel n =27
(1) 529(2) 427
(1) 129 (nausea af-ter paracetamol)(2) 027
None AE noneOther nonereported
kermark 1990 (1)Indomethacin spray1 (Elmetacin) 3-5 x 05-15 mL dailyn = 23(2) Indomethacincapsules 3 x 25 mgdaily n = 23(3) Placebo sprayand capsules n = 24
(1) 422(2) 022(3) 024
No usable data -reported for eventsnot participants
None reported AE (1) 1 (2) 1 (3)0Lost to follow-up(1) 1 (2) 2 (3) 3
Aoki 1984 (1) Piroxicam gel5 3 or 4 x 1 gdaily n = 84(2) Indomethacingel 1 3 or 4 x 1 gdaily n = 84(3) Placebo gel n =84
(1) 179(2) 270(3) 274
None None reported AE none23 excluded for pro-tocol violations (1)7 (2) 7 (3) 926 withdrew for rea-sons unrelated totreatment (1) 5 (2)13 (3) 8
Auclair 1989 (1) Niflumic acidgel 25 3 x 5 gdaily n = 117(2) Placebo gel n =110
All AEs(1) 5123(2) 6116Most commonly cu-taneous eruptions
No usable data None reported AE (1) 1123 (2) 011626 excluded fromefficacy analysis fornot meeting entrycriteria and protocolviolations
Billigmann 1996 (1) Ibuprofen mi-crogel 5 3 x 200mg daily n = 80(2) Placebo gel n =
(1) 1180(2) 480
None reported None reported AE (1) 280 (aller-gic rash dermatitis)No reason given (1)380 (2) 580
129Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
80 Symptom-free (1)180 (2) 180
Campbell 1994 (1) Ibuprofen cream5 (Proflex) 4 x 4rdquodaily n = 26(2) Placebo cream n= 25
No data (1) 126 (headache)(2) 025
No data AE noneExclusions 493 presented late 2missing forms 1 ap-peared twice 43 didnot return diaries
Chatterjee 1977 (1) BenzydamineHCl cream 3 3 xdaily n = 25(2) Placebo cream n= 25
None None None AE none1 participant lostto follow-up (groupnot reported)
Costantino 2011 (1) DHEP-hep-arin plaster (Flector-parin) x 1 daily n =142(2) DHEP plaster(Flector) x 1 daily n= 146(3) Placebo plastern = 142
Possibly or probablydrug-related(1) 2142(2) 1145(3) 6142
All mild and goneby 14 days ex-cept 1 (moderate) inplacebo group
Not reportedNo systemic GIevents
None AE noneOther(1) 0142(2) 2146 (proto-col deviation lost tofollow-up)(3) 2142 (lost tofollow-up)
Coudreuse 2010 (1) DHEP-hep-arin plaster (Flector-parin) n = 120(2) Placebo n = 120
(1) 1117(2) 7116
All AEs mild ormoderate resolvedspontaneously
(1) 1117(2) 0116
None AE(1) 1117 (increasedoedema)(2) 0116Other(1) 6117 (recovery3 no reason 3)(2) 3116 (recovery1 no reason 2)
Curioni 1985 (1) Ibuproxam n =20(2) Ketoprofen n =20(3) Etofenamate n =20
None None None AE noneOther none
Diebshlag 1990 (1) Ketorolac gel2 3 x 3 g daily n =13(2) Etofenamate gel5 3 x 3 g daily n =12
(1) 113(2) 112(3) 012
None None AE none1 ketorolac partic-ipant did not at-tend 15 day follow-up due to car acci-
130Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
(3) Placebo gel n =12
dent
Dreiser 1988 (1) Ibuprofen cream5 3 x 4 cm dailyn = 32 (3 x 10 cmfor large joints)(2) Placebo cream n= 32
No usable data None Not reported AE none4 placebo partici-pants lost to follow-up
Dreiser 1989 (1) Ketoprofen gel25 2 x 5 cm dailyn = 30(2) Placebo gel n =30
(1) 030(2) 230
None None reported AE (2) 230 (intol-erance)LoE (1) 130 (2) 130
Dreiser 1990 (1) Niflumic acidgel 25 3 x 5 gdaily n = 30(2) Placebo gel n =30
(1) 030(2) 330
None None AE (2) 130 (ery-thema)Exclusion 1 from(2) from efficacyanalysis for inade-quate baseline pain
Dreiser 1994 (1)Flurbiprofen plaster2 x 40 mg daily n =65(2) Placebo plastern = 66
(1) 265(2) 066
None None AE 0(1) 165 excludedfrom efficacy analy-sis for protocol vio-lation(2) 266 (1 LoE 1cured)
Fioravanti 1999 (1) DHEP lecithingel 3 x 5 g (= 65 mg)daily n = 50(2) DHEP gel 3 x 5g (= 65 mg) daily n= 50
(1) 050(2) 150
No data None reported AE noneOther none
Fujimaki 1985 (1) Piroxicam gel 05 3 or 4 x 1 g dailyn = 92(2) Indomethacingel 1 3 or 4 x 1 gdaily n = 90(3) Placebo gel n =89
(1) 183(2) 582(3) 282
(1) 083(2) 182 (nauseaand vomiting)(3) 082
None AE (1) 0 (2) 4 (3)0Unknown reasons(1) 2 (2) 1Did not return af-ter 1st visitirregularvisits (1) 6 (2) 6(3) 7
Gallacchi 1990 (1) Diclofenac hy-droxyethylpyrro-lidine gel 1 4 x 2
No AEs None None AE noneOther none
131Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
g daily n = 25 (Flec-tor gel)(2) Di-clofenac sodium 14 x 2 g daily n = 25(Voltaren Emugel)
Gonzaacutelez de Vega2013
(1) Traumeel gel 3 x2 g daily n = 140(2) Traumeel oint-ment n = 143(3) Diclofenac gel1 3 x 2 g daily n =137
No usable data
Infrequent mild tomoderate
No usable data None AE noneOther(1) 11140(2) 12137
Governali 1995 (1) Ketoprofen gel5 3 x 2-3 g dailyn = 15(2) Ketoprofencream 1 3 x 2-3 gdaily n = 15
No side effects None None AE noneOther none
Gualdi 1987 (1) Flunaxaprofengel 2 x 3-5 cm dailyn = 30(2) Ketoprofen gel 2x 3-5 cm daily n =30
(1) 130(2) 330
No data None reported AE noneOther none
Haig 1986 (1) Benzydaminecream 3 6 x dailyn = 21(2) Placebo cream n= 22
No AEs reported None None reported AE none reportedOther no data
Hoffmann 2012 (1) DHEP-hep-arin plaster (Flector-parin) x 1 daily n =121(2) DHEP (Flector)plaster x 1 daily n =115(3) Placebo plastern = 118
Most AEs were mi-nor local reactions(eg pruritus anderythema) in areaof plaster of mildto moderate inten-sity and resolvedwithout interrupt-ing treatment
No treatment-related systemic AEsrecorded
(2) 1 participanthad 3 SAEs nonejudged related tostudy medication
AE noneExclusions(1) 5121(2) 10115(3) 7118Excluded fromper protocol analysisdue to poor compli-ance or personal de-cision
Hofman 2000 (1) Diclofenacsodium gel 1 4 x2 cm daily n = 69(2) Lysine clonixi-
(1) 158(2) 161
None None AE noneLoE (1) 9 (2) 8
132Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
nate gel 5 4 x 2cm (225 mg) dailyn = 73
Hosie 1993 (1) Felbinac foam3 3 x 2 g daily+ placebo tablets 3x 1 daily n = 140(127 analysed for ef-ficacy)(2) Ibupro-fen tablets 3 x 400mg daily + placebofoam 3 x 2 g daily n= 147 (134 analysedfor efficacy)
(1) 1127(2) 3134
GI events (1) 14127 (2) 11134For (1) more mildnone definitely drugrelated for(2) definitely relatedto study drug
None AE noneExclusions (1) 13(2) 13 did not returnfor 7 day follow-up
Jenoure 1997 (1) DHEP plaster(Tissugel) 2 x dailyn = 44(2) Placebo plaster 2x daily n = 41
(1) 144(2) 141
No data None reported AE none reportedOther nonereported
Joussellin 2003 (1) DHEPplaster (Flector Tis-sugel 1) 1 x dailyn = 68(2) Placebo plaster 1x daily n = 66
(1) 168 (pruritus)(2) 366 (pruritus 2burning 1)
All AEs mild ormoderate
(1) 168 (allergic re-action)(2) 066
None reported AE(1) 066(2) 166Other(1) 366(2) 266
Julien 1989 (1) Ketoprofen gel25 2 x 5 cm (= 50mg) daily n = 30(2) Placebo gel n =30
(1) 130(2) 030
Not reported None AE noneOther none
Klainguti 2010 (1) DHEP-hep-arin plaster (Flector-parin) x 1 daily n =62(2) DHEP plaster(Flector) x 1 daily n= 61(3) Placebo n = 59
(1) 062(2) 061(3) 159All AEs mild in na-ture and resolvedspontaneously
(1) 165 (facial in-fection)(2) 161 (abdominalpain)(3) 059
None AE noneOther(1) 365(2) 161(3) 359
Kockelbergh 1985a (1) Ketoprofen gel25 2 x 5 cm (= 15mg) daily n = 38(2) Placebo gel n =
(1) 138(2) 126
Not reported None AE noneOther none
133Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
36
Kuehl 2011 (1) DETP 13plaster 2 x daily n =207(2) Placebo plastern = 211
(1) 16207(2) 12211
(1) 15207(2) 23211
None AE(1) 4207(2) 9211LoE(1) 21207(2) 25211Other(1) 58207(2) 62211
Li 2013 (1) DHEP plaster(Flector) 2 x daily n= 192(2) Placebo plaster 2x daily n = 192
(1) 4192(2) 3192
(1) 10192(2) 4192
None AE(1) 2192(2) 0192Other(1) 5192(2) 1192
Linde 1985 (1) Benzydamine3 cream 3 x dailyn = 50(2) Placebo gel n =50
(1) 440(2) 241
None None AE none(1) 6 (2) 6 excludedfrom 1st assessment(1) 3 (2) 4 ex-cluded from final as-sessment
Machen 2002 (1) Ibuprofen gel5 3 x daily n = 40(2) Placebo gel n =41
(1) 440(2) 241
None None AE none(1) 1 LoE 1 proto-col violation(2) 4 LoE
Mahler 2003 (1) DHEP + lethicingel 3 x 5 g daily n =52(2) DHEP gel 3 x 5g daily n = 48
(1) 152(2) 048
(1) 152(2) 048
None AE none5 lost to follow-up
Maziegraveres 2005b (1) Ketoprofen plas-ter 100 mg x 1 dailyn = 81(2) Placebo plastern = 82
At 21 days(1) 1281(2) 682
(1) 1381(2) 1482
None AE (1) 381(1) 781 (1 LoE 6cured)(2) 782 (5 LoE 2cured)
Maziegraveres 2005a (1) Ketoprofen plas-ter 100 mg x 1 dailyn = 87(2) Placebo plastern = 85
At 21 days(1) 2987(2) 2785
(1) 1187(2) 785
None AE (1) 987 (2) 685(1) 687 (2 LoE 4cured)(2) 585 (4 LoE 1cured)
134Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
McLatchie 1989 (1) Felbinac gel 33 x 3 cm daily n =118(2) Placebo gel n =113
(1) 3118(2) 2113Mild transient localirritation
None reported None AE noneOther none
Morris 1991 (1) Felbinac gel 33 x 1 cm daily n =41(2) Placebo gel n =43
None None None AE none(1) 4 (protocol vio-lations)(2) 1 (lost to follow-up)Exclusions 11 fromefficacy analysis be-cause evaluated by4 different investi-gators
NCT01255423 (1) Diclofenacsodium gel 1 x 4daily n = 104(2) Placebo gel x 4daily n = 100
(1) 1104(2) 3102
Total AE(1) 11104(2) 8102
None None
NCT01272934 (1) Diclofenacsodium gel 1 x 4daily n = 104(2) Placebo gel x 4daily n = 100
(1) 1102(2) 0103
Total AE (excludingSAE)(1) 6102(2) 3103
(1) 0102(2) 1103 (rupturedligaments in wrist)
AE see SAE
NCT01272947 (1) Diclofenacsodium gel 1 x 4daily n = 104(2) Placebo gel x 4daily n = 100
None (1) 2104(2) 2100
None None
Noret 1987 (1) Ketoprofen gel25 2 x 5 cm (75mg) daily n = 48(2) Placebo gel n =45
(1) 151(2) 047
None reported Not reported AE (1) 151 (skinallergy)(1) 1 LoE 1 unre-lated to trial(2) 1 LoE 1 unre-lated to trial
Parrini 1992 (1) Ketopro-fen foam 15 3 x 2g (200 mg) daily n= 83(2) Placebo foam n= 86
None None None AE noneOther none
135Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Picchio 1981 (1) Ibuprofen gel10 3 x daily n =20(2) Ketoprofen gel1 3 x daily n = 20
None None None AE noneOther not reported
Predel 2004 (1) Diclofenacsodium plaster 2 xdaily (140 mgplas-ter) n = 60(2) Placebo plastern = 60
12 participants ex-pe-rienced 16 mild AEswith no differencesbetween groups
None None AE (1) 160Other none
Predel 2012 (1) Diclofenac gel(Voltaren Emulgel232) 2 x 5 cmdaily n = 80(2) Diclofenac gel(Voltaren Emulgel232) 3 x 5 cmdaily n = 80(3) Placebo gel n =82
(1) 080(2) 180(3) 182All mild to moder-ate
(1) 280(2) 180(3) 382
None AE(1) 080(2) 080(3) 182Other(1) 180(2) 280(3) 282(protocol violationslost to follow-up)
Predel 2013a (1) Diclofenac 4spray gel 4 or 5sprays 3 x daily (96-120 mg diclofenacsodium) n = 118(2) Placebo spraygel n = 114
(1) 1118(2) 4114
(1) 5118(2) 4114All AEs mild re-versible
None AE(1) 1118(2) 1114Other(1) 3118(2) 43114
Predel 2013b (1) Diclofenac gel(Voltaren Emulgel)4 x 2 g daily n = 36(2) Placebo gel 4 x 2g daily n = 36
None (1) 036(2) 136
None AE noneOther none
Ramesh 1983 (1) Ibuprofen cream5 3 or 4 x 5-10 cmdaily n = 40(2) Placebo cream n= 40
(1) 140(2) 140
None reported Not reported AE (1) 140 (2) 140Other none
Rowbotham 2003 (1) DHEP plaster(Flector Tissuegel) 2x daily n = 191(2) Placebo plastern = 181
(1) 27191 (pruritis14)(2) 31181 (pruritis21)
(1) 21191(2) 22181
None re-ported (ldquovast major-ity mildrdquo)
AE none(1) 3191 (2) 4181(did not finish trialand complete dailydiaries)
136Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Russell 1991 (1) Piroxicam gel 05 4 x 5 mg daily n= 100(2) Placebo gel n =100
(1) 4102(2) 10102
GI or CNS events(1) 4 (2) 7Any AE (1) 7102(2) 15102
None reported AE (1) 1102 (2) 8102Other (1) 6 LoE 1ldquootherrdquo(2) 42 LoEExclusions7 did not complywith study medica-tion schedule 6 lostto follow-up 1 pro-tocol violation
Saillant 1998 (1) DHEPplaster (Flector Tis-sugel 1) 1 x dailyn = 70(2) Placebo plaster 1x daily n = 70
None None None AE noneOther(1) 570(2) 570
Sanguinetti 1989 (1) Felbinac gel 33 x daily n = 42(2) Placebo gel n =40
(1) 342(2) 140
None None reported AE noneOther nonereported
Sinniger 1981 (1) Fentiazac cream5 2 or 3 x daily n= 10(2) Placebo cream n= 10
ldquoNo untoward sideeffectsrdquo
None None AE noneOther nonereported
Spacca 2005 (1) DHEP lecithingel (Effigel) 3 x 5 gdaily n = 79(2) Placebo gel n =76
ldquoNo signsof cutaneous irrita-tion or sensitisationobservedrdquo
No AEs observed None AE noneOther nonereported
Sugioka 1984 (1) Piroxicam gel 05 3 or 4 x 1 g dailyn = 183(2) Indomethacingel 1 3 or 4 x 1 gdaily n = 183
(1) 5178(2) 26179
None reported None reported AE none reportedExclusions due toprotocol violations(1) 8 (2) 18Withdrawals(1) 11 (2) 12
Thorling 1990 (1) Naproxen gel10 2-6 x daily n =60(2) Placebo gel n =60
(1) 160(2) 060
None None AE none(1) 1 LoE 1 proto-col violation(2) 1 participant re-quest
137Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Tonutti 1994 (1) Ketoprofen gel5 3 x 2-3 g dailyn = 15(2) Etofenamate gel5 3 x 2-3 g n =15
None No AEs attributableto the medication
None AE NoneLoE (1) 1 (2) 2
Vecchiet 1991 (1) Meclofe-namic acid gel 5 2x 10 cm daily (2 g)n = 30(2) Placebo n = 30
Tolerabil-ity excellent or goodin nearly all partici-pants
No data None AE none reported(2) 5 lost to follow-up
Whitefield 2002 (1) Ibuprofen gel5 + placebo tablet3 x daily n = 50(2) Ibuprofen 400mg tablet + placebogel 3 x daily n = 50
No data 6 AEs reportednone judged relatedto study medication
None reported AE noneRecovered (1) 3 (2)2LoE (2) 1Lost to follow-up(1) 1 (2) 1
AE adverse event CNS central nervous system DHEP diclofenac epolamine GI gastrointestinal HCl hydrochloride LoE lackof efficacy n number SAE serious adverse event
Appendix 6 Concentration amount and frequency of dosing
Study Drug Concentration Quantity Frequency Estimated daily dose of topi-cal NSAID
Joussellin 2003 Diclofenac 1 Plaster 1 180 mg epolamine salt 140 mgNa salt
Li 2013 Diclofenac 1 Plaster 2 360 mg epolamine salt 280 mgNa salt
Rowbotham 2003 Diclofenac 1 Plaster 2 360 mg epolamine salt 280 mgNa salt
Saillant 1998 Diclofenac 1 Plaster 1 180 mg epolamine salt 140 mgNa salt
Coudreuse 2010 Diclofenac 1 Plaster 1 180 mg epolamine salt 140 mgNa salt
Klainguti 2010 Diclofenac 1 Plaster 1 180 mg epolamine salt 140 mgNa salt
138Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Predel 2004 Diclofenac - Plaster 2 280 mg Na salt
Predel 2012 Diclofenac 232 5 cm ribbon~ 2 g 2 or 3 92-138 mg (Na equiv) as di-ethylamine salt
Predel 2013b Diclofenac 116 2 g 4 92 mg (Na equiv) as diethy-lamine salt
Predel 2013a Diclofenac 4 4-5 sprays 3 96-120 mg Na salt
Costantino 2011 Diclofenac 1 Plaster 1 180 mg epolamine salt 140 mgNa salt
Fioravanti 1999 Diclofenac - 5 g 3 195 mg epolamine salt
Gallacchi 1990 Diclofenac 1 2 g 4 80 mg
Gonzaacutelez de Vega2013
Diclofenac 1 6 cm 2g 3 60 mg
Hoffmann 2012 Diclofenac 1 Plaster 1 180 mg epolamine salt 140 mgNa salt
Hofman 2000 Diclofenac 1 2 cm 4 probably 30-40 mg Na salt
Jenoure 1997 Diclofenac 1 Plaster 2 360 mg epolamine salt 280 mgNa salt
Mahler 2003 Diclofenac - 5 g 3 195 mg epolamine salt
NCT01255423 Diclofenac 1 - 4 -
NCT01272934 Diclofenac 1 - 4 -
NCT01272947 Diclofenac 1 - 4 -
Spacca 2005 Diclofenac - 5 g 3 195 mg epolamine salt
Campbell 1994 Ibuprofen 5 4 inch ribbon 4 Assume 800 mg
Dreiser 1988 Ibuprofen 5 4 cm ribbon(10 cm for largerjoints)
4 Assume up-800 mg
Ramesh 1983 Ibuprofen 5 5-10 cm ribbon 3-4 Assume 300-800 mg
Billigmann 1996 Ibuprofen 5 10 cm 4 g gel 3 600 mg
139Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Machen 2002 Ibuprofen 5 - 3 -
Picchio 1981 Ibuprofen 10 - 3 -
Whitefield 2002 Ibuprofen 5 - 3 -
Maziegraveres 2005b Ketoprofen - Plaster 1 100 mg
Maziegraveres 2005a Ketoprofen - Plaster 1 100 mg
Airaksinen 1993 Ketoprofen - 5 g 2 125 mg
Dreiser 1989 Ketoprofen 25 5 cm 2 100 mg
Julien 1989 Ketoprofen 25 5 cm 2 100 mg
Noret 1987 Ketoprofen 25 5 cm 75 g 2 375 mg
Curioni 1985 Ketoprofen No details 2 -
Governali 1995 Ketoprofen 5 2-3 g 3 300-450 mg
Governali 1995 Ketoprofen 1 2-3 g 3 60-90 mg
Gualdi 1987 Ketoprofen - 3-5 cm 2 -
Kockelbergh 1985 Ketoprofen 25 5 cm 2 100 mg
Parrini 1992 Ketoprofen 150 2 g 3 600 mg
Picchio 1981 Ketoprofen 1 - 3 -
Tonutti 1994 Ketoprofen 5 2-3 g 3 300-450 mg
kermark 1990 Indomethacin 1 05-15 mL 3-5 12-60 mg
Aoki 1984 Indomethacin 1 1 g 3-4 30-40 mg
Fujimaki 1985 Indomethacin 1 1 g 3-4 30-40 mg
Sugioka 1984 Indomethacin 1 1 g 3-4 30-40 mg
Aoki 1984 Piroxicam 05 1 g 3-4 15-20 mg
Fujimaki 1985 Piroxicam 05 1 g 3-4 15-20 mg
140Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Russell 1991 Piroxicam 05 5 mg 4 20 mg
Sugioka 1984 Piroxicam 05 1 g 3-4 15-20 mg
Chatterjee 1977 Benzydamine 3 - 3 -
Haig 1986 Benzydamine 3 - 6 -
Linde 1985 Benzydamine 3 - 3 -
Auclair 1989 Niflumic acid 25 10 cm 5 g 3 375 mg
Dreiser 1990 Niflumic acid 25 10 cm 5 g 3 375 mg
Curioni 1985 Ibuproxam gel 10 - 2 -
Curioni 1985 Etofenamate No details - 2 -
Diebshlag 1990 Etofenamate 5 3 g 3 450 mg
Tonutti 1994 Etofenamate 5 2-3 g 3 300-450 mg
Diebshlag 1990 Ketorolac 2 3 g 3 360 mg
Dreiser 1994 Flurbiprofen Patch 2 80 mg
Gualdi 1987 Flunoxaprofen - 3-5 cm 2 -
Hofman 2000 Lysine clonixinate 5 2 cm 4 90 mg
Hosie 1993 Felbinac 3 2 g 3 180 mg
McLatchie 1989 Felbinac 3 3 cm 3 -
Morris 1991 Felbinac 3 1 cm 3 -
Sanguinetti 1989 Felbinac 3 - 3 -
Sinniger 1981 Fentiazac 5 Varied according toinvolved areas
2-3 -
Thorling 1990 Naproxen 10 - 2-6 -
Vecchiet 1991 Meclofenamic acid 5 10 cm 4 g 2 400 mg
equiv equivalent Na sodium
141Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
F E E D B A C K
Query on formulations of topical NSAIDs particularly DMSO from Dr Chrubasik 11 April 2012
Summary
Dr Chrubasik highlighted this letter to the Editor postgradmedorgdoi103810pgm2011092482
DMSO [dimethyl sulphoxide] but also other additives eg nonivamide (which is a capsaicinoid added as drug enhancer) may contributeto the overall effect of topical NSAIDs Nonivamide certainly contributes to the analgesic effect and to adverse events (heat sensationburning pruritus etc) This has not been considered in the Cochrane review by Massey T Derry S Moore RA McQuay HJ TopicalNSAIDs for acute pain in adults but Dr Chrubasik believes should be done otherwise the effect size of the NSAID topicals is favoured
Reply
We have been asked by Dr Chrubasik to comment on a letter (Roth 2011) about the formulations of topical NSAIDs particularly howDMSO and other penetration enhancers can affect efficacy estimates or adverse event reporting in osteoarthritis It was suggested thatthe review of Topical NSAIDs for acute pain in adults did not consider this resulting in a bias towards the topical NSAIDThere are a number of points to be made here
1 Penetration enhancers are used in formulations of topical products to encourage local absorption through the skin and produce ahigh local concentration Topical NSAIDs use penetration enhancers and the result is high local concentration in joints for instancebut low systemic concentrations (Moore 2008) That is how they work Formulation is an important part of medicinal chemistry as awhole not just for topical agents
2 In our analysis of topical NSAIDs we were aware that a range of properties are or have been ascribed to the analgesia resultingfrom application of topical agents and which could contribute to overestimation of treatment effect of topical NSAID These includefeelings of heat or cold and even the act of rubbing itself For that reason we have chosen to include only double-blind studies wherethe placebo agent is identical to the active with the exception of course of the NSAID So heat cold rubbing and penetrationenhancers should be identical as best we can judge That leaves only the NSAID itself to provide any additional analgesic effect andit is that which we measure This is analogous for example to use of acupuncture say where the better studies show no differencebetween ldquotruerdquo acupuncture and ldquoshamrdquo acupuncture performed at nonspecific sites but better than non-treatment controls Theargument that we should only use high quality studies to evaluate evidence about pain interventions is well made
3 Overestimation of analgesic effect because of effects of enhancers themselves would be better made in direct comparisons oftopical and oral NSAIDs where local or even systemic effects would not be balanced in the oral study arm However our reviewconcentrated on placebo-controlled studies and had few studies with active controls Moreover the real test would be in chronicrather than acute conditions with long duration (12 week) outcomes using current best evidence rules (Moore 2010) includingimputation (Moore 2012) In their response to Rothrsquos letter the authors of the original review of products available in the USA showrather similar effect sizes of oral diclofenac and topical diclofenac with different penetration enhancers (Barthel 2011) in such studies
4 The Roth letter sought to differentiate between topical diclofenac preparations based on the penetration enhancers used Thatdifferent formulations may have different effect sizes is a fair point to make Two of the studies in our review of topical NSAIDs inacute conditions used diclofenac sodium 1 gel comparing it with either diclofenac epolamine gel (Gallacchi 1990 50 participants)or lysine clonixinate gel (Hofman 2000 142 participants) no difference between formulations was demonstrated It is difficult tomake any judgement for topical NSAIDs in acute conditions due to the relatively small number and particularly the small size ofstudies We did an analysis by drug and this showed that some topical NSAIDs were consistently beneficial irrespective offormulation while others had little or no efficacy This fits in with some theoretical considerations of molecular architecture andtissue penetration (Moore 2008) In chronic pain where there are larger studies and much more data we have considered formulation(Derry in preparation)
References
Roth SH Letter to the editor The importance of differentiating between topical NSAIDs Postgraduate Medicine 2011123251-2[103810pgm2011092482]Moore RA Derry S McQuay HJ Topical agents in the treatment of rheumatic pain Rheumatic Diseases Clinics of North America200834415-32
142Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Moore RA Eccleston C Derry S Wiffen P Bell RF Straube S et al ACTINPAIN Writing Group of the IASP Special Interest Groupon Systematic Reviews in Pain Relief Cochrane Pain Palliative and Supportive Care Systematic Review Group Editors ldquoEvidencerdquo inchronic pain - establishing best practice in the reporting of systematic reviews Pain 2010150386-9Moore RA Straube S Eccleston C Derry S Aldington D Wiffen P et al Estimate at your peril imputation methods for patientwithdrawal can bias efficacy outcomes in chronic pain trials using responder analyses Pain 2012153265-8Barthel HR Axford-Gatley RA Response to Roth Letter to the Editor Postgraduate Medicine 2011123253-4 [103810pgm2011092483]
Contributors
Feedback from Sigrun ChrubasikAuthors involved with responding Andrew Moore Sheena DerryFeedback Editor Kate Seers
Query submitted by Peter C Goslashtzsche 3 November 2015
Summary
Date of Submission 03-Nov-2015Name Peter C GoslashtzscheEmail Address pcgcochranedkAffiliation Nordic Cochrane CentreRole DirectorComment The authors found that the results were missing from 5900 patients Furthermore there was extreme heterogeneity in theirmeta-analyses eg I square was 92 for the diclofenac trials which were the most common ones and there was extreme funnel plotasymmetry with the largest trials showing the smallest effects (the authors didnrsquot show funnel plots but I constructed one for diclofenac)Moreover the trials were industry funded of relatively poor quality and the authors analysed published data not data from clinicalstudy reports and did not try to obtain all the missing trials and data from the manufacturersThe authors concluded that topical NSAIDs are effective in providing pain relief but also cautioned that the large amounts of unpublisheddata ldquocould influence results in updates of this reviewrdquo They certainly could I believe it is plain wrong to perform meta-analyses on theauthorsrsquo data When I most recently reviewed this area for the BMJ in 2010 I concluded that we donrsquot know whether topical NSAIDsare beneficial (1)1 Goslashtzsche PC NSAIDs Clin Evid (Online) 2010 Jun 28I agree with the conflict of interest statement belowI certify that I have no affiliations with or involvement in any organization or entity with a financial interest in the subject matter ofmy feedback
Reply
Response submitted by authors Sheena Derry and Professor Andrew Moore on 4 November 2015Peter has made these same points elsewhere [PubMed Commons PubPeer gt Cochrane Database Syst Rev October 2015 httpspubpeercompublicationsA9E5BEA36549727357F9FD14CC2537]As a preliminary it is important to stress that the number of trials available in pain has increased over time and newer trials are typicallybetter and larger This means we can move from answering the simple question of whether an intervention works to more importantand relevant questions such as how well the intervention works and for drugs particularly to examine effects of dose and formulationFormulation is now recognised to have profound effects For oral analgesics for example fast acting formulations demonstrate up todouble the analgesic effect for a given dose as our recently updated overview points out (Cochrane Database Syst Rev 2015 Sep 289CD008659 see also Pain 2014 Jan155(1)14-21)What was true for oral analgesics in acute pain is now true also for topical NSAIDs in acute pain where manufacturers have beenputting a lot of effort into trying to produce new formulations that work better This updated review sought to examine the influenceof formulation and drug although dose is somewhat more difficult There are good reasons why formulation may be important(Rheumatic Disease Clinics 2008 Vol 34 Issue 2 p 415-432)
143Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Our searches did identify a large number of unpublished trials We deplore this but are powerless to change things For a previousreview of topical NSAIDs we contacted all identified manufacturers and asked for published or unpublished data The yield was small(as others have found) but some unpublished studies were brought into the public domain Waiting for all studies and clinical trialreports of all studies would take forever and could probably never be achieved We believe that many of these unpublished studiesrelate to drugs andor formulations that have never been manufactured commercially While these would be of interest in determiningwhat does and doesnrsquot work and to direct future research they would probably have little clinical relevance because these formulationsare unlikely to come to marketPeterrsquos main issue appears to be heterogeneity Tests for heterogeneity are problematical anyway (Pain 2000 85415-24) and the I squareof 92 that he quotes was for all topical diclofenac formulations combined While we do give an overall summary for diclofenac wedemonstrated in the review that the different diclofenac formulations produced different results from one another using LrsquoAbbeacute plotsand in our detailed analyses showing large variations in efficacy between them In the circumstance a high I square for all combined(clinical heterogeneity) is to be expected but it is not relevant The bulk of the studies on diclofenac were published in the last fiveyears were of decent quality and moderate to large size There were older data for ketoprofen but again these showed major differencesbetween formulations Differences between formulations are highlighted throughout the reviewTrying to determine publication bias using funnel plots or other measures is something of a lost cause (Journal of Clinical Epidemiology200053207-16) It is especially so with small numbers of trials (Journal of Clinical Epidemiology 2000 53 477-484) and makingsense of funnel plots is anything but easy for most people (Journal of Clinical Epidemiology 2005 58 894-901) A useless methodseems an odd choice to make to criticise our reviewThere are very good scientific reasons why drug and formulation may play a big part in the efficacy of topical NSAIDs This is alsothe case for oral analgesics used in acute pain where formulation improvements generating rapid absorption confers greater efficacySimple lumping strategies may have been permissible in past systematic review methodology but a more forensic approach is needednow and in the future This is what we have attempted to do in this latest reviewComparisons with Peterrsquos 2010 review seem inappropriate since that review was based on other reviews that are now out of date
Contributors
Feedback from Peter C GoslashtzscheAuthors involved with responding Andrew Moore Sheena DerryFeedback Editor Kate Seers
W H A T rsquo S N E W
Last assessed as up-to-date 4 February 2015
Date Event Description
4 November 2015 Feedback has been incorporated Feedback has been incorporated See Feedback 2
H I S T O R Y
Protocol first published Issue 4 2008
Review first published Issue 6 2010
144Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Date Event Description
13 October 2015 Amended Small errors found in Summary Table B for total num-ber of participants with piroxicam and indomethacinand in text for NNT with diclofenac (other gel) Nochange to conclusions
6 October 2015 Amended Small error found in Summary Table B percentagesfor NSAID and placebo were the wrong way round forcomparison of all NSAIDs and placebo RR and textwere correct No change to results or conclusions
13 February 2015 New citation required but conclusions have notchanged
Conclusions not changed Results remain essentiallythe same but the focus has changed to examine drugand formulation combination
3 February 2015 New search has been performed Title changed from Topical NSAIDs for acute pain inadults to Topical NSAIDs for acute musculoskeletalpain in adults to increase specificityNew searches run and new studies identified Fourteennew included studies using diclofenac (3489 new par-ticipants a 63 increase) four new excluded studiesFifteen studies awaiting classification (completed noresults available but likely to satisfy inclusion criteria)
23 May 2014 Amended Error in data reported for clinical success in Hosie 1993was brought to our attention and has been corrected
12 June 2012 Feedback has been incorporated We have incorporated feedback received from DrSigrun Chrubasik and the authorrsquos response on DMSOand other additives
C O N T R I B U T I O N S O F A U T H O R S
For the earlier review Tom Massey and SD identified studies and carried out data extraction analysis and drafting RAM and HJMwere involved in planning acted as adjudicators and were involved with writing
For this update SD and RAM carried out searches data extraction and analysis All authors were involved with writing the review
145Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
D E C L A R A T I O N S O F I N T E R E S T
SD has no conflicts relating to this review or any similar product
RAM has no conflicts relating to this review or any similar product
PW has no conflicts relating to this review or any similar product
HG has no conflicts relating to this review or any similar product
MM has no conflicts relating to this review or any similar product
S O U R C E S O F S U P P O R T
Internal sources
bull Oxford Pain Relief Trust UKGeneral institutional support
External sources
bull No sources of support supplied
D I F F E R E N C E S B E T W E E N P R O T O C O L A N D R E V I E W
For this update in 2015 we have changed the title to specify musculoskeletal pain because topical NSAIDs are not normally used totreat visceral pain or headache We felt that the new title better reflected the content of the review We have also changed the focus ofthe review from pooled analysis of all topical NSAIDs and all studies of a particular NSAID to an examination of individual drug andits formulation This makes the review much more relevant We have expanded the rsquoRisk of biasrsquo assessment and added a rsquoSummaryof findingsrsquo table and PRISMA flow chart We have removed a number of sensitivity analyses because they were not appropriate giventhe current information on the impact of formulation on efficacy The sensitivity analyses have been superseded by the rsquoRisk of biasrsquoassessment and taken into account in the rsquoSummary of findingsrsquo tables
An earlier review in 2004 chose to exclude studies using benzydamine on the grounds that it was no longer considered to be an NSAID(Mason 2004a) Although the protocol for this review stated that we would not include benzydamine after further consultation wenow believe that it should be classified as an NSAID albeit with a different mode of action which is not fully understood (Quane1998) Thus we have reinstated studies using topical benzydamine
I N D E X T E R M S
Medical Subject Headings (MeSH)
Acute Disease Administration Topical Anti-Inflammatory Agents Non-Steroidal [lowastadministration amp dosage adverse effects] AthleticInjuries [drug therapy] Pain [lowastdrug therapy] Randomized Controlled Trials as Topic Sprains and Strains [drug therapy]
146Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
MeSH check words
Adult Humans
147Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
S U M M A R Y O F F I N D I N G S F O R T H E M A I N C O M P A R I S O N [Explanation]
Topical NSAIDs compared with topical placebo for acute musculoskeletal pain in adults
Patient or population adults with strains sprains or muscle pullSettings communityIntervention topical NSAID (topical diclofenac ibuprofen and ketoprofen gels only shown here for efficacy)Comparison topical placebo
Outcomes Probable outcome withintervention
Probable outcome withcomparator
RR NNT NNTp or NNH(95 CI)
No of studies partici-pants
Quality of the evidence(GRADE)
Comments
Topical diclofenac gel
(as Emulgel)
Clinical success (eg50 reduction in pain)
780 in 1000 200 in 1000 RR34 (27 to 55)NNT18 (15 to 21)
2 studies314 participants
High Consistent results in 2moderately sized recentstudies of high quality
Topical ibuprofen gel
Clinical success (eg
50 reduction in pain)
420 in 1000 160 in 1000 RR27 (17 to 42)NNT39 (27 to 67)
2 studies241 participants
Moderate Modest effect size andnumbers of participants
Topical ketoprofen gel
Clinical success (eg
50 reduction in pain)
720 in 1000 330 in 1000 RR22 (17 to 28)NNT25 (20 to 34)
5 studies348 participants
Moderate Modest effect size andnumbers of participantsbut studies small withnone recent
All topical NSAIDs
Local adverse events
46 in 1000 50 in 1000 RR10 (080 to 12)NNH not calculated
42 studies6125 participants
High Large number of studiesand participants with con-sistent results
All topical NSAIDs
Systemic adverse
events
32 in 1000 35 in 1000 RR10 (07 to 13)NNH not calculated
38 studies5372 participants
High Large number of studiesand participants with con-sistent results
All topical NSAIDs
Withdrawals - adverse
events
11 in 1000 11 in 1000 RR10 (07 to 17)NNH not calculated
42 studies5790 participants
High Large number of studiesand participants with con-sistent results4
To
pica
lN
SA
IDs
for
acu
tem
uscu
losk
ele
talp
ain
inad
ults
(Revie
w)
Co
pyrig
ht
copy2015
Th
eC
och
ran
eC
olla
bo
ratio
nP
ub
lished
by
Joh
nW
iley
ampS
on
sL
td
Serious adverse events 1 in total 0 in total Not calculated All data Low Small numbers of events
GRADE Working Group grades of evidenceHigh quality Further research is very unlikely to change our confidence in the estimate of effectModerate quality Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimateLow quality Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimateVery low quality We are very uncertain about the estimate
CI confidence interval RR risk ratio NNT number needed to treat for an additional beneficial outcome NNTp number needed to treatto prevent an event happening NNH number needed to treat for an additional harmful outcome
xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
5To
pica
lN
SA
IDs
for
acu
tem
uscu
losk
ele
talp
ain
inad
ults
(Revie
w)
Co
pyrig
ht
copy2015
Th
eC
och
ran
eC
olla
bo
ratio
nP
ub
lished
by
Joh
nW
iley
ampS
on
sL
td
B A C K G R O U N D
This review is an update of a review originally published in Issue 62010 on rsquoTopical NSAIDs for acute pain in adultsrsquo (Massey 2010)We have changed the title to specify musculoskeletal pain becausetopical NSAIDs are not normally used to treat visceral pain orheadache We felt that the new title better reflected the content ofthe review
The use of topical NSAIDs for pain relief has been a controversialsubject in analgesic practice In some parts of the world (includingmuch of Western Europe) they have been available for many yearsare widely available without prescription widely advertised usedextensively and evidence for their use is considered adequate Inother parts of the world they were regarded as little more thanplacebo with any apparent effect attributed to the process of rub-bing at the site of the affected area In some places (for examplethe US) their use was almost unknown until the mid-2010s InEngland 52 million prescriptions for topical NSAIDs were dis-pensed in 2013 (PACT 2014) mainly for formulations of ibupro-fen (245 million) piroxicam (118 million) and diclofenac (127million)
There is good evidence for the efficacy of topical NSAIDs in acuteand chronic musculoskeletal pain (Mason 2004a Mason 2004bMoore 1998a) In the US the Food and Drug Administration li-censed topical nonsteroidal products in 2007 and in England theNational Institute for Health and Care Excellence (NICE) recom-mended topical therapies as first line treatment in its guidelinesfor osteoarthritis in 2008 (NICE 2008) Earlier reviews of topicalanalgesics covered studies investigating the underlying science toexplain biological plausibility in addition to clinical trials (Anon2005 Moore 2008a)
This review is one of a series on topical analgesics including topicalcapsaicin at low and high doses (Derry 2012a Derry 2013) andtopical NSAIDs in chronic pain conditions (Derry 2012b) andsalicylate-containing rubefacients (Derry 2014)
Description of the condition
Acute pain is usually defined as pain of less than three monthsrsquo du-ration It is often associated with injury including trauma surgerymusculoskeletal injuries such as strains sprains and over-use in-juries or soft tissue injuries such as muscle soreness or cramps
Description of the intervention
Clinicians prescribe NSAIDs on a routine basis for a range of mildto moderate pain NSAIDs are the most commonly prescribedanalgesic medications worldwide and their efficacy for treatingacute pain has been well demonstrated (Moore 2003) They re-versibly inhibit cyclooxygenase (prostaglandin endoperoxide syn-thase) the enzyme mediating production of prostaglandins and
thromboxane A2 (FitzGerald 2001) Prostaglandins mediate a va-riety of physiological functions such as maintenance of the gastricmucosal barrier regulation of renal blood flow and regulation ofendothelial tone They also play an important role in inflamma-tory and nociceptive processes However relatively little is knownabout the mechanism of action of this class of compounds asidefrom their ability to inhibit cyclooxygenase-dependent prostanoidformation (Hawkey 1999)NSAIDs taken orally or intravenously are transported to all partsof the body in the blood and relatively high blood concentrationsare needed to achieve effective tissue concentrations at the site ofthe pain and inflammation These high concentrations throughoutthe body can give rise to a number of adverse events that canbe unpleasant (for example dyspepsia) or potentially serious (forexample gastrointestinal bleeding)A topical medication is a one applied to body surfaces such asthe skin or mucous membranes to treat ailments A large rangeof types of topical formulation may be used including but notlimited to creams foams gels lotions ointments and plastersThe exact formulation of a topical medication is often determinedby the speed of drug absorption required The need may be forslow absorption into the circulation to maintain low drug concen-trations and perhaps avoiding extensive first pass metabolism inthe liver plasters containing drug reservoirs may be used for thisas with transdermal opioids or contraceptive steroids For rapidabsorption the formulation is enhanced by substances to improveor assist skin penetration perhaps only to generate high concen-trations in tissues rather than in the blood gel formulations areuseful for this purpose which is why they are sometimes used fortopical NSAIDs
Topical NSAIDs
Topical NSAIDs are formulated for direct application to thepainful site and to produce a local pain-relieving effect whileavoiding body-wide distribution of the drug at physiologically ac-tive levels (McPherson 2013) This method of application (dos-ing) necessarily limits their use to more superficial painful condi-tions such as sprains strains and muscle or tendon soreness Theywould not for example be indicated for deep visceral pain orheadaches They are also not appropriate for use on broken skinso would not be used on open wounds (accidental or surgical)
How the intervention might work
For a topical formulation to be effective it must first penetratethe skin Only when the drug has entered the lower layers ofthe skin can it be absorbed by the blood and transported to thesite of action or penetrate deeper into areas where inflammationoccurs Individual drugs have different degrees of penetration Abalance between lipid and aqueous solubility is needed to optimisepenetration and use of prodrug esters has been suggested as a way
6Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
of enhancing permeability Formulation is also crucial to goodskin penetration Experiments with artificial membranes or humanepidermis suggest that creams are generally less effective than gelsor sprays but newer formulations such as microemulsions mayhave greater potentialOnce the drug has reached the site of action it must be presentat a sufficiently high concentration to inhibit cyclooxygenase en-zymes thereby reducing prostaglandin synthesis This in turn re-duces inflammation and relieves pain It is probable that in acuteconditions topical NSAIDs exert their action primarily by lo-cal reduction of symptoms independent of any systemic uptakeand delivery Tissue levels of NSAIDs applied topically certainlyreach levels high enough to inhibit cyclooxygenase-2 (Anon 2005Haroutiunian 2010 Moore 2008a) However plasma concentra-tions found after topical administration are only a fraction (usu-ally much less than 5) of the levels found in plasma followingoral administration Topical application can potentially limit sys-temic adverse events by increasing local effects and minimisingsystemic concentrations of the drug We know that the incidenceof upper gastrointestinal bleeding is low with chronic use of topi-cal NSAIDs (Evans 1995) although it has been reported partic-ularly in people with risk factors (Zimmerman 1995) We haveno certain knowledge of lower effects on cardiovascular events orrenal failure both of which have been associated with oral NSAIDuse
Why it is important to do this review
Since the last review in 2010 a number of new studies have beenpublished nearly all of which investigated various formulations ofdiclofenac These new studies are generally of higher quality thanmany of the earlier ones in this review and have the potential toinfluence the strength of its conclusions substantially Moreoverthe additional information allows for analysis based not only on aparticular drug but also on the formulation of that drug This canprovide better insight into whether formulation affects efficacy oftopical NSAIDs in acute musculoskeletal painAn updated review of evidence for topical NSAIDs was needed toinform choices made by consumers prescribers and commission-ers (purchasers of healthcare)
O B J E C T I V E S
To determine the efficacy and safety of topically applied NSAIDsin acute musculoskeletal pain in adults
M E T H O D S
Criteria for considering studies for this review
Types of studies
Randomised controlled double-blind studies comparing topicalNSAIDs with placebo (inert carrier) or other active treatment foracute pain with at least 10 participants per treatment arm andoutcomes close to seven days (minimum three days) We excludedstudies published only as short abstracts (which report insufficientdata to assess methods) or studying experimentally induced pain(which does not correlate well with clinical pain) Because a cross-over design is not appropriate for self limiting conditions such assprains strains and contusions we only considered parallel-groupdesigns
Types of participants
Adults (aged 16 years or more) with acute musculoskeletal pain ofat least moderate intensity resulting mainly from strains sprainsor sports injuries Typically for sports injuries the injury wouldhave occurred within 24 or 48 hours
Types of interventions
Included studies had at least one treatment arm using a topicalNSAID and a comparator arm using placebo (inert carrier withoutNSAID or other active treatment) The topical NSAID had to beapplied at least once daily We did not include salicylates in thisreview as they are no longer classified as topical NSAIDs and arecovered in a separate review (Derry 2014)
Types of outcome measures
We sought information on participant characteristics includingage sex and condition treated
Primary outcomes
The primary outcome was rsquoclinical successrsquo defined as at least a50 reduction in pain or equivalent measure such as a rsquovery goodrsquoor rsquoexcellentrsquo global assessment of treatment or rsquononersquo or rsquoslightrsquopain on rest or movement measured on a categorical scale (Moore1998a) We used the following hierarchy of outcomes to extractdata for the primary outcome
bull Participant reported reduction in pain of at least 50bull Participant reported global assessment of treatmentbull Pain on movementbull Pain at rest or spontaneous painbull Undefined rsquoimprovementrsquo
We used only participant reported outcomes of efficacy and notphysician or investigator reported outcomes
7Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Secondary outcomes
bull Numbers of participants with adverse events local andsystemic
bull Numbers of withdrawals all cause lack of efficacy andadverse events
We anticipated that outcomes would be reported after differentdurations of treatment and extracted data reported as close toseven days as possible with a minimum of three days We also ex-tracted data for outcomes reported after longer durations of treat-ment We anticipated that reporting of adverse events would varybetween studies with regard to the terminology used method ofascertainment and categories reported (for example occurring inat least 5 of participants or where there is a statistically signifi-cant difference between treatment groups) We took care to iden-tify these details where relevant
Search methods for identification of studies
Electronic searches
We searched the following databases without language restrictionbull CENTRAL (The Cochrane Library) Issue 4 2009 for the
original review and the Cochrane Register of Studies Online(CRSO) to 3 February 2015 for this update
bull MEDLINE (via Ovid) from inception to December 2009for the original review and from 2008 to 3 February 2015 forthis update
bull EMBASE (via Ovid) from inception to December 2009for the original review and from 2008 to 3 February 2015 forthis update
bull Oxford Pain Relief Database for the original review (Jadad1996a) This resource is no longer being updated
See Appendix 1 for the CENTRAL search strategy Appendix 2 forthe MEDLINE search strategy and Appendix 3 for the EMBASEsearch strategy
Searching other resources
We searched the reference lists of review articles and includedstudies We have previously asked manufacturers for details ofunpublished studies but did not make new requestsWe searched two clinical trial registries (clinicaltrialsgov (clinicaltrialsgov) and the World Health Organization In-ternational Clinical Trials Registry Platform (appswhointtrialsearch)) and asked personal contacts about ongoing and un-published studies
Data collection and analysis
Selection of studies
We screened the titles and abstracts of studies identified by thesearches to eliminate those that clearly did not satisfy the inclu-sion criteria and obtained full reports of the remaining studies todetermine inclusion in the review
Data extraction and management
Review authors were not blinded to the authorsrsquo names and in-stitutions journal of publication or study results at any stage ofthe review Two review authors independently selected the stud-ies for inclusion assessed methodological quality and risk of biasand extracted data We resolved disagreements and uncertaintiesthrough discussionWe abstracted information on participants interventions and out-comes from the original reports into a standard data extractionform One review author entered data suitable for meta-analysisinto Review Manager 5 (RevMan 2014) and another review au-thor checked it
Assessment of risk of bias in included studies
We used the Oxford Quality Score as the basis for inclusion lim-iting inclusion to studies that were randomised and double-blindas a minimum (Jadad 1996b)Two review authors independently assessed risk of bias for eachstudy using the criteria outlined in the Cochrane Handbook forSystematic Reviews of Interventions with any disagreements resolvedby discussion (Higgins 2011) We assessed the following for eachstudy
bull Random sequence generation (checking for possibleselection bias) We assessed the method used to generate theallocation sequence as low risk of bias (any truly randomprocess for example random number table computer randomnumber generator) unclear risk of bias (method used to generatesequence was not clearly stated) We excluded studies using anon-random process that were therefore at high risk of bias (forexample odd or even date of birth hospital or clinic recordnumber)
bull Allocation concealment (checking for possible selectionbias) The method used to conceal allocation to interventionsbefore assignment determines whether intervention allocationcould have been foreseen in advance of or during recruitmentor changed after assignment We assessed the methods as lowrisk of bias (for example telephone or central randomisationconsecutively numbered sealed opaque envelopes) unclear riskof bias (method was not clearly stated) We excluded studies thatdid not conceal allocation and were therefore at high risk of bias(for example open list)
8Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
bull Blinding of outcome assessment (checking for possibledetection bias) We assessed the methods used to blind studyparticipants and outcome assessors from knowledge of whichintervention a participant received We assessed the methods aslow risk of bias (study stated that it was blinded and describedthe method used to achieve blinding for example identicaltubes containing gel or identical plasters matched in appearanceand smell) unclear risk of bias (study stated that it was blindedbut did not provide an adequate description of how blinding wasachieved) We excluded studies that were not double-blind andwere therefore at high risk of bias
bull Incomplete outcome data (checking for possible attritionbias due to the amount nature and handling of incompleteoutcome data) We assessed the methods used to deal withincomplete data as low risk of bias (less than 10 ofparticipants did not complete the study or used rsquobaselineobservation carried forwardrsquo analysis or both) unclear risk ofbias (used rsquolast observation carried forwardrsquo analysis) or high riskof bias (used rsquocompleterrsquo analysis)
bull Size (checking for possible biases confounded by small size)Small studies have been shown to overestimate treatment effectsprobably due to methodological weaknesses (Dechartres 2013Nuumlesch 2010) We assessed studies as at low risk of bias if theyhad at least 200 participants at unclear risk if they had 50 to 200participants and at high risk if they had fewer than 50participants
Measures of treatment effect
We used risk ratio (RR) to establish statistical difference and num-bers needed to treat for an additional beneficial outcome (NNT)with 95 confidence intervals (CI) We pooled percentages as ab-solute measures of benefit or harmWe used the following terms to describe adverse outcomes in termsof harm or prevention of harm
bull When significantly fewer adverse outcomes occurred withtreatment than with control (placebo or active) we used the termthenumber needed to treat to prevent one event (NNTp)
bull When significantly more adverse outcomes occurred withtreatment compared with control (placebo or active) we usedthe term the number needed to treat for an additional harmfuloutcome or cause one event (NNH)
Unit of analysis issues
Randomisation was to the individual participant
Dealing with missing data
Wherever possible we used intention-to-treat (ITT) analysis wherethe ITT population consists of participants who were randomisedapplied at least one dose of the assigned study medication andprovided at least one post-baseline assessment We assigned miss-ing participants zero improvement
We also looked for information about methods of imputation formissing data
Assessment of heterogeneity
We examined heterogeneity visually using LrsquoAbbeacute plots (LrsquoAbbeacute1987) a visual method for assessing differences in results of indi-vidual studies and with the I2 statistic
Assessment of reporting biases
The aim of this review was to use dichotomous outcomes of knownutility and of value to patients (Moore 2013) The review did notdepend on what the authors of the original studies chose to reportor not Studies that did not report dichotomous results but onlyaverage pain data did not contribute to analysesWe assessed publication bias using a method designed to detect theamount of unpublished data with a null effect required to makeany result clinically irrelevant (usually taken to mean an NNT of10 or higher Moore 2008b)
Data synthesis
We pooled data only for comparisons and outcomes where therewere at least two studies and 200 participants (Moore 1998b)When two active treatment arms were compared with a placeboarm we took care to avoid double counting of participants in theplacebo arm if both active groups contributed to an analysis wesplit the placebo group between themWe calculated RRs with 95 CIs using the fixed-effect model(Morris 1995) A statistically significant benefit of topical NSAIDover control was assumed when the lower limit of the 95 CI ofthe RR was greater than one A statistically significant benefit ofcontrol over active treatment was assumed when the upper limitof the 95 CI was less than one We calculated NNTs with 95CIs using the pooled number of events by the method of Cookand Sackett (Cook 1995)Statistically significant differences between NNTs for differenttopical NSAIDs were tested using the z test (Tramer 1997) wherethere were sufficient data to do so and where the studies weresufficiently similar in types of participant outcome and durationto make such comparisons sensible
Subgroup analysis and investigation of heterogeneity
We carried out separate analyses for individual NSAIDs andwhere the data permitted for different formulations of individualNSAIDs
Sensitivity analysis
The earlier review included sensitivity analyses for various factorsthat are now covered by the assessment of risk of bias
9Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
R E S U L T S
Description of studies
Results of the search
New searches for this update identified 20 publications that wereexamined in further detail to determine inclusion status We alsoidentified 20 additional studies in clinical trial registries See Figure1
10Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Figure 1 Study flow diagram
11Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
We identified 14 new studies (11 publications three unpub-lished reports) satisfying our inclusion criteria (Costantino2011 Coudreuse 2010 Gonzaacutelez de Vega 2013 Hofman2000 Klainguti 2010 Kuehl 2011 Li 2013 NCT01255423NCT01272934 NCT01272947 Predel 2012 Predel 2013aPredel 2013b Saillant 1998) one post hoc analysis of a study thatwas included in the earlier review (Mueller 2010 in Predel 2004)and an additional publication and a pooled analysis that includednew data for another study from the earlier review (Lionberger2011 pooled analysis in Joussellin 2003)
Included studies
All except one of the new studies compared diclofenac withplacebo A number of different formulations were used includ-ing diclofenac epolamine (DHEP) with or without heparin (Flec-toparin Tissugel or Flector EP Tissugel) applied as a plaster (orpatch) diclofenac diethylamine (DDEA) applied as a gel and di-clofenac with lethicin applied as a spray gel The remaining studycompared diclofenac gel with traumeel a ldquofixed combination ofplant and mineral extractsrdquo applied as a gel or an ointmentThere were 47 studies in the original review 14 new studies wereincluded making a total of 61 studies in this updated reviewAll used a parallel group design Forty-four compared a topicalNSAID with placebo 13 a topical NSAID with an active com-parator (a different topical NSAID an oral NSAID the same top-ical NSAID in a different formulation or a compound of plantand mineral extracts) and four had both placebo and active com-parators In total 5311 participants were treated with a topicalNSAID 3470 with placebo and 220 with an oral NSAID Top-ical NSAIDs used were benzydamine diclofenac etofenamatefelbinac fentiazac flunoxaprophen flurbiprofen ibuprofen in-domethacin ketoprofen ketorolac lysine clonixinate meclofe-namic acid naproxen niflumic acid and piroxicam They wereapplied as creams gels sprays foams or plasters (patches) Topicalplacebos were the inert carriers without the active NSAID OralNSAIDs used were ibuprofen (as tablets) and indomethacin (ascapsules)Most studies enrolled participants who had sprains strains andcontusions usually as a result of sports injuries and treatmentwas started within a few hours or days Other studies enrolledparticipants with overuse-type injuries such as tendinitis and acutelow back pain where pain had been present for days or weeks butless than three monthsParticipants were treated for at least five days and up to threeweeks with most studies lasting seven to 14 days Participantswere usually assessed in clinic at intervals during treatment andsometimes also at home using daily patient diaries We used out-comes closest to seven days because many of these injuries are self
limiting with differences between active treatment and placebobeing diminished or lost after longer intervalsMost studies reported dichotomous outcomes suitable for a re-sponder analysis although group mean change (for pain or phys-ical function for example) was usually the primary outcomesHowever the definition of response varied both in the parametermeasured (for example pain pain on movement patient globalevaluation of treatment) and in the scale used to measure it (forexample a 3- 4- or 5-point scale for patient global evaluation)Details of included studies are in the Characteristics of includedstudies tableWe identified 14 completed but apparently unpublished studies ina clinical trial registry for which no results have been posted (4403participants NCT00351104 NCT00352625 NCT00426985NCT00640705 NCT00640939 NCT00680472NCT00680784 NCT00765700 NCT00869063NCT00869180 NCT00931866 NCT01874626NCT01957215 NCT02324270) We have placed these underCharacteristics of studies awaiting classification We also identi-fied three conference abstracts that relate to completed studies thatdo not appear to have been published but that may satisfy ourinclusion criteria One is likely to be the same study as one of theincluded studies identified in a clinical trial registry (Pallay 2013in NCT01272947) one relates to another study identified in theclinical trial registry that is awaiting classification (Ekman 2010in NCT00765700) while we could find no published or unpub-lished reports of the other Sarzi-Puttini 2014)We also identified three ongoing studies with an estimated en-rolment of 880 participants (NCT01945034 NCT02100670NCT02290821) Details are in the Characteristics of ongoingstudies table
Excluded studies
For the original review 25 studies were excluded after readingthe full paper For this update we excluded four new studies (Cesarone 2008 Coulibaly 2009 Kuwabara 2013 Vinciguerra2008) Details are in the Characteristics of excluded studies table
Risk of bias in included studies
All studies were randomised and double-blind One study scored25 (Sinniger 1981) 23 scored 35 23 scored 45 and 14 scored55 for methodological quality using the Oxford Quality Scale Abreakdown of the scores for individual studies is reported in theCharacteristics of included studies tableComments on potential biases in individual studies are reported inthe rsquoRisk of biasrsquo section of the Characteristics of included studiestable The findings are displayed in Figure 2 and Figure 3
12Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Figure 2 Risk of bias summary review authorsrsquo judgements about each risk of bias item for each includedstudy
13Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Figure 3 Methodological quality graph review authorsrsquo judgements about each methodological quality
item presented as percentages across all included studies
Allocation
All the studies were randomised but only 17 adequately describedthe method used to generate the random sequence Thirteen stud-ies adequately described the method used to conceal allocation ofthe sequence No studies were at high risk of bias for this item
Blinding
All studies were double-blind and 38 adequately described themethod used to maintain the blinding No studies were at highrisk of bias for this item
Incomplete outcome data
Thirty-six studies included all participants in the primary analysisor provided sufficient data to allow missing participants to be in-cluded as non-responders and were judged at low risk of bias Wejudged four studies to be at high risk of attrition bias (Campbell1994 Kuehl 2011 Maziegraveres 2005a Russell 1991) Three unpub-lished studies contributed only to adverse event analyses and ac-counted for all participants for these outcomes (NCT01255423NCT01272934 NCT01272947)
Other potential sources of bias
We judged one study that included more than 200 participantsin each treatment arm to be at low risk of bias from size butthis study had a very high attrition rate (see rsquoIncomplete outcome
data (attrition bias)rsquo) and did not report all the efficacy outcomesmeasured (Kuehl 2011) We judged 27 studies to be at high riskbecause they included fewer than 50 participants per treatmentarm
Effects of interventions
See Summary of findings for the main comparison TopicalNSAIDs compared with topical placebo for acute musculoskeletalpain in adultsThree studies did not contribute data suitable for analysis of at leastone outcome (Gonzaacutelez de Vega 2013 Gualdi 1987 Hoffmann2012)
1 Topical NSAID versus placebo
Details of efficacy outcomes in individual studies are in Appendix4 and of adverse events and withdrawals in Appendix 5 Appendix6 has details of the concentration of topical products the amountapplied the frequency of application and an estimation of thedaily dose of topical NSAID applied Not all studies providedsufficient information to allow calculation of daily dose appliedFor example for topical diclofenac the estimated doses appliedvaried between about 60 and 280 mg for topical ketoprofen 100to about 450 mg for topical ibuprofen 300 to 800 mg
Participants with clinical success
14Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Topical diclofenac versus placebo
Ten studies contributed to this analysis (Coudreuse 2010Joussellin 2003 Klainguti 2010 Li 2013 Predel 2004 Predel2012 Predel 2013a Predel 2013b Rowbotham 2003) of whichone (Predel 2012) had two active treatment arms A total of 1074participants were treated with topical diclofenac and 976 withplacebo (Analysis 11 Figure 4)
15Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Figure 4 Forest plot of comparison 2 Individual NSAID versus placebo outcome 21 Clinical success
16Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
bull The proportion of participants experiencing successfultreatment with topical diclofenac was 74 (8001074 range39 to 100)
bull The proportion of participants experiencing successfultreatment with placebo was 47 (461976 range 8 to 82)
bull The RR for treatment compared with placebo was 16(95 CI 15 to 17)
bull The NNT for successful treatment was 37 (32 to 43) Forevery four participants treated with topical diclofenac one wouldexperience successful treatment who would not have done sowith placebo
Effect of formulation
The effects of formulation are shown in Analysis 21 and Figure5
17Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Figure 5 LrsquoAbbeacute plot of clinical success in studies of topical diclofenac versus topical placebo The size ofthe symbol is proportional to the size of the study (inset scale) Dark blue Emulgel light blue spraygel red
Flector pink other patch or plaster
bull Four studies used a Flectorreg plaster (1030 participantsJoussellin 2003 Li 2013 Rowbotham 2003 Saillant 1998) TheRR for treatment compared with placebo was 15 (14 to 17)and the NNT was 47 (37 to 65)
bull Three studies used other makes of plaster (474 participantsCoudreuse 2010 Klainguti 2010 Predel 2004) The RR fortreatment compared with placebo was 16 (14 to 18) and theNNT was 32 (26 to 42)
bull Two studies used Voltaren Emulgel (314 participants Predel2012 Predel 2013b) The RR for treatment compared withplacebo was 38 (27 to 55) and the NNT was 18 (15 to 21)
bull One study used a spray gel (232 participants Predel2013a) The RR for treatment compared with placebo was 12(105 to 13) and the NNT was 80 (48 to 24)
Diclofenac as the gel formulation Emulgel was statistically more ef-ficacious than the plaster formulation as Flector plaster (z = 6360P value lt 000001)
Topical ibuprofen versus placebo
Five studies contributed to this analysis (Billigmann 1996Campbell 1994 Dreiser 1988 Machen 2002 Ramesh 1983) Atotal of 218 participants were treated with topical ibuprofen and218 with placebo (Analysis 11)
bull The proportion of participants experiencing successfultreatment with topical ibuprofen was 55 (120218 range 31to 81)
18Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
bull The proportion of participants experiencing successfultreatment with placebo was 33 (73218 range 13 to 76)
bull The RR of treatment compared with placebo was 16 (13to 20)
bull The NNT for successful treatment was 46 (33 to 80) Forevery five participants treated with topical ibuprofen one wouldexperience successful treatment who would not have done sowith placebo
Effect of formulation
The effects of formulation are shown in Analysis 31bull Three studies used cream formulations (195 participants
Campbell 1994 Dreiser 1988 Ramesh 1983) Although this isjust below our threshold for pooled analysis we have includedthis analysis for completeness and the results should beinterpreted with caution The RR for treatment compared withplacebo was 13 (103 to 16) and the NNT was 64 (34 to 41)
bull Two studies used gel formulations (241 participantsBilligmann 1996 Machen 2002) The RR for treatmentcompared with placebo was 27 (17 to 42) and the NNT was39 (27 to 67)
There was no statistically significant difference between the geland cream formulations (z = 1160 P value = 0246)
Topical ketoprofen versus placebo
Seven studies contributed to this analysis (Airaksinen 1993Dreiser 1989 Julien 1989 Kockelbergh 1985 Maziegraveres 2005bMaziegraveres 2005a Noret 1987) A total of 346 participants weretreated with topical ketoprofen and 337 with placebo (Analysis11)
bull The proportion of participants experiencing successfultreatment with topical ketoprofen was 73 (251346 range57 to 89)
bull The proportion of participants experiencing successfultreatment with placebo was 47 (157337 range 17 to 73)
bull The RR of treatment compared with placebo was 16 (14to 18)
bull The NNT for successful treatment was 39 (30 to 53) Forevery four participants treated with topical ketoprofen onewould experience successful treatment who would not have doneso with placebo
Effect of formulation
The effects of formulation are shown in Analysis 41 and Figure6
19Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Figure 6 LrsquoAbbeacute plot of clinical success in studies of topical ketoprofen versus topical placebo The size ofthe symbol is proportional to the size of the study (inset scale) Light blue ketoprofen gel pink ketoprofen
plaster
Two studies used a plaster formulation (335 participants Maziegraveres2005b Maziegraveres 2005a) The RR for treatment compared withplacebo was 12 (104 to 14) and the NNT was 82 (45 to 47)Five studies used gel formulations (348 participants Airaksinen1993 Dreiser 1989 Julien 1989 Kockelbergh 1985 Noret 1987)The RR for treatment compared with placebo was 22 (17 to 28)and the NNT was 25 (20 to 34)Ketoprofen as a gel formulation was statistically more efficaciousthan a plaster formulation (z = 3860 P value = 000014)
Topical piroxicam versus placebo
Three studies contributed to this analysis (Aoki 1984 Fujimaki1985 Russell 1991) A total of 255 participants were treated withtopical piroxicam and 249 with placebo (Analysis 11)
bull The proportion of participants experiencing successfultreatment with topical piroxicam was 68 (179255 range 53to 79)
bull The proportion of participants experiencing successfultreatment with placebo was 47 (118249 range 45 to 49)
bull The RR of treatment compared with placebo was 15 (13to 17)
bull The NNT for successful treatment was 44 (32 to 69) Forevery four participants treated with topical piroxicam one wouldexperience successful treatment who would not have done sowith placebo
Topical indomethacin versus placebo
20Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Three studies contributed to this analysis ( kermark 1990 Aoki1984 Fujimaki 1985) A total of 168 participants were treatedwith topical indomethacin and 173 with placebo (Analysis 11)
bull The proportion of participants experiencing successfultreatment with topical indomethacin was 58 (97168 range54 to 64)
bull The proportion of participants experiencing successfultreatment with placebo was 46 (79173 range 25 to 49)
bull The RR of treatment compared with placebo was 13 (103to 16)
bull The NNT for successful treatment was 83 (44 to 65) Forevery eight participants treated with topical indomethacin onewould experience successful treatment who would not have doneso with placebo
Topical benzydamine versus placebo
Three studies contributed to this analysis (Chatterjee 1977 Haig1986 Linde 1985) A total of 96 participants were treated withtopical benzydamine and 97 with placebo (Analysis 11) Al-though this is just below our threshold for pooled analysis we haveincluded this analysis for completeness and the results should beinterpreted with caution
bull The proportion of participants experiencing successfultreatment with topical benzydamine was 77 (7496 range70 to 86)
bull The proportion of participants experiencing successfultreatment with placebo was 67 (6597 range 48 to 80)
bull The RR of treatment compared with placebo was 12 (096to 14) There was no statistically significant difference betweentreatments (Figure 4)
Results for participants with clinical success with individual topi-cal NSAIDs where there were adequate data for analysis are sum-marised below in rsquoSummary of results Arsquo and Analysis 11 Analysis21 Analysis 31 and Analysis 41
Summary of results A Participants with clinical success
Comparison Studies Participants NSAID()
Placebo () Relative benefit (95CI)
NNT (95 CI)
Diclofenac -Flector plaster
4 1030 63 41 15 (14 to 17) 47 (37 to 65)
Diclofenac -other plaster
3 474 88 57 16 (14 to 18) 32 (26 to 42)
Diclofenac -Emulgel
2 314 78 20 38 (27 to 55) 18 (15 to 21)
Diclofenac -other gel
1 232 94 82 12 (11 to 13) 80 (48 to 24)
Ibuprofen -cream
3 195 71 56 13 (103 to 16) 64 (34 to 41)
Ibuprofen - gel 2 241 42 16 27 (17 to 42) 39 (27 to 67)
Ketoprofen -plaster
2 335 73 60 12 (104 to 14) 82 (45 to 47)
Ketoprofen - gel 5 348 72 33 22 (17 to 28) 25 (20 to 34)
Piroxicam 3 504 70 47 15 (13 to 17) 44 (32 to 69)
Indomethacin 3 341 58 46 13 (103 to 16) 83 (44 to 65)
21Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Benzydamine 3 193 77 67 12 (096 to 14) not calculated
Results for these two comparisons are derived from very smallamounts of data and are provided here for completeness Theyshould be interpreted with caution
Local adverse events
Local adverse events were irritation of the area to which the topicalNSAID was applied including redness or erythema and itch orpruritus Where reported these were usually described as mild andtransient
All topical NSAIDs versus placebo
Forty-two studies contributed to this analysis of which three com-pared two different drugs with placebo (Aoki 1984 Diebshlag1990 Fujimaki 1985) Three studies had two treatment armscomparing different formulations or application regimens for di-clofenac with placebo which have been combined for this analysis
(Costantino 2011 Klainguti 2010 Predel 2012) In total 3619participants were treated with topical NSAIDs and 3121 withplacebo (Analysis 51)
bull The proportion of participants experiencing a local adverseevent with a topical NSAID was 43 (1553619 range 0 to33)
bull The proportion of participants experiencing a local adverseevent with placebo was 46 (1453121 range 0 to 32)
bull The RR of topical NSAID compared with placebo was 098(080 to 12)
bull There was no significant difference between treatmentgroups so the NNH was not calculated
Individual topical NSAIDs versus placebo
Results for local adverse events with individual topical NSAIDswhere there were adequate data for analysis are in Summary ofresults B and Analysis 12
Summary of results B Participants with local adverse events
Comparison Studies Participants NSAID()
Placebo()
RR(95 CI)
NNH(95 CI)
All NSAIDs 42 6740 43 46 098 (080 to 12) Not calculated
Diclofenac 15 3271 31 43 078 (056 to 11) Not calculated
Ketoprofen 8 852 11 95 12 (083 to 17) Not calculated
Piroxicam 3 522 23 54 042 (017 to 11) Not calculated
Felbinac 3 397 30 15 19 (049 to 75) Not calculated
Indomethacin 3 354 63 22 27 (091 to 77) Not calculated
Ibuprofen 3 321 10 43 23 (098 to 54) Not calculated
Systemic adverse events
All topical NSAIDs versus placebo
Thirty-six studies contributed data on systemic adverse eventsof which three compared two different drugs with placebo (Aoki1984 Diebshlag 1990 Fujimaki 1985) Two studies had two treat-
22Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
ment arms comparing different formulations or application regi-mens for diclofenac with placebo which have been combined forthis analysis (Klainguti 2010 Predel 2012) In total 2956 partic-ipants were treated with a topical NSAID and 2620 with placebo(Analysis 52)
bull Twenty-three studies reported no systemic adverse events inany arm of the study
bull The proportion of participants experiencing a systemicadverse event with a topical NSAID was 31 (922956)
bull The proportion of participants experiencing a systemicadverse event with placebo was 35 (912620)
bull The RR of topical NSAID compared with placebo was 096(073 to 13)
bull There was no significant difference between treatmentgroups so the NNH was not calculated
A further six studies did not report the occurrence or oth-erwise of systemic adverse events (Billigmann 1996 Julien1989 Kockelbergh 1985 Noret 1987 Ramesh 1983 Vecchiet1991) while two studies did not report numbers of participantswith systemic adverse events ( kermark 1990 Auclair 1989)Costantino 2011 reported that there were no systemic gastroin-testinal adverse events Two studies reported only on total adverseevents without distinguishing between local and systemic events(NCT01255423 NCT01272934)
Serious adverse events
Two studies reported serious adverse events In Hoffmann 2012one participant experienced three serious adverse events none ofwhich was judged to be related to the study medication (diclofenacplaster) In NCT01272934 one participant using diclofenac gelruptured the ligaments of the wrist There was no statement aboutlikely relationship to the study medication but this seems unlikely
Adverse event withdrawals
Forty-two studies reported data relating to adverse event with-drawals of which three compared two different drugs with placebo(Aoki 1984 Diebshlag 1990 Fujimaki 1985) Two studies hadtwo treatment arms comparing different formulations or applica-tion regimens for diclofenac with placebo which have been com-bined for this analysis (Klainguti 2010 Predel 2012) In total3365 participants received a topical NSAID and 3040 placebo(Analysis 53)
bull Forty-four comparisons reported no adverse eventwithdrawals
bull The proportion of participants withdrawing from the studydue to an adverse event after treatment with a topical NSAIDwas 098 (333365)
bull The proportion of participants withdrawing from the studydue to an adverse event after treatment with placebo was 099(303040)
bull The RR of topical NSAID compared to placebo was 10(064 to 16)
bull There was no significant difference between treatmentgroups so the NNH was not calculated
Four studies did not specifically mention adverse event with-drawals (Haig 1986 Hoffmann 2012 Klainguti 2010 Vecchiet1991) while one reported that one participant withdrew with mildpruritus but did not state the treatment arm (Joussellin 2003)Ten studies specifically reported withdrawals due to lack of efficacy(Dreiser 1989 Dreiser 1994 Kuehl 2011 Machen 2002 Maziegraveres2005b Maziegraveres 2005a Noret 1987 Predel 2013a Russell 1991Thorling 1990) (Appendix 5) Numbers of participants withdraw-ing were generally low with rates of 6 or less except in Kuehl2011 where the rate was 10 with active treatment (diclofenacplaster) and 12 with placebo We did not carry out any analysisbecause the outcome was inconsistently reported
2 Topical NSAID versus active comparator
Details of efficacy outcomes in individual studies are in Appendix4 and of adverse events and withdrawals in Appendix 5
Participants with clinical success
Topical NSAID versus oral NSAID
bull kermark 1990 compared indomethacin spray withindomethacin capsules with response rates of 55 (1222) withspray and 23 (522) with capsules
bull Hosie 1993 compared felbinac foam with ibuprofen tabletswith response rates of 64 (81127) with felbinac foam and72 (96133) with ibuprofen tablets
bull Whitefield 2002 compared ibuprofen gel with ibuprofentablets with response rates of 60 (3050) with gel and 54(3650) with tablets
There were insufficient data for meta-analysis for any one of thesecomparisons felbinac is not known to be better than placebo
Topical NSAID versus different formulation of the same
topical NSAID
bull Fioravanti 1999 compared DHEP (diclofenac) gelformulated with and without lecithin with response rates of70 (3550) in both treatment arms
bull Mahler 2003 compared DHEP (diclofenac) gel formulatedwith and without lecithin with response rates of 89 (8292)with lecithin and 70 (6288) without lecithin
bull Gallacchi 1990 compared topical diclofenac formulated asFlectorreg gel and Emugelreg with response rates of 76 (1925)in both treatment arms
23Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
bull Governali 1995 compared topical ketoprofen cream withgel with response rates of 93 (1415) with cream and 27 (415) with gel
There were insufficient data for analysis
Topical NSAID versus different topical NSAID
Eight studies compared one topical NSAID versus at least oneother topical NSAID piroxicam versus indomethacin (Aoki1984 Fujimaki 1985 Sugioka 1984) ibuprofen versus ketopro-fen (Curioni 1985 Picchio 1981) ketoprofen versus etofena-mate (Curioni 1985 Tonutti 1994) ibuprofen versus etofenamate(Curioni 1985) ketorolac versus etofenamate (Diebshlag 1990)and diclofenac versus lysine clonixinate (Hofman 2000)There were sufficient data to compare only piroxicam with in-domethacin (Aoki 1984 Fujimaki 1985 Sugioka 1984 Analysis61)
bull The proportion of participants experiencing clinical successwith topical piroxicam was 56 (185330 range 49 to 78)
bull The proportion of participants experiencing clinical successwith topical indomethacin was 45 (140311 range 33 to64)
bull The RR of piroxicam compared with indomethacin was 12(11 to 14)
bull The NNT for successful treatment was 91 (53 to 30) Forevery nine participants treated with topical piroxicam onewould experience a clinical success who would not haveexperienced one with topical indomethacin
Topical NSAID versus different topical intervention
One study compared diclofenac gel with a herbal product calledTraumeel gel under double-blind conditions with response ratesfor being pain-free at seven days of 8137 (58) with diclofenacgel and 7140 (50) with Traumeel gel (Gonzaacutelez de Vega 2013)
Local adverse events
Topical NSAID versus oral NSAID
Two studies comparing a topical NSAID with an oral NSAIDprovided data on local adverse events ( kermark 1990 Hosie1993) There were five events with topical NSAID and three withoral NSAID which were too few for analysis
Topical NSAID versus different topical NSAID
All nine studies comparing one topical NSAID with at least oneother reported on local adverse events with a total of 48 events in1005 participants (48) There were sufficient data to compare
only piroxicam with indomethacin (Aoki 1984 Fujimaki 1985Sugioka 1984 Analysis 62)
bull The proportion of participants experiencing local adverseevents with topical piroxicam was 21 (7340 range 12 to28)
bull The proportion of participants experiencing local adverseevents with topical indomethacin was 10 (33331 range 29to 15)
bull The RR of piroxicam compared with indomethacin was021 (009 to 047)
bull The NNT to prevent a local adverse event was 13 (87 to23) For every thirteen participants treated with topicalpiroxicam one would not experience a local adverse event whowould have experienced one with topical indomethacin
Topical NSAID versus different topical intervention
Gonzaacutelez de Vega 2013 reported that adverse events were infre-quent and mild to moderate in intensity but did not distinguishbetween local and systemic events Numbers of participants expe-riencing any adverse event were 8147 with diclofenac gel and 14148 with Traumeel gel
Systemic adverse events
kermark 1990 reported numbers of events rather than numbersof participants with events while Tonutti 1994 and Whitefield2002 reported no adverse events attributable to the study medi-cation and Fioravanti 1999 Gallacchi 1990 Gualdi 1987 andSugioka 1984 did not mention systemic adverse events Gonzaacutelezde Vega 2013 did not distinguish between local and systemicevents In the remaining studies a total of 16 events were reportedin topical NSAID treatment arms (797 participants 2) and 11with ibuprofen tablets (134 participants 8)
Serious adverse events
No serious adverse events were reported in any treatment arm
Withdrawals
The only withdrawals reported due to adverse events were in stud-ies with placebo treatment arms ( kermark 1990 Fujimaki 1985)and have been reviewedThree studies reported withdrawals due to lack of efficacy (Gonzaacutelez de Vega 2013 Hofman 2000 Tonutti 1994) (Appendix5) There were insufficient data for analysisSome studies reported exclusions from analysis (efficacy or safetyor both) following randomisation mainly due to protocol vio-lations or loss to follow-up (Appendix 5) There is no reason tobelieve these exclusions would introduce systematic bias and thenumbers involved were not likely to influence results
24Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
D I S C U S S I O N
This updated review of topical NSAIDs for acute musculoskele-tal pain in adults differs from previous reviews Previously dataallowed only for comparison of individual topical NSAIDs withplacebo irrespective of formulation With a substantial amount ofnew data for diclofenac it is possible to distinguish effects of for-mulation for individual NSAIDs Because formulation chemistrycan substantially affect the rate and total amount of drug accessingsubcutaneous injured tissues the effect of formulation may be asimportant as the individual NSAID used Drug and formulationshould thus be considered together when assessing efficacy andthis is now possible
We have also included an assessment of the daily dose of NSAIDapplied to the skin This involved having information on the con-centration of NSAID in the preparation the amount used andthe frequency of use Not all studies reported all three but an esti-mation of topical doses applied was possible It varied by factors ofthree or four for each topical NSAID However for topical formu-lations the key issue is less the dose applied but the amount thatpenetrates locally (producing analgesic effect) and the amount en-tering the systemic circulation (producing potential harms) Bothwill depend on the exact formulation of the topical agent andwhether there is occlusion (Moore 2008a)
Summary of main results
This review included 61 studies comparing a topical NSAID withplacebo another topical NSAID or an oral NSAID In total5311 participants were treated with a topical NSAID 3470 withplacebo and 220 with an oral NSAID There were 63 moreparticipants than in the previous version of this review Conditionstreated were sprains strains and contusions mainly resulting fromsports injuries and overuse injuries such as tendinitisFormulations of topical diclofenac ibuprofen ketoprofen piroxi-cam and indomethacin demonstrated significantly higher rates ofclinical success than matching topical placebo lacking the NSAIDbenzydamine did not Three drug and formulation combinationshad NNTs for clinical success below 4 For diclofenac Emulgelreghad the lowest NNT of 18 (15 to 21) in two studies using atleast 50 pain intensity reduction as the outcome (high qualityevidence) Diclofenac plasters other than Flectorreg also had a lowNNT of 32 (26 to 42) based on good or excellent responses inrelatively recent studies (high quality evidence) Ketoprofen gelhad an NNT of 25 (20 to 34) from five studies in the 1980ssome with less well defined outcomes (moderate quality evidence)Ibuprofen gel had an NNT of 39 (27 to 67) from two stud-ies with outcomes of marked improvement or complete remission(moderate quality evidence) All other drug and formulation com-binations had NNT values above 4 indicating lesser efficacyThese results are better than alternative topical products that mightbe used for acute musculoskeletal pain There is no evidence tosupport the use of topical salicylate rubefacients (Derry 2014)
Treatment with a topical NSAID was not associated with anincrease in local adverse events (skin reactions) compared withplacebo (inert carrier) or in withdrawals due to adverse events(high quality evidence) The inert carrier was sometimes associ-ated with mild skin irritation but this rarely led to cessation oftreatment and quickly resolved Systemic adverse events were un-common and did not differ between topical NSAID and placebo(high quality evidence) Two participants experienced serious ad-verse events with diclofenac plaster and diclofenac gel but it isunlikely that these were related to the study medicationsThere were insufficient data directly comparing a topical NSAIDwith the same oral NSAID to draw conclusions about efficacyBased on very limited data for oral NSAIDs there were fewersystemic adverse events with topical than oral treatment Therewere sufficient data only for topical piroxicam compared withtopical indomethacin to compare one topical agent with anotherThese limited data suggested that piroxicam was more effectivethan indomethacin and was less likely to cause local adverse eventsIt is worth noting here that topical indomethacin did not givesignificantly better pain relief than placebo in two of the threestudies in this analysis
Overall completeness and applicability ofevidence
There is a tension between pooling studies to produce analyseswith larger numbers and the subsequent large increases in clinicaland statistical heterogeneity on the one hand and using the ap-proach of clinical homogeneity with subsequent smaller numbersof participants on the other hand Previous reviews have takenthe former approach that is useful in demonstrating that topicalNSAIDs rsquoworkrsquo by being significantly better than placebo Becauseof the substantial increase in the amount of data available in thisreview we have chosen to seek greater clinical homogeneity thisproduces results that are more relevant to patient and prescriberchoiceThere were too few studies comparing one topical NSAID versusanother or versus the same oral NSAID to allow meaningful directcomparisons between individual drugs or routes of administrationThe conditions treated in these studies are representative of thoselikely to be suitable for acute treatment with topical NSAIDs Themean age of participants in individual studies ranged from 25 to57 years and the nature of recruitment in many studies meant thatparticipants were actively engaged in sporting activities Neverthe-less older people in their 60s to 80s were also included in somestudies and the low levels of predominantly mild adverse eventsmeans that this route of administration of NSAIDs is suitable forall age groups able to manage the application processInformation from other sources mainly randomised studies last-ing 12 weeks or more in older populations with arthritis tendto confirm this A systematic review of topical NSAIDs in olderadults was difficult to interpret but suggested that the range of
25Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
withdrawal rates in these studies was similar with topical and oralNSAIDs (Makris 2010) It also claimed potentiation of warfarineffects but that was with topical salicylate not an NSAID Incontrast a pooled safety analysis of topical diclofenac in peopleaged 75 years or older reported minimal changes with a mean re-duction in haemoglobin of less than 1 gL with topical diclofenacand a mean systolic blood pressure reduction of almost 4 mm Hg(Roth 2012) Two large randomised 12-week studies comparingtopical with oral diclofenac in arthritis reported lower rates of gas-trointestinal adverse events with topical than oral especially severeevents but larger reductions in haemoglobin with oral diclofenac(Simon 2009 Tugwell 2004)The available evidence was limited by numbers to comment on rarebut potentially serious adverse events One example is the potentialfor photo-sensitivity reactions with topical ketoprofen Currentadvice from the Medicines and Healthcare products RegulatoryAgency in the UK is to avoid direct sunlight ultraviolet (UV) rayssunlamps and sunbeds while using topical ketoprofen and to seea healthcare professional or go to hospital if they experience a skinreaction to sunlight sunlamps or sunbeds (MHRA 2009)
Quality of the evidence
All included studies were both randomised and double-blind nonewas considered at high risk of methodological bias Many werecarried out in the 1980s and 1990s when methodological rigorand detailed reporting were not given such high priority and stud-ies did not always report details of the randomisation treatmentallocation and blinding processes More recent studies often didreport methodological details and tended to be larger (see Figure4 for a comparison of quality of reporting for different dates andNSAIDs) Our primary outcome of clinical success was not alwayswell-defined and was measured using different scales but againmore recent studies tended to report outcomes betterThe studies were conducted in different conditions with some-what different outcome definitions and duration and with differ-ent topical NSAIDs and formulations Moreover the small size ofmany of the studies is likely to result in considerable chance varia-tion (Counsell 1994 Moore 1998b) These factors would accountfor the high I2 values seen in several analyses Despite these sourcesof potential clinical heterogeneity most studies showed benefit oftopical NSAID over placeboThe design of studies to be able to demonstrate analgesic sensi-tivity is important in self limiting conditions such as strains andsprains Too long a duration and the condition results in sponta-neous resolution of painful symptoms while too short a durationmay be inadequate to show any effect The decision by trialists toconcentrate on outcomes closest to seven days of treatment appearsto be prudent and has been adopted in this and previous reviewsThere are potential differences in response to treatment betweenstrains and sprains and overuse-type injuries such as tendinitis
and future reviews may examine this At the present time thereare too few existing trials to explore any differences adequatelyBaseline pain may be a cause for concern Seven studies didnot report baseline pain levels (Billigmann 1996 Curioni 1985Haig 1986 NCT01255423 NCT01272934 NCT01272947Sinniger 1981) and a further 11 reported either mean levels ofless than moderate pain or a significant proportion of individualswith less than moderate pain ( kermark 1990 Aoki 1984 Auclair1989 Diebshlag 1990 Fujimaki 1985 Jenoure 1997 Linde 1985Picchio 1981 Ramesh 1983 Sugioka 1984 Whitefield 2002)using recognised scales Insufficient pain at baseline compromisesthe ability of a study to demonstrate any improvement All thenewly added studies reported baseline pain to be of at least mod-erate intensity
Potential biases in the review process
There has been greater interest in topical NSAIDs in recent yearsmainly because lower systemic drug levels reduce the risk of trou-blesome and severe adverse events particularly in the gastroin-testinal tract and renal and cardiovascular systems Most of theattention has been in chronic conditions such as osteoarthritiswith few studies in acute painful conditions Low levels of seriousadverse events with topical NSAIDs has been noted previously(Evans 1995) and the near absence of serious adverse events inthis review is unlikely to be due to any biases in the review processOne potential bias is that clinical trials for topical NSAIDs may nothave been published One previous review did find previously un-published trials (Moore 1998a) but a subsequent attempt that in-cluded extensive contacts with pharmaceutical companies revealedno additional data (Mason 2004a) While some old unpublishedstudies of topical NSAIDs in acute painful conditions may existthey constitute an unknown number of studies and participantswhose results are unknown and are likely to remain unknownFurthermore their relevance to current clinical practice may belimited as better formulations are developed New systems of trialregistration mean that we know what recent studies have been doneor are ongoing the number of studies and participants is knowneven if their results remain unknown We identified in Clinicaltri-alsgov three unpublished studies (612 participants) with adverseevent data but no dichotomous efficacy data 14 completed un-published studies (4403 participants) with no results posted andthree ongoing studies (880 participants)For the main topical NSAIDs of interest and where most informa-tion exists about 4200 participants in this review provided data onefficacy for diclofenac ibuprofen ketoprofen indomethacin andbenzydamine compared with placebo For efficacy there are un-known results from almost 5900 participants in studies known tohave been done but essentially unpublished Almost 6500 partic-ipants in this review provided information on local adverse eventsfor topical NSAIDs compared with placebo For local adverseevents the unknown results from known studies represents almost
26Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
5300 participants It is clear that identified unpublished but un-available study data amounts to a further potentially large increasein knowledge over and above the 60 increase in numbers ofparticipants already included in this updated reviewBased on efficacy data on known and available study results un-published trials showing no difference between any topical NSAIDand topical placebo and involving 5500 participants would haveto exist in order for the NNT to be as high as 9 at which pointthe effectiveness of topical NSAIDs would become clinically irrel-evant (Moore 2006) This amount of unpublished negative data isobviously available and while a negative result in all the identifiedstudies is unlikely knowledge would be greatly served by havingthese unpublished trial results availableWe have not yet attempted to obtain results from these clinicaltrials from the trial sponsors because this takes a considerableamount of time and may not be successful Moreover the studieswere spilt between nine different sponsors Cerimon Pharmaceu-ticals (five studies) Novartis (four studies) Endo Pharmaceuticals(three studies) GlaxoSmithKline (two studies) Hisamatsu Phar-maceutical (two studies) and one each from Actavis Pfizer Im-primis Pharmaceutical and Strategic Science amp Technologies
Agreements and disagreements with otherstudies or reviews
A review published in 2004 included some of the studies in thisreview and reported an NNT for all topical NSAIDs combinedof 38 (34 to 44) for clinical success equivalent to half pain reliefat seven days (Mason 2004a) That review found no differencebetween topical NSAID and placebo for local adverse events as didthis review In turn the Mason review was in broad agreement withthe original systematic review on topical NSAIDs (Moore 1998a)To our knowledge no previous review assembled sufficient trialdata to analyse results by both drug and formulation as was donehere
A U T H O R S rsquo C O N C L U S I O N S
Implications for practice
For people with acute musculoskeletal pain
Topical NSAIDs can provide good levels of pain relief in acuteconditions such as sprains strains and overuse injuries proba-bly similar to that provided by oral NSAIDs Gel formulations ofdiclofenac (as Emulgelreg) ibuprofen and ketoprofen and somediclofenac patches provide the best effects Adverse events are usu-ally minimal with topical NSAIDs
For clinicians
Topical diclofenac ibuprofen or ketoprofen gels provide goodpain relief for painful acute musculoskeletal conditions and arebetter tolerated than oral formulations These drugs and formu-lations are more likely to be cost effective than alternative topicalpreparations such as topical rubefacients
For policy makers
Topical NSAIDs are not associated with an increased incidence oflocal skin reactions compared with the inert carrier and while thecarrier may cause mild transient irritation it is rarely troublesomeTopical NSAIDs do not cause systemic (mainly gastrointestinal)problems commonly seen with oral NSAIDs making them partic-ularly useful for individuals unable to tolerate oral administrationor for whom it is contraindicated
For funders
Topical diclofenac ibuprofen or ketoprofen gels should be consid-ered for initial treatment of acute musculoskeletal painful condi-tions where there are no contraindications such as damaged skinThese drugs and formulations are more likely to be cost effectivethan alternative topical preparations such as topical rubefacientsBecause formulations of topical NSAIDs are likely to change overtime the relevant trials performed and reported in or before the1990s must be limited and may be questionable Funders mightwish to consider asking pharmaceutical companies without recenttrial evidence for their products to produce it
Implications for research
General
The general thrust of these findings is that gel formulations of top-ical diclofenac ibuprofen and ketoprofen work best but for somedrugs (ketoprofen for instance) studies were pre-1990 Thesestudies may not be relevant to products available now Becauseformulation can have a significant effect on efficacy formulationchanges should be accompanied by relevant randomised trials
Design
The design of the trials is generally good and the sports injurymodel appears to be reliable and reproducible Modern studieshave ensured that participants entering the trials have at least mod-erate pain and this helps sensitivity to detect an analgesic responseMajor changes to the design of these trials would not appear to beneeded
27Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Measurement (endpoints)
A major issue is not in the measurement of pain as most studiesespecially modern studies have used standard pain intensity andpain relief scales However reporting of average pain changes isinadequate and the use of responder analyses (at least 50 painintensity reduction or people experiencing mild or no pain) ispreferred
Comparison between active treatments
Indirect comparisons with placebo are probably as informative asuse of an active comparator
A C K N O W L E D G E M E N T S
The Oxford Pain Relief Trust the NHS Cochrane CollaborationProgramme Grant Scheme and the NIHR Biomedical ResearchCentre Programme provided support for the earlier review
The National Institute for Health Research (NIHR) is the largestsingle funder of the Cochrane Pain Palliative and SupportiveCare Review Group Disclaimer the views and opinions expressedherein are those of the authors and do not necessarily reflect thoseof the NIHR National Health Service (NHS) or the Departmentof Health
R E F E R E N C E S
References to studies included in this review
Airaksinen 1993 published data onlyAiraksinen O Venaumllaumlinen J Pietilaumlinen T Ketoprofen 25gel versus placebo gel in the treatment of acute soft tissueinjuries International Journal of Clinical PharmacologyTherapy and Toxicology 199331(11)561ndash3
kermark 1990 published data only
kermark C Forsskaringhl B Topical indomethacin in overuseinjuries in athletes A randomized double blind studycomparing Elmetacinreg with oral indomethacin andplacebo International Journal of Sports Medicine 199011(5)393ndash6
Aoki 1984 published data only
Aoki T Numajiri M Yamamoto M A well controlledcomparative study of piroxicam gel indomethacin geland placebo gel in the treatment of trauma JapanesePharmacology and Therapeutics 198412(12)101ndash17
Auclair 1989 published data onlyAuclair J Georges M Grapton X Gryp L DrsquoHooghe MMeisser RG et al A double-blind controlled multi-centerstudy of percutaneous niflumic acid gel and placebo in thetreatment of Achilles heel tendinitis Current TherapeuticResearch 198946(4)782ndash8
Billigmann 1996 published data onlyBilligmann PW Treatment of ankle distortion withibuprofen gel [Therapie von Sprunggelenksndashdistosionenmit ibuprofenndashmikrogel] Therapiewoche 199646(21)1187ndash92
Campbell 1994 published data onlyCampbell J Dunn T Evaluation of topical ibuprofen creamin the treatment of acute ankle sprains Journal of Accidentand Emergency Medicine 199411(3)178ndash82
Chatterjee 1977 published data onlyChatterjee DS A double-blind clinical study withbenzydamine 3 cream on soft tissue injuries in anoccupational health centre Journal of International MedicalResearch 19775(6)450ndash8
Costantino 2011 published data only
Costantino C Kwarecki J Samokhin AV Mautone GRovati S Diclofenac epolamine plus heparin plasterversus diclofenac epolamine plaster in mild to moderateankle sprain a randomized double-blind parallel-groupplacebo-controlled multicentre phase III trial ClinicalDrug Investigation 201131(1)15ndash26 [DOI 10216511585890-000000000-00000]
Coudreuse 2010 published data onlyCoudreuse JM de Vathaire F Effect of a plaster containingDHEP and heparin in acute ankle sprains with oedemaa randomized double-blind placebo-controlled clinicalstudy Current Medical Research and Opinion 201026(9)2221ndash8 [DOI 101185030079952010508020]
Curioni 1985 published data only
Curioni BG Di Domenica F Daolio P Spignoli GEvaluation of ibuproxam gel in traumatology [Valutazionedellrsquoefficacia terapeutica di ibudros gel in traumatologia]Clinica Europea 198524(3)456ndash60
Diebshlag 1990 published data only
Diebschlag W Nocker W Bullingham R A double-blindstudy of the efficacy of topical ketorolac tromethamine gelin the treatment of ankle sprain in comparison to placeboand etofenamate Journal of Clinical Pharmacology 199030
(1)82ndash9
Dreiser 1988 published data onlyDreiser RL Clinical trial of efficacy and tolerability oftopical ibuprofen in the treatment of tendinitis Le JournalInternational De Meacutedecine 198811915ndash31
Dreiser 1989 published data only
Dreiser RL Clinical trial - Fatsum gel FG-6 Supplied byMenarini 1989
Dreiser 1990 published data onlyDreiser RE Charlot J Lopez A Ditisheim A Clinicalevaluation of niflumic acid gel in the treatment ofuncomplicated ankle sprains Current Medical Research andOpinion 199012(2)93ndash9
28Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Dreiser 1994 published data only
Dreiser RL Roche R de Sahb R Thomas F Leutenegger EFlurbiprofen local action transcutaneous (LAT tm) clinicalevaluation in the treatment of acute ankle sprains EuropeanJournal of Rheumatology and Inflammation 199414(4)9ndash13
Fioravanti 1999 published data onlyFioravanti A Cicero MR Nerucci F Manopulo RMarcolongo R Double-blind controlled clinical study of theefficacy and tolerability of diclofenac-N-(2-hydroxyethyl)-pyrrolidine lecithin gel compared with diclofenac-N-(2-hydroxyethyl)-pyrrolidine gel in patients with periand extraarticular inflammatory diseases Drugs underExperimental and Clinical Research 199925(5)235ndash40
Fujimaki 1985 published data only
Fujimaki E et al Clinical evaluation of piroxicam gelversus indomethacin gel and placebo in the treatment ofmuscle pain a double-blind multicenter study JapanesePharmacology and Therapeutics 198512(12)119ndash37
Gallacchi 1990 published data only
Gallacchi G Mautone G Lauldi P Topical treatmentwith diclofenac hydroxyethylpyrrilidine (Flector gel 1)Clinical Trials Journal 199027(1)58ndash64
Gonzaacutelez de Vega 2013 published data only
Gonzaacutelez de Vega C Speed C Wolfarth B Gonzaacutelez JTraumeel vs diclofenac for reducing pain and improvingankle mobility after acute ankle sprain a multicentrerandomised blinded controlled and non-inferiority trialInternational Journal of Clinical Practice 201367(10)979ndash89 [DOI 101111ijcp12219]
Governali 1995 published data only
Governali E Casalini D A controlled clinical studyon ketoprofen gel 5 versus ketoprofen ointment 1in patients with post-traumatic lesions [Ricerda clinicacontrollata tra ketoprofene gel 5 e ketoprofene crema1 in pazienti con postumi di lesioni traumatiche]Riabilitazione 199528(1)61ndash9
Gualdi 1987 published data only
Gualdi A Bonollo L Martini A Forgione A Non-steroidal anti-inflammatory drugs for topical therapy intraumatology a double-blind study with flunoxaprofen andketoprofen [Antinflammatori no steroidi per uso topicoin traumatologia studio clinico con flunoxaprofene echetoprofene] Riforma Medica 1987102(10)401ndash4
Haig 1986 published data only
Haig G Portable thermogram technique for topicallyapplied benzydamine cream in acute soft-tissue injuriesInternation Journal of Tissue Reactions 19868(2)145ndash7
Hoffmann 2012 published data only
Hoffmann P Kopa ka P Gugliotta B Rovati S Efficacyand tolerability of DHEP-heparin plaster in reducingpain in mild-to-moderate muscle contusions a double-blind randomized trial Current Medical Researchand Opinion 201228(8)1313ndash21 [DOI 101185030079952012709182]
Hofman 2000 published data only
Hofman J Nasswetter G Cayetti LM Lysine clonixinate gelin soft tissue injuries Controlled randomized prospectivedouble-blind clinical trial with diclofenac [Clonixinatode lisina gel en lesiones de tejidos blandos ensayo clinicoprospectivo doble ciego al azar controlado con diclofenac]Prensa Medica Argentina 200087(5)513ndash20
Hosie 1993 published data only
Hosie GAC The topical NSAID felbinac versus oralibuprofen a comparison of efficacy in the treatment ofacute lower back injury British Journal of Clinical Research199345ndash17
Jenoure 1997 published data only
Jenoure PJ Rostan A Gremion G Meier JL Grossen RBielinki R et al Multicentre double-blind controlledclinical study on the efficacy of diclofenac epolamineTissugel plaster in patients with epicondylitis [Studiomulticentrico controllato in doppio cieco su diclofenacTissugel plaster in pazienti con epicondilite] MedicinaDello Sport 197750(3)285ndash92
Joussellin 2003 published data onlylowast Joussellin E Flector Tissugel for the treatment ofpainful ankle sprains [Flector Tissugel dans le traitementdes entorses douloureuses de la cheville] Journal deTraumatologie du Sport 2003201S5ndash9Lionberger DR Joussellin E Lanzarotti A Yanchick JMagelli M Diclofenac epolamine topical patch relieves painassociated with ankle sprain Journal of Pain Research 2011447ndash53 [DOI 102147JPRS15380]Lionberger DR Joussellin E Yanchick J Magelli MLanzarotti A Pooled analysis of clinical trial data evaluatingthe safety and effectiveness of diclofenac epolamine topicalpatch 13 for the treatment of acute ankle sprain OpenAccess Journal of Sports Medicine 2011275ndash84 [DOI102147OAJSMS17048]
Julien 1989 published data onlyJulien D Clinical trial - Fastum gel FG-8 Supplied byMenarini 1989
Klainguti 2010 published data onlyKlainguti A Forgacs A Berkes I Castellacci E A plastercontaining DHEP and heparin for mild to moderatecontusions and sprains with haematoma a double-blind randomized study Current Medical Researchand Opinion 201026(9)2243ndash51 [DOI 101185030079952010508022]
Kockelbergh 1985 published data onlyKockelbergh M Verspeelt P Caloine R Dermaux F Localanti inflammatory treatment with a ketoprofen gel currentclinical findings Journal Belge de Medecine Physique et deRehabilitation 19858(4)205ndash13
Kuehl 2011 published data only
Kuehl K Carr W Yanchick J Magelli M Rovati SAnalgesic efficacy and safety of the diclofenac epolaminetopical patch 13 (DETP) in minor soft tissue injuryInternational Journal of Sports Medicine 201132(8)635ndash43[DOI 101055s-0031-1275359]
29Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Li 2013 published data only
Li C Frangione V Rovati S Zheng Q Diclofenacepolamine medicated plaster in the treatment of minor softtissue injuries a multicenter randomized controlled trialCurrent Medical Research and Opinion 201329(9)1137ndash46[DOI 101185030079952013816669]
Linde 1985 published data only
Linde F Hvass I Juumlrgensen U Madsen F Treatment ofsprained ankles with 5 benzydamine creme A double-blind study [Ankelforstuvninger behandlet med benzydamin5 creme] Ugeskrift for Laeger 1985148(1)12ndash3
Machen 2002 published data onlyMachen J Whitefield M Efficacy of a proprietary ibuprofengel in soft tissue injuries a randomised double blindplacebo controlled study International Journal of ClinicalPharmacology 200256(2)102ndash6
Mahler 2003 published data only
Mahler P Mahler F Duruz H Ramazzina M Liguori VMautone G Double-blind randomized controlled studyon the efficacy and safety of a novel diclofenac epolaminegel formulated with lecithin for the treatment of sprainsstrains and contusions Drugs under Experimental andClinical Research 200329(1)45ndash52
Maziegraveres 2005a published data onlyMaziegraveres B Rouanet S Velicy J Scarsi C Reiner V Topicalketoprofen patch (100 mg) for the treatment of anklesprain a randomized double-blind placebo-controlledstudy American Journal of Sports Medicine 200533(4)515ndash23 [DOI 1011770363546504268135]
Maziegraveres 2005b published data onlyMaziegraveres B Rouanet S Guillon Y Scarsi C Reiner VTopical ketoprofen patch in the treatment of tendinitisa randomized double blind placebo controlled studyJournal of Rheumatology 200532(8)1563ndash70
McLatchie 1989 published data onlyMcLatchie GR McDonald M Lawrence GF RogmansD Lisai P Hibberd M Soft tissue trauma a randomisedcontrolled trial of the topical application of felbinac a newNSAID British Journal of Clinical Practice 198943(8)277ndash80
Morris 1991 published data only
Morris WD Scott HV Peters WA Ketelbey JW Felbinactopical gel for acute soft tissue sports injuries New ZealandJournal of Sports Medicine 19911945ndash7
NCT01255423 unpublished data only
Novartis (Sponsors) A randomized double-blind multi-center placebo-controlled parallel group study to evaluatethe efficacy and safety of diclofenac sodium topical gel(DSG) 1 applied 4 times daily in subjects with acuteankle sprain clinicaltrialsgovct2showNCT01255423(accessed 3 February 2015) 2012 [CTG NCT01255423]
NCT01272934 unpublished data only
Novartis (Sponsors) A randomized double-blind multi-center placebo-controlled parallel group study to evaluatethe efficacy and safety of diclofenac sodium topical gel
(DSG) 1 applied four times daily in subjects withacute ankle sprain clinicaltrialsgovct2showresultsNCT01272934 (accessed 3 February 2015) 2013 [CTGNCT01272934]
NCT01272947 unpublished data onlylowast Novartis (Sponsors) A randomized double-blindmulti-center placebo-controlled parallel group study toevaluate the efficacy and safety of diclofenac sodium topicalgel (DSG) 1 applied four times daily in subjects withacute blunt soft tissue injuriescontusions of the limbsclinicaltrialsgovct2showresultsNCT01272947 (accessed3 February 2015) 2012 [CTG NCT01272947]Pallay R Pabst H Giannetti B Burnett I Monnet JDiclofenac sodium topical gel (DSG) 1 for acute softtissue injuriescontusions of the limbs 2013 AnnualAssembly of the American Academy of Physical Medicineand Rehabilitation National Harbor MD United States2013 Vol 5 (9 Suppl 1)S204
Noret 1987 published data only
Noret A Roty V Allington N Hauters P Zuinen C PoelsR Ketoprofen gel as topical treatment for sport injuriesActa Therapeutica 198713367ndash78
Parrini 1992 published data onlyParrini M Cabitza P Arrigo A Vanasia M Efficacy andtolerability of ketoprofen lysine salt foam for topical use inthe treatment of traumatic pathologies of the locomotorapparatus [Efficacia e tollerabilita del ketoprofene sale dilisina schiuma per uso topico nel trattamento di alcunepatologie traumatiche dellrsquoapparato locomotore] La ClinicaTerapeutica 1992141(9)199ndash204
Picchio 1981 published data only
Picchio AA Volta S Longoni A Controlled clinical trial ofibuprofen for topical use in sport injuries [Studio clinicocontrollato sullrsquoimpiego dellrsquoibuprofen per uso topico intraumatologica sportiva] Medicina dello Sport 198134403ndash6
Predel 2004 published data only
Mueller EA Kirch W Reiter S Extent and time courseof pain intensity upon treatment with a topical diclofenacsodium patch versus placebo in acute traumatic injury basedon a validated end point post hoc analysis of a randomizedplacebo-controlled trial Expert Opinion on Pharmacotherapy201011(4)493ndash8 [DOI 10151714656560903535898]lowast Predel HG Koll R Pabst H Dieter R Gallacchi GGiannetti B et al Diclofenac patch for topical treatmentof acute impact injuries a randomised double blindplacebo controlled multicentre study British Journal ofSports Medicine 200438(3)318ndash23 [DOI 101136bjsm2003005017]
Predel 2012 published data onlyPredel HG Hamelsky S Gold M Giannetti B Efficacy andsafety of diclofenac diethylamine 232 gel in acute anklesprain Medicine and Science in Sports and Exercise 201244
(9)1629ndash36 [DOI 101249MSS0b013e318257ed41]
30Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Predel 2013a published data only
Predel HG Giannetti B Seigfried B Novellini R MenkeG A randomized double-blind placebo-controlledmulticentre study to evaluate the efficacy and safetyof diclofenac 4 spray gel in the treatment of acuteuncomplicated ankle sprain Journal of InternationalMedical Research 201341(4)1187ndash202 [DOI 1011770300060513487639]
Predel 2013b published data onlyPredel HG Giannetti B Pabst H Schaefer A Hug AMBurnett I Efficacy and safety of diclofenac diethylamine116 gel in acute neck pain a randomized double-blindplacebo-controlled study BMC Musculoskeletal Disorders201314250 [DOI 1011861471-2474-14-250]
Ramesh 1983 published data only
Ramesh N Steuber U Ibuprofen in a cream vehicle foraccidental and sport injuries [Dolgit creme bei unfallndash undsportverletzungen] Therapiewoche 1983334563ndash70
Rowbotham 2003 published data only
Rowbotham M Galer B Block J Backonja M FlectorTissugel efficacy and safety in minor sport injuries Avs controlled clinical trial Journal of Sports Traumatology200320IS15ndashIS20
Russell 1991 published data only
Russell AL Piroxicam 05 topical gel compared to placeboin the treatment of acute soft tissue injuries a double-blind study comparing efficacy and safety Clinical andInvestigative Medicine 199114(1)35ndash43
Saillant 1998 published data only
Lionberger DR Joussellin E Yanchick J Magelli MLanzarotti A Pooled analysis of clinical trial data evaluatingthe safety and effectiveness of diclofenac epolamine topicalpatch 13 for the treatment of acute ankle sprain OpenAccess Journal of Sports Medicine 2011275ndash84 [DOI102147OAJSMS17048]lowast Saillant G Study comparing the efficacy and toleranceof Flector Tissugelreg to that of a placebo in the treatmentof benign ankle sprains [Eacutetude comparant lrsquoefficaciteacute et latoleacuterance de FlectorTissugelreg agrave celles drsquoun placebo dans letraitement des entorses beacutenignes de la cheville] Medicinedu Sport 1998721ndash5
Sanguinetti 1989 published data onlySanguinetti C Treatment of soft tissue injury with BPAAgel Results of an Italian multicenter study vs placebo[Trattemento con BPAA gel dei traumi dei tessuti moli]Clinica Terapeutica 1989130(5)255ndash258
Sinniger 1981 published data onlySinniger M Blanchard P Controlled clinical trial withFentiazac cream in sport microtraumatology Journal ofInternational Medical Research 19819(4)300ndash2
Spacca 2005 published data onlySpacca G Cacchio A Forgcs A Monteforte PRovetta G Analgesic efficacy of a lecithin-vehiculateddiclofenac epolamine gel in shoulder periarthritis andlateral epicondylitis a placebo-controlled multicenter
randomized double-blind clinical trial Drugs underExperimental and Clinical Research 200531(4)147ndash54
Sugioka 1984 published data only
Sugioka Y Multicenter clinical evaluation of piroxicam gelvs indomethacin gel in the treatment of non-traumaticdiseases of tendon or muscle Japanese Pharmacology andTherapeutics 198412139ndash53
Thorling 1990 published data only
Thorling J Linden B Berg R Sandahl A A double blindcomparison of naproxen gel and placebo in the treatment ofsoft tissue injuries Current Medical Research and Opinion199012242ndash8
Tonutti 1994 published data only
Tonutti A The use of ketoprofen gel 5 (orudis gel) intraumatology controlled double-blind study vs etofenamate[Utilizzazione del ketoprofene gel 5 (orudis gel) nellapratica traumatologica studio in doppio cieco controllatoverso etofenamato] Ortopedia e Traumatologia Oggi 199414(3)119ndash25
Vecchiet 1991 published data only
Vecchiet L Colozzi A Effects of meclofenamic acid in thetreatment of lesions deriving from minor traumatologyClinical Journal of Pain 19917 Suppl 1S54ndash9
Whitefield 2002 published data only
Whitefield M OrsquoKane CJA Anderson S Comparativeefficacy of a proprietary topical ibuprofen gel and oralibuprofen in acute soft tissue injuries a randomised doubleblind study Journal of Clinical Pharmacy and Therapeutics200227(6)409ndash17
References to studies excluded from this review
Ambrus 1987 published data only
Ambrus P Boumlhmer D Mobilat ointment in acute sprains[Mobilat Salbe bei akuten Distorsionen] Fortschritte derMedizin 1987105(13)259ndash62
Anon 1993 unpublished data only
Anon Comparative clinical efficacy of Oruvail piroxicamand diclofenac gels in soft tissue injury Unpublished
Ascherl 1982 published data onlyAscherl R Schlemmer H Blumel G Lechner F Theeffectiveness of etofenamate in minor sports injuries ofthe knee and ankle joint A double blind study [DieWirksamkeit der Etofenamat in kleineren Sportverletzungenam Knie und Sprunggelenk Eine DoppelblindndashStudie]Fortschritte der Medizin 1982100(37)1729ndash34
Bagliani 1976 published data only
Bagliani A Montalbetti L Topical treatment ofthrombophlebitis with feprazone and benzydamineControlled clinical study [Il trattamento topico ditromboflebite con feprazone e benzidamina Studio clinicocontrollato] Minerva Medica 197667(14)880ndash4
Baracchi 1982 published data only
Baracchi G Messina Denaro S Piscini S Experience ofthe topical use of isobutylfenylproprionic acid (Ibuprofen)in traumatic inflammation A double-blind comparison
31Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
with placebo [Esperienza sullrsquoimpiego topico dellrsquoacidoisobutilndashfenilndashpropionico (ibuprofen) nella flogositraumatica Confronto a doppia cecita con placebo]Gazzetta Medica Italiana 1982141(12)691ndash4
Boumlhmer 1995 published data onlyBoumlhmer D Ambrus P Treatment of muscular injurieswith diclofenac-diethylammonium emugel [Behandlungvon muskelverletzungen mit diclofenacndashdiethylammoniumemugel] Sportverletzung Sportschaden 1995994ndash5
Burnham 1998 published data only
Burnham R Gregg R Healy P Steadward R Theeffectiveness of topical diclofenac for lateral epicondylitisClinical Journal of Sport Medicine 19988(2)78ndash81
Cesarone 2008 published data onlyCesarone MR Belcaro G Pellegrini L Ledda A VinciguerraG Ricci A et al Treatment of ankle sprain in patientswith vascular diseases of the lower limbs MinervaCardioangiologica 200856 (5 Suppl)39ndash46
Coulibaly 2009 published data only
Coulibaly SK Courau S Staiger C The sprained anklecomparative study of a comfreyroot extract ointment versusa diclofenac gel [Entorses de la cheville eacutetude comparativeentre un extrait agrave base de racines de consoude et lediclofeacutenac] Phytotheacuterpie 20097147ndash9 [DOI 101007s10298-009-0383-x]
Diebschlag 1985 published data only
Diebschlag W Benzydamine cream in post-traumaticoedema International Journal of Tissue Reactions 19857(3)219ndash23
Diebschlag 1986 published data onlyDiebschlag W Diclofenac in blunt traumatic ankle jointswelling Volumetric monitoring in a placebo controlleddouble blind trial [Diclofenac in stumpfen traumatischenSprunggelenk Schwellung Volumetrische Uumlberwachungin einer Placebondashkontrollierten DoppelblindndashStudie]Fortschritte der Medizin 1986104(21)437ndash40
Diebschlag 1992 published data onlyDiebschlag W Nocker W Lehmacher W Treatment of acutesprains of the ankle joint A comparison of the effectivenessand tolerance of two gel preparations containingindomethacin [Die Behandlung der akuten Verstauchungendes Sprunggelenks Ein Vergleich der Wirksamkeitund Vertraumlglichkeit von zwei GelndashZubereitungen mitIndometacin] Fortschritte der Medizin 1992110(6)64ndash72
Fantato 1971 published data only
Fantato S De Gregorio M Clinical evaluation of topicalbenzydamine in traumatology Arzneimittel-Forschung197121(10)1530ndash5
Galer 2000 published data onlyGaler BS Rowbotham M Perander J Devers A FriedmanE Topical diclofenac patch relieves minor sports injury painresults of a multicenter controlled clinical trial Journal ofPain and Symptom Management 200019(4)287ndash94
Hallmeier 1986 published data only
Hallmeier B Michelbach B Etofenamate under tapebandages - a controlled study [Etofenamat untertapendashverbaumlnden] Medizinische Welt 198637(43)1344ndash8
Hallmeier 1988 published data onlyHallmeier B Efficacy and tolerance of etofenamate anddiclofenac in acute sports injuries [Wirksamkeit undVertraumlglichkeit von Etofenamat und Diclofenac bei akutenSportverletzungen] Rheuma 19888183ndash6
Kaneko 1999 published data only
Kaneko M Shimojo H Saito H Onuma Y YamashitaK Clinical evaluation of felbinac patch (SELSPOT) onpost-traumatic disease clinical comparative study versuscommercially available patch Japanese Pharmacology andTherapeutics 19992775ndash85
Kockelbergh 1985b published data only
Kockelbergh M Verspeelt P Caloine R Dermaux F Localanti-inflammatory treatment with a ketoprofen gel currentclinical findings [Traitement antindashinflammatoire local parun gel de keacutetoprofegravene donneacutees cliniques reacutecentes] JournalBelge de Medecine Physique et de Rehabilitation 19858(4)205-13 (study 2)
Kuwabara 2013 published data only
Kuwabara Y Hamamoto H Hikake S Miwa Y Arandomized multi-center double-blind placebo-controlledphase IIIII trial to evaluate the eftopical patch in thetreatment of pain Journal of Pain 2013 Vol 14 (4 Suppl1)S73
Lee 1991 published data only
Lee EH Lee PY Ngai AT Chiu EH Treatment of acute softtissue trauma with a topical non-steroidal anti-inflammatorydrug (biphenylacetic acid 3 gel) Singapore MedicalJournal 199132(4)238ndash41
Link 1996 published data onlyLink R Balint G Pavlik G Otto J Krause W Topicaltreatment of soft tissue rheumatism and athletic injuriesEffectiveness and tolerance of a new ketoprofen gel[Topische Behandlung von WeichteilndashRheumatismus undSportverletzungen Wirksamkeit und Vertraumlglichkeit einesneuen Ketoprofen Gel] Fortschritte der Medizin 1996114(25)311ndash4
May 2007 published data onlyMay JJ Lovell G Hopkins WG Effectiveness of1 diclofenac gel in the treatment of wrist extensortenosynovitis in long distance kayakers Journal of Scienceand Medicine in Sport 200710(1)59ndash65
Oakland 1993 published data only
Oakland C Rapier C A comparison of the efficacy of thetopical NSAID felbinac and ultrasound in the treatmentof acute ankle injuries British Journal of Clinical Research1993489ndash96
Odaglia 1987 unpublished data only
Odaglia G Sereni G Sports minor traumatology results ofa double-blind controlled clinical study ketoprofen (fastumgel 25) versus placebo Menarini unpublished data
32Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Picardi 1993 published data only
Picardi E De Iasio R Efficacy of percutaneous anti-inflammatory drugs (NSAIDs) in swimmers and waterpoloplayers [Efficacia dei farmaci antinfiammatori (FANS) pervia percutanea in atleti di nuoto e pallanuoto] ClinicaTerapeutica 1993143(6)507ndash9
Taboada 1992 published data onlyTaboada A Controlled trial of piroxicam gel associated withultrasound in acute disturbances of the locomotive system[Experienca controlada con gel de piroxicam asociado aultrasonidos en afecciones agudas del aparato locomotor]Prensa Medica Argentina 199279(10)630ndash2
Vanderstraeten 1990 published data only
Vanderstraeten G Schuermans P Study on the effect ofetofenamate 10 cream in comparison with an oral NSAIDin strains and sprains due to sports injuries Acta BelgicaMedica Physica 199013(3)139ndash41
Vinciguerra 2008 published data only
Vinciguerra G Belcaro G Cesarone MR Errichi BM DiRenzo A Errichi S et al Management of uncomplicatedankle sprains with topical or oral ketoprofen treatmentA registry study Minerva Cardioangiologica 200856 (5
Suppl)47ndash53
Von Klug 1977 published data onlyVon Klug H Experience with a locally applied anti-inflammatory drug [Erfahrungen mit einem localanwendbaren antirheumatikum] Arzneimittel-ForschungDrug Research 1977271350ndash4
References to studies awaiting assessment
NCT00351104 published data onlyEndo Pharmaceuticals (Sponsors) A randomized double-blind placebo-controlled parallel group phase III study ofthe efficacy tolerability and safety of ketoprofen topicalpatch 20 (KTP) in the treatment of pain associated withgrade 1 or grade 2 ankle sprain or strain clinicaltrialsgovct2showNCT00351104 (accessed 3 February 2015) 2008[CTG NCT00351104]
NCT00352625 unpublished data only
Endo Pharmaceuticals (Sponsors) A randomized double-blind placebo-controlled parallel group phase III study ofthe efficacy tolerability and safety of ketoprofen topicalpatch 20 (KTP) in the treatment of pain associatedwith tendonitis or bursitis of the shoulder elbow or kneeclinicaltrialsgovct2showNCT00352625 (accessed 3February 2015) 2008 [CTG NCT00352625]
NCT00426985 unpublished data only
Endo Pharmaceuticals (Sponsors) A randomized double-blind placebo-controlled parallel group phase iii study ofthe efficacy tolerability and safety of ketoprofen topicalpatch 20 (KTP) in the treatment of pain associatedwith tendonitis or bursitis of the shoulder elbow or kneeclinicaltrialsgovct2showNCT00426985 (accessed 3February 2015) 2008 [CTG NCT00426985]
NCT00640705 unpublished data only
Cerimon Pharmaceuticals (Sponsors) A randomizeddouble-blind placebo-controlled study of the efficacyand safety of a diclofenac sodium patch for the topicaltreatment of pain due to mild to moderate ankle sprainclinicaltrialsgovct2showNCT00640705 (accessed 3February 2015) 2008 [CTG NCT00640705]
NCT00640939 unpublished data onlyCerimon Pharmaceuticals (Sponsors) A randomizeddouble-blind placebo-controlled study of the efficacy andsafety of a diclofenac sodium patch for the topical treatmentof pain due to mild to moderate tendonitis or bursitisclinicaltrialsgovct2showNCT00640939 (accessed 3February 2015) 2008 [CTG NCT00640939]
NCT00680472 unpublished data onlyHisamitsu Pharmaceutical Co Inc (Sponsors) Arandomized multicenter double-blind placebo-controlledtwo-week study to assess the efficacy and safety of HKT-500 in subjects with acute shoulder pain clinicaltrialsgovct2showNCT00680472 (accessed 3 February 2015) 2015[CTG NCT00680472]
NCT00680784 unpublished data only
Hisamitsu Pharmaceutical Co Inc (Sponsors) Arandomized multicenter double-blind placebo-controlledstudy to evaluate the efficacy and safety of HKT-500 inthe treatment of pain associated with Grade I or Grade IIankle sprain clinicaltrialsgovct2showNCT00680784(accessed 3 February 2015) 2015 [CTG NCT00680784]
NCT00765700 unpublished data onlyEkman E Skrepnik S Jones M Lawson K Schupp JEfficacy and safety of ketoprofen 10 cream in acute softtissue injuries (phase 3 study tdlp-110-001) Proceedings ofthe 13th World Congress on Pain 2010 Aug 29 - Sept 2Montreal CanadaImprimis Pharmaceuticals Inc (Sponsors) A randomizedmulticenter double-blind placebo-controlled parallel-group phase 3 study to assess the efficacy and safetyof Ketotransdeltrade (Ketoprofen topical cream 10) inthe treatment of pain associated with mild to moderateacute soft tissue injury clinicaltrialsgovct2showNCT00765700 (accessed 3 February 2015) 2013 [CTGNCT00765700]
NCT00869063 unpublished data onlyCerimon Pharmaceuticals (Sponsors) A randomizeddouble-blind placebo-controlled study of the efficacy andsafety of a diclofenac sodium patch for the topical treatmentof acute pain due to mild to moderate wrist sprain strainor contusion clinicaltrialsgovct2showNCT00869063(accessed 3 February 2015) 2010 [CTG NCT00869063]
NCT00869180 unpublished data onlyCerimon Pharmaceuticals (Sponsors) A randomizeddouble-blind placebo-controlled study of the efficacy andsafety of a diclofenac sodium patch for the topical treatmentof acute pain due to mild to moderate ankle sprainclinicaltrialsgovct2showNCT00869180 (accessed 3February 2015) 2010 [CTG NCT00869180]
33Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT00931866 unpublished data only
Cerimon Pharmaceuticals (Sponsors) A randomizeddouble-blind placebo-controlled study of the efficacy andsafety of a diclofenac sodium patch for the topical treatmentof acute pain due to mild to moderate soft tissue injuriesclinicaltrialsgovct2showNCT00931866 (accessed 3February 2015) 2010 [CTG NCT00931866]
NCT01874626 unpublished data only
Strategic Science amp Technologies LLC (Sponsors) Arandomized double-blind placebo-controlled multi-dosepivotal study to determine the efficacy and safety of SST0225 a topical ibuprofen cream in the treatment of painassociated with acute ankle sprain clinicaltrialsgovct2showNCT01874626 (accessed 3 February 2015) 2014[CTG NCT01874626]
NCT01957215 unpublished data only
GlaxoSmithKline (Sponsors) A clinical study to assess theefficacy of pain relief of topical indomethacin patch overplacebo in ankle sprain patients clinicaltrialsgovct2showNCT01957215 (accessed 3 February 2015) 2015 [CTGNCT01957215]
NCT02324270 unpublished data only
Actavis Inc (Sponsors) Randomized double-blindmultiple-center placebo-controlled study comparing thesafety and efficacy of generic diclofenac epolamine toFlectorreg patch in the treatment of acute pain due to minorankle sprain clinicaltrialsgovct2showNCT02324270(accessed 3 February 2015) 2014 [CTG NCT02324270]
Sarzi-Puttini 2014 published data only
Sarzi-Puttini P Atzeni F Damiani C Casale R BarbagalloM Cazzola M A double-blind randomised parallelgroup active controlled multicentre study to assess thetherapeutic non-inferiority of SKP-021 a 03 ketoprofenpatch versus diclofenac sodium patch in patients withacute inflammatory musculoskeletal injuries Proceeding ofthe Annual European Congress of Rheumatology of theEuropean League Against Rheumatism EULAR 2014 June11-14 Paris France 201473
References to ongoing studies
NCT01945034 unpublished data onlyPfizer (Sponsors) Placebo-controlled double-blindevaluation of the efficacy and safety of ibuprofen 5 topicalgel for the treatment of ankle sprain clinicaltrialsgovct2showNCT01945034 (accessed 3 February 2015) 2014[CTG NCT01945034]
NCT02100670 unpublished data onlyGlaxoSmithKline (Sponsors) A clinical study to assessthe efficacy and onset of pain relief of topical MFC51123diclofenac-menthol gel versus controls in ankle sprainclinicaltrialsgovct2showNCT02100670 (accessed 3February 2015) 2014 [ClinicalTiralsgov NCT02100670]
NCT02290821 unpublished data only
Novartis (Sponsors) A randomized double-blind placebo-controlled parallel group study to evaluate the efficacy andsafety of diclofenac sodium topical gel (DSG) 1 applied
four times daily in subjects with acute blunt soft tissueinjuriescontusions of the limbs clinicaltrialsgovct2showNCT02290821 (accessed 3 February 2015) 2015 [CTGNCT02290821]
Additional references
Anon 2005
Anon Topical analgesics a review of reviews and a bit ofperspective wwwjr2oxacukBandolierExtraforbandoTopextra3pdf 2005
Cook 1995
Cook RJ Sackett DL The number needed to treat aclinically useful measure of treatment effect BMJ 1995310
(6977)452ndash4
Counsell 1994Counsell CE Clarke MJ Slattery J Sandercock PA Themiracle of DICE therapy for acute stroke fact or fictionalproduct of subgroup analysis BMJ 1994309(6970)1677ndash81
Dechartres 2013Dechartres A Trinquart L Boutron I Ravaud P Influenceof trial sample size on treatment effect estimates meta-epidemiological study BMJ 2013346f2304 [DOI101136bmjf2304]
Derry 2012a
Derry S Moore RA Topical capsaicin (low concentration)for chronic neuropathic pain in adults Cochrane Databaseof Systematic Reviews 2012 Issue 9 [DOI 10100214651858CD010111]
Derry 2012bDerry S Moore RA Rabbie R Topical NSAIDs forchronic musculoskeletal pain in adults Cochrane Databaseof Systematic Reviews 2012 Issue 9 [DOI 10100214651858CD007400pub2]
Derry 2013Derry S Sven-Rice A Cole P Tan T Moore RA Topicalcapsaicin (high concentration) for chronic neuropathic painin adults Cochrane Database of Systematic Reviews 2013Issue 2 [DOI 10100214651858CD007393pub3]
Derry 2014
Derry S Matthews PR Wiffen PJ Moore RASalicylate-containing rubefacients for acute and chronicmusculoskeletal pain in adults Cochrane Database ofSystematic Reviews 2014 Issue 11 [DOI 10100214651858CD007403pub3]
Evans 1995Evans JM McMahon AD McGilchrist MM White GMurray FE McDevitt DG et al Topical non-steroidal anti-inflammatory drugs and admission to hospital for uppergastrointestinal bleeding and perforation a record linkagecase-control study BMJ 1995311(6996)22ndash6
FitzGerald 2001
FitzGerald GA Patrono C The coxibs selective inhibitorsof cyclooxygenase-2 New England Journal of Medicine2001345(6)433ndash42
34Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Haroutiunian 2010
Haroutiunian S Drennan DA Lipman AG Topical NSAIDtherapy for musculoskeletal pain Pain Medicine 201011
(4)535ndash49 [DOI 101111j1526-4637201000809x]
Hawkey 1999
Hawkey CJ Cox-2 inhibitors Lancet 1999353(9149)307ndash14
Higgins 2011
Higgins JPT Altman DG Sterne JAC Chapter 8 Assessingrisk of bias in included studies In Higgins JPT GreenS (editors) Cochrane Handbook for Systematic Reviewsof Interventions Version 510 [updated March 2011]The Cochrane Collaboration 2011 Available fromwwwcochrane-handbookorg
Jadad 1996a
Jadad AR Carroll D Moore A McQuay H Developing adatabase of published reports of randomised clinical trials inpain research Pain 199666(2-3)239ndash46
Jadad 1996b
Jadad AR Moore RA Carroll D Jenkinson C ReynoldsDJ Gavaghan DJ et al Assessing the quality of reports ofrandomized clinical trials is blinding necessary ControlledClinical Trials 199617(1)1ndash12
LrsquoAbbeacute 1987
LrsquoAbbeacute KA Detsky AS OrsquoRourke K Meta-analysis inclinical research Annals of Internal Medicine 1987107(2)224ndash33
Makris 2010Makris UE Kohler MJ Fraenkel L Adverse effects oftopical nonsteroidal antiinflammatory drugs in older adultswith osteoarthritis a systematic literature review Journalof Rheumatology 201037(6)1236ndash43 [DOI 103899jrheum090935]
Mason 2004a
Mason L Moore RA Edwards JE Derry S McQuay HJTopical NSAIDs for acute pain a meta-analysis BMCFamily Practice 2004510
Mason 2004bMason L Moore RA Edwards JE Derry S McQuayHJ Topical NSAIDs for chronic musculoskeletal painsystematic review and meta-analysis BMC MusculoskeletalDisorders 2004528
McPherson 2013McPherson ML Cimino NM Topical NSAIDformulations Pain Medicine 201314 Suppl 1S35ndash9[DOI 101111pme12288]
MHRA 2009
Medicines and Healthcare Products Regulatory AgencyTopical ketoprofen reminder on risk of photo-sensitivityreactions Drug Safety Update 2009 June Vol 2 issue 116
Moore 1998aMoore RA Tramegraver MR Carroll D Wiffen PJ McQuayHJ Quantitative systematic review of topically applied non-
steroidal anti-inflammatory drugs BMJ 1998316(7128)333ndash8
Moore 1998bMoore RA Gavaghan D Tramegraver MR Collins SL McQuayHJ Size is everything - large amounts of information areneeded to overcome random effects in estimating directionand magnitude of treatment effects Pain 199878(3)209ndash16
Moore 2003
Moore RA Edwards J Barden J McQuay HJ BandolierrsquosLittle Book of Pain Oxford Oxford University Press 2003[ISBN 0ndash19ndash263247ndash7]
Moore 2006
Moore RA Barden J Derry S McQuay HJ Managingpotential publication bias In McQuay HJ Kalso EMoore RA editor(s) Systematic Reviews in Pain ResearchMethodology Refined Seattle IASP Press 200615ndash23[ISBN 978ndash0ndash931092ndash69ndash5]
Moore 2008aMoore RA Derry S McQuay HJ Topical agents in thetreatment of rheumatic pain Rheumatic Diseases Clinics ofNorth America 200834(2)415ndash32
Moore 2008bMoore RA Barden J Derry S McQuay HJ Managingpotential publication bias In McQuay HJ Kalso EMoore RA editor(s) Systematic Reviews in Pain ResearchMethodology Refined Seattle IASP Press 200815ndash24[ISBN 978ndash0ndash931092ndash69ndash5]
Moore 2013
Moore RA Straube S Aldington D Pain measures andcut-offs - rsquono worse than mild painrsquo as a simple universaloutcome Anaesthesia 201368(4)400ndash12 [DOI 101111anae12148]
Morris 1995
Morris JA Gardner MJ Calculating confidence intervalsfor relative risk odds ratios and standardised ratios andrates In Gardner MJ Altman DG editor(s) Statistics withConfidence - Confidence Intervals and Statistical GuidelinesLondon British Medical Journal 199550ndash63
NICE 2008National Institute for Health and Care ExcellenceOsteoarthritis The care and management of osteoarthritisin adults 2008 wwwniceorgukCG059fullguideline
Nuumlesch 2010Nuumlesch E Trelle S Reichenbach S Rutjes AW TschannenB Altman DG et al Small study effects in meta-analysesof osteoarthritis trials meta-epidemiological study BMJ2010341c3515 [DOI 101136bmjc3515]
PACT 2014
Prescribing and Primary Care team Health and SocialCare Information Centre Prescrption Cost AnalysisEngland 2013 Health and Social Care Information CentreThe NHS Information Centre 2014298ndash301 [ISBN978ndash1ndash78386ndash089ndash0]
35Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Quane 1998
Quane PA Graham GG Zeigler JB Pharmacology ofbenzydamine Inflammopharmacology 19986(2)95ndash107
RevMan 2014The Nordic Cochrane Centre The Cochrane CollaborationReview Manager (RevMan) 53 Copenhagen The NordicCochrane Centre The Cochrane Collaboration 2014
Roth 2012Roth SH Fuller P Pooled safety analysis of diclofenacsodium topical solution 15 (ww) in the treatment ofosteoarthritis in patients aged 75 years or older ClinicalInterventions in Aging 20127127ndash37 [DOI 102147CIAS30884]
Simon 2009
Simon LS Grierson LM Naseer Z Bookman AA ZevShainhouse J Efficacy and safety of topical diclofenaccontaining dimethyl sulfoxide (DMSO) compared withthose of topical placebo DMSO vehicle and oral diclofenacfor knee osteoarthritis Pain 2009143(3)238ndash45 [DOI101016jpain200903008]
Tramer 1997
Tramer MR Reynolds DJ Moore RA McQuay HJ Impact
of covert duplicate publication on meta-analysis a casestudy BMJ 1997315(7109)635ndash40
Tugwell 2004Tugwell PS Wells GA Shainhouse JZ Equivalence studyof a topical diclofenac solution (pennsaid) compared withoral diclofenac in symptomatic treatment of osteoarthritisof the knee a randomized controlled trial Journal ofRheumatology 200431(10)2002ndash12
Zimmerman 1995
Zimmerman J Siguencia J Tsvang E Upper gastrointestinalhemorrhage associated with cutaneous application ofdiclofenac gel American Journal of Gastroenterology 199590(11)2032ndash4 [PUBMED 7485017]
References to other published versions of this review
Massey 2010
Massey T Derry S Moore RA McQuay HJ TopicalNSAIDs for acute pain in adults Cochrane Databaseof Systematic Reviews 2010 Issue 6 [DOI 10100214651858CD007402pub2]
lowast Indicates the major publication for the study
36Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
C H A R A C T E R I S T I C S O F S T U D I E S
Characteristics of included studies [ordered by study ID]
Airaksinen 1993
Methods R DB PC parallel groupsGel applied to the painful area twice daily for 7 daysAssessment at baseline 3 7 days
Participants Minor soft tissue injuries (lt 7 days)N = 56M 45 F 11Age not reportedMean baseline pain at rest 25-26 mm
Interventions Ketoprofen gel 2 x 5 g (125 mg) daily n = 29Placebo gel n = 27Rescue medication paracetamol 500 mgNo other treatment allowed
Outcomes PGE 5-point scale but reported as ldquoimprovedrdquo or ldquosame or worserdquo (responder = ldquoim-provedrdquo)Improvement in pain with movement 100 mm VAS reported as group meanAdverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Unclear risk No mention of early withdrawals ormethod of imputation
Size High risk lt 50 participants per treatment group
37Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Aoki 1984
Methods R DB PC AC parallel groupsGel applied to affected area 3 or 4 times daily with no occlusion for 7 daysAssessment at baseline 3 7 days
Participants Acute orthopaedic trauma (contusion distortion fracture lt 7 days)N = 252 (203 analysed for efficacy)M 98 F 105Age range 8 to 86 years 13 younger than 20 yearsBaseline pain mild in 35Exclusions 23 protocol violations 26 reasons ldquonot relatedrdquo to drug Equally distributedbetween groups
Interventions Piroxicam gel 05 1 g 3 to 4 x daily n = 84Indomethacin gel 1 1 g 3 to 4 x daily n = 84Placebo gel n = 84No other medication or initiation of physical therapy allowed
Outcomes PGE 5-point scale (responder = ldquobetterrdquo and ldquomuch betterrdquo)Adverse eventsWithdrawals and exclusions
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Low risk ldquokey code sealed until end of studyrdquo
Blinding (performance bias and detectionbias)All outcomes
Low risk Gels in ldquoidentical tubesrdquo
Incomplete outcome data (attrition bias)All outcomes
Unclear risk gt 10 withdrawals ldquounrelated to treat-mentrdquo and for ldquoprotocol violationsrdquo Nofurther details but no significant differ-ences between groups
Size Unclear risk 50 to 200 participants per treatment group
38Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Auclair 1989
Methods R DB PC parallel groupsGel massaged into skin over affected heel 3 times daily after cleaning with soap and waterfor up to 21 daysAssessment at baseline 7 21 days
Participants Acute Achilles heel tendinitis (not associated with continuous pain at rest or gt 1 monthhistory)N = 243 (227 analysed for efficacy)MF not reportedMean age 29 yearsBaseline pain ~ 10 had lt 26 mm on palpation of tendon ~ 30 had mild or no painon dorsiflexion of footExclusions failure to meet inclusion criteria major protocol violations failure to takestudy medication for full duration
Interventions Niflumic acid gel 25 3 x 5 g daily n = 117Placebo gel n = 110No other analgesics and anti-inflammatories physiotherapy or supportive measures al-lowed
Outcomes PGE 5-point scale (responder = ldquogoodrdquo or ldquovery goodrdquo)Pain improved or disappeared on dorsiflexionAdverse eventsWithdrawals and exclusions
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Unclear risk 10 excluded for ldquofailing to meet entry cri-teria and protocol violationsrdquo No furtherdetails
Size Unclear risk 50 to 200 participants per treatment group
39Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Billigmann 1996
Methods R DB PC parallel groupsGel applied 3 times daily with rubbingAssessed at baseline 3 5 7 days
Participants Distortion of ankle jointN = 160M and FAge 18+ yearsBaseline pain not reported
Interventions Ibuprofen microgel 5 3 x 10 cm (= 200 mg) daily n = 80Placebo gel n = 80
Outcomes Pain with movement VAS (responder = decreased by 20)Complete remissionAdverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Unclear risk Insufficient data to assess
Size Unclear risk 50 to 200 participants per treatment group
Campbell 1994
Methods R DB PC parallel groupsCream applied 4 times daily for 7 days (up to 14 days optional)Self assessed using daily diary for 7 days and up to 14 days
Participants Acute ankle sprain (lt 24 hours no fracture)N = 100 (51 analysed)M 33 F 18Mean age 29 years
40Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Campbell 1994 (Continued)
Baseline pain at rest gt 35 mm on walking 80 mmExclusions did not return diaries protocol exclusions (25 ibuprofen 24 placebo)
Interventions Ibuprofen cream 5 (Proflex) 4 x 4rdquo (10 cm) daily n = 26Placebo cream n = 25Advised to use rest and regular icing for 48 hours then walking and exerciseRescue medication paracetamol
Outcomes Improvement in walking ability 4-point scale (responder = ldquoimprovementrdquo)Pain on walking 100 mm VAS (mean data)Withdrawals and exclusions
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Low risk Randomisation carried out by sponsorTubes dispensed by hospital pharmacy whoheld the codes
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical creamrdquo
Incomplete outcome data (attrition bias)All outcomes
High risk gt 40 lost to follow-up Approximatelyequal between groups
Size High risk lt 50 participants per treatment arm as anal-ysed
Chatterjee 1977
Methods R DB PC parallel groupsCream applied to site of injury 3 times daily for 6 daysAssessment at baseline 2 6 days
Participants Soft tissue injuries (recent)N = 51MF not reportedAge not reportedBaseline pain on passive movement moderate or severe in all but 3 participants
41Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Chatterjee 1977 (Continued)
Interventions Benzydamine HCl cream 3 3 x daily n = 25Placebo cream n = 25(5 active 6 placebo participants also received ultrasound)No other topical agent allowed
Outcomes Pain on passive movement 4-point scale (responder = ldquoabsentrdquo or ldquoslightrdquo)Tenderness with pressure 4-point scale (responder = ldquoabsentrdquo or ldquoslightrdquo)Adverse eventsWithdrawals and exclusions
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk ldquopredetermined randomised schedulerdquo
Allocation concealment (selection bias) Low risk Sealed copy of schedule held by investiga-tor and duplicate copy kept by clinical trialco-ordinator Looked at only in event ofadverse reaction (not necessary)
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoindistinguishable in appearance andconsistencyrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk 2 lost to follow-up no withdrawals dueto adverse events Responder analysis
Size Unclear risk lt 50 participants per treatment arm
Costantino 2011
Methods R DB PC multicentre parallel groupPlaster applied daily for 7 daysPI assessment daily overall treatment efficacy and tolerability assessed at 3 and 7 days
Participants Grade I or II ankle sprain (lt 48 hours) with lateral external ligament involvementPI on movement ge 50100 oedema ge 20 mm difference between anklesN = 430M 249 F 175 (for analysis)Mean age 35 yearsBaseline PI on movement 72100 (SD 12)
42Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Costantino 2011 (Continued)
Interventions DHEPhep n = 142DHEP n = 146Placebo n = 142Rescue medication paracetamol to maximum 3 g dailyNo other treatments allowed
Outcomes Mean change in PI on movement from baseline to 3 daysMean reduction in oedema at 3 daysTolerability at 3 daysRescue medicationAdverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk Computer-generated
Allocation concealment (selection bias) Unclear risk ldquosealed envelopesrdquo
Blinding (performance bias and detectionbias)All outcomes
Low risk Shape colour size and application methodidentical for all plasters
Incomplete outcome data (attrition bias)All outcomes
Unclear risk Very few withdrawals but used LOCF
Size Unclear risk 50 to 200 participants per treatment arm
Coudreuse 2010
Methods R DB PC multicentre parallel groupPlaster applied daily for 7 daysPI assessed twice daily over 3 days then day 7 Overall treatment efficacy and tolerabilityassessed on days 3 7
Participants Ankle sprain (lt 48 hours) with lateral external ligament involvementPI on movement ge 50100 oedema ge 20 mm difference between anklesN = 240 (233 for analysis)M 148 F 86 6 unknown
Interventions DHEPhep n = 120Placebo n = 120Rescue medication paracetamol to maximum 4 g daily
43Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Coudreuse 2010 (Continued)
No other treatments allowed
Outcomes Global efficacy at 7 days 4-point scale (responder = ldquoexcellentrdquo or ldquogoodrdquo)Mean change in PI on movement at 6 hours and 7 daysMean change in oedema at 3 and 7 daysTolerabilityAdverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk Computer-generated
Allocation concealment (selection bias) Unclear risk ldquosealed envelopesrdquo
Blinding (performance bias and detectionbias)All outcomes
Low risk Appearance and odour identical for all plas-ters
Incomplete outcome data (attrition bias)All outcomes
Low risk ITT analysis lt 5 excluded for missingdata equal between groups
Size Unclear risk 50 to 200 participants per treatment arm
Curioni 1985
Methods R DB PC AC parallel groupsGel rubbed into affected area until absorbed twice daily for 10 daysAssessed at baseline and daily to 10 days
Participants Acute soft tissue injuriesN = 60M 33 F 27Median age 33 yearsBaseline pain not given
Interventions Ibuproxam gel 10 n = 20Ketoprofen gel n = 20Etofenamate gel n = 20
Outcomes PGE 4-point scale (ldquogoodrdquo or ldquoexcellentrdquo)Resolution of symptomsAdverse events
44Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Curioni 1985 (Continued)
Withdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk Medication supplied in identical tubes
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
Diebshlag 1990
Methods R DB PC AC parallel groupsGel applied 3 times daily without occlusion for 14 daysAssessment at baseline 2 3 4 8 15 days
Participants Ankle sprain (lt 24 hours)N = 37M 24 F 13Mean age 28 yearsBaseline pain slight to moderate
Interventions Ketorolac gel 2 3 x 3 g daily n = 13Etofenamate gel 5 3 x 3 g daily n = 12Placebo gel n = 12Rescue medication paracetamolNo other analgesic or anti-inflammatory medication ice packs or physiotherapy allowed
Outcomes Reduction in PI 100 mm VAS and 4-point scale (responder = ldquoimprovedrdquo)Adverse eventsWithdrawals and exclusions
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
45Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Diebshlag 1990 (Continued)
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Low risk ldquoMedication assignment supplied in asealed enveloperdquo Opened only if seriousparticipant event necessitation treatmentdisclosure occurred (not necessary)
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical appearancerdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
Dreiser 1988
Methods R DB PC parallel groupsCream applied 3 times dailyAssessment at baseline 7 days
Participants Acute tendinitis (lt 1 month)N = 64M 35 F 25Mean age 36 yearsBaseline spontaneous pain ge 60 mm
Interventions Ibuprofen cream 5 3 x 4 cm daily n = 32 (3 x 10 cm for large joints)Placebo cream n = 32No other topical systemic or physical treatment allowed
Outcomes PGE scale not reported (responder = ldquoimprovementrdquo or ldquocomplete relief rdquo)Improvement in pain VAS (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35Oxford Validity Score 1016
Risk of bias
Bias Authorsrsquo judgement Support for judgement
46Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Dreiser 1988 (Continued)
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Low risk Missing participants added back usingBOCF
Size High risk lt 50 participants per treatment arm
Dreiser 1989
Methods R DB PC parallel groupsGel applied twice daily to affected area with light massage then covered with standardcompressAssessed at baseline 3 7 days
Participants Uncomplicated recent ankle sprainN = 60M 36 F 24Mean age 33 yearsMean baseline pain 54 mm
Interventions Ketoprofen gel 25 2 x 5 cm daily n = 30Placebo gel n = 30No concomitant therapy other than simple oral analgesia allowed
Outcomes PGE 3-point scale (responder = ldquobetterrdquo)Improvement in pain VAS (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk ldquodrawing lotsrdquo
Allocation concealment (selection bias) Unclear risk Not described
47Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Dreiser 1989 (Continued)
Blinding (performance bias and detectionbias)All outcomes
Low risk Treatments ldquoidentical in every way exceptthat placebo did not contain active princi-plerdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participant per treatment arm
Dreiser 1990
Methods R DB PC parallel groupsGel lightly massaged into skin over affected area 3 times daily then covered with standardcompressAssessed at baseline 3 7 days
Participants Uncomplicated ankle sprain (lt 4 days)N = 60 (59 analysed)M 29 F 29 (not stated for 1 participant)Mean age 33 yearsBaseline pain ge moderately severeExclusions 1 participant had only moderate pain at baseline
Interventions Niflumic acid gel 25 3 x 5 g daily n = 30Placebo gel n = 30Concomitant treatment with systemic NSAIDs local therapies or physiotherapy werenot allowed
Outcomes PGE 4-point scale (responder = ldquocuredrdquo or ldquoimprovedrdquo)Improvement in pain VAS (mean data)Adverse eventsWithdrawals and exclusions
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
48Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Dreiser 1990 (Continued)
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
Dreiser 1994
Methods R DB PC parallel groupsPatch applied twice dailyAssessed at baseline 3 7 days
Participants Traumatic ankle sprain (lt 2 days)N = 131M 84 F 47Mean age 34 yearsBaseline pain ge 50 mm
Interventions Flurbiprofen patch 2 x 40 mg daily n = 65Placebo patch n = 66Rescue medication paracetamol Ice or light restraint allowedExclusions 1 from flurbiprofen group for protocol violation
Outcomes PGE 4-point scale (responder = ldquogoodrdquo or ldquovery goodrdquo)Improvement in pain VAS (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk Placebo patch was ldquonon-medicated (butotherwise identical)rdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
49Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Fioravanti 1999
Methods R DB AC parallel groupsGel lightly massaged into skin 3 times daily and kept dry for 6 to 8 hoursAssessed at baseline 3 10 days
Participants Peri and extra-articular inflammatory diseasesN = 100M 32 F 68Mean age 49 yearsBaseline spontaneous pain ge 40 mm
Interventions DHEP lecithin gel 3 x 5 g (= 65 mg) daily n = 50DHEP gel 3 x 5 g (= 65 mg) daily n = 50
Outcomes PGE 4-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo)Pain on movement meanAdverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size Unclear risk 50 participants per treatment arm
Fujimaki 1985
Methods R DB PC AC parallel groupsGel applied to affected area 3 or 4 times daily with no occlusion for up to 14 daysAssessed at baseline 7 14 days
Participants Muscle pain or inflammation in neck shoulder back chest and upper and lower ex-tremities or a combinationN = 271 (247 analysed)M 97 F 149
50Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Fujimaki 1985 (Continued)
Age lt 20 to 89 yearsBaseline pain mostly mild to moderateExclusions 24 due to protocol violations loss to follow-up
Interventions Piroxicam gel 05 1 g 3 to 4 x daily n = 92Indomethacin gel 1 1 g 3 to 4 x daily n = 90Placebo gel n = 89No concomitant oral or topical analgesic or anti-inflammatory medication allowed Nophysical therapy initiated after start of study
Outcomes PGE 5-point scale (responder = ldquobetterrdquo or ldquomuch betterrdquo)Physician rated improvement 5-point scale (responder = ldquomarked improvementrdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total =45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Low risk Cartons numbered randomly and numbersheld in a key code until study completion
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical tubesrdquo packed in numbered car-ton Gel bases slightly different in appear-ance so dispensing physician did not haveaccess to them
Incomplete outcome data (attrition bias)All outcomes
Unclear risk ~ 8 excluded from analysis for unknownreasons or lost to or inadequate follow-upApproximately equal between groups
Size Unclear risk 50 to 200 participants per treatment arm
Gallacchi 1990
Methods R DB AC parallel groupsGel applied to affected area 4 times daily with light massage for 14 daysAssessment at baseline 7 14 days
Participants Painful inflammatory conditionsN = 50M 20 F 30Mean age 50 years
51Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Gallacchi 1990 (Continued)
Baseline pain ge moderate severity
Interventions Diclofenac gel 1 2 g 4 x daily n = 25 (Flector)Diclofenac sodium 1 2 g 4 x daily n = 25 (Voltaren Emugel)No other medication that could interfere with test drugs allowed
Outcomes PGE 5-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo)Improvement in pain on pressure 4-point scale (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
Gonzaacutelez de Vega 2013
Methods R DB (for gels) SB (for ointment) AC multicentre parallel groups2 g (~ 6 cm of gel) applied over injured area x 3 daily for 14 daysAssessments at 0 4 7 14 days of treatment and day 42 follow-up
Participants Grade I or II ankle sprain (lt 24 hours) with lateral external ligament involvement PIon weight bearing ge 30100N = 449M = 308 F = 112 (for analysis)Mean age 28 years
Interventions Traumeel gel 3 x 2 g daily n = 140Traumeel ointment n = 143 (not analysed in this review)Diclofenac gel 1 3 x 2 g daily n = 137Rescue medication paracetamol to maximum 2 g daily
52Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Gonzaacutelez de Vega 2013 (Continued)
Outcomes Pain-free on day 7Global efficacy 5-point scale (responder = ldquogoodrdquo and ldquovery goodrdquo) on day 14Normal function on day 14 yes or no (responder = ldquoyesrdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 (gels) W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk Computer-generated
Allocation concealment (selection bias) Low risk Central allocation kits assigned in orderreceived and used envelopes (no furtherdetails)
Blinding (performance bias and detectionbias)All outcomes
Low risk For gel comparison ldquoidentical containersrdquo
Incomplete outcome data (attrition bias)All outcomes
Unclear risk Used LOCF Most withdrawals due to earlyrecovery (within 14 days) approximatelyequal between groups
Size Unclear risk 50 to 200 participants per treatment arm
Governali 1995
Methods R DB AC parallel groupsGel or cream applied 3 times daily for up to 14 daysAssessed at baseline 7 14 days
Participants Soft tissue injuries + 2 fracturesN = 30M = 21 F = 9Median age 38 yearsMean baseline pain on movement moderate to severe (28 scale 0 to 4)
Interventions Ketoprofen gel 5 3 x 2 to 3 g daily n = 15Ketoprofen cream 1 3 x 2 to 3 g daily n = 15
Outcomes PGE 5-point scale (responder = ldquogoodrdquo and ldquoexcellentrdquo)Adverse eventsWithdrawals
53Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Governali 1995 (Continued)
Notes Oxford Quality Score R1 DB1 W1 Total =35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Treatments were given in identical tubesand measurements made by blinded ob-servers but one was a cream and the othera gel
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
Gualdi 1987
Methods R DB AC parallel groupsGel applied twice daily for 10 daysAssessed at baseline 4 7 10 days
Participants Soft tissue injuriesN = 60M = 37 F = 23Mean age 32 years (range 13 to 78)Mean baseline pain on movement moderate to severe (22 scale 0 to 3)
Interventions Flunoxaprofen gel 2 x 3 to 5 cm daily n = 30Ketoprofen gel 2 x 3 to 5 cm daily n = 30
Outcomes Improvement in pain on pressure (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
54Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Gualdi 1987 (Continued)
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Unclear risk Insufficient data
Size High risk lt 50 participants per treatment arm
Haig 1986
Methods R DB PC parallel groupsCream applied lightly to affected area 6 times daily for 6 daysAssessed at baseline 2 4 6 days
Participants Soft tissue injuries (lt 24 hours)N = 43MF not reportedAge not reportedBaseline pain not reported
Interventions Benzydamine cream 3 6 x daily n = 21Placebo cream n = 22
Outcomes Pain on movement 4-point scale (responder = ldquoimprovedrdquo)Adverse events
Notes Oxford Quality Score R1 DB2 W0 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not reported
Allocation concealment (selection bias) Unclear risk Not reported
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquomatching placebordquo
55Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Haig 1986 (Continued)
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants apparently included inanalysis
Size High risk lt 50 participants per treatment arm
Hoffmann 2012
Methods R DB PC multicentre parallel groupPlaster applied x 1 daily (minimum of 20 consecutive hours of application per day) for14 daysPI assessed daily overall efficacy assessed at 7 and 14 days
Participants Unilateral mild to moderate muscle contusion of upper or lower limb (lt 72 hours) PIge 50100 superficial haematoma le 10 x 14 cm at injured siteN = 354M 126 F 228Mean age 39 years
Interventions DHEPhep plaster x 1 daily n = 121DHEP plaster x 1 daily n = 115Placebo plaster n = 118Rescue medication paracetamol 500 mg (no limit reported)
Outcomes PGE 5-point scale (responder = excellent or good) at 14 daysMean reduction in PI at 3 and 8 daysRescue medicationAdverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Method not reported
Allocation concealment (selection bias) Unclear risk Method not reported ldquoMedication packedand labelled in order to render all partici-pants and personnel fully blinded to treat-ment administeredrdquo
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoIndistinguishable regarding appearanceshape colour size and odourrdquo
56Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Hoffmann 2012 (Continued)
Incomplete outcome data (attrition bias)All outcomes
Unclear risk No responder analysis reported and impu-tation method not mentioned
Size Unclear risk 50 to 200 participants per treatment arm
Hofman 2000
Methods R DB AC parallel groupsGel applied to affected region 4 times daily with gentle massageAssessed at baseline 8 days in clinic and daily diary
Participants Soft tissue articular pain (le 15 days)N = 142M 19 F 123Mean age 57 yearsMean baseline PI moderate to severe
Interventions Diclofenac sodium gel 1 4 x 2 cm daily n = 69Lysine clonixinate gel 5 4 x 2 cm daily n = 73(2 cm = 225 mg)No other analgesic local treatment (including immobilisation bandaging) or acupunc-tureRescue mediation allowed after 2 applications if needed
Outcomes PGE 3-point scale (ldquogoodrdquo)PI participant diary (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk Diclofenac gel repackaged to maintain DBwith lysine clonixinate gel Minor differ-ences between gels only apparent when di-rectly compared
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
57Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Hofman 2000 (Continued)
Size Unclear risk 50 to 200 participants per treatment arm
Hosie 1993
Methods R DB AC parallel groupsFoam (approximately the size of a golf ball) applied and 1 tablet taken 3 times daily for7 days and up to 14 daysAssessed at baseline 7 14 (if necessary) days
Participants Acute lower back injury (lt 1 month)N = 287 (261 analysed for efficacy)M 151 F 136Mean age 37 years (range 18 to 63)Most participants had moderate to severe pain on movement one had noneExclusions 25 lost to follow up one assessed at 14 days but not 7 days
Interventions Felbinac foam 3 3 x 2 g daily + placebo tablets 3 x 1 daily n = 140 (127 analysed forefficacy)Ibuprofen tablets 3 x 400 mg daily + placebo foam 3 x 2 g daily n = 147 (134 analysedfor efficacy but one had no pain at baseline)No other oral injectable or topical analgesic or anti-inflammatory medication Ongoingphysiotherapy to continue without change
Outcomes Pain on movement 5-point scale (responder = ldquononerdquo or ldquomildrdquo)Spontaneous pain 5-point scale (responder = ldquononerdquo or ldquomildrdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquodouble dummyrdquo
Incomplete outcome data (attrition bias)All outcomes
Unclear risk Imputation method not mentioned
Size Unclear risk 50 to 200 participants per treatment arm
58Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Jenoure 1997
Methods R DB PC parallel groupsPlaster applied to skin over affected area twice daily and kept in place with an elasticbandageAssessed at baseline 7 14 days and after further 14 days without treatment
Participants Humero-radial epicondyl pain (tendinopathic) - nearly all tennis elbowN = 85M 54 F 31Mean age 45 yearsBaseline pain ldquomildrdquo in ~ 10 of placebo group and 29 of active group
Interventions DHEP plaster (Tissugel) x 2 daily n = 44Placebo plaster x 2 daily n = 41
Outcomes Pain on pressure 5-point scale (responder = ldquononerdquo or ldquomildrdquo)Spontaneous pain 5-point scale (responder = ldquono painrdquo)Adverse events
Notes Oxford Quality Score R1 DB2 W0 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not reported
Allocation concealment (selection bias) Unclear risk Not reported
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical characteristicsrdquo
Incomplete outcome data (attrition bias)All outcomes
Unclear risk No useable data
Size High risk lt 50 participants per treatment arm
Joussellin 2003
Methods R DB PC parallel groupsPlaster applied to skin over affected area once dailyAssessed at baseline 1 2 3 7 days
Participants Ankle sprain (lt 48 hours)N = 134M 72 F 62Age range 18 to 65 years
59Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Joussellin 2003 (Continued)
Baseline spontaneous pain ge 50 mm
Interventions DHEP plaster (Flector Tissugel 1) x 1 daily n = 68Placebo plaster x 1 daily n = 66Rescue medication paracetamol
Outcomes PGE 4-point scale (responder = ldquoexcellentrdquo)Pain on movement VAS (mean)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoIdentical in size appearance and usedsame formula as active patch without ac-tive ingredientrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in responder anal-ysis
Size Unclear risk 50 to 200 participants per treatment arm
Julien 1989
Methods R DB PC parallel groupsGel applied to affected area twice daily with light massageAssessed at baseline 3 7 days in clinic and daily diary
Participants TendinitisN = 60M 29 F 31Mean age 41 yearsBaseline pain gt 50 mm
Interventions Ketoprofen gel 25 2 x 5 cm (= 50 mg) daily n = 30Placebo gel 2 x 5 cm daily n = 30No concomitant therapy other than simple analgesia
60Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Julien 1989 (Continued)
Outcomes PGE 4-point scale (responder = ldquoimprovedrdquo or ldquorecoveredrdquo)Pain on movement 4-point scale (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB1 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk Random numbers table
Allocation concealment (selection bias) Unclear risk Randomisation code supplied by Menarinilaboratories remote from allocation
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in responder anal-ysis
Size High risk lt 50 participants per treatment arm
Klainguti 2010
Methods R DB PC multicentre parallel groupPlaster applied x 1 daily (minimum 20 consecutive hours of application daily) for 10daysPI assessed daily Overall treatment efficacy and tolerability assessed at days 3 and 7
Participants Unilateral mild to moderate muscle contusion or strain of upper or lower limb (lt 72hours) superficial haematoma le 140 cm2 PI ge 40100N = 185M 90 F 95Mean age 39 years
Interventions DHEPhep plaster x 1 daily n = 65DHEP plaster x 1 daily n = 61Placebo plaster x 1 daily n = 59
Outcomes Overall treatment efficacy at 3 days 5-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo)Resolution of haematoma at 10 days yes or noRescue medicationAdverse eventsWithdrawals
61Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Klainguti 2010 (Continued)
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk Computer-generated
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoIdentical in size appearance and odourrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk Withdrawals lt 5 All participants in-cluded in analysis
Size Unclear risk 50 to 200 participants per treatment arm
Kockelbergh 1985
Methods R DB PC parallel groupsGel applied twice dailyAssessed at baseline 3 7 days
Participants Acute soft tissue trauma (lt 24 hours)N = 74M 60 F 14Mean age 27 yearsBaseline pain gt 65 mm
Interventions Ketoprofen gel 25 2 x 5 cm (= 15 mg) daily n = 38Placebo gel 2 x 5 cm daily n = 36No concomitant treatmentRescue medication glafenine
Outcomes PGE 3-point scale (responder = ldquogoodrdquo)Spontaneous pain 100 mm VAS (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
62Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Kockelbergh 1985 (Continued)
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in responder anal-ysis
Size High risk lt 50 participants per treatment arm
Kuehl 2011
Methods R DB PC multicentre parallel groupsPlaster applied x 2 daily for 14 days or until resolutionPI assessed daily Overall treatment efficacy assessed at 14 days
Participants Mild or moderate sprain strain or contusion (lt 7 days) PI ge 510N = 418M 206 F 212Mean age 39 years
Interventions DHEP patch x 2 daily n = 207Placebo patch x 2 daily n = 211Rescue medication no analgesic allowed (or icewrapping) use = discontinuation
Outcomes Resolution of injury (VAS le 210)Overall tolerability 5-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk Computer-generated
Allocation concealment (selection bias) Unclear risk Not described
63Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Kuehl 2011 (Continued)
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical in appearance with same contentexcept diclofenacrdquo
Incomplete outcome data (attrition bias)All outcomes
High risk Used LOCF high withdrawal rate (43)Did not report all efficacy outcomes mea-sured
Size Low risk gt 200 participants per treatment arm
Li 2013
Methods R DB PC multicentre parallel groupsPlaster applied x 2 daily for 7 daysPI assessed daily overall treatment efficacy and tolerability assessed at 7 days
Participants Mild or moderate ankle or knee sprain muscle strain or contusion (lt 72 hours) PI ge
50100N = 384M 144 F 240Mean age 42 years
Interventions DHEP plaster x 2 daily n = 192Placebo plaster x 2 daily n = 192Rescue medication paracetamol to maximum 2 g daily
Outcomes ge 50 reduction in PI at 7 daysOverall treatment efficacy 5-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo)Overall tolerability 5-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo)Rescue medicationAdverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk Computer-generated
Allocation concealment (selection bias) Unclear risk Method not fully described ldquosealed en-velopesrdquo ldquoconcealed from investigators andparticipantsrdquo
64Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Li 2013 (Continued)
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical in texture size color and odorrdquo
Incomplete outcome data (attrition bias)All outcomes
Unclear risk Withdrawals 2 Methods states LOCFbut appears to describe BOCF for with-drawals
Size Unclear risk 50 to 200 participants per treatment arm
Linde 1985
Methods R DB PC parallel groupsCream applied x 3 daily for 5 days with elastic support for the first 3 daysAssessed at baseline 4 8 days
Participants Sprained ankle (lt 24 hours)N = 100M 58 F 42Mean age 28 yearsBaseline pain all participants had ldquowalking painrdquo
Interventions Benzydamine 3 cream x 3 daily n = 50Placebo gel x 3 daily n = 50
Outcomes Pain on movement responder = ldquofree of walking painrdquoAdverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35Paper describes a benzydamine cream and a placebo gel
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described Paper describes a benzy-damine cream and a placebo gel
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in responder anal-ysis
65Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Linde 1985 (Continued)
Size Unclear risk 50 participants per treatment arm
Machen 2002
Methods R DB PC parallel groupsGel gently (ldquominimal rubrdquo not vigorously) massaged into skin over affected site untilabsorbed 3 times daily until symptoms disappeared or for maximum of 7 daysAssessment at baseline and once daily using diary cards to 7 days
Participants Soft tissue injury (lt 2 weeks and untreated)N = 85 (81 analysed)M 42 F 39Mean age 41 yearsBaseline pain gt 50 mm4 placebo participants lost to follow-up
Interventions Ibuprofen gel 5 x 3 daily n = 40Placebo gel x 3 daily n = 41Initiation of other medication or physiotherapy not allowed during study
Outcomes PGE 5-point scale (responder = ldquomarked improvementrdquo or ldquocomplete clearancerdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk Gels had similar physical characteristicsand were supplied in identical tubes
Incomplete outcome data (attrition bias)All outcomes
Low risk lt 5 missing participants Remaining par-ticipants included in responder analysis
Size High risk lt 50 participants per treatment arm
66Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Mahler 2003
Methods R DB AC parallel groupsGel applied with gentle massage to affected area 3 times daily without occlusion for 10daysAssessed at baseline 3 10 days in clinic and daily patient diary
Participants First-degree ankle or knee sprains first-degree muscle strains and mild-to-moderatecontusionsn = 100M 69 F 31Mean age 32 yearsMean baseline pain with activity ge 65 mm
Interventions DHEP lethicin gel 3 x 5 g (= 65 mg) daily n = 52DHEP gel 3 x 5 g (= 65 mg) daily n = 48All participants treated with ice at site of inflammation for first 48 hours but no immo-bilisation allowedRescue medication paracetamol 500 mg if strictly necessary
Outcomes PGE 4-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo)Pain on movement 100-mm VAS (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 5
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk ldquoComputer-generated randomization listrdquo
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk Pharmaceutically inert colouring agentsadded to reference formulation so that gelswere indistinguishable
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in responder anal-ysis
Size High risk lt 50 participants in each treatment arm
67Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Maziegraveres 2005a
Methods R DB PC parallel groupsNew patch applied directly to skin over painful area each morningAssessed at baseline 3 7 14 days
Participants Symptomatic tendonitis in upper or lower limbs not requiring surgery (le 15 days)N = 172M 72 F 100Mean age 46 yearsBaseline pain with activity ge 40 mm
Interventions Ketoprofen patch 100 mg x 1 daily n = 87Placebo patch x 1 daily n = 85No analgesic or steroid by any route or other topical medication or physical therapyallowedRescue medication permitted but not within 12 hours of assessment
Outcomes PGE 4-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk ldquocomputer generated global randomizationcoderdquo
Allocation concealment (selection bias) Low risk ldquoThe randomization list and code en-velopes were prepared by the company ap-pointed for clinical supplies packaging Therandom code was disclosed only after studycompletion and database closurerdquo
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquothe same indistinguishable patch with noingredientrdquo
Incomplete outcome data (attrition bias)All outcomes
High risk LOCF withdrawals 12 All participantsincluded in analysis
Size Unclear risk 50 to 200 participants per treatment arm
68Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Maziegraveres 2005b
Methods R DB PC parallel groupsNew patch applied directly to skin over painful area each morningAssessed at baseline 3 7 14 days
Participants Painful benign ankle sprain (le 48 hours)N = 163M 83 F 80Mean age 37 yearsBaseline spontaneous pain ge 50 mm
Interventions Ketoprofen patch 100 mg x 1 daily n = 81Placebo patch x 1 daily n = 82No analgesic or steroid by any route or other topical medication or physical therapyallowedRescue medication permitted but not within 12 hours of assessment
Outcomes PGE 4-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk ldquocomputer-generated global randomiza-tion coderdquo
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk The same transdermal patch with no activeingredient
Incomplete outcome data (attrition bias)All outcomes
Unclear risk LOCF withdrawals 5 to 6 All partic-ipants included in analysis
Size Unclear risk 50 to 200 participants per treatment arm
McLatchie 1989
Methods R DB PC parallel groupsGel applied to injured site 3 times daily for 7 daysAssessment at baseline 4 7 days at clinic daily patient diary
Participants Acute soft tissue injury (lt 48 hours)N = 231
69Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
McLatchie 1989 (Continued)
M 143 F 88Mean age 33 yearsBaseline pain moderate to severe
Interventions Felbinac gel 3 3 x 3 cm daily n = 118Placebo gel 3 x 3 cm daily n = 113Rescue medication paracetamol
Outcomes Patient diary mean changeAdverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquotubes identical in all aspectsrdquo
Incomplete outcome data (attrition bias)All outcomes
Unclear risk No usable efficacy data
Size Unclear risk 50 to 200 participants per treatment arm
Morris 1991
Methods R DB PC parallel groupsGel applied to site of injury 3 times daily for 7 daysAssessed at baseline 7 days at clinic and daily patient diary
Participants Acute soft tissue injury (lt 3 days)N = 100 (84 analysed for efficacy)M 70 F 14Mean age 25 yearsBaseline pain moderate to severeExclusions 1 participant in placebo group lost to follow-up 15 protocol violations
Interventions Felbinac gel 3 3 x 1 cm daily n = 41Placebo gel n = 43Ice joint immobilisation bandaging and compression allowed
70Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Morris 1991 (Continued)
No concomitant oral NSAID occlusive dressing physiotherapy or liniments allowedRescue medication paracetamol
Outcomes PGE 5-point scale (responder = ldquogoodrdquo and ldquovery goodrdquo)Change in PI patient diary 10 cm VAS (mean data)Adverse eventsWithdrawals and exclusions
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Low risk ldquoRandomisation was undertaken at theproduction facility and a sealed copy of thelist supplied to the investigator for refer-ence only in defined circumstancesrdquo
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical tubes and outer boxesrdquo ldquoplacebowas a similarly constituted gelrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All relevant participants included in analy-sis
Size High risk lt 50 participants per treatment arm
NCT01255423
Methods R DB PC parallel group studyGel applied 4 x daily
Participants Acute lateral ankle sprain Grade I to II le 12 hoursN = 206Mean (plusmn SD) age 31 years (plusmn 13)M 87 F 119Baseline PI not reported
Interventions Diclofenac sodium gel 1 x 4 daily n = 104Placebo gel n = 102
Outcomes Mean PI on movementAdverse events
71Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT01255423 (Continued)
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Method of blinding not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants accounted for
Size Unclear risk 50 to 200 participants per treatment arm
NCT01272934
Methods R DB PC parallel group studyGel applied 4 x daily
Participants Acute lateral ankle sprain Grade I to II le 12 hoursN = 205M 101 F 104Mean (plusmn SD) age 32 years (plusmn 11)
Baseline PI not reported
Interventions Diclofenac sodium gel 1 x 4 daily n = 102Placebo gel x 4 daily n = 103
Outcomes Mean PI on movementAdverse events
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
72Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT01272934 (Continued)
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Method of blinding not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants accounted for
Size Unclear risk 50 to 200 participants per treatment arm
NCT01272947
Methods R DB PC parallel groupsGel applied 4 x daily
Participants Acute blunt soft tissue injuriescontusions of the limbs lt 3 hoursN = 204M 101 F 103Mean (plusmn SD) age 30 years (plusmn 11)
Baseline PI not reported
Interventions Diclofenac sodium gel 1 x 4 daily n = 104Placebo gel x 4 daily n = 100
Outcomes Mean PI on movementAdverse events
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Method of blinding not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants accounted for
73Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT01272947 (Continued)
Size Unclear risk 50 to 200 participants per treatment arm
Noret 1987
Methods R DB PC parallel groupsGel applied twice daily for 7 daysAssessment at baseline 3 8 days
Participants Minor sports injuries (lt 24 hours)N = 98 (93 analysed)M 71 F 27Mean age 29 yearsBaseline pain gt 60 mm
Interventions Ketoprofen gel 25 2 x 5 cm daily (= 15 mg) n = 48Placebo gel n = 45No other treatment given
Outcomes PGE 4-point scale (responder = ldquogoodrdquo and ldquoexcellentrdquo)Spontaneous pain 100 mm VAS (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Low risk ldquoallocated according to a randomization listand a corresponding code in a sealed enve-loperdquo
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
74Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Parrini 1992
Methods R DB PC parallel groupsFoam (the size of a walnut or a 1-second spray) applied with massage 3 x daily for 7days
Participants Articular trauma strains distortionsN = 169M 94 F 75Mean age 37 yearsMean baseline pain on movement 31 (scale 1 to 4)
Interventions Ketoprofen foam 15 3 x 2 g (= 600 mg) daily n = 83Placebo foam 3 x 2 g n = 86
Outcomes Pain on movement 4-point scale (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB1 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk ldquopatients were randomised according to themethod of random numbersrdquo [translated]
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Unclear risk No usable efficacy data
Size Unclear risk 50 to 200 participants per treatment arm
Picchio 1981
Methods R DB AC parallel groupsCream applied with slight massage until completely absorbed 3 times daily for up to 16daysAssessed at baseline 4 8 12 16 days
Participants Acute sports injuriesN = 40M 24 F 16Mean age 22 years (range 12 to 46)Most participants had mild to moderate baseline pain (12 and 9 with slight pain on
75Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Picchio 1981 (Continued)
movement)
Interventions Ibuprofen gel 10 3 x daily n = 20Ketoprofen gel 1 3 x daily n = 20
Outcomes Pain on movement (responder = ldquononerdquo)Adverse events
Notes Oxford Quality Score R1 DB2 W0 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk ldquotubes were identical in appearancerdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
Predel 2004
Methods R DB PC parallel groupsNew patch applied to injured area twice daily for 7 days Contact of patch with humidityor water to be avoidedAssessment at baseline 3 7 days
Participants Traumatic blunt soft tissue injuries (lt 3 hours no treatment)N = 120M 73 F 47Mean age 32 yearsBaseline pain gt 60 mm
Interventions Diclofenac sodium patch 2 x daily (140 mg per patch) n = 60Placebo patch 2 x daily n = 60NSAIDs analgesics psychotropic agents other topical preparations and bandages notallowed
Outcomes PGE 4-point scale (responder = rdquogoodldquo and rdquoexcellentldquo)Pain on movement 10 cm VAS (mean data)
76Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Predel 2004 (Continued)
Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk rdquocomputer generated block randomisationlist
Allocation concealment (selection bias) Low risk An independent statistician produced ran-domisation list and an independent con-tract research organisation packaged med-ication according to list Nobody else hadaccess to the randomisation list until thedatabase was closed
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoThe placebo patch was visually indistin-guishable from the active patchrdquo To avoidunblinding due to different smell anystudy nurse involved with medication wasnot involved in outcome assessment
Incomplete outcome data (attrition bias)All outcomes
Low risk Withdrawals lt 1 All participants in-cluded in analysis
Size Unclear risk 50 to 200 participants per treatment arm
Predel 2012
Methods R DB PC parallel groupsTreatment for 7 daysAssessed at baseline 1 3 5 and 8 days
Participants Grade I or II ankle sprain with lateral external ligament involvement lt 12 hoursPI on movement ge 50100N = 242M 152 F 90Mean age 32 years
Mean baseline pain on movement = 75100
Interventions Diclofenac gel (Voltaren Emulgel 232) 2 x 5 cm + 1 x 5 cm placebo gel daily n = 80Diclofenac gel (Voltaren Emulgel 232) 3 x 5 cm daily n = 80Placebo gel 3 x 5 cm daily n = 82Rescue medication paracetamol (to maximum 2 g daily)
77Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Predel 2012 (Continued)
No ice or bandages after randomisation
Outcomes ge 50 reduction in PI on movement at 5 daysPGE 5-point scale (responder = ldquogoodrdquo or rsquovery goodldquo)Patient satisfaction 5-point scale (responder = rdquogoodldquo or rsquovery goodrdquo or ldquoexcellentrdquo)Rescue medicationAdverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical in composition appearance tex-ture and smellrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk LOCF but withdrawals lt 3 in all treat-ment arms
Size Unclear risk 50 to 200 participants per treatment arm
Predel 2013a
Methods R DB PC parallel groupsTreatment for 14 daysAssessed at baseline 3 7 10 and 14 days (plusmn 1 day)
Participants Uncomplicated one-sided ankle sprain with swelling ge 12 mm (2 to 18 hours)PI on movement tenderness joint mobility (scale 0 to 3) summed as ge 5 and le 7N = 232M 126 F 106Mean age 29 years
Interventions Diclofenac 4 spray gel 3 x 4 to 5 sprays daily (96 to 120 mg diclofenac sodium) n =118Placebo spray gel 3 x 4 to 5 sprays daily n = 114Rescue medication paracetamol 500 mg (maximum 10 tablets per week)No ice or bandaging allowed
78Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Predel 2013a (Continued)
Outcomes None or slight PI on movement at 3 and 7 daysPGE 4-point scale (responder = ldquovery goodrdquo at 14 days)gt 50 reduction in swelling at 10 daysAdverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk ldquoplacebo (vehicle only no active ingredi-ent)rdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk LOCF for full or early cure Withdrawalslt 2
Size Unclear risk 50 to 200 participants per treatment arm
Predel 2013b
Methods R DB PC parallel groups2 g gel applied with fingertips over affected area and massaged in for 1 minute Treatmentfor 5 daysAssessed at baseline 2 3 5 days
Participants Uncomplicated neck pain originating from cervical joints and accompanying soft tissues(ge 12 hours but lt 3 months) PI ge 50100N = 72M 33 F 39Mean age 34 years
Interventions Diclofenac gel (Voltaren Emulgel) 4 x 2 g daily n = 36Placebo gel 4 x 2 g daily n = 36Rescue medication paracetamol up to 2 g dailyNo concomitant therapies allowed
Outcomes ge 50 decrease in PI on movement after 48 hoursPGE 5-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo at 5 daysAdverse eventsWithdrawals
79Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Predel 2013b (Continued)
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk Sponsor ldquoproduced randomisation listrdquo
Allocation concealment (selection bias) Low risk Automated remote system
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical in packaging labelling scheduleof administration appearance and odourrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk No missing data
Size High risk lt 50 participants per treatment arm
Ramesh 1983
Methods R DB PC parallel groupsCream applied to painful area and rubbed into skin over a large area for up to 10 daysAssessment at baseline 3 7 10 days
Participants Strains sprains contusions compressionsN = 80M 42 F 38Age 11 to 81 yearsBaseline pain 5 ibuprofen 2 placebo participants had no or slight pain
Interventions Ibuprofen cream 5 3 to 4 x 5 to 10 cm daily n = 40Placebo cream 3 to 4 x 5 to 10 cm daily n = 40Adjuvant therapy was not administered
Outcomes Pain on movement 4-point scale (responder = ldquononerdquo or ldquoslightrdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
80Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Ramesh 1983 (Continued)
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Low risk Randomisation key in sealed envelopeavailable for emergencies but opened onlyafter completion
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical appearance and odourrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
Rowbotham 2003
Methods R DB PC parallel groupsNew patches applied to the affected painful area for 12 consecutive hours twice dailyfor up to 14 daysAssessed at baseline 14 days in clinic and daily patient diary
Participants Minor sports injuries (sprains sprains contusions lt 72 hours)N = 372M 253 F 119Mean age 33 yearsBaseline pain at rest ge 510
Interventions DHEP patch (Flector Tissuegel) 2 x daily (equivalent to diclofenac sodium 140 mg perpatch) n = 191Placebo patch 2 x daily n = 181
Outcomes PGE 5-point scale (responder = ldquogoodrdquo and ldquoexcellentrdquo)Pain resolved lt moderate for 2 daysSpontaneous pain 10 cm VAS (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
81Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Rowbotham 2003 (Continued)
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk ldquoSystegraveme identiquerdquo without diclofenac
Incomplete outcome data (attrition bias)All outcomes
Unclear risk No information about imputation All par-ticipants included in analysis
Size Unclear risk 50 to 200 participants per treatment arm
Russell 1991
Methods R DB PC parallel groupsAffected area washed with soap and water and dried then gel applied and carefullyrubbed into skin 4 times daily for at least 7 daysAssessed at baseline 4 8 15 (if necessary) days at clinic and daily patient diary
Participants Acute soft tissue injuries (recent not recurrent)N = 214 (200 analysed)M 95 F 105Mean age 40 yearsBaseline pain gt 65 mm
Interventions Piroxicam gel 05 4 x 5 mg daily n = 100Placebo gel n = 100No other NSAIDs or analgesic drugs including liniments containing salicylates allowedAncillary therapy at the discretion of the investigator
Outcomes PGE 4-point scale (responder = ldquogoodrdquo and ldquoexcellentrdquo)Spontaneous pain mean reductionAdverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk ldquocomputer generated randomization coderdquo
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical base formulationrdquo
82Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Russell 1991 (Continued)
Incomplete outcome data (attrition bias)All outcomes
High risk High withdrawal rate (8 with piroxicam50 with placebo) No information aboutimputation
Size Unclear risk 50 to 100 participants per treatment group
Saillant 1998
Methods R DB PC parallel groupsTreatment for 7 daysAssessed at baseline 1 2 3 and 7 days
Participants Ankle sprain (lt 48 hours)N = 140M 72 F 62Age 18 to 65 yearsBaseline spontaneous pain ge 50 mm
Interventions (1) DHEP plaster (Flector Tissugel 1) 1 x daily n = 70(2) Placebo plaster 1 x daily n = 70Rescue medication paracetamol (to maximum 3 g daily)Ice allowed
Outcomes ge 30 decrease in PI at 7 daysPGE 4-point scale (responder = ldquoexcellentrdquo)No or low pain on passive stretchSingle foot leaning OK without painWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk Identical in size appearance and used sameformula as active patch without active in-gredient
Incomplete outcome data (attrition bias)All outcomes
Unclear risk Used LOCF for primary outcome With-drawals lt 7 and equal between groupsAll participants included in analysis
83Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Saillant 1998 (Continued)
Size Unclear risk 50 to 200 participants per treatment arm
Sanguinetti 1989
Methods R DB PC parallel groupsGel applied 3 times daily for 7 consecutive daysAssessment at baseline 7 days
Participants Soft tissue trauma (lt 48 hours)N = 82M = 47 F = 35Mean age 34 yearsBaseline pain moderate to severe
Interventions Felbinac gel 3 3 x daily n = 42Placebo gel 3 x daily n = 40No other NSAID steroid other topical application allowedRescue medication paracetamol felbinac is an active metabolite of the NSAID fenbufen
Outcomes PGE scale not reported (responder = ldquogoodrdquo and ldquovery goodrdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoindistinguishable in appearance colour orodourrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
84Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Sinniger 1981
Methods R DB PC parallel groupsCream applied 2 to 3 times daily with gentle massage or if massage not possible (toopainful) with protective dressingAssessment at baseline 5 10 days
Participants Minor soft tissue injuriesN = 20M 11 F 9Mean age 40 yearsBaseline pain not reported
Interventions Fentiazac cream 5 2 to 3 x daily n = 10Placebo cream 2 to 3 x daily n = 10All participants told to restNo other local and systemic treatments allowedRescue medication analgesic if actually needed
Outcomes Pain relief scale not reported (responder = total pain relief ) improvement in pain on movement pain scale not reported (mean data)Adverse events
Notes Oxford Quality Score R1 DB1 W0 Total = 25
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
85Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Spacca 2005
Methods R DB PC parallel groupsGel applied 3 times daily with gentle massage until complete absorption for up to 10daysAssessment at baseline 10 days in clinic and daily patient diary
Participants Shoulder periarthritis or lateral epicondylitis (lt 5 days)N = 155M 74 F 81Mean age 51 yearsBaseline pain with activity gt 70 mm
Interventions DHEP lecithin gel (Effigel) 3 x 5 g daily n = 79Placebo gel 3 x 5 g daily n = 76Rescue medication (paracetamol) allowed if pain unbearableNo other analgesic or anti-inflammatory drug allowed
Outcomes Improvement in pain 100 mm VAS (mean data)Adverse events
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Unclear risk No usable efficacy data
Size Unclear risk 50 to 200 participants per treatment arm
Sugioka 1984
Methods R DB AC parallel groupsGel applied to affected area 3 to 4 times daily without occlusion for 14 daysAssessed at baseline 7 14 days
Participants Non-traumatic diseases of muscle or tendonN = 366 (340 analysed for efficacy)M 115 F 202 (completers)
86Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Sugioka 1984 (Continued)
Age range 12 to 84 years (most 30 to 70)Baseline pain on movement ldquononerdquo or ldquomildrdquo in about 13 of participantsExclusions for protocol violations 8 piroxicam 18 indomethacin
Interventions Piroxicam gel 05 3 to 4 x 1 g daily n = 183Indomethacin gel 1 3 to 4 x 1 g daily n = 183No concomitant anti-inflammatory or analgesic drug including steroids or initiationof physical therapy allowed
Outcomes PGE 5-point scale (responder = ldquobetterrdquo or ldquomuch betterrdquo)Pain on movement 4-point scale (responder = ldquoreducedrdquo or ldquodisappearedrdquo)
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Low risk Key code sealed and retained until end ofstudy
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoboth packages were of the same appear-ance and indistinguishablerdquo and investiga-tors did not see contents
Incomplete outcome data (attrition bias)All outcomes
Low risk All relevant participants included in analy-sis
Size Unclear risk 50 to 200 participants per treatment arm
Thorling 1990
Methods R DB PC parallel groupsParticipants given specific instructions on how to apply gel (not reported) to affectedarea 2 to 6 times daily as requiredAssessment at baseline 3 7 days in clinic
Participants Soft tissue injuries (lt 48 hours)N = 120M 85 F 35Mean age 27 yearsBaseline pain moderate to severe
87Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Thorling 1990 (Continued)
Interventions Naproxen gel 10 2 to 6 x daily n = 60Placebo gel 2 to 6 x daily n = 60Rescue medication paracetamol 500 mg
Outcomes PGE 5-point scale (responder = ldquogoodrdquo and ldquovery goodrdquo)Pain on passive movement 4-point scale (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquosupplied in unmarked tubesrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size Unclear risk 50 to 200 participants per treatment arm
Tonutti 1994
Methods R DB AC parallel groupsGel applied 3 times daily for 2 to 3 weeksAssessed at baseline and intervals of 7 days
Participants Muscle or joint traumaN = 30M 20 F 10Mean age 34 years1 participant had injury of mild severity Mean baseline pain on active movement 28(scale 0 to 4)
Interventions Ketoprofen gel 5 3 x 2 to 3 g daily n = 15Etofenamate gel 5 3 x 2 to 3 g n = 15No concomitant treatment with NSAID aspirin steroid or physical therapy
88Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Tonutti 1994 (Continued)
Outcomes PGE 4-point scale (responder = ldquogoodrdquo and ldquoexcellentrdquo)Pain on movement 5-point scale (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not reported
Allocation concealment (selection bias) Unclear risk Not reported
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquothe two drugs were packed in indistin-guishable tubesrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
Vecchiet 1991
Methods R DB PC parallel groupsGel applied to the skin on and around painful area and gently rubbed in until absorbedtwice daily for up to 10 daysAssessed at baseline 5 10 days
Participants Soft tissue trauma (minor sports injuries)N = 60M 60Mean age 25 yearsMean baseline pain on active movement moderate
Interventions Meclofenamic acid gel 5 2 x 10 cm daily (= 4 g) n = 30Placebo gel 2 x 10 cm daily n = 30Both groups treated with ice rest and bandage for first 48 hours before starting testtreatmentRescue medication paracetamol
Outcomes PGE 4-point scale (responder = ldquogoodrdquo and ldquoexcellentrdquo)Pain on movement 4-point scale (mean data)Withdrawals
89Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Vecchiet 1991 (Continued)
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
Whitefield 2002
Methods R DB (double dummy) AC parallel groupsGel applied to affected site with gentle massage and 1 tablet taken 3 times daily for atleast 7 daysAssessed at baseline 7 14 (if necessary) days in clinic and daily patient diary
Participants Soft tissue injuries (lt 24 hours)N = 100M 95 F 5Mean age 26 years (range 18 to 50)Mean baseline pain on movement 22 cm
Interventions Ibuprofen gel 5 + placebo tablet 3 x daily n = 50Ibuprofen 400 mg tablet + placebo gel 3 x daily n = 50No other medication or physical therapy was prescribed and no other analgesics wereallowed
Outcomes PGE 3-point scale (responder = ldquoexcellentrdquo)Change in condition of injury site 5-point scale (responder = ldquocompletely betterrdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
90Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Whitefield 2002 (Continued)
Random sequence generation (selectionbias)
Unclear risk Not reported
Allocation concealment (selection bias) Unclear risk Not reported
Blinding (performance bias and detectionbias)All outcomes
Low risk Placebo tablets Identical in appearance toactive tablets Active and placebo gels hadsimilar physical characteristics and weresupplied in identical tubes
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size Unclear risk 50 participants per treatment arm
kermark 1990
Methods R DB (double dummy) PC AC parallel groupsSpray applied to affected area and capsules taken 3 times daily for 2 weeksAssessment at baseline 3 or 4 7 and 14 days
Participants Superficial overuse sports injuries (symptom onset 74 weeks)N = 70M 44 F 18 (completers)Mean age 30 yearsBaseline pain on palpation mostly slight to moderate
Interventions Elmetacin spray (indomethacin 1) 3 to 5 x 05 to 15 ml daily + placebo capsules n= 23Indomethacin capsules 3 x 25 mg daily + placebo spray n = 23Placebo spray and capsules n = 24Rescue medication paracetamol
Outcomes No pain on palpation (= responder)Participant improvement 100 mm VAS (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk ldquorandom number coderdquo
91Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
kermark 1990 (Continued)
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical in appearancerdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk Responder analysis lt 5 withdrawals
Size High risk lt 50 participants per treatment group
AC active control BOCF baseline observation carried forward DB double-blind DHEP diclofenac hydroxyethylpyrrolidine ordiclofenac epolamine F female HCl hydrochloride hep heparin LOCF last observation carried forward M male N numberof participants in study n number of participants in treatment arm PC placebo control PGE Participant Global Evaluation PIpain intensity R randomised VAS visual analogue scale W withdrawals
Characteristics of excluded studies [ordered by study ID]
Study Reason for exclusion
Ambrus 1987 No usable dichotomous data
Anon 1993 Not double-blind
Ascherl 1982 No usable dichotomous data
Bagliani 1976 Not an RCT
Baracchi 1982 No usable data
Burnham 1998 lt10 participantstreatment arm in first period of crossover study
Boumlhmer 1995 Active control invalid
Cesarone 2008 Not an RCT
Coulibaly 2009 Not double-blind
Diebschlag 1985 No usable dichotomous data
Diebschlag 1986 Inappropriate randomisation
Diebschlag 1992 No usable dichotomous data
92Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Fantato 1971 No usable dichotomous data
Galer 2000 No usable data
Hallmeier 1986 Not double-blind
Hallmeier 1988 Not double-blind
Kaneko 1999 Inappropriate randomisation - quasi-randomised
Kockelbergh 1985b Treatment not applied daily
Kuwabara 2013 Used NSAID-lidocaine combination (conference abstract)
Lee 1991 Not an RCT
Link 1996 No usable dichotomous data
May 2007 No usable dichotomous data
Oakland 1993 Inappropriate comparator
Odaglia 1987 Not an RCT
Picardi 1993 Not an RCT
Taboada 1992 Dose and duration of treatment unclear
Vanderstraeten 1990 Not double-blind
Vinciguerra 2008 Not an RCT
Von Klug 1977 Chronic and acute outcomes combined
RCT randomised controlled trial
Characteristics of studies awaiting assessment [ordered by study ID]
NCT00351104
Methods R DB PC parallel groups
Participants Ankle sprain or strain Grade I or II Age ge 18 yearsN = 220
93Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT00351104 (Continued)
Interventions Ketoprofen patch 20 x 1 dailyPlacebo patch x 1 dailyTreatment for 21 days
Outcomes Assessments days 3 7 21PI (0-10) at rest and during activitiesPGEFunctionRescue medication
Notes Sponsor Endo PharmaceuticalsEstimated completion February 2007 No results
NCT00352625
Methods R DB PC parallel groups
Participants Shoulder elbow or knee tendonitis or bursitis Age ge 18 yearsN = 330
Interventions Ketoprofen patch 20 x 1 dailyPlacebo patch x 1 dailyTreatment for 21 days
Outcomes Assessments days 3 7 21PI (0-10) at rest and during activitiesPGEFunctionRescue medication
Notes Sponsor Endo PharmaceuticalsEstimated completion February 2007 No results
NCT00426985
Methods R DB PC parallel groups
Participants Shoulder elbow or knee tendonitis or bursitis Age ge 18 years
Interventions Ketoprofen patch 20 x 1 dailyPlacebo patchTreatment for 21 days
Outcomes PI during activityPI at restPGERescue medication
94Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT00426985 (Continued)
Safety
Notes Study terminated May 2008 but ldquosufficient number of subjects accrued to conduct analysisrdquo No results
NCT00640705
Methods R DB PC parallel groups
Participants Ankle sprain or strain Grade I or II le 48 hours PI 510 to 910 Age 18 to 75 yearsN = 170
Interventions Diclofenac sodium patch (15 mg) x 1 dailyPlaceboTreatment for 7 days
Outcomes EfficacySafety
Notes Study terminated July 2008 (Sponsor decision - no further details) No results posted
NCT00640939
Methods R DB PC parallel groups
Participants Shoulder elbow or knee tendonitis or bursitis Age 18 to 75 yearsBaseline pain mild to moderate but states 510 to 910N = 308
Interventions Diclofenac sodium patch 1 (15 mg) x 1 dailyPlacebo patch x 1 dailyDuration of treatment not reported probably 14 days
Outcomes EfficacySafety
Notes Sponsor Cerimon PharmaceuticalsEstimated completion April 2008 No results
NCT00680472
Methods R DB PC parallel groups
Participants Acute unilateral shoulder pain requiring treatment for ge 2 weeks Age ge 18 yearsN = 368
95Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT00680472 (Continued)
Interventions Ketoprofen patch (HKT-500) x 1 dailyPlacebo x 1 dailyTreatment for 14 days
Outcomes PainSafety
Notes Sponsor Hisamitsu Pharmaceutical Co IncEstimated completion October 2008 No results
NCT00680784
Methods R DB PC parallel groups
Participants Acute benign ankle sprain Grade I-II le 48 hours Age ge 18 yearsN = 260
Interventions Ketoprofen patch (HKT-500) x 1 dailyPlacebo patch x 1 dailyTreatment for 14 days
Outcomes PainSafety
Notes Sponsor Hisamitsu Pharmaceutical Co IncEstimated completion November 2008 No results
NCT00765700
Methods R DB PC parallel groups
Participants Acute sprain or strain of upper and lower extremities mild to moderate injuries le 72 hours Age 18 to 70 yearsN = 364
Interventions Ketoprofen cream 10 1 g x 3 dailyPlacebo cream x 3 dailyTreatment for 7 days
Outcomes PainSafety
Notes Sponsor Imprimis Pharmaceuticals IncEstimated completion September 2009 No results
96Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT00869063
Methods R DB PC parallel groups
Participants Wrist sprain strain or contusion (mild to moderate) ldquorecentrdquo Age 17 to 75 yearsN = 214
Interventions Diclofenac sodium patch 1 x 1 dailyPlacebo patch x 1 dailyTreatment for 7 days
Outcomes Change in PI during activity at 3 and 7 daysSafety
Notes Sponsor Cerimon PharmaceuticalsEstimated completion September 2009 No results
NCT00869180
Methods R DB PC parallel groups
Participants Unilateral ankle sprain (mild or moderate) ldquorecentrdquo Age 17 to 75 yearsN = 219
Interventions Diclofenac sodium patch x 1 dailyPlacebo patch x 1 dailyTreatment for 7 days
Outcomes Change in PI during activity at 3 and 7 days
Notes Sponsor Cerimon PharmaceuticalsEstimated completion August 2009 No results
NCT00931866
Methods R DB PC parallel groups
Participants Unilateral soft tissue injury between mid-bicep to wrist or mid-thigh to ankle (mild to moderate) ldquorecentrdquo Age 18to 75 yearsN = 407
Interventions Diclofenac sodium patch x 1 dailyPlacebo patch x 1 dailyTreatment for 14 days
Outcomes Change in PI during activity at 7 and 14 daysPIPGERescue medicationSafety
97Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT00931866 (Continued)
Notes Sponsor Cerimon PharmaceuticalsEstimated completion December 2009 No results
NCT01874626
Methods R DB PC parallel groups
Participants Acute ankle sprain Grade I or II le 24 hours Age ge 16 years Baseline PI ge 510N = 305
Interventions Ibuprofen cream 200 mg in 27 g cream x 4 dailyPlacebo cream x 4 dailyTreatment for 7 days
Outcomes PI on movement daily to 7 daysPGE 7 daysSystemic and local adverse events
Notes Estimated completion January 2014 No results
NCT01957215
Methods R DB PC parallel groups
Participants Acute lateral ankle sprain Grade I-II le 24 hours PI ge 510 Age 18 to 65 yearsN = 270
Interventions Indomethacin patch x 2 dailyPlacebo patch x 2 dailyDuration of treatment not reported probably 7 days
Outcomes PI on movement dailyPR dailyPGERescue medication
Notes Estimated completion September 2014 No results
NCT02324270
Methods R DB PC and AC parallel groups
Participants Uncomplicated acute ankle sprain Grade I or II lt 48 hours PI gt 50100 Age 18 to 65 yearsN = 658
98Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT02324270 (Continued)
Interventions Diclofenac epolamine (Flector) 13 patchGeneric diclofenac epolamine 13 patchPlacebo patchDuration of treatment not reported
Outcomes Bioequivalence studyChange from baseline in PI (VAS) at 3 daysApplication site reactions
Notes Estimated completion December 2014 No results
Sarzi-Puttini 2014
Methods R DB AC parallel group non-inferiority studyDuration 7 days
Participants Acute musculoskeletal injury (mainly muscular joint tendon)N = 697M 271 F 426Mean age 52 years
Interventions Ketoprofen (SKP-021) patch ketoprofen 30 mgDiclofenac (Voltadola) patch diclofenac sodium 140 mgPatches applied twice daily for 7 days
Outcomes ge 50 reduction in PI from baseline to end of study (100 mm VAS)Patient overall ratingClinical symptoms (4-point scale)Time to responseAdverse events skin reactionsSerious adverse events
Notes ldquoThe analysis of the data of this trial showed that the two formulations were equally effective and well tolerated inthe treatment of acute musculoskeletal injuriesrdquo
AC active-controlled DB double-blind F female M male N number of participants in study PC placebo-controlled PGE PatientGlobal Evaluation of treatment PI pain intensity R randomised VAS visual analogue scale
99Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Characteristics of ongoing studies [ordered by study ID]
NCT01945034
Trial name or title Placebo-controlled double-blind evaluation of the efficacy and safety of ibuprofen 5 topical gel for thetreatment of ankle sprain
Methods R DB PC parallel groups
Participants Ankle sprain Grade I or II age ge 12 yearsEstimated enrolment 280
Interventions Ibuprofen gel 5 x 2 dailyIbuprofen gel 5 x 3 dailyPlacebo gelTreatment for 7 days with additional 3 days as needed
Outcomes PI on weight bearing and rest at 7 daysPGE at 10 daysRescue medicationSafety
Starting date November 2013
Contact information Pfizer
Notes Estimated completion February 2015Includes participants aged ge 12 years
NCT02100670
Trial name or title A clinical study to assess the efficacy and onset of pain relief of topical MFC51123 diclofenac-menthol gelversus controls in ankle sprain
Methods R DB PC and AC parallel groups
Participants Acute lateral ankle sprain Grade I or II le 24 hours PI ge 510 Age 16 to 65 yearsEstimated enrolment 400
Interventions Diclofenac sodium 1 + methanol 3 gelDiclofenac sodium 1 + methanol 009 gelMethanol 3 gelPlacebo + 009 methanol gelTreatment for 10 days with 4 g gel x 4 daily
Outcomes PRPID on movement (0 to 10 days)PGETime to complete recoveryAdverse events
100Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT02100670 (Continued)
Starting date November 2013
Contact information GSKClinicalSupportHDgskcom
Notes Estimated completion November 2014
NCT02290821
Trial name or title A randomised double-blind placebo-controlled parallel group study to evaluate the efficacy and safety ofdiclofenac sodium topical gel (DSG) 1 applied four times daily in subjects with acute blunt soft tissueinjuriescontusions of the limbs
Methods R DB PC parallel groups
Participants Acute blunt soft tissue injuries or contusions of the limbs le 6 hours (ldquofreshrdquo) Age ge 16 yearsEstimated enrolment 200
Interventions Diclofenac sodium gel 1 x 4 dailyPlacebo gelDuration of treatment not reported
Outcomes PainSafety
Starting date December 2014
Contact information Novartis
Notes Estimated completion August 2015
AC active control DB double-blind PC placebo-controlled PGE Patient Global Evaluation of treatment PI pain intensity PIDpain intensity difference PR pain relief R randomised
101Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
D A T A A N D A N A L Y S E S
Comparison 1 Individual NSAID versus placebo
Outcome or subgroup titleNo ofstudies
No ofparticipants Statistical method Effect size
1 Clinical success 29 Risk Ratio (M-H Fixed 95 CI) Subtotals only11 Diclofenac 10 2050 Risk Ratio (M-H Fixed 95 CI) 160 [149 172]12 Ibuprofen 5 436 Risk Ratio (M-H Fixed 95 CI) 164 [133 201]13 Ketoprofen 7 683 Risk Ratio (M-H Fixed 95 CI) 156 [137 177]14 Piroxicam 3 504 Risk Ratio (M-H Fixed 95 CI) 148 [127 173]15 Indomethacin 3 341 Risk Ratio (M-H Fixed 95 CI) 126 [103 155]16 Benzydamine 3 193 Risk Ratio (M-H Fixed 95 CI) 115 [096 138]
2 Local adverse events 33 Risk Ratio (M-H Fixed 95 CI) Subtotals only21 Diclofenac 15 3271 Risk Ratio (M-H Fixed 95 CI) 078 [056 110]22 Ibuprofen 3 321 Risk Ratio (M-H Fixed 95 CI) 230 [098 543]23 Ketoprofen 8 852 Risk Ratio (M-H Fixed 95 CI) 119 [083 170]24 Piroxicam 3 522 Risk Ratio (M-H Fixed 95 CI) 042 [017 108]25 Felbinac 3 397 Risk Ratio (M-H Fixed 95 CI) 191 [049 750]26 Indomethacin 3 354 Risk Ratio (M-H Fixed 95 CI) 265 [091 773]
Comparison 2 Diclofenac versus placebo (effect of formulation)
Outcome or subgroup titleNo ofstudies
No ofparticipants Statistical method Effect size
1 Clinical success 10 2050 Risk Ratio (M-H Fixed 95 CI) 160 [149 172]11 Plaster - Flector 4 1030 Risk Ratio (M-H Fixed 95 CI) 153 [136 171]12 Plaster - other 3 474 Risk Ratio (M-H Fixed 95 CI) 155 [137 175]13 Gel - Emulgel 2 314 Risk Ratio (M-H Fixed 95 CI) 384 [268 550]14 Gel - other 1 232 Risk Ratio (M-H Fixed 95 CI) 115 [105 127]
Comparison 3 Ibuprofen versus placebo (effect of formulation)
Outcome or subgroup titleNo ofstudies
No ofparticipants Statistical method Effect size
1 Clinical success 5 436 Risk Ratio (M-H Fixed 95 CI) 164 [133 201]11 Cream 3 195 Risk Ratio (M-H Fixed 95 CI) 128 [103 159]12 Gel 2 241 Risk Ratio (M-H Fixed 95 CI) 266 [169 421]
102Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Comparison 4 Ketoprofen versus placebo
Outcome or subgroup titleNo ofstudies
No ofparticipants Statistical method Effect size
1 Clinical success 7 683 Risk Ratio (M-H Fixed 95 CI) 156 [137 177]11 Plaster 2 335 Risk Ratio (M-H Fixed 95 CI) 121 [104 140]12 Gel 5 348 Risk Ratio (M-H Fixed 95 CI) 219 [174 275]
Comparison 5 All topical NSAIDs versus placebo
Outcome or subgroup titleNo ofstudies
No ofparticipants Statistical method Effect size
1 Local adverse events 42 6740 Risk Ratio (M-H Fixed 95 CI) 098 [080 121]2 Systemic adverse events 36 5576 Risk Ratio (M-H Fixed 95 CI) 096 [073 126]3 Adverse event withdrawals 42 6405 Risk Ratio (M-H Fixed 95 CI) 101 [064 159]
Comparison 6 Topical NSAID versus active comparator
Outcome or subgroup titleNo ofstudies
No ofparticipants Statistical method Effect size
1 Clinical success - topicalpiroxicam vs topicalindomethacin
3 641 Risk Ratio (M-H Fixed 95 CI) 124 [107 144]
2 Local adverse events -topical piroxicam vs topicalindomethacin
3 671 Risk Ratio (M-H Fixed 95 CI) 021 [009 047]
103Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Analysis 11 Comparison 1 Individual NSAID versus placebo Outcome 1 Clinical success
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 1 Individual NSAID versus placebo
Outcome 1 Clinical success
Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
1 Diclofenac
Coudreuse 2010 99117 82116 175 120 [ 104 138 ]
Joussellin 2003 3668 2466 52 146 [ 099 215 ]
Klainguti 2010 5662 4659 100 116 [ 099 136 ]
Li 2013 173192 114192 242 152 [ 134 172 ]
Predel 2004 5560 560 11 1100 [ 474 2555 ]
Predel 2012 (1) 5980 941 25 336 [ 186 607 ]
Predel 2012 (2) 5780 841 22 365 [ 193 690 ]
Predel 2013a 111118 93114 201 115 [ 105 127 ]
Predel 2013b 3636 736 16 487 [ 257 923 ]
Rowbotham 2003 75191 48181 104 148 [ 110 200 ]
Saillant 1998 4370 2570 53 172 [ 119 248 ]
Subtotal (95 CI) 1074 976 1000 160 [ 149 172 ]Total events 800 (NSAID) 461 (Placebo)
Heterogeneity Chi2 = 12176 df = 10 (Plt000001) I2 =92
Test for overall effect Z = 1278 (P lt 000001)
2 Ibuprofen
Billigmann 1996 2580 1080 136 250 [ 129 486 ]
Campbell 1994 2126 1925 264 106 [ 080 142 ]
Dreiser 1988 2632 1232 164 217 [ 134 349 ]
Machen 2002 2540 941 121 285 [ 152 532 ]
Ramesh 1983 2340 2340 314 100 [ 069 146 ]
Subtotal (95 CI) 218 218 1000 164 [ 133 201 ]Total events 120 (NSAID) 73 (Placebo)
Heterogeneity Chi2 = 2102 df = 4 (P = 000031) I2 =81
Test for overall effect Z = 465 (P lt 000001)
3 Ketoprofen
Airaksinen 1993 2429 1427 91 160 [ 107 238 ]
Dreiser 1989 1830 530 32 360 [ 154 844 ]
02 05 1 2 5
Favours placebo Favours NSAID
(Continued )
104Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
( Continued)Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Julien 1989 1830 630 38 300 [ 138 650 ]
Kockelbergh 1985 3038 2236 142 129 [ 095 176 ]
Mazi res 2005a 7281 6082 376 121 [ 104 141 ]
Mazi res 2005b 5087 4185 262 119 [ 090 158 ]
Noret 1987 3951 947 59 399 [ 218 733 ]
Subtotal (95 CI) 346 337 1000 156 [ 137 177 ]Total events 251 (NSAID) 157 (Placebo)
Heterogeneity Chi2 = 3104 df = 6 (P = 000002) I2 =81
Test for overall effect Z = 683 (P lt 000001)
4 Piroxicam
Aoki 1984 5672 3367 286 158 [ 120 207 ]
Fujimaki 1985 4483 4082 337 109 [ 080 147 ]
Russell 1991 79100 45100 377 176 [ 138 223 ]
Subtotal (95 CI) 255 249 1000 148 [ 127 173 ]Total events 179 (NSAID) 118 (Placebo)
Heterogeneity Chi2 = 624 df = 2 (P = 004) I2 =68
Test for overall effect Z = 499 (P lt 000001)
5 Indomethacin
Aoki 1984 4164 3367 413 130 [ 096 176 ]
Fujimaki 1985 4482 4082 513 110 [ 082 148 ]
kermark 1990 1222 624 74 218 [ 099 481 ]
Subtotal (95 CI) 168 173 1000 126 [ 103 155 ]Total events 97 (NSAID) 79 (Placebo)
Heterogeneity Chi2 = 269 df = 2 (P = 026) I2 =26
Test for overall effect Z = 221 (P = 0027)
6 Benzydamine
Chatterjee 1977 2125 1225 185 175 [ 112 272 ]
Haig 1986 1821 1322 196 145 [ 098 214 ]
Linde 1985 3550 4050 618 088 [ 070 110 ]
Subtotal (95 CI) 96 97 1000 115 [ 096 138 ]Total events 74 (NSAID) 65 (Placebo)
Heterogeneity Chi2 = 1033 df = 2 (P = 001) I2 =81
Test for overall effect Z = 152 (P = 013)
02 05 1 2 5
Favours placebo Favours NSAID
(1) Three times daily application
(2) Twice daily application
105Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Analysis 12 Comparison 1 Individual NSAID versus placebo Outcome 2 Local adverse events
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 1 Individual NSAID versus placebo
Outcome 2 Local adverse events
Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
1 Diclofenac
Coudreuse 2010 1117 7116 102 014 [ 002 113 ]
Jenoure 1997 144 141 15 093 [ 006 1442 ]
Joussellin 2003 168 366 44 032 [ 003 303 ]
Klainguti 2010 0126 159 30 016 [ 001 381 ]
Kuehl 2011 16207 12211 173 136 [ 066 280 ]
Li 2013 4192 3192 44 133 [ 030 588 ]
NCT01255423 1104 3102 44 033 [ 003 309 ]
NCT01272934 1102 0103 07 303 [ 012 7350 ]
NCT01272947 0104 0100 Not estimable
Predel 2012 (1) 1160 182 19 051 [ 003 809 ]
Predel 2013a 1120 4116 59 024 [ 003 213 ]
Predel 2013b 036 036 Not estimable
Rowbotham 2003 27191 31181 463 083 [ 051 133 ]
Saillant 1998 070 070 Not estimable
Spacca 2005 079 076 Not estimable
Subtotal (95 CI) 1720 1551 1000 078 [ 056 110 ]Total events 54 (NSAID) 66 (Placebo)
Heterogeneity Chi2 = 944 df = 10 (P = 049) I2 =00
Test for overall effect Z = 142 (P = 016)
2 Ibuprofen
Billigmann 1996 1180 480 573 275 [ 091 827 ]
Machen 2002 440 241 283 205 [ 040 1057 ]
Ramesh 1983 140 140 143 100 [ 006 1544 ]
Subtotal (95 CI) 160 161 1000 230 [ 098 543 ]Total events 16 (NSAID) 7 (Placebo)
Heterogeneity Chi2 = 048 df = 2 (P = 079) I2 =00
Test for overall effect Z = 190 (P = 0057)
3 Ketoprofen
002 01 1 10 50
Favours NSAID Favours placebo
(Continued )
106Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
( Continued)
Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Airaksinen 1993 529 427 99 116 [ 035 389 ]
Dreiser 1989 030 230 60 020 [ 001 400 ]
Julien 1989 130 030 12 300 [ 013 7083 ]
Kockelbergh 1985 138 136 24 095 [ 006 1459 ]
Mazi res 2005a 1281 682 142 202 [ 080 513 ]
Mazi res 2005b 2987 2785 651 105 [ 068 161 ]
Noret 1987 151 047 12 277 [ 012 6636 ]
Parrini 1992 083 086 Not estimable
Subtotal (95 CI) 429 423 1000 119 [ 083 170 ]Total events 49 (NSAID) 40 (Placebo)
Heterogeneity Chi2 = 357 df = 6 (P = 073) I2 =00
Test for overall effect Z = 095 (P = 034)
4 Piroxicam
Aoki 1984 179 274 147 047 [ 004 506 ]
Fujimaki 1985 183 282 143 049 [ 005 534 ]
Russell 1991 4102 10102 710 040 [ 013 123 ]
Subtotal (95 CI) 264 258 1000 042 [ 017 108 ]Total events 6 (NSAID) 14 (Placebo)
Heterogeneity Chi2 = 003 df = 2 (P = 098) I2 =00
Test for overall effect Z = 180 (P = 0072)
5 Felbinac
McLatchie 1989 3118 2113 666 144 [ 024 844 ]
Morris 1991 041 043 Not estimable
Sanguinetti 1989 342 140 334 286 [ 031 2634 ]
Subtotal (95 CI) 201 196 1000 191 [ 049 750 ]Total events 6 (NSAID) 3 (Placebo)
Heterogeneity Chi2 = 023 df = 1 (P = 063) I2 =00
Test for overall effect Z = 093 (P = 035)
6 Indomethacin
Aoki 1984 270 274 440 106 [ 015 730 ]
Fujimaki 1985 582 282 452 250 [ 050 1252 ]
kermark 1990 422 024 108 978 [ 056 17191 ]
Subtotal (95 CI) 174 180 1000 265 [ 091 773 ]Total events 11 (NSAID) 4 (Placebo)
Heterogeneity Chi2 = 167 df = 2 (P = 043) I2 =00
Test for overall effect Z = 179 (P = 0073)
002 01 1 10 50
Favours NSAID Favours placebo
(1) DHEP + heparin and DHEP combined
107Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Analysis 21 Comparison 2 Diclofenac versus placebo (effect of formulation) Outcome 1 Clinical success
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 2 Diclofenac versus placebo (effect of formulation)
Outcome 1 Clinical success
Study or subgroup Experimental Control Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
1 Plaster - Flector
Joussellin 2003 3668 2466 52 146 [ 099 215 ]
Li 2013 173192 114192 242 152 [ 134 172 ]
Rowbotham 2003 75191 48181 104 148 [ 110 200 ]
Saillant 1998 4370 2570 53 172 [ 119 248 ]
Subtotal (95 CI) 521 509 451 153 [ 136 171 ]Total events 327 (Experimental) 211 (Control)
Heterogeneity Chi2 = 051 df = 3 (P = 092) I2 =00
Test for overall effect Z = 720 (P lt 000001)
2 Plaster - other
Coudreuse 2010 99117 82116 175 120 [ 104 138 ]
Klainguti 2010 5662 4659 100 116 [ 099 136 ]
Predel 2004 5560 560 11 1100 [ 474 2555 ]
Subtotal (95 CI) 239 235 285 155 [ 137 175 ]Total events 210 (Experimental) 133 (Control)
Heterogeneity Chi2 = 4658 df = 2 (Plt000001) I2 =96
Test for overall effect Z = 699 (P lt 000001)
3 Gel - Emulgel
Predel 2012 (1) 116160 1782 48 350 [ 227 540 ]
Predel 2013b 3636 736 16 487 [ 257 923 ]
Subtotal (95 CI) 196 118 64 384 [ 268 550 ]Total events 152 (Experimental) 24 (Control)
Heterogeneity Chi2 = 070 df = 1 (P = 040) I2 =00
Test for overall effect Z = 734 (P lt 000001)
4 Gel - other
Predel 2013a 111118 93114 201 115 [ 105 127 ]
Subtotal (95 CI) 118 114 201 115 [ 105 127 ]Total events 111 (Experimental) 93 (Control)
02 05 1 2 5
Favours placebo Favours diclofenac
(Continued )
108Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
( Continued)Study or subgroup Experimental Control Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Heterogeneity not applicable
Test for overall effect Z = 284 (P = 00045)
Total (95 CI) 1074 976 1000 160 [ 149 172 ]Total events 800 (Experimental) 461 (Control)
Heterogeneity Chi2 = 12176 df = 9 (Plt000001) I2 =93
Test for overall effect Z = 1278 (P lt 000001)
Test for subgroup differences Chi2 = 5003 df = 3 (P = 000) I2 =94
02 05 1 2 5
Favours placebo Favours diclofenac
(1) twice daily and three times daily applications combined
Analysis 31 Comparison 3 Ibuprofen versus placebo (effect of formulation) Outcome 1 Clinical success
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 3 Ibuprofen versus placebo (effect of formulation)
Outcome 1 Clinical success
Study or subgroup Ibuprofen Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
1 Cream
Campbell 1994 2126 1925 264 106 [ 080 142 ]
Dreiser 1988 2632 1232 164 217 [ 134 349 ]
Ramesh 1983 2340 2340 314 100 [ 069 146 ]
Subtotal (95 CI) 98 97 742 128 [ 103 159 ]Total events 70 (Ibuprofen) 54 (Placebo)
Heterogeneity Chi2 = 791 df = 2 (P = 002) I2 =75
Test for overall effect Z = 222 (P = 0026)
2 Gel
Billigmann 1996 2580 1080 136 250 [ 129 486 ]
Machen 2002 2540 941 121 285 [ 152 532 ]
Subtotal (95 CI) 120 121 258 266 [ 169 421 ]Total events 50 (Ibuprofen) 19 (Placebo)
001 01 1 10 100
Favours placebo Favours ibuprofen
(Continued )
109Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
( Continued)Study or subgroup Ibuprofen Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Heterogeneity Chi2 = 008 df = 1 (P = 078) I2 =00
Test for overall effect Z = 420 (P = 0000026)
Total (95 CI) 218 218 1000 164 [ 133 201 ]Total events 120 (Ibuprofen) 73 (Placebo)
Heterogeneity Chi2 = 2102 df = 4 (P = 000031) I2 =81
Test for overall effect Z = 465 (P lt 000001)
Test for subgroup differences Chi2 = 805 df = 1 (P = 000) I2 =88
001 01 1 10 100
Favours placebo Favours ibuprofen
Analysis 41 Comparison 4 Ketoprofen versus placebo Outcome 1 Clinical success
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 4 Ketoprofen versus placebo
Outcome 1 Clinical success
Study or subgroup Diclofenac Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
1 Plaster
Mazi res 2005a 7281 6082 376 121 [ 104 141 ]
Mazi res 2005b 5087 4185 262 119 [ 090 158 ]
Subtotal (95 CI) 168 167 638 121 [ 104 140 ]Total events 122 (Diclofenac) 101 (Placebo)
Heterogeneity Chi2 = 002 df = 1 (P = 090) I2 =00
Test for overall effect Z = 249 (P = 0013)
2 Gel
Airaksinen 1993 2429 1427 91 160 [ 107 238 ]
Dreiser 1989 1830 530 32 360 [ 154 844 ]
Julien 1989 1830 630 38 300 [ 138 650 ]
Kockelbergh 1985 3038 2236 142 129 [ 095 176 ]
Noret 1987 3951 947 59 399 [ 218 733 ]
001 01 1 10 100
Favours placebo Favours diclofenac
(Continued )
110Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
( Continued)Study or subgroup Diclofenac Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Subtotal (95 CI) 178 170 362 219 [ 174 275 ]Total events 129 (Diclofenac) 56 (Placebo)
Heterogeneity Chi2 = 1937 df = 4 (P = 000067) I2 =79
Test for overall effect Z = 674 (P lt 000001)
Total (95 CI) 346 337 1000 156 [ 137 177 ]Total events 251 (Diclofenac) 157 (Placebo)
Heterogeneity Chi2 = 3104 df = 6 (P = 000002) I2 =81
Test for overall effect Z = 683 (P lt 000001)
Test for subgroup differences Chi2 = 1859 df = 1 (P = 000) I2 =95
001 01 1 10 100
Favours placebo Favours diclofenac
Analysis 51 Comparison 5 All topical NSAIDs versus placebo Outcome 1 Local adverse events
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 5 All topical NSAIDs versus placebo
Outcome 1 Local adverse events
Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Airaksinen 1993 529 427 26 116 [ 035 389 ]
Aoki 1984 (1) 270 137 08 106 [ 010 1128 ]
Aoki 1984 (2) 179 137 09 047 [ 003 728 ]
Auclair 1989 5123 6116 39 079 [ 025 251 ]
Billigmann 1996 1180 480 26 275 [ 091 827 ]
Costantino 2011 (3) 3287 6142 51 025 [ 006 097 ]
Coudreuse 2010 1117 7116 45 014 [ 002 113 ]
Diebshlag 1990 (4) 113 06 04 150 [ 007 3229 ]
Diebshlag 1990 (5) 112 06 04 162 [ 008 3466 ]
Dreiser 1989 030 230 16 020 [ 001 400 ]
002 01 1 10 50
Favours NSAID Favours placebo
(Continued )
111Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
( Continued)Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Dreiser 1990 030 330 22 014 [ 001 265 ]
Dreiser 1994 265 066 03 508 [ 025 10373 ]
Fujimaki 1985 (6) 183 141 09 049 [ 003 770 ]
Fujimaki 1985 (7) 582 141 09 250 [ 030 2070 ]
Haig 1986 021 022 Not estimable
Jenoure 1997 144 141 07 093 [ 006 1442 ]
Joussellin 2003 168 366 19 032 [ 003 303 ]
Julien 1989 130 030 03 300 [ 013 7083 ]
Klainguti 2010 (8) 0126 159 13 016 [ 001 381 ]
Kockelbergh 1985 138 136 07 095 [ 006 1459 ]
Kuehl 2011 16207 12211 76 136 [ 066 280 ]
Li 2013 4192 3192 19 133 [ 030 588 ]
Linde 1985 050 050 Not estimable
Machen 2002 440 241 13 205 [ 040 1057 ]
Mazi res 2005a 1281 682 38 202 [ 080 513 ]
Mazi res 2005b 2987 2785 174 105 [ 068 161 ]
McLatchie 1989 3118 2113 13 144 [ 024 844 ]
Morris 1991 041 043 Not estimable
NCT01255423 1104 3102 19 033 [ 003 309 ]
NCT01272934 1102 0103 03 303 [ 012 7350 ]
NCT01272947 0104 0100 Not estimable
Noret 1987 151 047 03 277 [ 012 6636 ]
Parrini 1992 083 086 Not estimable
Predel 2012 (9) 1160 182 08 051 [ 003 809 ]
Predel 2013a 1120 4116 26 024 [ 003 213 ]
Predel 2013b 036 036 Not estimable
Ramesh 1983 140 140 06 100 [ 006 1544 ]
Rowbotham 2003 27191 31181 203 083 [ 051 133 ]
Russell 1991 4102 10102 64 040 [ 013 123 ]
Saillant 1998 070 070 Not estimable
Sanguinetti 1989 342 140 07 286 [ 031 2634 ]
002 01 1 10 50
Favours NSAID Favours placebo
(Continued )
112Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
( Continued)Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Sinniger 1981 010 010 Not estimable
Spacca 2005 079 076 Not estimable
Thorling 1990 160 060 03 300 [ 012 7220 ]
kermark 1990 422 024 03 978 [ 056 17191 ]
Total (95 CI) 3619 3121 1000 098 [ 080 121 ]Total events 155 (NSAID) 145 (Placebo)
Heterogeneity Chi2 = 3354 df = 35 (P = 054) I2 =00
Test for overall effect Z = 017 (P = 087)
Test for subgroup differences Not applicable
002 01 1 10 50
Favours NSAID Favours placebo
(1) indomethacin
(2) piroxicam
(3) DHEP + heparin and DHEP combined
(4) ketorolac
(5) etofenamate
(6) piroxicam
(7) indomethacin
(8) DHEP + heparin and DHEP combined
(9) Twice and three times daily application combined
113Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Analysis 52 Comparison 5 All topical NSAIDs versus placebo Outcome 2 Systemic adverse events
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 5 All topical NSAIDs versus placebo
Outcome 2 Systemic adverse events
Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Airaksinen 1993 129 027 05 280 [ 012 6593 ]
Aoki 1984 (1) 079 037 Not estimable
Aoki 1984 (2) 070 037 Not estimable
Campbell 1994 126 025 05 289 [ 012 6775 ]
Chatterjee 1977 025 025 Not estimable
Coudreuse 2010 1117 0116 05 297 [ 012 7227 ]
Diebshlag 1990 (3) 013 06 Not estimable
Diebshlag 1990 (4) 012 06 Not estimable
Dreiser 1988 032 032 Not estimable
Dreiser 1989 030 030 Not estimable
Dreiser 1990 030 030 Not estimable
Dreiser 1994 065 066 Not estimable
Fujimaki 1985 (5) 083 041 Not estimable
Fujimaki 1985 (6) 182 041 07 152 [ 006 3647 ]
Haig 1986 021 022 Not estimable
Hoffmann 2012 0115 0118 Not estimable
Joussellin 2003 168 066 05 291 [ 012 7025 ]
Klainguti 2010 (7) 2126 059 07 236 [ 012 4844 ]
Kuehl 2011 15207 23211 236 066 [ 036 124 ]
Li 2013 10192 4192 41 250 [ 080 783 ]
Linde 1985 050 050 Not estimable
Machen 2002 040 041 Not estimable
Mazi res 2005a 1187 785 73 154 [ 062 377 ]
Mazi res 2005b 1381 1482 144 094 [ 047 187 ]
McLatchie 1989 0118 0113 Not estimable
Morris 1991 041 043 Not estimable
001 01 1 10 100
Favours topical NSAID Favours placebo
(Continued )
114Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
( Continued)Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
NCT01272947 2104 2100 21 096 [ 014 670 ]
Parrini 1992 083 086 Not estimable
Predel 2004 060 060 Not estimable
Predel 2012 (8) 3160 382 41 051 [ 011 248 ]
Predel 2013a 6120 8116 84 073 [ 026 203 ]
Predel 2013b 036 136 16 033 [ 001 792 ]
Rowbotham 2003 21191 22181 234 090 [ 052 159 ]
Russell 1991 4102 7102 73 057 [ 017 189 ]
Saillant 1998 070 070 Not estimable
Sanguinetti 1989 042 040 Not estimable
Sinniger 1981 010 010 Not estimable
Spacca 2005 079 076 Not estimable
Thorling 1990 060 060 Not estimable
Total (95 CI) 2956 2620 1000 096 [ 073 126 ]Total events 92 (NSAID) 91 (Placebo)
Heterogeneity Chi2 = 946 df = 15 (P = 085) I2 =00
Test for overall effect Z = 030 (P = 076)
Test for subgroup differences Not applicable
001 01 1 10 100
Favours topical NSAID Favours placebo
(1) piroxicam
(2) indomethacin
(3) ketorolac
(4) etofenamate
(5) piroxicam
(6) indomethacin
(7) DHEP + heparin and DHEP combined
(8) Twice and three times daily application combined
115Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Analysis 53 Comparison 5 All topical NSAIDs versus placebo Outcome 3 Adverse event withdrawals
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 5 All topical NSAIDs versus placebo
Outcome 3 Adverse event withdrawals
Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Airaksinen 1993 029 027 Not estimable
Aoki 1984 072 034 Not estimable
Aoki 1984 064 033 Not estimable
Auclair 1989 1123 0116 14 283 [ 012 6879 ]
Billigmann 1996 280 080 14 500 [ 024 10253 ]
Campbell 1994 026 025 Not estimable
Chatterjee 1977 025 025 Not estimable
Costantino 2011 0188 0142 Not estimable
Coudreuse 2010 0117 0116 Not estimable
Diebshlag 1990 013 06 Not estimable
Diebshlag 1990 012 06 Not estimable
Dreiser 1988 032 032 Not estimable
Dreiser 1989 030 230 69 020 [ 001 400 ]
Dreiser 1990 030 130 42 033 [ 001 787 ]
Dreiser 1994 065 066 Not estimable
Fujimaki 1985 083 041 Not estimable
Fujimaki 1985 482 041 18 455 [ 025 8261 ]
Julien 1989 030 030 Not estimable
Kockelbergh 1985 038 036 Not estimable
Kuehl 2011 4207 9211 248 045 [ 014 145 ]
Li 2013 2192 0192 14 500 [ 024 10347 ]
Linde 1985 050 050 Not estimable
Machen 2002 040 041 Not estimable
Mazi res 2005a 987 685 169 147 [ 055 394 ]
Mazi res 2005b 381 082 14 709 [ 037 13503 ]
McLatchie 1989 0118 0113 Not estimable
001 01 1 10 100
Favours topical NSAID Favours placebo
(Continued )
116Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
( Continued)Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Morris 1991 041 043 Not estimable
NCT01255423 0104 0102 Not estimable
NCT01272934 0102 1103 41 034 [ 001 817 ]
NCT01272947 0104 0100 Not estimable
Noret 1987 148 045 14 282 [ 012 6740 ]
Parrini 1992 060 060 Not estimable
Predel 2004 160 060 14 300 [ 012 7220 ]
Predel 2012 2160 182 37 103 [ 009 1114 ]
Predel 2013a 1120 1116 28 097 [ 006 1527 ]
Predel 2013b 036 036 Not estimable
Ramesh 1983 140 140 28 100 [ 006 1544 ]
Rowbotham 2003 0191 0181 Not estimable
Russell 1991 1102 8102 222 013 [ 002 098 ]
Saillant 1998 070 070 Not estimable
Sanguinetti 1989 042 040 Not estimable
Sinniger 1981 010 010 Not estimable
Spacca 2005 079 076 Not estimable
Thorling 1990 060 060 Not estimable
kermark 1990 122 024 13 326 [ 014 7610 ]
Total (95 CI) 3365 3040 1000 101 [ 064 159 ]Total events 33 (NSAID) 30 (Placebo)
Heterogeneity Chi2 = 1502 df = 16 (P = 052) I2 =00
Test for overall effect Z = 004 (P = 097)
Test for subgroup differences Not applicable
001 01 1 10 100
Favours topical NSAID Favours placebo
117Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Analysis 61 Comparison 6 Topical NSAID versus active comparator Outcome 1 Clinical success - topical
piroxicam vs topical indomethacin
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 6 Topical NSAID versus active comparator
Outcome 1 Clinical success - topical piroxicam vs topical indomethacin
Study or subgroup Piroxicam Indomethacin Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Aoki 1984 5672 4164 301 121 [ 097 151 ]
Fujimaki 1985 4483 4482 307 099 [ 074 131 ]
Sugioka 1984 85175 55165 392 146 [ 112 190 ]
Total (95 CI) 330 311 1000 124 [ 107 144 ]Total events 185 (Piroxicam) 140 (Indomethacin)
Heterogeneity Chi2 = 390 df = 2 (P = 014) I2 =49
Test for overall effect Z = 278 (P = 00055)
Test for subgroup differences Not applicable
01 02 05 1 2 5 10
Favours indomethacin Favours piroxicam
118Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Analysis 62 Comparison 6 Topical NSAID versus active comparator Outcome 2 Local adverse events -
topical piroxicam vs topical indomethacin
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 6 Topical NSAID versus active comparator
Outcome 2 Local adverse events - topical piroxicam vs topical indomethacin
Study or subgroup Piroxicam Indomethacin Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Aoki 1984 179 270 64 044 [ 004 478 ]
Fujimaki 1985 183 582 152 020 [ 002 165 ]
Sugioka 1984 5178 26179 784 019 [ 008 049 ]
Total (95 CI) 340 331 1000 021 [ 009 047 ]Total events 7 (Piroxicam) 33 (Indomethacin)
Heterogeneity Chi2 = 041 df = 2 (P = 081) I2 =00
Test for overall effect Z = 382 (P = 000013)
Test for subgroup differences Not applicable
002 01 1 10 50
Favours piroxicam Favours indomethacin
A P P E N D I C E S
Appendix 1 CENTRAL search strategy (via CRSO) for 2015 update
1 MESH DESCRIPTOR Anti-Inflammatory Agents Non-Steroidal EXPLODE ALL TREES (13419)2 (bufexamac OR bufexine OR calmaderm OR ekzemase OR dicoflenac OR solaraze OR pennsaid OR voltarol OR emugel OR
voltarene OR voltarol OR optha OR voltaren OR etofenamate OR afrolate OR algesalona OR bayro OR deiron OR etofen ORflexium OR flogoprofen OR rheuma-gel OR rheumon OR traumalix OR traumon OR zenavan OR felbinac OR dolinac OR flexfreeOR napageln OR target OR traxam OR fentiazac OR domureuma OR fentiazaco OR norvedan OR riscalon OR fepradinol ORdalgen OR flexidol OR cocresol OR rangozona OR reuflodol OR pinazone OR zepelin OR flufenamic OR dignodolin OR rheumaOR lindofluid OR sastridex OR lunoxaprofen OR priaxim OR flubiprofen OR fenomel OR ocufen OR ocuflur OR ldquoTrans ActLATrdquo OR tulip OR ibuprofen OR cuprofen OR ldquodeep relief rdquo OR fenbid OR ibu-cream OR ibugel OR ibuleve OR ibumousse ORibuspray OR ldquonurofen gelrdquo OR proflex OR motrin OR advil OR radian OR ralgex OR ibutop OR indomethacin OR indocin ORindospray OR isonixin OR nixyn OR ketoprofen OR tiloket OR oruvail OR powergel OR solpaflex OR ketorolac OR acular ORtrometamol OR meclofenamic OR naproxen OR naprosyn OR niflumic OR actol OR flunir OR niflactol topico OR niflugel ORnifluril OR oxyphenbutazone OR californit OR diflamil OR otone OR tanderil OR piketoprofen OR calmatel OR triparsean ORpiroxicam OR feldene OR pranoprofen OR oftalar OR pranox OR suxibuzone OR danilon OR flamilon OR ufenamate OR fenazolOR flector OR benzydamine) TIABKY (25220)
3 1 OR 2 (32484)4 MESH DESCRIPTOR Administration Topical EXPLODE ALL TREES (2169)
119Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
5 (topical OR cutaneous OR dermal OR transcutaneous OR transdermal OR percutaneous OR skin OR massage ORembrocation OR gel OR ointment OR aerosol OR cream OR cregraveme OR lotion OR mouse OR foam OR liniment OR spray OR rubOR balm OR salve OR emulsion OR oil OR patch OR plaster)TIABKY (67940)
6 4 OR 5 (70486)7 MESH DESCRIPTOR Athletic Injuries EXPLODE ALL TREES (411)8 (strain OR sprain OR contusion OR distortion OR compression OR ldquosports injurrdquo OR ldquosoft tissue injurrdquo OR tendnitis OR
ldquomuscle painrdquo OR periarthritis OR epicondylitis OR tenosynovitis)TIABKY (9158)9 7 OR 8 (9448)
10 MESH DESCRIPTOR pain EXPLODE ALL TREES (29943)11 (pain OR analgesi)TIABKY (74815)12 10 OR 11 (80041)13 3 AND 6 AND 9 AND 12 (110)
Appendix 2 MEDLINE search strategy via Ovid (for 2015 update)
1 exp Anti-inflammatory Agents non-steroidal (162888)2 (bufexamac OR bufexine OR calmaderm OR ekzemase OR diclofenac OR solaraze OR pennsaid OR voltarol OR emugel OR
voltarene OR voltarol OR optha OR voltaren OR etofenamate OR afrolate OR algesalona OR bayro OR deiron OR etofen ORflexium OR flogoprofen OR rheuma-gel OR rheumon OR traumalix OR traumon OR zenavan OR felbinac OR dolinac OR flexfreeOR napageln OR target OR traxam OR fentiazac OR domureuma OR fentiazaco OR norvedan OR riscalon OR fepradinol ORdalgen OR flexidol OR cocresol OR rangozona OR reuflodol OR pinazone OR zepelin OR flufenamic OR dignodolin OR rheumaOR lindofluid OR sastridex OR lunoxaprofen OR priaxim OR flubiprofen OR fenomel OR ocufen OR ocuflur OR ldquoTrans ActLATrdquo OR tulip OR ibuprofen OR cuprofen OR ldquodeep relief rdquo OR fenbid OR ibu-cream OR ibugel OR ibuleve OR ibumousse ORibuspray OR ldquonurofen gelrdquo OR proflex OR motrin OR advil OR radian OR ralgex OR ibutop OR indomethacin OR indocin ORindospray OR isonixin OR nixyn OR ketoprofen OR tiloket OR oruvail OR powergel OR solpaflex OR ketorolac OR acular ORtrometamol OR meclofenamic OR naproxen OR naprosyn OR niflumic OR actol OR flunir OR niflactol topico OR niflugel ORnifluril OR oxyphenbutazone OR californit OR diflamil OR otone OR tanderil OR piketoprofen OR calmatel OR triparsean ORpiroxicam OR feldene OR pranoprofen OR oftalar OR pranox OR suxibuzone OR danilon OR flamilon OR ufenamate OR fenazolOR flector OR benzydamine)mp (558284)
3 1 OR 2 (664691)4 exp Administration Topical (69697)5 (topical OR cutaneous OR dermal OR transcutaneous OR transdermal OR percutaneous OR skin OR massage OR
embrocation OR gel OR ointment OR aerosol OR cream OR cregraveme OR lotion OR mouse OR foam OR liniment OR spray OR rubOR balm OR salve OR emulsion OR oil OR patch OR plaster)mp (1982395)
6 4 OR 5 (1997720)7 exp Athletic Injuries (21464)8 (strain OR sprain OR contusion OR distortion OR compression OR ldquosports injurrdquo OR ldquosoft tissue injurrdquo OR tendnitis OR
ldquomuscle painrdquo OR periarthritis OR epicondylitis OR tenosynovitis)mp (420133)9 7 OR 8 (439228)
10 Pain (113906)11 (pain OR analgesi)mp (585249)12 10 or 11 (585249)13 randomized controlled trialpt (401171)14 randomizedab (296222)15 placeboab (155341)16 drug therapyfs (1789858)17 randomlyab (207517)18 trialab (308477)19 groupsab (1318386)20 OR13-1921 3 AND 6 AND 9 AND 12 AND 20 (139)22 limit 21 to yr=ldquo2008-Currentrdquo (56)
120Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Appendix 3 EMBASE search strategy via Ovid (for 2015 update)
1 exp Anti-inflammatory Agents non-steroidal (452266)2 (bufexamac OR bufexine OR calmaderm OR ekzemase OR dicoflenac OR solaraze OR pennsaid OR voltarol OR emugel OR
voltarene OR voltarol OR optha OR voltaren OR etofenamate OR afrolate OR algesalona OR bayro OR deiron OR etofen ORflexium OR flogoprofen OR rheuma-gel OR rheumon OR traumalix OR traumon OR zenavan OR felbinac OR dolinac OR flexfreeOR napageln OR target OR traxam OR fentiazac OR domureuma OR fentiazaco OR norvedan OR riscalon OR fepradinol ORdalgen OR flexidol OR cocresol OR rangozona OR reuflodol OR pinazone OR zepelin OR flufenamic OR dignodolin OR rheumaOR lindofluid OR sastridex OR lunoxaprofen OR priaxim OR flubiprofen OR fenomel OR ocufen OR ocuflur OR ldquoTrans ActLATrdquo OR tulip OR ibuprofen OR cuprofen OR ldquodeep relief rdquo OR fenbid OR ibu-cream OR ibugel OR ibuleve OR ibumousse ORibuspray OR ldquonurofen gelrdquo OR proflex OR motrin OR advil OR radian OR ralgex OR ibutop OR indomethacin OR indocin ORindospray OR isonixin OR nixyn OR ketoprofen OR tiloket OR oruvail OR powergel OR solpaflex OR ketorolac OR acular ORtrometamol OR meclofenamic OR naproxen OR naprosyn OR niflumic OR actol OR flunir OR niflactol topico OR niflugel ORnifluril OR oxyphenbutazone OR californit OR diflamil OR otone OR tanderil OR piketoprofen OR calmatel OR triparsean ORpiroxicam OR feldene OR pranoprofen OR oftalar OR pranox OR suxibuzone OR danilon OR flamilon OR ufenamate OR fenazolOR flector OR benzydamine)mp (790435)
3 1 OR 2 (1108683)4 exp Administration Topical (68490)5 (topical OR cutaneous OR dermal OR transcutaneous OR transdermal OR percutaneous OR skin OR massage OR
embrocation OR gel OR ointment OR aerosol OR cream OR cregraveme OR lotion OR mouse OR foam OR liniment OR spray OR rubOR balm OR salve OR emulsion OR oil OR patch OR plaster)mp (3208844)
6 4 OR 5 (3208844)7 exp Athletic Injuries (24675)8 (strain OR sprain OR contusion OR distortion OR compression OR ldquosports injurrdquo OR ldquosoft tissue injurrdquo OR tendnitis OR
ldquomuscle painrdquo OR periarthritis OR epicondylitis OR tenosynovitis)mp (804322)9 7 OR 8 (825164)
10 Pain (216406)11 (pain OR analgesi)mp (971593)12 10 OR 11 (971593)13 clinical trialssh (841252)14 controlled clinical trialssh (389335)15 randomized controlled trialsh (357169)16 double-blind proceduresh (118945)17 (clin adj25 trial)ab (331002)18 ((doubl or trebl or tripl) adj25 (blind or mask))ab (141726)19 placeboab (203390)20 randomab (906176)21 OR13-20 (1731380)22 3 AND 6 AND 9 AND 10 AND 19 (439)23 limit 22 to yr=ldquo2008-Currentrdquo (204)
Appendix 4 Summary of outcomes efficacy
Study ID Treatment Clinical response Other response
Airaksinen 1993 (1) Ketoprofen gel 2 x 5 g (125mg) daily n = 29(2) Placebo gel n = 27
PGE rdquoimprovedldquo at 7 days(1) 2429(2) 1427
No additional data
121Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
kermark 1990 (1) Indomethacin spray 1 (El-metacin) 3-5 x 05-15 mL dailyn = 23(2) Indomethacin capsules 3 x 25mg daily n = 23(3) Placebo spray and capsules n= 24
No pain on palpation at 7 days(1) 1222(2) 522(3) 624
Patient assessment of improve-ment at 7 days (scale 0 to 100)(1) 57(2) 49(3) 30
Aoki 1984 (1) Piroxicam gel 5 3 or 4 x 1 gdaily n = 84(2) Indomethacin gel 1 3 or 4x 1 g daily n = 84(3) Placebo gel n = 84
PGE (5 point) rdquobetter or muchbetterldquo at 7 days(1) 5672(2) 4164(3) 3367
Pain on movement rdquoreducedldquo orrdquodisappearedldquo at 7 days(1) 4861(2) 3860(3) 3563
Auclair 1989 (1) Niflumic acid gel 25 3 x 5g daily n = 117(2) Placebo gel n = 110
PGE (5 point) rdquogood or verygoodldquo at 7 days(1) 69117(2) 54110
Pain on palpation rdquoimprovedldquo at7 days(1) 69117(2) 53110
Billigmann 1996 (1) Ibuprofen microgel 5 3 x200 mg daily n = 80(2) Placebo gel n = 80
Complete remission(1) 2580(2) 1080
Improvement in pain with move-ment of 20 at 7 days(1) 6580(2) 5580
Campbell 1994 (1) Ibuprofen cream 5 (Proflex)4 x 4rdquo daily n = 26(2) Placebo cream n = 25
Improvement in walking ability (4point) at 7 days(1) 2126(2) 1925
No additional data
Chatterjee 1977 (1) Benzydamine HCl cream 33 x daily n = 25(2) Placebo cream n = 25
Pain on movement ldquoabsentslightrdquoat 6 days(1) 2125(2) 1225
Tenderness with pressure ldquoabsentslightrdquo at 6 days(1) 2125(2) 1225
Costantino 2011 (1) DHEP-heparin plaster (Flec-torparin) x 1 daily n = 142(2) DHEP plaster (Flector) x 1daily n = 146(3) Placebo plaster n = 142
No responder analysis for efficacyOverall treatment efficacy not re-ported
Mean reduction from baseline forPI on movement at 3 days (fromgraph)(1) 24 mm(2) 19 mm(3) 14 mm
Coudreuse 2010 (1) DHEP-heparin plaster (Flec-torparin) n = 120(2) Placebo n = 120
PGE ldquogoodrdquo or ldquoexcellentrdquo at 7days(1) 99117(2) 82116These are results for physician-re-ported judgement - not partici-pant
Mean reduction from baseline forPI on movement over 6 hours(1) about 30(2) about 20Overall greater in DHEP-heparingroup than placebo over 7 days
122Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Curioni 1985 (1) Ibuproxam n = 20(2) Ketoprofen n = 20(3) Etofenamate n = 20
Resolution of symptoms by 7 days(1) 1520(2) 1320(3) 1320
PGE ldquogoodrdquo or ldquoexcellentrdquo at 10days(1) 1920(2) not reported(3) 1620
Diebshlag 1990 (1) Ketorolac gel 2 3 x 3 g dailyn = 13(2) Etofenamate gel 5 3 x 3 gdaily n = 12(3) Placebo gel n = 12
Improvement in pain at 7 days(1) 1213(2) 1012(3) 912
No additional data
Dreiser 1988 (1) Ibuprofen cream 5 3 x 4 cmdaily n = 32 (3 x 10 cm for largejoints)(2) Placebo cream n = 32
PGE ldquoimprovementrdquo or ldquocom-plete relief rdquo at 7 days(1) 2632(2) 1232
(1) significantly better than (2)for mean improvement in sponta-neous pain movement pain restpain tenderness to pressure (VAS)
Dreiser 1989 (1) Ketoprofen gel 25 2 x 5 cmdaily n = 30(2) Placebo gel n = 30
PGE (3 point) ldquobetterrdquo at 7 days(1) 1830(2) 530
(1) significantly better than (2) formean improvement in pain (restand movement) (VAS)
Dreiser 1990 (1) Niflumic acid gel 25 3 x 5g daily n = 30(2) Placebo gel n = 30
PGE (4 point) ldquocuredrdquo or ldquoim-provedrdquo at 7 days(1) 2330(2) 1030
(1) significantly better than (2) formean improvement in pain (VAS)
Dreiser 1994 (1) Flurbiprofen plaster 2 x 40 mgdaily n = 65(2) Placebo plaster n = 66
PGE (4 point) ldquogoodrdquo or ldquoverygoodrdquo at 7 days(1) 4865(2) 4166
(1) significantly better than (2)for mean improvement in sponta-neous pain but not pain on move-ment or palpation (VAS)
Fioravanti 1999 (1) DHEP lecithin gel 3 x 5 g (=65 mg) daily n = 50(2) DHEP gel 3 x 5 g (= 65 mg)daily n = 50
PGE (4 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 10 days(1) 3550(2) 3550
(1) significantly better than (2)for mean improvement in spon-taneous pain at 7 days but notfor pain on movement at 10 days(VAS)
Fujimaki 1985 (1) Piroxicam gel 05 3 or 4 x 1g daily n = 92(2) Indomethacin gel 1 3 or 4 x1 g daily n = 90(3) Placebo gel n = 89
PGE (5 point) ldquobetterrdquo or ldquomuchbetterrdquo at end of treatment at 14days(1) 4483(2) 4482(3) 4082
No additional data
Gallacchi 1990 (1) Diclofenac hy-droxyethylpyrrolidine gel 1 4 x2 g daily n = 25 (Flector gel)(2) Diclofenac sodium 1 4 x 2 g
PGE (5 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 14 days(1) 1925(2) 1925
No significant difference betweengroups for pain on applied pres-sure at 7 and 14 days
123Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
daily n = 25 (Voltaren Emugel)
Gonzaacutelez de Vega 2013 (1) Traumeel gel 3 x 2 g daily n =140(2) Traumeel ointment n = 143(3) Diclofenac gel 1 3 x 2 g dailyn = 137
Pain-free at 7 days(1) 7140(3) 8137PGE (5 point) ldquogoodrdquo or ldquoverygoodrdquo(1) 128140(3) 127137
Normal function (5 point) scoreof 0 or 1 at 14 days(1) 133140(3) 131137
Governali 1995 (1) Ketoprofen gel 5 3 x 2-3 gdaily n = 15(2) Ketoprofen cream 1 3 x 2-3 g daily n = 15
PGE (5 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 7 days(1) 1415(2) 415
No additional data
Gualdi 1987 (1) Flunaxaprofen gel 2 x 3-5 cmdaily n = 30(2) Ketoprofen gel 2 x 3-5 cmdaily n = 30
No dichotomous data No significant difference betweengroups for pain on movement at 7days
Haig 1986 (1) Benzydamine cream 3 6 xdaily n = 21(2) Placebo cream n = 22
Pain on movement ldquoimprovedrdquo by6 days(1) 1821(2) 1322
No additional data
Hoffmann 2012 (1) DHEP-heparin plaster (Flec-torparin) x 1 daily n = 121(2) DHEP (Flector) plaster x 1daily n = 115(3) Placebo plaster n = 118
PGE (5 point) ldquogoodrdquo or ldquoexcel-lentrdquo over 80 in all groups
Mean reduction from baseline inPI (100 mm VAS)Day 3(1) 191 mm(2) 114 mm(3) 52 mmDay 8(1) 443 mm(2) 372 mm(3) 296 mm
Hofman 2000 (1) Diclofenac sodium gel 1 4x 2 cm daily n = 69(2) Lysine clonixinate gel 5 4 x2 cm (225 mg) daily n = 73
PGE (3 point) at 8 days ldquogoodrdquo(1) 3869(2) 3673
No significant difference betweentreatments for any pain outcomes
Hosie 1993 (1) Felbinac foam 3 3 x 2 g daily+ placebo tablets 3 x 1 daily n =140 (127 analysed for efficacy)(2) Ibuprofen tablets 3 x 400 mgdaily + placebo foam 3 x 2 g dailyn = 147 (134 analysed for efficacy)
Pain on movement ldquononerdquo orldquomildrdquo at 7 days(1) 81127(2) 96133
Spontaneous pain ldquononerdquo orldquomildrdquo at 7 days(1) 99127(2) 108134
124Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Jenoure 1997 (1) DHEP plaster (Tissugel) 2 xdaily n = 44(2) Placebo plaster 2 x daily n =41
Baseline pain in two groups notbalanced and data in table and fig-ure do not agree so efficacy out-comes not used
No additional data
Joussellin 2003 (1) DHEP plaster (Flector Tis-sugel 1) 1 x daily n = 68(2) Placebo plaster 1 x daily n =66
PGE (4 point) ldquoexcellentrdquo at 7 days(1) 3668(2) 2466ge 30 reduction in PI at 7 days(1) 2568(2) 1166
(1) significantly better than (2) formean pain on movement at 6 days
Julien 1989 (1) Ketoprofen gel 25 2 x 5 cm(= 50 mg) daily n = 30(2) Placebo gel n = 30
PGE (4 point) ldquorecoveredrdquo at 7days(1) 1830(2) 630
PGE (4 point) ldquorecoveredrdquo or ldquoim-provedrdquo at 7 days(1) 2530(2) 1330
Klainguti 2010 (1) DHEP-heparin plaster (Flec-torparin) x 1 daily n = 62(2) DHEP plaster (Flector) x 1daily n = 61(3) Placebo n = 59
Overall treatment efficacy (partic-ipant and investigator) (5 point)ldquogoodrdquo or ldquoexcellentrdquo at 3 days(1) 5662(2) not reported(3) 4659
Mean reduction in PI at day 3(from graph)(1) 33 mm(2) Not reported(3) 24 mm
Kockelbergh 1985 (1) Ketoprofen gel 25 2 x 5 cm(= 15 mg) daily n = 38(2) Placebo gel n = 36
PGE (3 point) ldquogoodrdquo at 7 days(1) 3038(2) 2236
(1) and (2) slightly better than (3)for mean spontaneous pain at 7days
Kuehl 2011 (1) DETP 13 plaster 2 x dailyn = 207(2) Placebo plaster n = 211
No dichotomous data Percentage reduction from base-line at last application(1) 73(2) 62
Li 2013 (1) DHEP plaster (Flector) 2 xdaily n = 192(2) Placebo plaster 2 x daily n =192
ge 50 pain reduction at day 7(posthoc analysis)(1) 173192(2) 114192
Overall treatment efficacy (5point) ldquogoodrdquo or ldquoexcellentrdquo(1) 161192(2) 81192Participant and investigator ratingMean plusmn SD reduction in PI onmovement at 7 days(1) 538 plusmn 170(2) 370 plusmn 183
Linde 1985 (1) Benzydamine 3 cream 3 xdaily n = 50(2) Placebo gel n = 50
No pain on movement (walking)at 8 days(1) 3550(2) 4050
No additional data
125Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Machen 2002 (1) Ibuprofen gel 5 3 x daily n= 40(2) Placebo gel n = 41
PGE (5 point) ldquomarked improve-mentrdquo or ldquocomplete clearancerdquo at7 days(1) 2540(2) 941
Clinically meaningful (ge 30 mm)pain relief at day 7(1) 3040(2) 1641
Mahler 2003 (1) DHEP + lethicin gel 3 x 5 gdaily n = 52(2) DHEP gel 3 x 5 g daily n = 48
PGE (4 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 10 days(1) 4952(2) 3948
Mean reduction in pain on move-ment at 3 and 10 days significantlygreater with (1) than (2)
Maziegraveres 2005b (1) Ketoprofen plaster 100 mg x1 daily n = 81(2) Placebo plaster n = 82
PGE (4 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 14 days(1) 7281(2) 6082
All mean efficacy measures im-proved more for (1) than (2) mostwere statistically significant
Maziegraveres 2005a (1) Ketoprofen plaster 100 mg x1 daily n = 87(2) Placebo plaster n = 85
PGE (4 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 14 days(1) 5087(2) 4185
All mean efficacy measures im-proved more for (1) than (2) mostwere statistically significant
McLatchie 1989 (1) Felbinac gel 3 3 x 3 cm dailyn = 118(2) Placebo gel n = 113
No dichotomous data Patient daily self-assessment formean pain on rest movement atnight interference with normaland leisure activities show betterefficacy for (1) than (2) from day2 (VAS)
Morris 1991 (1) Felbinac gel 3 3 x 1 cm dailyn = 41(2) Placebo gel n = 43
PGE (5 point) ldquogoodrdquo or ldquoverygoodrdquo at 7 days(1) 2341(2) 2743
(1) better than (2) for meanimprovement in symptoms andsporting function at 7 days
NCT01255423 (1) Diclofenac sodium gel 1 x 4daily n = 104(2) Placebo gel x 4 daily n = 100
No dichotomous data VAS (mean plusmn SD) at 72 hours(baseline PI not reported)(1) 374 plusmn 252(2) 388 plusmn 241
NCT01272934 (1) Diclofenac sodium gel 1 x 4daily n = 104(2) Placebo gel x 4 daily n = 100
No dichotomous data VAS (mean plusmn SD) at 72 hours(baseline PI not reported)(1) 256 plusmn 159(2) 612 plusmn 166
NCT01272947 (1) Diclofenac sodium gel 1 x 4daily n = 104(2) Placebo gel x 4 daily n = 100
No dichotomous data VAS (mean plusmn SD) at 24 hours(baseline PI not reported)(1) 331 plusmn 214(2) 654 plusmn 169
126Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Noret 1987 (1) Ketoprofen gel 25 2 x 5 cm(75 mg) daily n = 48(2) Placebo gel n = 45
PGE (4 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 8 days(1) 3948(2) 945
Decrease in mean spontaneouspain significantly greater in (1)than (2) by 3 days
Parrini 1992 (1) Ketoprofen foam 15 3 x 2 g(200 mg) daily n = 83(2) Placebo foam n = 86
No dichotomous data Mean pain on movement andpressure significantly decreased by7 days in (1) compared with (2)
Picchio 1981 (1) Ibuprofen gel 10 3 x daily n = 20(2) Ketoprofen gel 1 3 x daily n= 20
No pain on movement at 8 days(1) 320(2) 020
Spontaneous pain ldquononerdquo at 8days(1) 620(2) 020
Predel 2004 (1) Diclofenac sodium plaster 2 xdaily (140 mgplaster) n = 60(2) Placebo plaster n = 60
PGE (4 point) ldquogoodrdquo ldquoexcellentrdquoat 7 days(1) 5560(2) 560
(1) better than (2) for reductionin tenderness pain and speed ofpain reduction
Predel 2012 (1) Diclofenac gel (VoltarenEmulgel 232) 2 x 5 cm daily n= 80(2)Diclofenac gel (Voltaren Emulgel232 ) 3 x 5 cm daily n = 80(3) Placebo gel n = 82
ge 50 red in PI on movement at5 days(1) 5780(2) 5980(3) 1782
PGE efficacy (5 point) ldquogoodrdquo orldquovery goodrdquo at 8 days(1) 6880(2) 7380(3) 2482
Predel 2013a (1) Diclofenac 4 spray gel 4 or5 sprays 3 x daily (96-120 mg di-clofenac sodium) n = 118(2) Placebo spray gel n = 114
None or slight PI on movement at7-8 days(1) 111118(2) 93114
None or slight PI on movement at3-4 days(1) 76118(2) 58114
Predel 2013b (1) Diclofenac gel (VoltarenEmulgel) 4 x 2 g daily n = 36(2) Placebo gel 4 x 2 g daily n =36
PGE (5 point) ldquogoodrdquo or ldquoexcel-lentrdquo(1) 3636(2) 736
ge 50 reduction in PI on move-ment after 48 hours(1) 3436(2) 336
Ramesh 1983 (1) Ibuprofen cream 5 3 or 4 x5-10 cm daily n = 40(2) Placebo cream n = 40
Pain on movement (4 point)ldquononerdquo or ldquoslightrdquo at 7 days(1) 2340(2) 2340
Physician global assessment at 10days ldquogoodrdquo(1) 2940(2) 1640
Rowbotham 2003 (1) Diclofenac epolamine plaster(Flector Tissuegel) 2 x daily n =191(2) Placebo plaster n = 181
Pain intensity le 210 for 2 daysor 4 consecutive evaluations by 7days(1) 75191(2) 48181
Mean pain on rest significantlybetter with (1) than (2) after 7 days
127Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Russell 1991 (1) Piroxicam gel 05 4 x 5 mgdaily n = 100(2) Placebo gel n = 100
PGE (4 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 8 days(1) 79100(2) 45100
Statistically greater reduction inmean pain on movement at 8 dayswith (1) than (2)
Saillant 1998 (1) DHEP plaster (Flector Tis-sugel 1) 1 x daily n = 70(2) Placebo plaster 1 x daily n =70
PGE (4 point) ldquoexcellentrdquo at 7 days(1) 4370(2) 2570
ge 30 decrease in PI at 7 days(1) 6470(2) 5070
Sanguinetti 1989 (1) Felbinac gel 3 3 x daily n =42(2) Placebo gel n = 40
PGE ldquogoodrdquo or ldquovery goodrdquo at 7days(1) 3442(2) 1140
(1) better than (2) by 2 days
Sinniger 1981 (1) Fentiazac cream 5 2 or 3 xdaily n = 10(2) Placebo cream n = 10
Complete pain relief within 10days(1) 710(2) 110
Improvement in active pain onmovement at 5 days(1) 67(2) 32
Spacca 2005 (1) DHEP lecithin gel (Effigel) 3x 5 g daily n = 79(2) Placebo gel n = 76
No dichotomous data Mean pain scores improved morerapidly in (1) than (2) - statisticallysignificant at 3 and 6 days
Sugioka 1984 (1) Piroxicam gel 05 3 or 4 x 1g daily n = 183(2) Indomethacin gel 1 3 or 4 x1 g daily n = 183
PGE (5 point) ldquobetterrdquo or ldquomuchbetterrdquo at 14 days(1) 85175(2) 55165
Pain on movement ldquoreducedrdquo orldquodisappearedrdquo at 7 days(1) 77175(2) 63165
Thorling 1990 (1) Naproxen gel 10 2-6 x dailyn = 60(2) Placebo gel n = 60
PGE (5 point) ldquogoodrdquo or ldquoverygoodrdquo at 7 days(1) 3860(2) 2760
Participants using naproxen im-proved more rapidly and had sig-nificantly lower severity scores byday 3
Tonutti 1994 (1) Ketoprofen gel 5 3 x 2-3 gdaily n = 15(2) Etofenamate gel 5 3 x 2-3g n = 15
PGE (4 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 7 days(1) 1015(2) 1115
Significant reductions in pain onmovement by 7 days in bothgroups
Vecchiet 1991 (1) Meclofenamic acid gel 5 2 x10 cm daily (2 g) n = 30(2) Placebo n = 30
PGE (4 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 10 days(1) 3030(2) 1930
(1) significantly better than (2)for mean improvement in sponta-neous pain movement pain func-tional restriction
Whitefield 2002 (1) Ibuprofen gel 5 + placebotablet 3 x daily n = 50(2) Ibuprofen 400 mg tablet +placebo gel 3 x daily n = 50
Participant satisfied at 7 days(1) 3050(2) 3650
ldquoCompletely betterrdquo at 14 days(1) 2450(2) 3050
128Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
DHEP diclofenac epolamine HCl hydrochloride n number PGE participant global evaluation PI pain intensity SD standarddeviation VAS visual analogue scale
Appendix 5 Summary of outcomes adverse events and withdrawals
Study ID Treatment Local AEs Systemic AEs Serious AEs Withdrawals
Airaksinen 1993 (1) Ketoprofen gel 2x 5 g (125 mg) dailyn = 29(2) Placebo gel n =27
(1) 529(2) 427
(1) 129 (nausea af-ter paracetamol)(2) 027
None AE noneOther nonereported
kermark 1990 (1)Indomethacin spray1 (Elmetacin) 3-5 x 05-15 mL dailyn = 23(2) Indomethacincapsules 3 x 25 mgdaily n = 23(3) Placebo sprayand capsules n = 24
(1) 422(2) 022(3) 024
No usable data -reported for eventsnot participants
None reported AE (1) 1 (2) 1 (3)0Lost to follow-up(1) 1 (2) 2 (3) 3
Aoki 1984 (1) Piroxicam gel5 3 or 4 x 1 gdaily n = 84(2) Indomethacingel 1 3 or 4 x 1 gdaily n = 84(3) Placebo gel n =84
(1) 179(2) 270(3) 274
None None reported AE none23 excluded for pro-tocol violations (1)7 (2) 7 (3) 926 withdrew for rea-sons unrelated totreatment (1) 5 (2)13 (3) 8
Auclair 1989 (1) Niflumic acidgel 25 3 x 5 gdaily n = 117(2) Placebo gel n =110
All AEs(1) 5123(2) 6116Most commonly cu-taneous eruptions
No usable data None reported AE (1) 1123 (2) 011626 excluded fromefficacy analysis fornot meeting entrycriteria and protocolviolations
Billigmann 1996 (1) Ibuprofen mi-crogel 5 3 x 200mg daily n = 80(2) Placebo gel n =
(1) 1180(2) 480
None reported None reported AE (1) 280 (aller-gic rash dermatitis)No reason given (1)380 (2) 580
129Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
80 Symptom-free (1)180 (2) 180
Campbell 1994 (1) Ibuprofen cream5 (Proflex) 4 x 4rdquodaily n = 26(2) Placebo cream n= 25
No data (1) 126 (headache)(2) 025
No data AE noneExclusions 493 presented late 2missing forms 1 ap-peared twice 43 didnot return diaries
Chatterjee 1977 (1) BenzydamineHCl cream 3 3 xdaily n = 25(2) Placebo cream n= 25
None None None AE none1 participant lostto follow-up (groupnot reported)
Costantino 2011 (1) DHEP-hep-arin plaster (Flector-parin) x 1 daily n =142(2) DHEP plaster(Flector) x 1 daily n= 146(3) Placebo plastern = 142
Possibly or probablydrug-related(1) 2142(2) 1145(3) 6142
All mild and goneby 14 days ex-cept 1 (moderate) inplacebo group
Not reportedNo systemic GIevents
None AE noneOther(1) 0142(2) 2146 (proto-col deviation lost tofollow-up)(3) 2142 (lost tofollow-up)
Coudreuse 2010 (1) DHEP-hep-arin plaster (Flector-parin) n = 120(2) Placebo n = 120
(1) 1117(2) 7116
All AEs mild ormoderate resolvedspontaneously
(1) 1117(2) 0116
None AE(1) 1117 (increasedoedema)(2) 0116Other(1) 6117 (recovery3 no reason 3)(2) 3116 (recovery1 no reason 2)
Curioni 1985 (1) Ibuproxam n =20(2) Ketoprofen n =20(3) Etofenamate n =20
None None None AE noneOther none
Diebshlag 1990 (1) Ketorolac gel2 3 x 3 g daily n =13(2) Etofenamate gel5 3 x 3 g daily n =12
(1) 113(2) 112(3) 012
None None AE none1 ketorolac partic-ipant did not at-tend 15 day follow-up due to car acci-
130Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
(3) Placebo gel n =12
dent
Dreiser 1988 (1) Ibuprofen cream5 3 x 4 cm dailyn = 32 (3 x 10 cmfor large joints)(2) Placebo cream n= 32
No usable data None Not reported AE none4 placebo partici-pants lost to follow-up
Dreiser 1989 (1) Ketoprofen gel25 2 x 5 cm dailyn = 30(2) Placebo gel n =30
(1) 030(2) 230
None None reported AE (2) 230 (intol-erance)LoE (1) 130 (2) 130
Dreiser 1990 (1) Niflumic acidgel 25 3 x 5 gdaily n = 30(2) Placebo gel n =30
(1) 030(2) 330
None None AE (2) 130 (ery-thema)Exclusion 1 from(2) from efficacyanalysis for inade-quate baseline pain
Dreiser 1994 (1)Flurbiprofen plaster2 x 40 mg daily n =65(2) Placebo plastern = 66
(1) 265(2) 066
None None AE 0(1) 165 excludedfrom efficacy analy-sis for protocol vio-lation(2) 266 (1 LoE 1cured)
Fioravanti 1999 (1) DHEP lecithingel 3 x 5 g (= 65 mg)daily n = 50(2) DHEP gel 3 x 5g (= 65 mg) daily n= 50
(1) 050(2) 150
No data None reported AE noneOther none
Fujimaki 1985 (1) Piroxicam gel 05 3 or 4 x 1 g dailyn = 92(2) Indomethacingel 1 3 or 4 x 1 gdaily n = 90(3) Placebo gel n =89
(1) 183(2) 582(3) 282
(1) 083(2) 182 (nauseaand vomiting)(3) 082
None AE (1) 0 (2) 4 (3)0Unknown reasons(1) 2 (2) 1Did not return af-ter 1st visitirregularvisits (1) 6 (2) 6(3) 7
Gallacchi 1990 (1) Diclofenac hy-droxyethylpyrro-lidine gel 1 4 x 2
No AEs None None AE noneOther none
131Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
g daily n = 25 (Flec-tor gel)(2) Di-clofenac sodium 14 x 2 g daily n = 25(Voltaren Emugel)
Gonzaacutelez de Vega2013
(1) Traumeel gel 3 x2 g daily n = 140(2) Traumeel oint-ment n = 143(3) Diclofenac gel1 3 x 2 g daily n =137
No usable data
Infrequent mild tomoderate
No usable data None AE noneOther(1) 11140(2) 12137
Governali 1995 (1) Ketoprofen gel5 3 x 2-3 g dailyn = 15(2) Ketoprofencream 1 3 x 2-3 gdaily n = 15
No side effects None None AE noneOther none
Gualdi 1987 (1) Flunaxaprofengel 2 x 3-5 cm dailyn = 30(2) Ketoprofen gel 2x 3-5 cm daily n =30
(1) 130(2) 330
No data None reported AE noneOther none
Haig 1986 (1) Benzydaminecream 3 6 x dailyn = 21(2) Placebo cream n= 22
No AEs reported None None reported AE none reportedOther no data
Hoffmann 2012 (1) DHEP-hep-arin plaster (Flector-parin) x 1 daily n =121(2) DHEP (Flector)plaster x 1 daily n =115(3) Placebo plastern = 118
Most AEs were mi-nor local reactions(eg pruritus anderythema) in areaof plaster of mildto moderate inten-sity and resolvedwithout interrupt-ing treatment
No treatment-related systemic AEsrecorded
(2) 1 participanthad 3 SAEs nonejudged related tostudy medication
AE noneExclusions(1) 5121(2) 10115(3) 7118Excluded fromper protocol analysisdue to poor compli-ance or personal de-cision
Hofman 2000 (1) Diclofenacsodium gel 1 4 x2 cm daily n = 69(2) Lysine clonixi-
(1) 158(2) 161
None None AE noneLoE (1) 9 (2) 8
132Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
nate gel 5 4 x 2cm (225 mg) dailyn = 73
Hosie 1993 (1) Felbinac foam3 3 x 2 g daily+ placebo tablets 3x 1 daily n = 140(127 analysed for ef-ficacy)(2) Ibupro-fen tablets 3 x 400mg daily + placebofoam 3 x 2 g daily n= 147 (134 analysedfor efficacy)
(1) 1127(2) 3134
GI events (1) 14127 (2) 11134For (1) more mildnone definitely drugrelated for(2) definitely relatedto study drug
None AE noneExclusions (1) 13(2) 13 did not returnfor 7 day follow-up
Jenoure 1997 (1) DHEP plaster(Tissugel) 2 x dailyn = 44(2) Placebo plaster 2x daily n = 41
(1) 144(2) 141
No data None reported AE none reportedOther nonereported
Joussellin 2003 (1) DHEPplaster (Flector Tis-sugel 1) 1 x dailyn = 68(2) Placebo plaster 1x daily n = 66
(1) 168 (pruritus)(2) 366 (pruritus 2burning 1)
All AEs mild ormoderate
(1) 168 (allergic re-action)(2) 066
None reported AE(1) 066(2) 166Other(1) 366(2) 266
Julien 1989 (1) Ketoprofen gel25 2 x 5 cm (= 50mg) daily n = 30(2) Placebo gel n =30
(1) 130(2) 030
Not reported None AE noneOther none
Klainguti 2010 (1) DHEP-hep-arin plaster (Flector-parin) x 1 daily n =62(2) DHEP plaster(Flector) x 1 daily n= 61(3) Placebo n = 59
(1) 062(2) 061(3) 159All AEs mild in na-ture and resolvedspontaneously
(1) 165 (facial in-fection)(2) 161 (abdominalpain)(3) 059
None AE noneOther(1) 365(2) 161(3) 359
Kockelbergh 1985a (1) Ketoprofen gel25 2 x 5 cm (= 15mg) daily n = 38(2) Placebo gel n =
(1) 138(2) 126
Not reported None AE noneOther none
133Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
36
Kuehl 2011 (1) DETP 13plaster 2 x daily n =207(2) Placebo plastern = 211
(1) 16207(2) 12211
(1) 15207(2) 23211
None AE(1) 4207(2) 9211LoE(1) 21207(2) 25211Other(1) 58207(2) 62211
Li 2013 (1) DHEP plaster(Flector) 2 x daily n= 192(2) Placebo plaster 2x daily n = 192
(1) 4192(2) 3192
(1) 10192(2) 4192
None AE(1) 2192(2) 0192Other(1) 5192(2) 1192
Linde 1985 (1) Benzydamine3 cream 3 x dailyn = 50(2) Placebo gel n =50
(1) 440(2) 241
None None AE none(1) 6 (2) 6 excludedfrom 1st assessment(1) 3 (2) 4 ex-cluded from final as-sessment
Machen 2002 (1) Ibuprofen gel5 3 x daily n = 40(2) Placebo gel n =41
(1) 440(2) 241
None None AE none(1) 1 LoE 1 proto-col violation(2) 4 LoE
Mahler 2003 (1) DHEP + lethicingel 3 x 5 g daily n =52(2) DHEP gel 3 x 5g daily n = 48
(1) 152(2) 048
(1) 152(2) 048
None AE none5 lost to follow-up
Maziegraveres 2005b (1) Ketoprofen plas-ter 100 mg x 1 dailyn = 81(2) Placebo plastern = 82
At 21 days(1) 1281(2) 682
(1) 1381(2) 1482
None AE (1) 381(1) 781 (1 LoE 6cured)(2) 782 (5 LoE 2cured)
Maziegraveres 2005a (1) Ketoprofen plas-ter 100 mg x 1 dailyn = 87(2) Placebo plastern = 85
At 21 days(1) 2987(2) 2785
(1) 1187(2) 785
None AE (1) 987 (2) 685(1) 687 (2 LoE 4cured)(2) 585 (4 LoE 1cured)
134Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
McLatchie 1989 (1) Felbinac gel 33 x 3 cm daily n =118(2) Placebo gel n =113
(1) 3118(2) 2113Mild transient localirritation
None reported None AE noneOther none
Morris 1991 (1) Felbinac gel 33 x 1 cm daily n =41(2) Placebo gel n =43
None None None AE none(1) 4 (protocol vio-lations)(2) 1 (lost to follow-up)Exclusions 11 fromefficacy analysis be-cause evaluated by4 different investi-gators
NCT01255423 (1) Diclofenacsodium gel 1 x 4daily n = 104(2) Placebo gel x 4daily n = 100
(1) 1104(2) 3102
Total AE(1) 11104(2) 8102
None None
NCT01272934 (1) Diclofenacsodium gel 1 x 4daily n = 104(2) Placebo gel x 4daily n = 100
(1) 1102(2) 0103
Total AE (excludingSAE)(1) 6102(2) 3103
(1) 0102(2) 1103 (rupturedligaments in wrist)
AE see SAE
NCT01272947 (1) Diclofenacsodium gel 1 x 4daily n = 104(2) Placebo gel x 4daily n = 100
None (1) 2104(2) 2100
None None
Noret 1987 (1) Ketoprofen gel25 2 x 5 cm (75mg) daily n = 48(2) Placebo gel n =45
(1) 151(2) 047
None reported Not reported AE (1) 151 (skinallergy)(1) 1 LoE 1 unre-lated to trial(2) 1 LoE 1 unre-lated to trial
Parrini 1992 (1) Ketopro-fen foam 15 3 x 2g (200 mg) daily n= 83(2) Placebo foam n= 86
None None None AE noneOther none
135Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Picchio 1981 (1) Ibuprofen gel10 3 x daily n =20(2) Ketoprofen gel1 3 x daily n = 20
None None None AE noneOther not reported
Predel 2004 (1) Diclofenacsodium plaster 2 xdaily (140 mgplas-ter) n = 60(2) Placebo plastern = 60
12 participants ex-pe-rienced 16 mild AEswith no differencesbetween groups
None None AE (1) 160Other none
Predel 2012 (1) Diclofenac gel(Voltaren Emulgel232) 2 x 5 cmdaily n = 80(2) Diclofenac gel(Voltaren Emulgel232) 3 x 5 cmdaily n = 80(3) Placebo gel n =82
(1) 080(2) 180(3) 182All mild to moder-ate
(1) 280(2) 180(3) 382
None AE(1) 080(2) 080(3) 182Other(1) 180(2) 280(3) 282(protocol violationslost to follow-up)
Predel 2013a (1) Diclofenac 4spray gel 4 or 5sprays 3 x daily (96-120 mg diclofenacsodium) n = 118(2) Placebo spraygel n = 114
(1) 1118(2) 4114
(1) 5118(2) 4114All AEs mild re-versible
None AE(1) 1118(2) 1114Other(1) 3118(2) 43114
Predel 2013b (1) Diclofenac gel(Voltaren Emulgel)4 x 2 g daily n = 36(2) Placebo gel 4 x 2g daily n = 36
None (1) 036(2) 136
None AE noneOther none
Ramesh 1983 (1) Ibuprofen cream5 3 or 4 x 5-10 cmdaily n = 40(2) Placebo cream n= 40
(1) 140(2) 140
None reported Not reported AE (1) 140 (2) 140Other none
Rowbotham 2003 (1) DHEP plaster(Flector Tissuegel) 2x daily n = 191(2) Placebo plastern = 181
(1) 27191 (pruritis14)(2) 31181 (pruritis21)
(1) 21191(2) 22181
None re-ported (ldquovast major-ity mildrdquo)
AE none(1) 3191 (2) 4181(did not finish trialand complete dailydiaries)
136Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Russell 1991 (1) Piroxicam gel 05 4 x 5 mg daily n= 100(2) Placebo gel n =100
(1) 4102(2) 10102
GI or CNS events(1) 4 (2) 7Any AE (1) 7102(2) 15102
None reported AE (1) 1102 (2) 8102Other (1) 6 LoE 1ldquootherrdquo(2) 42 LoEExclusions7 did not complywith study medica-tion schedule 6 lostto follow-up 1 pro-tocol violation
Saillant 1998 (1) DHEPplaster (Flector Tis-sugel 1) 1 x dailyn = 70(2) Placebo plaster 1x daily n = 70
None None None AE noneOther(1) 570(2) 570
Sanguinetti 1989 (1) Felbinac gel 33 x daily n = 42(2) Placebo gel n =40
(1) 342(2) 140
None None reported AE noneOther nonereported
Sinniger 1981 (1) Fentiazac cream5 2 or 3 x daily n= 10(2) Placebo cream n= 10
ldquoNo untoward sideeffectsrdquo
None None AE noneOther nonereported
Spacca 2005 (1) DHEP lecithingel (Effigel) 3 x 5 gdaily n = 79(2) Placebo gel n =76
ldquoNo signsof cutaneous irrita-tion or sensitisationobservedrdquo
No AEs observed None AE noneOther nonereported
Sugioka 1984 (1) Piroxicam gel 05 3 or 4 x 1 g dailyn = 183(2) Indomethacingel 1 3 or 4 x 1 gdaily n = 183
(1) 5178(2) 26179
None reported None reported AE none reportedExclusions due toprotocol violations(1) 8 (2) 18Withdrawals(1) 11 (2) 12
Thorling 1990 (1) Naproxen gel10 2-6 x daily n =60(2) Placebo gel n =60
(1) 160(2) 060
None None AE none(1) 1 LoE 1 proto-col violation(2) 1 participant re-quest
137Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Tonutti 1994 (1) Ketoprofen gel5 3 x 2-3 g dailyn = 15(2) Etofenamate gel5 3 x 2-3 g n =15
None No AEs attributableto the medication
None AE NoneLoE (1) 1 (2) 2
Vecchiet 1991 (1) Meclofe-namic acid gel 5 2x 10 cm daily (2 g)n = 30(2) Placebo n = 30
Tolerabil-ity excellent or goodin nearly all partici-pants
No data None AE none reported(2) 5 lost to follow-up
Whitefield 2002 (1) Ibuprofen gel5 + placebo tablet3 x daily n = 50(2) Ibuprofen 400mg tablet + placebogel 3 x daily n = 50
No data 6 AEs reportednone judged relatedto study medication
None reported AE noneRecovered (1) 3 (2)2LoE (2) 1Lost to follow-up(1) 1 (2) 1
AE adverse event CNS central nervous system DHEP diclofenac epolamine GI gastrointestinal HCl hydrochloride LoE lackof efficacy n number SAE serious adverse event
Appendix 6 Concentration amount and frequency of dosing
Study Drug Concentration Quantity Frequency Estimated daily dose of topi-cal NSAID
Joussellin 2003 Diclofenac 1 Plaster 1 180 mg epolamine salt 140 mgNa salt
Li 2013 Diclofenac 1 Plaster 2 360 mg epolamine salt 280 mgNa salt
Rowbotham 2003 Diclofenac 1 Plaster 2 360 mg epolamine salt 280 mgNa salt
Saillant 1998 Diclofenac 1 Plaster 1 180 mg epolamine salt 140 mgNa salt
Coudreuse 2010 Diclofenac 1 Plaster 1 180 mg epolamine salt 140 mgNa salt
Klainguti 2010 Diclofenac 1 Plaster 1 180 mg epolamine salt 140 mgNa salt
138Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Predel 2004 Diclofenac - Plaster 2 280 mg Na salt
Predel 2012 Diclofenac 232 5 cm ribbon~ 2 g 2 or 3 92-138 mg (Na equiv) as di-ethylamine salt
Predel 2013b Diclofenac 116 2 g 4 92 mg (Na equiv) as diethy-lamine salt
Predel 2013a Diclofenac 4 4-5 sprays 3 96-120 mg Na salt
Costantino 2011 Diclofenac 1 Plaster 1 180 mg epolamine salt 140 mgNa salt
Fioravanti 1999 Diclofenac - 5 g 3 195 mg epolamine salt
Gallacchi 1990 Diclofenac 1 2 g 4 80 mg
Gonzaacutelez de Vega2013
Diclofenac 1 6 cm 2g 3 60 mg
Hoffmann 2012 Diclofenac 1 Plaster 1 180 mg epolamine salt 140 mgNa salt
Hofman 2000 Diclofenac 1 2 cm 4 probably 30-40 mg Na salt
Jenoure 1997 Diclofenac 1 Plaster 2 360 mg epolamine salt 280 mgNa salt
Mahler 2003 Diclofenac - 5 g 3 195 mg epolamine salt
NCT01255423 Diclofenac 1 - 4 -
NCT01272934 Diclofenac 1 - 4 -
NCT01272947 Diclofenac 1 - 4 -
Spacca 2005 Diclofenac - 5 g 3 195 mg epolamine salt
Campbell 1994 Ibuprofen 5 4 inch ribbon 4 Assume 800 mg
Dreiser 1988 Ibuprofen 5 4 cm ribbon(10 cm for largerjoints)
4 Assume up-800 mg
Ramesh 1983 Ibuprofen 5 5-10 cm ribbon 3-4 Assume 300-800 mg
Billigmann 1996 Ibuprofen 5 10 cm 4 g gel 3 600 mg
139Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Machen 2002 Ibuprofen 5 - 3 -
Picchio 1981 Ibuprofen 10 - 3 -
Whitefield 2002 Ibuprofen 5 - 3 -
Maziegraveres 2005b Ketoprofen - Plaster 1 100 mg
Maziegraveres 2005a Ketoprofen - Plaster 1 100 mg
Airaksinen 1993 Ketoprofen - 5 g 2 125 mg
Dreiser 1989 Ketoprofen 25 5 cm 2 100 mg
Julien 1989 Ketoprofen 25 5 cm 2 100 mg
Noret 1987 Ketoprofen 25 5 cm 75 g 2 375 mg
Curioni 1985 Ketoprofen No details 2 -
Governali 1995 Ketoprofen 5 2-3 g 3 300-450 mg
Governali 1995 Ketoprofen 1 2-3 g 3 60-90 mg
Gualdi 1987 Ketoprofen - 3-5 cm 2 -
Kockelbergh 1985 Ketoprofen 25 5 cm 2 100 mg
Parrini 1992 Ketoprofen 150 2 g 3 600 mg
Picchio 1981 Ketoprofen 1 - 3 -
Tonutti 1994 Ketoprofen 5 2-3 g 3 300-450 mg
kermark 1990 Indomethacin 1 05-15 mL 3-5 12-60 mg
Aoki 1984 Indomethacin 1 1 g 3-4 30-40 mg
Fujimaki 1985 Indomethacin 1 1 g 3-4 30-40 mg
Sugioka 1984 Indomethacin 1 1 g 3-4 30-40 mg
Aoki 1984 Piroxicam 05 1 g 3-4 15-20 mg
Fujimaki 1985 Piroxicam 05 1 g 3-4 15-20 mg
140Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Russell 1991 Piroxicam 05 5 mg 4 20 mg
Sugioka 1984 Piroxicam 05 1 g 3-4 15-20 mg
Chatterjee 1977 Benzydamine 3 - 3 -
Haig 1986 Benzydamine 3 - 6 -
Linde 1985 Benzydamine 3 - 3 -
Auclair 1989 Niflumic acid 25 10 cm 5 g 3 375 mg
Dreiser 1990 Niflumic acid 25 10 cm 5 g 3 375 mg
Curioni 1985 Ibuproxam gel 10 - 2 -
Curioni 1985 Etofenamate No details - 2 -
Diebshlag 1990 Etofenamate 5 3 g 3 450 mg
Tonutti 1994 Etofenamate 5 2-3 g 3 300-450 mg
Diebshlag 1990 Ketorolac 2 3 g 3 360 mg
Dreiser 1994 Flurbiprofen Patch 2 80 mg
Gualdi 1987 Flunoxaprofen - 3-5 cm 2 -
Hofman 2000 Lysine clonixinate 5 2 cm 4 90 mg
Hosie 1993 Felbinac 3 2 g 3 180 mg
McLatchie 1989 Felbinac 3 3 cm 3 -
Morris 1991 Felbinac 3 1 cm 3 -
Sanguinetti 1989 Felbinac 3 - 3 -
Sinniger 1981 Fentiazac 5 Varied according toinvolved areas
2-3 -
Thorling 1990 Naproxen 10 - 2-6 -
Vecchiet 1991 Meclofenamic acid 5 10 cm 4 g 2 400 mg
equiv equivalent Na sodium
141Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
F E E D B A C K
Query on formulations of topical NSAIDs particularly DMSO from Dr Chrubasik 11 April 2012
Summary
Dr Chrubasik highlighted this letter to the Editor postgradmedorgdoi103810pgm2011092482
DMSO [dimethyl sulphoxide] but also other additives eg nonivamide (which is a capsaicinoid added as drug enhancer) may contributeto the overall effect of topical NSAIDs Nonivamide certainly contributes to the analgesic effect and to adverse events (heat sensationburning pruritus etc) This has not been considered in the Cochrane review by Massey T Derry S Moore RA McQuay HJ TopicalNSAIDs for acute pain in adults but Dr Chrubasik believes should be done otherwise the effect size of the NSAID topicals is favoured
Reply
We have been asked by Dr Chrubasik to comment on a letter (Roth 2011) about the formulations of topical NSAIDs particularly howDMSO and other penetration enhancers can affect efficacy estimates or adverse event reporting in osteoarthritis It was suggested thatthe review of Topical NSAIDs for acute pain in adults did not consider this resulting in a bias towards the topical NSAIDThere are a number of points to be made here
1 Penetration enhancers are used in formulations of topical products to encourage local absorption through the skin and produce ahigh local concentration Topical NSAIDs use penetration enhancers and the result is high local concentration in joints for instancebut low systemic concentrations (Moore 2008) That is how they work Formulation is an important part of medicinal chemistry as awhole not just for topical agents
2 In our analysis of topical NSAIDs we were aware that a range of properties are or have been ascribed to the analgesia resultingfrom application of topical agents and which could contribute to overestimation of treatment effect of topical NSAID These includefeelings of heat or cold and even the act of rubbing itself For that reason we have chosen to include only double-blind studies wherethe placebo agent is identical to the active with the exception of course of the NSAID So heat cold rubbing and penetrationenhancers should be identical as best we can judge That leaves only the NSAID itself to provide any additional analgesic effect andit is that which we measure This is analogous for example to use of acupuncture say where the better studies show no differencebetween ldquotruerdquo acupuncture and ldquoshamrdquo acupuncture performed at nonspecific sites but better than non-treatment controls Theargument that we should only use high quality studies to evaluate evidence about pain interventions is well made
3 Overestimation of analgesic effect because of effects of enhancers themselves would be better made in direct comparisons oftopical and oral NSAIDs where local or even systemic effects would not be balanced in the oral study arm However our reviewconcentrated on placebo-controlled studies and had few studies with active controls Moreover the real test would be in chronicrather than acute conditions with long duration (12 week) outcomes using current best evidence rules (Moore 2010) includingimputation (Moore 2012) In their response to Rothrsquos letter the authors of the original review of products available in the USA showrather similar effect sizes of oral diclofenac and topical diclofenac with different penetration enhancers (Barthel 2011) in such studies
4 The Roth letter sought to differentiate between topical diclofenac preparations based on the penetration enhancers used Thatdifferent formulations may have different effect sizes is a fair point to make Two of the studies in our review of topical NSAIDs inacute conditions used diclofenac sodium 1 gel comparing it with either diclofenac epolamine gel (Gallacchi 1990 50 participants)or lysine clonixinate gel (Hofman 2000 142 participants) no difference between formulations was demonstrated It is difficult tomake any judgement for topical NSAIDs in acute conditions due to the relatively small number and particularly the small size ofstudies We did an analysis by drug and this showed that some topical NSAIDs were consistently beneficial irrespective offormulation while others had little or no efficacy This fits in with some theoretical considerations of molecular architecture andtissue penetration (Moore 2008) In chronic pain where there are larger studies and much more data we have considered formulation(Derry in preparation)
References
Roth SH Letter to the editor The importance of differentiating between topical NSAIDs Postgraduate Medicine 2011123251-2[103810pgm2011092482]Moore RA Derry S McQuay HJ Topical agents in the treatment of rheumatic pain Rheumatic Diseases Clinics of North America200834415-32
142Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Moore RA Eccleston C Derry S Wiffen P Bell RF Straube S et al ACTINPAIN Writing Group of the IASP Special Interest Groupon Systematic Reviews in Pain Relief Cochrane Pain Palliative and Supportive Care Systematic Review Group Editors ldquoEvidencerdquo inchronic pain - establishing best practice in the reporting of systematic reviews Pain 2010150386-9Moore RA Straube S Eccleston C Derry S Aldington D Wiffen P et al Estimate at your peril imputation methods for patientwithdrawal can bias efficacy outcomes in chronic pain trials using responder analyses Pain 2012153265-8Barthel HR Axford-Gatley RA Response to Roth Letter to the Editor Postgraduate Medicine 2011123253-4 [103810pgm2011092483]
Contributors
Feedback from Sigrun ChrubasikAuthors involved with responding Andrew Moore Sheena DerryFeedback Editor Kate Seers
Query submitted by Peter C Goslashtzsche 3 November 2015
Summary
Date of Submission 03-Nov-2015Name Peter C GoslashtzscheEmail Address pcgcochranedkAffiliation Nordic Cochrane CentreRole DirectorComment The authors found that the results were missing from 5900 patients Furthermore there was extreme heterogeneity in theirmeta-analyses eg I square was 92 for the diclofenac trials which were the most common ones and there was extreme funnel plotasymmetry with the largest trials showing the smallest effects (the authors didnrsquot show funnel plots but I constructed one for diclofenac)Moreover the trials were industry funded of relatively poor quality and the authors analysed published data not data from clinicalstudy reports and did not try to obtain all the missing trials and data from the manufacturersThe authors concluded that topical NSAIDs are effective in providing pain relief but also cautioned that the large amounts of unpublisheddata ldquocould influence results in updates of this reviewrdquo They certainly could I believe it is plain wrong to perform meta-analyses on theauthorsrsquo data When I most recently reviewed this area for the BMJ in 2010 I concluded that we donrsquot know whether topical NSAIDsare beneficial (1)1 Goslashtzsche PC NSAIDs Clin Evid (Online) 2010 Jun 28I agree with the conflict of interest statement belowI certify that I have no affiliations with or involvement in any organization or entity with a financial interest in the subject matter ofmy feedback
Reply
Response submitted by authors Sheena Derry and Professor Andrew Moore on 4 November 2015Peter has made these same points elsewhere [PubMed Commons PubPeer gt Cochrane Database Syst Rev October 2015 httpspubpeercompublicationsA9E5BEA36549727357F9FD14CC2537]As a preliminary it is important to stress that the number of trials available in pain has increased over time and newer trials are typicallybetter and larger This means we can move from answering the simple question of whether an intervention works to more importantand relevant questions such as how well the intervention works and for drugs particularly to examine effects of dose and formulationFormulation is now recognised to have profound effects For oral analgesics for example fast acting formulations demonstrate up todouble the analgesic effect for a given dose as our recently updated overview points out (Cochrane Database Syst Rev 2015 Sep 289CD008659 see also Pain 2014 Jan155(1)14-21)What was true for oral analgesics in acute pain is now true also for topical NSAIDs in acute pain where manufacturers have beenputting a lot of effort into trying to produce new formulations that work better This updated review sought to examine the influenceof formulation and drug although dose is somewhat more difficult There are good reasons why formulation may be important(Rheumatic Disease Clinics 2008 Vol 34 Issue 2 p 415-432)
143Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Our searches did identify a large number of unpublished trials We deplore this but are powerless to change things For a previousreview of topical NSAIDs we contacted all identified manufacturers and asked for published or unpublished data The yield was small(as others have found) but some unpublished studies were brought into the public domain Waiting for all studies and clinical trialreports of all studies would take forever and could probably never be achieved We believe that many of these unpublished studiesrelate to drugs andor formulations that have never been manufactured commercially While these would be of interest in determiningwhat does and doesnrsquot work and to direct future research they would probably have little clinical relevance because these formulationsare unlikely to come to marketPeterrsquos main issue appears to be heterogeneity Tests for heterogeneity are problematical anyway (Pain 2000 85415-24) and the I squareof 92 that he quotes was for all topical diclofenac formulations combined While we do give an overall summary for diclofenac wedemonstrated in the review that the different diclofenac formulations produced different results from one another using LrsquoAbbeacute plotsand in our detailed analyses showing large variations in efficacy between them In the circumstance a high I square for all combined(clinical heterogeneity) is to be expected but it is not relevant The bulk of the studies on diclofenac were published in the last fiveyears were of decent quality and moderate to large size There were older data for ketoprofen but again these showed major differencesbetween formulations Differences between formulations are highlighted throughout the reviewTrying to determine publication bias using funnel plots or other measures is something of a lost cause (Journal of Clinical Epidemiology200053207-16) It is especially so with small numbers of trials (Journal of Clinical Epidemiology 2000 53 477-484) and makingsense of funnel plots is anything but easy for most people (Journal of Clinical Epidemiology 2005 58 894-901) A useless methodseems an odd choice to make to criticise our reviewThere are very good scientific reasons why drug and formulation may play a big part in the efficacy of topical NSAIDs This is alsothe case for oral analgesics used in acute pain where formulation improvements generating rapid absorption confers greater efficacySimple lumping strategies may have been permissible in past systematic review methodology but a more forensic approach is needednow and in the future This is what we have attempted to do in this latest reviewComparisons with Peterrsquos 2010 review seem inappropriate since that review was based on other reviews that are now out of date
Contributors
Feedback from Peter C GoslashtzscheAuthors involved with responding Andrew Moore Sheena DerryFeedback Editor Kate Seers
W H A T rsquo S N E W
Last assessed as up-to-date 4 February 2015
Date Event Description
4 November 2015 Feedback has been incorporated Feedback has been incorporated See Feedback 2
H I S T O R Y
Protocol first published Issue 4 2008
Review first published Issue 6 2010
144Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Date Event Description
13 October 2015 Amended Small errors found in Summary Table B for total num-ber of participants with piroxicam and indomethacinand in text for NNT with diclofenac (other gel) Nochange to conclusions
6 October 2015 Amended Small error found in Summary Table B percentagesfor NSAID and placebo were the wrong way round forcomparison of all NSAIDs and placebo RR and textwere correct No change to results or conclusions
13 February 2015 New citation required but conclusions have notchanged
Conclusions not changed Results remain essentiallythe same but the focus has changed to examine drugand formulation combination
3 February 2015 New search has been performed Title changed from Topical NSAIDs for acute pain inadults to Topical NSAIDs for acute musculoskeletalpain in adults to increase specificityNew searches run and new studies identified Fourteennew included studies using diclofenac (3489 new par-ticipants a 63 increase) four new excluded studiesFifteen studies awaiting classification (completed noresults available but likely to satisfy inclusion criteria)
23 May 2014 Amended Error in data reported for clinical success in Hosie 1993was brought to our attention and has been corrected
12 June 2012 Feedback has been incorporated We have incorporated feedback received from DrSigrun Chrubasik and the authorrsquos response on DMSOand other additives
C O N T R I B U T I O N S O F A U T H O R S
For the earlier review Tom Massey and SD identified studies and carried out data extraction analysis and drafting RAM and HJMwere involved in planning acted as adjudicators and were involved with writing
For this update SD and RAM carried out searches data extraction and analysis All authors were involved with writing the review
145Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
D E C L A R A T I O N S O F I N T E R E S T
SD has no conflicts relating to this review or any similar product
RAM has no conflicts relating to this review or any similar product
PW has no conflicts relating to this review or any similar product
HG has no conflicts relating to this review or any similar product
MM has no conflicts relating to this review or any similar product
S O U R C E S O F S U P P O R T
Internal sources
bull Oxford Pain Relief Trust UKGeneral institutional support
External sources
bull No sources of support supplied
D I F F E R E N C E S B E T W E E N P R O T O C O L A N D R E V I E W
For this update in 2015 we have changed the title to specify musculoskeletal pain because topical NSAIDs are not normally used totreat visceral pain or headache We felt that the new title better reflected the content of the review We have also changed the focus ofthe review from pooled analysis of all topical NSAIDs and all studies of a particular NSAID to an examination of individual drug andits formulation This makes the review much more relevant We have expanded the rsquoRisk of biasrsquo assessment and added a rsquoSummaryof findingsrsquo table and PRISMA flow chart We have removed a number of sensitivity analyses because they were not appropriate giventhe current information on the impact of formulation on efficacy The sensitivity analyses have been superseded by the rsquoRisk of biasrsquoassessment and taken into account in the rsquoSummary of findingsrsquo tables
An earlier review in 2004 chose to exclude studies using benzydamine on the grounds that it was no longer considered to be an NSAID(Mason 2004a) Although the protocol for this review stated that we would not include benzydamine after further consultation wenow believe that it should be classified as an NSAID albeit with a different mode of action which is not fully understood (Quane1998) Thus we have reinstated studies using topical benzydamine
I N D E X T E R M S
Medical Subject Headings (MeSH)
Acute Disease Administration Topical Anti-Inflammatory Agents Non-Steroidal [lowastadministration amp dosage adverse effects] AthleticInjuries [drug therapy] Pain [lowastdrug therapy] Randomized Controlled Trials as Topic Sprains and Strains [drug therapy]
146Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
MeSH check words
Adult Humans
147Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Serious adverse events 1 in total 0 in total Not calculated All data Low Small numbers of events
GRADE Working Group grades of evidenceHigh quality Further research is very unlikely to change our confidence in the estimate of effectModerate quality Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimateLow quality Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimateVery low quality We are very uncertain about the estimate
CI confidence interval RR risk ratio NNT number needed to treat for an additional beneficial outcome NNTp number needed to treatto prevent an event happening NNH number needed to treat for an additional harmful outcome
xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
5To
pica
lN
SA
IDs
for
acu
tem
uscu
losk
ele
talp
ain
inad
ults
(Revie
w)
Co
pyrig
ht
copy2015
Th
eC
och
ran
eC
olla
bo
ratio
nP
ub
lished
by
Joh
nW
iley
ampS
on
sL
td
B A C K G R O U N D
This review is an update of a review originally published in Issue 62010 on rsquoTopical NSAIDs for acute pain in adultsrsquo (Massey 2010)We have changed the title to specify musculoskeletal pain becausetopical NSAIDs are not normally used to treat visceral pain orheadache We felt that the new title better reflected the content ofthe review
The use of topical NSAIDs for pain relief has been a controversialsubject in analgesic practice In some parts of the world (includingmuch of Western Europe) they have been available for many yearsare widely available without prescription widely advertised usedextensively and evidence for their use is considered adequate Inother parts of the world they were regarded as little more thanplacebo with any apparent effect attributed to the process of rub-bing at the site of the affected area In some places (for examplethe US) their use was almost unknown until the mid-2010s InEngland 52 million prescriptions for topical NSAIDs were dis-pensed in 2013 (PACT 2014) mainly for formulations of ibupro-fen (245 million) piroxicam (118 million) and diclofenac (127million)
There is good evidence for the efficacy of topical NSAIDs in acuteand chronic musculoskeletal pain (Mason 2004a Mason 2004bMoore 1998a) In the US the Food and Drug Administration li-censed topical nonsteroidal products in 2007 and in England theNational Institute for Health and Care Excellence (NICE) recom-mended topical therapies as first line treatment in its guidelinesfor osteoarthritis in 2008 (NICE 2008) Earlier reviews of topicalanalgesics covered studies investigating the underlying science toexplain biological plausibility in addition to clinical trials (Anon2005 Moore 2008a)
This review is one of a series on topical analgesics including topicalcapsaicin at low and high doses (Derry 2012a Derry 2013) andtopical NSAIDs in chronic pain conditions (Derry 2012b) andsalicylate-containing rubefacients (Derry 2014)
Description of the condition
Acute pain is usually defined as pain of less than three monthsrsquo du-ration It is often associated with injury including trauma surgerymusculoskeletal injuries such as strains sprains and over-use in-juries or soft tissue injuries such as muscle soreness or cramps
Description of the intervention
Clinicians prescribe NSAIDs on a routine basis for a range of mildto moderate pain NSAIDs are the most commonly prescribedanalgesic medications worldwide and their efficacy for treatingacute pain has been well demonstrated (Moore 2003) They re-versibly inhibit cyclooxygenase (prostaglandin endoperoxide syn-thase) the enzyme mediating production of prostaglandins and
thromboxane A2 (FitzGerald 2001) Prostaglandins mediate a va-riety of physiological functions such as maintenance of the gastricmucosal barrier regulation of renal blood flow and regulation ofendothelial tone They also play an important role in inflamma-tory and nociceptive processes However relatively little is knownabout the mechanism of action of this class of compounds asidefrom their ability to inhibit cyclooxygenase-dependent prostanoidformation (Hawkey 1999)NSAIDs taken orally or intravenously are transported to all partsof the body in the blood and relatively high blood concentrationsare needed to achieve effective tissue concentrations at the site ofthe pain and inflammation These high concentrations throughoutthe body can give rise to a number of adverse events that canbe unpleasant (for example dyspepsia) or potentially serious (forexample gastrointestinal bleeding)A topical medication is a one applied to body surfaces such asthe skin or mucous membranes to treat ailments A large rangeof types of topical formulation may be used including but notlimited to creams foams gels lotions ointments and plastersThe exact formulation of a topical medication is often determinedby the speed of drug absorption required The need may be forslow absorption into the circulation to maintain low drug concen-trations and perhaps avoiding extensive first pass metabolism inthe liver plasters containing drug reservoirs may be used for thisas with transdermal opioids or contraceptive steroids For rapidabsorption the formulation is enhanced by substances to improveor assist skin penetration perhaps only to generate high concen-trations in tissues rather than in the blood gel formulations areuseful for this purpose which is why they are sometimes used fortopical NSAIDs
Topical NSAIDs
Topical NSAIDs are formulated for direct application to thepainful site and to produce a local pain-relieving effect whileavoiding body-wide distribution of the drug at physiologically ac-tive levels (McPherson 2013) This method of application (dos-ing) necessarily limits their use to more superficial painful condi-tions such as sprains strains and muscle or tendon soreness Theywould not for example be indicated for deep visceral pain orheadaches They are also not appropriate for use on broken skinso would not be used on open wounds (accidental or surgical)
How the intervention might work
For a topical formulation to be effective it must first penetratethe skin Only when the drug has entered the lower layers ofthe skin can it be absorbed by the blood and transported to thesite of action or penetrate deeper into areas where inflammationoccurs Individual drugs have different degrees of penetration Abalance between lipid and aqueous solubility is needed to optimisepenetration and use of prodrug esters has been suggested as a way
6Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
of enhancing permeability Formulation is also crucial to goodskin penetration Experiments with artificial membranes or humanepidermis suggest that creams are generally less effective than gelsor sprays but newer formulations such as microemulsions mayhave greater potentialOnce the drug has reached the site of action it must be presentat a sufficiently high concentration to inhibit cyclooxygenase en-zymes thereby reducing prostaglandin synthesis This in turn re-duces inflammation and relieves pain It is probable that in acuteconditions topical NSAIDs exert their action primarily by lo-cal reduction of symptoms independent of any systemic uptakeand delivery Tissue levels of NSAIDs applied topically certainlyreach levels high enough to inhibit cyclooxygenase-2 (Anon 2005Haroutiunian 2010 Moore 2008a) However plasma concentra-tions found after topical administration are only a fraction (usu-ally much less than 5) of the levels found in plasma followingoral administration Topical application can potentially limit sys-temic adverse events by increasing local effects and minimisingsystemic concentrations of the drug We know that the incidenceof upper gastrointestinal bleeding is low with chronic use of topi-cal NSAIDs (Evans 1995) although it has been reported partic-ularly in people with risk factors (Zimmerman 1995) We haveno certain knowledge of lower effects on cardiovascular events orrenal failure both of which have been associated with oral NSAIDuse
Why it is important to do this review
Since the last review in 2010 a number of new studies have beenpublished nearly all of which investigated various formulations ofdiclofenac These new studies are generally of higher quality thanmany of the earlier ones in this review and have the potential toinfluence the strength of its conclusions substantially Moreoverthe additional information allows for analysis based not only on aparticular drug but also on the formulation of that drug This canprovide better insight into whether formulation affects efficacy oftopical NSAIDs in acute musculoskeletal painAn updated review of evidence for topical NSAIDs was needed toinform choices made by consumers prescribers and commission-ers (purchasers of healthcare)
O B J E C T I V E S
To determine the efficacy and safety of topically applied NSAIDsin acute musculoskeletal pain in adults
M E T H O D S
Criteria for considering studies for this review
Types of studies
Randomised controlled double-blind studies comparing topicalNSAIDs with placebo (inert carrier) or other active treatment foracute pain with at least 10 participants per treatment arm andoutcomes close to seven days (minimum three days) We excludedstudies published only as short abstracts (which report insufficientdata to assess methods) or studying experimentally induced pain(which does not correlate well with clinical pain) Because a cross-over design is not appropriate for self limiting conditions such assprains strains and contusions we only considered parallel-groupdesigns
Types of participants
Adults (aged 16 years or more) with acute musculoskeletal pain ofat least moderate intensity resulting mainly from strains sprainsor sports injuries Typically for sports injuries the injury wouldhave occurred within 24 or 48 hours
Types of interventions
Included studies had at least one treatment arm using a topicalNSAID and a comparator arm using placebo (inert carrier withoutNSAID or other active treatment) The topical NSAID had to beapplied at least once daily We did not include salicylates in thisreview as they are no longer classified as topical NSAIDs and arecovered in a separate review (Derry 2014)
Types of outcome measures
We sought information on participant characteristics includingage sex and condition treated
Primary outcomes
The primary outcome was rsquoclinical successrsquo defined as at least a50 reduction in pain or equivalent measure such as a rsquovery goodrsquoor rsquoexcellentrsquo global assessment of treatment or rsquononersquo or rsquoslightrsquopain on rest or movement measured on a categorical scale (Moore1998a) We used the following hierarchy of outcomes to extractdata for the primary outcome
bull Participant reported reduction in pain of at least 50bull Participant reported global assessment of treatmentbull Pain on movementbull Pain at rest or spontaneous painbull Undefined rsquoimprovementrsquo
We used only participant reported outcomes of efficacy and notphysician or investigator reported outcomes
7Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Secondary outcomes
bull Numbers of participants with adverse events local andsystemic
bull Numbers of withdrawals all cause lack of efficacy andadverse events
We anticipated that outcomes would be reported after differentdurations of treatment and extracted data reported as close toseven days as possible with a minimum of three days We also ex-tracted data for outcomes reported after longer durations of treat-ment We anticipated that reporting of adverse events would varybetween studies with regard to the terminology used method ofascertainment and categories reported (for example occurring inat least 5 of participants or where there is a statistically signifi-cant difference between treatment groups) We took care to iden-tify these details where relevant
Search methods for identification of studies
Electronic searches
We searched the following databases without language restrictionbull CENTRAL (The Cochrane Library) Issue 4 2009 for the
original review and the Cochrane Register of Studies Online(CRSO) to 3 February 2015 for this update
bull MEDLINE (via Ovid) from inception to December 2009for the original review and from 2008 to 3 February 2015 forthis update
bull EMBASE (via Ovid) from inception to December 2009for the original review and from 2008 to 3 February 2015 forthis update
bull Oxford Pain Relief Database for the original review (Jadad1996a) This resource is no longer being updated
See Appendix 1 for the CENTRAL search strategy Appendix 2 forthe MEDLINE search strategy and Appendix 3 for the EMBASEsearch strategy
Searching other resources
We searched the reference lists of review articles and includedstudies We have previously asked manufacturers for details ofunpublished studies but did not make new requestsWe searched two clinical trial registries (clinicaltrialsgov (clinicaltrialsgov) and the World Health Organization In-ternational Clinical Trials Registry Platform (appswhointtrialsearch)) and asked personal contacts about ongoing and un-published studies
Data collection and analysis
Selection of studies
We screened the titles and abstracts of studies identified by thesearches to eliminate those that clearly did not satisfy the inclu-sion criteria and obtained full reports of the remaining studies todetermine inclusion in the review
Data extraction and management
Review authors were not blinded to the authorsrsquo names and in-stitutions journal of publication or study results at any stage ofthe review Two review authors independently selected the stud-ies for inclusion assessed methodological quality and risk of biasand extracted data We resolved disagreements and uncertaintiesthrough discussionWe abstracted information on participants interventions and out-comes from the original reports into a standard data extractionform One review author entered data suitable for meta-analysisinto Review Manager 5 (RevMan 2014) and another review au-thor checked it
Assessment of risk of bias in included studies
We used the Oxford Quality Score as the basis for inclusion lim-iting inclusion to studies that were randomised and double-blindas a minimum (Jadad 1996b)Two review authors independently assessed risk of bias for eachstudy using the criteria outlined in the Cochrane Handbook forSystematic Reviews of Interventions with any disagreements resolvedby discussion (Higgins 2011) We assessed the following for eachstudy
bull Random sequence generation (checking for possibleselection bias) We assessed the method used to generate theallocation sequence as low risk of bias (any truly randomprocess for example random number table computer randomnumber generator) unclear risk of bias (method used to generatesequence was not clearly stated) We excluded studies using anon-random process that were therefore at high risk of bias (forexample odd or even date of birth hospital or clinic recordnumber)
bull Allocation concealment (checking for possible selectionbias) The method used to conceal allocation to interventionsbefore assignment determines whether intervention allocationcould have been foreseen in advance of or during recruitmentor changed after assignment We assessed the methods as lowrisk of bias (for example telephone or central randomisationconsecutively numbered sealed opaque envelopes) unclear riskof bias (method was not clearly stated) We excluded studies thatdid not conceal allocation and were therefore at high risk of bias(for example open list)
8Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
bull Blinding of outcome assessment (checking for possibledetection bias) We assessed the methods used to blind studyparticipants and outcome assessors from knowledge of whichintervention a participant received We assessed the methods aslow risk of bias (study stated that it was blinded and describedthe method used to achieve blinding for example identicaltubes containing gel or identical plasters matched in appearanceand smell) unclear risk of bias (study stated that it was blindedbut did not provide an adequate description of how blinding wasachieved) We excluded studies that were not double-blind andwere therefore at high risk of bias
bull Incomplete outcome data (checking for possible attritionbias due to the amount nature and handling of incompleteoutcome data) We assessed the methods used to deal withincomplete data as low risk of bias (less than 10 ofparticipants did not complete the study or used rsquobaselineobservation carried forwardrsquo analysis or both) unclear risk ofbias (used rsquolast observation carried forwardrsquo analysis) or high riskof bias (used rsquocompleterrsquo analysis)
bull Size (checking for possible biases confounded by small size)Small studies have been shown to overestimate treatment effectsprobably due to methodological weaknesses (Dechartres 2013Nuumlesch 2010) We assessed studies as at low risk of bias if theyhad at least 200 participants at unclear risk if they had 50 to 200participants and at high risk if they had fewer than 50participants
Measures of treatment effect
We used risk ratio (RR) to establish statistical difference and num-bers needed to treat for an additional beneficial outcome (NNT)with 95 confidence intervals (CI) We pooled percentages as ab-solute measures of benefit or harmWe used the following terms to describe adverse outcomes in termsof harm or prevention of harm
bull When significantly fewer adverse outcomes occurred withtreatment than with control (placebo or active) we used the termthenumber needed to treat to prevent one event (NNTp)
bull When significantly more adverse outcomes occurred withtreatment compared with control (placebo or active) we usedthe term the number needed to treat for an additional harmfuloutcome or cause one event (NNH)
Unit of analysis issues
Randomisation was to the individual participant
Dealing with missing data
Wherever possible we used intention-to-treat (ITT) analysis wherethe ITT population consists of participants who were randomisedapplied at least one dose of the assigned study medication andprovided at least one post-baseline assessment We assigned miss-ing participants zero improvement
We also looked for information about methods of imputation formissing data
Assessment of heterogeneity
We examined heterogeneity visually using LrsquoAbbeacute plots (LrsquoAbbeacute1987) a visual method for assessing differences in results of indi-vidual studies and with the I2 statistic
Assessment of reporting biases
The aim of this review was to use dichotomous outcomes of knownutility and of value to patients (Moore 2013) The review did notdepend on what the authors of the original studies chose to reportor not Studies that did not report dichotomous results but onlyaverage pain data did not contribute to analysesWe assessed publication bias using a method designed to detect theamount of unpublished data with a null effect required to makeany result clinically irrelevant (usually taken to mean an NNT of10 or higher Moore 2008b)
Data synthesis
We pooled data only for comparisons and outcomes where therewere at least two studies and 200 participants (Moore 1998b)When two active treatment arms were compared with a placeboarm we took care to avoid double counting of participants in theplacebo arm if both active groups contributed to an analysis wesplit the placebo group between themWe calculated RRs with 95 CIs using the fixed-effect model(Morris 1995) A statistically significant benefit of topical NSAIDover control was assumed when the lower limit of the 95 CI ofthe RR was greater than one A statistically significant benefit ofcontrol over active treatment was assumed when the upper limitof the 95 CI was less than one We calculated NNTs with 95CIs using the pooled number of events by the method of Cookand Sackett (Cook 1995)Statistically significant differences between NNTs for differenttopical NSAIDs were tested using the z test (Tramer 1997) wherethere were sufficient data to do so and where the studies weresufficiently similar in types of participant outcome and durationto make such comparisons sensible
Subgroup analysis and investigation of heterogeneity
We carried out separate analyses for individual NSAIDs andwhere the data permitted for different formulations of individualNSAIDs
Sensitivity analysis
The earlier review included sensitivity analyses for various factorsthat are now covered by the assessment of risk of bias
9Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
R E S U L T S
Description of studies
Results of the search
New searches for this update identified 20 publications that wereexamined in further detail to determine inclusion status We alsoidentified 20 additional studies in clinical trial registries See Figure1
10Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Figure 1 Study flow diagram
11Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
We identified 14 new studies (11 publications three unpub-lished reports) satisfying our inclusion criteria (Costantino2011 Coudreuse 2010 Gonzaacutelez de Vega 2013 Hofman2000 Klainguti 2010 Kuehl 2011 Li 2013 NCT01255423NCT01272934 NCT01272947 Predel 2012 Predel 2013aPredel 2013b Saillant 1998) one post hoc analysis of a study thatwas included in the earlier review (Mueller 2010 in Predel 2004)and an additional publication and a pooled analysis that includednew data for another study from the earlier review (Lionberger2011 pooled analysis in Joussellin 2003)
Included studies
All except one of the new studies compared diclofenac withplacebo A number of different formulations were used includ-ing diclofenac epolamine (DHEP) with or without heparin (Flec-toparin Tissugel or Flector EP Tissugel) applied as a plaster (orpatch) diclofenac diethylamine (DDEA) applied as a gel and di-clofenac with lethicin applied as a spray gel The remaining studycompared diclofenac gel with traumeel a ldquofixed combination ofplant and mineral extractsrdquo applied as a gel or an ointmentThere were 47 studies in the original review 14 new studies wereincluded making a total of 61 studies in this updated reviewAll used a parallel group design Forty-four compared a topicalNSAID with placebo 13 a topical NSAID with an active com-parator (a different topical NSAID an oral NSAID the same top-ical NSAID in a different formulation or a compound of plantand mineral extracts) and four had both placebo and active com-parators In total 5311 participants were treated with a topicalNSAID 3470 with placebo and 220 with an oral NSAID Top-ical NSAIDs used were benzydamine diclofenac etofenamatefelbinac fentiazac flunoxaprophen flurbiprofen ibuprofen in-domethacin ketoprofen ketorolac lysine clonixinate meclofe-namic acid naproxen niflumic acid and piroxicam They wereapplied as creams gels sprays foams or plasters (patches) Topicalplacebos were the inert carriers without the active NSAID OralNSAIDs used were ibuprofen (as tablets) and indomethacin (ascapsules)Most studies enrolled participants who had sprains strains andcontusions usually as a result of sports injuries and treatmentwas started within a few hours or days Other studies enrolledparticipants with overuse-type injuries such as tendinitis and acutelow back pain where pain had been present for days or weeks butless than three monthsParticipants were treated for at least five days and up to threeweeks with most studies lasting seven to 14 days Participantswere usually assessed in clinic at intervals during treatment andsometimes also at home using daily patient diaries We used out-comes closest to seven days because many of these injuries are self
limiting with differences between active treatment and placebobeing diminished or lost after longer intervalsMost studies reported dichotomous outcomes suitable for a re-sponder analysis although group mean change (for pain or phys-ical function for example) was usually the primary outcomesHowever the definition of response varied both in the parametermeasured (for example pain pain on movement patient globalevaluation of treatment) and in the scale used to measure it (forexample a 3- 4- or 5-point scale for patient global evaluation)Details of included studies are in the Characteristics of includedstudies tableWe identified 14 completed but apparently unpublished studies ina clinical trial registry for which no results have been posted (4403participants NCT00351104 NCT00352625 NCT00426985NCT00640705 NCT00640939 NCT00680472NCT00680784 NCT00765700 NCT00869063NCT00869180 NCT00931866 NCT01874626NCT01957215 NCT02324270) We have placed these underCharacteristics of studies awaiting classification We also identi-fied three conference abstracts that relate to completed studies thatdo not appear to have been published but that may satisfy ourinclusion criteria One is likely to be the same study as one of theincluded studies identified in a clinical trial registry (Pallay 2013in NCT01272947) one relates to another study identified in theclinical trial registry that is awaiting classification (Ekman 2010in NCT00765700) while we could find no published or unpub-lished reports of the other Sarzi-Puttini 2014)We also identified three ongoing studies with an estimated en-rolment of 880 participants (NCT01945034 NCT02100670NCT02290821) Details are in the Characteristics of ongoingstudies table
Excluded studies
For the original review 25 studies were excluded after readingthe full paper For this update we excluded four new studies (Cesarone 2008 Coulibaly 2009 Kuwabara 2013 Vinciguerra2008) Details are in the Characteristics of excluded studies table
Risk of bias in included studies
All studies were randomised and double-blind One study scored25 (Sinniger 1981) 23 scored 35 23 scored 45 and 14 scored55 for methodological quality using the Oxford Quality Scale Abreakdown of the scores for individual studies is reported in theCharacteristics of included studies tableComments on potential biases in individual studies are reported inthe rsquoRisk of biasrsquo section of the Characteristics of included studiestable The findings are displayed in Figure 2 and Figure 3
12Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Figure 2 Risk of bias summary review authorsrsquo judgements about each risk of bias item for each includedstudy
13Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Figure 3 Methodological quality graph review authorsrsquo judgements about each methodological quality
item presented as percentages across all included studies
Allocation
All the studies were randomised but only 17 adequately describedthe method used to generate the random sequence Thirteen stud-ies adequately described the method used to conceal allocation ofthe sequence No studies were at high risk of bias for this item
Blinding
All studies were double-blind and 38 adequately described themethod used to maintain the blinding No studies were at highrisk of bias for this item
Incomplete outcome data
Thirty-six studies included all participants in the primary analysisor provided sufficient data to allow missing participants to be in-cluded as non-responders and were judged at low risk of bias Wejudged four studies to be at high risk of attrition bias (Campbell1994 Kuehl 2011 Maziegraveres 2005a Russell 1991) Three unpub-lished studies contributed only to adverse event analyses and ac-counted for all participants for these outcomes (NCT01255423NCT01272934 NCT01272947)
Other potential sources of bias
We judged one study that included more than 200 participantsin each treatment arm to be at low risk of bias from size butthis study had a very high attrition rate (see rsquoIncomplete outcome
data (attrition bias)rsquo) and did not report all the efficacy outcomesmeasured (Kuehl 2011) We judged 27 studies to be at high riskbecause they included fewer than 50 participants per treatmentarm
Effects of interventions
See Summary of findings for the main comparison TopicalNSAIDs compared with topical placebo for acute musculoskeletalpain in adultsThree studies did not contribute data suitable for analysis of at leastone outcome (Gonzaacutelez de Vega 2013 Gualdi 1987 Hoffmann2012)
1 Topical NSAID versus placebo
Details of efficacy outcomes in individual studies are in Appendix4 and of adverse events and withdrawals in Appendix 5 Appendix6 has details of the concentration of topical products the amountapplied the frequency of application and an estimation of thedaily dose of topical NSAID applied Not all studies providedsufficient information to allow calculation of daily dose appliedFor example for topical diclofenac the estimated doses appliedvaried between about 60 and 280 mg for topical ketoprofen 100to about 450 mg for topical ibuprofen 300 to 800 mg
Participants with clinical success
14Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Topical diclofenac versus placebo
Ten studies contributed to this analysis (Coudreuse 2010Joussellin 2003 Klainguti 2010 Li 2013 Predel 2004 Predel2012 Predel 2013a Predel 2013b Rowbotham 2003) of whichone (Predel 2012) had two active treatment arms A total of 1074participants were treated with topical diclofenac and 976 withplacebo (Analysis 11 Figure 4)
15Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Figure 4 Forest plot of comparison 2 Individual NSAID versus placebo outcome 21 Clinical success
16Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
bull The proportion of participants experiencing successfultreatment with topical diclofenac was 74 (8001074 range39 to 100)
bull The proportion of participants experiencing successfultreatment with placebo was 47 (461976 range 8 to 82)
bull The RR for treatment compared with placebo was 16(95 CI 15 to 17)
bull The NNT for successful treatment was 37 (32 to 43) Forevery four participants treated with topical diclofenac one wouldexperience successful treatment who would not have done sowith placebo
Effect of formulation
The effects of formulation are shown in Analysis 21 and Figure5
17Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Figure 5 LrsquoAbbeacute plot of clinical success in studies of topical diclofenac versus topical placebo The size ofthe symbol is proportional to the size of the study (inset scale) Dark blue Emulgel light blue spraygel red
Flector pink other patch or plaster
bull Four studies used a Flectorreg plaster (1030 participantsJoussellin 2003 Li 2013 Rowbotham 2003 Saillant 1998) TheRR for treatment compared with placebo was 15 (14 to 17)and the NNT was 47 (37 to 65)
bull Three studies used other makes of plaster (474 participantsCoudreuse 2010 Klainguti 2010 Predel 2004) The RR fortreatment compared with placebo was 16 (14 to 18) and theNNT was 32 (26 to 42)
bull Two studies used Voltaren Emulgel (314 participants Predel2012 Predel 2013b) The RR for treatment compared withplacebo was 38 (27 to 55) and the NNT was 18 (15 to 21)
bull One study used a spray gel (232 participants Predel2013a) The RR for treatment compared with placebo was 12(105 to 13) and the NNT was 80 (48 to 24)
Diclofenac as the gel formulation Emulgel was statistically more ef-ficacious than the plaster formulation as Flector plaster (z = 6360P value lt 000001)
Topical ibuprofen versus placebo
Five studies contributed to this analysis (Billigmann 1996Campbell 1994 Dreiser 1988 Machen 2002 Ramesh 1983) Atotal of 218 participants were treated with topical ibuprofen and218 with placebo (Analysis 11)
bull The proportion of participants experiencing successfultreatment with topical ibuprofen was 55 (120218 range 31to 81)
18Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
bull The proportion of participants experiencing successfultreatment with placebo was 33 (73218 range 13 to 76)
bull The RR of treatment compared with placebo was 16 (13to 20)
bull The NNT for successful treatment was 46 (33 to 80) Forevery five participants treated with topical ibuprofen one wouldexperience successful treatment who would not have done sowith placebo
Effect of formulation
The effects of formulation are shown in Analysis 31bull Three studies used cream formulations (195 participants
Campbell 1994 Dreiser 1988 Ramesh 1983) Although this isjust below our threshold for pooled analysis we have includedthis analysis for completeness and the results should beinterpreted with caution The RR for treatment compared withplacebo was 13 (103 to 16) and the NNT was 64 (34 to 41)
bull Two studies used gel formulations (241 participantsBilligmann 1996 Machen 2002) The RR for treatmentcompared with placebo was 27 (17 to 42) and the NNT was39 (27 to 67)
There was no statistically significant difference between the geland cream formulations (z = 1160 P value = 0246)
Topical ketoprofen versus placebo
Seven studies contributed to this analysis (Airaksinen 1993Dreiser 1989 Julien 1989 Kockelbergh 1985 Maziegraveres 2005bMaziegraveres 2005a Noret 1987) A total of 346 participants weretreated with topical ketoprofen and 337 with placebo (Analysis11)
bull The proportion of participants experiencing successfultreatment with topical ketoprofen was 73 (251346 range57 to 89)
bull The proportion of participants experiencing successfultreatment with placebo was 47 (157337 range 17 to 73)
bull The RR of treatment compared with placebo was 16 (14to 18)
bull The NNT for successful treatment was 39 (30 to 53) Forevery four participants treated with topical ketoprofen onewould experience successful treatment who would not have doneso with placebo
Effect of formulation
The effects of formulation are shown in Analysis 41 and Figure6
19Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Figure 6 LrsquoAbbeacute plot of clinical success in studies of topical ketoprofen versus topical placebo The size ofthe symbol is proportional to the size of the study (inset scale) Light blue ketoprofen gel pink ketoprofen
plaster
Two studies used a plaster formulation (335 participants Maziegraveres2005b Maziegraveres 2005a) The RR for treatment compared withplacebo was 12 (104 to 14) and the NNT was 82 (45 to 47)Five studies used gel formulations (348 participants Airaksinen1993 Dreiser 1989 Julien 1989 Kockelbergh 1985 Noret 1987)The RR for treatment compared with placebo was 22 (17 to 28)and the NNT was 25 (20 to 34)Ketoprofen as a gel formulation was statistically more efficaciousthan a plaster formulation (z = 3860 P value = 000014)
Topical piroxicam versus placebo
Three studies contributed to this analysis (Aoki 1984 Fujimaki1985 Russell 1991) A total of 255 participants were treated withtopical piroxicam and 249 with placebo (Analysis 11)
bull The proportion of participants experiencing successfultreatment with topical piroxicam was 68 (179255 range 53to 79)
bull The proportion of participants experiencing successfultreatment with placebo was 47 (118249 range 45 to 49)
bull The RR of treatment compared with placebo was 15 (13to 17)
bull The NNT for successful treatment was 44 (32 to 69) Forevery four participants treated with topical piroxicam one wouldexperience successful treatment who would not have done sowith placebo
Topical indomethacin versus placebo
20Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Three studies contributed to this analysis ( kermark 1990 Aoki1984 Fujimaki 1985) A total of 168 participants were treatedwith topical indomethacin and 173 with placebo (Analysis 11)
bull The proportion of participants experiencing successfultreatment with topical indomethacin was 58 (97168 range54 to 64)
bull The proportion of participants experiencing successfultreatment with placebo was 46 (79173 range 25 to 49)
bull The RR of treatment compared with placebo was 13 (103to 16)
bull The NNT for successful treatment was 83 (44 to 65) Forevery eight participants treated with topical indomethacin onewould experience successful treatment who would not have doneso with placebo
Topical benzydamine versus placebo
Three studies contributed to this analysis (Chatterjee 1977 Haig1986 Linde 1985) A total of 96 participants were treated withtopical benzydamine and 97 with placebo (Analysis 11) Al-though this is just below our threshold for pooled analysis we haveincluded this analysis for completeness and the results should beinterpreted with caution
bull The proportion of participants experiencing successfultreatment with topical benzydamine was 77 (7496 range70 to 86)
bull The proportion of participants experiencing successfultreatment with placebo was 67 (6597 range 48 to 80)
bull The RR of treatment compared with placebo was 12 (096to 14) There was no statistically significant difference betweentreatments (Figure 4)
Results for participants with clinical success with individual topi-cal NSAIDs where there were adequate data for analysis are sum-marised below in rsquoSummary of results Arsquo and Analysis 11 Analysis21 Analysis 31 and Analysis 41
Summary of results A Participants with clinical success
Comparison Studies Participants NSAID()
Placebo () Relative benefit (95CI)
NNT (95 CI)
Diclofenac -Flector plaster
4 1030 63 41 15 (14 to 17) 47 (37 to 65)
Diclofenac -other plaster
3 474 88 57 16 (14 to 18) 32 (26 to 42)
Diclofenac -Emulgel
2 314 78 20 38 (27 to 55) 18 (15 to 21)
Diclofenac -other gel
1 232 94 82 12 (11 to 13) 80 (48 to 24)
Ibuprofen -cream
3 195 71 56 13 (103 to 16) 64 (34 to 41)
Ibuprofen - gel 2 241 42 16 27 (17 to 42) 39 (27 to 67)
Ketoprofen -plaster
2 335 73 60 12 (104 to 14) 82 (45 to 47)
Ketoprofen - gel 5 348 72 33 22 (17 to 28) 25 (20 to 34)
Piroxicam 3 504 70 47 15 (13 to 17) 44 (32 to 69)
Indomethacin 3 341 58 46 13 (103 to 16) 83 (44 to 65)
21Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Benzydamine 3 193 77 67 12 (096 to 14) not calculated
Results for these two comparisons are derived from very smallamounts of data and are provided here for completeness Theyshould be interpreted with caution
Local adverse events
Local adverse events were irritation of the area to which the topicalNSAID was applied including redness or erythema and itch orpruritus Where reported these were usually described as mild andtransient
All topical NSAIDs versus placebo
Forty-two studies contributed to this analysis of which three com-pared two different drugs with placebo (Aoki 1984 Diebshlag1990 Fujimaki 1985) Three studies had two treatment armscomparing different formulations or application regimens for di-clofenac with placebo which have been combined for this analysis
(Costantino 2011 Klainguti 2010 Predel 2012) In total 3619participants were treated with topical NSAIDs and 3121 withplacebo (Analysis 51)
bull The proportion of participants experiencing a local adverseevent with a topical NSAID was 43 (1553619 range 0 to33)
bull The proportion of participants experiencing a local adverseevent with placebo was 46 (1453121 range 0 to 32)
bull The RR of topical NSAID compared with placebo was 098(080 to 12)
bull There was no significant difference between treatmentgroups so the NNH was not calculated
Individual topical NSAIDs versus placebo
Results for local adverse events with individual topical NSAIDswhere there were adequate data for analysis are in Summary ofresults B and Analysis 12
Summary of results B Participants with local adverse events
Comparison Studies Participants NSAID()
Placebo()
RR(95 CI)
NNH(95 CI)
All NSAIDs 42 6740 43 46 098 (080 to 12) Not calculated
Diclofenac 15 3271 31 43 078 (056 to 11) Not calculated
Ketoprofen 8 852 11 95 12 (083 to 17) Not calculated
Piroxicam 3 522 23 54 042 (017 to 11) Not calculated
Felbinac 3 397 30 15 19 (049 to 75) Not calculated
Indomethacin 3 354 63 22 27 (091 to 77) Not calculated
Ibuprofen 3 321 10 43 23 (098 to 54) Not calculated
Systemic adverse events
All topical NSAIDs versus placebo
Thirty-six studies contributed data on systemic adverse eventsof which three compared two different drugs with placebo (Aoki1984 Diebshlag 1990 Fujimaki 1985) Two studies had two treat-
22Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
ment arms comparing different formulations or application regi-mens for diclofenac with placebo which have been combined forthis analysis (Klainguti 2010 Predel 2012) In total 2956 partic-ipants were treated with a topical NSAID and 2620 with placebo(Analysis 52)
bull Twenty-three studies reported no systemic adverse events inany arm of the study
bull The proportion of participants experiencing a systemicadverse event with a topical NSAID was 31 (922956)
bull The proportion of participants experiencing a systemicadverse event with placebo was 35 (912620)
bull The RR of topical NSAID compared with placebo was 096(073 to 13)
bull There was no significant difference between treatmentgroups so the NNH was not calculated
A further six studies did not report the occurrence or oth-erwise of systemic adverse events (Billigmann 1996 Julien1989 Kockelbergh 1985 Noret 1987 Ramesh 1983 Vecchiet1991) while two studies did not report numbers of participantswith systemic adverse events ( kermark 1990 Auclair 1989)Costantino 2011 reported that there were no systemic gastroin-testinal adverse events Two studies reported only on total adverseevents without distinguishing between local and systemic events(NCT01255423 NCT01272934)
Serious adverse events
Two studies reported serious adverse events In Hoffmann 2012one participant experienced three serious adverse events none ofwhich was judged to be related to the study medication (diclofenacplaster) In NCT01272934 one participant using diclofenac gelruptured the ligaments of the wrist There was no statement aboutlikely relationship to the study medication but this seems unlikely
Adverse event withdrawals
Forty-two studies reported data relating to adverse event with-drawals of which three compared two different drugs with placebo(Aoki 1984 Diebshlag 1990 Fujimaki 1985) Two studies hadtwo treatment arms comparing different formulations or applica-tion regimens for diclofenac with placebo which have been com-bined for this analysis (Klainguti 2010 Predel 2012) In total3365 participants received a topical NSAID and 3040 placebo(Analysis 53)
bull Forty-four comparisons reported no adverse eventwithdrawals
bull The proportion of participants withdrawing from the studydue to an adverse event after treatment with a topical NSAIDwas 098 (333365)
bull The proportion of participants withdrawing from the studydue to an adverse event after treatment with placebo was 099(303040)
bull The RR of topical NSAID compared to placebo was 10(064 to 16)
bull There was no significant difference between treatmentgroups so the NNH was not calculated
Four studies did not specifically mention adverse event with-drawals (Haig 1986 Hoffmann 2012 Klainguti 2010 Vecchiet1991) while one reported that one participant withdrew with mildpruritus but did not state the treatment arm (Joussellin 2003)Ten studies specifically reported withdrawals due to lack of efficacy(Dreiser 1989 Dreiser 1994 Kuehl 2011 Machen 2002 Maziegraveres2005b Maziegraveres 2005a Noret 1987 Predel 2013a Russell 1991Thorling 1990) (Appendix 5) Numbers of participants withdraw-ing were generally low with rates of 6 or less except in Kuehl2011 where the rate was 10 with active treatment (diclofenacplaster) and 12 with placebo We did not carry out any analysisbecause the outcome was inconsistently reported
2 Topical NSAID versus active comparator
Details of efficacy outcomes in individual studies are in Appendix4 and of adverse events and withdrawals in Appendix 5
Participants with clinical success
Topical NSAID versus oral NSAID
bull kermark 1990 compared indomethacin spray withindomethacin capsules with response rates of 55 (1222) withspray and 23 (522) with capsules
bull Hosie 1993 compared felbinac foam with ibuprofen tabletswith response rates of 64 (81127) with felbinac foam and72 (96133) with ibuprofen tablets
bull Whitefield 2002 compared ibuprofen gel with ibuprofentablets with response rates of 60 (3050) with gel and 54(3650) with tablets
There were insufficient data for meta-analysis for any one of thesecomparisons felbinac is not known to be better than placebo
Topical NSAID versus different formulation of the same
topical NSAID
bull Fioravanti 1999 compared DHEP (diclofenac) gelformulated with and without lecithin with response rates of70 (3550) in both treatment arms
bull Mahler 2003 compared DHEP (diclofenac) gel formulatedwith and without lecithin with response rates of 89 (8292)with lecithin and 70 (6288) without lecithin
bull Gallacchi 1990 compared topical diclofenac formulated asFlectorreg gel and Emugelreg with response rates of 76 (1925)in both treatment arms
23Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
bull Governali 1995 compared topical ketoprofen cream withgel with response rates of 93 (1415) with cream and 27 (415) with gel
There were insufficient data for analysis
Topical NSAID versus different topical NSAID
Eight studies compared one topical NSAID versus at least oneother topical NSAID piroxicam versus indomethacin (Aoki1984 Fujimaki 1985 Sugioka 1984) ibuprofen versus ketopro-fen (Curioni 1985 Picchio 1981) ketoprofen versus etofena-mate (Curioni 1985 Tonutti 1994) ibuprofen versus etofenamate(Curioni 1985) ketorolac versus etofenamate (Diebshlag 1990)and diclofenac versus lysine clonixinate (Hofman 2000)There were sufficient data to compare only piroxicam with in-domethacin (Aoki 1984 Fujimaki 1985 Sugioka 1984 Analysis61)
bull The proportion of participants experiencing clinical successwith topical piroxicam was 56 (185330 range 49 to 78)
bull The proportion of participants experiencing clinical successwith topical indomethacin was 45 (140311 range 33 to64)
bull The RR of piroxicam compared with indomethacin was 12(11 to 14)
bull The NNT for successful treatment was 91 (53 to 30) Forevery nine participants treated with topical piroxicam onewould experience a clinical success who would not haveexperienced one with topical indomethacin
Topical NSAID versus different topical intervention
One study compared diclofenac gel with a herbal product calledTraumeel gel under double-blind conditions with response ratesfor being pain-free at seven days of 8137 (58) with diclofenacgel and 7140 (50) with Traumeel gel (Gonzaacutelez de Vega 2013)
Local adverse events
Topical NSAID versus oral NSAID
Two studies comparing a topical NSAID with an oral NSAIDprovided data on local adverse events ( kermark 1990 Hosie1993) There were five events with topical NSAID and three withoral NSAID which were too few for analysis
Topical NSAID versus different topical NSAID
All nine studies comparing one topical NSAID with at least oneother reported on local adverse events with a total of 48 events in1005 participants (48) There were sufficient data to compare
only piroxicam with indomethacin (Aoki 1984 Fujimaki 1985Sugioka 1984 Analysis 62)
bull The proportion of participants experiencing local adverseevents with topical piroxicam was 21 (7340 range 12 to28)
bull The proportion of participants experiencing local adverseevents with topical indomethacin was 10 (33331 range 29to 15)
bull The RR of piroxicam compared with indomethacin was021 (009 to 047)
bull The NNT to prevent a local adverse event was 13 (87 to23) For every thirteen participants treated with topicalpiroxicam one would not experience a local adverse event whowould have experienced one with topical indomethacin
Topical NSAID versus different topical intervention
Gonzaacutelez de Vega 2013 reported that adverse events were infre-quent and mild to moderate in intensity but did not distinguishbetween local and systemic events Numbers of participants expe-riencing any adverse event were 8147 with diclofenac gel and 14148 with Traumeel gel
Systemic adverse events
kermark 1990 reported numbers of events rather than numbersof participants with events while Tonutti 1994 and Whitefield2002 reported no adverse events attributable to the study medi-cation and Fioravanti 1999 Gallacchi 1990 Gualdi 1987 andSugioka 1984 did not mention systemic adverse events Gonzaacutelezde Vega 2013 did not distinguish between local and systemicevents In the remaining studies a total of 16 events were reportedin topical NSAID treatment arms (797 participants 2) and 11with ibuprofen tablets (134 participants 8)
Serious adverse events
No serious adverse events were reported in any treatment arm
Withdrawals
The only withdrawals reported due to adverse events were in stud-ies with placebo treatment arms ( kermark 1990 Fujimaki 1985)and have been reviewedThree studies reported withdrawals due to lack of efficacy (Gonzaacutelez de Vega 2013 Hofman 2000 Tonutti 1994) (Appendix5) There were insufficient data for analysisSome studies reported exclusions from analysis (efficacy or safetyor both) following randomisation mainly due to protocol vio-lations or loss to follow-up (Appendix 5) There is no reason tobelieve these exclusions would introduce systematic bias and thenumbers involved were not likely to influence results
24Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
D I S C U S S I O N
This updated review of topical NSAIDs for acute musculoskele-tal pain in adults differs from previous reviews Previously dataallowed only for comparison of individual topical NSAIDs withplacebo irrespective of formulation With a substantial amount ofnew data for diclofenac it is possible to distinguish effects of for-mulation for individual NSAIDs Because formulation chemistrycan substantially affect the rate and total amount of drug accessingsubcutaneous injured tissues the effect of formulation may be asimportant as the individual NSAID used Drug and formulationshould thus be considered together when assessing efficacy andthis is now possible
We have also included an assessment of the daily dose of NSAIDapplied to the skin This involved having information on the con-centration of NSAID in the preparation the amount used andthe frequency of use Not all studies reported all three but an esti-mation of topical doses applied was possible It varied by factors ofthree or four for each topical NSAID However for topical formu-lations the key issue is less the dose applied but the amount thatpenetrates locally (producing analgesic effect) and the amount en-tering the systemic circulation (producing potential harms) Bothwill depend on the exact formulation of the topical agent andwhether there is occlusion (Moore 2008a)
Summary of main results
This review included 61 studies comparing a topical NSAID withplacebo another topical NSAID or an oral NSAID In total5311 participants were treated with a topical NSAID 3470 withplacebo and 220 with an oral NSAID There were 63 moreparticipants than in the previous version of this review Conditionstreated were sprains strains and contusions mainly resulting fromsports injuries and overuse injuries such as tendinitisFormulations of topical diclofenac ibuprofen ketoprofen piroxi-cam and indomethacin demonstrated significantly higher rates ofclinical success than matching topical placebo lacking the NSAIDbenzydamine did not Three drug and formulation combinationshad NNTs for clinical success below 4 For diclofenac Emulgelreghad the lowest NNT of 18 (15 to 21) in two studies using atleast 50 pain intensity reduction as the outcome (high qualityevidence) Diclofenac plasters other than Flectorreg also had a lowNNT of 32 (26 to 42) based on good or excellent responses inrelatively recent studies (high quality evidence) Ketoprofen gelhad an NNT of 25 (20 to 34) from five studies in the 1980ssome with less well defined outcomes (moderate quality evidence)Ibuprofen gel had an NNT of 39 (27 to 67) from two stud-ies with outcomes of marked improvement or complete remission(moderate quality evidence) All other drug and formulation com-binations had NNT values above 4 indicating lesser efficacyThese results are better than alternative topical products that mightbe used for acute musculoskeletal pain There is no evidence tosupport the use of topical salicylate rubefacients (Derry 2014)
Treatment with a topical NSAID was not associated with anincrease in local adverse events (skin reactions) compared withplacebo (inert carrier) or in withdrawals due to adverse events(high quality evidence) The inert carrier was sometimes associ-ated with mild skin irritation but this rarely led to cessation oftreatment and quickly resolved Systemic adverse events were un-common and did not differ between topical NSAID and placebo(high quality evidence) Two participants experienced serious ad-verse events with diclofenac plaster and diclofenac gel but it isunlikely that these were related to the study medicationsThere were insufficient data directly comparing a topical NSAIDwith the same oral NSAID to draw conclusions about efficacyBased on very limited data for oral NSAIDs there were fewersystemic adverse events with topical than oral treatment Therewere sufficient data only for topical piroxicam compared withtopical indomethacin to compare one topical agent with anotherThese limited data suggested that piroxicam was more effectivethan indomethacin and was less likely to cause local adverse eventsIt is worth noting here that topical indomethacin did not givesignificantly better pain relief than placebo in two of the threestudies in this analysis
Overall completeness and applicability ofevidence
There is a tension between pooling studies to produce analyseswith larger numbers and the subsequent large increases in clinicaland statistical heterogeneity on the one hand and using the ap-proach of clinical homogeneity with subsequent smaller numbersof participants on the other hand Previous reviews have takenthe former approach that is useful in demonstrating that topicalNSAIDs rsquoworkrsquo by being significantly better than placebo Becauseof the substantial increase in the amount of data available in thisreview we have chosen to seek greater clinical homogeneity thisproduces results that are more relevant to patient and prescriberchoiceThere were too few studies comparing one topical NSAID versusanother or versus the same oral NSAID to allow meaningful directcomparisons between individual drugs or routes of administrationThe conditions treated in these studies are representative of thoselikely to be suitable for acute treatment with topical NSAIDs Themean age of participants in individual studies ranged from 25 to57 years and the nature of recruitment in many studies meant thatparticipants were actively engaged in sporting activities Neverthe-less older people in their 60s to 80s were also included in somestudies and the low levels of predominantly mild adverse eventsmeans that this route of administration of NSAIDs is suitable forall age groups able to manage the application processInformation from other sources mainly randomised studies last-ing 12 weeks or more in older populations with arthritis tendto confirm this A systematic review of topical NSAIDs in olderadults was difficult to interpret but suggested that the range of
25Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
withdrawal rates in these studies was similar with topical and oralNSAIDs (Makris 2010) It also claimed potentiation of warfarineffects but that was with topical salicylate not an NSAID Incontrast a pooled safety analysis of topical diclofenac in peopleaged 75 years or older reported minimal changes with a mean re-duction in haemoglobin of less than 1 gL with topical diclofenacand a mean systolic blood pressure reduction of almost 4 mm Hg(Roth 2012) Two large randomised 12-week studies comparingtopical with oral diclofenac in arthritis reported lower rates of gas-trointestinal adverse events with topical than oral especially severeevents but larger reductions in haemoglobin with oral diclofenac(Simon 2009 Tugwell 2004)The available evidence was limited by numbers to comment on rarebut potentially serious adverse events One example is the potentialfor photo-sensitivity reactions with topical ketoprofen Currentadvice from the Medicines and Healthcare products RegulatoryAgency in the UK is to avoid direct sunlight ultraviolet (UV) rayssunlamps and sunbeds while using topical ketoprofen and to seea healthcare professional or go to hospital if they experience a skinreaction to sunlight sunlamps or sunbeds (MHRA 2009)
Quality of the evidence
All included studies were both randomised and double-blind nonewas considered at high risk of methodological bias Many werecarried out in the 1980s and 1990s when methodological rigorand detailed reporting were not given such high priority and stud-ies did not always report details of the randomisation treatmentallocation and blinding processes More recent studies often didreport methodological details and tended to be larger (see Figure4 for a comparison of quality of reporting for different dates andNSAIDs) Our primary outcome of clinical success was not alwayswell-defined and was measured using different scales but againmore recent studies tended to report outcomes betterThe studies were conducted in different conditions with some-what different outcome definitions and duration and with differ-ent topical NSAIDs and formulations Moreover the small size ofmany of the studies is likely to result in considerable chance varia-tion (Counsell 1994 Moore 1998b) These factors would accountfor the high I2 values seen in several analyses Despite these sourcesof potential clinical heterogeneity most studies showed benefit oftopical NSAID over placeboThe design of studies to be able to demonstrate analgesic sensi-tivity is important in self limiting conditions such as strains andsprains Too long a duration and the condition results in sponta-neous resolution of painful symptoms while too short a durationmay be inadequate to show any effect The decision by trialists toconcentrate on outcomes closest to seven days of treatment appearsto be prudent and has been adopted in this and previous reviewsThere are potential differences in response to treatment betweenstrains and sprains and overuse-type injuries such as tendinitis
and future reviews may examine this At the present time thereare too few existing trials to explore any differences adequatelyBaseline pain may be a cause for concern Seven studies didnot report baseline pain levels (Billigmann 1996 Curioni 1985Haig 1986 NCT01255423 NCT01272934 NCT01272947Sinniger 1981) and a further 11 reported either mean levels ofless than moderate pain or a significant proportion of individualswith less than moderate pain ( kermark 1990 Aoki 1984 Auclair1989 Diebshlag 1990 Fujimaki 1985 Jenoure 1997 Linde 1985Picchio 1981 Ramesh 1983 Sugioka 1984 Whitefield 2002)using recognised scales Insufficient pain at baseline compromisesthe ability of a study to demonstrate any improvement All thenewly added studies reported baseline pain to be of at least mod-erate intensity
Potential biases in the review process
There has been greater interest in topical NSAIDs in recent yearsmainly because lower systemic drug levels reduce the risk of trou-blesome and severe adverse events particularly in the gastroin-testinal tract and renal and cardiovascular systems Most of theattention has been in chronic conditions such as osteoarthritiswith few studies in acute painful conditions Low levels of seriousadverse events with topical NSAIDs has been noted previously(Evans 1995) and the near absence of serious adverse events inthis review is unlikely to be due to any biases in the review processOne potential bias is that clinical trials for topical NSAIDs may nothave been published One previous review did find previously un-published trials (Moore 1998a) but a subsequent attempt that in-cluded extensive contacts with pharmaceutical companies revealedno additional data (Mason 2004a) While some old unpublishedstudies of topical NSAIDs in acute painful conditions may existthey constitute an unknown number of studies and participantswhose results are unknown and are likely to remain unknownFurthermore their relevance to current clinical practice may belimited as better formulations are developed New systems of trialregistration mean that we know what recent studies have been doneor are ongoing the number of studies and participants is knowneven if their results remain unknown We identified in Clinicaltri-alsgov three unpublished studies (612 participants) with adverseevent data but no dichotomous efficacy data 14 completed un-published studies (4403 participants) with no results posted andthree ongoing studies (880 participants)For the main topical NSAIDs of interest and where most informa-tion exists about 4200 participants in this review provided data onefficacy for diclofenac ibuprofen ketoprofen indomethacin andbenzydamine compared with placebo For efficacy there are un-known results from almost 5900 participants in studies known tohave been done but essentially unpublished Almost 6500 partic-ipants in this review provided information on local adverse eventsfor topical NSAIDs compared with placebo For local adverseevents the unknown results from known studies represents almost
26Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
5300 participants It is clear that identified unpublished but un-available study data amounts to a further potentially large increasein knowledge over and above the 60 increase in numbers ofparticipants already included in this updated reviewBased on efficacy data on known and available study results un-published trials showing no difference between any topical NSAIDand topical placebo and involving 5500 participants would haveto exist in order for the NNT to be as high as 9 at which pointthe effectiveness of topical NSAIDs would become clinically irrel-evant (Moore 2006) This amount of unpublished negative data isobviously available and while a negative result in all the identifiedstudies is unlikely knowledge would be greatly served by havingthese unpublished trial results availableWe have not yet attempted to obtain results from these clinicaltrials from the trial sponsors because this takes a considerableamount of time and may not be successful Moreover the studieswere spilt between nine different sponsors Cerimon Pharmaceu-ticals (five studies) Novartis (four studies) Endo Pharmaceuticals(three studies) GlaxoSmithKline (two studies) Hisamatsu Phar-maceutical (two studies) and one each from Actavis Pfizer Im-primis Pharmaceutical and Strategic Science amp Technologies
Agreements and disagreements with otherstudies or reviews
A review published in 2004 included some of the studies in thisreview and reported an NNT for all topical NSAIDs combinedof 38 (34 to 44) for clinical success equivalent to half pain reliefat seven days (Mason 2004a) That review found no differencebetween topical NSAID and placebo for local adverse events as didthis review In turn the Mason review was in broad agreement withthe original systematic review on topical NSAIDs (Moore 1998a)To our knowledge no previous review assembled sufficient trialdata to analyse results by both drug and formulation as was donehere
A U T H O R S rsquo C O N C L U S I O N S
Implications for practice
For people with acute musculoskeletal pain
Topical NSAIDs can provide good levels of pain relief in acuteconditions such as sprains strains and overuse injuries proba-bly similar to that provided by oral NSAIDs Gel formulations ofdiclofenac (as Emulgelreg) ibuprofen and ketoprofen and somediclofenac patches provide the best effects Adverse events are usu-ally minimal with topical NSAIDs
For clinicians
Topical diclofenac ibuprofen or ketoprofen gels provide goodpain relief for painful acute musculoskeletal conditions and arebetter tolerated than oral formulations These drugs and formu-lations are more likely to be cost effective than alternative topicalpreparations such as topical rubefacients
For policy makers
Topical NSAIDs are not associated with an increased incidence oflocal skin reactions compared with the inert carrier and while thecarrier may cause mild transient irritation it is rarely troublesomeTopical NSAIDs do not cause systemic (mainly gastrointestinal)problems commonly seen with oral NSAIDs making them partic-ularly useful for individuals unable to tolerate oral administrationor for whom it is contraindicated
For funders
Topical diclofenac ibuprofen or ketoprofen gels should be consid-ered for initial treatment of acute musculoskeletal painful condi-tions where there are no contraindications such as damaged skinThese drugs and formulations are more likely to be cost effectivethan alternative topical preparations such as topical rubefacientsBecause formulations of topical NSAIDs are likely to change overtime the relevant trials performed and reported in or before the1990s must be limited and may be questionable Funders mightwish to consider asking pharmaceutical companies without recenttrial evidence for their products to produce it
Implications for research
General
The general thrust of these findings is that gel formulations of top-ical diclofenac ibuprofen and ketoprofen work best but for somedrugs (ketoprofen for instance) studies were pre-1990 Thesestudies may not be relevant to products available now Becauseformulation can have a significant effect on efficacy formulationchanges should be accompanied by relevant randomised trials
Design
The design of the trials is generally good and the sports injurymodel appears to be reliable and reproducible Modern studieshave ensured that participants entering the trials have at least mod-erate pain and this helps sensitivity to detect an analgesic responseMajor changes to the design of these trials would not appear to beneeded
27Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Measurement (endpoints)
A major issue is not in the measurement of pain as most studiesespecially modern studies have used standard pain intensity andpain relief scales However reporting of average pain changes isinadequate and the use of responder analyses (at least 50 painintensity reduction or people experiencing mild or no pain) ispreferred
Comparison between active treatments
Indirect comparisons with placebo are probably as informative asuse of an active comparator
A C K N O W L E D G E M E N T S
The Oxford Pain Relief Trust the NHS Cochrane CollaborationProgramme Grant Scheme and the NIHR Biomedical ResearchCentre Programme provided support for the earlier review
The National Institute for Health Research (NIHR) is the largestsingle funder of the Cochrane Pain Palliative and SupportiveCare Review Group Disclaimer the views and opinions expressedherein are those of the authors and do not necessarily reflect thoseof the NIHR National Health Service (NHS) or the Departmentof Health
R E F E R E N C E S
References to studies included in this review
Airaksinen 1993 published data onlyAiraksinen O Venaumllaumlinen J Pietilaumlinen T Ketoprofen 25gel versus placebo gel in the treatment of acute soft tissueinjuries International Journal of Clinical PharmacologyTherapy and Toxicology 199331(11)561ndash3
kermark 1990 published data only
kermark C Forsskaringhl B Topical indomethacin in overuseinjuries in athletes A randomized double blind studycomparing Elmetacinreg with oral indomethacin andplacebo International Journal of Sports Medicine 199011(5)393ndash6
Aoki 1984 published data only
Aoki T Numajiri M Yamamoto M A well controlledcomparative study of piroxicam gel indomethacin geland placebo gel in the treatment of trauma JapanesePharmacology and Therapeutics 198412(12)101ndash17
Auclair 1989 published data onlyAuclair J Georges M Grapton X Gryp L DrsquoHooghe MMeisser RG et al A double-blind controlled multi-centerstudy of percutaneous niflumic acid gel and placebo in thetreatment of Achilles heel tendinitis Current TherapeuticResearch 198946(4)782ndash8
Billigmann 1996 published data onlyBilligmann PW Treatment of ankle distortion withibuprofen gel [Therapie von Sprunggelenksndashdistosionenmit ibuprofenndashmikrogel] Therapiewoche 199646(21)1187ndash92
Campbell 1994 published data onlyCampbell J Dunn T Evaluation of topical ibuprofen creamin the treatment of acute ankle sprains Journal of Accidentand Emergency Medicine 199411(3)178ndash82
Chatterjee 1977 published data onlyChatterjee DS A double-blind clinical study withbenzydamine 3 cream on soft tissue injuries in anoccupational health centre Journal of International MedicalResearch 19775(6)450ndash8
Costantino 2011 published data only
Costantino C Kwarecki J Samokhin AV Mautone GRovati S Diclofenac epolamine plus heparin plasterversus diclofenac epolamine plaster in mild to moderateankle sprain a randomized double-blind parallel-groupplacebo-controlled multicentre phase III trial ClinicalDrug Investigation 201131(1)15ndash26 [DOI 10216511585890-000000000-00000]
Coudreuse 2010 published data onlyCoudreuse JM de Vathaire F Effect of a plaster containingDHEP and heparin in acute ankle sprains with oedemaa randomized double-blind placebo-controlled clinicalstudy Current Medical Research and Opinion 201026(9)2221ndash8 [DOI 101185030079952010508020]
Curioni 1985 published data only
Curioni BG Di Domenica F Daolio P Spignoli GEvaluation of ibuproxam gel in traumatology [Valutazionedellrsquoefficacia terapeutica di ibudros gel in traumatologia]Clinica Europea 198524(3)456ndash60
Diebshlag 1990 published data only
Diebschlag W Nocker W Bullingham R A double-blindstudy of the efficacy of topical ketorolac tromethamine gelin the treatment of ankle sprain in comparison to placeboand etofenamate Journal of Clinical Pharmacology 199030
(1)82ndash9
Dreiser 1988 published data onlyDreiser RL Clinical trial of efficacy and tolerability oftopical ibuprofen in the treatment of tendinitis Le JournalInternational De Meacutedecine 198811915ndash31
Dreiser 1989 published data only
Dreiser RL Clinical trial - Fatsum gel FG-6 Supplied byMenarini 1989
Dreiser 1990 published data onlyDreiser RE Charlot J Lopez A Ditisheim A Clinicalevaluation of niflumic acid gel in the treatment ofuncomplicated ankle sprains Current Medical Research andOpinion 199012(2)93ndash9
28Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Dreiser 1994 published data only
Dreiser RL Roche R de Sahb R Thomas F Leutenegger EFlurbiprofen local action transcutaneous (LAT tm) clinicalevaluation in the treatment of acute ankle sprains EuropeanJournal of Rheumatology and Inflammation 199414(4)9ndash13
Fioravanti 1999 published data onlyFioravanti A Cicero MR Nerucci F Manopulo RMarcolongo R Double-blind controlled clinical study of theefficacy and tolerability of diclofenac-N-(2-hydroxyethyl)-pyrrolidine lecithin gel compared with diclofenac-N-(2-hydroxyethyl)-pyrrolidine gel in patients with periand extraarticular inflammatory diseases Drugs underExperimental and Clinical Research 199925(5)235ndash40
Fujimaki 1985 published data only
Fujimaki E et al Clinical evaluation of piroxicam gelversus indomethacin gel and placebo in the treatment ofmuscle pain a double-blind multicenter study JapanesePharmacology and Therapeutics 198512(12)119ndash37
Gallacchi 1990 published data only
Gallacchi G Mautone G Lauldi P Topical treatmentwith diclofenac hydroxyethylpyrrilidine (Flector gel 1)Clinical Trials Journal 199027(1)58ndash64
Gonzaacutelez de Vega 2013 published data only
Gonzaacutelez de Vega C Speed C Wolfarth B Gonzaacutelez JTraumeel vs diclofenac for reducing pain and improvingankle mobility after acute ankle sprain a multicentrerandomised blinded controlled and non-inferiority trialInternational Journal of Clinical Practice 201367(10)979ndash89 [DOI 101111ijcp12219]
Governali 1995 published data only
Governali E Casalini D A controlled clinical studyon ketoprofen gel 5 versus ketoprofen ointment 1in patients with post-traumatic lesions [Ricerda clinicacontrollata tra ketoprofene gel 5 e ketoprofene crema1 in pazienti con postumi di lesioni traumatiche]Riabilitazione 199528(1)61ndash9
Gualdi 1987 published data only
Gualdi A Bonollo L Martini A Forgione A Non-steroidal anti-inflammatory drugs for topical therapy intraumatology a double-blind study with flunoxaprofen andketoprofen [Antinflammatori no steroidi per uso topicoin traumatologia studio clinico con flunoxaprofene echetoprofene] Riforma Medica 1987102(10)401ndash4
Haig 1986 published data only
Haig G Portable thermogram technique for topicallyapplied benzydamine cream in acute soft-tissue injuriesInternation Journal of Tissue Reactions 19868(2)145ndash7
Hoffmann 2012 published data only
Hoffmann P Kopa ka P Gugliotta B Rovati S Efficacyand tolerability of DHEP-heparin plaster in reducingpain in mild-to-moderate muscle contusions a double-blind randomized trial Current Medical Researchand Opinion 201228(8)1313ndash21 [DOI 101185030079952012709182]
Hofman 2000 published data only
Hofman J Nasswetter G Cayetti LM Lysine clonixinate gelin soft tissue injuries Controlled randomized prospectivedouble-blind clinical trial with diclofenac [Clonixinatode lisina gel en lesiones de tejidos blandos ensayo clinicoprospectivo doble ciego al azar controlado con diclofenac]Prensa Medica Argentina 200087(5)513ndash20
Hosie 1993 published data only
Hosie GAC The topical NSAID felbinac versus oralibuprofen a comparison of efficacy in the treatment ofacute lower back injury British Journal of Clinical Research199345ndash17
Jenoure 1997 published data only
Jenoure PJ Rostan A Gremion G Meier JL Grossen RBielinki R et al Multicentre double-blind controlledclinical study on the efficacy of diclofenac epolamineTissugel plaster in patients with epicondylitis [Studiomulticentrico controllato in doppio cieco su diclofenacTissugel plaster in pazienti con epicondilite] MedicinaDello Sport 197750(3)285ndash92
Joussellin 2003 published data onlylowast Joussellin E Flector Tissugel for the treatment ofpainful ankle sprains [Flector Tissugel dans le traitementdes entorses douloureuses de la cheville] Journal deTraumatologie du Sport 2003201S5ndash9Lionberger DR Joussellin E Lanzarotti A Yanchick JMagelli M Diclofenac epolamine topical patch relieves painassociated with ankle sprain Journal of Pain Research 2011447ndash53 [DOI 102147JPRS15380]Lionberger DR Joussellin E Yanchick J Magelli MLanzarotti A Pooled analysis of clinical trial data evaluatingthe safety and effectiveness of diclofenac epolamine topicalpatch 13 for the treatment of acute ankle sprain OpenAccess Journal of Sports Medicine 2011275ndash84 [DOI102147OAJSMS17048]
Julien 1989 published data onlyJulien D Clinical trial - Fastum gel FG-8 Supplied byMenarini 1989
Klainguti 2010 published data onlyKlainguti A Forgacs A Berkes I Castellacci E A plastercontaining DHEP and heparin for mild to moderatecontusions and sprains with haematoma a double-blind randomized study Current Medical Researchand Opinion 201026(9)2243ndash51 [DOI 101185030079952010508022]
Kockelbergh 1985 published data onlyKockelbergh M Verspeelt P Caloine R Dermaux F Localanti inflammatory treatment with a ketoprofen gel currentclinical findings Journal Belge de Medecine Physique et deRehabilitation 19858(4)205ndash13
Kuehl 2011 published data only
Kuehl K Carr W Yanchick J Magelli M Rovati SAnalgesic efficacy and safety of the diclofenac epolaminetopical patch 13 (DETP) in minor soft tissue injuryInternational Journal of Sports Medicine 201132(8)635ndash43[DOI 101055s-0031-1275359]
29Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Li 2013 published data only
Li C Frangione V Rovati S Zheng Q Diclofenacepolamine medicated plaster in the treatment of minor softtissue injuries a multicenter randomized controlled trialCurrent Medical Research and Opinion 201329(9)1137ndash46[DOI 101185030079952013816669]
Linde 1985 published data only
Linde F Hvass I Juumlrgensen U Madsen F Treatment ofsprained ankles with 5 benzydamine creme A double-blind study [Ankelforstuvninger behandlet med benzydamin5 creme] Ugeskrift for Laeger 1985148(1)12ndash3
Machen 2002 published data onlyMachen J Whitefield M Efficacy of a proprietary ibuprofengel in soft tissue injuries a randomised double blindplacebo controlled study International Journal of ClinicalPharmacology 200256(2)102ndash6
Mahler 2003 published data only
Mahler P Mahler F Duruz H Ramazzina M Liguori VMautone G Double-blind randomized controlled studyon the efficacy and safety of a novel diclofenac epolaminegel formulated with lecithin for the treatment of sprainsstrains and contusions Drugs under Experimental andClinical Research 200329(1)45ndash52
Maziegraveres 2005a published data onlyMaziegraveres B Rouanet S Velicy J Scarsi C Reiner V Topicalketoprofen patch (100 mg) for the treatment of anklesprain a randomized double-blind placebo-controlledstudy American Journal of Sports Medicine 200533(4)515ndash23 [DOI 1011770363546504268135]
Maziegraveres 2005b published data onlyMaziegraveres B Rouanet S Guillon Y Scarsi C Reiner VTopical ketoprofen patch in the treatment of tendinitisa randomized double blind placebo controlled studyJournal of Rheumatology 200532(8)1563ndash70
McLatchie 1989 published data onlyMcLatchie GR McDonald M Lawrence GF RogmansD Lisai P Hibberd M Soft tissue trauma a randomisedcontrolled trial of the topical application of felbinac a newNSAID British Journal of Clinical Practice 198943(8)277ndash80
Morris 1991 published data only
Morris WD Scott HV Peters WA Ketelbey JW Felbinactopical gel for acute soft tissue sports injuries New ZealandJournal of Sports Medicine 19911945ndash7
NCT01255423 unpublished data only
Novartis (Sponsors) A randomized double-blind multi-center placebo-controlled parallel group study to evaluatethe efficacy and safety of diclofenac sodium topical gel(DSG) 1 applied 4 times daily in subjects with acuteankle sprain clinicaltrialsgovct2showNCT01255423(accessed 3 February 2015) 2012 [CTG NCT01255423]
NCT01272934 unpublished data only
Novartis (Sponsors) A randomized double-blind multi-center placebo-controlled parallel group study to evaluatethe efficacy and safety of diclofenac sodium topical gel
(DSG) 1 applied four times daily in subjects withacute ankle sprain clinicaltrialsgovct2showresultsNCT01272934 (accessed 3 February 2015) 2013 [CTGNCT01272934]
NCT01272947 unpublished data onlylowast Novartis (Sponsors) A randomized double-blindmulti-center placebo-controlled parallel group study toevaluate the efficacy and safety of diclofenac sodium topicalgel (DSG) 1 applied four times daily in subjects withacute blunt soft tissue injuriescontusions of the limbsclinicaltrialsgovct2showresultsNCT01272947 (accessed3 February 2015) 2012 [CTG NCT01272947]Pallay R Pabst H Giannetti B Burnett I Monnet JDiclofenac sodium topical gel (DSG) 1 for acute softtissue injuriescontusions of the limbs 2013 AnnualAssembly of the American Academy of Physical Medicineand Rehabilitation National Harbor MD United States2013 Vol 5 (9 Suppl 1)S204
Noret 1987 published data only
Noret A Roty V Allington N Hauters P Zuinen C PoelsR Ketoprofen gel as topical treatment for sport injuriesActa Therapeutica 198713367ndash78
Parrini 1992 published data onlyParrini M Cabitza P Arrigo A Vanasia M Efficacy andtolerability of ketoprofen lysine salt foam for topical use inthe treatment of traumatic pathologies of the locomotorapparatus [Efficacia e tollerabilita del ketoprofene sale dilisina schiuma per uso topico nel trattamento di alcunepatologie traumatiche dellrsquoapparato locomotore] La ClinicaTerapeutica 1992141(9)199ndash204
Picchio 1981 published data only
Picchio AA Volta S Longoni A Controlled clinical trial ofibuprofen for topical use in sport injuries [Studio clinicocontrollato sullrsquoimpiego dellrsquoibuprofen per uso topico intraumatologica sportiva] Medicina dello Sport 198134403ndash6
Predel 2004 published data only
Mueller EA Kirch W Reiter S Extent and time courseof pain intensity upon treatment with a topical diclofenacsodium patch versus placebo in acute traumatic injury basedon a validated end point post hoc analysis of a randomizedplacebo-controlled trial Expert Opinion on Pharmacotherapy201011(4)493ndash8 [DOI 10151714656560903535898]lowast Predel HG Koll R Pabst H Dieter R Gallacchi GGiannetti B et al Diclofenac patch for topical treatmentof acute impact injuries a randomised double blindplacebo controlled multicentre study British Journal ofSports Medicine 200438(3)318ndash23 [DOI 101136bjsm2003005017]
Predel 2012 published data onlyPredel HG Hamelsky S Gold M Giannetti B Efficacy andsafety of diclofenac diethylamine 232 gel in acute anklesprain Medicine and Science in Sports and Exercise 201244
(9)1629ndash36 [DOI 101249MSS0b013e318257ed41]
30Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Predel 2013a published data only
Predel HG Giannetti B Seigfried B Novellini R MenkeG A randomized double-blind placebo-controlledmulticentre study to evaluate the efficacy and safetyof diclofenac 4 spray gel in the treatment of acuteuncomplicated ankle sprain Journal of InternationalMedical Research 201341(4)1187ndash202 [DOI 1011770300060513487639]
Predel 2013b published data onlyPredel HG Giannetti B Pabst H Schaefer A Hug AMBurnett I Efficacy and safety of diclofenac diethylamine116 gel in acute neck pain a randomized double-blindplacebo-controlled study BMC Musculoskeletal Disorders201314250 [DOI 1011861471-2474-14-250]
Ramesh 1983 published data only
Ramesh N Steuber U Ibuprofen in a cream vehicle foraccidental and sport injuries [Dolgit creme bei unfallndash undsportverletzungen] Therapiewoche 1983334563ndash70
Rowbotham 2003 published data only
Rowbotham M Galer B Block J Backonja M FlectorTissugel efficacy and safety in minor sport injuries Avs controlled clinical trial Journal of Sports Traumatology200320IS15ndashIS20
Russell 1991 published data only
Russell AL Piroxicam 05 topical gel compared to placeboin the treatment of acute soft tissue injuries a double-blind study comparing efficacy and safety Clinical andInvestigative Medicine 199114(1)35ndash43
Saillant 1998 published data only
Lionberger DR Joussellin E Yanchick J Magelli MLanzarotti A Pooled analysis of clinical trial data evaluatingthe safety and effectiveness of diclofenac epolamine topicalpatch 13 for the treatment of acute ankle sprain OpenAccess Journal of Sports Medicine 2011275ndash84 [DOI102147OAJSMS17048]lowast Saillant G Study comparing the efficacy and toleranceof Flector Tissugelreg to that of a placebo in the treatmentof benign ankle sprains [Eacutetude comparant lrsquoefficaciteacute et latoleacuterance de FlectorTissugelreg agrave celles drsquoun placebo dans letraitement des entorses beacutenignes de la cheville] Medicinedu Sport 1998721ndash5
Sanguinetti 1989 published data onlySanguinetti C Treatment of soft tissue injury with BPAAgel Results of an Italian multicenter study vs placebo[Trattemento con BPAA gel dei traumi dei tessuti moli]Clinica Terapeutica 1989130(5)255ndash258
Sinniger 1981 published data onlySinniger M Blanchard P Controlled clinical trial withFentiazac cream in sport microtraumatology Journal ofInternational Medical Research 19819(4)300ndash2
Spacca 2005 published data onlySpacca G Cacchio A Forgcs A Monteforte PRovetta G Analgesic efficacy of a lecithin-vehiculateddiclofenac epolamine gel in shoulder periarthritis andlateral epicondylitis a placebo-controlled multicenter
randomized double-blind clinical trial Drugs underExperimental and Clinical Research 200531(4)147ndash54
Sugioka 1984 published data only
Sugioka Y Multicenter clinical evaluation of piroxicam gelvs indomethacin gel in the treatment of non-traumaticdiseases of tendon or muscle Japanese Pharmacology andTherapeutics 198412139ndash53
Thorling 1990 published data only
Thorling J Linden B Berg R Sandahl A A double blindcomparison of naproxen gel and placebo in the treatment ofsoft tissue injuries Current Medical Research and Opinion199012242ndash8
Tonutti 1994 published data only
Tonutti A The use of ketoprofen gel 5 (orudis gel) intraumatology controlled double-blind study vs etofenamate[Utilizzazione del ketoprofene gel 5 (orudis gel) nellapratica traumatologica studio in doppio cieco controllatoverso etofenamato] Ortopedia e Traumatologia Oggi 199414(3)119ndash25
Vecchiet 1991 published data only
Vecchiet L Colozzi A Effects of meclofenamic acid in thetreatment of lesions deriving from minor traumatologyClinical Journal of Pain 19917 Suppl 1S54ndash9
Whitefield 2002 published data only
Whitefield M OrsquoKane CJA Anderson S Comparativeefficacy of a proprietary topical ibuprofen gel and oralibuprofen in acute soft tissue injuries a randomised doubleblind study Journal of Clinical Pharmacy and Therapeutics200227(6)409ndash17
References to studies excluded from this review
Ambrus 1987 published data only
Ambrus P Boumlhmer D Mobilat ointment in acute sprains[Mobilat Salbe bei akuten Distorsionen] Fortschritte derMedizin 1987105(13)259ndash62
Anon 1993 unpublished data only
Anon Comparative clinical efficacy of Oruvail piroxicamand diclofenac gels in soft tissue injury Unpublished
Ascherl 1982 published data onlyAscherl R Schlemmer H Blumel G Lechner F Theeffectiveness of etofenamate in minor sports injuries ofthe knee and ankle joint A double blind study [DieWirksamkeit der Etofenamat in kleineren Sportverletzungenam Knie und Sprunggelenk Eine DoppelblindndashStudie]Fortschritte der Medizin 1982100(37)1729ndash34
Bagliani 1976 published data only
Bagliani A Montalbetti L Topical treatment ofthrombophlebitis with feprazone and benzydamineControlled clinical study [Il trattamento topico ditromboflebite con feprazone e benzidamina Studio clinicocontrollato] Minerva Medica 197667(14)880ndash4
Baracchi 1982 published data only
Baracchi G Messina Denaro S Piscini S Experience ofthe topical use of isobutylfenylproprionic acid (Ibuprofen)in traumatic inflammation A double-blind comparison
31Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
with placebo [Esperienza sullrsquoimpiego topico dellrsquoacidoisobutilndashfenilndashpropionico (ibuprofen) nella flogositraumatica Confronto a doppia cecita con placebo]Gazzetta Medica Italiana 1982141(12)691ndash4
Boumlhmer 1995 published data onlyBoumlhmer D Ambrus P Treatment of muscular injurieswith diclofenac-diethylammonium emugel [Behandlungvon muskelverletzungen mit diclofenacndashdiethylammoniumemugel] Sportverletzung Sportschaden 1995994ndash5
Burnham 1998 published data only
Burnham R Gregg R Healy P Steadward R Theeffectiveness of topical diclofenac for lateral epicondylitisClinical Journal of Sport Medicine 19988(2)78ndash81
Cesarone 2008 published data onlyCesarone MR Belcaro G Pellegrini L Ledda A VinciguerraG Ricci A et al Treatment of ankle sprain in patientswith vascular diseases of the lower limbs MinervaCardioangiologica 200856 (5 Suppl)39ndash46
Coulibaly 2009 published data only
Coulibaly SK Courau S Staiger C The sprained anklecomparative study of a comfreyroot extract ointment versusa diclofenac gel [Entorses de la cheville eacutetude comparativeentre un extrait agrave base de racines de consoude et lediclofeacutenac] Phytotheacuterpie 20097147ndash9 [DOI 101007s10298-009-0383-x]
Diebschlag 1985 published data only
Diebschlag W Benzydamine cream in post-traumaticoedema International Journal of Tissue Reactions 19857(3)219ndash23
Diebschlag 1986 published data onlyDiebschlag W Diclofenac in blunt traumatic ankle jointswelling Volumetric monitoring in a placebo controlleddouble blind trial [Diclofenac in stumpfen traumatischenSprunggelenk Schwellung Volumetrische Uumlberwachungin einer Placebondashkontrollierten DoppelblindndashStudie]Fortschritte der Medizin 1986104(21)437ndash40
Diebschlag 1992 published data onlyDiebschlag W Nocker W Lehmacher W Treatment of acutesprains of the ankle joint A comparison of the effectivenessand tolerance of two gel preparations containingindomethacin [Die Behandlung der akuten Verstauchungendes Sprunggelenks Ein Vergleich der Wirksamkeitund Vertraumlglichkeit von zwei GelndashZubereitungen mitIndometacin] Fortschritte der Medizin 1992110(6)64ndash72
Fantato 1971 published data only
Fantato S De Gregorio M Clinical evaluation of topicalbenzydamine in traumatology Arzneimittel-Forschung197121(10)1530ndash5
Galer 2000 published data onlyGaler BS Rowbotham M Perander J Devers A FriedmanE Topical diclofenac patch relieves minor sports injury painresults of a multicenter controlled clinical trial Journal ofPain and Symptom Management 200019(4)287ndash94
Hallmeier 1986 published data only
Hallmeier B Michelbach B Etofenamate under tapebandages - a controlled study [Etofenamat untertapendashverbaumlnden] Medizinische Welt 198637(43)1344ndash8
Hallmeier 1988 published data onlyHallmeier B Efficacy and tolerance of etofenamate anddiclofenac in acute sports injuries [Wirksamkeit undVertraumlglichkeit von Etofenamat und Diclofenac bei akutenSportverletzungen] Rheuma 19888183ndash6
Kaneko 1999 published data only
Kaneko M Shimojo H Saito H Onuma Y YamashitaK Clinical evaluation of felbinac patch (SELSPOT) onpost-traumatic disease clinical comparative study versuscommercially available patch Japanese Pharmacology andTherapeutics 19992775ndash85
Kockelbergh 1985b published data only
Kockelbergh M Verspeelt P Caloine R Dermaux F Localanti-inflammatory treatment with a ketoprofen gel currentclinical findings [Traitement antindashinflammatoire local parun gel de keacutetoprofegravene donneacutees cliniques reacutecentes] JournalBelge de Medecine Physique et de Rehabilitation 19858(4)205-13 (study 2)
Kuwabara 2013 published data only
Kuwabara Y Hamamoto H Hikake S Miwa Y Arandomized multi-center double-blind placebo-controlledphase IIIII trial to evaluate the eftopical patch in thetreatment of pain Journal of Pain 2013 Vol 14 (4 Suppl1)S73
Lee 1991 published data only
Lee EH Lee PY Ngai AT Chiu EH Treatment of acute softtissue trauma with a topical non-steroidal anti-inflammatorydrug (biphenylacetic acid 3 gel) Singapore MedicalJournal 199132(4)238ndash41
Link 1996 published data onlyLink R Balint G Pavlik G Otto J Krause W Topicaltreatment of soft tissue rheumatism and athletic injuriesEffectiveness and tolerance of a new ketoprofen gel[Topische Behandlung von WeichteilndashRheumatismus undSportverletzungen Wirksamkeit und Vertraumlglichkeit einesneuen Ketoprofen Gel] Fortschritte der Medizin 1996114(25)311ndash4
May 2007 published data onlyMay JJ Lovell G Hopkins WG Effectiveness of1 diclofenac gel in the treatment of wrist extensortenosynovitis in long distance kayakers Journal of Scienceand Medicine in Sport 200710(1)59ndash65
Oakland 1993 published data only
Oakland C Rapier C A comparison of the efficacy of thetopical NSAID felbinac and ultrasound in the treatmentof acute ankle injuries British Journal of Clinical Research1993489ndash96
Odaglia 1987 unpublished data only
Odaglia G Sereni G Sports minor traumatology results ofa double-blind controlled clinical study ketoprofen (fastumgel 25) versus placebo Menarini unpublished data
32Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Picardi 1993 published data only
Picardi E De Iasio R Efficacy of percutaneous anti-inflammatory drugs (NSAIDs) in swimmers and waterpoloplayers [Efficacia dei farmaci antinfiammatori (FANS) pervia percutanea in atleti di nuoto e pallanuoto] ClinicaTerapeutica 1993143(6)507ndash9
Taboada 1992 published data onlyTaboada A Controlled trial of piroxicam gel associated withultrasound in acute disturbances of the locomotive system[Experienca controlada con gel de piroxicam asociado aultrasonidos en afecciones agudas del aparato locomotor]Prensa Medica Argentina 199279(10)630ndash2
Vanderstraeten 1990 published data only
Vanderstraeten G Schuermans P Study on the effect ofetofenamate 10 cream in comparison with an oral NSAIDin strains and sprains due to sports injuries Acta BelgicaMedica Physica 199013(3)139ndash41
Vinciguerra 2008 published data only
Vinciguerra G Belcaro G Cesarone MR Errichi BM DiRenzo A Errichi S et al Management of uncomplicatedankle sprains with topical or oral ketoprofen treatmentA registry study Minerva Cardioangiologica 200856 (5
Suppl)47ndash53
Von Klug 1977 published data onlyVon Klug H Experience with a locally applied anti-inflammatory drug [Erfahrungen mit einem localanwendbaren antirheumatikum] Arzneimittel-ForschungDrug Research 1977271350ndash4
References to studies awaiting assessment
NCT00351104 published data onlyEndo Pharmaceuticals (Sponsors) A randomized double-blind placebo-controlled parallel group phase III study ofthe efficacy tolerability and safety of ketoprofen topicalpatch 20 (KTP) in the treatment of pain associated withgrade 1 or grade 2 ankle sprain or strain clinicaltrialsgovct2showNCT00351104 (accessed 3 February 2015) 2008[CTG NCT00351104]
NCT00352625 unpublished data only
Endo Pharmaceuticals (Sponsors) A randomized double-blind placebo-controlled parallel group phase III study ofthe efficacy tolerability and safety of ketoprofen topicalpatch 20 (KTP) in the treatment of pain associatedwith tendonitis or bursitis of the shoulder elbow or kneeclinicaltrialsgovct2showNCT00352625 (accessed 3February 2015) 2008 [CTG NCT00352625]
NCT00426985 unpublished data only
Endo Pharmaceuticals (Sponsors) A randomized double-blind placebo-controlled parallel group phase iii study ofthe efficacy tolerability and safety of ketoprofen topicalpatch 20 (KTP) in the treatment of pain associatedwith tendonitis or bursitis of the shoulder elbow or kneeclinicaltrialsgovct2showNCT00426985 (accessed 3February 2015) 2008 [CTG NCT00426985]
NCT00640705 unpublished data only
Cerimon Pharmaceuticals (Sponsors) A randomizeddouble-blind placebo-controlled study of the efficacyand safety of a diclofenac sodium patch for the topicaltreatment of pain due to mild to moderate ankle sprainclinicaltrialsgovct2showNCT00640705 (accessed 3February 2015) 2008 [CTG NCT00640705]
NCT00640939 unpublished data onlyCerimon Pharmaceuticals (Sponsors) A randomizeddouble-blind placebo-controlled study of the efficacy andsafety of a diclofenac sodium patch for the topical treatmentof pain due to mild to moderate tendonitis or bursitisclinicaltrialsgovct2showNCT00640939 (accessed 3February 2015) 2008 [CTG NCT00640939]
NCT00680472 unpublished data onlyHisamitsu Pharmaceutical Co Inc (Sponsors) Arandomized multicenter double-blind placebo-controlledtwo-week study to assess the efficacy and safety of HKT-500 in subjects with acute shoulder pain clinicaltrialsgovct2showNCT00680472 (accessed 3 February 2015) 2015[CTG NCT00680472]
NCT00680784 unpublished data only
Hisamitsu Pharmaceutical Co Inc (Sponsors) Arandomized multicenter double-blind placebo-controlledstudy to evaluate the efficacy and safety of HKT-500 inthe treatment of pain associated with Grade I or Grade IIankle sprain clinicaltrialsgovct2showNCT00680784(accessed 3 February 2015) 2015 [CTG NCT00680784]
NCT00765700 unpublished data onlyEkman E Skrepnik S Jones M Lawson K Schupp JEfficacy and safety of ketoprofen 10 cream in acute softtissue injuries (phase 3 study tdlp-110-001) Proceedings ofthe 13th World Congress on Pain 2010 Aug 29 - Sept 2Montreal CanadaImprimis Pharmaceuticals Inc (Sponsors) A randomizedmulticenter double-blind placebo-controlled parallel-group phase 3 study to assess the efficacy and safetyof Ketotransdeltrade (Ketoprofen topical cream 10) inthe treatment of pain associated with mild to moderateacute soft tissue injury clinicaltrialsgovct2showNCT00765700 (accessed 3 February 2015) 2013 [CTGNCT00765700]
NCT00869063 unpublished data onlyCerimon Pharmaceuticals (Sponsors) A randomizeddouble-blind placebo-controlled study of the efficacy andsafety of a diclofenac sodium patch for the topical treatmentof acute pain due to mild to moderate wrist sprain strainor contusion clinicaltrialsgovct2showNCT00869063(accessed 3 February 2015) 2010 [CTG NCT00869063]
NCT00869180 unpublished data onlyCerimon Pharmaceuticals (Sponsors) A randomizeddouble-blind placebo-controlled study of the efficacy andsafety of a diclofenac sodium patch for the topical treatmentof acute pain due to mild to moderate ankle sprainclinicaltrialsgovct2showNCT00869180 (accessed 3February 2015) 2010 [CTG NCT00869180]
33Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT00931866 unpublished data only
Cerimon Pharmaceuticals (Sponsors) A randomizeddouble-blind placebo-controlled study of the efficacy andsafety of a diclofenac sodium patch for the topical treatmentof acute pain due to mild to moderate soft tissue injuriesclinicaltrialsgovct2showNCT00931866 (accessed 3February 2015) 2010 [CTG NCT00931866]
NCT01874626 unpublished data only
Strategic Science amp Technologies LLC (Sponsors) Arandomized double-blind placebo-controlled multi-dosepivotal study to determine the efficacy and safety of SST0225 a topical ibuprofen cream in the treatment of painassociated with acute ankle sprain clinicaltrialsgovct2showNCT01874626 (accessed 3 February 2015) 2014[CTG NCT01874626]
NCT01957215 unpublished data only
GlaxoSmithKline (Sponsors) A clinical study to assess theefficacy of pain relief of topical indomethacin patch overplacebo in ankle sprain patients clinicaltrialsgovct2showNCT01957215 (accessed 3 February 2015) 2015 [CTGNCT01957215]
NCT02324270 unpublished data only
Actavis Inc (Sponsors) Randomized double-blindmultiple-center placebo-controlled study comparing thesafety and efficacy of generic diclofenac epolamine toFlectorreg patch in the treatment of acute pain due to minorankle sprain clinicaltrialsgovct2showNCT02324270(accessed 3 February 2015) 2014 [CTG NCT02324270]
Sarzi-Puttini 2014 published data only
Sarzi-Puttini P Atzeni F Damiani C Casale R BarbagalloM Cazzola M A double-blind randomised parallelgroup active controlled multicentre study to assess thetherapeutic non-inferiority of SKP-021 a 03 ketoprofenpatch versus diclofenac sodium patch in patients withacute inflammatory musculoskeletal injuries Proceeding ofthe Annual European Congress of Rheumatology of theEuropean League Against Rheumatism EULAR 2014 June11-14 Paris France 201473
References to ongoing studies
NCT01945034 unpublished data onlyPfizer (Sponsors) Placebo-controlled double-blindevaluation of the efficacy and safety of ibuprofen 5 topicalgel for the treatment of ankle sprain clinicaltrialsgovct2showNCT01945034 (accessed 3 February 2015) 2014[CTG NCT01945034]
NCT02100670 unpublished data onlyGlaxoSmithKline (Sponsors) A clinical study to assessthe efficacy and onset of pain relief of topical MFC51123diclofenac-menthol gel versus controls in ankle sprainclinicaltrialsgovct2showNCT02100670 (accessed 3February 2015) 2014 [ClinicalTiralsgov NCT02100670]
NCT02290821 unpublished data only
Novartis (Sponsors) A randomized double-blind placebo-controlled parallel group study to evaluate the efficacy andsafety of diclofenac sodium topical gel (DSG) 1 applied
four times daily in subjects with acute blunt soft tissueinjuriescontusions of the limbs clinicaltrialsgovct2showNCT02290821 (accessed 3 February 2015) 2015 [CTGNCT02290821]
Additional references
Anon 2005
Anon Topical analgesics a review of reviews and a bit ofperspective wwwjr2oxacukBandolierExtraforbandoTopextra3pdf 2005
Cook 1995
Cook RJ Sackett DL The number needed to treat aclinically useful measure of treatment effect BMJ 1995310
(6977)452ndash4
Counsell 1994Counsell CE Clarke MJ Slattery J Sandercock PA Themiracle of DICE therapy for acute stroke fact or fictionalproduct of subgroup analysis BMJ 1994309(6970)1677ndash81
Dechartres 2013Dechartres A Trinquart L Boutron I Ravaud P Influenceof trial sample size on treatment effect estimates meta-epidemiological study BMJ 2013346f2304 [DOI101136bmjf2304]
Derry 2012a
Derry S Moore RA Topical capsaicin (low concentration)for chronic neuropathic pain in adults Cochrane Databaseof Systematic Reviews 2012 Issue 9 [DOI 10100214651858CD010111]
Derry 2012bDerry S Moore RA Rabbie R Topical NSAIDs forchronic musculoskeletal pain in adults Cochrane Databaseof Systematic Reviews 2012 Issue 9 [DOI 10100214651858CD007400pub2]
Derry 2013Derry S Sven-Rice A Cole P Tan T Moore RA Topicalcapsaicin (high concentration) for chronic neuropathic painin adults Cochrane Database of Systematic Reviews 2013Issue 2 [DOI 10100214651858CD007393pub3]
Derry 2014
Derry S Matthews PR Wiffen PJ Moore RASalicylate-containing rubefacients for acute and chronicmusculoskeletal pain in adults Cochrane Database ofSystematic Reviews 2014 Issue 11 [DOI 10100214651858CD007403pub3]
Evans 1995Evans JM McMahon AD McGilchrist MM White GMurray FE McDevitt DG et al Topical non-steroidal anti-inflammatory drugs and admission to hospital for uppergastrointestinal bleeding and perforation a record linkagecase-control study BMJ 1995311(6996)22ndash6
FitzGerald 2001
FitzGerald GA Patrono C The coxibs selective inhibitorsof cyclooxygenase-2 New England Journal of Medicine2001345(6)433ndash42
34Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Haroutiunian 2010
Haroutiunian S Drennan DA Lipman AG Topical NSAIDtherapy for musculoskeletal pain Pain Medicine 201011
(4)535ndash49 [DOI 101111j1526-4637201000809x]
Hawkey 1999
Hawkey CJ Cox-2 inhibitors Lancet 1999353(9149)307ndash14
Higgins 2011
Higgins JPT Altman DG Sterne JAC Chapter 8 Assessingrisk of bias in included studies In Higgins JPT GreenS (editors) Cochrane Handbook for Systematic Reviewsof Interventions Version 510 [updated March 2011]The Cochrane Collaboration 2011 Available fromwwwcochrane-handbookorg
Jadad 1996a
Jadad AR Carroll D Moore A McQuay H Developing adatabase of published reports of randomised clinical trials inpain research Pain 199666(2-3)239ndash46
Jadad 1996b
Jadad AR Moore RA Carroll D Jenkinson C ReynoldsDJ Gavaghan DJ et al Assessing the quality of reports ofrandomized clinical trials is blinding necessary ControlledClinical Trials 199617(1)1ndash12
LrsquoAbbeacute 1987
LrsquoAbbeacute KA Detsky AS OrsquoRourke K Meta-analysis inclinical research Annals of Internal Medicine 1987107(2)224ndash33
Makris 2010Makris UE Kohler MJ Fraenkel L Adverse effects oftopical nonsteroidal antiinflammatory drugs in older adultswith osteoarthritis a systematic literature review Journalof Rheumatology 201037(6)1236ndash43 [DOI 103899jrheum090935]
Mason 2004a
Mason L Moore RA Edwards JE Derry S McQuay HJTopical NSAIDs for acute pain a meta-analysis BMCFamily Practice 2004510
Mason 2004bMason L Moore RA Edwards JE Derry S McQuayHJ Topical NSAIDs for chronic musculoskeletal painsystematic review and meta-analysis BMC MusculoskeletalDisorders 2004528
McPherson 2013McPherson ML Cimino NM Topical NSAIDformulations Pain Medicine 201314 Suppl 1S35ndash9[DOI 101111pme12288]
MHRA 2009
Medicines and Healthcare Products Regulatory AgencyTopical ketoprofen reminder on risk of photo-sensitivityreactions Drug Safety Update 2009 June Vol 2 issue 116
Moore 1998aMoore RA Tramegraver MR Carroll D Wiffen PJ McQuayHJ Quantitative systematic review of topically applied non-
steroidal anti-inflammatory drugs BMJ 1998316(7128)333ndash8
Moore 1998bMoore RA Gavaghan D Tramegraver MR Collins SL McQuayHJ Size is everything - large amounts of information areneeded to overcome random effects in estimating directionand magnitude of treatment effects Pain 199878(3)209ndash16
Moore 2003
Moore RA Edwards J Barden J McQuay HJ BandolierrsquosLittle Book of Pain Oxford Oxford University Press 2003[ISBN 0ndash19ndash263247ndash7]
Moore 2006
Moore RA Barden J Derry S McQuay HJ Managingpotential publication bias In McQuay HJ Kalso EMoore RA editor(s) Systematic Reviews in Pain ResearchMethodology Refined Seattle IASP Press 200615ndash23[ISBN 978ndash0ndash931092ndash69ndash5]
Moore 2008aMoore RA Derry S McQuay HJ Topical agents in thetreatment of rheumatic pain Rheumatic Diseases Clinics ofNorth America 200834(2)415ndash32
Moore 2008bMoore RA Barden J Derry S McQuay HJ Managingpotential publication bias In McQuay HJ Kalso EMoore RA editor(s) Systematic Reviews in Pain ResearchMethodology Refined Seattle IASP Press 200815ndash24[ISBN 978ndash0ndash931092ndash69ndash5]
Moore 2013
Moore RA Straube S Aldington D Pain measures andcut-offs - rsquono worse than mild painrsquo as a simple universaloutcome Anaesthesia 201368(4)400ndash12 [DOI 101111anae12148]
Morris 1995
Morris JA Gardner MJ Calculating confidence intervalsfor relative risk odds ratios and standardised ratios andrates In Gardner MJ Altman DG editor(s) Statistics withConfidence - Confidence Intervals and Statistical GuidelinesLondon British Medical Journal 199550ndash63
NICE 2008National Institute for Health and Care ExcellenceOsteoarthritis The care and management of osteoarthritisin adults 2008 wwwniceorgukCG059fullguideline
Nuumlesch 2010Nuumlesch E Trelle S Reichenbach S Rutjes AW TschannenB Altman DG et al Small study effects in meta-analysesof osteoarthritis trials meta-epidemiological study BMJ2010341c3515 [DOI 101136bmjc3515]
PACT 2014
Prescribing and Primary Care team Health and SocialCare Information Centre Prescrption Cost AnalysisEngland 2013 Health and Social Care Information CentreThe NHS Information Centre 2014298ndash301 [ISBN978ndash1ndash78386ndash089ndash0]
35Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Quane 1998
Quane PA Graham GG Zeigler JB Pharmacology ofbenzydamine Inflammopharmacology 19986(2)95ndash107
RevMan 2014The Nordic Cochrane Centre The Cochrane CollaborationReview Manager (RevMan) 53 Copenhagen The NordicCochrane Centre The Cochrane Collaboration 2014
Roth 2012Roth SH Fuller P Pooled safety analysis of diclofenacsodium topical solution 15 (ww) in the treatment ofosteoarthritis in patients aged 75 years or older ClinicalInterventions in Aging 20127127ndash37 [DOI 102147CIAS30884]
Simon 2009
Simon LS Grierson LM Naseer Z Bookman AA ZevShainhouse J Efficacy and safety of topical diclofenaccontaining dimethyl sulfoxide (DMSO) compared withthose of topical placebo DMSO vehicle and oral diclofenacfor knee osteoarthritis Pain 2009143(3)238ndash45 [DOI101016jpain200903008]
Tramer 1997
Tramer MR Reynolds DJ Moore RA McQuay HJ Impact
of covert duplicate publication on meta-analysis a casestudy BMJ 1997315(7109)635ndash40
Tugwell 2004Tugwell PS Wells GA Shainhouse JZ Equivalence studyof a topical diclofenac solution (pennsaid) compared withoral diclofenac in symptomatic treatment of osteoarthritisof the knee a randomized controlled trial Journal ofRheumatology 200431(10)2002ndash12
Zimmerman 1995
Zimmerman J Siguencia J Tsvang E Upper gastrointestinalhemorrhage associated with cutaneous application ofdiclofenac gel American Journal of Gastroenterology 199590(11)2032ndash4 [PUBMED 7485017]
References to other published versions of this review
Massey 2010
Massey T Derry S Moore RA McQuay HJ TopicalNSAIDs for acute pain in adults Cochrane Databaseof Systematic Reviews 2010 Issue 6 [DOI 10100214651858CD007402pub2]
lowast Indicates the major publication for the study
36Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
C H A R A C T E R I S T I C S O F S T U D I E S
Characteristics of included studies [ordered by study ID]
Airaksinen 1993
Methods R DB PC parallel groupsGel applied to the painful area twice daily for 7 daysAssessment at baseline 3 7 days
Participants Minor soft tissue injuries (lt 7 days)N = 56M 45 F 11Age not reportedMean baseline pain at rest 25-26 mm
Interventions Ketoprofen gel 2 x 5 g (125 mg) daily n = 29Placebo gel n = 27Rescue medication paracetamol 500 mgNo other treatment allowed
Outcomes PGE 5-point scale but reported as ldquoimprovedrdquo or ldquosame or worserdquo (responder = ldquoim-provedrdquo)Improvement in pain with movement 100 mm VAS reported as group meanAdverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Unclear risk No mention of early withdrawals ormethod of imputation
Size High risk lt 50 participants per treatment group
37Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Aoki 1984
Methods R DB PC AC parallel groupsGel applied to affected area 3 or 4 times daily with no occlusion for 7 daysAssessment at baseline 3 7 days
Participants Acute orthopaedic trauma (contusion distortion fracture lt 7 days)N = 252 (203 analysed for efficacy)M 98 F 105Age range 8 to 86 years 13 younger than 20 yearsBaseline pain mild in 35Exclusions 23 protocol violations 26 reasons ldquonot relatedrdquo to drug Equally distributedbetween groups
Interventions Piroxicam gel 05 1 g 3 to 4 x daily n = 84Indomethacin gel 1 1 g 3 to 4 x daily n = 84Placebo gel n = 84No other medication or initiation of physical therapy allowed
Outcomes PGE 5-point scale (responder = ldquobetterrdquo and ldquomuch betterrdquo)Adverse eventsWithdrawals and exclusions
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Low risk ldquokey code sealed until end of studyrdquo
Blinding (performance bias and detectionbias)All outcomes
Low risk Gels in ldquoidentical tubesrdquo
Incomplete outcome data (attrition bias)All outcomes
Unclear risk gt 10 withdrawals ldquounrelated to treat-mentrdquo and for ldquoprotocol violationsrdquo Nofurther details but no significant differ-ences between groups
Size Unclear risk 50 to 200 participants per treatment group
38Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Auclair 1989
Methods R DB PC parallel groupsGel massaged into skin over affected heel 3 times daily after cleaning with soap and waterfor up to 21 daysAssessment at baseline 7 21 days
Participants Acute Achilles heel tendinitis (not associated with continuous pain at rest or gt 1 monthhistory)N = 243 (227 analysed for efficacy)MF not reportedMean age 29 yearsBaseline pain ~ 10 had lt 26 mm on palpation of tendon ~ 30 had mild or no painon dorsiflexion of footExclusions failure to meet inclusion criteria major protocol violations failure to takestudy medication for full duration
Interventions Niflumic acid gel 25 3 x 5 g daily n = 117Placebo gel n = 110No other analgesics and anti-inflammatories physiotherapy or supportive measures al-lowed
Outcomes PGE 5-point scale (responder = ldquogoodrdquo or ldquovery goodrdquo)Pain improved or disappeared on dorsiflexionAdverse eventsWithdrawals and exclusions
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Unclear risk 10 excluded for ldquofailing to meet entry cri-teria and protocol violationsrdquo No furtherdetails
Size Unclear risk 50 to 200 participants per treatment group
39Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Billigmann 1996
Methods R DB PC parallel groupsGel applied 3 times daily with rubbingAssessed at baseline 3 5 7 days
Participants Distortion of ankle jointN = 160M and FAge 18+ yearsBaseline pain not reported
Interventions Ibuprofen microgel 5 3 x 10 cm (= 200 mg) daily n = 80Placebo gel n = 80
Outcomes Pain with movement VAS (responder = decreased by 20)Complete remissionAdverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Unclear risk Insufficient data to assess
Size Unclear risk 50 to 200 participants per treatment group
Campbell 1994
Methods R DB PC parallel groupsCream applied 4 times daily for 7 days (up to 14 days optional)Self assessed using daily diary for 7 days and up to 14 days
Participants Acute ankle sprain (lt 24 hours no fracture)N = 100 (51 analysed)M 33 F 18Mean age 29 years
40Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Campbell 1994 (Continued)
Baseline pain at rest gt 35 mm on walking 80 mmExclusions did not return diaries protocol exclusions (25 ibuprofen 24 placebo)
Interventions Ibuprofen cream 5 (Proflex) 4 x 4rdquo (10 cm) daily n = 26Placebo cream n = 25Advised to use rest and regular icing for 48 hours then walking and exerciseRescue medication paracetamol
Outcomes Improvement in walking ability 4-point scale (responder = ldquoimprovementrdquo)Pain on walking 100 mm VAS (mean data)Withdrawals and exclusions
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Low risk Randomisation carried out by sponsorTubes dispensed by hospital pharmacy whoheld the codes
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical creamrdquo
Incomplete outcome data (attrition bias)All outcomes
High risk gt 40 lost to follow-up Approximatelyequal between groups
Size High risk lt 50 participants per treatment arm as anal-ysed
Chatterjee 1977
Methods R DB PC parallel groupsCream applied to site of injury 3 times daily for 6 daysAssessment at baseline 2 6 days
Participants Soft tissue injuries (recent)N = 51MF not reportedAge not reportedBaseline pain on passive movement moderate or severe in all but 3 participants
41Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Chatterjee 1977 (Continued)
Interventions Benzydamine HCl cream 3 3 x daily n = 25Placebo cream n = 25(5 active 6 placebo participants also received ultrasound)No other topical agent allowed
Outcomes Pain on passive movement 4-point scale (responder = ldquoabsentrdquo or ldquoslightrdquo)Tenderness with pressure 4-point scale (responder = ldquoabsentrdquo or ldquoslightrdquo)Adverse eventsWithdrawals and exclusions
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk ldquopredetermined randomised schedulerdquo
Allocation concealment (selection bias) Low risk Sealed copy of schedule held by investiga-tor and duplicate copy kept by clinical trialco-ordinator Looked at only in event ofadverse reaction (not necessary)
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoindistinguishable in appearance andconsistencyrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk 2 lost to follow-up no withdrawals dueto adverse events Responder analysis
Size Unclear risk lt 50 participants per treatment arm
Costantino 2011
Methods R DB PC multicentre parallel groupPlaster applied daily for 7 daysPI assessment daily overall treatment efficacy and tolerability assessed at 3 and 7 days
Participants Grade I or II ankle sprain (lt 48 hours) with lateral external ligament involvementPI on movement ge 50100 oedema ge 20 mm difference between anklesN = 430M 249 F 175 (for analysis)Mean age 35 yearsBaseline PI on movement 72100 (SD 12)
42Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Costantino 2011 (Continued)
Interventions DHEPhep n = 142DHEP n = 146Placebo n = 142Rescue medication paracetamol to maximum 3 g dailyNo other treatments allowed
Outcomes Mean change in PI on movement from baseline to 3 daysMean reduction in oedema at 3 daysTolerability at 3 daysRescue medicationAdverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk Computer-generated
Allocation concealment (selection bias) Unclear risk ldquosealed envelopesrdquo
Blinding (performance bias and detectionbias)All outcomes
Low risk Shape colour size and application methodidentical for all plasters
Incomplete outcome data (attrition bias)All outcomes
Unclear risk Very few withdrawals but used LOCF
Size Unclear risk 50 to 200 participants per treatment arm
Coudreuse 2010
Methods R DB PC multicentre parallel groupPlaster applied daily for 7 daysPI assessed twice daily over 3 days then day 7 Overall treatment efficacy and tolerabilityassessed on days 3 7
Participants Ankle sprain (lt 48 hours) with lateral external ligament involvementPI on movement ge 50100 oedema ge 20 mm difference between anklesN = 240 (233 for analysis)M 148 F 86 6 unknown
Interventions DHEPhep n = 120Placebo n = 120Rescue medication paracetamol to maximum 4 g daily
43Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Coudreuse 2010 (Continued)
No other treatments allowed
Outcomes Global efficacy at 7 days 4-point scale (responder = ldquoexcellentrdquo or ldquogoodrdquo)Mean change in PI on movement at 6 hours and 7 daysMean change in oedema at 3 and 7 daysTolerabilityAdverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk Computer-generated
Allocation concealment (selection bias) Unclear risk ldquosealed envelopesrdquo
Blinding (performance bias and detectionbias)All outcomes
Low risk Appearance and odour identical for all plas-ters
Incomplete outcome data (attrition bias)All outcomes
Low risk ITT analysis lt 5 excluded for missingdata equal between groups
Size Unclear risk 50 to 200 participants per treatment arm
Curioni 1985
Methods R DB PC AC parallel groupsGel rubbed into affected area until absorbed twice daily for 10 daysAssessed at baseline and daily to 10 days
Participants Acute soft tissue injuriesN = 60M 33 F 27Median age 33 yearsBaseline pain not given
Interventions Ibuproxam gel 10 n = 20Ketoprofen gel n = 20Etofenamate gel n = 20
Outcomes PGE 4-point scale (ldquogoodrdquo or ldquoexcellentrdquo)Resolution of symptomsAdverse events
44Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Curioni 1985 (Continued)
Withdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk Medication supplied in identical tubes
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
Diebshlag 1990
Methods R DB PC AC parallel groupsGel applied 3 times daily without occlusion for 14 daysAssessment at baseline 2 3 4 8 15 days
Participants Ankle sprain (lt 24 hours)N = 37M 24 F 13Mean age 28 yearsBaseline pain slight to moderate
Interventions Ketorolac gel 2 3 x 3 g daily n = 13Etofenamate gel 5 3 x 3 g daily n = 12Placebo gel n = 12Rescue medication paracetamolNo other analgesic or anti-inflammatory medication ice packs or physiotherapy allowed
Outcomes Reduction in PI 100 mm VAS and 4-point scale (responder = ldquoimprovedrdquo)Adverse eventsWithdrawals and exclusions
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
45Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Diebshlag 1990 (Continued)
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Low risk ldquoMedication assignment supplied in asealed enveloperdquo Opened only if seriousparticipant event necessitation treatmentdisclosure occurred (not necessary)
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical appearancerdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
Dreiser 1988
Methods R DB PC parallel groupsCream applied 3 times dailyAssessment at baseline 7 days
Participants Acute tendinitis (lt 1 month)N = 64M 35 F 25Mean age 36 yearsBaseline spontaneous pain ge 60 mm
Interventions Ibuprofen cream 5 3 x 4 cm daily n = 32 (3 x 10 cm for large joints)Placebo cream n = 32No other topical systemic or physical treatment allowed
Outcomes PGE scale not reported (responder = ldquoimprovementrdquo or ldquocomplete relief rdquo)Improvement in pain VAS (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35Oxford Validity Score 1016
Risk of bias
Bias Authorsrsquo judgement Support for judgement
46Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Dreiser 1988 (Continued)
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Low risk Missing participants added back usingBOCF
Size High risk lt 50 participants per treatment arm
Dreiser 1989
Methods R DB PC parallel groupsGel applied twice daily to affected area with light massage then covered with standardcompressAssessed at baseline 3 7 days
Participants Uncomplicated recent ankle sprainN = 60M 36 F 24Mean age 33 yearsMean baseline pain 54 mm
Interventions Ketoprofen gel 25 2 x 5 cm daily n = 30Placebo gel n = 30No concomitant therapy other than simple oral analgesia allowed
Outcomes PGE 3-point scale (responder = ldquobetterrdquo)Improvement in pain VAS (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk ldquodrawing lotsrdquo
Allocation concealment (selection bias) Unclear risk Not described
47Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Dreiser 1989 (Continued)
Blinding (performance bias and detectionbias)All outcomes
Low risk Treatments ldquoidentical in every way exceptthat placebo did not contain active princi-plerdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participant per treatment arm
Dreiser 1990
Methods R DB PC parallel groupsGel lightly massaged into skin over affected area 3 times daily then covered with standardcompressAssessed at baseline 3 7 days
Participants Uncomplicated ankle sprain (lt 4 days)N = 60 (59 analysed)M 29 F 29 (not stated for 1 participant)Mean age 33 yearsBaseline pain ge moderately severeExclusions 1 participant had only moderate pain at baseline
Interventions Niflumic acid gel 25 3 x 5 g daily n = 30Placebo gel n = 30Concomitant treatment with systemic NSAIDs local therapies or physiotherapy werenot allowed
Outcomes PGE 4-point scale (responder = ldquocuredrdquo or ldquoimprovedrdquo)Improvement in pain VAS (mean data)Adverse eventsWithdrawals and exclusions
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
48Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Dreiser 1990 (Continued)
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
Dreiser 1994
Methods R DB PC parallel groupsPatch applied twice dailyAssessed at baseline 3 7 days
Participants Traumatic ankle sprain (lt 2 days)N = 131M 84 F 47Mean age 34 yearsBaseline pain ge 50 mm
Interventions Flurbiprofen patch 2 x 40 mg daily n = 65Placebo patch n = 66Rescue medication paracetamol Ice or light restraint allowedExclusions 1 from flurbiprofen group for protocol violation
Outcomes PGE 4-point scale (responder = ldquogoodrdquo or ldquovery goodrdquo)Improvement in pain VAS (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk Placebo patch was ldquonon-medicated (butotherwise identical)rdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
49Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Fioravanti 1999
Methods R DB AC parallel groupsGel lightly massaged into skin 3 times daily and kept dry for 6 to 8 hoursAssessed at baseline 3 10 days
Participants Peri and extra-articular inflammatory diseasesN = 100M 32 F 68Mean age 49 yearsBaseline spontaneous pain ge 40 mm
Interventions DHEP lecithin gel 3 x 5 g (= 65 mg) daily n = 50DHEP gel 3 x 5 g (= 65 mg) daily n = 50
Outcomes PGE 4-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo)Pain on movement meanAdverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size Unclear risk 50 participants per treatment arm
Fujimaki 1985
Methods R DB PC AC parallel groupsGel applied to affected area 3 or 4 times daily with no occlusion for up to 14 daysAssessed at baseline 7 14 days
Participants Muscle pain or inflammation in neck shoulder back chest and upper and lower ex-tremities or a combinationN = 271 (247 analysed)M 97 F 149
50Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Fujimaki 1985 (Continued)
Age lt 20 to 89 yearsBaseline pain mostly mild to moderateExclusions 24 due to protocol violations loss to follow-up
Interventions Piroxicam gel 05 1 g 3 to 4 x daily n = 92Indomethacin gel 1 1 g 3 to 4 x daily n = 90Placebo gel n = 89No concomitant oral or topical analgesic or anti-inflammatory medication allowed Nophysical therapy initiated after start of study
Outcomes PGE 5-point scale (responder = ldquobetterrdquo or ldquomuch betterrdquo)Physician rated improvement 5-point scale (responder = ldquomarked improvementrdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total =45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Low risk Cartons numbered randomly and numbersheld in a key code until study completion
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical tubesrdquo packed in numbered car-ton Gel bases slightly different in appear-ance so dispensing physician did not haveaccess to them
Incomplete outcome data (attrition bias)All outcomes
Unclear risk ~ 8 excluded from analysis for unknownreasons or lost to or inadequate follow-upApproximately equal between groups
Size Unclear risk 50 to 200 participants per treatment arm
Gallacchi 1990
Methods R DB AC parallel groupsGel applied to affected area 4 times daily with light massage for 14 daysAssessment at baseline 7 14 days
Participants Painful inflammatory conditionsN = 50M 20 F 30Mean age 50 years
51Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Gallacchi 1990 (Continued)
Baseline pain ge moderate severity
Interventions Diclofenac gel 1 2 g 4 x daily n = 25 (Flector)Diclofenac sodium 1 2 g 4 x daily n = 25 (Voltaren Emugel)No other medication that could interfere with test drugs allowed
Outcomes PGE 5-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo)Improvement in pain on pressure 4-point scale (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
Gonzaacutelez de Vega 2013
Methods R DB (for gels) SB (for ointment) AC multicentre parallel groups2 g (~ 6 cm of gel) applied over injured area x 3 daily for 14 daysAssessments at 0 4 7 14 days of treatment and day 42 follow-up
Participants Grade I or II ankle sprain (lt 24 hours) with lateral external ligament involvement PIon weight bearing ge 30100N = 449M = 308 F = 112 (for analysis)Mean age 28 years
Interventions Traumeel gel 3 x 2 g daily n = 140Traumeel ointment n = 143 (not analysed in this review)Diclofenac gel 1 3 x 2 g daily n = 137Rescue medication paracetamol to maximum 2 g daily
52Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Gonzaacutelez de Vega 2013 (Continued)
Outcomes Pain-free on day 7Global efficacy 5-point scale (responder = ldquogoodrdquo and ldquovery goodrdquo) on day 14Normal function on day 14 yes or no (responder = ldquoyesrdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 (gels) W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk Computer-generated
Allocation concealment (selection bias) Low risk Central allocation kits assigned in orderreceived and used envelopes (no furtherdetails)
Blinding (performance bias and detectionbias)All outcomes
Low risk For gel comparison ldquoidentical containersrdquo
Incomplete outcome data (attrition bias)All outcomes
Unclear risk Used LOCF Most withdrawals due to earlyrecovery (within 14 days) approximatelyequal between groups
Size Unclear risk 50 to 200 participants per treatment arm
Governali 1995
Methods R DB AC parallel groupsGel or cream applied 3 times daily for up to 14 daysAssessed at baseline 7 14 days
Participants Soft tissue injuries + 2 fracturesN = 30M = 21 F = 9Median age 38 yearsMean baseline pain on movement moderate to severe (28 scale 0 to 4)
Interventions Ketoprofen gel 5 3 x 2 to 3 g daily n = 15Ketoprofen cream 1 3 x 2 to 3 g daily n = 15
Outcomes PGE 5-point scale (responder = ldquogoodrdquo and ldquoexcellentrdquo)Adverse eventsWithdrawals
53Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Governali 1995 (Continued)
Notes Oxford Quality Score R1 DB1 W1 Total =35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Treatments were given in identical tubesand measurements made by blinded ob-servers but one was a cream and the othera gel
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
Gualdi 1987
Methods R DB AC parallel groupsGel applied twice daily for 10 daysAssessed at baseline 4 7 10 days
Participants Soft tissue injuriesN = 60M = 37 F = 23Mean age 32 years (range 13 to 78)Mean baseline pain on movement moderate to severe (22 scale 0 to 3)
Interventions Flunoxaprofen gel 2 x 3 to 5 cm daily n = 30Ketoprofen gel 2 x 3 to 5 cm daily n = 30
Outcomes Improvement in pain on pressure (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
54Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Gualdi 1987 (Continued)
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Unclear risk Insufficient data
Size High risk lt 50 participants per treatment arm
Haig 1986
Methods R DB PC parallel groupsCream applied lightly to affected area 6 times daily for 6 daysAssessed at baseline 2 4 6 days
Participants Soft tissue injuries (lt 24 hours)N = 43MF not reportedAge not reportedBaseline pain not reported
Interventions Benzydamine cream 3 6 x daily n = 21Placebo cream n = 22
Outcomes Pain on movement 4-point scale (responder = ldquoimprovedrdquo)Adverse events
Notes Oxford Quality Score R1 DB2 W0 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not reported
Allocation concealment (selection bias) Unclear risk Not reported
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquomatching placebordquo
55Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Haig 1986 (Continued)
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants apparently included inanalysis
Size High risk lt 50 participants per treatment arm
Hoffmann 2012
Methods R DB PC multicentre parallel groupPlaster applied x 1 daily (minimum of 20 consecutive hours of application per day) for14 daysPI assessed daily overall efficacy assessed at 7 and 14 days
Participants Unilateral mild to moderate muscle contusion of upper or lower limb (lt 72 hours) PIge 50100 superficial haematoma le 10 x 14 cm at injured siteN = 354M 126 F 228Mean age 39 years
Interventions DHEPhep plaster x 1 daily n = 121DHEP plaster x 1 daily n = 115Placebo plaster n = 118Rescue medication paracetamol 500 mg (no limit reported)
Outcomes PGE 5-point scale (responder = excellent or good) at 14 daysMean reduction in PI at 3 and 8 daysRescue medicationAdverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Method not reported
Allocation concealment (selection bias) Unclear risk Method not reported ldquoMedication packedand labelled in order to render all partici-pants and personnel fully blinded to treat-ment administeredrdquo
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoIndistinguishable regarding appearanceshape colour size and odourrdquo
56Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Hoffmann 2012 (Continued)
Incomplete outcome data (attrition bias)All outcomes
Unclear risk No responder analysis reported and impu-tation method not mentioned
Size Unclear risk 50 to 200 participants per treatment arm
Hofman 2000
Methods R DB AC parallel groupsGel applied to affected region 4 times daily with gentle massageAssessed at baseline 8 days in clinic and daily diary
Participants Soft tissue articular pain (le 15 days)N = 142M 19 F 123Mean age 57 yearsMean baseline PI moderate to severe
Interventions Diclofenac sodium gel 1 4 x 2 cm daily n = 69Lysine clonixinate gel 5 4 x 2 cm daily n = 73(2 cm = 225 mg)No other analgesic local treatment (including immobilisation bandaging) or acupunc-tureRescue mediation allowed after 2 applications if needed
Outcomes PGE 3-point scale (ldquogoodrdquo)PI participant diary (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk Diclofenac gel repackaged to maintain DBwith lysine clonixinate gel Minor differ-ences between gels only apparent when di-rectly compared
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
57Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Hofman 2000 (Continued)
Size Unclear risk 50 to 200 participants per treatment arm
Hosie 1993
Methods R DB AC parallel groupsFoam (approximately the size of a golf ball) applied and 1 tablet taken 3 times daily for7 days and up to 14 daysAssessed at baseline 7 14 (if necessary) days
Participants Acute lower back injury (lt 1 month)N = 287 (261 analysed for efficacy)M 151 F 136Mean age 37 years (range 18 to 63)Most participants had moderate to severe pain on movement one had noneExclusions 25 lost to follow up one assessed at 14 days but not 7 days
Interventions Felbinac foam 3 3 x 2 g daily + placebo tablets 3 x 1 daily n = 140 (127 analysed forefficacy)Ibuprofen tablets 3 x 400 mg daily + placebo foam 3 x 2 g daily n = 147 (134 analysedfor efficacy but one had no pain at baseline)No other oral injectable or topical analgesic or anti-inflammatory medication Ongoingphysiotherapy to continue without change
Outcomes Pain on movement 5-point scale (responder = ldquononerdquo or ldquomildrdquo)Spontaneous pain 5-point scale (responder = ldquononerdquo or ldquomildrdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquodouble dummyrdquo
Incomplete outcome data (attrition bias)All outcomes
Unclear risk Imputation method not mentioned
Size Unclear risk 50 to 200 participants per treatment arm
58Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Jenoure 1997
Methods R DB PC parallel groupsPlaster applied to skin over affected area twice daily and kept in place with an elasticbandageAssessed at baseline 7 14 days and after further 14 days without treatment
Participants Humero-radial epicondyl pain (tendinopathic) - nearly all tennis elbowN = 85M 54 F 31Mean age 45 yearsBaseline pain ldquomildrdquo in ~ 10 of placebo group and 29 of active group
Interventions DHEP plaster (Tissugel) x 2 daily n = 44Placebo plaster x 2 daily n = 41
Outcomes Pain on pressure 5-point scale (responder = ldquononerdquo or ldquomildrdquo)Spontaneous pain 5-point scale (responder = ldquono painrdquo)Adverse events
Notes Oxford Quality Score R1 DB2 W0 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not reported
Allocation concealment (selection bias) Unclear risk Not reported
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical characteristicsrdquo
Incomplete outcome data (attrition bias)All outcomes
Unclear risk No useable data
Size High risk lt 50 participants per treatment arm
Joussellin 2003
Methods R DB PC parallel groupsPlaster applied to skin over affected area once dailyAssessed at baseline 1 2 3 7 days
Participants Ankle sprain (lt 48 hours)N = 134M 72 F 62Age range 18 to 65 years
59Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Joussellin 2003 (Continued)
Baseline spontaneous pain ge 50 mm
Interventions DHEP plaster (Flector Tissugel 1) x 1 daily n = 68Placebo plaster x 1 daily n = 66Rescue medication paracetamol
Outcomes PGE 4-point scale (responder = ldquoexcellentrdquo)Pain on movement VAS (mean)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoIdentical in size appearance and usedsame formula as active patch without ac-tive ingredientrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in responder anal-ysis
Size Unclear risk 50 to 200 participants per treatment arm
Julien 1989
Methods R DB PC parallel groupsGel applied to affected area twice daily with light massageAssessed at baseline 3 7 days in clinic and daily diary
Participants TendinitisN = 60M 29 F 31Mean age 41 yearsBaseline pain gt 50 mm
Interventions Ketoprofen gel 25 2 x 5 cm (= 50 mg) daily n = 30Placebo gel 2 x 5 cm daily n = 30No concomitant therapy other than simple analgesia
60Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Julien 1989 (Continued)
Outcomes PGE 4-point scale (responder = ldquoimprovedrdquo or ldquorecoveredrdquo)Pain on movement 4-point scale (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB1 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk Random numbers table
Allocation concealment (selection bias) Unclear risk Randomisation code supplied by Menarinilaboratories remote from allocation
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in responder anal-ysis
Size High risk lt 50 participants per treatment arm
Klainguti 2010
Methods R DB PC multicentre parallel groupPlaster applied x 1 daily (minimum 20 consecutive hours of application daily) for 10daysPI assessed daily Overall treatment efficacy and tolerability assessed at days 3 and 7
Participants Unilateral mild to moderate muscle contusion or strain of upper or lower limb (lt 72hours) superficial haematoma le 140 cm2 PI ge 40100N = 185M 90 F 95Mean age 39 years
Interventions DHEPhep plaster x 1 daily n = 65DHEP plaster x 1 daily n = 61Placebo plaster x 1 daily n = 59
Outcomes Overall treatment efficacy at 3 days 5-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo)Resolution of haematoma at 10 days yes or noRescue medicationAdverse eventsWithdrawals
61Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Klainguti 2010 (Continued)
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk Computer-generated
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoIdentical in size appearance and odourrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk Withdrawals lt 5 All participants in-cluded in analysis
Size Unclear risk 50 to 200 participants per treatment arm
Kockelbergh 1985
Methods R DB PC parallel groupsGel applied twice dailyAssessed at baseline 3 7 days
Participants Acute soft tissue trauma (lt 24 hours)N = 74M 60 F 14Mean age 27 yearsBaseline pain gt 65 mm
Interventions Ketoprofen gel 25 2 x 5 cm (= 15 mg) daily n = 38Placebo gel 2 x 5 cm daily n = 36No concomitant treatmentRescue medication glafenine
Outcomes PGE 3-point scale (responder = ldquogoodrdquo)Spontaneous pain 100 mm VAS (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
62Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Kockelbergh 1985 (Continued)
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in responder anal-ysis
Size High risk lt 50 participants per treatment arm
Kuehl 2011
Methods R DB PC multicentre parallel groupsPlaster applied x 2 daily for 14 days or until resolutionPI assessed daily Overall treatment efficacy assessed at 14 days
Participants Mild or moderate sprain strain or contusion (lt 7 days) PI ge 510N = 418M 206 F 212Mean age 39 years
Interventions DHEP patch x 2 daily n = 207Placebo patch x 2 daily n = 211Rescue medication no analgesic allowed (or icewrapping) use = discontinuation
Outcomes Resolution of injury (VAS le 210)Overall tolerability 5-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk Computer-generated
Allocation concealment (selection bias) Unclear risk Not described
63Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Kuehl 2011 (Continued)
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical in appearance with same contentexcept diclofenacrdquo
Incomplete outcome data (attrition bias)All outcomes
High risk Used LOCF high withdrawal rate (43)Did not report all efficacy outcomes mea-sured
Size Low risk gt 200 participants per treatment arm
Li 2013
Methods R DB PC multicentre parallel groupsPlaster applied x 2 daily for 7 daysPI assessed daily overall treatment efficacy and tolerability assessed at 7 days
Participants Mild or moderate ankle or knee sprain muscle strain or contusion (lt 72 hours) PI ge
50100N = 384M 144 F 240Mean age 42 years
Interventions DHEP plaster x 2 daily n = 192Placebo plaster x 2 daily n = 192Rescue medication paracetamol to maximum 2 g daily
Outcomes ge 50 reduction in PI at 7 daysOverall treatment efficacy 5-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo)Overall tolerability 5-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo)Rescue medicationAdverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk Computer-generated
Allocation concealment (selection bias) Unclear risk Method not fully described ldquosealed en-velopesrdquo ldquoconcealed from investigators andparticipantsrdquo
64Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Li 2013 (Continued)
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical in texture size color and odorrdquo
Incomplete outcome data (attrition bias)All outcomes
Unclear risk Withdrawals 2 Methods states LOCFbut appears to describe BOCF for with-drawals
Size Unclear risk 50 to 200 participants per treatment arm
Linde 1985
Methods R DB PC parallel groupsCream applied x 3 daily for 5 days with elastic support for the first 3 daysAssessed at baseline 4 8 days
Participants Sprained ankle (lt 24 hours)N = 100M 58 F 42Mean age 28 yearsBaseline pain all participants had ldquowalking painrdquo
Interventions Benzydamine 3 cream x 3 daily n = 50Placebo gel x 3 daily n = 50
Outcomes Pain on movement responder = ldquofree of walking painrdquoAdverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35Paper describes a benzydamine cream and a placebo gel
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described Paper describes a benzy-damine cream and a placebo gel
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in responder anal-ysis
65Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Linde 1985 (Continued)
Size Unclear risk 50 participants per treatment arm
Machen 2002
Methods R DB PC parallel groupsGel gently (ldquominimal rubrdquo not vigorously) massaged into skin over affected site untilabsorbed 3 times daily until symptoms disappeared or for maximum of 7 daysAssessment at baseline and once daily using diary cards to 7 days
Participants Soft tissue injury (lt 2 weeks and untreated)N = 85 (81 analysed)M 42 F 39Mean age 41 yearsBaseline pain gt 50 mm4 placebo participants lost to follow-up
Interventions Ibuprofen gel 5 x 3 daily n = 40Placebo gel x 3 daily n = 41Initiation of other medication or physiotherapy not allowed during study
Outcomes PGE 5-point scale (responder = ldquomarked improvementrdquo or ldquocomplete clearancerdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk Gels had similar physical characteristicsand were supplied in identical tubes
Incomplete outcome data (attrition bias)All outcomes
Low risk lt 5 missing participants Remaining par-ticipants included in responder analysis
Size High risk lt 50 participants per treatment arm
66Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Mahler 2003
Methods R DB AC parallel groupsGel applied with gentle massage to affected area 3 times daily without occlusion for 10daysAssessed at baseline 3 10 days in clinic and daily patient diary
Participants First-degree ankle or knee sprains first-degree muscle strains and mild-to-moderatecontusionsn = 100M 69 F 31Mean age 32 yearsMean baseline pain with activity ge 65 mm
Interventions DHEP lethicin gel 3 x 5 g (= 65 mg) daily n = 52DHEP gel 3 x 5 g (= 65 mg) daily n = 48All participants treated with ice at site of inflammation for first 48 hours but no immo-bilisation allowedRescue medication paracetamol 500 mg if strictly necessary
Outcomes PGE 4-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo)Pain on movement 100-mm VAS (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 5
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk ldquoComputer-generated randomization listrdquo
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk Pharmaceutically inert colouring agentsadded to reference formulation so that gelswere indistinguishable
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in responder anal-ysis
Size High risk lt 50 participants in each treatment arm
67Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Maziegraveres 2005a
Methods R DB PC parallel groupsNew patch applied directly to skin over painful area each morningAssessed at baseline 3 7 14 days
Participants Symptomatic tendonitis in upper or lower limbs not requiring surgery (le 15 days)N = 172M 72 F 100Mean age 46 yearsBaseline pain with activity ge 40 mm
Interventions Ketoprofen patch 100 mg x 1 daily n = 87Placebo patch x 1 daily n = 85No analgesic or steroid by any route or other topical medication or physical therapyallowedRescue medication permitted but not within 12 hours of assessment
Outcomes PGE 4-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk ldquocomputer generated global randomizationcoderdquo
Allocation concealment (selection bias) Low risk ldquoThe randomization list and code en-velopes were prepared by the company ap-pointed for clinical supplies packaging Therandom code was disclosed only after studycompletion and database closurerdquo
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquothe same indistinguishable patch with noingredientrdquo
Incomplete outcome data (attrition bias)All outcomes
High risk LOCF withdrawals 12 All participantsincluded in analysis
Size Unclear risk 50 to 200 participants per treatment arm
68Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Maziegraveres 2005b
Methods R DB PC parallel groupsNew patch applied directly to skin over painful area each morningAssessed at baseline 3 7 14 days
Participants Painful benign ankle sprain (le 48 hours)N = 163M 83 F 80Mean age 37 yearsBaseline spontaneous pain ge 50 mm
Interventions Ketoprofen patch 100 mg x 1 daily n = 81Placebo patch x 1 daily n = 82No analgesic or steroid by any route or other topical medication or physical therapyallowedRescue medication permitted but not within 12 hours of assessment
Outcomes PGE 4-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk ldquocomputer-generated global randomiza-tion coderdquo
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk The same transdermal patch with no activeingredient
Incomplete outcome data (attrition bias)All outcomes
Unclear risk LOCF withdrawals 5 to 6 All partic-ipants included in analysis
Size Unclear risk 50 to 200 participants per treatment arm
McLatchie 1989
Methods R DB PC parallel groupsGel applied to injured site 3 times daily for 7 daysAssessment at baseline 4 7 days at clinic daily patient diary
Participants Acute soft tissue injury (lt 48 hours)N = 231
69Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
McLatchie 1989 (Continued)
M 143 F 88Mean age 33 yearsBaseline pain moderate to severe
Interventions Felbinac gel 3 3 x 3 cm daily n = 118Placebo gel 3 x 3 cm daily n = 113Rescue medication paracetamol
Outcomes Patient diary mean changeAdverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquotubes identical in all aspectsrdquo
Incomplete outcome data (attrition bias)All outcomes
Unclear risk No usable efficacy data
Size Unclear risk 50 to 200 participants per treatment arm
Morris 1991
Methods R DB PC parallel groupsGel applied to site of injury 3 times daily for 7 daysAssessed at baseline 7 days at clinic and daily patient diary
Participants Acute soft tissue injury (lt 3 days)N = 100 (84 analysed for efficacy)M 70 F 14Mean age 25 yearsBaseline pain moderate to severeExclusions 1 participant in placebo group lost to follow-up 15 protocol violations
Interventions Felbinac gel 3 3 x 1 cm daily n = 41Placebo gel n = 43Ice joint immobilisation bandaging and compression allowed
70Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Morris 1991 (Continued)
No concomitant oral NSAID occlusive dressing physiotherapy or liniments allowedRescue medication paracetamol
Outcomes PGE 5-point scale (responder = ldquogoodrdquo and ldquovery goodrdquo)Change in PI patient diary 10 cm VAS (mean data)Adverse eventsWithdrawals and exclusions
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Low risk ldquoRandomisation was undertaken at theproduction facility and a sealed copy of thelist supplied to the investigator for refer-ence only in defined circumstancesrdquo
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical tubes and outer boxesrdquo ldquoplacebowas a similarly constituted gelrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All relevant participants included in analy-sis
Size High risk lt 50 participants per treatment arm
NCT01255423
Methods R DB PC parallel group studyGel applied 4 x daily
Participants Acute lateral ankle sprain Grade I to II le 12 hoursN = 206Mean (plusmn SD) age 31 years (plusmn 13)M 87 F 119Baseline PI not reported
Interventions Diclofenac sodium gel 1 x 4 daily n = 104Placebo gel n = 102
Outcomes Mean PI on movementAdverse events
71Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT01255423 (Continued)
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Method of blinding not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants accounted for
Size Unclear risk 50 to 200 participants per treatment arm
NCT01272934
Methods R DB PC parallel group studyGel applied 4 x daily
Participants Acute lateral ankle sprain Grade I to II le 12 hoursN = 205M 101 F 104Mean (plusmn SD) age 32 years (plusmn 11)
Baseline PI not reported
Interventions Diclofenac sodium gel 1 x 4 daily n = 102Placebo gel x 4 daily n = 103
Outcomes Mean PI on movementAdverse events
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
72Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT01272934 (Continued)
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Method of blinding not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants accounted for
Size Unclear risk 50 to 200 participants per treatment arm
NCT01272947
Methods R DB PC parallel groupsGel applied 4 x daily
Participants Acute blunt soft tissue injuriescontusions of the limbs lt 3 hoursN = 204M 101 F 103Mean (plusmn SD) age 30 years (plusmn 11)
Baseline PI not reported
Interventions Diclofenac sodium gel 1 x 4 daily n = 104Placebo gel x 4 daily n = 100
Outcomes Mean PI on movementAdverse events
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Method of blinding not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants accounted for
73Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT01272947 (Continued)
Size Unclear risk 50 to 200 participants per treatment arm
Noret 1987
Methods R DB PC parallel groupsGel applied twice daily for 7 daysAssessment at baseline 3 8 days
Participants Minor sports injuries (lt 24 hours)N = 98 (93 analysed)M 71 F 27Mean age 29 yearsBaseline pain gt 60 mm
Interventions Ketoprofen gel 25 2 x 5 cm daily (= 15 mg) n = 48Placebo gel n = 45No other treatment given
Outcomes PGE 4-point scale (responder = ldquogoodrdquo and ldquoexcellentrdquo)Spontaneous pain 100 mm VAS (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Low risk ldquoallocated according to a randomization listand a corresponding code in a sealed enve-loperdquo
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
74Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Parrini 1992
Methods R DB PC parallel groupsFoam (the size of a walnut or a 1-second spray) applied with massage 3 x daily for 7days
Participants Articular trauma strains distortionsN = 169M 94 F 75Mean age 37 yearsMean baseline pain on movement 31 (scale 1 to 4)
Interventions Ketoprofen foam 15 3 x 2 g (= 600 mg) daily n = 83Placebo foam 3 x 2 g n = 86
Outcomes Pain on movement 4-point scale (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB1 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk ldquopatients were randomised according to themethod of random numbersrdquo [translated]
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Unclear risk No usable efficacy data
Size Unclear risk 50 to 200 participants per treatment arm
Picchio 1981
Methods R DB AC parallel groupsCream applied with slight massage until completely absorbed 3 times daily for up to 16daysAssessed at baseline 4 8 12 16 days
Participants Acute sports injuriesN = 40M 24 F 16Mean age 22 years (range 12 to 46)Most participants had mild to moderate baseline pain (12 and 9 with slight pain on
75Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Picchio 1981 (Continued)
movement)
Interventions Ibuprofen gel 10 3 x daily n = 20Ketoprofen gel 1 3 x daily n = 20
Outcomes Pain on movement (responder = ldquononerdquo)Adverse events
Notes Oxford Quality Score R1 DB2 W0 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk ldquotubes were identical in appearancerdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
Predel 2004
Methods R DB PC parallel groupsNew patch applied to injured area twice daily for 7 days Contact of patch with humidityor water to be avoidedAssessment at baseline 3 7 days
Participants Traumatic blunt soft tissue injuries (lt 3 hours no treatment)N = 120M 73 F 47Mean age 32 yearsBaseline pain gt 60 mm
Interventions Diclofenac sodium patch 2 x daily (140 mg per patch) n = 60Placebo patch 2 x daily n = 60NSAIDs analgesics psychotropic agents other topical preparations and bandages notallowed
Outcomes PGE 4-point scale (responder = rdquogoodldquo and rdquoexcellentldquo)Pain on movement 10 cm VAS (mean data)
76Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Predel 2004 (Continued)
Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk rdquocomputer generated block randomisationlist
Allocation concealment (selection bias) Low risk An independent statistician produced ran-domisation list and an independent con-tract research organisation packaged med-ication according to list Nobody else hadaccess to the randomisation list until thedatabase was closed
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoThe placebo patch was visually indistin-guishable from the active patchrdquo To avoidunblinding due to different smell anystudy nurse involved with medication wasnot involved in outcome assessment
Incomplete outcome data (attrition bias)All outcomes
Low risk Withdrawals lt 1 All participants in-cluded in analysis
Size Unclear risk 50 to 200 participants per treatment arm
Predel 2012
Methods R DB PC parallel groupsTreatment for 7 daysAssessed at baseline 1 3 5 and 8 days
Participants Grade I or II ankle sprain with lateral external ligament involvement lt 12 hoursPI on movement ge 50100N = 242M 152 F 90Mean age 32 years
Mean baseline pain on movement = 75100
Interventions Diclofenac gel (Voltaren Emulgel 232) 2 x 5 cm + 1 x 5 cm placebo gel daily n = 80Diclofenac gel (Voltaren Emulgel 232) 3 x 5 cm daily n = 80Placebo gel 3 x 5 cm daily n = 82Rescue medication paracetamol (to maximum 2 g daily)
77Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Predel 2012 (Continued)
No ice or bandages after randomisation
Outcomes ge 50 reduction in PI on movement at 5 daysPGE 5-point scale (responder = ldquogoodrdquo or rsquovery goodldquo)Patient satisfaction 5-point scale (responder = rdquogoodldquo or rsquovery goodrdquo or ldquoexcellentrdquo)Rescue medicationAdverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical in composition appearance tex-ture and smellrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk LOCF but withdrawals lt 3 in all treat-ment arms
Size Unclear risk 50 to 200 participants per treatment arm
Predel 2013a
Methods R DB PC parallel groupsTreatment for 14 daysAssessed at baseline 3 7 10 and 14 days (plusmn 1 day)
Participants Uncomplicated one-sided ankle sprain with swelling ge 12 mm (2 to 18 hours)PI on movement tenderness joint mobility (scale 0 to 3) summed as ge 5 and le 7N = 232M 126 F 106Mean age 29 years
Interventions Diclofenac 4 spray gel 3 x 4 to 5 sprays daily (96 to 120 mg diclofenac sodium) n =118Placebo spray gel 3 x 4 to 5 sprays daily n = 114Rescue medication paracetamol 500 mg (maximum 10 tablets per week)No ice or bandaging allowed
78Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Predel 2013a (Continued)
Outcomes None or slight PI on movement at 3 and 7 daysPGE 4-point scale (responder = ldquovery goodrdquo at 14 days)gt 50 reduction in swelling at 10 daysAdverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk ldquoplacebo (vehicle only no active ingredi-ent)rdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk LOCF for full or early cure Withdrawalslt 2
Size Unclear risk 50 to 200 participants per treatment arm
Predel 2013b
Methods R DB PC parallel groups2 g gel applied with fingertips over affected area and massaged in for 1 minute Treatmentfor 5 daysAssessed at baseline 2 3 5 days
Participants Uncomplicated neck pain originating from cervical joints and accompanying soft tissues(ge 12 hours but lt 3 months) PI ge 50100N = 72M 33 F 39Mean age 34 years
Interventions Diclofenac gel (Voltaren Emulgel) 4 x 2 g daily n = 36Placebo gel 4 x 2 g daily n = 36Rescue medication paracetamol up to 2 g dailyNo concomitant therapies allowed
Outcomes ge 50 decrease in PI on movement after 48 hoursPGE 5-point scale (responder = ldquogoodrdquo or ldquoexcellentrdquo at 5 daysAdverse eventsWithdrawals
79Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Predel 2013b (Continued)
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk Sponsor ldquoproduced randomisation listrdquo
Allocation concealment (selection bias) Low risk Automated remote system
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical in packaging labelling scheduleof administration appearance and odourrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk No missing data
Size High risk lt 50 participants per treatment arm
Ramesh 1983
Methods R DB PC parallel groupsCream applied to painful area and rubbed into skin over a large area for up to 10 daysAssessment at baseline 3 7 10 days
Participants Strains sprains contusions compressionsN = 80M 42 F 38Age 11 to 81 yearsBaseline pain 5 ibuprofen 2 placebo participants had no or slight pain
Interventions Ibuprofen cream 5 3 to 4 x 5 to 10 cm daily n = 40Placebo cream 3 to 4 x 5 to 10 cm daily n = 40Adjuvant therapy was not administered
Outcomes Pain on movement 4-point scale (responder = ldquononerdquo or ldquoslightrdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
80Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Ramesh 1983 (Continued)
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Low risk Randomisation key in sealed envelopeavailable for emergencies but opened onlyafter completion
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical appearance and odourrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
Rowbotham 2003
Methods R DB PC parallel groupsNew patches applied to the affected painful area for 12 consecutive hours twice dailyfor up to 14 daysAssessed at baseline 14 days in clinic and daily patient diary
Participants Minor sports injuries (sprains sprains contusions lt 72 hours)N = 372M 253 F 119Mean age 33 yearsBaseline pain at rest ge 510
Interventions DHEP patch (Flector Tissuegel) 2 x daily (equivalent to diclofenac sodium 140 mg perpatch) n = 191Placebo patch 2 x daily n = 181
Outcomes PGE 5-point scale (responder = ldquogoodrdquo and ldquoexcellentrdquo)Pain resolved lt moderate for 2 daysSpontaneous pain 10 cm VAS (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
81Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Rowbotham 2003 (Continued)
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk ldquoSystegraveme identiquerdquo without diclofenac
Incomplete outcome data (attrition bias)All outcomes
Unclear risk No information about imputation All par-ticipants included in analysis
Size Unclear risk 50 to 200 participants per treatment arm
Russell 1991
Methods R DB PC parallel groupsAffected area washed with soap and water and dried then gel applied and carefullyrubbed into skin 4 times daily for at least 7 daysAssessed at baseline 4 8 15 (if necessary) days at clinic and daily patient diary
Participants Acute soft tissue injuries (recent not recurrent)N = 214 (200 analysed)M 95 F 105Mean age 40 yearsBaseline pain gt 65 mm
Interventions Piroxicam gel 05 4 x 5 mg daily n = 100Placebo gel n = 100No other NSAIDs or analgesic drugs including liniments containing salicylates allowedAncillary therapy at the discretion of the investigator
Outcomes PGE 4-point scale (responder = ldquogoodrdquo and ldquoexcellentrdquo)Spontaneous pain mean reductionAdverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk ldquocomputer generated randomization coderdquo
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical base formulationrdquo
82Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Russell 1991 (Continued)
Incomplete outcome data (attrition bias)All outcomes
High risk High withdrawal rate (8 with piroxicam50 with placebo) No information aboutimputation
Size Unclear risk 50 to 100 participants per treatment group
Saillant 1998
Methods R DB PC parallel groupsTreatment for 7 daysAssessed at baseline 1 2 3 and 7 days
Participants Ankle sprain (lt 48 hours)N = 140M 72 F 62Age 18 to 65 yearsBaseline spontaneous pain ge 50 mm
Interventions (1) DHEP plaster (Flector Tissugel 1) 1 x daily n = 70(2) Placebo plaster 1 x daily n = 70Rescue medication paracetamol (to maximum 3 g daily)Ice allowed
Outcomes ge 30 decrease in PI at 7 daysPGE 4-point scale (responder = ldquoexcellentrdquo)No or low pain on passive stretchSingle foot leaning OK without painWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk Identical in size appearance and used sameformula as active patch without active in-gredient
Incomplete outcome data (attrition bias)All outcomes
Unclear risk Used LOCF for primary outcome With-drawals lt 7 and equal between groupsAll participants included in analysis
83Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Saillant 1998 (Continued)
Size Unclear risk 50 to 200 participants per treatment arm
Sanguinetti 1989
Methods R DB PC parallel groupsGel applied 3 times daily for 7 consecutive daysAssessment at baseline 7 days
Participants Soft tissue trauma (lt 48 hours)N = 82M = 47 F = 35Mean age 34 yearsBaseline pain moderate to severe
Interventions Felbinac gel 3 3 x daily n = 42Placebo gel 3 x daily n = 40No other NSAID steroid other topical application allowedRescue medication paracetamol felbinac is an active metabolite of the NSAID fenbufen
Outcomes PGE scale not reported (responder = ldquogoodrdquo and ldquovery goodrdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoindistinguishable in appearance colour orodourrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
84Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Sinniger 1981
Methods R DB PC parallel groupsCream applied 2 to 3 times daily with gentle massage or if massage not possible (toopainful) with protective dressingAssessment at baseline 5 10 days
Participants Minor soft tissue injuriesN = 20M 11 F 9Mean age 40 yearsBaseline pain not reported
Interventions Fentiazac cream 5 2 to 3 x daily n = 10Placebo cream 2 to 3 x daily n = 10All participants told to restNo other local and systemic treatments allowedRescue medication analgesic if actually needed
Outcomes Pain relief scale not reported (responder = total pain relief ) improvement in pain on movement pain scale not reported (mean data)Adverse events
Notes Oxford Quality Score R1 DB1 W0 Total = 25
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
85Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Spacca 2005
Methods R DB PC parallel groupsGel applied 3 times daily with gentle massage until complete absorption for up to 10daysAssessment at baseline 10 days in clinic and daily patient diary
Participants Shoulder periarthritis or lateral epicondylitis (lt 5 days)N = 155M 74 F 81Mean age 51 yearsBaseline pain with activity gt 70 mm
Interventions DHEP lecithin gel (Effigel) 3 x 5 g daily n = 79Placebo gel 3 x 5 g daily n = 76Rescue medication (paracetamol) allowed if pain unbearableNo other analgesic or anti-inflammatory drug allowed
Outcomes Improvement in pain 100 mm VAS (mean data)Adverse events
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Unclear risk No usable efficacy data
Size Unclear risk 50 to 200 participants per treatment arm
Sugioka 1984
Methods R DB AC parallel groupsGel applied to affected area 3 to 4 times daily without occlusion for 14 daysAssessed at baseline 7 14 days
Participants Non-traumatic diseases of muscle or tendonN = 366 (340 analysed for efficacy)M 115 F 202 (completers)
86Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Sugioka 1984 (Continued)
Age range 12 to 84 years (most 30 to 70)Baseline pain on movement ldquononerdquo or ldquomildrdquo in about 13 of participantsExclusions for protocol violations 8 piroxicam 18 indomethacin
Interventions Piroxicam gel 05 3 to 4 x 1 g daily n = 183Indomethacin gel 1 3 to 4 x 1 g daily n = 183No concomitant anti-inflammatory or analgesic drug including steroids or initiationof physical therapy allowed
Outcomes PGE 5-point scale (responder = ldquobetterrdquo or ldquomuch betterrdquo)Pain on movement 4-point scale (responder = ldquoreducedrdquo or ldquodisappearedrdquo)
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Low risk Key code sealed and retained until end ofstudy
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoboth packages were of the same appear-ance and indistinguishablerdquo and investiga-tors did not see contents
Incomplete outcome data (attrition bias)All outcomes
Low risk All relevant participants included in analy-sis
Size Unclear risk 50 to 200 participants per treatment arm
Thorling 1990
Methods R DB PC parallel groupsParticipants given specific instructions on how to apply gel (not reported) to affectedarea 2 to 6 times daily as requiredAssessment at baseline 3 7 days in clinic
Participants Soft tissue injuries (lt 48 hours)N = 120M 85 F 35Mean age 27 yearsBaseline pain moderate to severe
87Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Thorling 1990 (Continued)
Interventions Naproxen gel 10 2 to 6 x daily n = 60Placebo gel 2 to 6 x daily n = 60Rescue medication paracetamol 500 mg
Outcomes PGE 5-point scale (responder = ldquogoodrdquo and ldquovery goodrdquo)Pain on passive movement 4-point scale (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquosupplied in unmarked tubesrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size Unclear risk 50 to 200 participants per treatment arm
Tonutti 1994
Methods R DB AC parallel groupsGel applied 3 times daily for 2 to 3 weeksAssessed at baseline and intervals of 7 days
Participants Muscle or joint traumaN = 30M 20 F 10Mean age 34 years1 participant had injury of mild severity Mean baseline pain on active movement 28(scale 0 to 4)
Interventions Ketoprofen gel 5 3 x 2 to 3 g daily n = 15Etofenamate gel 5 3 x 2 to 3 g n = 15No concomitant treatment with NSAID aspirin steroid or physical therapy
88Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Tonutti 1994 (Continued)
Outcomes PGE 4-point scale (responder = ldquogoodrdquo and ldquoexcellentrdquo)Pain on movement 5-point scale (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not reported
Allocation concealment (selection bias) Unclear risk Not reported
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquothe two drugs were packed in indistin-guishable tubesrdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
Vecchiet 1991
Methods R DB PC parallel groupsGel applied to the skin on and around painful area and gently rubbed in until absorbedtwice daily for up to 10 daysAssessed at baseline 5 10 days
Participants Soft tissue trauma (minor sports injuries)N = 60M 60Mean age 25 yearsMean baseline pain on active movement moderate
Interventions Meclofenamic acid gel 5 2 x 10 cm daily (= 4 g) n = 30Placebo gel 2 x 10 cm daily n = 30Both groups treated with ice rest and bandage for first 48 hours before starting testtreatmentRescue medication paracetamol
Outcomes PGE 4-point scale (responder = ldquogoodrdquo and ldquoexcellentrdquo)Pain on movement 4-point scale (mean data)Withdrawals
89Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Vecchiet 1991 (Continued)
Notes Oxford Quality Score R1 DB1 W1 Total = 35
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Unclear risk Not described
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Unclear risk Not described
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size High risk lt 50 participants per treatment arm
Whitefield 2002
Methods R DB (double dummy) AC parallel groupsGel applied to affected site with gentle massage and 1 tablet taken 3 times daily for atleast 7 daysAssessed at baseline 7 14 (if necessary) days in clinic and daily patient diary
Participants Soft tissue injuries (lt 24 hours)N = 100M 95 F 5Mean age 26 years (range 18 to 50)Mean baseline pain on movement 22 cm
Interventions Ibuprofen gel 5 + placebo tablet 3 x daily n = 50Ibuprofen 400 mg tablet + placebo gel 3 x daily n = 50No other medication or physical therapy was prescribed and no other analgesics wereallowed
Outcomes PGE 3-point scale (responder = ldquoexcellentrdquo)Change in condition of injury site 5-point scale (responder = ldquocompletely betterrdquo)Adverse eventsWithdrawals
Notes Oxford Quality Score R1 DB2 W1 Total = 45
Risk of bias
Bias Authorsrsquo judgement Support for judgement
90Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Whitefield 2002 (Continued)
Random sequence generation (selectionbias)
Unclear risk Not reported
Allocation concealment (selection bias) Unclear risk Not reported
Blinding (performance bias and detectionbias)All outcomes
Low risk Placebo tablets Identical in appearance toactive tablets Active and placebo gels hadsimilar physical characteristics and weresupplied in identical tubes
Incomplete outcome data (attrition bias)All outcomes
Low risk All participants included in analysis
Size Unclear risk 50 participants per treatment arm
kermark 1990
Methods R DB (double dummy) PC AC parallel groupsSpray applied to affected area and capsules taken 3 times daily for 2 weeksAssessment at baseline 3 or 4 7 and 14 days
Participants Superficial overuse sports injuries (symptom onset 74 weeks)N = 70M 44 F 18 (completers)Mean age 30 yearsBaseline pain on palpation mostly slight to moderate
Interventions Elmetacin spray (indomethacin 1) 3 to 5 x 05 to 15 ml daily + placebo capsules n= 23Indomethacin capsules 3 x 25 mg daily + placebo spray n = 23Placebo spray and capsules n = 24Rescue medication paracetamol
Outcomes No pain on palpation (= responder)Participant improvement 100 mm VAS (mean data)Adverse eventsWithdrawals
Notes Oxford Quality Score R2 DB2 W1 Total = 55
Risk of bias
Bias Authorsrsquo judgement Support for judgement
Random sequence generation (selectionbias)
Low risk ldquorandom number coderdquo
91Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
kermark 1990 (Continued)
Allocation concealment (selection bias) Unclear risk Not described
Blinding (performance bias and detectionbias)All outcomes
Low risk ldquoidentical in appearancerdquo
Incomplete outcome data (attrition bias)All outcomes
Low risk Responder analysis lt 5 withdrawals
Size High risk lt 50 participants per treatment group
AC active control BOCF baseline observation carried forward DB double-blind DHEP diclofenac hydroxyethylpyrrolidine ordiclofenac epolamine F female HCl hydrochloride hep heparin LOCF last observation carried forward M male N numberof participants in study n number of participants in treatment arm PC placebo control PGE Participant Global Evaluation PIpain intensity R randomised VAS visual analogue scale W withdrawals
Characteristics of excluded studies [ordered by study ID]
Study Reason for exclusion
Ambrus 1987 No usable dichotomous data
Anon 1993 Not double-blind
Ascherl 1982 No usable dichotomous data
Bagliani 1976 Not an RCT
Baracchi 1982 No usable data
Burnham 1998 lt10 participantstreatment arm in first period of crossover study
Boumlhmer 1995 Active control invalid
Cesarone 2008 Not an RCT
Coulibaly 2009 Not double-blind
Diebschlag 1985 No usable dichotomous data
Diebschlag 1986 Inappropriate randomisation
Diebschlag 1992 No usable dichotomous data
92Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Fantato 1971 No usable dichotomous data
Galer 2000 No usable data
Hallmeier 1986 Not double-blind
Hallmeier 1988 Not double-blind
Kaneko 1999 Inappropriate randomisation - quasi-randomised
Kockelbergh 1985b Treatment not applied daily
Kuwabara 2013 Used NSAID-lidocaine combination (conference abstract)
Lee 1991 Not an RCT
Link 1996 No usable dichotomous data
May 2007 No usable dichotomous data
Oakland 1993 Inappropriate comparator
Odaglia 1987 Not an RCT
Picardi 1993 Not an RCT
Taboada 1992 Dose and duration of treatment unclear
Vanderstraeten 1990 Not double-blind
Vinciguerra 2008 Not an RCT
Von Klug 1977 Chronic and acute outcomes combined
RCT randomised controlled trial
Characteristics of studies awaiting assessment [ordered by study ID]
NCT00351104
Methods R DB PC parallel groups
Participants Ankle sprain or strain Grade I or II Age ge 18 yearsN = 220
93Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT00351104 (Continued)
Interventions Ketoprofen patch 20 x 1 dailyPlacebo patch x 1 dailyTreatment for 21 days
Outcomes Assessments days 3 7 21PI (0-10) at rest and during activitiesPGEFunctionRescue medication
Notes Sponsor Endo PharmaceuticalsEstimated completion February 2007 No results
NCT00352625
Methods R DB PC parallel groups
Participants Shoulder elbow or knee tendonitis or bursitis Age ge 18 yearsN = 330
Interventions Ketoprofen patch 20 x 1 dailyPlacebo patch x 1 dailyTreatment for 21 days
Outcomes Assessments days 3 7 21PI (0-10) at rest and during activitiesPGEFunctionRescue medication
Notes Sponsor Endo PharmaceuticalsEstimated completion February 2007 No results
NCT00426985
Methods R DB PC parallel groups
Participants Shoulder elbow or knee tendonitis or bursitis Age ge 18 years
Interventions Ketoprofen patch 20 x 1 dailyPlacebo patchTreatment for 21 days
Outcomes PI during activityPI at restPGERescue medication
94Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT00426985 (Continued)
Safety
Notes Study terminated May 2008 but ldquosufficient number of subjects accrued to conduct analysisrdquo No results
NCT00640705
Methods R DB PC parallel groups
Participants Ankle sprain or strain Grade I or II le 48 hours PI 510 to 910 Age 18 to 75 yearsN = 170
Interventions Diclofenac sodium patch (15 mg) x 1 dailyPlaceboTreatment for 7 days
Outcomes EfficacySafety
Notes Study terminated July 2008 (Sponsor decision - no further details) No results posted
NCT00640939
Methods R DB PC parallel groups
Participants Shoulder elbow or knee tendonitis or bursitis Age 18 to 75 yearsBaseline pain mild to moderate but states 510 to 910N = 308
Interventions Diclofenac sodium patch 1 (15 mg) x 1 dailyPlacebo patch x 1 dailyDuration of treatment not reported probably 14 days
Outcomes EfficacySafety
Notes Sponsor Cerimon PharmaceuticalsEstimated completion April 2008 No results
NCT00680472
Methods R DB PC parallel groups
Participants Acute unilateral shoulder pain requiring treatment for ge 2 weeks Age ge 18 yearsN = 368
95Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT00680472 (Continued)
Interventions Ketoprofen patch (HKT-500) x 1 dailyPlacebo x 1 dailyTreatment for 14 days
Outcomes PainSafety
Notes Sponsor Hisamitsu Pharmaceutical Co IncEstimated completion October 2008 No results
NCT00680784
Methods R DB PC parallel groups
Participants Acute benign ankle sprain Grade I-II le 48 hours Age ge 18 yearsN = 260
Interventions Ketoprofen patch (HKT-500) x 1 dailyPlacebo patch x 1 dailyTreatment for 14 days
Outcomes PainSafety
Notes Sponsor Hisamitsu Pharmaceutical Co IncEstimated completion November 2008 No results
NCT00765700
Methods R DB PC parallel groups
Participants Acute sprain or strain of upper and lower extremities mild to moderate injuries le 72 hours Age 18 to 70 yearsN = 364
Interventions Ketoprofen cream 10 1 g x 3 dailyPlacebo cream x 3 dailyTreatment for 7 days
Outcomes PainSafety
Notes Sponsor Imprimis Pharmaceuticals IncEstimated completion September 2009 No results
96Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT00869063
Methods R DB PC parallel groups
Participants Wrist sprain strain or contusion (mild to moderate) ldquorecentrdquo Age 17 to 75 yearsN = 214
Interventions Diclofenac sodium patch 1 x 1 dailyPlacebo patch x 1 dailyTreatment for 7 days
Outcomes Change in PI during activity at 3 and 7 daysSafety
Notes Sponsor Cerimon PharmaceuticalsEstimated completion September 2009 No results
NCT00869180
Methods R DB PC parallel groups
Participants Unilateral ankle sprain (mild or moderate) ldquorecentrdquo Age 17 to 75 yearsN = 219
Interventions Diclofenac sodium patch x 1 dailyPlacebo patch x 1 dailyTreatment for 7 days
Outcomes Change in PI during activity at 3 and 7 days
Notes Sponsor Cerimon PharmaceuticalsEstimated completion August 2009 No results
NCT00931866
Methods R DB PC parallel groups
Participants Unilateral soft tissue injury between mid-bicep to wrist or mid-thigh to ankle (mild to moderate) ldquorecentrdquo Age 18to 75 yearsN = 407
Interventions Diclofenac sodium patch x 1 dailyPlacebo patch x 1 dailyTreatment for 14 days
Outcomes Change in PI during activity at 7 and 14 daysPIPGERescue medicationSafety
97Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT00931866 (Continued)
Notes Sponsor Cerimon PharmaceuticalsEstimated completion December 2009 No results
NCT01874626
Methods R DB PC parallel groups
Participants Acute ankle sprain Grade I or II le 24 hours Age ge 16 years Baseline PI ge 510N = 305
Interventions Ibuprofen cream 200 mg in 27 g cream x 4 dailyPlacebo cream x 4 dailyTreatment for 7 days
Outcomes PI on movement daily to 7 daysPGE 7 daysSystemic and local adverse events
Notes Estimated completion January 2014 No results
NCT01957215
Methods R DB PC parallel groups
Participants Acute lateral ankle sprain Grade I-II le 24 hours PI ge 510 Age 18 to 65 yearsN = 270
Interventions Indomethacin patch x 2 dailyPlacebo patch x 2 dailyDuration of treatment not reported probably 7 days
Outcomes PI on movement dailyPR dailyPGERescue medication
Notes Estimated completion September 2014 No results
NCT02324270
Methods R DB PC and AC parallel groups
Participants Uncomplicated acute ankle sprain Grade I or II lt 48 hours PI gt 50100 Age 18 to 65 yearsN = 658
98Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT02324270 (Continued)
Interventions Diclofenac epolamine (Flector) 13 patchGeneric diclofenac epolamine 13 patchPlacebo patchDuration of treatment not reported
Outcomes Bioequivalence studyChange from baseline in PI (VAS) at 3 daysApplication site reactions
Notes Estimated completion December 2014 No results
Sarzi-Puttini 2014
Methods R DB AC parallel group non-inferiority studyDuration 7 days
Participants Acute musculoskeletal injury (mainly muscular joint tendon)N = 697M 271 F 426Mean age 52 years
Interventions Ketoprofen (SKP-021) patch ketoprofen 30 mgDiclofenac (Voltadola) patch diclofenac sodium 140 mgPatches applied twice daily for 7 days
Outcomes ge 50 reduction in PI from baseline to end of study (100 mm VAS)Patient overall ratingClinical symptoms (4-point scale)Time to responseAdverse events skin reactionsSerious adverse events
Notes ldquoThe analysis of the data of this trial showed that the two formulations were equally effective and well tolerated inthe treatment of acute musculoskeletal injuriesrdquo
AC active-controlled DB double-blind F female M male N number of participants in study PC placebo-controlled PGE PatientGlobal Evaluation of treatment PI pain intensity R randomised VAS visual analogue scale
99Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Characteristics of ongoing studies [ordered by study ID]
NCT01945034
Trial name or title Placebo-controlled double-blind evaluation of the efficacy and safety of ibuprofen 5 topical gel for thetreatment of ankle sprain
Methods R DB PC parallel groups
Participants Ankle sprain Grade I or II age ge 12 yearsEstimated enrolment 280
Interventions Ibuprofen gel 5 x 2 dailyIbuprofen gel 5 x 3 dailyPlacebo gelTreatment for 7 days with additional 3 days as needed
Outcomes PI on weight bearing and rest at 7 daysPGE at 10 daysRescue medicationSafety
Starting date November 2013
Contact information Pfizer
Notes Estimated completion February 2015Includes participants aged ge 12 years
NCT02100670
Trial name or title A clinical study to assess the efficacy and onset of pain relief of topical MFC51123 diclofenac-menthol gelversus controls in ankle sprain
Methods R DB PC and AC parallel groups
Participants Acute lateral ankle sprain Grade I or II le 24 hours PI ge 510 Age 16 to 65 yearsEstimated enrolment 400
Interventions Diclofenac sodium 1 + methanol 3 gelDiclofenac sodium 1 + methanol 009 gelMethanol 3 gelPlacebo + 009 methanol gelTreatment for 10 days with 4 g gel x 4 daily
Outcomes PRPID on movement (0 to 10 days)PGETime to complete recoveryAdverse events
100Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
NCT02100670 (Continued)
Starting date November 2013
Contact information GSKClinicalSupportHDgskcom
Notes Estimated completion November 2014
NCT02290821
Trial name or title A randomised double-blind placebo-controlled parallel group study to evaluate the efficacy and safety ofdiclofenac sodium topical gel (DSG) 1 applied four times daily in subjects with acute blunt soft tissueinjuriescontusions of the limbs
Methods R DB PC parallel groups
Participants Acute blunt soft tissue injuries or contusions of the limbs le 6 hours (ldquofreshrdquo) Age ge 16 yearsEstimated enrolment 200
Interventions Diclofenac sodium gel 1 x 4 dailyPlacebo gelDuration of treatment not reported
Outcomes PainSafety
Starting date December 2014
Contact information Novartis
Notes Estimated completion August 2015
AC active control DB double-blind PC placebo-controlled PGE Patient Global Evaluation of treatment PI pain intensity PIDpain intensity difference PR pain relief R randomised
101Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
D A T A A N D A N A L Y S E S
Comparison 1 Individual NSAID versus placebo
Outcome or subgroup titleNo ofstudies
No ofparticipants Statistical method Effect size
1 Clinical success 29 Risk Ratio (M-H Fixed 95 CI) Subtotals only11 Diclofenac 10 2050 Risk Ratio (M-H Fixed 95 CI) 160 [149 172]12 Ibuprofen 5 436 Risk Ratio (M-H Fixed 95 CI) 164 [133 201]13 Ketoprofen 7 683 Risk Ratio (M-H Fixed 95 CI) 156 [137 177]14 Piroxicam 3 504 Risk Ratio (M-H Fixed 95 CI) 148 [127 173]15 Indomethacin 3 341 Risk Ratio (M-H Fixed 95 CI) 126 [103 155]16 Benzydamine 3 193 Risk Ratio (M-H Fixed 95 CI) 115 [096 138]
2 Local adverse events 33 Risk Ratio (M-H Fixed 95 CI) Subtotals only21 Diclofenac 15 3271 Risk Ratio (M-H Fixed 95 CI) 078 [056 110]22 Ibuprofen 3 321 Risk Ratio (M-H Fixed 95 CI) 230 [098 543]23 Ketoprofen 8 852 Risk Ratio (M-H Fixed 95 CI) 119 [083 170]24 Piroxicam 3 522 Risk Ratio (M-H Fixed 95 CI) 042 [017 108]25 Felbinac 3 397 Risk Ratio (M-H Fixed 95 CI) 191 [049 750]26 Indomethacin 3 354 Risk Ratio (M-H Fixed 95 CI) 265 [091 773]
Comparison 2 Diclofenac versus placebo (effect of formulation)
Outcome or subgroup titleNo ofstudies
No ofparticipants Statistical method Effect size
1 Clinical success 10 2050 Risk Ratio (M-H Fixed 95 CI) 160 [149 172]11 Plaster - Flector 4 1030 Risk Ratio (M-H Fixed 95 CI) 153 [136 171]12 Plaster - other 3 474 Risk Ratio (M-H Fixed 95 CI) 155 [137 175]13 Gel - Emulgel 2 314 Risk Ratio (M-H Fixed 95 CI) 384 [268 550]14 Gel - other 1 232 Risk Ratio (M-H Fixed 95 CI) 115 [105 127]
Comparison 3 Ibuprofen versus placebo (effect of formulation)
Outcome or subgroup titleNo ofstudies
No ofparticipants Statistical method Effect size
1 Clinical success 5 436 Risk Ratio (M-H Fixed 95 CI) 164 [133 201]11 Cream 3 195 Risk Ratio (M-H Fixed 95 CI) 128 [103 159]12 Gel 2 241 Risk Ratio (M-H Fixed 95 CI) 266 [169 421]
102Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Comparison 4 Ketoprofen versus placebo
Outcome or subgroup titleNo ofstudies
No ofparticipants Statistical method Effect size
1 Clinical success 7 683 Risk Ratio (M-H Fixed 95 CI) 156 [137 177]11 Plaster 2 335 Risk Ratio (M-H Fixed 95 CI) 121 [104 140]12 Gel 5 348 Risk Ratio (M-H Fixed 95 CI) 219 [174 275]
Comparison 5 All topical NSAIDs versus placebo
Outcome or subgroup titleNo ofstudies
No ofparticipants Statistical method Effect size
1 Local adverse events 42 6740 Risk Ratio (M-H Fixed 95 CI) 098 [080 121]2 Systemic adverse events 36 5576 Risk Ratio (M-H Fixed 95 CI) 096 [073 126]3 Adverse event withdrawals 42 6405 Risk Ratio (M-H Fixed 95 CI) 101 [064 159]
Comparison 6 Topical NSAID versus active comparator
Outcome or subgroup titleNo ofstudies
No ofparticipants Statistical method Effect size
1 Clinical success - topicalpiroxicam vs topicalindomethacin
3 641 Risk Ratio (M-H Fixed 95 CI) 124 [107 144]
2 Local adverse events -topical piroxicam vs topicalindomethacin
3 671 Risk Ratio (M-H Fixed 95 CI) 021 [009 047]
103Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Analysis 11 Comparison 1 Individual NSAID versus placebo Outcome 1 Clinical success
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 1 Individual NSAID versus placebo
Outcome 1 Clinical success
Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
1 Diclofenac
Coudreuse 2010 99117 82116 175 120 [ 104 138 ]
Joussellin 2003 3668 2466 52 146 [ 099 215 ]
Klainguti 2010 5662 4659 100 116 [ 099 136 ]
Li 2013 173192 114192 242 152 [ 134 172 ]
Predel 2004 5560 560 11 1100 [ 474 2555 ]
Predel 2012 (1) 5980 941 25 336 [ 186 607 ]
Predel 2012 (2) 5780 841 22 365 [ 193 690 ]
Predel 2013a 111118 93114 201 115 [ 105 127 ]
Predel 2013b 3636 736 16 487 [ 257 923 ]
Rowbotham 2003 75191 48181 104 148 [ 110 200 ]
Saillant 1998 4370 2570 53 172 [ 119 248 ]
Subtotal (95 CI) 1074 976 1000 160 [ 149 172 ]Total events 800 (NSAID) 461 (Placebo)
Heterogeneity Chi2 = 12176 df = 10 (Plt000001) I2 =92
Test for overall effect Z = 1278 (P lt 000001)
2 Ibuprofen
Billigmann 1996 2580 1080 136 250 [ 129 486 ]
Campbell 1994 2126 1925 264 106 [ 080 142 ]
Dreiser 1988 2632 1232 164 217 [ 134 349 ]
Machen 2002 2540 941 121 285 [ 152 532 ]
Ramesh 1983 2340 2340 314 100 [ 069 146 ]
Subtotal (95 CI) 218 218 1000 164 [ 133 201 ]Total events 120 (NSAID) 73 (Placebo)
Heterogeneity Chi2 = 2102 df = 4 (P = 000031) I2 =81
Test for overall effect Z = 465 (P lt 000001)
3 Ketoprofen
Airaksinen 1993 2429 1427 91 160 [ 107 238 ]
Dreiser 1989 1830 530 32 360 [ 154 844 ]
02 05 1 2 5
Favours placebo Favours NSAID
(Continued )
104Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
( Continued)Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Julien 1989 1830 630 38 300 [ 138 650 ]
Kockelbergh 1985 3038 2236 142 129 [ 095 176 ]
Mazi res 2005a 7281 6082 376 121 [ 104 141 ]
Mazi res 2005b 5087 4185 262 119 [ 090 158 ]
Noret 1987 3951 947 59 399 [ 218 733 ]
Subtotal (95 CI) 346 337 1000 156 [ 137 177 ]Total events 251 (NSAID) 157 (Placebo)
Heterogeneity Chi2 = 3104 df = 6 (P = 000002) I2 =81
Test for overall effect Z = 683 (P lt 000001)
4 Piroxicam
Aoki 1984 5672 3367 286 158 [ 120 207 ]
Fujimaki 1985 4483 4082 337 109 [ 080 147 ]
Russell 1991 79100 45100 377 176 [ 138 223 ]
Subtotal (95 CI) 255 249 1000 148 [ 127 173 ]Total events 179 (NSAID) 118 (Placebo)
Heterogeneity Chi2 = 624 df = 2 (P = 004) I2 =68
Test for overall effect Z = 499 (P lt 000001)
5 Indomethacin
Aoki 1984 4164 3367 413 130 [ 096 176 ]
Fujimaki 1985 4482 4082 513 110 [ 082 148 ]
kermark 1990 1222 624 74 218 [ 099 481 ]
Subtotal (95 CI) 168 173 1000 126 [ 103 155 ]Total events 97 (NSAID) 79 (Placebo)
Heterogeneity Chi2 = 269 df = 2 (P = 026) I2 =26
Test for overall effect Z = 221 (P = 0027)
6 Benzydamine
Chatterjee 1977 2125 1225 185 175 [ 112 272 ]
Haig 1986 1821 1322 196 145 [ 098 214 ]
Linde 1985 3550 4050 618 088 [ 070 110 ]
Subtotal (95 CI) 96 97 1000 115 [ 096 138 ]Total events 74 (NSAID) 65 (Placebo)
Heterogeneity Chi2 = 1033 df = 2 (P = 001) I2 =81
Test for overall effect Z = 152 (P = 013)
02 05 1 2 5
Favours placebo Favours NSAID
(1) Three times daily application
(2) Twice daily application
105Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Analysis 12 Comparison 1 Individual NSAID versus placebo Outcome 2 Local adverse events
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 1 Individual NSAID versus placebo
Outcome 2 Local adverse events
Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
1 Diclofenac
Coudreuse 2010 1117 7116 102 014 [ 002 113 ]
Jenoure 1997 144 141 15 093 [ 006 1442 ]
Joussellin 2003 168 366 44 032 [ 003 303 ]
Klainguti 2010 0126 159 30 016 [ 001 381 ]
Kuehl 2011 16207 12211 173 136 [ 066 280 ]
Li 2013 4192 3192 44 133 [ 030 588 ]
NCT01255423 1104 3102 44 033 [ 003 309 ]
NCT01272934 1102 0103 07 303 [ 012 7350 ]
NCT01272947 0104 0100 Not estimable
Predel 2012 (1) 1160 182 19 051 [ 003 809 ]
Predel 2013a 1120 4116 59 024 [ 003 213 ]
Predel 2013b 036 036 Not estimable
Rowbotham 2003 27191 31181 463 083 [ 051 133 ]
Saillant 1998 070 070 Not estimable
Spacca 2005 079 076 Not estimable
Subtotal (95 CI) 1720 1551 1000 078 [ 056 110 ]Total events 54 (NSAID) 66 (Placebo)
Heterogeneity Chi2 = 944 df = 10 (P = 049) I2 =00
Test for overall effect Z = 142 (P = 016)
2 Ibuprofen
Billigmann 1996 1180 480 573 275 [ 091 827 ]
Machen 2002 440 241 283 205 [ 040 1057 ]
Ramesh 1983 140 140 143 100 [ 006 1544 ]
Subtotal (95 CI) 160 161 1000 230 [ 098 543 ]Total events 16 (NSAID) 7 (Placebo)
Heterogeneity Chi2 = 048 df = 2 (P = 079) I2 =00
Test for overall effect Z = 190 (P = 0057)
3 Ketoprofen
002 01 1 10 50
Favours NSAID Favours placebo
(Continued )
106Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
( Continued)
Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Airaksinen 1993 529 427 99 116 [ 035 389 ]
Dreiser 1989 030 230 60 020 [ 001 400 ]
Julien 1989 130 030 12 300 [ 013 7083 ]
Kockelbergh 1985 138 136 24 095 [ 006 1459 ]
Mazi res 2005a 1281 682 142 202 [ 080 513 ]
Mazi res 2005b 2987 2785 651 105 [ 068 161 ]
Noret 1987 151 047 12 277 [ 012 6636 ]
Parrini 1992 083 086 Not estimable
Subtotal (95 CI) 429 423 1000 119 [ 083 170 ]Total events 49 (NSAID) 40 (Placebo)
Heterogeneity Chi2 = 357 df = 6 (P = 073) I2 =00
Test for overall effect Z = 095 (P = 034)
4 Piroxicam
Aoki 1984 179 274 147 047 [ 004 506 ]
Fujimaki 1985 183 282 143 049 [ 005 534 ]
Russell 1991 4102 10102 710 040 [ 013 123 ]
Subtotal (95 CI) 264 258 1000 042 [ 017 108 ]Total events 6 (NSAID) 14 (Placebo)
Heterogeneity Chi2 = 003 df = 2 (P = 098) I2 =00
Test for overall effect Z = 180 (P = 0072)
5 Felbinac
McLatchie 1989 3118 2113 666 144 [ 024 844 ]
Morris 1991 041 043 Not estimable
Sanguinetti 1989 342 140 334 286 [ 031 2634 ]
Subtotal (95 CI) 201 196 1000 191 [ 049 750 ]Total events 6 (NSAID) 3 (Placebo)
Heterogeneity Chi2 = 023 df = 1 (P = 063) I2 =00
Test for overall effect Z = 093 (P = 035)
6 Indomethacin
Aoki 1984 270 274 440 106 [ 015 730 ]
Fujimaki 1985 582 282 452 250 [ 050 1252 ]
kermark 1990 422 024 108 978 [ 056 17191 ]
Subtotal (95 CI) 174 180 1000 265 [ 091 773 ]Total events 11 (NSAID) 4 (Placebo)
Heterogeneity Chi2 = 167 df = 2 (P = 043) I2 =00
Test for overall effect Z = 179 (P = 0073)
002 01 1 10 50
Favours NSAID Favours placebo
(1) DHEP + heparin and DHEP combined
107Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Analysis 21 Comparison 2 Diclofenac versus placebo (effect of formulation) Outcome 1 Clinical success
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 2 Diclofenac versus placebo (effect of formulation)
Outcome 1 Clinical success
Study or subgroup Experimental Control Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
1 Plaster - Flector
Joussellin 2003 3668 2466 52 146 [ 099 215 ]
Li 2013 173192 114192 242 152 [ 134 172 ]
Rowbotham 2003 75191 48181 104 148 [ 110 200 ]
Saillant 1998 4370 2570 53 172 [ 119 248 ]
Subtotal (95 CI) 521 509 451 153 [ 136 171 ]Total events 327 (Experimental) 211 (Control)
Heterogeneity Chi2 = 051 df = 3 (P = 092) I2 =00
Test for overall effect Z = 720 (P lt 000001)
2 Plaster - other
Coudreuse 2010 99117 82116 175 120 [ 104 138 ]
Klainguti 2010 5662 4659 100 116 [ 099 136 ]
Predel 2004 5560 560 11 1100 [ 474 2555 ]
Subtotal (95 CI) 239 235 285 155 [ 137 175 ]Total events 210 (Experimental) 133 (Control)
Heterogeneity Chi2 = 4658 df = 2 (Plt000001) I2 =96
Test for overall effect Z = 699 (P lt 000001)
3 Gel - Emulgel
Predel 2012 (1) 116160 1782 48 350 [ 227 540 ]
Predel 2013b 3636 736 16 487 [ 257 923 ]
Subtotal (95 CI) 196 118 64 384 [ 268 550 ]Total events 152 (Experimental) 24 (Control)
Heterogeneity Chi2 = 070 df = 1 (P = 040) I2 =00
Test for overall effect Z = 734 (P lt 000001)
4 Gel - other
Predel 2013a 111118 93114 201 115 [ 105 127 ]
Subtotal (95 CI) 118 114 201 115 [ 105 127 ]Total events 111 (Experimental) 93 (Control)
02 05 1 2 5
Favours placebo Favours diclofenac
(Continued )
108Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
( Continued)Study or subgroup Experimental Control Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Heterogeneity not applicable
Test for overall effect Z = 284 (P = 00045)
Total (95 CI) 1074 976 1000 160 [ 149 172 ]Total events 800 (Experimental) 461 (Control)
Heterogeneity Chi2 = 12176 df = 9 (Plt000001) I2 =93
Test for overall effect Z = 1278 (P lt 000001)
Test for subgroup differences Chi2 = 5003 df = 3 (P = 000) I2 =94
02 05 1 2 5
Favours placebo Favours diclofenac
(1) twice daily and three times daily applications combined
Analysis 31 Comparison 3 Ibuprofen versus placebo (effect of formulation) Outcome 1 Clinical success
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 3 Ibuprofen versus placebo (effect of formulation)
Outcome 1 Clinical success
Study or subgroup Ibuprofen Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
1 Cream
Campbell 1994 2126 1925 264 106 [ 080 142 ]
Dreiser 1988 2632 1232 164 217 [ 134 349 ]
Ramesh 1983 2340 2340 314 100 [ 069 146 ]
Subtotal (95 CI) 98 97 742 128 [ 103 159 ]Total events 70 (Ibuprofen) 54 (Placebo)
Heterogeneity Chi2 = 791 df = 2 (P = 002) I2 =75
Test for overall effect Z = 222 (P = 0026)
2 Gel
Billigmann 1996 2580 1080 136 250 [ 129 486 ]
Machen 2002 2540 941 121 285 [ 152 532 ]
Subtotal (95 CI) 120 121 258 266 [ 169 421 ]Total events 50 (Ibuprofen) 19 (Placebo)
001 01 1 10 100
Favours placebo Favours ibuprofen
(Continued )
109Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
( Continued)Study or subgroup Ibuprofen Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Heterogeneity Chi2 = 008 df = 1 (P = 078) I2 =00
Test for overall effect Z = 420 (P = 0000026)
Total (95 CI) 218 218 1000 164 [ 133 201 ]Total events 120 (Ibuprofen) 73 (Placebo)
Heterogeneity Chi2 = 2102 df = 4 (P = 000031) I2 =81
Test for overall effect Z = 465 (P lt 000001)
Test for subgroup differences Chi2 = 805 df = 1 (P = 000) I2 =88
001 01 1 10 100
Favours placebo Favours ibuprofen
Analysis 41 Comparison 4 Ketoprofen versus placebo Outcome 1 Clinical success
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 4 Ketoprofen versus placebo
Outcome 1 Clinical success
Study or subgroup Diclofenac Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
1 Plaster
Mazi res 2005a 7281 6082 376 121 [ 104 141 ]
Mazi res 2005b 5087 4185 262 119 [ 090 158 ]
Subtotal (95 CI) 168 167 638 121 [ 104 140 ]Total events 122 (Diclofenac) 101 (Placebo)
Heterogeneity Chi2 = 002 df = 1 (P = 090) I2 =00
Test for overall effect Z = 249 (P = 0013)
2 Gel
Airaksinen 1993 2429 1427 91 160 [ 107 238 ]
Dreiser 1989 1830 530 32 360 [ 154 844 ]
Julien 1989 1830 630 38 300 [ 138 650 ]
Kockelbergh 1985 3038 2236 142 129 [ 095 176 ]
Noret 1987 3951 947 59 399 [ 218 733 ]
001 01 1 10 100
Favours placebo Favours diclofenac
(Continued )
110Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
( Continued)Study or subgroup Diclofenac Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Subtotal (95 CI) 178 170 362 219 [ 174 275 ]Total events 129 (Diclofenac) 56 (Placebo)
Heterogeneity Chi2 = 1937 df = 4 (P = 000067) I2 =79
Test for overall effect Z = 674 (P lt 000001)
Total (95 CI) 346 337 1000 156 [ 137 177 ]Total events 251 (Diclofenac) 157 (Placebo)
Heterogeneity Chi2 = 3104 df = 6 (P = 000002) I2 =81
Test for overall effect Z = 683 (P lt 000001)
Test for subgroup differences Chi2 = 1859 df = 1 (P = 000) I2 =95
001 01 1 10 100
Favours placebo Favours diclofenac
Analysis 51 Comparison 5 All topical NSAIDs versus placebo Outcome 1 Local adverse events
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 5 All topical NSAIDs versus placebo
Outcome 1 Local adverse events
Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Airaksinen 1993 529 427 26 116 [ 035 389 ]
Aoki 1984 (1) 270 137 08 106 [ 010 1128 ]
Aoki 1984 (2) 179 137 09 047 [ 003 728 ]
Auclair 1989 5123 6116 39 079 [ 025 251 ]
Billigmann 1996 1180 480 26 275 [ 091 827 ]
Costantino 2011 (3) 3287 6142 51 025 [ 006 097 ]
Coudreuse 2010 1117 7116 45 014 [ 002 113 ]
Diebshlag 1990 (4) 113 06 04 150 [ 007 3229 ]
Diebshlag 1990 (5) 112 06 04 162 [ 008 3466 ]
Dreiser 1989 030 230 16 020 [ 001 400 ]
002 01 1 10 50
Favours NSAID Favours placebo
(Continued )
111Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
( Continued)Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Dreiser 1990 030 330 22 014 [ 001 265 ]
Dreiser 1994 265 066 03 508 [ 025 10373 ]
Fujimaki 1985 (6) 183 141 09 049 [ 003 770 ]
Fujimaki 1985 (7) 582 141 09 250 [ 030 2070 ]
Haig 1986 021 022 Not estimable
Jenoure 1997 144 141 07 093 [ 006 1442 ]
Joussellin 2003 168 366 19 032 [ 003 303 ]
Julien 1989 130 030 03 300 [ 013 7083 ]
Klainguti 2010 (8) 0126 159 13 016 [ 001 381 ]
Kockelbergh 1985 138 136 07 095 [ 006 1459 ]
Kuehl 2011 16207 12211 76 136 [ 066 280 ]
Li 2013 4192 3192 19 133 [ 030 588 ]
Linde 1985 050 050 Not estimable
Machen 2002 440 241 13 205 [ 040 1057 ]
Mazi res 2005a 1281 682 38 202 [ 080 513 ]
Mazi res 2005b 2987 2785 174 105 [ 068 161 ]
McLatchie 1989 3118 2113 13 144 [ 024 844 ]
Morris 1991 041 043 Not estimable
NCT01255423 1104 3102 19 033 [ 003 309 ]
NCT01272934 1102 0103 03 303 [ 012 7350 ]
NCT01272947 0104 0100 Not estimable
Noret 1987 151 047 03 277 [ 012 6636 ]
Parrini 1992 083 086 Not estimable
Predel 2012 (9) 1160 182 08 051 [ 003 809 ]
Predel 2013a 1120 4116 26 024 [ 003 213 ]
Predel 2013b 036 036 Not estimable
Ramesh 1983 140 140 06 100 [ 006 1544 ]
Rowbotham 2003 27191 31181 203 083 [ 051 133 ]
Russell 1991 4102 10102 64 040 [ 013 123 ]
Saillant 1998 070 070 Not estimable
Sanguinetti 1989 342 140 07 286 [ 031 2634 ]
002 01 1 10 50
Favours NSAID Favours placebo
(Continued )
112Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
( Continued)Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Sinniger 1981 010 010 Not estimable
Spacca 2005 079 076 Not estimable
Thorling 1990 160 060 03 300 [ 012 7220 ]
kermark 1990 422 024 03 978 [ 056 17191 ]
Total (95 CI) 3619 3121 1000 098 [ 080 121 ]Total events 155 (NSAID) 145 (Placebo)
Heterogeneity Chi2 = 3354 df = 35 (P = 054) I2 =00
Test for overall effect Z = 017 (P = 087)
Test for subgroup differences Not applicable
002 01 1 10 50
Favours NSAID Favours placebo
(1) indomethacin
(2) piroxicam
(3) DHEP + heparin and DHEP combined
(4) ketorolac
(5) etofenamate
(6) piroxicam
(7) indomethacin
(8) DHEP + heparin and DHEP combined
(9) Twice and three times daily application combined
113Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Analysis 52 Comparison 5 All topical NSAIDs versus placebo Outcome 2 Systemic adverse events
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 5 All topical NSAIDs versus placebo
Outcome 2 Systemic adverse events
Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Airaksinen 1993 129 027 05 280 [ 012 6593 ]
Aoki 1984 (1) 079 037 Not estimable
Aoki 1984 (2) 070 037 Not estimable
Campbell 1994 126 025 05 289 [ 012 6775 ]
Chatterjee 1977 025 025 Not estimable
Coudreuse 2010 1117 0116 05 297 [ 012 7227 ]
Diebshlag 1990 (3) 013 06 Not estimable
Diebshlag 1990 (4) 012 06 Not estimable
Dreiser 1988 032 032 Not estimable
Dreiser 1989 030 030 Not estimable
Dreiser 1990 030 030 Not estimable
Dreiser 1994 065 066 Not estimable
Fujimaki 1985 (5) 083 041 Not estimable
Fujimaki 1985 (6) 182 041 07 152 [ 006 3647 ]
Haig 1986 021 022 Not estimable
Hoffmann 2012 0115 0118 Not estimable
Joussellin 2003 168 066 05 291 [ 012 7025 ]
Klainguti 2010 (7) 2126 059 07 236 [ 012 4844 ]
Kuehl 2011 15207 23211 236 066 [ 036 124 ]
Li 2013 10192 4192 41 250 [ 080 783 ]
Linde 1985 050 050 Not estimable
Machen 2002 040 041 Not estimable
Mazi res 2005a 1187 785 73 154 [ 062 377 ]
Mazi res 2005b 1381 1482 144 094 [ 047 187 ]
McLatchie 1989 0118 0113 Not estimable
Morris 1991 041 043 Not estimable
001 01 1 10 100
Favours topical NSAID Favours placebo
(Continued )
114Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
( Continued)Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
NCT01272947 2104 2100 21 096 [ 014 670 ]
Parrini 1992 083 086 Not estimable
Predel 2004 060 060 Not estimable
Predel 2012 (8) 3160 382 41 051 [ 011 248 ]
Predel 2013a 6120 8116 84 073 [ 026 203 ]
Predel 2013b 036 136 16 033 [ 001 792 ]
Rowbotham 2003 21191 22181 234 090 [ 052 159 ]
Russell 1991 4102 7102 73 057 [ 017 189 ]
Saillant 1998 070 070 Not estimable
Sanguinetti 1989 042 040 Not estimable
Sinniger 1981 010 010 Not estimable
Spacca 2005 079 076 Not estimable
Thorling 1990 060 060 Not estimable
Total (95 CI) 2956 2620 1000 096 [ 073 126 ]Total events 92 (NSAID) 91 (Placebo)
Heterogeneity Chi2 = 946 df = 15 (P = 085) I2 =00
Test for overall effect Z = 030 (P = 076)
Test for subgroup differences Not applicable
001 01 1 10 100
Favours topical NSAID Favours placebo
(1) piroxicam
(2) indomethacin
(3) ketorolac
(4) etofenamate
(5) piroxicam
(6) indomethacin
(7) DHEP + heparin and DHEP combined
(8) Twice and three times daily application combined
115Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Analysis 53 Comparison 5 All topical NSAIDs versus placebo Outcome 3 Adverse event withdrawals
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 5 All topical NSAIDs versus placebo
Outcome 3 Adverse event withdrawals
Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Airaksinen 1993 029 027 Not estimable
Aoki 1984 072 034 Not estimable
Aoki 1984 064 033 Not estimable
Auclair 1989 1123 0116 14 283 [ 012 6879 ]
Billigmann 1996 280 080 14 500 [ 024 10253 ]
Campbell 1994 026 025 Not estimable
Chatterjee 1977 025 025 Not estimable
Costantino 2011 0188 0142 Not estimable
Coudreuse 2010 0117 0116 Not estimable
Diebshlag 1990 013 06 Not estimable
Diebshlag 1990 012 06 Not estimable
Dreiser 1988 032 032 Not estimable
Dreiser 1989 030 230 69 020 [ 001 400 ]
Dreiser 1990 030 130 42 033 [ 001 787 ]
Dreiser 1994 065 066 Not estimable
Fujimaki 1985 083 041 Not estimable
Fujimaki 1985 482 041 18 455 [ 025 8261 ]
Julien 1989 030 030 Not estimable
Kockelbergh 1985 038 036 Not estimable
Kuehl 2011 4207 9211 248 045 [ 014 145 ]
Li 2013 2192 0192 14 500 [ 024 10347 ]
Linde 1985 050 050 Not estimable
Machen 2002 040 041 Not estimable
Mazi res 2005a 987 685 169 147 [ 055 394 ]
Mazi res 2005b 381 082 14 709 [ 037 13503 ]
McLatchie 1989 0118 0113 Not estimable
001 01 1 10 100
Favours topical NSAID Favours placebo
(Continued )
116Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
( Continued)Study or subgroup NSAID Placebo Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Morris 1991 041 043 Not estimable
NCT01255423 0104 0102 Not estimable
NCT01272934 0102 1103 41 034 [ 001 817 ]
NCT01272947 0104 0100 Not estimable
Noret 1987 148 045 14 282 [ 012 6740 ]
Parrini 1992 060 060 Not estimable
Predel 2004 160 060 14 300 [ 012 7220 ]
Predel 2012 2160 182 37 103 [ 009 1114 ]
Predel 2013a 1120 1116 28 097 [ 006 1527 ]
Predel 2013b 036 036 Not estimable
Ramesh 1983 140 140 28 100 [ 006 1544 ]
Rowbotham 2003 0191 0181 Not estimable
Russell 1991 1102 8102 222 013 [ 002 098 ]
Saillant 1998 070 070 Not estimable
Sanguinetti 1989 042 040 Not estimable
Sinniger 1981 010 010 Not estimable
Spacca 2005 079 076 Not estimable
Thorling 1990 060 060 Not estimable
kermark 1990 122 024 13 326 [ 014 7610 ]
Total (95 CI) 3365 3040 1000 101 [ 064 159 ]Total events 33 (NSAID) 30 (Placebo)
Heterogeneity Chi2 = 1502 df = 16 (P = 052) I2 =00
Test for overall effect Z = 004 (P = 097)
Test for subgroup differences Not applicable
001 01 1 10 100
Favours topical NSAID Favours placebo
117Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Analysis 61 Comparison 6 Topical NSAID versus active comparator Outcome 1 Clinical success - topical
piroxicam vs topical indomethacin
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 6 Topical NSAID versus active comparator
Outcome 1 Clinical success - topical piroxicam vs topical indomethacin
Study or subgroup Piroxicam Indomethacin Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Aoki 1984 5672 4164 301 121 [ 097 151 ]
Fujimaki 1985 4483 4482 307 099 [ 074 131 ]
Sugioka 1984 85175 55165 392 146 [ 112 190 ]
Total (95 CI) 330 311 1000 124 [ 107 144 ]Total events 185 (Piroxicam) 140 (Indomethacin)
Heterogeneity Chi2 = 390 df = 2 (P = 014) I2 =49
Test for overall effect Z = 278 (P = 00055)
Test for subgroup differences Not applicable
01 02 05 1 2 5 10
Favours indomethacin Favours piroxicam
118Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Analysis 62 Comparison 6 Topical NSAID versus active comparator Outcome 2 Local adverse events -
topical piroxicam vs topical indomethacin
Review Topical NSAIDs for acute musculoskeletal pain in adults
Comparison 6 Topical NSAID versus active comparator
Outcome 2 Local adverse events - topical piroxicam vs topical indomethacin
Study or subgroup Piroxicam Indomethacin Risk Ratio Weight Risk Ratio
nN nN M-HFixed95 CI M-HFixed95 CI
Aoki 1984 179 270 64 044 [ 004 478 ]
Fujimaki 1985 183 582 152 020 [ 002 165 ]
Sugioka 1984 5178 26179 784 019 [ 008 049 ]
Total (95 CI) 340 331 1000 021 [ 009 047 ]Total events 7 (Piroxicam) 33 (Indomethacin)
Heterogeneity Chi2 = 041 df = 2 (P = 081) I2 =00
Test for overall effect Z = 382 (P = 000013)
Test for subgroup differences Not applicable
002 01 1 10 50
Favours piroxicam Favours indomethacin
A P P E N D I C E S
Appendix 1 CENTRAL search strategy (via CRSO) for 2015 update
1 MESH DESCRIPTOR Anti-Inflammatory Agents Non-Steroidal EXPLODE ALL TREES (13419)2 (bufexamac OR bufexine OR calmaderm OR ekzemase OR dicoflenac OR solaraze OR pennsaid OR voltarol OR emugel OR
voltarene OR voltarol OR optha OR voltaren OR etofenamate OR afrolate OR algesalona OR bayro OR deiron OR etofen ORflexium OR flogoprofen OR rheuma-gel OR rheumon OR traumalix OR traumon OR zenavan OR felbinac OR dolinac OR flexfreeOR napageln OR target OR traxam OR fentiazac OR domureuma OR fentiazaco OR norvedan OR riscalon OR fepradinol ORdalgen OR flexidol OR cocresol OR rangozona OR reuflodol OR pinazone OR zepelin OR flufenamic OR dignodolin OR rheumaOR lindofluid OR sastridex OR lunoxaprofen OR priaxim OR flubiprofen OR fenomel OR ocufen OR ocuflur OR ldquoTrans ActLATrdquo OR tulip OR ibuprofen OR cuprofen OR ldquodeep relief rdquo OR fenbid OR ibu-cream OR ibugel OR ibuleve OR ibumousse ORibuspray OR ldquonurofen gelrdquo OR proflex OR motrin OR advil OR radian OR ralgex OR ibutop OR indomethacin OR indocin ORindospray OR isonixin OR nixyn OR ketoprofen OR tiloket OR oruvail OR powergel OR solpaflex OR ketorolac OR acular ORtrometamol OR meclofenamic OR naproxen OR naprosyn OR niflumic OR actol OR flunir OR niflactol topico OR niflugel ORnifluril OR oxyphenbutazone OR californit OR diflamil OR otone OR tanderil OR piketoprofen OR calmatel OR triparsean ORpiroxicam OR feldene OR pranoprofen OR oftalar OR pranox OR suxibuzone OR danilon OR flamilon OR ufenamate OR fenazolOR flector OR benzydamine) TIABKY (25220)
3 1 OR 2 (32484)4 MESH DESCRIPTOR Administration Topical EXPLODE ALL TREES (2169)
119Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
5 (topical OR cutaneous OR dermal OR transcutaneous OR transdermal OR percutaneous OR skin OR massage ORembrocation OR gel OR ointment OR aerosol OR cream OR cregraveme OR lotion OR mouse OR foam OR liniment OR spray OR rubOR balm OR salve OR emulsion OR oil OR patch OR plaster)TIABKY (67940)
6 4 OR 5 (70486)7 MESH DESCRIPTOR Athletic Injuries EXPLODE ALL TREES (411)8 (strain OR sprain OR contusion OR distortion OR compression OR ldquosports injurrdquo OR ldquosoft tissue injurrdquo OR tendnitis OR
ldquomuscle painrdquo OR periarthritis OR epicondylitis OR tenosynovitis)TIABKY (9158)9 7 OR 8 (9448)
10 MESH DESCRIPTOR pain EXPLODE ALL TREES (29943)11 (pain OR analgesi)TIABKY (74815)12 10 OR 11 (80041)13 3 AND 6 AND 9 AND 12 (110)
Appendix 2 MEDLINE search strategy via Ovid (for 2015 update)
1 exp Anti-inflammatory Agents non-steroidal (162888)2 (bufexamac OR bufexine OR calmaderm OR ekzemase OR diclofenac OR solaraze OR pennsaid OR voltarol OR emugel OR
voltarene OR voltarol OR optha OR voltaren OR etofenamate OR afrolate OR algesalona OR bayro OR deiron OR etofen ORflexium OR flogoprofen OR rheuma-gel OR rheumon OR traumalix OR traumon OR zenavan OR felbinac OR dolinac OR flexfreeOR napageln OR target OR traxam OR fentiazac OR domureuma OR fentiazaco OR norvedan OR riscalon OR fepradinol ORdalgen OR flexidol OR cocresol OR rangozona OR reuflodol OR pinazone OR zepelin OR flufenamic OR dignodolin OR rheumaOR lindofluid OR sastridex OR lunoxaprofen OR priaxim OR flubiprofen OR fenomel OR ocufen OR ocuflur OR ldquoTrans ActLATrdquo OR tulip OR ibuprofen OR cuprofen OR ldquodeep relief rdquo OR fenbid OR ibu-cream OR ibugel OR ibuleve OR ibumousse ORibuspray OR ldquonurofen gelrdquo OR proflex OR motrin OR advil OR radian OR ralgex OR ibutop OR indomethacin OR indocin ORindospray OR isonixin OR nixyn OR ketoprofen OR tiloket OR oruvail OR powergel OR solpaflex OR ketorolac OR acular ORtrometamol OR meclofenamic OR naproxen OR naprosyn OR niflumic OR actol OR flunir OR niflactol topico OR niflugel ORnifluril OR oxyphenbutazone OR californit OR diflamil OR otone OR tanderil OR piketoprofen OR calmatel OR triparsean ORpiroxicam OR feldene OR pranoprofen OR oftalar OR pranox OR suxibuzone OR danilon OR flamilon OR ufenamate OR fenazolOR flector OR benzydamine)mp (558284)
3 1 OR 2 (664691)4 exp Administration Topical (69697)5 (topical OR cutaneous OR dermal OR transcutaneous OR transdermal OR percutaneous OR skin OR massage OR
embrocation OR gel OR ointment OR aerosol OR cream OR cregraveme OR lotion OR mouse OR foam OR liniment OR spray OR rubOR balm OR salve OR emulsion OR oil OR patch OR plaster)mp (1982395)
6 4 OR 5 (1997720)7 exp Athletic Injuries (21464)8 (strain OR sprain OR contusion OR distortion OR compression OR ldquosports injurrdquo OR ldquosoft tissue injurrdquo OR tendnitis OR
ldquomuscle painrdquo OR periarthritis OR epicondylitis OR tenosynovitis)mp (420133)9 7 OR 8 (439228)
10 Pain (113906)11 (pain OR analgesi)mp (585249)12 10 or 11 (585249)13 randomized controlled trialpt (401171)14 randomizedab (296222)15 placeboab (155341)16 drug therapyfs (1789858)17 randomlyab (207517)18 trialab (308477)19 groupsab (1318386)20 OR13-1921 3 AND 6 AND 9 AND 12 AND 20 (139)22 limit 21 to yr=ldquo2008-Currentrdquo (56)
120Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Appendix 3 EMBASE search strategy via Ovid (for 2015 update)
1 exp Anti-inflammatory Agents non-steroidal (452266)2 (bufexamac OR bufexine OR calmaderm OR ekzemase OR dicoflenac OR solaraze OR pennsaid OR voltarol OR emugel OR
voltarene OR voltarol OR optha OR voltaren OR etofenamate OR afrolate OR algesalona OR bayro OR deiron OR etofen ORflexium OR flogoprofen OR rheuma-gel OR rheumon OR traumalix OR traumon OR zenavan OR felbinac OR dolinac OR flexfreeOR napageln OR target OR traxam OR fentiazac OR domureuma OR fentiazaco OR norvedan OR riscalon OR fepradinol ORdalgen OR flexidol OR cocresol OR rangozona OR reuflodol OR pinazone OR zepelin OR flufenamic OR dignodolin OR rheumaOR lindofluid OR sastridex OR lunoxaprofen OR priaxim OR flubiprofen OR fenomel OR ocufen OR ocuflur OR ldquoTrans ActLATrdquo OR tulip OR ibuprofen OR cuprofen OR ldquodeep relief rdquo OR fenbid OR ibu-cream OR ibugel OR ibuleve OR ibumousse ORibuspray OR ldquonurofen gelrdquo OR proflex OR motrin OR advil OR radian OR ralgex OR ibutop OR indomethacin OR indocin ORindospray OR isonixin OR nixyn OR ketoprofen OR tiloket OR oruvail OR powergel OR solpaflex OR ketorolac OR acular ORtrometamol OR meclofenamic OR naproxen OR naprosyn OR niflumic OR actol OR flunir OR niflactol topico OR niflugel ORnifluril OR oxyphenbutazone OR californit OR diflamil OR otone OR tanderil OR piketoprofen OR calmatel OR triparsean ORpiroxicam OR feldene OR pranoprofen OR oftalar OR pranox OR suxibuzone OR danilon OR flamilon OR ufenamate OR fenazolOR flector OR benzydamine)mp (790435)
3 1 OR 2 (1108683)4 exp Administration Topical (68490)5 (topical OR cutaneous OR dermal OR transcutaneous OR transdermal OR percutaneous OR skin OR massage OR
embrocation OR gel OR ointment OR aerosol OR cream OR cregraveme OR lotion OR mouse OR foam OR liniment OR spray OR rubOR balm OR salve OR emulsion OR oil OR patch OR plaster)mp (3208844)
6 4 OR 5 (3208844)7 exp Athletic Injuries (24675)8 (strain OR sprain OR contusion OR distortion OR compression OR ldquosports injurrdquo OR ldquosoft tissue injurrdquo OR tendnitis OR
ldquomuscle painrdquo OR periarthritis OR epicondylitis OR tenosynovitis)mp (804322)9 7 OR 8 (825164)
10 Pain (216406)11 (pain OR analgesi)mp (971593)12 10 OR 11 (971593)13 clinical trialssh (841252)14 controlled clinical trialssh (389335)15 randomized controlled trialsh (357169)16 double-blind proceduresh (118945)17 (clin adj25 trial)ab (331002)18 ((doubl or trebl or tripl) adj25 (blind or mask))ab (141726)19 placeboab (203390)20 randomab (906176)21 OR13-20 (1731380)22 3 AND 6 AND 9 AND 10 AND 19 (439)23 limit 22 to yr=ldquo2008-Currentrdquo (204)
Appendix 4 Summary of outcomes efficacy
Study ID Treatment Clinical response Other response
Airaksinen 1993 (1) Ketoprofen gel 2 x 5 g (125mg) daily n = 29(2) Placebo gel n = 27
PGE rdquoimprovedldquo at 7 days(1) 2429(2) 1427
No additional data
121Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
kermark 1990 (1) Indomethacin spray 1 (El-metacin) 3-5 x 05-15 mL dailyn = 23(2) Indomethacin capsules 3 x 25mg daily n = 23(3) Placebo spray and capsules n= 24
No pain on palpation at 7 days(1) 1222(2) 522(3) 624
Patient assessment of improve-ment at 7 days (scale 0 to 100)(1) 57(2) 49(3) 30
Aoki 1984 (1) Piroxicam gel 5 3 or 4 x 1 gdaily n = 84(2) Indomethacin gel 1 3 or 4x 1 g daily n = 84(3) Placebo gel n = 84
PGE (5 point) rdquobetter or muchbetterldquo at 7 days(1) 5672(2) 4164(3) 3367
Pain on movement rdquoreducedldquo orrdquodisappearedldquo at 7 days(1) 4861(2) 3860(3) 3563
Auclair 1989 (1) Niflumic acid gel 25 3 x 5g daily n = 117(2) Placebo gel n = 110
PGE (5 point) rdquogood or verygoodldquo at 7 days(1) 69117(2) 54110
Pain on palpation rdquoimprovedldquo at7 days(1) 69117(2) 53110
Billigmann 1996 (1) Ibuprofen microgel 5 3 x200 mg daily n = 80(2) Placebo gel n = 80
Complete remission(1) 2580(2) 1080
Improvement in pain with move-ment of 20 at 7 days(1) 6580(2) 5580
Campbell 1994 (1) Ibuprofen cream 5 (Proflex)4 x 4rdquo daily n = 26(2) Placebo cream n = 25
Improvement in walking ability (4point) at 7 days(1) 2126(2) 1925
No additional data
Chatterjee 1977 (1) Benzydamine HCl cream 33 x daily n = 25(2) Placebo cream n = 25
Pain on movement ldquoabsentslightrdquoat 6 days(1) 2125(2) 1225
Tenderness with pressure ldquoabsentslightrdquo at 6 days(1) 2125(2) 1225
Costantino 2011 (1) DHEP-heparin plaster (Flec-torparin) x 1 daily n = 142(2) DHEP plaster (Flector) x 1daily n = 146(3) Placebo plaster n = 142
No responder analysis for efficacyOverall treatment efficacy not re-ported
Mean reduction from baseline forPI on movement at 3 days (fromgraph)(1) 24 mm(2) 19 mm(3) 14 mm
Coudreuse 2010 (1) DHEP-heparin plaster (Flec-torparin) n = 120(2) Placebo n = 120
PGE ldquogoodrdquo or ldquoexcellentrdquo at 7days(1) 99117(2) 82116These are results for physician-re-ported judgement - not partici-pant
Mean reduction from baseline forPI on movement over 6 hours(1) about 30(2) about 20Overall greater in DHEP-heparingroup than placebo over 7 days
122Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Curioni 1985 (1) Ibuproxam n = 20(2) Ketoprofen n = 20(3) Etofenamate n = 20
Resolution of symptoms by 7 days(1) 1520(2) 1320(3) 1320
PGE ldquogoodrdquo or ldquoexcellentrdquo at 10days(1) 1920(2) not reported(3) 1620
Diebshlag 1990 (1) Ketorolac gel 2 3 x 3 g dailyn = 13(2) Etofenamate gel 5 3 x 3 gdaily n = 12(3) Placebo gel n = 12
Improvement in pain at 7 days(1) 1213(2) 1012(3) 912
No additional data
Dreiser 1988 (1) Ibuprofen cream 5 3 x 4 cmdaily n = 32 (3 x 10 cm for largejoints)(2) Placebo cream n = 32
PGE ldquoimprovementrdquo or ldquocom-plete relief rdquo at 7 days(1) 2632(2) 1232
(1) significantly better than (2)for mean improvement in sponta-neous pain movement pain restpain tenderness to pressure (VAS)
Dreiser 1989 (1) Ketoprofen gel 25 2 x 5 cmdaily n = 30(2) Placebo gel n = 30
PGE (3 point) ldquobetterrdquo at 7 days(1) 1830(2) 530
(1) significantly better than (2) formean improvement in pain (restand movement) (VAS)
Dreiser 1990 (1) Niflumic acid gel 25 3 x 5g daily n = 30(2) Placebo gel n = 30
PGE (4 point) ldquocuredrdquo or ldquoim-provedrdquo at 7 days(1) 2330(2) 1030
(1) significantly better than (2) formean improvement in pain (VAS)
Dreiser 1994 (1) Flurbiprofen plaster 2 x 40 mgdaily n = 65(2) Placebo plaster n = 66
PGE (4 point) ldquogoodrdquo or ldquoverygoodrdquo at 7 days(1) 4865(2) 4166
(1) significantly better than (2)for mean improvement in sponta-neous pain but not pain on move-ment or palpation (VAS)
Fioravanti 1999 (1) DHEP lecithin gel 3 x 5 g (=65 mg) daily n = 50(2) DHEP gel 3 x 5 g (= 65 mg)daily n = 50
PGE (4 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 10 days(1) 3550(2) 3550
(1) significantly better than (2)for mean improvement in spon-taneous pain at 7 days but notfor pain on movement at 10 days(VAS)
Fujimaki 1985 (1) Piroxicam gel 05 3 or 4 x 1g daily n = 92(2) Indomethacin gel 1 3 or 4 x1 g daily n = 90(3) Placebo gel n = 89
PGE (5 point) ldquobetterrdquo or ldquomuchbetterrdquo at end of treatment at 14days(1) 4483(2) 4482(3) 4082
No additional data
Gallacchi 1990 (1) Diclofenac hy-droxyethylpyrrolidine gel 1 4 x2 g daily n = 25 (Flector gel)(2) Diclofenac sodium 1 4 x 2 g
PGE (5 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 14 days(1) 1925(2) 1925
No significant difference betweengroups for pain on applied pres-sure at 7 and 14 days
123Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
daily n = 25 (Voltaren Emugel)
Gonzaacutelez de Vega 2013 (1) Traumeel gel 3 x 2 g daily n =140(2) Traumeel ointment n = 143(3) Diclofenac gel 1 3 x 2 g dailyn = 137
Pain-free at 7 days(1) 7140(3) 8137PGE (5 point) ldquogoodrdquo or ldquoverygoodrdquo(1) 128140(3) 127137
Normal function (5 point) scoreof 0 or 1 at 14 days(1) 133140(3) 131137
Governali 1995 (1) Ketoprofen gel 5 3 x 2-3 gdaily n = 15(2) Ketoprofen cream 1 3 x 2-3 g daily n = 15
PGE (5 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 7 days(1) 1415(2) 415
No additional data
Gualdi 1987 (1) Flunaxaprofen gel 2 x 3-5 cmdaily n = 30(2) Ketoprofen gel 2 x 3-5 cmdaily n = 30
No dichotomous data No significant difference betweengroups for pain on movement at 7days
Haig 1986 (1) Benzydamine cream 3 6 xdaily n = 21(2) Placebo cream n = 22
Pain on movement ldquoimprovedrdquo by6 days(1) 1821(2) 1322
No additional data
Hoffmann 2012 (1) DHEP-heparin plaster (Flec-torparin) x 1 daily n = 121(2) DHEP (Flector) plaster x 1daily n = 115(3) Placebo plaster n = 118
PGE (5 point) ldquogoodrdquo or ldquoexcel-lentrdquo over 80 in all groups
Mean reduction from baseline inPI (100 mm VAS)Day 3(1) 191 mm(2) 114 mm(3) 52 mmDay 8(1) 443 mm(2) 372 mm(3) 296 mm
Hofman 2000 (1) Diclofenac sodium gel 1 4x 2 cm daily n = 69(2) Lysine clonixinate gel 5 4 x2 cm (225 mg) daily n = 73
PGE (3 point) at 8 days ldquogoodrdquo(1) 3869(2) 3673
No significant difference betweentreatments for any pain outcomes
Hosie 1993 (1) Felbinac foam 3 3 x 2 g daily+ placebo tablets 3 x 1 daily n =140 (127 analysed for efficacy)(2) Ibuprofen tablets 3 x 400 mgdaily + placebo foam 3 x 2 g dailyn = 147 (134 analysed for efficacy)
Pain on movement ldquononerdquo orldquomildrdquo at 7 days(1) 81127(2) 96133
Spontaneous pain ldquononerdquo orldquomildrdquo at 7 days(1) 99127(2) 108134
124Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Jenoure 1997 (1) DHEP plaster (Tissugel) 2 xdaily n = 44(2) Placebo plaster 2 x daily n =41
Baseline pain in two groups notbalanced and data in table and fig-ure do not agree so efficacy out-comes not used
No additional data
Joussellin 2003 (1) DHEP plaster (Flector Tis-sugel 1) 1 x daily n = 68(2) Placebo plaster 1 x daily n =66
PGE (4 point) ldquoexcellentrdquo at 7 days(1) 3668(2) 2466ge 30 reduction in PI at 7 days(1) 2568(2) 1166
(1) significantly better than (2) formean pain on movement at 6 days
Julien 1989 (1) Ketoprofen gel 25 2 x 5 cm(= 50 mg) daily n = 30(2) Placebo gel n = 30
PGE (4 point) ldquorecoveredrdquo at 7days(1) 1830(2) 630
PGE (4 point) ldquorecoveredrdquo or ldquoim-provedrdquo at 7 days(1) 2530(2) 1330
Klainguti 2010 (1) DHEP-heparin plaster (Flec-torparin) x 1 daily n = 62(2) DHEP plaster (Flector) x 1daily n = 61(3) Placebo n = 59
Overall treatment efficacy (partic-ipant and investigator) (5 point)ldquogoodrdquo or ldquoexcellentrdquo at 3 days(1) 5662(2) not reported(3) 4659
Mean reduction in PI at day 3(from graph)(1) 33 mm(2) Not reported(3) 24 mm
Kockelbergh 1985 (1) Ketoprofen gel 25 2 x 5 cm(= 15 mg) daily n = 38(2) Placebo gel n = 36
PGE (3 point) ldquogoodrdquo at 7 days(1) 3038(2) 2236
(1) and (2) slightly better than (3)for mean spontaneous pain at 7days
Kuehl 2011 (1) DETP 13 plaster 2 x dailyn = 207(2) Placebo plaster n = 211
No dichotomous data Percentage reduction from base-line at last application(1) 73(2) 62
Li 2013 (1) DHEP plaster (Flector) 2 xdaily n = 192(2) Placebo plaster 2 x daily n =192
ge 50 pain reduction at day 7(posthoc analysis)(1) 173192(2) 114192
Overall treatment efficacy (5point) ldquogoodrdquo or ldquoexcellentrdquo(1) 161192(2) 81192Participant and investigator ratingMean plusmn SD reduction in PI onmovement at 7 days(1) 538 plusmn 170(2) 370 plusmn 183
Linde 1985 (1) Benzydamine 3 cream 3 xdaily n = 50(2) Placebo gel n = 50
No pain on movement (walking)at 8 days(1) 3550(2) 4050
No additional data
125Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Machen 2002 (1) Ibuprofen gel 5 3 x daily n= 40(2) Placebo gel n = 41
PGE (5 point) ldquomarked improve-mentrdquo or ldquocomplete clearancerdquo at7 days(1) 2540(2) 941
Clinically meaningful (ge 30 mm)pain relief at day 7(1) 3040(2) 1641
Mahler 2003 (1) DHEP + lethicin gel 3 x 5 gdaily n = 52(2) DHEP gel 3 x 5 g daily n = 48
PGE (4 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 10 days(1) 4952(2) 3948
Mean reduction in pain on move-ment at 3 and 10 days significantlygreater with (1) than (2)
Maziegraveres 2005b (1) Ketoprofen plaster 100 mg x1 daily n = 81(2) Placebo plaster n = 82
PGE (4 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 14 days(1) 7281(2) 6082
All mean efficacy measures im-proved more for (1) than (2) mostwere statistically significant
Maziegraveres 2005a (1) Ketoprofen plaster 100 mg x1 daily n = 87(2) Placebo plaster n = 85
PGE (4 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 14 days(1) 5087(2) 4185
All mean efficacy measures im-proved more for (1) than (2) mostwere statistically significant
McLatchie 1989 (1) Felbinac gel 3 3 x 3 cm dailyn = 118(2) Placebo gel n = 113
No dichotomous data Patient daily self-assessment formean pain on rest movement atnight interference with normaland leisure activities show betterefficacy for (1) than (2) from day2 (VAS)
Morris 1991 (1) Felbinac gel 3 3 x 1 cm dailyn = 41(2) Placebo gel n = 43
PGE (5 point) ldquogoodrdquo or ldquoverygoodrdquo at 7 days(1) 2341(2) 2743
(1) better than (2) for meanimprovement in symptoms andsporting function at 7 days
NCT01255423 (1) Diclofenac sodium gel 1 x 4daily n = 104(2) Placebo gel x 4 daily n = 100
No dichotomous data VAS (mean plusmn SD) at 72 hours(baseline PI not reported)(1) 374 plusmn 252(2) 388 plusmn 241
NCT01272934 (1) Diclofenac sodium gel 1 x 4daily n = 104(2) Placebo gel x 4 daily n = 100
No dichotomous data VAS (mean plusmn SD) at 72 hours(baseline PI not reported)(1) 256 plusmn 159(2) 612 plusmn 166
NCT01272947 (1) Diclofenac sodium gel 1 x 4daily n = 104(2) Placebo gel x 4 daily n = 100
No dichotomous data VAS (mean plusmn SD) at 24 hours(baseline PI not reported)(1) 331 plusmn 214(2) 654 plusmn 169
126Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Noret 1987 (1) Ketoprofen gel 25 2 x 5 cm(75 mg) daily n = 48(2) Placebo gel n = 45
PGE (4 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 8 days(1) 3948(2) 945
Decrease in mean spontaneouspain significantly greater in (1)than (2) by 3 days
Parrini 1992 (1) Ketoprofen foam 15 3 x 2 g(200 mg) daily n = 83(2) Placebo foam n = 86
No dichotomous data Mean pain on movement andpressure significantly decreased by7 days in (1) compared with (2)
Picchio 1981 (1) Ibuprofen gel 10 3 x daily n = 20(2) Ketoprofen gel 1 3 x daily n= 20
No pain on movement at 8 days(1) 320(2) 020
Spontaneous pain ldquononerdquo at 8days(1) 620(2) 020
Predel 2004 (1) Diclofenac sodium plaster 2 xdaily (140 mgplaster) n = 60(2) Placebo plaster n = 60
PGE (4 point) ldquogoodrdquo ldquoexcellentrdquoat 7 days(1) 5560(2) 560
(1) better than (2) for reductionin tenderness pain and speed ofpain reduction
Predel 2012 (1) Diclofenac gel (VoltarenEmulgel 232) 2 x 5 cm daily n= 80(2)Diclofenac gel (Voltaren Emulgel232 ) 3 x 5 cm daily n = 80(3) Placebo gel n = 82
ge 50 red in PI on movement at5 days(1) 5780(2) 5980(3) 1782
PGE efficacy (5 point) ldquogoodrdquo orldquovery goodrdquo at 8 days(1) 6880(2) 7380(3) 2482
Predel 2013a (1) Diclofenac 4 spray gel 4 or5 sprays 3 x daily (96-120 mg di-clofenac sodium) n = 118(2) Placebo spray gel n = 114
None or slight PI on movement at7-8 days(1) 111118(2) 93114
None or slight PI on movement at3-4 days(1) 76118(2) 58114
Predel 2013b (1) Diclofenac gel (VoltarenEmulgel) 4 x 2 g daily n = 36(2) Placebo gel 4 x 2 g daily n =36
PGE (5 point) ldquogoodrdquo or ldquoexcel-lentrdquo(1) 3636(2) 736
ge 50 reduction in PI on move-ment after 48 hours(1) 3436(2) 336
Ramesh 1983 (1) Ibuprofen cream 5 3 or 4 x5-10 cm daily n = 40(2) Placebo cream n = 40
Pain on movement (4 point)ldquononerdquo or ldquoslightrdquo at 7 days(1) 2340(2) 2340
Physician global assessment at 10days ldquogoodrdquo(1) 2940(2) 1640
Rowbotham 2003 (1) Diclofenac epolamine plaster(Flector Tissuegel) 2 x daily n =191(2) Placebo plaster n = 181
Pain intensity le 210 for 2 daysor 4 consecutive evaluations by 7days(1) 75191(2) 48181
Mean pain on rest significantlybetter with (1) than (2) after 7 days
127Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Russell 1991 (1) Piroxicam gel 05 4 x 5 mgdaily n = 100(2) Placebo gel n = 100
PGE (4 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 8 days(1) 79100(2) 45100
Statistically greater reduction inmean pain on movement at 8 dayswith (1) than (2)
Saillant 1998 (1) DHEP plaster (Flector Tis-sugel 1) 1 x daily n = 70(2) Placebo plaster 1 x daily n =70
PGE (4 point) ldquoexcellentrdquo at 7 days(1) 4370(2) 2570
ge 30 decrease in PI at 7 days(1) 6470(2) 5070
Sanguinetti 1989 (1) Felbinac gel 3 3 x daily n =42(2) Placebo gel n = 40
PGE ldquogoodrdquo or ldquovery goodrdquo at 7days(1) 3442(2) 1140
(1) better than (2) by 2 days
Sinniger 1981 (1) Fentiazac cream 5 2 or 3 xdaily n = 10(2) Placebo cream n = 10
Complete pain relief within 10days(1) 710(2) 110
Improvement in active pain onmovement at 5 days(1) 67(2) 32
Spacca 2005 (1) DHEP lecithin gel (Effigel) 3x 5 g daily n = 79(2) Placebo gel n = 76
No dichotomous data Mean pain scores improved morerapidly in (1) than (2) - statisticallysignificant at 3 and 6 days
Sugioka 1984 (1) Piroxicam gel 05 3 or 4 x 1g daily n = 183(2) Indomethacin gel 1 3 or 4 x1 g daily n = 183
PGE (5 point) ldquobetterrdquo or ldquomuchbetterrdquo at 14 days(1) 85175(2) 55165
Pain on movement ldquoreducedrdquo orldquodisappearedrdquo at 7 days(1) 77175(2) 63165
Thorling 1990 (1) Naproxen gel 10 2-6 x dailyn = 60(2) Placebo gel n = 60
PGE (5 point) ldquogoodrdquo or ldquoverygoodrdquo at 7 days(1) 3860(2) 2760
Participants using naproxen im-proved more rapidly and had sig-nificantly lower severity scores byday 3
Tonutti 1994 (1) Ketoprofen gel 5 3 x 2-3 gdaily n = 15(2) Etofenamate gel 5 3 x 2-3g n = 15
PGE (4 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 7 days(1) 1015(2) 1115
Significant reductions in pain onmovement by 7 days in bothgroups
Vecchiet 1991 (1) Meclofenamic acid gel 5 2 x10 cm daily (2 g) n = 30(2) Placebo n = 30
PGE (4 point) ldquogoodrdquo or ldquoexcel-lentrdquo at 10 days(1) 3030(2) 1930
(1) significantly better than (2)for mean improvement in sponta-neous pain movement pain func-tional restriction
Whitefield 2002 (1) Ibuprofen gel 5 + placebotablet 3 x daily n = 50(2) Ibuprofen 400 mg tablet +placebo gel 3 x daily n = 50
Participant satisfied at 7 days(1) 3050(2) 3650
ldquoCompletely betterrdquo at 14 days(1) 2450(2) 3050
128Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
DHEP diclofenac epolamine HCl hydrochloride n number PGE participant global evaluation PI pain intensity SD standarddeviation VAS visual analogue scale
Appendix 5 Summary of outcomes adverse events and withdrawals
Study ID Treatment Local AEs Systemic AEs Serious AEs Withdrawals
Airaksinen 1993 (1) Ketoprofen gel 2x 5 g (125 mg) dailyn = 29(2) Placebo gel n =27
(1) 529(2) 427
(1) 129 (nausea af-ter paracetamol)(2) 027
None AE noneOther nonereported
kermark 1990 (1)Indomethacin spray1 (Elmetacin) 3-5 x 05-15 mL dailyn = 23(2) Indomethacincapsules 3 x 25 mgdaily n = 23(3) Placebo sprayand capsules n = 24
(1) 422(2) 022(3) 024
No usable data -reported for eventsnot participants
None reported AE (1) 1 (2) 1 (3)0Lost to follow-up(1) 1 (2) 2 (3) 3
Aoki 1984 (1) Piroxicam gel5 3 or 4 x 1 gdaily n = 84(2) Indomethacingel 1 3 or 4 x 1 gdaily n = 84(3) Placebo gel n =84
(1) 179(2) 270(3) 274
None None reported AE none23 excluded for pro-tocol violations (1)7 (2) 7 (3) 926 withdrew for rea-sons unrelated totreatment (1) 5 (2)13 (3) 8
Auclair 1989 (1) Niflumic acidgel 25 3 x 5 gdaily n = 117(2) Placebo gel n =110
All AEs(1) 5123(2) 6116Most commonly cu-taneous eruptions
No usable data None reported AE (1) 1123 (2) 011626 excluded fromefficacy analysis fornot meeting entrycriteria and protocolviolations
Billigmann 1996 (1) Ibuprofen mi-crogel 5 3 x 200mg daily n = 80(2) Placebo gel n =
(1) 1180(2) 480
None reported None reported AE (1) 280 (aller-gic rash dermatitis)No reason given (1)380 (2) 580
129Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
80 Symptom-free (1)180 (2) 180
Campbell 1994 (1) Ibuprofen cream5 (Proflex) 4 x 4rdquodaily n = 26(2) Placebo cream n= 25
No data (1) 126 (headache)(2) 025
No data AE noneExclusions 493 presented late 2missing forms 1 ap-peared twice 43 didnot return diaries
Chatterjee 1977 (1) BenzydamineHCl cream 3 3 xdaily n = 25(2) Placebo cream n= 25
None None None AE none1 participant lostto follow-up (groupnot reported)
Costantino 2011 (1) DHEP-hep-arin plaster (Flector-parin) x 1 daily n =142(2) DHEP plaster(Flector) x 1 daily n= 146(3) Placebo plastern = 142
Possibly or probablydrug-related(1) 2142(2) 1145(3) 6142
All mild and goneby 14 days ex-cept 1 (moderate) inplacebo group
Not reportedNo systemic GIevents
None AE noneOther(1) 0142(2) 2146 (proto-col deviation lost tofollow-up)(3) 2142 (lost tofollow-up)
Coudreuse 2010 (1) DHEP-hep-arin plaster (Flector-parin) n = 120(2) Placebo n = 120
(1) 1117(2) 7116
All AEs mild ormoderate resolvedspontaneously
(1) 1117(2) 0116
None AE(1) 1117 (increasedoedema)(2) 0116Other(1) 6117 (recovery3 no reason 3)(2) 3116 (recovery1 no reason 2)
Curioni 1985 (1) Ibuproxam n =20(2) Ketoprofen n =20(3) Etofenamate n =20
None None None AE noneOther none
Diebshlag 1990 (1) Ketorolac gel2 3 x 3 g daily n =13(2) Etofenamate gel5 3 x 3 g daily n =12
(1) 113(2) 112(3) 012
None None AE none1 ketorolac partic-ipant did not at-tend 15 day follow-up due to car acci-
130Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
(3) Placebo gel n =12
dent
Dreiser 1988 (1) Ibuprofen cream5 3 x 4 cm dailyn = 32 (3 x 10 cmfor large joints)(2) Placebo cream n= 32
No usable data None Not reported AE none4 placebo partici-pants lost to follow-up
Dreiser 1989 (1) Ketoprofen gel25 2 x 5 cm dailyn = 30(2) Placebo gel n =30
(1) 030(2) 230
None None reported AE (2) 230 (intol-erance)LoE (1) 130 (2) 130
Dreiser 1990 (1) Niflumic acidgel 25 3 x 5 gdaily n = 30(2) Placebo gel n =30
(1) 030(2) 330
None None AE (2) 130 (ery-thema)Exclusion 1 from(2) from efficacyanalysis for inade-quate baseline pain
Dreiser 1994 (1)Flurbiprofen plaster2 x 40 mg daily n =65(2) Placebo plastern = 66
(1) 265(2) 066
None None AE 0(1) 165 excludedfrom efficacy analy-sis for protocol vio-lation(2) 266 (1 LoE 1cured)
Fioravanti 1999 (1) DHEP lecithingel 3 x 5 g (= 65 mg)daily n = 50(2) DHEP gel 3 x 5g (= 65 mg) daily n= 50
(1) 050(2) 150
No data None reported AE noneOther none
Fujimaki 1985 (1) Piroxicam gel 05 3 or 4 x 1 g dailyn = 92(2) Indomethacingel 1 3 or 4 x 1 gdaily n = 90(3) Placebo gel n =89
(1) 183(2) 582(3) 282
(1) 083(2) 182 (nauseaand vomiting)(3) 082
None AE (1) 0 (2) 4 (3)0Unknown reasons(1) 2 (2) 1Did not return af-ter 1st visitirregularvisits (1) 6 (2) 6(3) 7
Gallacchi 1990 (1) Diclofenac hy-droxyethylpyrro-lidine gel 1 4 x 2
No AEs None None AE noneOther none
131Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
g daily n = 25 (Flec-tor gel)(2) Di-clofenac sodium 14 x 2 g daily n = 25(Voltaren Emugel)
Gonzaacutelez de Vega2013
(1) Traumeel gel 3 x2 g daily n = 140(2) Traumeel oint-ment n = 143(3) Diclofenac gel1 3 x 2 g daily n =137
No usable data
Infrequent mild tomoderate
No usable data None AE noneOther(1) 11140(2) 12137
Governali 1995 (1) Ketoprofen gel5 3 x 2-3 g dailyn = 15(2) Ketoprofencream 1 3 x 2-3 gdaily n = 15
No side effects None None AE noneOther none
Gualdi 1987 (1) Flunaxaprofengel 2 x 3-5 cm dailyn = 30(2) Ketoprofen gel 2x 3-5 cm daily n =30
(1) 130(2) 330
No data None reported AE noneOther none
Haig 1986 (1) Benzydaminecream 3 6 x dailyn = 21(2) Placebo cream n= 22
No AEs reported None None reported AE none reportedOther no data
Hoffmann 2012 (1) DHEP-hep-arin plaster (Flector-parin) x 1 daily n =121(2) DHEP (Flector)plaster x 1 daily n =115(3) Placebo plastern = 118
Most AEs were mi-nor local reactions(eg pruritus anderythema) in areaof plaster of mildto moderate inten-sity and resolvedwithout interrupt-ing treatment
No treatment-related systemic AEsrecorded
(2) 1 participanthad 3 SAEs nonejudged related tostudy medication
AE noneExclusions(1) 5121(2) 10115(3) 7118Excluded fromper protocol analysisdue to poor compli-ance or personal de-cision
Hofman 2000 (1) Diclofenacsodium gel 1 4 x2 cm daily n = 69(2) Lysine clonixi-
(1) 158(2) 161
None None AE noneLoE (1) 9 (2) 8
132Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
nate gel 5 4 x 2cm (225 mg) dailyn = 73
Hosie 1993 (1) Felbinac foam3 3 x 2 g daily+ placebo tablets 3x 1 daily n = 140(127 analysed for ef-ficacy)(2) Ibupro-fen tablets 3 x 400mg daily + placebofoam 3 x 2 g daily n= 147 (134 analysedfor efficacy)
(1) 1127(2) 3134
GI events (1) 14127 (2) 11134For (1) more mildnone definitely drugrelated for(2) definitely relatedto study drug
None AE noneExclusions (1) 13(2) 13 did not returnfor 7 day follow-up
Jenoure 1997 (1) DHEP plaster(Tissugel) 2 x dailyn = 44(2) Placebo plaster 2x daily n = 41
(1) 144(2) 141
No data None reported AE none reportedOther nonereported
Joussellin 2003 (1) DHEPplaster (Flector Tis-sugel 1) 1 x dailyn = 68(2) Placebo plaster 1x daily n = 66
(1) 168 (pruritus)(2) 366 (pruritus 2burning 1)
All AEs mild ormoderate
(1) 168 (allergic re-action)(2) 066
None reported AE(1) 066(2) 166Other(1) 366(2) 266
Julien 1989 (1) Ketoprofen gel25 2 x 5 cm (= 50mg) daily n = 30(2) Placebo gel n =30
(1) 130(2) 030
Not reported None AE noneOther none
Klainguti 2010 (1) DHEP-hep-arin plaster (Flector-parin) x 1 daily n =62(2) DHEP plaster(Flector) x 1 daily n= 61(3) Placebo n = 59
(1) 062(2) 061(3) 159All AEs mild in na-ture and resolvedspontaneously
(1) 165 (facial in-fection)(2) 161 (abdominalpain)(3) 059
None AE noneOther(1) 365(2) 161(3) 359
Kockelbergh 1985a (1) Ketoprofen gel25 2 x 5 cm (= 15mg) daily n = 38(2) Placebo gel n =
(1) 138(2) 126
Not reported None AE noneOther none
133Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
36
Kuehl 2011 (1) DETP 13plaster 2 x daily n =207(2) Placebo plastern = 211
(1) 16207(2) 12211
(1) 15207(2) 23211
None AE(1) 4207(2) 9211LoE(1) 21207(2) 25211Other(1) 58207(2) 62211
Li 2013 (1) DHEP plaster(Flector) 2 x daily n= 192(2) Placebo plaster 2x daily n = 192
(1) 4192(2) 3192
(1) 10192(2) 4192
None AE(1) 2192(2) 0192Other(1) 5192(2) 1192
Linde 1985 (1) Benzydamine3 cream 3 x dailyn = 50(2) Placebo gel n =50
(1) 440(2) 241
None None AE none(1) 6 (2) 6 excludedfrom 1st assessment(1) 3 (2) 4 ex-cluded from final as-sessment
Machen 2002 (1) Ibuprofen gel5 3 x daily n = 40(2) Placebo gel n =41
(1) 440(2) 241
None None AE none(1) 1 LoE 1 proto-col violation(2) 4 LoE
Mahler 2003 (1) DHEP + lethicingel 3 x 5 g daily n =52(2) DHEP gel 3 x 5g daily n = 48
(1) 152(2) 048
(1) 152(2) 048
None AE none5 lost to follow-up
Maziegraveres 2005b (1) Ketoprofen plas-ter 100 mg x 1 dailyn = 81(2) Placebo plastern = 82
At 21 days(1) 1281(2) 682
(1) 1381(2) 1482
None AE (1) 381(1) 781 (1 LoE 6cured)(2) 782 (5 LoE 2cured)
Maziegraveres 2005a (1) Ketoprofen plas-ter 100 mg x 1 dailyn = 87(2) Placebo plastern = 85
At 21 days(1) 2987(2) 2785
(1) 1187(2) 785
None AE (1) 987 (2) 685(1) 687 (2 LoE 4cured)(2) 585 (4 LoE 1cured)
134Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
McLatchie 1989 (1) Felbinac gel 33 x 3 cm daily n =118(2) Placebo gel n =113
(1) 3118(2) 2113Mild transient localirritation
None reported None AE noneOther none
Morris 1991 (1) Felbinac gel 33 x 1 cm daily n =41(2) Placebo gel n =43
None None None AE none(1) 4 (protocol vio-lations)(2) 1 (lost to follow-up)Exclusions 11 fromefficacy analysis be-cause evaluated by4 different investi-gators
NCT01255423 (1) Diclofenacsodium gel 1 x 4daily n = 104(2) Placebo gel x 4daily n = 100
(1) 1104(2) 3102
Total AE(1) 11104(2) 8102
None None
NCT01272934 (1) Diclofenacsodium gel 1 x 4daily n = 104(2) Placebo gel x 4daily n = 100
(1) 1102(2) 0103
Total AE (excludingSAE)(1) 6102(2) 3103
(1) 0102(2) 1103 (rupturedligaments in wrist)
AE see SAE
NCT01272947 (1) Diclofenacsodium gel 1 x 4daily n = 104(2) Placebo gel x 4daily n = 100
None (1) 2104(2) 2100
None None
Noret 1987 (1) Ketoprofen gel25 2 x 5 cm (75mg) daily n = 48(2) Placebo gel n =45
(1) 151(2) 047
None reported Not reported AE (1) 151 (skinallergy)(1) 1 LoE 1 unre-lated to trial(2) 1 LoE 1 unre-lated to trial
Parrini 1992 (1) Ketopro-fen foam 15 3 x 2g (200 mg) daily n= 83(2) Placebo foam n= 86
None None None AE noneOther none
135Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Picchio 1981 (1) Ibuprofen gel10 3 x daily n =20(2) Ketoprofen gel1 3 x daily n = 20
None None None AE noneOther not reported
Predel 2004 (1) Diclofenacsodium plaster 2 xdaily (140 mgplas-ter) n = 60(2) Placebo plastern = 60
12 participants ex-pe-rienced 16 mild AEswith no differencesbetween groups
None None AE (1) 160Other none
Predel 2012 (1) Diclofenac gel(Voltaren Emulgel232) 2 x 5 cmdaily n = 80(2) Diclofenac gel(Voltaren Emulgel232) 3 x 5 cmdaily n = 80(3) Placebo gel n =82
(1) 080(2) 180(3) 182All mild to moder-ate
(1) 280(2) 180(3) 382
None AE(1) 080(2) 080(3) 182Other(1) 180(2) 280(3) 282(protocol violationslost to follow-up)
Predel 2013a (1) Diclofenac 4spray gel 4 or 5sprays 3 x daily (96-120 mg diclofenacsodium) n = 118(2) Placebo spraygel n = 114
(1) 1118(2) 4114
(1) 5118(2) 4114All AEs mild re-versible
None AE(1) 1118(2) 1114Other(1) 3118(2) 43114
Predel 2013b (1) Diclofenac gel(Voltaren Emulgel)4 x 2 g daily n = 36(2) Placebo gel 4 x 2g daily n = 36
None (1) 036(2) 136
None AE noneOther none
Ramesh 1983 (1) Ibuprofen cream5 3 or 4 x 5-10 cmdaily n = 40(2) Placebo cream n= 40
(1) 140(2) 140
None reported Not reported AE (1) 140 (2) 140Other none
Rowbotham 2003 (1) DHEP plaster(Flector Tissuegel) 2x daily n = 191(2) Placebo plastern = 181
(1) 27191 (pruritis14)(2) 31181 (pruritis21)
(1) 21191(2) 22181
None re-ported (ldquovast major-ity mildrdquo)
AE none(1) 3191 (2) 4181(did not finish trialand complete dailydiaries)
136Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Russell 1991 (1) Piroxicam gel 05 4 x 5 mg daily n= 100(2) Placebo gel n =100
(1) 4102(2) 10102
GI or CNS events(1) 4 (2) 7Any AE (1) 7102(2) 15102
None reported AE (1) 1102 (2) 8102Other (1) 6 LoE 1ldquootherrdquo(2) 42 LoEExclusions7 did not complywith study medica-tion schedule 6 lostto follow-up 1 pro-tocol violation
Saillant 1998 (1) DHEPplaster (Flector Tis-sugel 1) 1 x dailyn = 70(2) Placebo plaster 1x daily n = 70
None None None AE noneOther(1) 570(2) 570
Sanguinetti 1989 (1) Felbinac gel 33 x daily n = 42(2) Placebo gel n =40
(1) 342(2) 140
None None reported AE noneOther nonereported
Sinniger 1981 (1) Fentiazac cream5 2 or 3 x daily n= 10(2) Placebo cream n= 10
ldquoNo untoward sideeffectsrdquo
None None AE noneOther nonereported
Spacca 2005 (1) DHEP lecithingel (Effigel) 3 x 5 gdaily n = 79(2) Placebo gel n =76
ldquoNo signsof cutaneous irrita-tion or sensitisationobservedrdquo
No AEs observed None AE noneOther nonereported
Sugioka 1984 (1) Piroxicam gel 05 3 or 4 x 1 g dailyn = 183(2) Indomethacingel 1 3 or 4 x 1 gdaily n = 183
(1) 5178(2) 26179
None reported None reported AE none reportedExclusions due toprotocol violations(1) 8 (2) 18Withdrawals(1) 11 (2) 12
Thorling 1990 (1) Naproxen gel10 2-6 x daily n =60(2) Placebo gel n =60
(1) 160(2) 060
None None AE none(1) 1 LoE 1 proto-col violation(2) 1 participant re-quest
137Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Tonutti 1994 (1) Ketoprofen gel5 3 x 2-3 g dailyn = 15(2) Etofenamate gel5 3 x 2-3 g n =15
None No AEs attributableto the medication
None AE NoneLoE (1) 1 (2) 2
Vecchiet 1991 (1) Meclofe-namic acid gel 5 2x 10 cm daily (2 g)n = 30(2) Placebo n = 30
Tolerabil-ity excellent or goodin nearly all partici-pants
No data None AE none reported(2) 5 lost to follow-up
Whitefield 2002 (1) Ibuprofen gel5 + placebo tablet3 x daily n = 50(2) Ibuprofen 400mg tablet + placebogel 3 x daily n = 50
No data 6 AEs reportednone judged relatedto study medication
None reported AE noneRecovered (1) 3 (2)2LoE (2) 1Lost to follow-up(1) 1 (2) 1
AE adverse event CNS central nervous system DHEP diclofenac epolamine GI gastrointestinal HCl hydrochloride LoE lackof efficacy n number SAE serious adverse event
Appendix 6 Concentration amount and frequency of dosing
Study Drug Concentration Quantity Frequency Estimated daily dose of topi-cal NSAID
Joussellin 2003 Diclofenac 1 Plaster 1 180 mg epolamine salt 140 mgNa salt
Li 2013 Diclofenac 1 Plaster 2 360 mg epolamine salt 280 mgNa salt
Rowbotham 2003 Diclofenac 1 Plaster 2 360 mg epolamine salt 280 mgNa salt
Saillant 1998 Diclofenac 1 Plaster 1 180 mg epolamine salt 140 mgNa salt
Coudreuse 2010 Diclofenac 1 Plaster 1 180 mg epolamine salt 140 mgNa salt
Klainguti 2010 Diclofenac 1 Plaster 1 180 mg epolamine salt 140 mgNa salt
138Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Predel 2004 Diclofenac - Plaster 2 280 mg Na salt
Predel 2012 Diclofenac 232 5 cm ribbon~ 2 g 2 or 3 92-138 mg (Na equiv) as di-ethylamine salt
Predel 2013b Diclofenac 116 2 g 4 92 mg (Na equiv) as diethy-lamine salt
Predel 2013a Diclofenac 4 4-5 sprays 3 96-120 mg Na salt
Costantino 2011 Diclofenac 1 Plaster 1 180 mg epolamine salt 140 mgNa salt
Fioravanti 1999 Diclofenac - 5 g 3 195 mg epolamine salt
Gallacchi 1990 Diclofenac 1 2 g 4 80 mg
Gonzaacutelez de Vega2013
Diclofenac 1 6 cm 2g 3 60 mg
Hoffmann 2012 Diclofenac 1 Plaster 1 180 mg epolamine salt 140 mgNa salt
Hofman 2000 Diclofenac 1 2 cm 4 probably 30-40 mg Na salt
Jenoure 1997 Diclofenac 1 Plaster 2 360 mg epolamine salt 280 mgNa salt
Mahler 2003 Diclofenac - 5 g 3 195 mg epolamine salt
NCT01255423 Diclofenac 1 - 4 -
NCT01272934 Diclofenac 1 - 4 -
NCT01272947 Diclofenac 1 - 4 -
Spacca 2005 Diclofenac - 5 g 3 195 mg epolamine salt
Campbell 1994 Ibuprofen 5 4 inch ribbon 4 Assume 800 mg
Dreiser 1988 Ibuprofen 5 4 cm ribbon(10 cm for largerjoints)
4 Assume up-800 mg
Ramesh 1983 Ibuprofen 5 5-10 cm ribbon 3-4 Assume 300-800 mg
Billigmann 1996 Ibuprofen 5 10 cm 4 g gel 3 600 mg
139Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Machen 2002 Ibuprofen 5 - 3 -
Picchio 1981 Ibuprofen 10 - 3 -
Whitefield 2002 Ibuprofen 5 - 3 -
Maziegraveres 2005b Ketoprofen - Plaster 1 100 mg
Maziegraveres 2005a Ketoprofen - Plaster 1 100 mg
Airaksinen 1993 Ketoprofen - 5 g 2 125 mg
Dreiser 1989 Ketoprofen 25 5 cm 2 100 mg
Julien 1989 Ketoprofen 25 5 cm 2 100 mg
Noret 1987 Ketoprofen 25 5 cm 75 g 2 375 mg
Curioni 1985 Ketoprofen No details 2 -
Governali 1995 Ketoprofen 5 2-3 g 3 300-450 mg
Governali 1995 Ketoprofen 1 2-3 g 3 60-90 mg
Gualdi 1987 Ketoprofen - 3-5 cm 2 -
Kockelbergh 1985 Ketoprofen 25 5 cm 2 100 mg
Parrini 1992 Ketoprofen 150 2 g 3 600 mg
Picchio 1981 Ketoprofen 1 - 3 -
Tonutti 1994 Ketoprofen 5 2-3 g 3 300-450 mg
kermark 1990 Indomethacin 1 05-15 mL 3-5 12-60 mg
Aoki 1984 Indomethacin 1 1 g 3-4 30-40 mg
Fujimaki 1985 Indomethacin 1 1 g 3-4 30-40 mg
Sugioka 1984 Indomethacin 1 1 g 3-4 30-40 mg
Aoki 1984 Piroxicam 05 1 g 3-4 15-20 mg
Fujimaki 1985 Piroxicam 05 1 g 3-4 15-20 mg
140Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
(Continued)
Russell 1991 Piroxicam 05 5 mg 4 20 mg
Sugioka 1984 Piroxicam 05 1 g 3-4 15-20 mg
Chatterjee 1977 Benzydamine 3 - 3 -
Haig 1986 Benzydamine 3 - 6 -
Linde 1985 Benzydamine 3 - 3 -
Auclair 1989 Niflumic acid 25 10 cm 5 g 3 375 mg
Dreiser 1990 Niflumic acid 25 10 cm 5 g 3 375 mg
Curioni 1985 Ibuproxam gel 10 - 2 -
Curioni 1985 Etofenamate No details - 2 -
Diebshlag 1990 Etofenamate 5 3 g 3 450 mg
Tonutti 1994 Etofenamate 5 2-3 g 3 300-450 mg
Diebshlag 1990 Ketorolac 2 3 g 3 360 mg
Dreiser 1994 Flurbiprofen Patch 2 80 mg
Gualdi 1987 Flunoxaprofen - 3-5 cm 2 -
Hofman 2000 Lysine clonixinate 5 2 cm 4 90 mg
Hosie 1993 Felbinac 3 2 g 3 180 mg
McLatchie 1989 Felbinac 3 3 cm 3 -
Morris 1991 Felbinac 3 1 cm 3 -
Sanguinetti 1989 Felbinac 3 - 3 -
Sinniger 1981 Fentiazac 5 Varied according toinvolved areas
2-3 -
Thorling 1990 Naproxen 10 - 2-6 -
Vecchiet 1991 Meclofenamic acid 5 10 cm 4 g 2 400 mg
equiv equivalent Na sodium
141Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
F E E D B A C K
Query on formulations of topical NSAIDs particularly DMSO from Dr Chrubasik 11 April 2012
Summary
Dr Chrubasik highlighted this letter to the Editor postgradmedorgdoi103810pgm2011092482
DMSO [dimethyl sulphoxide] but also other additives eg nonivamide (which is a capsaicinoid added as drug enhancer) may contributeto the overall effect of topical NSAIDs Nonivamide certainly contributes to the analgesic effect and to adverse events (heat sensationburning pruritus etc) This has not been considered in the Cochrane review by Massey T Derry S Moore RA McQuay HJ TopicalNSAIDs for acute pain in adults but Dr Chrubasik believes should be done otherwise the effect size of the NSAID topicals is favoured
Reply
We have been asked by Dr Chrubasik to comment on a letter (Roth 2011) about the formulations of topical NSAIDs particularly howDMSO and other penetration enhancers can affect efficacy estimates or adverse event reporting in osteoarthritis It was suggested thatthe review of Topical NSAIDs for acute pain in adults did not consider this resulting in a bias towards the topical NSAIDThere are a number of points to be made here
1 Penetration enhancers are used in formulations of topical products to encourage local absorption through the skin and produce ahigh local concentration Topical NSAIDs use penetration enhancers and the result is high local concentration in joints for instancebut low systemic concentrations (Moore 2008) That is how they work Formulation is an important part of medicinal chemistry as awhole not just for topical agents
2 In our analysis of topical NSAIDs we were aware that a range of properties are or have been ascribed to the analgesia resultingfrom application of topical agents and which could contribute to overestimation of treatment effect of topical NSAID These includefeelings of heat or cold and even the act of rubbing itself For that reason we have chosen to include only double-blind studies wherethe placebo agent is identical to the active with the exception of course of the NSAID So heat cold rubbing and penetrationenhancers should be identical as best we can judge That leaves only the NSAID itself to provide any additional analgesic effect andit is that which we measure This is analogous for example to use of acupuncture say where the better studies show no differencebetween ldquotruerdquo acupuncture and ldquoshamrdquo acupuncture performed at nonspecific sites but better than non-treatment controls Theargument that we should only use high quality studies to evaluate evidence about pain interventions is well made
3 Overestimation of analgesic effect because of effects of enhancers themselves would be better made in direct comparisons oftopical and oral NSAIDs where local or even systemic effects would not be balanced in the oral study arm However our reviewconcentrated on placebo-controlled studies and had few studies with active controls Moreover the real test would be in chronicrather than acute conditions with long duration (12 week) outcomes using current best evidence rules (Moore 2010) includingimputation (Moore 2012) In their response to Rothrsquos letter the authors of the original review of products available in the USA showrather similar effect sizes of oral diclofenac and topical diclofenac with different penetration enhancers (Barthel 2011) in such studies
4 The Roth letter sought to differentiate between topical diclofenac preparations based on the penetration enhancers used Thatdifferent formulations may have different effect sizes is a fair point to make Two of the studies in our review of topical NSAIDs inacute conditions used diclofenac sodium 1 gel comparing it with either diclofenac epolamine gel (Gallacchi 1990 50 participants)or lysine clonixinate gel (Hofman 2000 142 participants) no difference between formulations was demonstrated It is difficult tomake any judgement for topical NSAIDs in acute conditions due to the relatively small number and particularly the small size ofstudies We did an analysis by drug and this showed that some topical NSAIDs were consistently beneficial irrespective offormulation while others had little or no efficacy This fits in with some theoretical considerations of molecular architecture andtissue penetration (Moore 2008) In chronic pain where there are larger studies and much more data we have considered formulation(Derry in preparation)
References
Roth SH Letter to the editor The importance of differentiating between topical NSAIDs Postgraduate Medicine 2011123251-2[103810pgm2011092482]Moore RA Derry S McQuay HJ Topical agents in the treatment of rheumatic pain Rheumatic Diseases Clinics of North America200834415-32
142Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Moore RA Eccleston C Derry S Wiffen P Bell RF Straube S et al ACTINPAIN Writing Group of the IASP Special Interest Groupon Systematic Reviews in Pain Relief Cochrane Pain Palliative and Supportive Care Systematic Review Group Editors ldquoEvidencerdquo inchronic pain - establishing best practice in the reporting of systematic reviews Pain 2010150386-9Moore RA Straube S Eccleston C Derry S Aldington D Wiffen P et al Estimate at your peril imputation methods for patientwithdrawal can bias efficacy outcomes in chronic pain trials using responder analyses Pain 2012153265-8Barthel HR Axford-Gatley RA Response to Roth Letter to the Editor Postgraduate Medicine 2011123253-4 [103810pgm2011092483]
Contributors
Feedback from Sigrun ChrubasikAuthors involved with responding Andrew Moore Sheena DerryFeedback Editor Kate Seers
Query submitted by Peter C Goslashtzsche 3 November 2015
Summary
Date of Submission 03-Nov-2015Name Peter C GoslashtzscheEmail Address pcgcochranedkAffiliation Nordic Cochrane CentreRole DirectorComment The authors found that the results were missing from 5900 patients Furthermore there was extreme heterogeneity in theirmeta-analyses eg I square was 92 for the diclofenac trials which were the most common ones and there was extreme funnel plotasymmetry with the largest trials showing the smallest effects (the authors didnrsquot show funnel plots but I constructed one for diclofenac)Moreover the trials were industry funded of relatively poor quality and the authors analysed published data not data from clinicalstudy reports and did not try to obtain all the missing trials and data from the manufacturersThe authors concluded that topical NSAIDs are effective in providing pain relief but also cautioned that the large amounts of unpublisheddata ldquocould influence results in updates of this reviewrdquo They certainly could I believe it is plain wrong to perform meta-analyses on theauthorsrsquo data When I most recently reviewed this area for the BMJ in 2010 I concluded that we donrsquot know whether topical NSAIDsare beneficial (1)1 Goslashtzsche PC NSAIDs Clin Evid (Online) 2010 Jun 28I agree with the conflict of interest statement belowI certify that I have no affiliations with or involvement in any organization or entity with a financial interest in the subject matter ofmy feedback
Reply
Response submitted by authors Sheena Derry and Professor Andrew Moore on 4 November 2015Peter has made these same points elsewhere [PubMed Commons PubPeer gt Cochrane Database Syst Rev October 2015 httpspubpeercompublicationsA9E5BEA36549727357F9FD14CC2537]As a preliminary it is important to stress that the number of trials available in pain has increased over time and newer trials are typicallybetter and larger This means we can move from answering the simple question of whether an intervention works to more importantand relevant questions such as how well the intervention works and for drugs particularly to examine effects of dose and formulationFormulation is now recognised to have profound effects For oral analgesics for example fast acting formulations demonstrate up todouble the analgesic effect for a given dose as our recently updated overview points out (Cochrane Database Syst Rev 2015 Sep 289CD008659 see also Pain 2014 Jan155(1)14-21)What was true for oral analgesics in acute pain is now true also for topical NSAIDs in acute pain where manufacturers have beenputting a lot of effort into trying to produce new formulations that work better This updated review sought to examine the influenceof formulation and drug although dose is somewhat more difficult There are good reasons why formulation may be important(Rheumatic Disease Clinics 2008 Vol 34 Issue 2 p 415-432)
143Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Our searches did identify a large number of unpublished trials We deplore this but are powerless to change things For a previousreview of topical NSAIDs we contacted all identified manufacturers and asked for published or unpublished data The yield was small(as others have found) but some unpublished studies were brought into the public domain Waiting for all studies and clinical trialreports of all studies would take forever and could probably never be achieved We believe that many of these unpublished studiesrelate to drugs andor formulations that have never been manufactured commercially While these would be of interest in determiningwhat does and doesnrsquot work and to direct future research they would probably have little clinical relevance because these formulationsare unlikely to come to marketPeterrsquos main issue appears to be heterogeneity Tests for heterogeneity are problematical anyway (Pain 2000 85415-24) and the I squareof 92 that he quotes was for all topical diclofenac formulations combined While we do give an overall summary for diclofenac wedemonstrated in the review that the different diclofenac formulations produced different results from one another using LrsquoAbbeacute plotsand in our detailed analyses showing large variations in efficacy between them In the circumstance a high I square for all combined(clinical heterogeneity) is to be expected but it is not relevant The bulk of the studies on diclofenac were published in the last fiveyears were of decent quality and moderate to large size There were older data for ketoprofen but again these showed major differencesbetween formulations Differences between formulations are highlighted throughout the reviewTrying to determine publication bias using funnel plots or other measures is something of a lost cause (Journal of Clinical Epidemiology200053207-16) It is especially so with small numbers of trials (Journal of Clinical Epidemiology 2000 53 477-484) and makingsense of funnel plots is anything but easy for most people (Journal of Clinical Epidemiology 2005 58 894-901) A useless methodseems an odd choice to make to criticise our reviewThere are very good scientific reasons why drug and formulation may play a big part in the efficacy of topical NSAIDs This is alsothe case for oral analgesics used in acute pain where formulation improvements generating rapid absorption confers greater efficacySimple lumping strategies may have been permissible in past systematic review methodology but a more forensic approach is needednow and in the future This is what we have attempted to do in this latest reviewComparisons with Peterrsquos 2010 review seem inappropriate since that review was based on other reviews that are now out of date
Contributors
Feedback from Peter C GoslashtzscheAuthors involved with responding Andrew Moore Sheena DerryFeedback Editor Kate Seers
W H A T rsquo S N E W
Last assessed as up-to-date 4 February 2015
Date Event Description
4 November 2015 Feedback has been incorporated Feedback has been incorporated See Feedback 2
H I S T O R Y
Protocol first published Issue 4 2008
Review first published Issue 6 2010
144Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
Date Event Description
13 October 2015 Amended Small errors found in Summary Table B for total num-ber of participants with piroxicam and indomethacinand in text for NNT with diclofenac (other gel) Nochange to conclusions
6 October 2015 Amended Small error found in Summary Table B percentagesfor NSAID and placebo were the wrong way round forcomparison of all NSAIDs and placebo RR and textwere correct No change to results or conclusions
13 February 2015 New citation required but conclusions have notchanged
Conclusions not changed Results remain essentiallythe same but the focus has changed to examine drugand formulation combination
3 February 2015 New search has been performed Title changed from Topical NSAIDs for acute pain inadults to Topical NSAIDs for acute musculoskeletalpain in adults to increase specificityNew searches run and new studies identified Fourteennew included studies using diclofenac (3489 new par-ticipants a 63 increase) four new excluded studiesFifteen studies awaiting classification (completed noresults available but likely to satisfy inclusion criteria)
23 May 2014 Amended Error in data reported for clinical success in Hosie 1993was brought to our attention and has been corrected
12 June 2012 Feedback has been incorporated We have incorporated feedback received from DrSigrun Chrubasik and the authorrsquos response on DMSOand other additives
C O N T R I B U T I O N S O F A U T H O R S
For the earlier review Tom Massey and SD identified studies and carried out data extraction analysis and drafting RAM and HJMwere involved in planning acted as adjudicators and were involved with writing
For this update SD and RAM carried out searches data extraction and analysis All authors were involved with writing the review
145Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
D E C L A R A T I O N S O F I N T E R E S T
SD has no conflicts relating to this review or any similar product
RAM has no conflicts relating to this review or any similar product
PW has no conflicts relating to this review or any similar product
HG has no conflicts relating to this review or any similar product
MM has no conflicts relating to this review or any similar product
S O U R C E S O F S U P P O R T
Internal sources
bull Oxford Pain Relief Trust UKGeneral institutional support
External sources
bull No sources of support supplied
D I F F E R E N C E S B E T W E E N P R O T O C O L A N D R E V I E W
For this update in 2015 we have changed the title to specify musculoskeletal pain because topical NSAIDs are not normally used totreat visceral pain or headache We felt that the new title better reflected the content of the review We have also changed the focus ofthe review from pooled analysis of all topical NSAIDs and all studies of a particular NSAID to an examination of individual drug andits formulation This makes the review much more relevant We have expanded the rsquoRisk of biasrsquo assessment and added a rsquoSummaryof findingsrsquo table and PRISMA flow chart We have removed a number of sensitivity analyses because they were not appropriate giventhe current information on the impact of formulation on efficacy The sensitivity analyses have been superseded by the rsquoRisk of biasrsquoassessment and taken into account in the rsquoSummary of findingsrsquo tables
An earlier review in 2004 chose to exclude studies using benzydamine on the grounds that it was no longer considered to be an NSAID(Mason 2004a) Although the protocol for this review stated that we would not include benzydamine after further consultation wenow believe that it should be classified as an NSAID albeit with a different mode of action which is not fully understood (Quane1998) Thus we have reinstated studies using topical benzydamine
I N D E X T E R M S
Medical Subject Headings (MeSH)
Acute Disease Administration Topical Anti-Inflammatory Agents Non-Steroidal [lowastadministration amp dosage adverse effects] AthleticInjuries [drug therapy] Pain [lowastdrug therapy] Randomized Controlled Trials as Topic Sprains and Strains [drug therapy]
146Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd
MeSH check words
Adult Humans
147Topical NSAIDs for acute musculoskeletal pain in adults (Review)
Copyright copy 2015 The Cochrane Collaboration Published by John Wiley amp Sons Ltd