Opioids in chronic neuropathic pain - A systematic review ...

1

Opioids in chronic neuropathic pain - A systematic review and meta-analysis of

efficacy, tolerability and safety in randomized placebo-controlled studies of at least

four weeks duration

C. Sommer 1, P. Welsch 2, P. Klose 3, R. Schaefert 4, F. Petzke 5, W. Häuser 6,7

1 Neurologische Klinik, Universitätsklinikum Würzburg, Würzburg, Germany

2 Stichting Rugzorg Nederland, Ede, The Netherlands

3 Abteilung für Naturheilkunde und Integrative Medizin, Kliniken Essen-Mitte, Essen,

Germany

4 Klinik für Allgemeine Innere Medizin und Psychosomatik, Universitätsklinikum

Heidelberg, Heidelberg, Germany

5 Schmerz-Tagesklinik und Ambulanz, Universitätsmedizin Göttingen, Göttingen,

Germany

6 Klinik und Poliklinik für Psychosomatische Medizin und Psychotherapie, Technische

Universität München, Germany

7 Innere Medizin I, Klinikum Saarbrücken gGmbH, Germany

Korrespondenzadresse:

PD Dr. med. Winfried Häuser

Klinikum Saarbrücken gGmbH

Innere Medizin 1

Winterberg 1

D-66119 Saarbrücken

Germany

Tel: +49 681 9632020

Fax: +49 681 9632022

Email: [email protected]

mailto:[email protected]

2

Background: The efficacy and safety of opioid therapy in chronic neuropathic pain

(CNP) is under debate. We updated a recent Cochrane systematic review on the

efficacy, tolerability and safety of opioids in CNP.

Methods: We screened Medline, Scopus, and the Cochrane Library (through October

2013), as well as reference sections of original studies and systematic reviews of

randomized controlled trials (RCTs) of opioids in CNP. We included double-blind

randomized placebo-controlled studies of at least four weeks duration. Risk

differences (RD) of categorical data and standardized mean differences (SMD) of

continuous variables were calculated using a random effects model.

Results: We included twelve RCTs with 1192 participants. The included diagnostic

entities were painful diabetic neuropathy (four studies), postherpetic neuralgia (three

studies), mixed polyneuropathic pain (two studies) and lumbar root, spinal cord injury

and post-amputation pain in one study each. Mean study duration was six (4 - 12)

weeks. Four studies tested morphine, three studies tramadol, two studies oxycodone

and one study tapentadol. These are the pooled results of studies with a parallel or

cross-over design: Opioids were superior to placebo in reducing pain intensity (SMD -

0.64 [95% confidence interval [CI] -0.81, -0.46]; p < 0.0001; eleven studies with 1040

participants). Opioids were not superior to placebo in 50% pain reduction (RD 0.16

[95% CI -0.04, 0.35], p = 0.11; one study with 93 participants). Opioids were not

superior to placebo in reports of much or very much improvement (RD 0.17 [95% CI -

0.01, 0.36]; p = 0.07; one study with 53 participants). Opioids were superior to

placebo in improving physical functioning (SMD -0.28 [95% CI -0.43, -0.13]; p <

0.0001; seven studies with 680 participants). Patients dropped out less frequently

with opioids than with placebo due to lack of efficacy (RD -0.07 [95% CI -0.13, -0.02],

p=0.008, six studies with 656 participants). Patients dropped out more frequently with

opioids than with placebo due to adverse events (RD 0.08 [95% CI 0.05, 0.12], p <

0.0001; ten studies with 1018 participants; number needed to harm 11 [95% CI 8 -

17]). There was no significant difference between opioids and placebo in the

frequency of serious adverse events and of deaths.

Conclusions: In short-term studies (4 - 12 weeks) in CNP, opioids were superior to

placebo in terms of efficacy and inferior in terms of tolerability. Opioids and placebo

did not differ in terms of safety. The conclusion on the safety of opioids compared to

placebo in CNP is limited by the low number of serious adverse events and deaths.

Short-term opioid therapy may be considered in selected patients with CNP.

3

The English full-text version of this article is available at SpringerLink (under

“Supplemental”). This article is published with free access at Springerlink.com.

Key words: Systematic review; meta-analysis; opioids; chronic neuropathic pain

4

Introduction

Chronic neuropathic pain (CNP) may result from a large variety of insults to the

peripheral or central somatosensory nervous system, including trauma, inflammation,

ischemia, and metabolic and neoplastic disorders. Common examples of peripheral

neuropathic pain include diabetic and postsurgical neuropathy. Central neuropathic

pain includes post-stroke pain, pain in multiple sclerosis, and pain after spinal cord

injury. The main clinical characteristics of neuropathic pain are a burning, aching, or

shooting quality, and abnormal sensitivity of the painful site to normally innocuous

stimuli (14). Estimates of the prevalence of chronic pain with neuropathic

characteristics range from 6.9% - 10% in the general population (22). However,

specific conditions with neuropathic pain can be rare, like postherpetic neuralgia with

a prevalence of 0.09% and trigeminal neuralgia with 0,07% (22), which makes it

difficult to collect large homogenous study cohorts.

The drug therapy of CNP is under debate (1,7,10). There is a lack of definitive

evidence regarding the efficacy of opioids in reducing neuropathic pain (14). A recent

systematic Cochrane reviews on opioids in neuropathic pain searched the literature

until October 2012. The review included short-term studies (< 4 weeks) into the

analyses of efficacy and harms and excluded studies with tramadol and tapentadol

(14).

For the revision of the German 2008 guideline on the long-term administration of

opioids in chronic non-cancer pain (CNCP) (LONTS) (19) we updated the search of

literature of systematic review on opioids in CNP. The objectives of this review were

to determine the efficacy, tolerability and safety of opioids (including tapentadol and

tramadol) compared to placebo in patients of any age with CNP in randomized

placebo-controlled studies with a duration of four or more weeks.

.

5

Methods

The review was performed according to the PRISMA (Preferred

Reporting Items for Systematic Reviews and Meta-Analyses) statement (15) and the

recommendations of the Cochrane Collaboration (13).

Criteria for considering studies for this review

Types of studies

We included fully published randomised double-blind controlled trials (RCT) that

compared any opioid to placebo (pure or pseudo) for therapeutic purposes in CNP.

We included studies with a parallel design and an enriched enrolment withdrawal

(EERW) design. Studies with a cross-over design were included if a.) separated data

of the two periods were reported or b.) data were presented which excluded

statistically significant carry-over effects or c.) statistical adjustments were carried out

in case of a significant carry-over effect.

Study duration should be at least 4 weeks (maintenance phase for parallel and cross-

over design; double blind withdrawal phase for EERW design). Studies should

include at least 10 patients per treatment arm.

We grouped outcome measures according to the timing of post-randomization follow-

up: short-term (four to twelve weeks), intermediate (12 - 26 weeks) and long-term

(longer than 26 weeks).

We had no restriction on the language of the publication.

We excluded studies which conducted a tapering phase after open-label run-in and a

consecutive double-blind parallel design with responders of the open-label run-in

period. We excluded studies with a duration of maintenance or withdrawal period of

less than four weeks, with an experimental design (i.e. if the primary purpose was to

study pain mechanisms and not pain relief) and studies and studies which were only

published as abstracts. We excluded studies in which different dosages of one opioid

were compared without a control group.

Types of participants

We included men and women of all ages and races or ethnicities diagnosed with

central or peripheral neuropathic pain of any etiology of least three months duration.

Types of interventions

We considered trials with opioids given by oral and transdermal routes.

6

We included studies in which opioids were combined with abuse deterrent

formulations (ADFs) (e.g. naloxone). We also included studies with tramadol, a

centrally acting, synthetic opioid analgesic with two complementary mechanisms of

action: binding of parent and M1 metabolite to μ-opioid receptors and inhibition of

reuptake of norepinephrine and serotonin. We included studies with tapentadol, a

drug with two mechanisms of action: μ-receptor agonist and norepinephrine reuptake

inhibitor. The reason for including both drugs into this review was that they are

classified as opioids by German medicine agencies.

We considered trials with the following comparisons:

1. Opioids compared to placebo;

We excluded trials that examined opioids given by intravenous route and intrathecal

implantable pumps, due to the invasive nature of the therapy and its limited clinical

relevance in the outpatient setting. We did not assess the effectiveness of opioids

used in neuraxial implantable pumps as this has been discussed elsewhere (18). We

excluded studies in which drugs other than opioid agonists were combined with

opioids (e.g. tramadol with acetaminophen) except as a rescue medication because it

is not possible to detangle the effects of opioids from the other analgesic. We

excluded studies in which a defined opioid was compared to the same opioid with

abuse deterrent formulations (ADFs) (e.g. oxycodone with and without naloxone) or

in which two opioids combined were compared with a single opioid without a placebo

group. We excluded studies in which opioids and placebo were compared as add-on

to other drug therapies or vice versa. We excluded studies in which opioids were

compared to non-pharmacological treatments. We excluded studies with

propoxyphene because the drug has been withdrawn from the market (United States

Food and Drug Administration NEWS RELEASE vom 19.11.2010). We excluded

studies with opioid receptor agonist / NMDA antagonists (e.g. levorphanol) because

these drugs are not available in Germany.

Types of outcome measures

The selection of outcomes was based on the recommendations of the ACTINPAIN

writing group of the International Association for the Study of Pain (IASP) Special

Interest Group (SIG) on Systematic Reviews in Pain Relief and the Cochrane Pain,

Palliative and Supportive Care Systematic Review Group editors for reporting meta-

analyses of RCTs in chronic pain (16,17). We included pain intensity as additional

7

outcome because most studies conducted before 2005 did not report responder

analyses (14).

Efficacy

1. Pain intensity ratings

2. Proportion of patients reporting 50% pain relief (responders)

3. Global improvement: Number of patients reporting to be much or very much

improved

4. Function: Examples of functional impairment outcomes that could be extracted

were as follows: Brief Pain Inventory (BPI); Fibromyalgia Impact Questionnaire

Subscale Physical Function (FIQ); Multidimensional Pain Inventory (MPI, physical

function); Western Ontario and McMaster Universities Arthritis Index (WOMAC);

Neck Disability Index (NDI); Oswestry Disability Index (ODI); Pain Disability Index

(PDI), physical disability; Roland Disability Questionnaire (RDQ); Short Form (SF)-36

or SF-12 (physical functioning scale). In case both, generic and disease specific

instruments were used, disease specific instruments were preferred (e.g. FIQ over

PDI, WOMAC over SF-36 physical functioning scale)

5. Proportion of patients who withdrew due to lack of efficacy

Tolerability

1. Proportion of patients who withdrew because of adverse events

Safety

1. Proportion of patients who experienced any serious adverse events (SAE).

2. Proportion of patients who died during study

We excluded studies in which the primary outcome measure was not one of the five

outcomes of efficacy as defined above.

Search methods for identification of studies

Electronic searches

The review updated and expanded the search of literature of the first version of the

German guideline on the long-term administration of opiods (LONTS), which

searched the literature until October 2008 (19). The updated and expanded search

included CENTRAL, Medline and Scopus from October 2008 to October 2013 and all

types of CNCP. The search was conducted by PK. Our search included all

languages.

Searching other resources

8

The search strategy has been detailed in another paper of this issue (20).

Data collection and analysis

Selection of studies

Two authors (PW, WH) independently screened titles, abstracts, and keywords of

trials that we identified by the search strategies to determine if the references met the

inclusion criteria. We obtained the full text of trials that either appeared to meet

criteria or for which we considered their inclusion was uncertain. We screened these

articles for inclusion and we resolved any disagreements through discussion.

Data extraction

Three pairs of authors (CS, WH; FP, WH; RS, WH) independently extracted data,

using the standardized forms on inclusion and exclusion criteria of studies,

characteristics of participants, intervention group, clinical setting, interventions,

country of study, and sponsorship of study. If data were not available in a format that

was appropriate for data extraction, we did not contact the authors of the trial for

further clarification. We resolved any disagreements through discussion.

Dealing with missing data

If both, baseline observation carried forward (BOCF) data as well as last observation

carried forward (LOCF) data were reported for intention-to-treat (ITT) analysis, we

preferred BOCF data (17).

Where means or standard deviations (SD) were missing, we calculated them from t-

values, CIs or standard errors, where reported in articles (13). Where missing SDs

could not be calculated fromt values, the study was excluded from analysis.

Measures of treatment effect

The effect measures of choice were absolute risk difference (RD) for dichotomous

data and standardised mean difference (SMD) for continuous data (pain intensity,

physical functioning) using a random effects model (method inverse variance). For

subgroup analyses of dichotomous outcomes we calculated risk ratios (RR). We

expressed uncertainty using 95% confidence intervals (CIs). The threshold for

“appreciable benefit” or “appreciable harm” was set for categorical variables by a

relative risk reduction (RRR) or relative risk increase (RRI) >= 25% (4). We used

Cohen’s categories to evaluate the magnitude of the effect size, calculated by SMD,

with Hedge’ s g of 0.2 = small, 0.5 = medium and 0.8 = large (6). We labelled g < 0.2

to be a ’not substantial’ effect size. We assumed a minimally important difference if

there was small effect size (9).

9

The numbers needed to treat for an additional beneficial outcome (NNTB) and the

numbers needed to treat for an additional harm (NNTH) for dichotomous variables

(50% pain reduction, PGIC, drop out due to adverse events, serious adverse events,

death) by a calculator provided by the Cochrane Musculoskeletal Group.

Subgroup comparisons were performed by the test of interaction (1).

Unit of analysis issues

In the case of multiple opioid arms compared with one placebo group, we adjusted

the number of participants in the placebo group according to the number of

participants in the different opioid arms for continuous outcomes.

Data synthesis

We pooled data from trials comparing opioids to controls by a random-effects

methods method inverse variance. We used the I² statistic to describe the percentage

variability of effect estimates that is due to heterogeneity. I² values above 50%

indicate high heterogeneity, between 25% and 50% moderate heterogeneity, and

below 25% low heterogeneity (12).

The risk of bias in each trial was assessed independently by two pairs of authors

(CS, WH; FP; WH; RS, WH) using eight domains recommended by the Cochrane

Collaboration: selection bias, performance bias, detection bias, attrition bias,

reporting bias, selection bias, performance bias, detection bias and funding bias (19).

We defined a high quality study (low risk of bias) that fulfilled six to eight, a moderate

quality study (moderate risk of bias) that fulfilled three to five and a low quality study

(high risk of bias) that fulfilled zero to two of the eight validity criteria (4). We used

Grading of Recommendations Assessment, Development and Evaluation (GRADE)

approach to assess the overall quality evidence (13,), defined as the extent of

confidence into the estimates of treatment benefits and harms. Quality ratings were

made separately for each of the eight outcomes. The quality of evidence was

downgraded by one level for each of the following factors that were encountered:

Limitations of study design: > 50% of the participants of low quality studies

Inconsistency of results: I² > 50%

Indirectness: We assessed whether the question being addressed in this systematic

review was different from the available evidence regarding the population in routine

clinical care if patients with clinically relevant somatic disease and / or major mental

10

disorders (history of substance abuse or major depression) were excluded in > 50%

of participants

Imprecision: There was only one trial or when there was more than one trial, the total

number was < 400 patients or when the 95% CI of the effect size included zero

We categorized the quality of evidence as follows:

· High: further research is very unlikely to change the confidence in the estimate of

effect.

· Moderate: further research is likely to have an important impact in the confidence in

the estimate of effect.

· Low: further research is very likely to have an important impact on our confidence in

the estimate of effect and is likely to change the estimate.

· Very low: any estimate of effect is very uncertain.

Assessment of reporting biases

For analyses without substantial heterogeneity (I² < 50%) and at least 10 studies, we

used the Egger intercept test and the Begg rank correlation test at the significance

level p< 0.05. The Begg test examines the rank correlation between standardized

intervention effect and its standard error. An asymmetric funnel plot would give rise to

such a correlation and may be indicative of publication bias (3). In the Egger test, the

standard normal deviate is regressed on precision, defined as the inverse of the

standard error. The intercept in this regression corresponds to the slope in a

weighted regression of the effect size on the standard error (8).

Subgroup analysis

Subgroups were a priori planned to assess the variations in effect size

(heterogeneity) for all types of opioids pooled together compared to placebo groups

pooled together for the outcomes (pain intensity and drop out due to adverse events)

for different types of chronic neuropathic pain, different types of opioids (pure opioids

versus opioids with additional modes of action [tramadol, tapentadol], studies with

and without industry sponsoring and treatment duration (short-term, intermediate-

term and long-term studies). At least two studies should be available for subgroup

analysis.

Sensitivity analysis

11

We performed sensitivity analysis of all types of opioids pooled together compared to

placebo groups pooled together for the outcomes in studies in which we extracted

means and / or SDs from figures or calculated SD from p-values.

Software

Comprehensive meta-analysis (Biostat, Englewood, NJ, USA) and RevMan Analysis

(RevMan 5.2) software of the Cochrane Collaboration were used for statistical

analyses.

Results

Search

After removing duplicates, the literature search produced unique 12601 citations.

Through screening, 12589 records were excluded. Twelve full-text articles were

assessed for eligibility. No study was excluded after full-text review. Twelve studies

with 12 treatment arms were included in the meta-analysis (see Figure 1).

Study characteristics (see table 1 and supplementary table 1).

Study design: The 12 studies included 1192 (35 - 395) patients. Median study

duration was 6 (4 - 12) weeks. No study had a duration > 12 weeks. Eight (66.7%)

studies had a cross-over, 3 (25%) studies had a parallel and one (8.3%) study had an

enriched enrolment randomized withdrawal design. Nine (75.0%) studies were

conducted in North America and three studies (25.0%) in Europe. Six studies (50%)

were funded by the manufacturer of one of the drugs tested.

Participants: Participants were diagnosed with painful diabetic neuropathy (four

studies), postherpetic neuralgia (three studies), mixed polyneuropathic pain (two

studies) and in one study each with lumbar root, spinal cord injury and

postamputation pain. Only adults were included into the studies. The range of the

mean ages of participants in the studies was 51 - 71 years. The participants were

predominantly Caucasians, the gender ratio was nearly balanced. Nine (80%) studies

each excluded patients with current and/or a history of substance abuse and/or

current major mental disorders and all studies excluded patients with clinically

relevant medical diseases.

Interventions: Four studies tested morphine, three studies tramadol, two studies

oxycodone and one study tapentadol. All studies used a flexible dosage of the opioid.

The opioids were administered by extended release formulations in all studies. Two

12

(16.7%) studies did not report on or prohibited rescue medication. Eight (66.7%)

studies allowed rescue medication (acetaminophen, NSAIDs, short-acting opioids).

Quality of evidence

Risk of bias could not be properly assessed in all studies due to poor method

reporting. Two studies each (16.7%) had a low or high and eight studies (75.0%) had

a moderate study quality (see Figure 2, Figure 3 for risk of bias summary and graph).

Detailed information regarding risk of bias assessments of every study are given in

the Electronic Supplementary Material table 2.

Synthesis of results

Parallel and cross-over design (Results are reported with 95% CI)

Eleven studies with 1040 participants were entered into an analysis of mean pain

reduction at the end of the study. Opioids were superior to placebo (SMD -0.64 [-

0.81, -0.46]; p < 0.0001; I² = 45) (moderate quality evidence). According to Cohen’s

categories the effects size was moderate (see Electronic Supplementary Material

figure 1).

One study with 93 participants was entered into an analysis of 50% pain reduction at

the end of the study. Opioid was not superior to placebo (RD 0.16 [-0.04, 0.35], p =

0.11, I² = 51) (low quality evidence) (see Electronic Supplementary Material figure 2).

One study with 35 participants was entered into an analysis of reports to be much or

very much improved at the end of the study. Opioids were not superior to placebo

(RD 0.17 [-0.01, 0.36]; p = 0.07) (very low quality evidence) (see Electronic

Supplementary Material figure 3).

Seven studies with 680 participants were entered into an analysis of physical function

at the end of the study. Opioids were superior to placebo (SMD -0.28 [-0.43, -0.13]; p

= 0.0003]; I² = 0) (moderate quality evidence) (see Electronic Supplementary Material

figure 4). According to Cohen’s categories, the effect size was small. Two studies

(appendix references 1,12), which reported no significant differences between opioid

and placebo, could not be entered in meta-analysis because no means, SDs, p- or t-

values were reported.

Patients dropped out less frequently with opioids than with placebo due to lack of

efficacy (RD -0.07 [ -0.13, -0.02], p=0.008, six studies with 656 participants)

(moderate quality evidence) (see Electronic supplementary figure 5). 11/330 (3.3%)

13

of patients dropped out in opioid and 44/326 (13.5%) dropped out in placebo group

(NNTB 10 [95% CI 7-17]). According to the predefined criteria, there was an

appreciable additional benefit by opioids (RRR 75% [53% - 87%]).

Ten studies with 1018 participants were entered into an analysis of dropping out due

to adverse events. Patients dropped out more frequently with opioids than with

placebo (RD 0.08 [0.05, 0.12], p < 0.0001, I² = 0) (moderate quality evidence) (see

Electronic Supplementary Material figure 6). 70/524 (13.3%) of patients dropped out

in the opioid groups and 20/494 (4.0%) in the placebo groups due to adverse events

(NNH 11 [95% CI 8 - 17]). According to the predefined criteria, there was no

appreciable additional harm by opioids (RRI 230% [104% - 434%]).

Three studies with 357 participants were entered into an analysis of serious adverse

events. There was no significant difference between opioids and placebo (RD -0.01 [-

0.09, 0.07] p = 0.78; I² = 63) (very low quality evidence) (see Electronic

Supplementary Material figure 7).

One study with 159 participants was entered into an analysis of deaths. 1/82 patients

died in opioid and 0/77 patients died in placebo groups (p = 0.48) (very low quality

evidence).

EERW design

One study with 200 participants was entered into an analysis of 50% pain reduction

maintenance from randomization to the end of study. Opioid was superior to placebo

(RD 0.23 [0.09, 0.36] p = 0.01 (very low quality evidence). 55/93 (59.1%) of patients

in the opioid group and 39/107 (36.4%) of patients in the placebo group reported a

50% pain reduction maintenance (NNTB 4 [95% CI 3 - 11) (see Electronic

Supplementary Material figure 8). According to the predefined criteria, there was an

appreciable additional benefit by opioids (RRI 62% [20% - 119%]).

One study with 357 participants was entered into an analysis of reports to be much or

very much improved at the end of the study. Opioid was superior to placebo (RD 0.26

[0.16, 0.36]; p < 0.0001) (very low quality evidence) (see Electronic Supplementary

Material figure 9). 116/180 (64.4%) of patients in the opioid group and 68/177

(38.4%) of patients in the placebo group reported to be much or very much improved

(NNTB 4 [95% CI 3 - 6). According to the predefined criteria, there was an

appreciable additional benefit by opioids (RRI 68% [35% - 108%].

One study with 389 participants was entered into an analysis of dropping out due to

adverse events. Patients dropped out more frequently with opioids than with placebo

14

(RD 0.07 [0.01, 0.13], p = 0.03) (very low quality evidence) (see Electronic

Supplementary Material figure10). 29/196 (14.8%) of patients dropped out in opioid

group and 15/193 (7.8%) in placebo group (NNTH 14 [95% CI 8 - 128]). According to

the predefined criteria, there was an appreciable additional harm by opioids (RRI

90% [5% - 244%]).

One study with 389 participants was entered into an analysis of serious adverse

events. There were more SAE in the opioid than in the placebo groups (RD 0.04

[0.00, 0.07]; p = 0.05) (very low quality evidence) (see Electronic Supplementary

Material figure 11). 10/196 (5.1%) of patients in opiod group and 3/193 (1.6%) of

patients in placebo group reported serious adverse events. (NNTH 28 [95% CI 28-

14400]). According to the predefined criteria, there was no appreciable additional

harm by opioids (RRI 228% [-8% - 1000%]) because the confidence interval included

zero..

No deaths were reported in both groups.

Subgroup and sensitivity analyses

In three parallel or cross-over studies of painful diabetic neuropathy, opioids were

superior in pain reduction and in improving physical function and inferior in tolerability

if compared to placebo (see table 2).

In three parallel or cross-over studies of chronic postherpetic neuralgia, opioids were

superior in pain reduction, but not in improving physical function compared to

placebo. Opioids were inferior in tolerability compared to placebo (see table 2).

Studies with and without industry sponsoring did not differ in their effect sizes on pain

(z=0.74, p=0.46).. The drop out rate due to adverse events was higher in studies

without industry sponsoring than without industry sponsoring (z=3.41, p=0.0006).(see

table 3).

Publication bias

The Kendall tau of the Begg rank correlation test of the outcome dropping out due to

adverse events was not significant (tau = -0.27, P two-tailed = 0.24). The Egger

intercept of the outcome pain was not significant (intercept = -2.44, p two-tailed =

0.27).

15

Discussion

Summary of main results

In short-term studies in patients with CNP (4 - 12 weeks), opioids were superior to

placebo in terms of efficacy and inferior in terms of tolerability. Opioids and placebo

did not differ in terms of safety. Sufficient data (> one RCT, > 400 patients included)

were only available for painful diabetic neuropathy and postherpetic neuralgia pain.

No studies with opioids are available for other types of chronic polyneuropathic pain,

e.g. in HIV, alcoholic or by chemotherapy-induced PNP. The evidence on opioids on

central neuropathic pain syndromes assessed in this review (spinal cord injury and

postamputation pain) is based on one study each with small sample size. There are

no RCTs with opioids available on other types of central CNP such as post stroke

pain. In short, the efficacy of opioids in all types of CNP has not been demonstrated

until now.

Comparison with other systematic reviews

Our results on the efficacy in terms of pain reduction and tolerability of opioids in

CNP are mainly in line with those of a Cochrane review which included fourteen

studies (845 participants, average 60 participants per study with duration lasting 1 to

12 weeks; most studies lasted less than six weeks). The authors did not include

studies with tramadol and tapentadol (14). Seven RCTs were included in the

Cochrane and in our review. Meta-analysis of the Cochrane group demonstrated at

least 33% pain relief in 57% of participants receiving an opioid versus 34% of those

receiving placebo. The overall risk difference was 0.25 (95% CI 0.13 to 0.37, P <

0.0001), translating to a number needed to treat for an additional beneficial outcome

(NNTB) of 4.0 (95% CI 2.7, 7.7). When the number of participants achieving at least

50% pain relief was analyzed, the overall point estimate of risk difference between

opioids (47%) and placebo (30%) was 0.17 (95% CI 0.02, 0.33, P = 0.03), translating

to an NNTB of 5.9 (3.0, 50.0). The authors did not find improvement in physical

functioning. More participants withdrew from opioid treatment due to adverse events

(13%) than from placebo (4%) (NNTH 12.5; 95% CI 8.3, 25.0). The authors

concluded that the studies demonstrated significant efficacy of opioids over placebo

in pain reduction in CNP (14).

16

In contrast to the Cochrane review, we found a moderate effect size of opioids on

physical function. However, two studies which did not find a significant difference

between opioids and placebo could not be included into meta-analysis. In conclusion,

despite of efficacy in terms of pain reduction, the overall efficay to improve clinical

status in neuropathic pain is subject to considerable uncertainty (14).

In addition, RCTs of up to six weeks duration do not necessarily provide information

about long-term use, which is important in treatment of a chronic condition. In

particular, concern has been raised about the lack of evidence on potential problems

with long-term use of opioids in the treatment of neuropathic pain (such as safety

issues, addiction and misuse) (21,23).

The presently available systematic reviews of placebo-controlled studies on CNP are

not very helpful in guiding patients and physicians in the choice of first- and second-

line treatment. Head-to-head comparisons of opioids with other drugs have been

rarely conducted. Morphine was compared with antidepressants (two studies), an

anticonvulsant (one study) and an antiarrhythmic (one study) in different CNP

syndromes. There was no difference between the opioids, antidepressants and

anticonvulsants tested in terms of efficacy and tolerability (24. Therefore, evidence-

based recommendations of first- or second-line drug therapies in CNP are not

possible. To provide a superior evidence base for future treatment guidelines,

additional studies must be conducted in which existing drugs are directly compared

with each other and administered in various combinations. Traditional RCTs may

ultimately not be the method of choice to answer all these questions; alternative

approaches should be developed and evaluated, e.g. systematic comparative

effectiveness studies of health care registry data (7).

Limitations

Only double-blind randomized placebo-controlled studies were included in this meta-

analysis, representing a high level in evidence based medicine. However, the

methodological quality of the majority of included studies was only moderate. There

was a high risk of selective reporting causing a relevant reporting bias.

We may have missed published studies by our search strategy. Since clinical trial

registries were searched, this statement is mainly valid for studies conducted before

the establishment of these registries. In addition, we may have underestimated the

quality of studies because we did not ask the authors for missing details.

17

The conclusion on the safety of opioids compared to placebo is limited by the low

number of serious adverse events and deaths.

Conclusions for clinical practice

Opioids may be considered in the short-term treatment (4 - 12 weeks) of CNP

syndromes. There are no RCTs with opioids available for some neuropathic

conditions like, poststroke pain, multiple sclerosis-related and non-diabetic PNP pain

(e.g. HIV, alcoholic). Diverse pharmacological treatments of CNP have become

available, and interpreting the data on their efficacy and safety involves substantial

complexity and ambiguities (7). Recent guidelines (update of the NeuPSIG

pharmacological guidelines for management of neuropathic pain; European

Federation of Neurological Societies Task Force) did not recommend first- and

second-line drug treatment options, but a multifactorial evaluation that carefully

considers the clinical importance of the improvements shown by patients and the

benefits and risks of each treatment in view of the other available treatments (2,7).

The German National Patient-Centered Guideline on neuropathy in type 2 diabetes

recommended morphine, oxycodone and tramadol if opioid therapy is considered in

painful diabetic neuropathy (4).

Long-term open-label studies demonstrated that a minority of patients with chronic

non-cancer pain (including CNP) treated with opioids experienced a sustained (> 1

year) response with no or tolerable side effects (12,24). These data support the

notion that long-term ( 26 months) opioid therapy may be offered to sustained

responders to short-term opioid therapy, especially when alternative treatment

options fail, show only partial efficacy, or have significant side effects. However, the

potential benefits must be carefully balanced against potential risks of long-term

opioid therapy (12).

Acknowledgements: We thank Professor Sorgatz (Essen) for reviewing our data on

drop outs due to lack of efficacy

18

Table 1: Overview of the randomized controlled trials in chronic neuropathic pain

included into the systematic review (grouped by type of opioid in alphabetical order)

Morphine

Reference Year Countries of study centers

Study design Population type Number of patients randomized

Interventions and control group

Duration of trial (titration and maintenance)

Gilron 2005 Canada

Cross over Postherpetic neuralgia and painful diabetic neuropathy

57

Morphine up to 120 mg/d flexible oral Active placebo (lorazepam) flexible

No reports on wash-out periods 4 weeks each

Khoromi 2007 USA

Cross over Lumbar root pain 56

Morphine flexible 15 - 90 mg/d oral Nortriptyline flexible 25 - 100 mg/d oral Benztropine flexible (0.25 -1 mg/d) (active placebo)

Duration screening and wash-out not reported 3 weeks titration 2 weeks maintenance 2 weeks tapering

Raja 2002 USA

Cross over Postherpetic neuralgia 50

Morphine flexible 15 - 240 mg/d oral or methadone 5 - 75 mg/d or placebo oral flexible

Duration wash-out: 1 week

Approximately 8 weeks titration, maintenance and dose tapering for each period

1 week drug-free wash-out

Wu 2008 USA

Cross over Postamputation (stump and phantom pain) 60

Morphine 15 - 180 mg/d flexible oral vs. flexible oral placebo

4 weeks titration

2 weeks,

19

maintenance

2 weeks dose tapering for each period

1 week drug-free wash-out

Oxycodone


Study design

Population type Number of patients randomized


Duration of trial (titration and maintenance)

Gimbel 2003 USA

Parallel Painful diabetic neuropathy 159

Oxycodone oral flexible up to 20 mg/d oral Placebo

Screening 7 days 6 weeks titration and maintenance 1-week tapering optional

Watson 1998 USA

Cross over Postherpetic neuralgia 38

Oxycodone oral flexible 20 - 160 mg/d Placebo

Duration wash out and screening not reported 4 weeks each No wash out between periods

Watson 2003 USA

Cross over Painful diabetic polyneuropathy 45

Oxycodone oral flexible 20 - 80 mg/d Active placebo flexible 0,5 – 2mg mg/d

2 - 7 days wash out 4 weeks each No wash out between periods Up to one year open label

20

Tapentadol


Study design Population type Number of patients randomized


Duration of trial (weeks)

Schwartz Canada, USA 2011

Enriched-enrollment randomized withdrawal

Painful diabetic polyneuropathy 395

Tapentadol 200 - 500 mg/d flexible oral Placebo

Duration screening not reported 3 weeks open-label titration phase 12 week double blind withdrawal

Tramadol


Study design

Population type Number of patients randomized


Duration of trial (weeks)

Boureau France 2003

Parallel Postherpetic neuralgia pain 125

Tramadol flexible 100 - 400 mg/d oral Placebo

Duration of screening not reported 6 weeks titration and maintenance

Harati 1998 USA

Parallel Painful diabetic polyneuropathy 131


Duration of wash-out not reported

6 weeks titration and maintenance

6 months open label

Norrbrink 2009 Sweden

Parellel Neuropathic pain after spinal cord injury 35


Duration of wash-out not reported

6 weeks titration and maintenance

Sindrup 1999 Denmark

Cross-over Polyneuropathy of different etiologies

Tramadol flexible 100 - 400 mg/d oral

Up to 1 week wash-out

21

45

Placebo 4 weeks each period

1 week wash-out

22

Table 2: Effect sizes of opioids on selected outcome variables (parallel and cross-over

design) in different types of chronic neuropathic pain

Outcome

title

Number

of

studies

Number

of

patients

Effect size

[95% CI])

Test for

overall

effect

p-value

Heteroge

neity

I² [%]

Painful diabetic neuropathy

01 Pain 3 380 SMD -0.74 (-1.06, -0.43)

<

0.0001

55

02 Drop out

due to

adverse

events

3 380 RD 0.07 [0.01, 0.13]

0.01 10

Postherpetic neuralgia

01 Pain 3 323 SMD -0.58 [-0.85, -0.31]

<

0.0001

29

02 Function 1 122 SMD -0.13 [-0.49, 0.22]

0.47

03 Drop out

due to

adverse

events

3 323 RD 0.09 [0.03, 0.14]

0.001 0

Mixed polyneuropathy and postherpetic neuralgia

01 Pain 2 177 SMD -0.54 [-1.03, -0.06]

0.03 62

02 Drop out

due to

adverse

events

2 170 RD 0.10 [0.02, 0.18]

0.002 0

23

Abbreviations: CI: Confidence interval; PNP= Polyneuropathy; RD = Risk

difference; SMD: Standardized mean difference;

24

Table 3: Effect sizes of opioids on selected outcome variables (parallel and cross-over

design) in studies with and without industry sponsoring

Outcome

title

Number

of

studies

Number

of

patients

Effect size

[95% CI])

Test for

overall

effect

p-value

Heteroge

neity

I² [%]

Without industry sponsoring

01 Pain 5 391 SMD -0.59 (-0.92, -0.26)

0.0004 58

02 Drop out

due to

adverse

events

5 354 RR 3.41 (1.64, 7.06)

0.001 0

With industry sponsoring

01 Pain 6 649 SMD -0.67(-0.88, -0.46]

<

0.0001

40

02 Drop out

due to

adverse

events

6 664 RR 2.23 (1.16, 4.27)

0.02 0

Abbreviations: CI: Confidence interval; RR = Relative Risk; SMD: Standardized

mean difference;

25

Figure 1: PRISMA Flow Diagram

Records identified through

database searching (n = 17 591)

CENTRAL: (n=3688) Medline: (n=6944) Scopus: (n=6959)

Scre

enin

g

Incl

ud

ed

Elig

ibili

ty

Iden

tifi

cati

on

Additional records identified through hand searching

(n = 52)

Records after duplicates removed (n = 12 601 )

Records screened (n = 12601 )

Records excluded (n =12 589)

Full-text articles assessed for eligibility

(n = 12 )

Full-text articles excluded, with reasons (n = 0)

Studies included in qualitative synthesis

(n = 12)

Studies included in quantitative synthesis

(meta-analysis) (n =12)

26

Figure 2: Risk of bias graph

Random sequence generation (selection bias)

Allocation concealment (selection bias)

Blinding of participants and personnel (performance bias)

Blinding of outcome assessment (detection bias)

Incomplete outcome data (attrition bias)

Selective reporting (reporting bias)

Selection bias

Funding bias

0% 25% 50% 75% 100%

Low risk of bias Unclear risk of bias High risk of bias

27

Figure 3: Risk of bias summary

Ran

do

m s

equ

en

ce

gen

era

tion

(sele

ction

bia

s)

Boureau 2003 +

Gilron 2005 ?

Gimbel 2003 +

Harati 1998 +

Khoromi 2007 +

Norrbrink 2007 ?

Raja 2002 +

Schwartz 2011 ?

Sindrup 1999 +

Watson 1998 +

Watson 2003 +

Wu 2008 +

Allo

cation c

oncealm

ent (s

ele

ction b

ias)

?

+

+

+

+

+

+

+

+

?

+

+

Blin

din

g o

f part

icip

ants

and p

ers

onnel (p

erf

orm

ance b

ias)

+

?

+

+

+

+

+

?

?

?

?

+

Blin

din

g o

f outc

om

e a

ssessm

ent (d

ete

ction b

ias)

?

?

?

?

+

?

?

?

?

?

?

?

Incom

ple

te o

utc

om

e d

ata

(a

ttrition

bia

s)

?

–

?

+

?

?

?

?

–

–

?

–

Sele

ctive r

eport

ing (

report

ing b

ias)

–

–

?

+

–

–

–

–

–

–

?

–

Sele

ction b

ias

+

+

+

+

+

–

+

+

+

+

+

+

Fu

nd

ing b

ias

–

+

–

–

+

+

+

–

–

–

–

+

28

References

1. Altman DG, Bland JM. Interaction revisited: the difference between two

estimates. BMJ 2003;326(7382):219.

2. Attal N, Cruccu G, Baron R, Haanpää M, Hansson P, Jensen TS, Nurmikko T;

European Federation of Neurological Societies. EFNS guidelines on the

pharmacological treatment of neuropathic pain: 2010 revision. Eur J Neurol

2010;17:1113-e88.

3. Begg CB, Mazumdar M. Operating characteristics of a rank correlation test for

publication bias. Biometrics 1994;50:1088–1101.

4. Bundesärztekammer (BÄK), Kassenärztliche Bundesvereinigung (KBV),

Arbeitsgemeinschaft der Wissenschaftlichen Medizinischen Fachgesellschaften

(AWMF). Nationale VersorgungsLeitlinie Neuropathie bei Diabetes im

Erwachsenenalter, 2011. http://www.diabetes.versorgungsleitlinien.de. Accessed

August 15, 2013.

5. Chaparro LE, Furlan AD, Deshpande A, Mailis-Gagnon A, Atlas S, Turk DC.

Opioids compared to placebo or other treatments for chronic low-back pain.

Cochrane Database Syst Rev 2013;8:CD004959.

6. Cohen J. Statistical Power Analysis for the Behavoral Sciences. Hillsdale, NJ:

Lawrence Erlbaum Associates; 1988.

7. Dworkin RH, O'Connor AB, Audette J, Baron R, Gourlay GK, Haanpää ML, Kent

JL, Krane EJ, Lebel AA, Levy RM, Mackey SC, Mayer J, Miaskowski C, Raja SN,

Rice AS, Schmader KE, Stacey B, Stanos S, Treede RD, Turk DC, Walco GA, Wells

CD. Recommendations for the pharmacological management of neuropathic pain: an

overview and literature update. Mayo Clin Proc 2010;85:S3-14.

8. Egger M, Smith GD, Schneider M. Bias in meta-analysis detected by a simple,

graphical test. BMJ 1997;315:629–634.

9. Fayers PM, Hays RD. Don't middle your MIDs: regression to the mean shrinks

estimates of minimally important differences. Qual Life Res 2014;23(1):1-4.

10. Freynhagen R, Geisslinger G, Schug SA. Opioids for chronic non-cancer pain.

BMJ 2013; 346:f2937

11. Guyatt GH, Oxman AD, Kunz R, et al; GRADE Working Group. Going from

evidence to recommendations. BMJ 2008;336(7652):1049–1051.

12. Häuser W, Bernardy K, Maier C. Long-term opioid therapy in chronic non-cancer

pain: A systematic review and meta-analysis of efficacy and harms in open-label

29

extension trials with a study duration of at least 26 weeks duration. Schmerz 2014,

submitted.

13. Higgins JPT, Green S. Cochrane Handbook for systematic reviews of

intervention. Version 5.1.0. http://handbook.cochrane.org/

14. McNicol ED, Midbari A, Eisenberg E. Opioids for neuropathic pain. Cochrane

Database Syst Rev 2013;8:CD006146.

15. Moher D, Liberati A, Tetzlaff J, Altman G and the PRISMA Group. Preferred

reporting items for systematic reviews and meta-analyses: The PRISMA Statement.

Ann Intern Med 2009;51:1-7.

16. Moore AR, Eccleston C, Derry S, Wiffen P, Bell RF, Straube S, McQuay H;

ACTINPAIN Writing Group of the IASP Special Interest Group on Systematic

Reviews in Pain Relief; Cochrane Pain, Palliative and Supportive Care Systematic

Review Group Editors. "Evidence" in chronic pain-establishing best practice in the

reporting of systematic reviews. Pain 2010;150:386-9.

17. Moore RA, Derry S, Wiffen PJ. Challenges in design and interpretation of chronic

pain trials. Br J Anaesth 2013;111:38-45.

18. Noble M, Treadwell JR, Tregear SJ, Coates VH, Wiffen PJ, Akafomo C,

Schoelles KM. Long-term opioid management for chronic noncancer pain. Cochrane

Database Syst Rev 2010;(1):CD006605.

19. Reinecke H, Sorgatz H; German Society for the Study of Pain (DGSS). [S3

guideline LONTS. Long-term administration of opioids for non-tumor pain]. Schmerz

2009;23:440-7.

20. Schaefert R, Welsch P, Klose P, Petzke F, Sommer C, Häuser W. Opioids in

chronic osteoarthritis pain - A systematic review and meta-analysis of efficacy and

harms in randomized placebo-controlled studies of at least four weeks duration.

Schmerz 2014; submitted.

21. Stannard C. Opioids in the UK: what's the problem? BMJ 2013;347:f5108.

18. Turk DC, Wilson HD, Cahana A. Treatment of chronic non-cancer pain. Lancet

2011;377:2226–35.

22. van Hecke O, Austin SK, Khan RA, Smith BH, Torrance N. Neuropathic pain in

the general population: A systematic review of epidemiological studies. Pain

2014;155: 654-62.

http://handbook.cochrane.org/

30

23. Watson CP, Watt-Watson J, Chipman M. The long-term safety and efficacy of

opioids: a survey of 84 selected patients with intractable chronic noncancer pain.

Pain Res Manag 2010;15:213-7.

24. Welsch P, Sommer C, Schiltenwolf M, Häuser W. Opioids in chronic non-cancer

pain: Are opioids superior to non-opioid analgesics? A systematic review and meta-

analysis of efficacy and harms of randomized head-to head comparisons of opioids

versus non-opioid analgesics in studies of at least four weeks duration. Schmerz

2014, in press.

31

Appendix references

1. Boureau F, Legallicier P, Kabir-Ahmadi M. Tramadol in post-herpetic neuralgia: a

randomized, double- blind, placebo-controlled trial. Pain 2003;104:323-331.

2. Gimbel JS, Richards P, Portenoy RK. Controlled-release oxycodone for pain in

diabetic neuropathy - a randomized controlled trial. Neurology 2003;60:927-934.

3. Gilron I, Bailey JM, Tu D, Holden RR, Weaver DF, Houlden RL. Morphine,

gabapentin, or their combination for neuropathic pain. NEJM 2005;352:1324–34.

4. Harati Y, Gooch C, Swenson M, Edelman S, Greene D, Raskin P, Donofrio P,

Cornblath D, Sachdeo R, Siu CO, Kamin M. Double-blind randomized trial of

tramadol for the treatment of the pain of diabetic neuropathy. Neurology 1998;50:

1842-1846.

5. Khoromi S, Cui L, Nackers L, Max MB. Morphine, nortriptyline and their

combination vs. placebo in patients with chronic lumbar root pain. Pain 2007;130:66–

75.

6. Norrbrink C, Lundeberg T. Tramadol in neuropathic pain after spinal cord injury: A

randomized, double-blind, placebo- controlled trial. Clin J Pain 2009;25:177-184.

7. Raja SN, Haythornthwaite JA, Pappagallo M, Clark MR, Travison TG, Sabeen S,

Royall RM, Max MB. Opioids versus antidepressants in postherpetic neuralgia - a

randomized, placebo-controlled trial. Neurology 2002;59:1015-1021.

8. Schwartz S, Etropolski M, Shapiro DY, Okamoto A, Lange R, Haeussler J,

Rauschkolb C. Safety and efficacy of tapentadol ER in patients with painful diabetic

peripheral neuropathy: results of a randomized-withdrawal,placebo-controlled trial.

Curr Med Res Opin 2011;27:151-62.

9. Sindrup SH, Andersen G, Madsen C, Smith T, Brosen K, Jensen TS. Tramadol

relieves pain and allodynia in polyneuropathy: A randomised, double-blind, controlled

trial. Pain 1999;83:85-90.

10. Watson CP, Babul N. Efficacy of oxycodone in neuropathic pain: a randomized

trial in postherpetic neuralgia. Neurology 1998;50:1837-41.

11. Watson CP, Moulin D, Watt-Watson J, Gordon A, Eisenhoffer J. Controlled-

release oxycodone relieves neuropathic pain: a randomized controlled trial in painful

diabetic neuropathy. Pain 2003;105:71-8.

12. Wu CL, Agarwal S, Tella PK, Klick B, Clark MR, Haythornthwaite JA. Morphine

versus mexiletine for treatment of postamputation pain: a randomized, placebo

controlled crossover trial. Anesthesiology 2008;109:289-96.

32

Electronic supplementary material tables Table 1: Characteristics of included studies Boureau 2013

Methods Disease: Postherpetic neuralgia

Study setting: 77 general practitioners in France

Study design: Parallel

Study duration: Duration of screening not reported, 1 week titration and 5 weeks maintenance

Participants Inclusion criteria: Aged 18 – 85 years, had to suffer from PHN for at least 3 months and for a maximum of 1 year, and to present with spontaneous pain rated at least 40 mm on a standard 100 mm Visual Analogue Scale (VAS)

Exclusion criteria: Symptoms or past history of depression, immune-depression, seizures, illicit drug abuse or central nervous system depressant drug abuse, cerebral tumour or recent cranial traumatism, severe hepatic, renal, cardiac or respiratory pathology, hypersensitivity to tramadol or to opioids. Pregnant or breast-feeding women, and women who risked becoming pregnant during the study period could not be included. Other exclusion criteria concerned patients treated with monoamine oxidase inhibitors within 15 days prior to the inclusion visit and patients treated with antidepressants, anticonvulsants, opioid analgesics or local/general anaesthetics within 7 days prior to the inclusion visit. Patients could not be included if they were likely to receive any treatment known to interfere with the studied drug or to be able to interfere with the study design (neurological surgery, anaesthetic blocks, local treatments of pain, antidepressants, anticonvulsants, anti-vitamin K, enzymatic inductors, psychoactive agents with the exception of clonazepam, if prescribed for insomnia for more than 15 days and at a constant dose, central and peripheral analgesics

Tramadol: N=53 (per protocol); mean age 65.7 years; 62.3% female; race not reported. Pain baseline 60.8 (±12.1)

33

Placebo: N=55 (per protocol); mean age 67.9 years; 80 % female; race not reported. Pain baseline 60.0 (±13.6)

Interventions Study medication: Upward titration up to tramadol 400mg/d (mean dosage 275 (±90) mg/d

Rescue medication: Acetaminophen if given at the maximal daily dose of 3 g

Allowed co-therapies: Concomitant treatments other than those defined in the exclusion criteria were authorized. During the study all patients took at least one concomitant medication differing from acetaminophen.

Outcomes Pain: Average daily pain intensity during the past 24 h VAS 0-100

Responder: No 50% pain reduction rates reported

PGIC: Not assessed

Function: Nottingham scale *

Withdrawal due to lack of efficacy: Not reported

Withdrawal due to adverse events: Reported

Serious adverse events: Reported **

Death: Not explicitly stated

Notes * Mean (SD) scores on the Nottingham scale decreased in both groups from day 1 (10.6 (7.2) in the tramadol group and 12.4 (7.0) in the placebo group) to day 43 (5.7 (6.0) in the tramadol group and 6.7 (7.0) in the placebo group) with no significant difference between groups

** Serious adverse events occurred in 3 patients; no information provided in which group the SAE occurred

34

Gilron 2005

Methods Disease: Postherpetic neuralgia and painful diabetic neuropathy

Study setting: 1 site in Canada

Study design: Cross over

Study duration: 4 weeks in each period

Participants Inclusion criteria: Patients with diabetic neuropathy had distal, symmetric, sensory diabetic polyneuropathy as determined on the basis of their medical history and either an unequivocal decrease in response to pinprick, temperature, or vibration in both feet or bilaterally decreased or absent ankle-jerk reflexes. Patients with postherpetic neuralgia had had an eruption of herpes zoster rash not more recently than six months before enrollment. General criteria for inclusion were daily moderate pain for three months or more, an age of 18 to 89 years, a serum alanine aminotransferase or aspartate aminotransferase level less than 1.2 times the normal level, a creatinine level less than 1.5 times the upper limit of the normal range, and sufficient language skills to communicate with research staff

Exclusion criteria: Hypersensitivity to study medications, another painful condition as severe as the diabetic neuropathy or postherpetic neuralgia, a recent myocardial infarction, unstable angina or congestive heart failure, any central neurologic disorder (including seizures), a serious mood disorder, a history of serious drug or alcohol abuse, pregnancy, lactation, and lack of a primary care physician

Total sample: N=57; median 60 years; 49 % female; 97% white (demographics of N=35 patients with diabetic neuropathy). Pain baseline 5.7 (±1.7)

Interventions Study medication: Morphine up to 120 mg/d flexible (mean 45 mg/d) vs. placebo

Rescue medication: No information provided

Allowed co-therapies: NSAIDS,

35

acetaminophen, antidepressants, anticonvulsants

Outcomes Pain: Average pain intensity last 24 hours NRS 0-10


PGIC: Not assessed

Function: SF-36 physical functioning



Serious adverse events: Not reported


Notes "The primary analysis showed no significant main effects of either sequence or treatment period, but the effects of drug treatment (P<0.001) and carryover (P=0.04) were statistically significant. An exploratory analysis of all differences in pairwise carryover effects showed a difference only between morphine and placebo (P=0.005); the effect of morphine is more likely to carry over to the next treatment period than that of placebo. In the linear mixed model, all treatment contrasts were adjusted for all observed carry-over effects

36

Gimbel 2003

Methods Disease: Painful diabetic neuropathy

Study setting: 18 sites in US


Study duration: Up to 6 weeks titration, 1-week taper optional

Participants Inclusion criteria: History of pain in both feet defined as an average pain intensity ≥ 5 on an 11-point NRS the three days prior to randomisation, absence of any opioid therapy for 3 days prior randomisation

Exclusion criteria: Unstable or poorly controlled diabetes, chronic pain unrelated to diabetic polyneuropathy, history of substance or alcohol abuse, elevated serum creatinine and liver enzymes, history of cancer, hypersensitivity to oxycodone, rapidly increasing pain or neurology deficits within the previous 3 months, autonomic neuropathy or gastrointestinal dysfunction, a need for selective surgery in the study period, treatment with longacting opioids or >3 doses short acting opioids prior two weeks

Oxycodone: N=82; mean age 58.8 years; 51.6% female; 80.5% white. Pain baseline 4.9 (±2.1)

Placebo: N=77; mean age 59.0 years; 45.1% female; 87.8 % white. Pain baseline 5.9 (±2.0)

Interventions Study medication: Upward titration up to 120 mg oxycodone/d (overall daily dosage 37 ± 11 mg/d)

Rescue medication: No opioid rescue was allowed

Allowed co-therapies: Stable dosages of acetaminophen and NSAIDs allowed as well as analgesic co-medications

Outcomes Pain: Average daily pain intensity during the past 24 h NRS 0-10


37

PGIC: Not assessed

Function: Brief Pain Inventory physical functioning score

Withdrawal due to lack of efficacy: Reported


Serious adverse events: Reported

Death: Reported

Notes

38

Harati 1998


Study setting: 6 sites US


Study duration: Wash out phase of 7 - 21 days before randomization, 6 weeks tapering and maintenance

Participants Inclusion criteria: Patients with diabetes mellitus and acceptable glycemic control (total glycosylated hemoglobin of <14% in an assay where the upper limit for nondiabetic individuals is <6.1%). All patients had distal symmetric diabetic neuropathy documented by clinical, electrophysiologic, or quantitative sensory testing and were required to have pain on a daily basis in the lower extremities due to diabetic neuropathy for the previous 3 months. Before entering the double-blind treatment phase of the study, patients were required to have moderate pain, as assessed using a Likert pain rating scale, in the absence of any analgesic therapy. All patients were 18 years of age or older and required to sign an informed consent before participation in the study. At least moderate pain on a daily basis in the lower extremities due to diabetic polyneuropathy for at least 3 previous months

Exclusion criteria: Patients were excluded from participation for any of the following reasons: (1) known contraindication to or prior use of tramadol; (2) peripheral neuropathy attributable to other causes such as alcoholism, connective tissue disease, or toxic exposure; (3) pain more severe than the neuropathic pain; (4) severe depression; (5) estimated creatinine clearance <30 mL/ min; (6) clinically significant medical conditions, profound autonomic dysfunction, or brittle diabetes; (7) history of narcotic or alcohol abuse; (8) use of multiple daily doses of narcotic analgesics or mexiletine on a regular basis; (9) evidence of amputations (including toes), open ulcers, or Charcot joint

Tramadol: N=65 mean age 59 years; 40% female; race not reported; pain intensity baseline

39

2.5 (±0.1)

Placebo: N=66; mean age 59 years; 41 % female; race not reported; pain intensity baseline 2.6 (± 0.1)

Interventions Study medication: Tramadol flexible 100-400 mg/d (average dosage 210 ± 113 mg/day); Placebo

Rescue medication: No rescue medication allowed

Allowed co-therapies: No information provided. Anticonvulsants and antidepressants were washed out.

Outcomes Pain: Average pain intensity NRS 0-4 at clinical visit


PGIC: Much or very much improved


Function: Subscale physcial functioning of the SF-36




Notes

40

Khoromi 2007

Methods Disease: Chronic lumbar root pain

Study setting: 1 site in US


Study duration: 4 weeks titration, 2 weeks maintenance, 2 weeks dose tapering for each period

Participants Inclusion criteria: 1.) Evidence of lumbar radiculopathy, including pain in one or both buttocks or legs for 3 months or greater for at least 5 days a week and at least one of the following features on the side corresponding to leg pain: a. Sharp and shooting pain below the knee; b. Pain evoked by straight leg raising to 60 degrees or less; c. Decreased or absent ankle reflex; d. Weakness of muscles below the knee. e. Sensory loss in L5/S1 distribution; f. Electromyographic evidence for L4, L5, or S1 root denervation; g. Imaging (MRI, CT/myelogram) evidence of nerve root compression in the lower lumbar region; 2.) Average leg pain of at least 4/10 for the past month on a numerical scale of 0 to 10 where 0 represents no pain and 10 represents the worst possible pain; 3.) Willingness to refrain from making changes in non-study medications taken for sciatica, and 4.) Age between 18 and 65 years at the start of the study

Exclusion criteria: 1.) Serious medical illness involving other organ systems such as unstable angina disease, advanced diabetes, and cancer; 2.) Prostatic disease requiring usage of urological medications; 3.) Pregnancy or lactation; 4.) History of depression requiring treatment with antidepressants within the 6 months preceding study participation or a score of 20 or greater on the Beck Depression Inventory at the screening visit; 5.) History of narcotic or alcohol abuse; 6.) Narrow angle glaucoma; 7.) Seizure disorder; 8.) Fibromyalgia; 9.) Pain of greater intensity in any other location than the low back or leg; 10.) Polyneuropathy and peripheral vascular disease associated with symptoms of numbness, or burning pain in the lower extremities; 11.) Allergy

41

to morphine, nortriptyline or benztropine; 12.) Evidence for multisomatoform disorder as assessed by a 15 item questionnaire, the PHQ-15; 13.) Unwillingness to be tapered off of opioids and then maintained drug free for two weeks prior to randomization to study medications for participants on maintenance opioid regimen at the time of study enrollment

Total study sample: N=56; Completers: Mean age 52.5 years; 100% Female; Race nor reported. Pain baseline 5.0 (± 2.25)

Interventions Study medication: Morphine flexible 15-90 mg/d (average 62 ± 29 mg/d); active placebo (benztopine) 0.25-1 mg/d

Rescue medication: Anti-inflammatory medications and acetaminophen

Allowed co-therapies: Patients also were asked to refrain from making changes in their analgesic medication regimen, or taking any opioids, SSRIs and tricylic medications outside of the protocol during the study

Outcomes Pain: Average overall pain intensity NRS 0-10


PGIC: Not assessed

Function: Oswestry disability index





Notes "The primary analysis for the 28 study completers showed no significant period or carry over effect between any two possible sequences of treatment."

42

Norrbrink 2007

Methods Disease: Neuropathic pain after spinal cord injury

Study setting: 4 sites Sweden


Study duration: Duration wash out not reported, 4 weeks tapering and maintenance

Participants Inclusion criteria: Traumatic or nontraumatic injury and pain diagnosed in a previous annual health control as being of at or below level neuropathic pain in origin were selected from the database. Patients between 18 and 70 years of age with no known cognitive dysfunction and currently not using tramadol; SCI for more than 12 months, pain classified as neuropathic pain at or below the level of lesion14 and of a duration of more than 6 months, fluency in Swedish, and no known cognitive impairment. Pain intensity ratings of either the general or the worst pain intensity in the last week were required to be equivalent to or more than 3 on Borg’s Category Ratio (CR-10) scale, a combined numerical and verbal rating scale. The patients had to be naive to treatment with tramadol and have had no signs of intolerance to treatment with opioids in the past. If patients were currently using opioids or antidepressants, inclusion was decided on an individual basis. The patients had to be naive to treatment with tramadol and have had no signs of intolerance to treatment with opioids in the past

Exclusion criteria: Patients who were pregnant or lactating were excluded

Tramadol: N=23; mean age 51.4 years; 82.6% female; race not reported. Pain baseline median 3

Placebo: N=12; mean age 51.2 years; 75% % female; race not reported. Pain baseline median 5

Interventions Study medication: Tramadol flexible 100-400 mg/d (Median maximum dose 250 mg/d); Placebo

Rescue medication: Other analgesics (no

43

detailed infromation provided) allowed

Allowed co-therapies: No information provided

Outcomes Pain: Present pain intensity NRS 0-4 at clinical visit *


PGIC: Much or very much improved

Function: Multidimensional Pain Inventory Subscale pain interference *





Notes * No SDs and precise p-values reported; differences between tramadol and placebo were reported to be "significant" (p<0.05). p=0.05 chosen for calculating SD

44

Raja 2002

Methods Disease: Chronic postherpetic neuralgia



Study duration: Approximately 8 weeks titration, maintenance and dose tapering for each period, 1 week drug-free wash-out

Participants Inclusion criteria: Age 18 years, pain persisting for 3 months after the resolution of the cutaneous lesions, and typical pain intensity of 4/10 (numerical rating scale) during the previous week

Exclusion criteria: History of substance abuse or an allergic reaction to an opioid or a TCA, a myocardial infarction in the previous 3 months, cardiac conduction defects, severe pulmonary disease, angle-closure glaucoma, pregnancy, dementia, or encephalopathy were excluded, as were HIV-positive patients and those with life expectancy of 6 months. In addition, patients on monoamine oxidase inhibitors or with severe depression precluding withdrawal from antidepressants were excluded

Total study sample: N=76; Mean age 71 years; 55.3% Female; 88.2% white. Pain baseline opioids 6.5 (±1.9); Pain baseline placebo 6 (±2)

Interventions Study medication: Morphine 15-240 mg/d (average 91 mg/d) or or methadone flexible (average15 mg) or placebo

Rescue medication: Anti-inflammatory medications and acetaminophen (no maximum dosage reported)


Outcomes Pain: Average overall pain intensity during the previous 24 hours NRS 0-10


PGIC: Not assessed

Function: Interference and general activity

45

subscale of the Multidimensional Pain Inventory




Death:Not explicitly stated

Notes "Analysis of the treatment sequence did not reveal significant carry-over or order effects"

46

Schwartz 2011


Study setting: 87 sites in US, 16 sites in Canada

Study design: Enriched enrollment randomized withdrawal

Study duration: 3 weeks open label, 12 weeks double-blind withdrawal

Participants Inclusion criteria: Age ≥ 18 years; type 1 or 2 diabetes mellitus; HbA1c no more than 11% for at least 3 months; Patients taking analgesic medications for at least 3 months prior to screening and dissatisfied with their current therapy; Patients requiring opioid treatment must be taking daily doses of opioid-based analgesic, equivalent to <160 mg of oral morphine; Baseline score of ≥ 5 on an 11-point numeric rating scale, calculated as the average pain intensity during the last 3 days prior to randomization

Exclusion criteria: History of alcohol and/or drug abuse; other chronic pain condition; significant disorder (e.g. pulmonary, psychiatric); moderate to severe hepatic impairment, severely impaired renal function; seizure disorder or epilepsy; uncontrolled hypertension; severely impaired renal function; treatment with neuroleptics, monoamine oxidase inhibitors, serotonin norepinephrine reuptake inhibitors (SNRI), tricyclic antidepressants, anticonvulsants, or anti-parkinsonian drugs; treatment with any other analgesic therapy than investigational medication or rescue medication during the trial

Placebo: N=196; mean age 60.6 years; 39.9% female; 69.9% white. Severe pain baseline 11.2%

Tapentadol: N=199; mean age 59.9 years; 39.3% female; 69.9% white. Severe pain baseline 12.4%

Interventions Study medication: Titration to individually optimal dosage tapentadol 200-500 mg/d, patients with at least 1-point pain reduction were randomised 1:1 to receive optimal fixed dose of tapentadol or placebo

47

Rescue medication: Tapentadol

Allowed co-therapies: Use of selective serotonine reuptake inhibitors allowed of patients were on stable dose for at least 30 days prior to screening

Outcomes Pain: Change from baseline in average pain intensity NRS 0-10

Responder: Maintenance of 50% pain reduction NRS 0-10

PGIC: Maintenance of much or very much improved

Function: Not assessed




Death: Reported

Notes

48

Sindrup 1999

Methods Diasease: Polyneuropathic pain of various etiology (no details on etiology provided)

Study setting: 2 sites in Denmark


Study duration: (1) Up to 1 week wash-out; (2) cross-over treatment sequence for 4 plus 4 weeks with tramadol and placebo 4 weeks each period; (3) a washout period of at least one week seperated the two treatment periods

Participants Inclusion criteria: More than 6 months, polyneuropathy diagnosis confirmed by electrophysiological tests (slowing of nerve conduction or reduction of amplitude of sensory action potential), a pain rating off-medication of at least 4 on a 11-point NRS, and age 20±80 years

Exclusion criteria: Causes of pain other than polyneuropathy, previous allergic reactions towards tramadol, intolerance to tramadol or other opioids, treatment with MAO inhibitors, pregnancy or breast-feeding period, epilepsy and severe terminal illness

Total sample: N=34; mean age 58 years; 44.1% female; race and pain baseline not reported

Interventions Study medication: Tramadol flexible 100-400 mg/d; Average dosage not reported: The final tramadol dose was 400 mg/day in 23 patients, 300 mg/day in four patients and 200 mg/day in seven patients. Placebo

Rescue medication: Acetaminophen up to 3g/d


Outcomes Pain: Average pain intensity last 24 hours NRS 0-10 *

Responder: 50% pain reduction **

PGIC: Not assessed

Function: Not reported

49





Notes * Mean and SDs calculated from raw data presented in table 2

** Only NNT, but no absolute values for both groups reported

*** Period and hang-over effects were not present for pain (p=0.40 and 0.38)

50

Watson 1998

Methods Diasease: Postherpetic neuralgia



Study duration: Duration screening and wash-out not reported, 4 weeks each, no wash-out between cross over periods

Participants Inclusion criteria: History of postherpetic neuralgia ≥ 3months and pain of at least moderate intensity for at least half of the days

Exclusion criteria: Intolerance to oxycodone, a history of drug or alcohol abuse, or significant pain of alternate etiology

Total sample: N=54; evaluable N=36; mean age 63.0 years; 47.2% female; race and pain intensity baseline not reported

Interventions Study medication: Oxycodone flexible 20-160 mg/d (mean daily dosage 45 mg/d), Placebo

Rescue medication: Not reported

Allowed co-therapies: Stable doses of antidepressants, NSAIDs and acetaminophen

Outcomes Pain: Average daily pain intensity diary VAS 0-100


PGIC: Not assessed

Function: Disability as rated by the investigator (none, mild, moderate and severe disability)*





Notes * Not used for meta-analysis, because no patient-

51

reported outcome

"Analysis of treatment sequence revealed no significant carryover effect for the primary variables."

52

Watson 2003


Study setting: 2 sites in Canada


Study duration: 2-7 days wash out, 4 weeks each, no wash-out between cross over periods, up to one year open label

Participants Inclusion criteria: Stable glycemic control suffering painful symmetrical distal sensory neuropathy were enrolled in the study. Patients had at least moderate pain in the lower extremities assessed at the screening visit on a 5-point categorical scale (0, none; 1, mild; 2, moderate; 3, severe; 4, excruciating), a medical history of moderate daily pain based on the patient’s recall over the previous 3 months, one or more symptoms of diabetic neuropathy (including paresthesia, dysesthesia, hyperesthesia, hyperalgesia, and allodynia) and signs of reduced sensation, strength or tendon reflexes not attributable to any other cause.

Exclusion criteria: Intolerance to oxycodone, a history of drug or alcohol abuse, or significant pain of alternate etiology

Total sample: N=54; evaluable N= 36; mean age 63.0 years; 47.2% female; race not reported. Pain baseline 67.0 ± 14.9

Interventions Study medication: Oxycodone flexible 20-80 mg/d (mean daily dosage 40 mg/d), Active placebo flexible 0,5 – 2mg mg/d Benztropine

Rescue medication: Acetaminophen, 325–650 mg 4 to 6 times/d

Allowed co-therapies: Stable doses of antidepressants, anticonvulsants, or non-opioid analgesics

Outcomes Pain: Average daily pain intensity diary VAS 0-100


53

PGIC: Not assessed

Function: Pain Disability Index





Notes "Analysis of treatment sequence revealed no significant carryover effect for the primary variables."

54

Wu 2008

Methods Disease: Postamputation pain



Study duration: 4 weeks titration, 2 weeks maintenance, 2 weeks dose tapering for each period, 1 week drug-free wash-out

Participants Inclusion criteria: Adults (aged 18 yrs) and presence of persistent postamputation pain rated as greater than 3 on a 0–10 numerical rating scale for a period of 6 months or longer

Exclusion criteria: History of allergic reaction to any of the study drugs (i.e., morphine and mexiletine), cardiac conduction defects (e.g., second-degree or complete heart block), myocardial infarction within 3 months of evaluation, severe pulmonary disease, current history of alcohol or substance abuse, seizures, dementia, encephalopathy, current pregnancy or breast-feeding, chronic hepatic disease, hepatic or renal failure, any hematologic disease associated with leukopenia or thrombocytopenia, or the presence of any terminal disease with a life expectancy of less than 6 months

Total sample: N=60; mean age 63.4 years; 21.7% female; 85% white. Pain baseline 6.8 (extracted from figure)

Interventions Study medication: Morphine 15-180 mg/d flexible (mean 112 mg/d) vs. placebo

Rescue medication: NSAIDS (no information on maximum dosage provided

Allowed co-therapies: No information provided.

Outcomes Pain: Average pain intensity NRS 0-10

Responder: 50% pain reduction

PGIC: Not assessed

Function: Multidimensional PIan Inventory interference subscale; detailed results not

55

reported *


Withdrawal due to adverse events: Not reported (only total drop out rate reported)

Serious adverse events: Moderate or severe side effects reported (data not used for analysis)


Notes "we concluded that no significant drug carry-over effects, independent of a time effect, occurred."

* "There were no differences between groups with regard to the effects of study drug on self-reported levels of overall functional activity and pain-related interference in daily activities as assessed by the Multidimensional Pain Inventory

56

Table 2: Support for risk of bias judgement Boureau 2003

Bias Authors'

judgement Support for judgement


Low risk

"computer-generated four-block centralized randomization list"


Unclear risk

We had insufficient information to permit judgement


Low risk

"Both treatments were identical with regard to appearance"


Unclear risk

We had insufficient information to permit judgement. Outcome assessors could be biased by the side effects of tramadol


Unclear risk ITT, method not reported

Selective reporting (reporting bias) High risk

No protocol reported by the authors; means and SDs of Nottingham subscales not available for meta-analysis; SAE not reported

Selection bias Low risk

No significant baseline differences in demographic and clinical variables between the groups

Funding bias High risk

No information on study sponsoring provided; 2 of 3 study authors affiliated with pharmaceutical company

57

Gilron 2005

Bias Authors'



Unclear risk



Low risk

"At the commencement of the trial, a pharmacist at the Kingston General Hospital in Kingston, Ontario, Canada, prepared a concealed allocation schedule randomly”


Unclear risk



Unclear risk

We had insufficient information to permit judgement. Outcome assessors could be unblinded by the side effects of morphine and gabapentin


High risk No ITT


No study protocol provided by authors; SAE not reported


Cross over design

Funding bias Low risk

Study funded by Canadian Institutes of Helath Research. No author affiliated with manufacturer of the drug

58

Gimbel 2003

Bias Authors'



Low risk

“A computer generated randomization schedule with permuted blocks of size 4 was used to assign subjects to study treatment”


Low risk

Randomized information sealed at sponsor site


Low risk

Placebo described as being identical to opioid


Unclear risk

We had insufficient information to permit judgement. Outcomes assessors could be biased by the side effects profile of oxycodone


Unclear risk ITT by LOCF

Selective reporting (reporting bias) Unclear risk

No protocol reported by the authors


No significant baseline differences in demographic and clinical variables between the groups


Stuy funded by pharmaceutical company

59

Harati 1998

Bias Authors'



Low risk

"Computer random number generator"


Low risk

"The randomization code was not revealed to patients, investigators and clinical stuff or study monitors"


Low risk

"Tramadol and placebo were administered as identical capsules"


Unclear risk

Outcome assessors could be unblinded by the side effects of tramadol


Low risk ITT by LOCF

Selective reporting (reporting bias) Low risk

Protocol reported by the authors


The demographic and baseline characteristics of the two treatment groups were similar


The study was sponsored by the manufacturer of the drug. 2 of 9 authors were affiliated with the manufacturer of the drug

60

Khoromi 2007

Bias Authors'



Low risk

“Patients were assigned by random numbers within blocks of four to one of four treatment sequences specified by a Latin square”


Low risk

“Randomization was performed by the NIH Pharmaceutical Development Service”


Low risk

“During the MS Contin treatment period, each blue pill containedMS Contin 15 mg and each pink pill contained inert placebo”


Low risk

"The rate of guessing by the nurses was above the rate for chance only (> 25%), but did not reach a high percentage. “Patients and research staff were blinded to the randomization order”


Unclear risk

ITT-analysis, method not reported


The trial was registered at clinicaltrials.gov (NCT00009672); the primary and the secondary outcomes were consistent in the protocol compared with the publication. SAE not reported


Cross over design


No funding by pharmaceutical industry: the study was sponsored by the National Institute of Dental and Craniofacial Health

61

Norrbrink 2007

Bias Authors'



Unclear risk



Low risk "Sealed and coded envelopes"


Low risk

" The active and placebo drug were identical in appearance"


Unclear risk

No details provided. Outcomes assessors could be biased by the side effects profile of tramadol


Unclear risk ITT-analysis, LOCF


No protocol reported by the authors; imputation method used to calculate SDs

Selection bias High risk

Baseline pain values higher in placebo than in tramadol group


Public funding; no authors affiliated with industry

62

Raja 2002

Bias Authors'



Low risk

"The randomization sequence was computer generated by the biostatistician"


Low risk

" After meeting eligibility criteria, patients were assigned identification numbers by the investigator, and the pharmacy was informed. Study medications were mailed directly to the patients by the pharmacy in a box marked Study Drug A, B, or C, depending on the treatment period


Low risk

"The pharmacist formulated the study drugs in identical gel capsules to maintain the blinding."


Unclear risk

"All investigators were blinded to the drug treatments during the study." Outcomes assessors could be biased by the side effects profile of morphine


Unclear risk ITT-analysis according to LCOF


No protocol reported by the authors. SAE not reported


Cross over design


No funding by pharmaceutical industry reported; no authors affiliated with industry. Study supported by NIH

63

Schwartz 2011

Bias Authors'



Unclear risk



Low risk

Randomization was implemented by interactive voice response system


Unclear risk



Unclear risk

We had insufficient information to permit judgement. Outcomes assessors could be bias based on the side effects profile of tapentadol


Unclear risk

ITT-analysis according to LOCF; BOCF analysis performed, but data not reported


NCT00455520; EuroQol-5 (EQ-5D) Health Status Index and sleep questionnaire data as reported in the protocol were not published


No significant baseline differences in demographic and clinical variables between the three groups


Funding by pharmaceutical industry; 3 of 4 authors affiliated with industry

64

Sindrup 1999

Bias Authors'



Low risk

"Treatment sequence was random via computer generated randomisation code"


Low risk

"Sealed envelopes with treatment sequence for each patient"


Unclear risk

We had insufficient information to permit judgement.


Unclear risk

We had insufficient information to permit judgement. Outcomes assessors could be bias based on the side effects profile of tramadol


High risk No ITT-analysis


No protocol published by the authors; some data for meta-analysis extracted from figures


Cross-over design


Study sponsored by manufacturer of the drug

65

Watson 1998

Bias Authors'



Low risk

Treatment assignment was retained in an opaque envelope


Unclear risk



Unclear risk



Unclear risk

No information provided. Outcomes assessors could be bias based on the side effects profile of oxycodone


High risk No ITT-analysis


No protocol reported by the authors; SAE not reported


Cross over


Funding by pharmaceutical industry; 1 of 2 authors affiliated with manufacturer of the drug

66

Watson 2003

Bias Authors'



Low risk



Low risk

"Patient's treatment assignment was retained in an opaque envelope"


Unclear risk



Unclear risk

No information provided. Outcomes assessors could be bias based on the side effects profile of oxycodone


Unclear risk

ITT-analysis, method not reported

Selective reporting (reporting bias) Unclear risk

No protocol reported by the authors


Cross-over design


Funding by pharmaceutical industry

67

Wu 2008

Bias Authors'



Low risk

"The randomization sequence was computer generated by a biostatistician"


Low risk

"The sequence of drug and placebo treatment periods for each subject was provided in sealed envelopes to the investigational pharmacy and the monitoring committee"


Low risk

"Mexiletine and placebo were similarly packaged in sealed capsules that were identical in appearance"


Unclear risk

No information provided. Outcomes assessors could be bias based on the side effects profile of morphine and mexitilene


High risk

ITT-analysis, method not reported; Only completer data for 50% pain reduction reported


NCT00383682; the primary and the secondary outcomes were consistent in the protocol compared with the publication. However, the standard deviations and p-values of the secondary outcomes function and quality of life not reported. SAE not reported


No significant baseline differences in demographic and clinical variables between the two groups


No funding by pharmaceutical industry; grant by NIH

68

Evidence report – Forest Plots of standardised mean differences and risk differences of opioids compared to placebo on selected

outcomes

Parallel or cross over design

Figure 1 (Electronic Supplementary Material): Effect estimates (standardised mean

differences) of mean pain intensity reduction at end of treatment

Study or Subgroup

1.1.5 Morphine

Gilron 2005

Khoromi 2007

Raja 2002

Wu 2008Subtotal (95% CI)

Heterogeneity: Tau² = 0.02; Chi² = 4.47, df = 3 (P = 0.21); I² = 33%

Test for overall effect: Z = 3.68 (P = 0.0002)

1.1.6 Oxycodone

Gimbel 2003

Watson 1998

Watson 2003Subtotal (95% CI)


Test for overall effect: Z = 5.50 (P < 0.00001)

1.1.9 Tramadol

Boureau 2003

Harati 1998

Norrbrink 2007

Sindrup 1999Subtotal (95% CI)



Total (95% CI)



Test for subgroup differences: Chi² = 1.79, df = 2 (P = 0.41), I² = 0%

Mean

3.3

3.8

4.4

-2.8

-2

35

26.3

25.3

1.4

3

4.24

SD

1.3

2.5

2.4

2

2.1

25

24.7

23

0.8

1.7

2.49

Total

44

28

64

50186

82

38

45165

63

65

23

34185

536

Mean

3.9

3.9

6

-1.4

-1

54

46.7

33.6

2.2

5.5

6.16

SD

1.3

2.4

2

2.7

2

25

26.9

25.4

0.8

1.7

2.3

Total

43

28

56

43170

77

38

45160

62

66

12

34174

504

Weight

9.2%

7.2%

10.7%

9.4%36.5%

12.3%

8.3%

9.1%29.8%

11.2%

10.9%

4.0%

7.7%33.7%

100.0%

IV, Random, 95% CI

-0.46 [-0.88, -0.03]

-0.04 [-0.56, 0.48]

-0.72 [-1.09, -0.35]

-0.59 [-1.01, -0.17]-0.49 [-0.75, -0.23]

-0.48 [-0.80, -0.17]

-0.75 [-1.22, -0.29]

-0.78 [-1.21, -0.35]-0.63 [-0.85, -0.40]

-0.34 [-0.69, 0.01]

-0.99 [-1.36, -0.63]

-1.44 [-2.22, -0.65]

-0.79 [-1.29, -0.30]-0.82 [-1.24, -0.40]

-0.64 [-0.81, -0.46]

Opioids Placebo Std. Mean Difference Std. Mean Difference

IV, Random, 95% CI

-2 -1 0 1 2Favours opioids Favours placebo

69

Figure 2 (Electronic Supplementary Material): Effect estimates (risk difference) of at least

50% pain reduction at end of treatment

Study or Subgroup

1.2.1 Morphine

Wu 2008Subtotal (95% CI)

Total events

Heterogeneity: Not applicable


Total (95% CI)

Total events



Test for subgroup differences: Not applicable

Events

23

23

23

Total

5050

50

Events

13

13

13

Total

4343

43

Weight

100.0%100.0%

100.0%

IV, Random, 95% CI

0.16 [-0.04, 0.35]0.16 [-0.04, 0.35]

0.16 [-0.04, 0.35]

Opioids Placebo Risk Difference Risk Difference

IV, Random, 95% CI

-0.5 -0.25 0 0.25 0.5Opioids Placebo

70

Figure 3 (Electronic Supplementary Material): Effect estimates (risk difference) of reports to

be much or very much improved at end of treatment

Study or Subgroup

1.3.3 Tramadol

Norrbrink 2007Subtotal (95% CI)

Total events



Total (95% CI)

Total events




Events

4

4

4

Total

2323

23

Events

0

0

0

Total

1212

12

Weight

100.0%100.0%

100.0%

IV, Random, 95% CI

0.17 [-0.01, 0.36]0.17 [-0.01, 0.36]

0.17 [-0.01, 0.36]


IV, Random, 95% CI

-0.5 -0.25 0 0.25 0.5Placebo Opioids

71


differences) of physical function improvement at end of treatment

Study or Subgroup

1.4.5 Morphine

Gilron 2005

Khoromi 2007

Raja 2002Subtotal (95% CI)



1.4.6 Oxycodone

Gimbel 2003




1.4.9 Tramadol

Harati 1998

Norrbrink 2007Subtotal (95% CI)



Total (95% CI)




Mean

-57.8

25.7

2.3

-2.4

16.8

-64.3

2.45

SD

26.5

16.5

1.5

2.5

15.6

30.6

1.7

Total

44

28

66138

82

45127

65

2388

353

Mean

-51.7

30.5

2.5

-1.9

25.2

-55.1

3.64

SD

23

15.9

1.5

2.5

16.7

32.5

1.7

Total

43

28

56127

77

45122

66

1278

327

Weight

12.9%

8.3%

18.1%39.4%

23.7%

13.0%36.7%

19.4%

4.5%23.9%

100.0%

IV, Random, 95% CI

-0.24 [-0.67, 0.18]

-0.29 [-0.82, 0.23]

-0.13 [-0.49, 0.22]-0.20 [-0.44, 0.04]

-0.20 [-0.51, 0.11]

-0.52 [-0.94, -0.10]-0.32 [-0.63, -0.02]

-0.29 [-0.63, 0.05]

-0.68 [-1.40, 0.03]-0.36 [-0.67, -0.05]

-0.28 [-0.43, -0.13]

Opioids Placebo Std. Mean Difference Std. Mean Difference

IV, Random, 95% CI

-4 -2 0 2 4Opioids Placebo

72

Figure 5 (Electronic Supplementary Material): Effect estimates (risk difference) of dropping

out due to lack of efficacy during study

Study or Subgroup

1.8.1 Morphine

Khoromi 2007Subtotal (95% CI)

Total events



1.8.2 Oxycodone

Gimbel 2003

Watson 1998


Total events



1.8.4 Tramadol

Harati 1998


Total events



Total (95% CI)

Total events




Events

0

0

1

0

1

2

9

0

9

11

Total

5555

82

38

45165

65

45110

330

Events

3

3

11

1

7

19

22

0

22

44

Total

5555

77

38

45160

66

45111

326

Weight

19.0%19.0%

16.9%

18.7%

12.5%48.1%

9.8%

23.1%32.8%

100.0%

IV, Random, 95% CI

-0.05 [-0.12, 0.01]-0.05 [-0.12, 0.01]

-0.13 [-0.21, -0.05]

-0.03 [-0.10, 0.04]

-0.13 [-0.25, -0.02]-0.09 [-0.17, -0.01]

-0.19 [-0.34, -0.05]

0.00 [-0.04, 0.04]-0.09 [-0.27, 0.10]

-0.07 [-0.13, -0.02]


IV, Random, 95% CI

-1 -0.5 0 0.5 1Favours placebo Favours opioid

73


out due to adverse events during study

Study or Subgroup

1.5.4 Morphine

Gilron 2005

Khoromi 2007

Raja 2002Subtotal (95% CI)

Total events



1.5.5 Oxycodone

Gimbel 2003

Watson 1998


Total events



1.5.8 Tramadol

Boureau 2003

Harati 1998

Norrbrink 2007


Total events



Total (95% CI)

Total events



Test for subgroup differences: Chi² = 2.95, df = 2 (P = 0.23), I² = 32.1%

Events

5

5

7

17

7

5

7

19

6

9

11

8

34

70

Total

44

55

66165

82

38

45165

63

65

23

43194

524

Events

1

1

1

3

4

3

4

11

0

1

2

3

6

20

Total

43

55

56154

77

38

45160

62

66

12

40180

494

Weight

9.1%

14.1%

14.7%38.0%

16.2%

5.2%

5.5%26.9%

16.6%

12.5%

1.1%

4.9%35.2%

100.0%

IV, Random, 95% CI

0.09 [-0.01, 0.19]

0.07 [-0.01, 0.16]

0.09 [0.01, 0.17]0.08 [0.03, 0.13]

0.03 [-0.04, 0.11]

0.05 [-0.08, 0.19]

0.07 [-0.07, 0.20]0.04 [-0.02, 0.10]

0.10 [0.02, 0.17]

0.12 [0.03, 0.21]

0.31 [0.02, 0.61]

0.11 [-0.03, 0.25]0.11 [0.06, 0.17]

0.08 [0.05, 0.12]


IV, Random, 95% CI

-0.5 0 0.25 0.5Placebo Opioid

74

Figure 7 (Electronic Supplementary Material): Effect estimates (risk difference) of serious

adverse events during study

Study or Subgroup

1.6.4 Oxycodone

Gimbel 2003


Total events



1.6.6 Tramadol

Boureau 2003Subtotal (95% CI)

Total events



Total (95% CI)

Total events



Test for subgroup differences: Chi² = 5.03, df = 1 (P = 0.02), I² = 80.1%

Events

5

1

6

3

3

9

Total

82

45127

5353

180

Events

9

3

12

0

0

12

Total

77

45122

5555

177

Weight

31.1%

32.2%63.3%

36.7%36.7%

100.0%

M-H, Random, 95% CI

-0.06 [-0.14, 0.03]

-0.04 [-0.13, 0.04]-0.05 [-0.11, 0.01]

0.06 [-0.01, 0.13]0.06 [-0.01, 0.13]

-0.01 [-0.09, 0.07]


M-H, Random, 95% CI

-0.5 -0.25 0 0.25 0.5Favours placebo Favours opioid

75

Forest Plots of standardised mean differences and risk differences of opioids compared to placebo on selected outcomes

EERW design


differences) of at least 50% pain reduction maintenance

Study or Subgroup

2.3.1 Tapentadol

Schwartz 2011Subtotal (95% CI)

Total events



Total (95% CI)

Total events




Events

55

55

55

Total

9393

93

Events

39

39

39

Total

107107

107

Weight

100.0%100.0%

100.0%

IV, Random, 95% CI

0.23 [0.09, 0.36]0.23 [0.09, 0.36]

0.23 [0.09, 0.36]

Opioid Placebo Risk Difference Risk Difference

IV, Random, 95% CI

-4 -2 0 2 4Favours placebo Favours opioid

76

Figure 9 (Electronic Supplementary Material): Effect estimates (risk difference) of reports to

be much or very much improved at end of treatment

Study or Subgroup

2.6.1 Tapentadol


Total events



Total (95% CI)

Total events




Events

116

116

116

Total

180180

180

Events

68

68

68

Total

177177

177

Weight

100.0%100.0%

100.0%

IV, Random, 95% CI

0.26 [0.16, 0.36]0.26 [0.16, 0.36]

0.26 [0.16, 0.36]


IV, Random, 95% CI

-0.5 -0.25 0 0.25 0.5Placebo Opioid

77


out due to adverse events during study

Study or Subgroup

2.8.7 Tapentadol


Total events



Total (95% CI)

Total events




Events

29

29

29

Total

196196

196

Events

15

15

15

Total

193193

193

Weight

100.0%100.0%

100.0%

IV, Random, 95% CI

0.07 [0.01, 0.13]0.07 [0.01, 0.13]

0.07 [0.01, 0.13]


IV, Random, 95% CI

-0.5 -0.25 0 0.25 0.5Placebo Opioid

78

Figure 11 (Electronic Supplementary Material): Effect estimates (risk difference) of serious

adverse events during study

Study or Subgroup

2.9.6 Tapentadol


Total events



Total (95% CI)

Total events




Events

10

10

10

Total

196196

196

Events

3

3

3

Total

193193

193

Weight

100.0%100.0%

100.0%

IV, Random, 95% CI

0.04 [0.00, 0.07]0.04 [0.00, 0.07]

0.04 [0.00, 0.07]


IV, Random, 95% CI

-0.5 -0.25 0 0.25 0.5Placebo Opioid

Opioids in chronic neuropathic pain - A systematic review ...

Documents

Transcript of Opioids in chronic neuropathic pain - A systematic review ...