Effectiveness of Exercise Therapy in Treatment of Originally … 2014.pdf · 2020. 3. 14. ·...

doi: 10.2522/ptj.20130310Originally published online July 31, 2014

2014; 94:1697-1708.PHYS THER. Ron Clijsen, Janine Fuchs and Jan TaeymansSystematic Review and Meta-AnalysisPatients With Patellofemoral Pain Syndrome: Effectiveness of Exercise Therapy in Treatment of

http://ptjournal.apta.org/content/94/12/1697found online at: The online version of this article, along with updated information and services, can be

Online-Only Material 30310.DC1.html

http://ptjournal.apta.org/content/suppl/2014/12/01/ptj.201

Collections

Therapeutic Exercise Systematic Reviews/Meta-analyses

Injuries and Conditions: Lower Extremity Injuries and Conditions: Knee

Evidence-Based Practice in the following collection(s): This article, along with others on similar topics, appears

e-Letters

"Responses" in the online version of this article. "Submit a response" in the right-hand menu under

or click onhere To submit an e-Letter on this article, click

E-mail alerts to receive free e-mail alerts hereSign up

by Dienst E-bronnen on December 8, 2014http://ptjournal.apta.org/Downloaded from by Dienst E-bronnen on December 8, 2014http://ptjournal.apta.org/Downloaded from

http://ptjournal.apta.org/content/94/12/1697http://ptjournal.apta.org/content/suppl/2014/12/01/ptj.20130310.DC1.htmlhttp://ptjournal.apta.org/content/suppl/2014/12/01/ptj.20130310.DC1.htmlhttp://ptjournal.apta.org/cgi/collection/evidencebased_practicehttp://ptjournal.apta.org/cgi/collection/injuries_and_conditions_kneehttp://ptjournal.apta.org/cgi/collection/injuries_and_conditions_lower_extremityhttp://ptjournal.apta.org/cgi/collection/systematic_reviewshttp://ptjournal.apta.org/cgi/collection/therapeutic_exercisehttp://ptjournal.apta.org/letters/submit/ptjournal;94/12/1697http://ptjournal.apta.org/subscriptions/etoc.xhtmlhttp://ptjournal.apta.org/http://ptjournal.apta.org/

Effectiveness of Exercise Therapy inTreatment of Patients WithPatellofemoral Pain Syndrome:Systematic Review and Meta-AnalysisRon Clijsen, Janine Fuchs, Jan Taeymans

Background and Purpose. This systematic review and meta-analysis wasaccomplished to determine whether exercise therapy is an effective intervention toreduce pain and patient-reported measures of activity limitations and participationrestrictions (PRMALP) in patients with patellofemoral pain.

Data Sources and Study Selection. Randomized controlled trials in Englishand German languages published in the MEDLINE, Physiotherapy Evidence Database(PEDro), International Clinical Trials Registry Platform, and Cochrane databases weresearched. Eligibility was assessed in 2 stages. The methodological quality of thestudies was rated using the PEDro scale. Data were pooled using random-effectsmeta-analysis, allowing for variability among studies. For clinical use, overall estimateswere re-expressed in the original visual analog scale scores. Significance was set at5%.

Data Extraction and Data Synthesis. Fifteen studies, with a total of 748participants, were included and pooled for the meta-analysis. Six studies comparedthe effect of exercise therapy with a control group receiving neither exercise therapynor another intervention. Four studies compared the effect of exercise therapy versusadditive therapy, and 5 studies compared different exercise interventions. In bothcomparisons, exercise therapy resulted in strong pain reduction and improvement ofPRMALP effects. Significant short-term effects (�12 weeks) of exercise therapy werefound for pain and PRMALP, whereas long-term effects (�26 weeks) were observedfor PRMALP only.

Limitations and Conclusion. The 15 studies included in this analysis were ofvariable quality. Large-scale, high-quality randomized controlled trials are needed tofurther the evaluation of the possible effects of different exercise therapy modalitieson patellofemoral pain. This meta-analysis presents evidence that exercise therapyhas a strong pain-reducing effect and decreases PRMALP in patients with patellofemo-ral pain. However, the question of which exercise modality yields the strongestreducing effect on pain and PRMALP remains unanswered.

R. Clijsen, PhD, Department ofHealth Sciences, University ofApplied Sciences and Arts ofSouthern Switzerland, Weststrasse8, CH-7302 Landquart, Switzer-land, and Health Department,University College Physiotherapy,“Thim van der Laan,” Landquart,Switzerland. Address all corre-spondence to Dr Clijsen at:[email protected].

J. Fuchs, PT, Health Department,University College Physiotherapy,“Thim van der Laan,” Landquart,Switzerland.

J. Taeymans, PhD, Health Depart-ment, Bern University of AppliedSciences, Bern, Switzerland.

[Clijsen R, Fuchs J, Taeymans J.Effectiveness of exercise therapy intreatment of patients with patel-lofemoral pain syndrome: system-atic review and meta-analysis.Phys Ther. 2014;94:1697–1708.]

© 2014 American Physical TherapyAssociation

Published Ahead of Print:July 31, 2014

Accepted: July 19, 2014Submitted: July 26, 2013

Research Report

Post a Rapid Response tothis article at:ptjournal.apta.org

December 2014 Volume 94 Number 12 Physical Therapy f 1697 by Dienst E-bronnen on December 8, 2014http://ptjournal.apta.org/Downloaded from

http://ptjournal.apta.org/

Patellofemoral pain syndrome(PFPS) is a common musculo-skeletal disorder in physicallyactive individuals aged 15 to 30years.1,2 The prevalence ranges from21% to 45% in active adolescents andfrom 15% to 33% in adults3 and ishigher in females than in males.2,4

However, there are limited epidemi-ological incidence data.2,5,6 The mostreported complaint is retropatellarpain or diffuse peripatellar pain dur-ing activities such as running,ascending and descending stairs,squatting, and sitting with flexedknees for prolonged periods oftime.7,8 Other common symptomsare crepitation and the giving-wayphenomenon.9–11

The etiology of PFPS appears to bemultifactorial. Several causative fac-tors relating to the malalignment ofthe lower extremity kinematics aredescribed in the literature. Altered orexcessive foot pronation may resultin a compensatory internal rotationof the tibia and increased valgusstress.12–15 The quadriceps musclestabilizes the gliding of the patellathrough the femoral groove. Animbalance between the force of thevastus medialis obliquus muscle andother quadriceps muscle groups canlead to a lateral displacement of thepatella, resulting in patellofemoraljoint stress on the lateral facet.

Biomechanical studies have demon-strated that patients with patel-lofemoral pain exhibit greaterfrontal-plane16 and transverse-plane17,18 hip motion during activi-ties of daily living. It has beenhypothesized that increased frontal-and transverse-plane hip motion mayaffect the lateral forces acting on thepatella.13 Furthermore, internal rota-tion of the femur during the stancephase is related to lateral patellar dis-placement and thus may be a con-tributor to altered patellofemoraljoint kinematics.13 Hamstring muscletightness, iliotibial band tightness,patellar retinaculum tightness, andoveractivity also may increase patel-lofemoral joint stress.19–21 Althoughdifferent hypotheses of the biome-chanical and neurophysiologicalmechanisms of the beneficial effectsof exercise therapy exist, there isconsiderable individual variation.For example, malalignment of thepatellofemoral joint does not neces-sarily correlate with pain,22 indicat-ing that no particular exercise pro-gram is effective for all patients withPFPS.1,7,11,19,23

Currently, the most accepted theorybehind PFPS relates the symptoms toexcessive patellofemoral jointstress13,22,24 modifying the soft tissuehomeostasis.15,22,25 The hyperinner-vation and vascularization resultingfrom the ischemic tissue can causethe pain sensation.

Because no consensus exists on thedefinition of a clinical diagnosis orclassification,26 no validation of clin-ical tests is possible. Therefore, PFPSis often a diagnosis based on theexclusion of other pathologies.24

A conservative approach has beenrecommended as a first-line treat-ment for patients with PFPS.9,19,27

The most successful rehabilitationshould progress without increasingsymptoms.28,29 Currently, consensusregarding the most appropriate con-

servative treatment for PFPS is stilllacking. A widely accepted interven-tion to treat PFPS is exercise ther-apy.19,23,30–32 Good clinical resultshave been achieved with differentvariations of quadriceps musclestrengthening.11 Exercise therapy isoften combined with static quadri-ceps muscle stretching exercises(Tab. 1).

The aim of this systematic reviewwas to establish an overview of thecurrently available evidence for theeffectiveness of exercise therapy onpain and functional improvement,which was defined as “patient-reported measures of activity limita-tions and participation restrictions”(PRMALP), in patients with PFPS.33

Furthermore, this study aimed toreach more conclusive results bysummarizing the available studies ina meta-analysis.

MethodData Sources and SearchStrategyThe systematic literature search wasperformed according to the Pre-ferred Reporting Items for System-atic Reviews and Meta-Analyses(PRISMA) statement.34 An electronicsearch was conducted during theperiod of January to August 2011and updated at the end ofDecember 2013 on the followingdatabases: MEDLINE (PubMed),Cochrane Database of SystematicReviews, Physiotherapy EvidenceDatabase (PEDro),35 and the Interna-tional Clinical Trials Registry Plat-form. A manual search of the refer-ence lists of retrieved publicationswas completed to screen for topic-related studies.

The sensitive search strategy wasestablished combining the followingMeSH-listed key words in the searchalgorithm: (“patellofemoral pain syn-drome” OR “anterior knee pain” OR“chondromalacia patellae”) AND(“exercise therapy” OR “therapy”).

Available WithThis Article atptjournal.apta.org

• eFigure 1: Flowchart Describingthe Systematic Review Procedure

• eFigure 2: Forest Plot of the 6Studies Evaluating the Short-TermEffect of Exercise Therapy VersusNo Exercise on Pain

• eFigure 3. Forest Plot of 2Studies Evaluating the Long-TermEffect of Exercise on PainCompared With No ExerciseTherapy

Exercise Therapy for Patients With Patellofemoral Pain

1698 f Physical Therapy Volume 94 Number 12 December 2014 by Dienst E-bronnen on December 8, 2014http://ptjournal.apta.org/Downloaded from

http://ptjournal.apta.org/content/94/12/1697/suppl/DC1http://ptjournal.apta.org/content/94/12/1697/suppl/DC1http://ptjournal.apta.org/content/94/12/1697/suppl/DC1http://ptjournal.apta.org/

Table 1.RCTs Comparing Exercise Therapy and No Exercisea

StudyPEDroScore

Participants(Age: X�SD in

Years) Intervention GroupControl Group (or

Intervention 2) Outcome Variables Main Results

Clark et al,23

20007 17 f, 25 m

EX group age:29.5�6.2

CON group age:27.1�7.2

Duration: 12 wkEX: 6 sessions; warm-up,

squatting on the wall, sit-to-stand, proprioceptivebalance work, EX forgluteus medius andmaximus muscles,progressive step-down

Home program: dailyexercises

Duration: 6 sessionsAdvice and (1)

explanation aboutanterior knee pain, (2)footwear and sportingactivities, (3) pain-controlling drugs, (4)stress relaxation, (5) dietand weight advice, and(6) prognosis and self-help

VAS and WOMACscores after 12 wkand 52 wk

After 12 wk, pain reductionand PRMALPimprovement in bothgroups (P�.0001), nosignificant difference inpain and PRMALPimprovement betweengroups

After 52 wk, pain reduction(P�.005) and PRMALPimprovement (P�.007) inboth groups, nodifference in pain andPRMALP improvementbetween groups (P�.05)

Herrington andAl-Sherhi,48

2007

6 45 mAge: 26.9�5.6

Duration: 6 wkEX: 3 sessions per week for

6 wk; warm-up, OKC in aseated position, CKC onleg press

No intervention VAS and functionalperformance(modified Kujalascale) after 6 wkduring differentactivities

Pain reduction in EX groups(P�.015), differencebetween groups in pain(P�.01), PRMALPimprovement in EXgroups (P�.05), nodifference between EXgroups (P�.05)

van Linschotenet al,43 2009

6 84 f, 47 mEX group age:

24.7�8.6CON group age:

23.3�7.8

Duration: 12 wkEX: 9 sessions in 6 wk under

supervision; daily EX for 25min over 12 wk; warm-upon ergometer; differentmuscular EX forquadriceps, adductor, andgluteal muscles; flexibilityEX for thigh muscles

Duration: 12 wkAdvice: rest during periods

of time, refraining frompain-provoking activitiesand usual care

VAS during rest andactivity andfunctional disability(Kujala scale) after12 and 52 wk

Pain reduction in EX groupafter 12 wk (P�.02) and52 wk (P�.02), differencebetween groups in pain(P�.01) after 12 wk and52 wk, higher functionscore in EX group after12 wk (P�.04), nodifference in PRMALPbetween EX and CONgroups (P�.09) after 52wk

Song et al,49

20098 69 f, 20 m




8 wk; leg press or leg presscombined with isometrichip adduction

Duration: 8 wkAdvice: health educational

material regarding PFPS

VAS-U and VAS-W andfunctional ability(Lysholm scale) after8 wk

Pain reduction in EX group(P�.005), difference inpain between EX andCON groups (P�.008),PRMALP improvement inEX group versus CONgroup (P�.005)

Fukuda et al,47

20108 70 f



Duration: 4 wkEX: 12 sessions in 4 wk;

strengthening andstretching, strengtheningof the hip abductor andlateral rotator muscles

No intervention NPRS and functionalimprovement (LEFS)after 4 wk duringdifferent activities

Pain reduction in EX group(P�.008), difference inpain between EX andCON groups (P�.01),PRMALP improvement(LEFS) in EX group(P�.05)

Khayambashiet al,38 2012

5 28 fEX group age:


30.5�4.8


8 wk; 5-min warm-up, 20-min hip strengthening,5-min cool-down

No intervention VAS and WOMACafter 8 wk and 26wk

Pain reduction in EX group(P�.001); after 26 wk,pain reduction in EXgroup compared withbaseline level (P�.001),PRMALP improvement inEX group compared withCON group (P�.001)

a RCT�randomized controlled trial, f�female, m�male, EX�exercise, CON�control, OKC�open kinetic chain, CKC�closed kinetic chain, VAS�visualanalog scale, PRMALP�patient-reported measures of activity limitations and participation restrictions, WOMAC�Western Ontario and McMaster UniversitiesOsteoarthritis Index, PFPS�patellofemoral pain syndrome, LEFS�Lower Extremity Functional Scale, NPRS�numerical pain rating scale, VAS-U�VAS usualscore, VAS-W�VAS worse score.




Research Question and StudySelectionTo establish the research question,the recommendations from the PICOmodel (Population: patients withpatellofemoral pain syndrome; Inter-vention: exercise therapy; Compara-tor: no exercise or exercise withadditive intervention [electrical mus-cle stimulation (EMS) or splints];

Outcome: pain and PRMALP) wereused.

Randomized controlled trials (RCTs)regarding the effect of exercise ther-apy in patients with PFPS weresourced. Studies comparing partici-pants with pathological conditionsand those with nonpathological con-ditions and studies with compari-

sons based on taping, chiropracticinterventions, foot orthoses, and sur-gical treatments were excluded.Studies including participants withother knee pathologies were notincluded. Full texts of articles writ-ten in German or English wereincluded. No restrictions wereimposed on the year of publication.

Table 2.RCTs Comparing Exercise and Exercise With Additive Therapy (EMS or Splints)a

StudyPEDroScore

Participants(Age: X�SD in

Years) Intervention GroupControl Group (or

Intervention 2) Outcome Variables Main Results

Dursunet al,46

2001



36.9�9.2

Duration: 12 wkEX: 5 days per week forthe first 4 wk and 3

days per weekthereafter;strengthening of thequadriceps andvastus medialisobliquus muscles,flexibility training,proprioceptiontraining, endurancetraining

Duration: 12 wkEMG � EX program: 3 days

per week for 4 wk

VAS and FIQ after 4, 8,and 12 wk

Significant (P�.0001)improvements (VASand FIQ) in bothgroups, no differencebetween groups(P�.05)

Schneideret al,50

2001


23CON group age:

21

Duration: 8 wkEX: 2 sessions per week

of 1 h duration for 8wk; exercise basedon proprioceptiveneuromuscularfacilitation

Duration: 8 wkKnee splint: 15 min 3�/d

combined with exercisetherapy to strengthen theischiocrural muscles

VAS after exposure andBessette and Hunterscale at rest after 4and 8 wk

Pain reduction in bothEX group (P�.003)and EX � splint group(P�.0005), differencesbetween groups inVAS (P�.0065) andBessette and Hunter(P�.0047) scores infavor of EX � splintgroup

Yip andNg,42

2006

6 16 f, 10 mAge: 32.5�8.8

Duration: 8 wkEX: home program for

15 min daily;quadriceps musclestrengthening,balance andproprioceptiontraining, plyometricand agility training,flexibility training

Duration: 8 wkEX: home program like

intervention 1,additionally EMGbiofeedback system

Patellofemoral PainSyndrome SeverityScale after 4 and8 wk

Pain reduction (P�.088),no difference in painbetween groups(P�.05)

Bily et al,41

20085 14 m, 24 f



Duration: 12 wkEX: 7 sessions per week

for 2 wk, 3 sessionsper week thereafterunder supervision;isometric, concentric,and eccentric legraises and pulls,stepping, squatting,balance exercises,static stretching

Duration: 12 wkEX program (like

intervention group) �daily EMS of thequadriceps muscles, 2sessions for 20 min/d

VAS (maximum) andKujala patellofemoralscore after 12 and52 wk

Pain reduction (P�.003)and PRMALPimprovement(P�.001) in bothgroups withoutsignificant differencebetween groups

a RCT�randomized controlled trial, f�female, m�male, EX�exercise, CON�control, EMS�electrical muscle stimulation, EMG�electromyography,VAS�visual analog scale.




Identification of Literature andQuality AssessmentEligibility, based on the a priori setinclusion and exclusion criteria, wasassessed in 2 stages. First, studieswere screened on titles andabstracts. In a second step, full textsof the eligible publications were ana-lyzed. Additional information wasrequested from the correspondingauthors in case of missing data.

The methodological quality of thestudies was assessed using the PEDrorating scale. For studies retrievedfrom PEDro, the mentioned PEDroscores were accepted. If the scorewas not available, the quality wasevaluated by 2 reviewers (M.K. andF.J.) independently using the PEDrorating scale. Disagreements wereresolved in a consensus meeting.

Data Synthesis andMeta-AnalysisThe primary outcome measure forthis meta-analysis was pain, assessedwith a visual analog scale. From theretrieved studies, PRMALP data eval-uated with standardized scoringscales (Kujala patellofemoral score,Western Ontario and McMaster Uni-versities Osteoarthritis Index[WOMAC], Lysholm scale, Func-tional Index Questionnaire [FIQ],Lower Extremity Functional Scale[LEFS]) were extracted as a second-ary outcome. Most studies presentedVAS and PRMALP data as means andstandard deviations, making poolingof the data possible.

A meta-analysis with a random-effects model (specified a priori),accounting for possible between-studies heterogeneity, was used tocalculate the overall effect size ofexercise therapy on patellofemoralpain and PRMALP. Effect sizes of theRCTs included in the present meta-analysis were expressed as Hedges’ gto correct for small-scale studies’effect of overestimating the trueeffect size. Ninety-five percent confi-

dence intervals (95% CIs) were cal-culated for the individual studies andthe overall estimates of the differentmeta-analyses. The Q statistic and itsP value were calculated to assessbetween-studies heterogeneity, andI2 was used to express the degree ofheterogeneity. For the purpose ofbetter clinical understanding of theresults for pain, the overall estimateswere re-expressed as the originalVAS scores.36 Publication bias wasassessed by visually analyzing thefunnel plot and by formal testing forfunnel plot asymmetry37 using the“trim and fill” and the “fail ’n safe”algorithms scores. All meta-analyticcalculations and plots were con-ducted using CMA-2 software (Com-prehensive Meta-Analysis, version 2,Biostat, Englewood, New Jersey).

ResultsFlow of the Search ProcedureThe initial literature search resultedin a total of 285 studies. Two hun-dred two publications wereexcluded after screening the titleand abstract. Out of the remaining83 studies, 27 had to be excludedbecause of insufficient information,4 were excluded for justifiable rea-sons (healthy versus unhealthy par-ticipants, no isolated effect eluci-dated), and 39 duplicates wereremoved, resulting in 13 studiesmeeting the inclusion criteria. Thefinal update search (December2013) yielded 2 more publica-tions.38,39 Thus, 15 studies wereincluded in the present meta-analysis(eFig. 1, available at ptjournal.apta.org).

Study Design and Characteristicsof PopulationTables 1, 2, and 3 depict the generalcharacteristics of the 15 includedstudies by different comparisons (ie,therapy versus no therapy, exerciseversus exercise with additive ther-apy, and knee extension exercisesversus other forms of exercises,respectively).

Exercise characteristics. Exercisemodalities, regimens, and settingsvaried widely across the trials. In 4studies, patients were instructed toexercise daily.40–43 In 9 studies, thetherapy program consisted of 3 to 5weekly exercise sessions.20,38,39,44–49

Two studies reported 2 sessions orless per week.23,50 Four studies usedhome-based exercise therapy inter-ventions.20,23,43,44 From the 15included studies, only 6 providedsufficient information regarding theexercise therapy load.20,38,39,42,43,50

Exercise therapy versus no exer-cise. Exercise therapy was basedon quadriceps muscle strength train-ing in open or closed kineticchain,23,43,47–49 hip strength train-ing,23,38,43,47 adductor training,43,49

and proprioceptive and flexibilitytraining.23,43 Control group partici-pants received advice on pain ori-gin23,43,49 or no information atall38,47,48 (Tab. 1).

Meta-AnalysesShort-term and long-term effectsof exercise therapy versus noexercise on pain. A total of 6 stud-ies providing data on 349 partici-pants (exercise group: n�189; con-trol group: n�160) were included inthe meta-analysis.23,38,43,47–49 Nega-tive overall estimates represent painreduction and thus favor exercisetherapy. The mean overall effect size(Hedges’ g) for postintervention val-ues was �1.18 (95% CI��1.86,�0.51), favoring exercise therapy.Between-studies heterogeneity forpostintervention values was signifi-cant (Q�39.57, P�.0001) and high(I2�87.4%). A subgroup analysisshowed that part of this heterogene-ity could be explained by sex and thetype of intervention (hip training,knee training, or combined). For thepurpose of a better clinical under-standing, these values werere-expressed as the original painscale scores.36 An overall estimate(Hedges’ g) of �1.18 represents a



http://ptjournal.apta.org/content/94/12/1697/suppl/DC1http://ptjournal.apta.org/

Table 3.RCTs Comparing Knee Extension Exercises and Other Forms of Exercisesa

StudyPEDroScore

Participants(Age: X�SD

in Years) Intervention 1 Intervention 2OutcomeVariables Main Results

Witvrouwet al,20

2000

6 40 f, 20 mAge: 20.3


5 wk of 30–45 minduration; OKC maximum,static quadriceps musclecontraction, SLR, short-arcmovements, legadduction, stretching

Home EX: participantsadvised to maintain theirmuscle strength


5 wk of 30–45 minduration; CKC leg presses,knee bends, bicycling,rowing machine exercises,step-up and step-down,jumping exercises,stretching

Home EX: participantsadvised to maintain theirmuscle strength

VAS andfunctionalperformance(Kujala scale)after 5 and12 wk

Pain reduction in the CKCgroup (P�.05), no painreduction in the OKCgroup (P�.05); PRMALPimprovement in bothgroups; after 5 wk, CKCgroup (P�.002) and OKCgroup (P�.001); after 12wk, CKC group (P�.001)and OKC group(P�.0004); differencebetween groups for pain(P�.024); no differencebetween groups forPRMALP (P�.05)

BakhtiaryandFatemi,40

2008

6 32 fGroup 1 age:

22.3�1.7Group 2 age:

21.8�0.6

Duration: 3 wkEX: 42 sessions in 3 wk

(2 sessions per day); SLRexercises (OKC) in supineposition until 45° hipflexion and hold it for3–4 s


(2 sessions per day); semi-squat exercises (CKC)until 15°�20° of kneeflexion, hold for 3–4 s

VAS and MIVCFafter 3 wk

Pain reduction in bothgroups, with no differencebetween groups (P�.13);increased muscle force inthe semi-squat group(P�.01)

Nakagawaet al,45

2008

7 10 f, 4 mAge: 23.6�5.9


6 wk; quadriceps musclestrengthening andstretching


6 wk; quadriceps musclestrengthening andstretching supplementedby strengthening andfunctional trainingfocused on the transversusabdominis, hip abductor,and lateral rotator muscles

VAS-U andVAS-W after6 wk

Pain reduction in group2 (P�.05), no decrease inpain in group 1; increasedeccentric isokinetic kneeextensor peak torque forboth intervention groups(P�.04) and control group(P�.02)

Balci etal,44

2009

5 40 fGroup 1 age:

39.1�8.0Group 2 age:

36.1�8.7


under supervision;functional squat exerciseswith hip internal rotationand 0°�45° flexion intervalof the knee

Home EX program: 6 wk;SLR in supine position,strengthening hipadductor muscles in side-lying position, isometric


under supervision;functional squat exercisewith hip external rotationand 0°�45° flexioninterval of the knee

Home EX program: 6 wk;SLR in supine position,strengthening hipadductor muscles in side-lying position, isometric

VAS in rest andactivity after4 wk and10 wk

Functionalcapacity after4 wk and10 wk

Pain reduction in bothgroups (P�.01), nodifference betweengroups; PRMALPimprovement in bothgroups (P�.001) andcontrol group (P�.001),no difference betweengroups

Dolak etal,39

2011

6 33 fGroup 1 age:

26�6Group 2 age:

25�5

Duration: 8 wkEX (weeks 2–4): 3 sessions

per week consisting ofshort-arc squats, SLR, andterminal knee extensionswith progression of theresistance

EX (weeks 5–8): single-legbalance with either frontor diagonal pull, wallslides, step-downs, calfraises, lunges, squats,flexibility and balancetraining

Duration: 8 wkEX (weeks 2–4): standing

hip abduction, side-lyinghip abduction, seated hipexternal rotation,combined hip abductionand external rotation withprogression of theresistance

EX (weeks 5–8): the sameintervention as the othergroup

VAS, functionalimprovement(LEFS) andisometricstrength ofHABD, HER,and KE after4 wk, 8 wk,and 12 wk

Pain reduction in group 1after 8 wk (P�.028), ingroup 2 after 4 and 8 wk(P�.001, P�.003);improvement in LEFSscores in both groups after4 wk (P�.006) and 8 wk(P�.006); increased HABDstrength after 8 wk(P�.001); increased HERstrength after 8 wk inboth groups (P�.004); inboth groups, no increasein KE strength (P�.39)

a RCT�randomized controlled trial, f�female, m�male, EX�exercise, OKC�open kinetic chain, CKC�closed kinetic chain, VAS�visual analog scale, VAS-U�VAS usual score, VAS-W�VAS worse score, SLR�straight leg raise, LEFS�Lower Extremity Functional Scale, PRMALP�patient-reported measures of activitylimitations and participation restrictions, MIVCF�maximal isometric voluntary contraction force, HABD�hip abductors, HER�hip extensors, KE�kneeextensor.




pain reduction of 40 mm on the100-mm VAS pain scale afterre-expression (eFig. 2, available atptjournal.apta.org).

To evaluate the long-term (�26weeks) effect of exercise therapy onpain, 2 studies providing data on 197patients comparing exercise therapywith no exercise therapy23,43 couldbe analyzed. The mean overall effectsize (Hedges’ g) for postinterventionvalues was �0.48 (95% CI��0.80,�0.16), yielding a statisticallynonsignificant pain reduction afterre-expression on the VAS pain scaleof 10 mm (eFig. 3, available atptjournal.apta.org).

Short-term PRMALP improve-ment exercise therapy versus notherapy. The meta-analysis on theeffectiveness of exercise therapyversus no therapy on the short-term (�12 weeks) PRMALP improve-ment revealed a negative overallestimate (Hedges’ g) of �1.30(95% CI��2.11, �0.50), indicatingPRMALP improvement favoringexercise therapy. Between-studiesheterogeneity for postinterventionvalues was significant (Q�57.99,P�.0001) and high (I2�91.4%).Again, subgroup analysis showedthat part of this heterogeneity couldbe explained by sex and the type ofintervention (Fig. 1).

Three studies were selected for asubgroup analysis, comparing theoutcome of exercise therapy onlong-term (�26 weeks) PRMALPimprovement.23,38,43 The overallsummary estimate (Hedges’ g) of thepooled standard deviation for post-treatment values on PRMALP was�0.99 (95% CI��2.28, �1.51),favoring exercise therapy (P�.131)(Fig. 2).

Exercise therapy versus additivetherapy. Two included studiesexamined the efficacy of biofeed-back supplementation in addition toexercise therapy in PFPS treat-ment.42,46 The study by Bily et al41

Study Subgroup Within Study Time Point Statistics for Each Study Hedges’ g and 95% CI

–4.00 –2.00 0.00 2.00 4.00

Hedges’g

StandardError Variance

LowerLimit

UpperLimit z Value P Value

Clark et al,23 2000Herrington and Al-Sherhi,48 2007van Linschoten et al,43 2009Song et al,49 2009Fukuda et al,47 2010Khayambashi et al,38 2012

FMMFMFMFF

126

12848

–0.082–3.438–0.342–0.922–0.721–2.896–1.301

0.3250.4770.1750.2510.2650.5330.410

0.1060.2280.0310.0630.0700.2840.168

–0.719–4.374–0.685–1.413–1.240–3.941–2.105

0.555–2.503 0.001–0.431–0.201–1.851–0.497

–0.253–7.202–1.956–3.681–2.719–5.430–3.172

.801

.000

.050

.000

.007

.000

.002

FavorsExercise Therapy

FavorsNo Exercise Therapy

Figure 1.Forest plot of 6 trials evaluating the short-term patient-reported measures of activity limitations and participation restrictions(PRMALP) improvement of exercise therapy versus no therapy in the treatment of patellofemoral pain syndrome. 95% CI�95%confidence interval, F�female, M�male.

Study Subgroup Within Study Time Point Statistics for Each Study Hedges’ g and 95% CI

–4.00 –2.00 0.00 2.00 4.00

Hedges’g

StandardError Variance

LowerLimit

UpperLimit z Value P Value

Clark et al,23 2000van Linschoten et al,43 2009Khayambashi et al,38 2012

FMFMF

525226

–0.078–0.315–2.820–0.993

0.3760.1800.5260.658

0.1410.0330.2770.433

–0.815–0.668–3.851–2.282

0.658 0.039–1.789 0.297

–0.209–1.744–5.362–1.509

.835

.081

.000

.131

FavorsExercise Therapy

FavorsNo Exercise Therapy

Figure 2.Forest plot of 3 studies evaluating the short-term patient-reported measures of activity limitations and participation restrictions(PRMALP) improvement of exercise therapy versus no therapy in the treatment of patellofemoral pain syndrome. 95% CI�95%confidence interval, F�female, M�male.



http://ptjournal.apta.org/content/94/12/1697/suppl/DC1http://ptjournal.apta.org/content/94/12/1697/suppl/DC1http://ptjournal.apta.org/

evaluated the efficacy of a trainingprogram including electrical musclestimulation in which the electrodeswere positioned on the vastus medi-alis and lateralis muscles. The appli-cation was performed daily41,42 or3 times per week.46 Schneider et al50

evaluated the use of a knee splint.The comparative exercise therapyconsisted of strengthening, flexibility,proprioceptive, and endurance train-ing41,42,46 and proprioceptive neuro-muscular facilitation50 (Tab. 2).

Four studies providing data after 4, 8,and 12 weeks of intervention in 81patients receiving an exercise inter-vention and 81 patients receivingexercise plus an additive therapy(EMS or splints) were included toevaluate the short-term (�12 weeks)pain effect on PFPS.41,42,46,50 Toavoid erroneous deflating of the vari-ance, the number of participants wasdivided by the number of measure-ment sessions for meta-analysis. Theoverall summary estimate (Hedges’g) for posttreatment values was 0.12(95% CI��0.29, 0.53, P�.57). Theobserved moderate heterogeneitywas not significant (Q�11.92,P�.103, I2�41.3%).

Of the above-mentioned studies,only 2 reported on PRMALP in 48patients receiving an exercise inter-vention and 45 patients receivingexercise plus an additive ther-apy.41,46 The overall summary esti-mate for posttreatment values was�0.60 (95% CI��1.00, �0.19,P�.004), favoring exercise therapyonly. No heterogeneity wasobserved (Q�0.07, P�.79, I2�0%).

Effect of different exercise inter-ventions on pain. Two studiescompared the effect of closed andopen kinetic chain exercises.20,40

Open chain exercises relative to thehip are based on straight leg raises,whereas closed chain exercisesrelate to leg press or squat exercises.

The additional effect of hip strengthtraining to quadriceps musclestrength training was evaluated in 2studies.39,45 The focus was set on thehip joint abductors and lateral rota-tors (Tab. 3).

Two subgroup analyses were per-formed to compare different exer-cise interventions. The first compar-ison was between closed and openkinetic chain exercises.20,40 Ninety-two participants were included. Themean overall effect size (Hedges’ g)for the postintervention values was0.08, although this value was not sta-tistically significant (95% CI��0.32,0.48, P�.695). No heterogeneitywas observed (Q�0.23, P�.634),I2�0%).

The second comparison wasbetween patients with exercise ther-apy focused on quadriceps musclestrength training and exercise ther-apy focused on hip muscle strengthtraining. The mean overall effect size(Hedges’ g) for postintervention val-ues was 0.32 (95% CI��0.33, 0.98,P�.337). The results of these 2 sub-group analyses do not provide con-vincing evidence to make conclusivedecisions on favoring a specificintervention over another. No signif-icant heterogeneity was observed(Q�1.32, P�.250, I2�24.33%).

Methodological quality. Thequality of the included studies variedwidely according to the PEDro ratingscale, with scores ranging from 4positive items (out of 10 items)50 to 8positive items (out of 10 items).49

Twelve studies described the proce-dure of the randomization: 6 usedshuffled sealed envelopes,20,41,47–49 4made use of a computer-generatedrandomization list,23,39,40,43 1 used adrawing lot,42 1 did the randomiza-tion manually,38 and 3 did notexplain the randomization proce-dure in detail.44,46,50 Unfortunately,the corresponding authors did notreact on the request to describe the

randomization procedure. Six stud-ies23,42,45,47–49 reported that ablinded assessor was involved for alloutcome measurements.

Funnel plots are a way of detectingwhether selective publication ofsmall-scale studies with positiveresults and nonpublication of small-scale studies with negative resultsinfluence the outcome of a meta-analysis (Fig. 3). If all points areevenly distributed on both sides ofthe solid vertical line, it indicatesno publication bias. One outlier onthe left side could be detected.48

This finding was considered through-out the statistical analysis. The exclu-sion decreased the heterogeneity,although the significance of exercisetherapy was still confirmed.

Not only visual inspection of the fun-nel plot can help to interpret theresults but also mathematical mea-sures. The “fail-safe N” statistic canbe calculated to define the numberof new unpublished or unretrievednonsignificant or “null result” studiesthat would be required to change theoverall estimate value toward a non-significant value.37 Calculations haveindicated that it would take 89 unpub-lished null result studies to decreasethe overall estimate toward 0.

DiscussionThis systematic review and meta-analysis of 15 studies totaling 748participants (females: n�539; males:n�209) assessed the available evi-dence for the use of exercise therapyin the treatment of PFPS. Based onthe results of the present study, exer-cise therapy appears to be an impor-tant strategy to achieve pain andPRMALP relief in patients with PFPS.

A few reasons could explain the con-vincing results observed by Khayam-bashi et al38 and Herrington andAl-Sherhi.48 As a result of the inter-vention of Khayambashi et al,38 par-ticipants improved in hip abduction




and external rotation strength.Excessive hip adduction and internalrotation have been postulated toinfluence patellofemoral joint kine-matics.13 Therefore, alterations inmuscular hip strength might haveresulted in a decrease in patellofemo-ral joint loading and in pain. In addi-tion, the homogeneity of the studysample, with inclusion of female par-ticipants only, could have led to thestrong pain-relieving and PRMALP-reducing effects, as females are morelikely to have stronger internal rota-tor muscles and weaker externalrotator muscles compared withmales, in whom this relationship isinverse. Nevertheless, the study sup-ports the importance of hip strength-ening as a viable intervention to treatPFPS and, therefore, can be animportant factor in the rehabilitationprogram. Possible reasons for thestrong results achieved by Her-rington and Al-Sherhi48 may be theshort follow-up time and the inclu-sion of male participants. Further-more, it has to be noted that pain forat least 1 month was one of the inclu-sion criteria in this study, whereas inmany of the other studies, partici-pants fulfilled the inclusion criteriaif the symptoms of anterior kneepain occurred for at least2 months.23,38,41–44,47,50,51

Closed kinetic chain exercises aremore associated with activities ofdaily living than open kinetic chainexercises. Two studies20,40 demon-strated a significant decrease in painin patients treated with closedkinetic chain exercises (as definedby the authors) as well as openkinetic chain exercises, with a ten-dency toward few significantly bet-ter functional results in the grouptreated with closed kinetic chainexercises. The overall estimate(Hedges’ g�0.08) was not significantand, therefore, did not support theevidence of either closed kineticchain or open kinetic chain exer-cises. Furthermore, the analysis

included 2 studies with a sample sizeof 92 participants.20,40 No valid con-clusion can be drawn from theresults and the study limitations.

Balci et al44 confirmed the pain-reducing effects of closed kineticchain exercises in patients with PFPSindependent of internal or externalhip rotation. Dolak et al39 and Naka-gawa et al45 confirmed the impor-tance of functional training in studiescomparing knee exercise and otherforms of exercise. Pain reductionwas achieved in both exercisegroups. However, the groups treatedwith supplemented hip and trunkstrength training achieved strongerpain-reducing results. A possiblereason may be improved lowerextremity joint alignment leading todecreased patellofemoral jointpressure.

For better clinical understandingthe Hedges’ g, effect sizes werere-expressed in terms of VAS painscore reduction (ranging from �10to �40 mm). Several studies havebeen published on VAS score reduc-tion posttreatment and the clinicaleffectiveness of a treatment interven-tion. Grilo et al52 stated that the min-imum clinically significant difference

in pain intensity is achieved by a20-mm decrease in VAS pain score inpatients with acute rheumatic condi-tions. Todd et al53 found that a VASscore reduction of 13 mm was per-ceived as clinically significant inpatients with acute pain from atrauma. A clinically relevant 10-mmVAS score reduction was found forthe long-term follow-up studies inthe present meta-analysis. Theabsence of statistical significancecould be a result of the low numberof studies and study participants.Therefore, the observed effects jus-tify the conclusion that exercisetherapy can have a short-term andlong-term pain- and PRMALP-relieving effect in patients withPFPS.

Reduction in pain can be observed incontrol groups, too. There are differ-ent arguments that could explainthis phenomenon. The observedimprovement in the studies by Clarket al23 and van Linschoten et al43

could have been due to the naturalcourse of the disease because thefollow-up was done after 52 weeks.Another possible explanation couldrelate to the Hawthorne effect. TheHawthorne effect refers to a phe-nomenon in which human behavior

Figure 3.Funnel plot of included studies.




or performance is altered as a resultof being part of an experiment orstudy.54

The results of the studies by Biley etal,41 Yip and Ng,42 and Schneider etal50 demonstrate that exercise ther-apy with an additive therapy (EMS orEMG biofeedback) provides no addi-tional benefit for patients with PFPS,which is important in the context ofallocation of resources.

Although patellofemoral pain syn-drome is one of the most prevalentmusculoskeletal injuries encoun-tered in an active population, aunique definition of the syndrome isstill not available. The exact patho-mechanism is an unsolved issue aswell. However, various theories con-cerning the etiology have beenproposed.

Because of the complex etiology ofPFPS in humans, the “universal PFPSpatient” does not exist. Therefore, itis difficult to find consensus con-cerning the most appropriate conser-vative treatment. Although exercisetherapy as a PFPS treatment is widelyaccepted, a comprehensive treat-ment program should adapt to theindividual needs of each patient.

Quality AssessmentThe PEDro summed item score waschosen to assess the quality of theRCTs included in this meta-analysisbecause of its documented reliabil-ity55 and its widespread use in phys-ical therapy. To analyze a possibleeffect of study quality on the overallestimate, a meta-regression was per-formed. Individual studies’ effectsizes (Hedges’ g) were regressedover their study quality (PEDroscore), indicating an increasingeffect size of exercise therapy andincreasing study quality (B�0.237,P�.0009). Details of the randomiza-tion procedure were missing in 3studies with a total of 140 partici-pants.44,46,50 The corresponding

authors did not answer our requestsfor more information. Sensitivityanalysis excluding the 3 studies didnot influence the overall effect size.However, the PEDro scale does notinclude field-specific items. A moreresearch-specific tool (eg, GRADE)56

or a component approach to analyz-ing the risk of bias36 could have influ-enced the overall effect estimate.57

Several authors discussed possibleexplanations for the effectiveness ofexercise programs in relieving painin patients with PFPS. Most studiesassociated the increase in functionalmuscle strength, altered sensorimo-tor behavior, and the restoration ofpatella alignment with a reduction ofstress on the patellofemoral joint.The improvements of motor controlmotion and patellofemoral joint per-formance appear to be importantfactors in the management of patell-ofemoral pain syndrome. Althoughthere is available evidence for theeffectiveness of exercise programs inthe treatment of patients with PFPS,there is general agreement that nosingle intervention is superior to oth-ers and no particular program iseffective for all patients.

The main aim of this review was toassess the pain-reducing effect ofexercise therapy in patients withpatellofemoral pain. Among otherreasons, we chose pain and PRMALPas outcome measures because theycorrelate with human well-being aswell as with the adaptation in dailylife.

Besides the strengths of this study,2 major limitations should be dis-cussed. First, a language bias maybe present because of the restrictionto publications in German and Eng-lish languages. Second, publicationbias and citation bias cannot beexcluded.

In conclusion, this meta-analysis of15 RCTs provides evidence that

exercise therapy favors reduction ofpain and PRMALP in patients withPFPS. Studies comparing exerciseintervention with no intervention aswell as studies comparing exerciseintervention with exercise com-bined with an additive intervention(EMG or knee splint) resulted in clin-ically important effects. Further-more, the results were independentof time. Short-term (�12 weeks) aswell as long-term (�26 weeks)follow-up periods have confirmedthe evidence in support of exercisetherapy. However, there is notenough evidence to prefer one spe-cific exercise intervention overanother.

Dr Taeymans provided concept/idea/re-search design. All authors provided writing.Dr Clijsen and Dr Taeymans provided datacollection and analysis. Dr Clijsen providedconsultation (including review of manuscriptbefore submission). The authors expresstheir gratitude to Mayer Katharina for PEDrorating of those studies not listed in the PEDrodatabase.

DOI: 10.2522/ptj.20130310

References1 Kannus P, Natri A, Paakkala T, Jarvinen M.

An outcome study of chronic patellofemo-ral pain syndrome: seven-year follow-up ofpatients in a randomized, controlled trial.J Bone Joint Surg Am. 1999;81:355–363.

2 Boling M, Padua D, Marshall S, et al. Gen-der differences in the incidence and prev-alence of patellofemoral pain syndrome.Scand J Med Sci Sports. 2010;20:725–730.

3 Lindberg U, Lysholm J, Gillquist J. The cor-relation between arthroscopic findingsand the patellofemoral pain syndrome.Arthroscopy. 1986;2:103–107.

4 Witvrouw E, Lysens R, Bellemans J, et al.Intrinsic risk factors for the developmentof anterior knee pain in an athletic popu-lation: a two-year prospective study. Am JSports Med. 2000;28:480–489.

5 Roush JR, Curtis Bay R. Prevalence of ante-rior knee pain in 18–35 year-old females.Int J Sports Phys Ther. 2012;7:396–401.

6 Tenforde AS, Sayres LC, McCurdy ML,et al. Overuse injuries in high school run-ners: lifetime prevalence and preventionstrategies. PM&R. 2011;3:125–131.

7 Thomee R. A comprehensive treatmentapproach for patellofemoral pain syn-drome in young women. Phys Ther. 1997;77:1690–1703.




8 Fredericson M, Powers CM. Practical man-agement of patellofemoral pain. ClinJ Sport Med. 2002;12:36–38.

9 Cutbill JW, Ladly KO, Bray RC, et al. Ante-rior knee pain: a review. Clin J Sport Med.1997;7:40–45.

10 Nissen CW, Cullen MC, Hewett TE, NoyesFR. Physical and arthroscopic examinationtechniques of the patellofemoral joint.J Orthop Sports Phys Ther. 1998;28:277–285.

11 Powers CM. Rehabilitation of patellofemo-ral joint disorders: a critical review.J Orthop Sports Phys Ther. 1998;28:345–354.

12 Shelton GL, Thigpen LK. Rehabilitation ofpatellofemoral dysfunction: a review of lit-erature. J Orthop Sports Phys Ther. 1991;14:243–249.

13 Powers CM. The influence of alteredlower-extremity kinematics on patell-ofemoral joint dysfunction: a theoreticalperspective. J Orthop Sports Phys Ther.2003;33:639–646.

14 Tiberio D. The effect of excessive subtalarjoint pronation on patellofemoral mechan-ics: a theoretical model. J Orthop SportsPhys Ther. 1987;9:160–165.

15 Witvrouw E, Werner S, Mikkelsen C, et al.Clinical classification of patellofemoralpain syndrome: guidelines for non-operative treatment. Knee Surg SportsTraumatol Arthrosc. 2005;13:122–130.

16 Willson JD, Davis IS. Lower extremitymechanics of females with and withoutpatellofemoral pain across activities withprogressively greater task demands. ClinBiomech (Bristol, Avon). 2008;23:203–211.

17 Souza RB, Powers CM. Predictors of hipinternal rotation during running: an evalu-ation of hip strength and femoral structurein women with and without patellofemo-ral pain. Am J Sports Med. 2009;37:579–587.

18 Souza RB, Powers CM. Differences in hipkinematics, muscle strength, and muscleactivation between subjects with andwithout patellofemoral pain. J OrthopSports Phys Ther. 2009;39:12–19.

19 Thomee R, Augustsson J, Karlsson J. Patel-lofemoral pain syndrome: a review of cur-rent issues. Sports Med. 1999;28:245–262.

20 Witvrouw E, Lysens R, Bellemans J, et al.Open versus closed kinetic chain exer-cises for patellofemoral pain: a prospec-tive, randomized study. Am J Sports Med.2000;28:687–694.

21 Holmes SW Jr, Clancy WG Jr. Clinical clas-sification of patellofemoral pain and dys-function. J Orthop Sports Phys Ther. 1998;28:299–306.

22 Sanchis-Alfonso V. Patellofemoral pain[article in German]. Orthopade. 2008;37:835–836, 838–840.

23 Clark DI, Downing N, Mitchell J, et al.Physiotherapy for anterior knee pain: arandomised controlled trial. Ann RheumDis. 2000;59:700–704.

24 Naslund J, Naslund UB, Odenbring S, Lun-deberg T. Comparison of symptoms andclinical findings in subgroups of individu-als with patellofemoral pain. PhysiotherTheory Pract. 2006;22:105–118.

25 Dye SF. The pathophysiology of patell-ofemoral pain: a tissue homeostasis per-spective. Clin Orthop Relat Res. 2005;436:100–110.

26 Patellofemoral semantics: the Tower ofBabel. The International PatellofemoralStudy Group. Am J Knee Surg. 1997;10:92–95.

27 Juhn MS. Patellofemoral pain syndrome: areview and guidelines for treatment [erra-tum in: Am Fam Physician. 2000;61:960,965]. Am Fam Physician. 1999;60:2012–2022.

28 Wilk KE. Challenging tradition in the treat-ment of patellofemoral disorders. J OrthopSports Phys Ther. 1998;28:275–276.

29 McConnell J. The physical therapist’sapproach to patellofemoral disorders. ClinSports Med. 2002;21:363–387.

30 Arroll B, Ellis-Pegler E, Edwards A, SutcliffeG. Patellofemoral pain syndrome: a criticalreview of the clinical trials on nonopera-tive therapy. Am J Sports Med. 1997;25:207–212.

31 Boling MC, Bolgla LA, Mattacola CG, et al.Outcomes of a weight-bearing rehabilita-tion program for patients diagnosed withpatellofemoral pain syndrome. Arch PhysMed Rehabil. 2006;87:1428–1435.

32 Crossley K, Bennell K, Green S, et al. Phys-ical therapy for patellofemoral pain: a ran-domized, double-blinded, placebo-controlled trial. Am J Sports Med. 2002;30:857–865.

33 Stucki G, Cieza A, Ewert T, et al. Applica-tion of the International Classification ofFunctioning, Disability and Health (ICF)in clinical practice. Disabil Rehabil. 2002;24:281–282.

34 Knobloch K, Yoon U, Vogt PM. PreferredReporting Items for Systematic Reviewsand Meta-analyses (PRISMA) statement andpublication bias. J CraniomaxillofacSurg. 2011;39:91–92.

35 Physiotherapy Evidence Database (PEDro).Available at: http://www.pedro.org.au.

36 Cochrane Handbook for SystematicReviews of Interventions. The CochraneCollaboration. 2012. Available at: http://www.cochrane.org/handbook.

37 Sterne JA, Egger M, Smith GD. Systematicreviews in health care: investigating anddealing with publication and other biasesin meta-analysis. BMJ. 2001;323:101–105.

38 Khayambashi K, Mohammadkhani Z,Ghaznavi K, et al. The effects of isolatedhip abductor and external rotator musclestrengthening on pain, health status, andhip strength in females with patellofemo-ral pain: a randomized controlled trial.J Orthop Sports Phys Ther. 2012;42:22–29.

39 Dolak KL, Silkman C, Medina McKeon J,et al. Hip strengthening prior to functionalexercises reduces pain sooner than quad-riceps strengthening in females with patel-lofemoral pain syndrome: a randomizedclinical trial [erratum in: J Orthop SportsPhys Ther. 2011;41:700]. J Orthop SportsPhys Ther. 2011;41:560–570.

40 Bakhtiary AH, Fatemi E. Open versusclosed kinetic chain exercises for patellarchondromalacia. Br J Sports Med. 2008;42:99–102.

41 Bily W, Trimmel L, Modlin M, et al. Train-ing program and additional electric mus-cle stimulation for patellofemoral pain syn-drome: a pilot study. Arch Phys MedRehabil. 2008;89:1230–1236.

42 Yip SL, Ng GY. Biofeedback supplementa-tion to physiotherapy exercise pro-gramme for rehabilitation of patellofemo-ral pain syndrome: a randomizedcontrolled pilot study. Clin Rehabil. 2006;20:1050–1057.

43 van Linschoten R, van Middelkoop M,Berger MY, et al. Supervised exercise ther-apy versus usual care for patellofemoralpain syndrome: an open label randomisedcontrolled trial. BMJ. 2009;339:b4074.

44 Balci P, Tunay VB, Baltaci G, Atay AO. Theeffects of two different closed kineticchain exercises on muscle strength andproprioception in patients with patell-ofemoral pain syndrome [article in Turk-ish]. Acta Orthop Traumatol Turc. 2009;43:419–425.

45 Nakagawa TH, Muniz TB, Baldon Rde M,et al. The effect of additional strengthen-ing of hip abductor and lateral rotator mus-cles in patellofemoral pain syndrome: arandomized controlled pilot study. ClinRehabil. 2008;22:1051–1060.

46 Dursun N, Dursun E, Kilic Z. Electromyo-graphic biofeedback-controlled exerciseversus conservative care for patellofemo-ral pain syndrome. Arch Phys Med Reha-bil. 2001;82:1692–1695.

47 Fukuda TY, Rossetto FM, Magalhaes E,et al. Short-term effects of hip abductorsand lateral rotators strengthening infemales with patellofemoral pain syn-drome: a randomized controlled clinicaltrial. J Orthop Sports Phys Ther. 2010;40:736–742.

48 Herrington L, Al-Sherhi A. A controlledtrial of weight-bearing versus non-weight-bearing exercises for patellofemoral pain.J Orthop Sports Phys Ther. 2007;37:155–160.

49 Song CY, Lin YF, Wei TC, et al. Surplusvalue of hip adduction in leg-press exer-cise in patients with patellofemoral painsyndrome: a randomized controlled trial.Phys Ther. 2009;89:409–418.

50 Schneider F, Labs K, Wagner S. Chronicpatellofemoral pain syndrome: alterna-tives for cases of therapy resistance. KneeSurg Sports Traumatol Arthrosc. 2001;9:290–295.

51 Kettunen JA, Harilainen A, Sandelin J, et al.Knee arthroscopy and exercise versusexercise only for chronic patellofemoralpain syndrome: a randomized controlledtrial. BMC Med. 2007;5:38.




52 Grilo RM, Treves R, Preux PM, et al. Clin-ically relevant VAS pain score change inpatients with acute rheumatic conditions.Joint Bone Spine. 2007;74:358–361.

53 Todd KH, Funk KG, Funk JP, Bonacci R.Clinical significance of reported changesin pain severity. Ann Emerg Med. 1996;27:485–489.

54 Adair J. The Hawthorne effect: a reconsid-eration of the methodological artifact.J Appl Psychol. 1984;69:334–345.

55 Tooth L, Bennett S, McCluskey A, et al.Appraising the quality of randomized con-trolled trials: inter-rater reliability for theOTseeker evidence database. J Eval ClinPract. 2005;11:547–555.

56 Guyatt GH, Oxman AD, Vist G, et al.GRADE guidelines: 4; rating the quality ofevidence—study limitations (risk of bias).J Clin Epidemiol. 2011;64:407–415.

57 Juni P, Witschi A, Bloch R, Egger M. Thehazards of scoring the quality of clinicaltrials for meta-analysis. JAMA. 1999;282:1054–1060.




doi: 10.2522/ptj.20130310Originally published online July 31, 2014

2014; 94:1697-1708.PHYS THER. Ron Clijsen, Janine Fuchs and Jan TaeymansSystematic Review and Meta-AnalysisPatients With Patellofemoral Pain Syndrome: Effectiveness of Exercise Therapy in Treatment of

References

http://ptjournal.apta.org/content/94/12/1697#BIBLfor free at: This article cites 55 articles, 14 of which you can access

Information Subscription http://ptjournal.apta.org/subscriptions/

Permissions and Reprints http://ptjournal.apta.org/site/misc/terms.xhtml

Information for Authors http://ptjournal.apta.org/site/misc/ifora.xhtml

by Dienst E-bronnen on December 8, 2014http://ptjournal.apta.org/Downloaded from
http://ptjournal.apta.org/content/94/12/1697#BIBLhttp://ptjournal.apta.org/subscriptions/http://ptjournal.apta.org/site/misc/terms.xhtmlhttp://ptjournal.apta.org/site/misc/ifora.xhtmlhttp://ptjournal.apta.org/

Effectiveness of Exercise Therapy in Treatment of Originally … 2014.pdf · 2020. 3. 14. ·...

Documents

Transcript of Effectiveness of Exercise Therapy in Treatment of Originally … 2014.pdf · 2020. 3. 14. ·...