Patellar taping and bracing for the treatment of chronic knee painA systematic review and meta-analysis.код для вставкиСкачать
Arthritis & Rheumatism (Arthritis Care & Research) Vol. 59, No. 1, January 15, 2008, pp 73– 83 DOI 10.1002/art.23242 © 2008, American College of Rheumatology ORIGINAL ARTICLE Patellar Taping and Bracing for the Treatment of Chronic Knee Pain: A Systematic Review and Meta-Analysis STUART J. WARDEN,1 RANA S. HINMAN,2 MARK A. WATSON, JR.,3 KEITH G. AVIN,3 ANDREA E. BIALOCERKOWSKI,2 AND KAY M. CROSSLEY2 Objective. To evaluate the evidence for patellar taping and bracing in the management of chronic knee pain. Methods. Randomized or quasi-randomized studies assessing patellar taping or bracing effects on chronic knee pain were sourced from 7 electronic databases (to November 2006), and assessed using the Physiotherapy Evidence Database scale. Weighted mean differences were determined, and pooled estimates of taping and bracing effects were obtained using random-effects models. Results. Of 16 eligible trials, 13 investigated patellar taping or bracing effects in individuals with anterior knee pain, and 3 investigated taping effects in individuals with knee osteoarthritis (OA). The methodologic quality of the taping studies was signiﬁcantly higher than the bracing studies (mean ⴞ SD 4.8 ⴞ 2.1 versus 2.8 ⴞ 0.8; P < 0.05). On a 100-mm scale, tape applied to exert a medially-directed force on the patella decreased chronic knee pain compared with no tape by 16.1 mm (95% conﬁdence interval [95% CI] ⴚ22.2, ⴚ10.0; P < 0.001) and sham tape by 10.9 mm (95% CI ⴚ18.4, ⴚ3.4; P < 0.001). For anterior knee pain and OA, medially-directed tape decreased pain compared with no tape by 14.7 mm (95% CI ⴚ22.8, ⴚ6.9; P < 0.001) and 20.1 mm (95% CI ⴚ26.0, ⴚ14.3; P < 0.001), respectively. There was disputable evidence from low-quality studies for patellar bracing beneﬁts. Conclusion. There was evidence that tape applied to exert a medially-directed force on the patella produces a clinically meaningful change in chronic knee pain. There was limited evidence to demonstrate the efﬁcacy of patellar bracing. These outcomes were limited by the presence of high heterogeneity between study outcomes and signiﬁcant publication bias. INTRODUCTION Chronic knee pain is a leading cause of disability, and accounts for a large proportion of visits to health professionals. The most frequent presentation in younger people (age ⬍50 years) is anterior knee pain, characterized by a gradual onset of poorly-localized pain under or around the 1 Stuart J. Warden, PT, PhD, FACSM: School of Health and Rehabilitation Sciences, Indiana University, Indianapolis, and Centre for Health, Exercise and Sports Medicine, School of Physiotherapy, The University of Melbourne, Melbourne, Victoria, Australia; 2Rana S. Hinman, PT, PhD, Andrea E. Bialocerkowski, PT, PhD, Kay M. Crossley, PT, PhD: The University of Melbourne, Melbourne, Victoria, Australia; 3 Mark A. Watson, Jr., DPT, Keith G. Avin, DPT: School of Health and Rehabilitation Sciences, Indiana University, Indianapolis. Address correspondence to Stuart J. Warden, PT, PhD, FACSM, Department of Physical Therapy, School of Health and Rehabilitation Sciences, Indiana University, Indianapolis, IN 46202. E-mail: firstname.lastname@example.org. Submitted for publication March 2, 2007; accepted in revised form June 11, 2007. patellofemoral joint. It is aggravated by activities including squatting, stair climbing and descent, and prolonged sitting with the knee bent (1). In contrast, the most common cause of knee pain in older individuals (age ⬎50 years) is osteoarthritis (OA). Knee OA is a chronic disease affecting articular cartilage and subchondral bone, typically affecting the patellofemoral and tibiofemoral joints concurrently (2). Pain associated with knee OA is insidious in onset, usually aggravated by movement or weight bearing, and relieved by rest. The mainstay of treatment for chronic knee pain is management of symptoms (3,4). Two frequently-used treatments are therapeutic taping and bracing of the patella. Taping involves applying adhesive, rigid, strapping tape to glide, tilt, and/or rotate the patella, whereas bracing involves applying an external, nonadhesive device that also aims to modify patella position (the latter is distinct from valgus and varus bracing used to treat tibiofemoral joint OA). Both treatments aim to reduce pain by increasing the patellofemoral contact area, thereby decreasing joint stress. They are ideal treatments for chronic knee pain as they are simple, inexpensive, and associated with negligi73 74 ble adverse effects. This type of treatment decreases the burden associated with treating chronic knee pain because patients can be taught to self-tape or self-brace, increasing their responsibility in management. Although they are clinically popular and recommended (3,5), the effectiveness of patellar taping and bracing in the management of chronic knee pain is still debated (4,6,7). It is not currently known whether patellar taping and bracing effects differ: 1) with the direction of force applied to the patella, 2) between anterior knee pain and OA, 3) from sham effects, or 4) between the 2 respective interventions. This study addresses these questions. Studies of anterior knee pain and knee OA were both included because they are the most common causes of chronic knee pain treated by patellar taping and bracing, and because patellar taping and bracing are thought to produce similar effects across these 2 diagnoses. MATERIALS AND METHODS Inclusion and exclusion criteria. The protocol was established a priori. Randomized or quasi-randomized studies assessing patellar taping or bracing effects on chronic knee pain were sourced. There were no restrictions on the cause, severity, or history of knee pain, nor on the underlying pathology. There were no restrictions on followup duration; studies reporting both immediate (same-day) and longer-term effects of patellar taping and bracing were included. Studies evaluating combined effects of taping or bracing with a concurrent intervention were included if the isolated effects of taping or bracing could be elucidated. Dissertations, conference proceedings, and studies in non-English languages were excluded. Experts in the ﬁeld were not contacted to obtain possible unpublished studies, as these may be subject to bias. Search strategy. Relevant studies were identiﬁed by sequentially searching the following databases: Medline (1980 to November 2006); the Cumulative Index to Nursing and Allied Health Literature (CINAHL) (1982 to November 2006); Evidence Based Medicine (EBM) Reviews (to November 2006); the ISI Web of Knowledge CrossSearch (1980 to November 2006); SPORTDiscus (1980 to November 2006); Expanded Academic ASAP (1980 to November 2006); and the Physiotherapy Evidence Database (PEDro) (to November 2006). Our search dates began at 1980, or as soon as possible thereafter if the database was launched after 1980, because patellar taping and bracing for chronic knee pain were not used prior to this date. For all databases except PEDro, the following search strategy was used with database-speciﬁc truncation terms: (knee$ OR patell$) AND pain AND (tape$ OR taping OR brace$ OR bracing). PEDro was searched with the terms “patell* AND tap*”, “patell* AND brac*”, “knee* AND tap*”, and “knee* AND brac*” in the abstract and title ﬁeld. Bibliographies of potentially eligible studies were searched recursively until no other potentially eligible studies were identiﬁed. Warden et al Inclusion determination. Eligibility was assessed by 2 independent reviewers (SJW and KMC), with consensus reached by discussing conﬂicts with a third investigator (RSH). Assessments were performed in 2 stages. First, titles and abstracts were assessed, then potentially eligible studies and studies whose titles and abstracts provided insufﬁcient information to determine inclusion suitability were obtained in full text and assessed. Assessment of study quality. Studies were blindly evaluated for methodologic quality by 2 independent reviewers (MAW and KGA), with disagreements resolved by a third assessor (SJW). Quality was assessed using the PEDro scale, which is an 11-item checklist where 1 point is awarded for each of the following satisﬁed items (except for the ﬁrst item, which pertains to external validity and is not counted towards the ﬁnal score): 1) eligibility criteria were speciﬁed, 2) subjects were randomly allocated to groups or treatment order, 3) allocation was concealed, 4) groups were similar at baseline, 5) subjects were blinded, 6) therapists who administered the treatment were blinded, 7) assessors were blinded, 8) measures of key outcomes were obtained from more than 85% of subjects, 9) data were analyzed by intention to treat, 10) statistical comparisons were conducted between groups, and 11) point measures and measures of variability were provided (8). Only studies scoring ⱖ3 points out of 10 were considered to be of sufﬁcient quality to warrant data extraction (9,10). Data extraction. Study design, number of participants, participant characteristics and diagnoses, interventions investigated, and method of pain assessments were abstracted from each study by a single reviewer (SJW). The same reviewer abstracted in duplicate the quantitative pain outcome data. When a study provided data from more than 1 pain scale, the data used for analysis were selected according to the hierarchy of pain-related outcomes in OA (11,12). In this hierarchy, global pain (measured on a visual analog scale [VAS] or Likert scale) takes precedence over pain during walking (VAS or Likert scale), which in turn takes precedence over the Western Ontario and McMaster Universities Osteoarthritis Index pain subscale and pain during activities other than walking (VAS or Likert scale). For crossover studies, it was assumed that subjects completed all arms of the study as allocated. This is a valid assumption because typically these studies involved within-day testing between groups, thus allowing for followup and maintenance of group allocation. Similarly, it was assumed in crossover study designs that there were minimal carry-over effects for successive interventions, being that cutaneous sensibility following tape removal returns to baseline within 5 minutes (13). Included studies used longer periods than this between successive interventions and, consequently, data from all periods of eligible crossover trials were included in our analyses. Data synthesis. Pain scores were converted to percentages of the maximum possible score and reported as millimeters on a 100-mm analog scale. The normalization of Effect of Patellar Taping and Bracing on Chronic Knee Pain Figure 1. Flow of studies through selection process. PFJ ⫽ patellofemoral joint. data to a 100-point scale is a frequent approach in metaanalyses of intervention effects on pain (10,14 –16). The mean difference and SEM difference between comparison interventions were determined. As crossover studies typically did not report variation of paired differences, the standard errors of their difference scores were calculated assuming a conservative correlation of 0.5 in order to ensure appropriate weighting in analyses (17). Effect sizes for comparative interventions were derived by dividing the mean differences by the pooled SD (18). Data were entered in Cochrane Collaboration’s Review Manager program (RevMan version 4.2, Cochrane Collaboration, Oxford, UK), and results of comparable studies were pooled in meta-analyses. Weighted mean differences between interventions and 95% conﬁdence intervals (95% CIs) expressed as millimeters on a 100-mm analog scale were determined, weighted by the inverse of the variance for each study. Weighted mean differences on a 100-mm analog scale were used for the primary analysis because of the relative uniformity of the pain measures across studies, and to facilitate clinical interpretation. Weighted mean differences between interventions and 95% CIs for unitless effect sizes were also determined, weighted by the inverse of the variance for each study. Pooled estimates of the effects of patellar taping and bracing were obtained using random-effects models in order to minimize the inﬂuences of heterogeneity (17). The I2 quantity was used to test heterogeneity between trials in each analysis (19). Moderate-to-high heterogeneity (I2 ⱖ 50%) was explored using sensitivity analyses. Publication bias was determined by assessing funnel plot asymmetry using a linear regression approach (20,21). RESULTS Search results. Searching retrieved 462 unique articles, of which 16 fuﬁlled the inclusion criteria (Figure 1) (11,22–36). These 16 articles were retrieved from the ﬁrst 3 databases searched (Medline [n ⫽ 12], CINAHL [n ⫽ 1], EBM Reviews [n ⫽ 3]). No additional eligible studies were retrieved from succeeding electronic database or bibliography searches. The 16 eligible studies are described in Table 1. Seven of the studies were randomized controlled trials and 9 were randomized crossover trials. Thirteen studies investigated patellar taping or bracing effects in anterior knee pain, and 3 investigated taping effects in knee OA. Diagnoses were 75 primarily made by physicians, orthopedists, rheumatologists, and physical therapists. Consistent with clinical practice, diagnosis of anterior knee pain was always based on clinical examination, whereas knee OA diagnosis was determined both clinically and radiologically according to the American College of Rheumatology criteria (37). Criteria for the clinical diagnosis of anterior knee pain predominantly included an insidious onset of retropatellar or peripatellar pain aggravated by palpation of the patellar facets (28,31), external compression of the patellofemoral joint (27,28,33), and/or activities that stress the patellofemoral joint (i.e., stair ascent and descent, running, and prolonged sitting with the knee bent) (22,25,28,31,34 –36). The age of participants in the studies reﬂected clinical presentations, with studies on anterior knee pain and knee OA including participants who were ⬍50 and ⬎50 years of age, respectively. Most studies (11 [69%] of 16) investigated independent taping or bracing effects (13,22,24 –29,33,34,36). The remaining studies (5 [31%] of 16) investigated taping or bracing effects with ⱖ1 cointerventions, including exercise (23,30,31,35) and other physical therapy treatments (23,32). Only 1 study directly compared patellar taping effects with bracing effects (24). All taping studies included a group in which tape with a medially-directed force was applied to the patella (medially-directed tape). Tape was used to exert a medial glide and/or tilt force on the patella, with or without an anterior tilt or rotation force, and with or without tape to unload the infrapatellar fat pad. Comparative groups in taping studies included groups in which tape was applied without exerting appreciable force on the patella (sham tape); tape with a laterally-directed force was applied to the patella (laterally-directed tape); or no tape was applied (no tape). All bracing studies included a group in which the brace was designed to generate a medially-directed force on the patella (medially-directed brace). Comparative groups included groups in which either an elastic knee sleeve or a simple infrapatellar knee strap was used (sham brace) or no brace was applied (no brace). Methodologic quality of included studies. The 2 reviewers scored 176 quality criteria and agreed on 138 (78%). The intraclass correlation coefﬁcient (2,1) for the total PEDro score for each study was 0.73. All disagreements were resolved by discussion with the third assessor. The methodologic quality of the studies was moderate, ranging from 2– 8 points out of 10 (mean ⫾ SD score 4.4 ⫾ 2.0) (Table 2). Taping studies scored signiﬁcantly higher than bracing studies (mean ⫾ SD score 4.8 ⫾ 2.1 versus 2.8 ⫾ 0.8, respectively; P ⬍ 0.05 by Mann-Whitney U test). Most studies (11 [69%] of 16) clearly deﬁned their selection criteria and the source of their participants. However, not all criteria on the PEDro scale could be satisﬁed for the chosen interventions and study designs. Although it is possible to achieve subject and assessor blinding when studying patellar taping and bracing effects, it is not possible to achieve therapist blinding, because treatments for these conditions require therapist skill and/or instruction. 76 Warden et al Table 1. Description of included studies* Study (ref.) Trial design Diagnosis† Participants Intervention groups Christou (22) Crossover AKP, by physician 15 female patients Mean ⫾ SD age 26.3 ⫾ 1.5 Symptom duration NR Medially-directed tape Laterally-directed tape Sham tape No tape Medially-directed tape, education, and exercise Medially-directed tape and education No tape; education and exercise No tape; education Medially-directed tape Medially-directed brace No tape or brace Clark et al (23) Controlled AKP, by orthopedist, rheumatologist, or GP 81 patients (45 men, 36 women) Age range 6–40 years Symptom duration ⱖ3 months Conway et al (24) Crossover AKP, by orthopedist and/or physiotherapist 30 Air Force cadets (21 men, 9 women) Mean age 20.1 years Symptom duration NR Cowan et al (25) Crossover AKP, by physical therapist Medially-directed tape Sham tape No tape Cushnaghan et al (26) Crossover OA, by rheumatologist Finestone et al (27) Controlled AKP, by orthopedist and physician 10 patients (3 men, 7 women) Mean age ⫾ SD 22.7 ⫾ 8.0 Symptom duration NR 14 patients (4 men, 10 women) Mean age (range) years 70.4 (55–84) Mean (range) symptom duration 8.3 (1–20) years 59 male infantry recruits (84 knees) Ages NR Symptom duration NR Handﬁeld and Kramer (28) Crossover AKP, by physical therapist 36 patients (10 men, 26 women) Mean age ⫾ SD 29 ⫾ 12 years Symptom duration NR Medially-directed tape Sham tape Hinman et al (29) Controlled OA, by physical therapist 87 patients (30 men, 57 women) Mean age ⫾ SD 69 ⫾ 8 Mean ⫾ SD symptom duration 9 ⫾ 10 years Medially-directed tape Sham tape No tape Pain outcome‡ Modiﬁed McGill analog (0–5) pain questionnaire during isokinetic leg press exercise VAS (2 10-cm horizontal lines with anchors of no pain and ‘extreme pain’) during climbing stairs and walking on ﬂat area after 12 weeks of intervention Category rating scale 1–8 (1 ⫽ no pain, 8 ⫽ excruciating pain) during maximal concentric and eccentric isokinetic quadriceps contractions VAS (10 cm) for average pain during stairstepping task Medially-directed tape Laterally-directed tape Sham tape VAS (10 cm) for overall pain after 4 days of intervention Medially-directed brace Sham-brace No brace Category rating scale 1–4 (1 ⫽ discomfort, 4 ⫽ activity-limiting pain) during infantry training VAS (10-cm horizontal line with anchors of no pain and worst pain ever experienced) during isokinetic knee extensor strength test at 60 degrees/second VAS (10-cm horizontal line numbered in 1-cm increments) for pain on movement after 3 weeks of intervention (continued) Effect of Patellar Taping and Bracing on Chronic Knee Pain 77 Table 1. Description of included studies* (Continued) Study (ref.) Trial design Diagnosis† Participants Hinman et al (13) Crossover OA, by physical therapist 18 patients (6 men, 12 women) Mean age ⫾ SD 66.9 ⫾ 6.5 years Symptom duration ⫽ knee pain on most days of previous month (average pain ⬎3/10) 25 patients (8 men, 17 women) Mean age (range) 29 (14–40) years Mean (range) symptom duration 2.5 (0.1–15) years 129 patients (53 men, 76 women) Mean age (range) 35 (18–60) years Symptom duration ⱖ3 weeks Kowall et al (30) Controlled AKP, by orthopedist and/or physical therapist Lun et al (31) Controlled AKP, by physician Miller et al (32) Controlled AKP, by physician 59 Air Force cadets (48 men, 11 women) Ages NR Symptom duration NR Ng and Cheng (33) Crossover AKP, by physician and physical therapist 15 patients (8 men, 7 women) Mean age ⫾ SD 32 ⫾ 6.6 Symptom duration NR Powers et al (34) Crossover AKP, by physical therapist 15 female patients Mean age ⫾ SD 31.1 ⫾ 7.5 Symptom duration NR Because more than half of the studies (9 [56%] of 16) used a crossover (within-subject) study design, these studies were not able to satisfy concealed allocation because it was known a priori that all eligible participants would be exposed to each intervention. Similarly, crossover study designs could not meet the criterion of baseline comparability because the repeated testing of participants in these studies theoretically inﬂuenced pain outcomes for succeeding interventions. We assumed minimal carry-over Intervention groups Pain outcome‡ Medially-directed tape Sham tape No tape VAS (10-cm horizontal line) for pain during walking Medially-directed tape and physical therapy No tape; physical therapy VAS (10-point scale) for pain during activity after 4 weeks of intervention Medially-directed brace Medially-directed brace and exercise Sham brace and exercise No brace; exercise Medially-directed brace and physical therapy Sham brace and physical therapy No brace; physical therapy Medially-directed tape No tape VAS (10 cm) for pain during sport activity after 12 weeks of intervention Medially-directed brace 1 Medially-directed brace 2 No brace VAS (horizontal line with anchors of no pain and worst pain in life) for pain with activity after 2–3 weeks intervention VAS (10-cm horizontal line with anchors of no pain and maximum pain) immediately following singleleg stand for 5 seconds with additional 20% body weight added VAS (10-point scale) for pain during aggravating activity (unilateral squat or deep knee bend) (continued) effects for successive interventions in crossover studies (see data extraction in Materials and Methods); however, their potential presence negated the ability of crossover studies to meet the PEDro criterion of baseline comparability. Many crossover studies also did not satisfy the criteria for adequate followup (6 [67%] of 9) or intent-totreat (8 [89%] of 9). It is expected that these criteria were met; study periods were conducive with both followup and the maintenance of group allocation because same-day 78 Warden et al Table 1. Description of included studies* (Continued) Study (ref.) Trial design Diagnosis† Participants Intervention groups Pain outcome‡ VAS (10-cm horizontal line with anchors of no pain and worst pain possible) for average pain over previous 24 hours after 3 weeks of intervention 11-point numerical pain-rating scale (0 ⫽ no pain, 10 ⫽ worst pain imaginable) for pain during single step-down Whittingham et al (35) Controlled AKP, by physical therapist 30 Army recruits (24 men, 6 women) Mean age ⫾ SD 18.7 ⫾ 1.2 Symptom duration NR Medially-directed tape and exercise Sham tape and exercise No tape; exercise Wilson et al (36) Crossover AKP, by physical therapist, hospital consultant, or physician 71 patients (39 men, 32 women) Mean age ⫾ SD 33.8 ⫾ 10.2 Symptom duration ⱖ1 month Medially-directed tape Sham tape No tape * AKP ⫽ anterior knee pain; NR ⫽ not reported; GP ⫽ general practitioner; VAS ⫽ visual analog scale; OA ⫽ osteoarthritis. † Diagnosis of AKP was determined clinically, whereas diagnosis of OA was determined clinically and radiologically according to the American College of Rheumatology criteria (37). ‡ Pain outcome used for data extraction and meta-analyses in the current review. testing was employed. However, the reports typically did not specify that ⬎85% of subjects completed the tasks nor that participants were analyzed according to their initial group allocation. Effect of patellar taping on chronic knee pain. One taping study (24) was of insufﬁcient quality to warrant data extraction, and another (30) did not provide a means of determining variability for its pain outcome measures. The remaining 10 studies investigated the immediate (same-day; 6 studies) and short-term (3–12 weeks; 4 studies) effects of patellar taping. As there was relative homogeneity between studies in terms of interventions and pain outcome measures (Table 1), similar studies were combined for meta-analyses. Medially-directed tape compared with no tape (8 studies, 288 participants) and sham tape (8 studies, 242 participants) decreased reported pain by 16.1 mm (95% CI ⫺22.2, ⫺10.0; P ⬍ 0.001; I2 ⫽ 79%) (Figure 2A) and 10.9 mm (95% CI ⫺18.4, ⫺3.4; P ⬍ 0.001; I2 ⫽ 87%) (Figure 2B), respectively. Sensitivity analyses revealed that 2 studies (23,36) with opposing ﬁndings to the pooled outcomes accounted for the statistical heterogeneities. Removal of these studies did not alter the outcomes of the analyses. Outcomes and statistical heterogeneities were not inﬂuenced by the inclusion of studies investigating taping effects with cointerventions, or by elevation of the PEDro score inclusion criterion to 5. However, there was a publication bias, as evident by signiﬁcant funnel plot asymmetries (all P ⬍ 0.02). For anterior knee pain, medially-directed tape compared with no tape (6 studies, 212 participants) decreased reported pain by 14.7 mm (95% CI ⫺22.8, ⫺6.9; P ⬍ 0.001) (Figure 2C). There was no difference between mediallydirected tape and sham tape in this population (5 studies, 152 participants), although effects did favor medially-di- rected tape (⫺9.1 mm, 95% CI ⫺19.9, 1.8; P ⫽ 0.10) (Figure 2D). For OA, medially-directed tape compared with no tape (2 studies, 76 participants) decreased reported pain by 20.1 mm (95% CI ⫺26.0, ⫺14.3; P ⬍ 0.001) (Figure 2C), and medially-directed tape compared with sham tape (3 studies, 90 participants) decreased reported pain by 13.3 mm (95% CI ⫺18.1, ⫺8.4; P ⬍ 0.001) (Figure 2D). Sham tape compared with no tape (6 studies, 192 participants) decreased reported pain by 10.4 mm (95% CI ⫺15.7, ⫺5.1; P ⬍ 0.001; I2 ⫽ 67%) (Figure 3A). For anterior knee pain (4 studies, 116 participants) and OA (2 studies, 76 participants), sham tape compared with no tape decreased reported pain by 12.0 mm (95% CI ⫺19.7, ⫺4.3; P ⬍ 0.01) and 7.2 mm (95% CI ⫺11.9, ⫺2.5; P ⬍ 0.01), respectively (Figure 3B). Laterally-directed tape did not differ from medially-directed tape (3 studies, 100 participants) or from sham tape (3 studies, 100 participants), with effects favoring medially-directed tape (6.9 mm, 95% CI ⫺10.2, 24.0; P ⫽ 0.43) (Figure 3C) and sham tape (5.5 mm, 95% CI ⫺1.6, 12.6; P ⫽ 0.13) (Figure 3D). No studies compared laterally-directed tape with no tape. Effect of patellar bracing on anterior knee pain. Two patellar bracing studies (24,27) were of insufﬁcient quality to warrant data extraction. The remaining 3 studies investigated immediate (same-day; 1 study) and short-term (3–12 weeks; 2 studies) effects of patellar bracing on anterior knee pain. No studies investigated patellar bracing effects on pain associated with knee OA. Medially-directed brace compared with no brace (3 studies, 119 participants) decreased reported pain associated with anterior knee pain by 14.6 mm (95% CI ⫺25.5, ⫺3.8; P ⬍ 0.01; I2 ⫽ 76%) (Figure 4A). Sensitivity analyses revealed that the differences between immediate and short-term effects accounted for the statistical heterogeneity. Medially-directed brace did not differ from sham brace (2 studies, 94 partic- Effect of Patellar Taping and Bracing on Chronic Knee Pain 79 Table 2. Quality scores for eligible studies* PEDro scale item† Author (ref.) Christou (22) Clark et al (23) Conway et al (24) Cowan et al (25) Cushnaghan et al (26) Finestone et al (27) Handﬁeld and Kramer (28) Hinman et al (29) Hinman et al (13) Kowall et al (30) Lun et al (31) Miller et al (32) Ng and Cheng (33) Powers et al (34) Whittingham et al (35) Wilson et al (36) 1 ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ 2 ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ 3 4 5 ⫹ 6 7 8 9 10 11 Total score‡ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ 5 7 2 3 4 2 4 3 8 4 3 4 3 3 8 7 ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ * PEDro ⫽ Physiotherapy Evidence Database; ⴙ ⴝ the item was clearly satisﬁed. † Column numbers correspond to the following items on the PEDro scale: 1, eligibility criteria were speciﬁed; 2, subjects were randomly allocated to groups or treatment order; 3, allocation was concealed; 4, groups were similar at baseline; 5, subjects were blinded; 6, therapists who administered the treatment were blinded; 7, assessors were blinded; 8, measures of key outcomes were obtained from more than 85% of subjects; 9, data were analyzed by intention to treat; 10, statistical comparisons were conducted between groups; and 11, point measures and measures of variability were provided. ‡ Total score (out of a possible 10) was determined by counting the number of items that were satisﬁed, except for the ﬁrst item, which pertains to external validity. ipants) (1.3 mm, 95% CI ⫺9.8, 7.2; P ⫽ 0.76; I2 ⫽ 0%) (Figure 4B). Similarly, sham brace did not differ from no brace (2 studies, 98 participants) (2.5 mm, 95% CI ⫺10.6, 5.7; P ⫽ 0.55; I2 ⫽ 0%) (Figure 4C). Only 1 eligible study (24) directly compared patellar taping and bracing effects; however, it was of insufﬁcient quality to warrant data extraction. DISCUSSION This systematic review and meta-analysis provides evidence for the beneﬁts of patellar taping, disputable evidence for the beneﬁts of patellar bracing, and no evidence of the comparative efﬁcacies of patellar taping and bracing in the management of chronic knee pain. Some acknowledged liberties in pooling data were taken in acquiring these outcomes, such as the inclusion of studies investigating multiple diagnoses (anterior knee pain and knee OA) and the inclusion of data from all periods of eligible crossover trials. This approach may have contributed to the elevated levels of heterogeneity observed in the outcomes, which, combined with the presence of signiﬁcant publication bias, indicates the need for additional highquality trials in this area of inquiry. However, the current review provides the most up-to-date and detailed analysis of the effects of patellar taping and bracing on chronic knee pain. Following its initial description in 1986 (38), patellar taping has become clinically accepted as an intervention for the management of chronic knee pain. This systematic review and meta-analysis provides evidence for this use. The results of 10 moderate-quality studies (mean ⫾ SD PEDro score 5.2 ⫾ 2.1) demonstrate that on a 100-mm scale, medially-directed patellar tape reduces chronic knee pain by 16 mm compared with no tape. This effect was irrespective of the time course of tape application; pain reductions were observed both immediately (sameday) following tape application and after repeated applications over the short term (3–12 weeks), with reductions in pain of 17 mm and 14 mm, respectively. Similarly, the reduction in pain with tape use was irrespective of diagnosis; medially-directed tape reduced pain associated with anterior knee pain and knee OA by 15 mm and 20 mm, respectively. Approximately 50% of the beneﬁt of medially-directed tape on chronic knee pain was explained by sham treatment effects. Tape that does not exert appreciable force on the patella (sham tape) reduced chronic knee pain by 10 mm compared with no tape. This effect was not inﬂuenced by the duration of tape application or diagnosis, and it suggests that sensory and/or placebo effects associated with tape are sufﬁcient to modify chronic knee pain. However, medially-directed tape reduced chronic knee pain by 11 mm compared with sham tape, irrespective of the duration of tape application. This indicates that while sham tape does beneﬁt chronic knee pain, greater beneﬁts are gained if tape exerts a medially-directed force on the patella. Medially-directed patellar tape is commonly applied clinically to treat chronic knee pain. This ﬁts with the accepted theory of patellar malalignment in both anterior knee pain and knee OA, where there is lateral displacement of the patella relative to the femoral trochlear groove, resulting in increased peak patellofemoral contact pressures and loading of the lateral facet (39,40). Reﬂecting the clinical popularity of medially-directed tape, all taping 80 Warden et al Figure 2. Effects of medially-directed patellar tape on chronic knee pain. Immediate and short-term effects of medially-directed tape compared with A, no tape and B, sham tape. Effects of medially-directed tape on anterior knee pain and knee osteoarthritis compared with C, no tape and D, sham tape. WMD ⫽ weighted mean difference; 95% CI ⫽ 95% conﬁdence interval. Figure 3. Effects of sham and laterally-directed patellar tape on chronic knee pain. A, Immediate and short-term effects of sham tape compared with no tape. B, Effects of sham tape on anterior knee pain and knee osteoarthritis compared with no tape. Effects of laterally-directed tape on anterior knee pain and knee osteoarthritis compared with C, medially-directed tape and D, sham tape. WMD ⫽ weighted mean difference; 95% CI ⫽ 95% conﬁdence interval. studies in the current review included a group in which medially-directed tape was used, whereas only 3 studies investigated laterally-directed tape. Due to the small number of studies investigating laterally-directed tape, no de- ﬁnitive conclusions can be made regarding its beneﬁts or its effects compared with medially-directed tape. The reduction in chronic knee pain with medially-directed tape is clinically signiﬁcant. Medially-directed tape Effect of Patellar Taping and Bracing on Chronic Knee Pain Figure 4. Effects of patellar bracing on anterior knee pain. Immediate and short-term effects of medially-directed bracing compared with A, no brace and B, sham brace. C, Short-term effects of sham bracing versus no bracing. WMD ⫽ weighted mean difference; 95% CI ⫽ 95% conﬁdence interval. reduced the pain associated with anterior knee pain and OA by 15 mm and 20 mm, respectively. These effect sizes compare favorably with the minimal clinically important differences of 15–20 mm and 17.5 mm required to detect treatment effects on pain in individuals with anterior knee pain and OA, respectively (41,42). Thus, medially-directed tape used in isolation may generate clinically important changes in chronic knee pain. However, these changes have only been investigated in a limited number of short-term studies (ⱕ12 weeks) with relatively small sample sizes, and the long-term effects of tape on chronic knee pain have not been established. Nevertheless, this short-term investigation ﬁts with the clinical use of tape as a temporary pain-relieving technique for the management of chronic knee pain. In contrast to the evidence for the beneﬁts of patellar tape, there was disputable evidence from 3 low-quality studies (mean ⫾ SD PEDro score 3.3 ⫾ 0.6) for the beneﬁts of patellar bracing. Bracing designed to generate a medially-directed force on the patella reduced anterior knee pain by 15 mm on a 100-mm scale compared with no brace. This outcome was attributable to the immediate effects reported for 2 braces in 1 study (34), with no differences being found in 2 studies (31,32) on the short-term effects of medially-directed bracing. These latter 2 studies also found no difference between medially-directed bracing and sham bracing. To our knowledge, no studies have assessed the immediate sham effects of bracing, or compared patellar bracing and taping effects in chronic knee pain. Therefore, there is a need for further well-designed studies on the effects of patellar bracing on chronic knee pain, particularly in knee OA, for which no studies are currently available. 81 Our conclusions differ from previous systematic reviews that have reported inconclusive evidence for patellar taping efﬁcacy in the management of chronic knee pain (43– 46). However, previous studies were not speciﬁc to patellar taping and bracing, did not perform meta-analyses, were limited to only anterior knee pain, did not include all eligible controlled studies, and were published prior to more recent controlled studies on the effects of patellar taping and bracing on chronic knee pain. We used a systematic methodology to eliminate potential sources of bias; however, our conclusions are inﬂuenced by publication bias, as is indicated by signiﬁcant funnel plot asymmetry. This asymmetry indicates that negative studies investigating patellar taping and bracing effects are less likely to be published, and that small studies are more likely to produce larger effect sizes. Potentially contributing to this publication bias was our noninclusion of dissertations and conference proceedings; however, these are infrequently subjected to extensive peer review, and there is no systematic and comprehensive method of searching for these sources. The presence of publication bias in the current review indicates the need for additional high-quality trials of patellar taping and bracing effects on chronic knee pain. In addition to publication bias, the outcomes of our analyses may have been inﬂuenced by the inclusion of studies reported only in English (21). However, our original search found only 3 studies published in non-English languages (47– 49). Based on their English abstracts, these studies would not have altered the conclusions reached in this review, as their ﬁndings support those of our pooled results. Our meta-analysis found that tape applied to exert a medially-directed force on the patella produces a clinically meaningful change in chronic knee pain resulting from either anterior knee pain or knee OA. This effect occurs immediately and persists at least short term (ⱕ12 weeks). Placebo and/or other effects (possibly including sensory effects) contribute to beneﬁcial tape effects, but these only explain approximately half of the reduction in pain associated with medially-directed tape. There is insufﬁcient evidence to make deﬁnitive conclusions regarding in which direction forces should be applied to the patella to optimally reduce pain (medially or laterally), or to establish the efﬁcacy of patellar bracing. Similarly, there is insufﬁcient evidence from quality studies directly comparing patellar taping and bracing effects to establish their comparative beneﬁts in the management of chronic knee pain. These conclusions need to be considered in light of signiﬁcant limitations, including low numbers of studies investigating relatively small numbers of participants, high heterogeneity between study outcomes, and publication bias, all of which indicate the need for additional high-quality studies of taping and bracing effects on chronic knee pain. AUTHOR CONTRIBUTIONS Dr. Warden had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study design. Warden, Hinman, Watson, Avin, Bialocerkowski, Crossley. 82 Acquisition of data. Warden, Watson, Avin. Analysis and interpretation of data. Warden, Hinman, Watson, Avin, Bialocerkowski, Crossley. Manuscript preparation. Warden, Hinman, Watson, Avin, Bialocerkowski, Crossley. Statistical analysis. Warden, Hinman, Bialocerkowski, Crossley. REFERENCES 1. Gerbino PG II, Grifﬁn ED, d’Hemecourt PA, Kim T, Kocher MS, Zurakowski D, et al. Patellofemoral pain syndrome: evaluation of location and intensity of pain. Clin J Pain 2006;22: 154 –9. 2. Duncan RC, Hay EM, Saklatvala J, Croft PR. Prevalence of radiographic osteoarthritis: it all depends on your point of view. Rheumatology (Oxford) 2006;45:757– 60. 3. American College of Rheumatology Subcommittee on Osteoarthritis Guidelines. Recommendations for the medical management of osteoarthritis of the hip and knee: 2000 update. Arthritis Rheum 2000;43:1905–15. 4. Dixit S, DiFiori JP, Burton M, Mines B. Management of patellofemoral pain syndrome. Am Fam Physician 2007;75:194 – 202. 5. Fulkerson JP. Diagnosis and treatment of patients with patellofemoral pain. Am J Sports Med 2002;30:447–56. 6. Crossley K, Cowan SM, Bennell KL, McConnell J. Patellar taping: is clinical success supported by scientiﬁc evidence? Man Ther 2000;5:142–50. 7. Jordan KM, Arden NK, Doherty M, Bannwarth B, Bijlsma JW, Dieppe P, et al, and the Standing Committee for International Clinical Studies Including Therapeutic Trials. EULAR Recommendations 2003: an evidence based approach to the management of knee osteoarthritis. Report of a Task Force of the Standing Committee for International Clinical Studies Including Therapeutic Trials (ESCISIT). Ann Rheum Dis 2003;62: 1145–55. 8. Maher CG, Sherrington C, Herbert RD, Moseley AM, Elkins M. Reliability of the PEDro scale for rating quality of randomized controlled trials. Phys Ther 2003;83:713–21. 9. Ferreira PH, Ferreira ML, Maher CG, Herbert RD, Refshauge K. Speciﬁc stabilisation exercise for spinal and pelvic pain: a systematic review. Aust J Physiother 2006;52:79 – 88. 10. Herbert RD, Gabriel M. Effects of stretching before and after exercising on muscle soreness and risk of injury: systematic review. BMJ 2002;325:468. 11. McAlindon TE, LaValley MP, Gulin JP, Felson DT. Glucosamine and chondroitin for treatment of osteoarthritis: a systematic quality assessment and meta-analysis. JAMA 2000; 283:1469 –75. 12. Reichenbach S, Sterchi R, Scherer M, Trelle S, Burgi E, Burgi U, et al. Meta-analysis: chondroitin for osteoarthritis of the knee or hip. Ann Intern Med 2007;146:580 –90. 13. Hinman RS, Bennell KL, Crossley KM, McConnell J. Immediate effects of adhesive tape on pain and disability in individuals with knee osteoarthritis. Rheumatology (Oxford) 2003; 42:865–9. 14. Bjordal JM, Ljunggren AE, Klovning A, Slordal L. Nonsteroidal antiinﬂammatory drugs, including cyclo-oxygenase-2 inhibitors, in osteoarthritic knee pain: meta-analysis of randomised placebo controlled trials. BMJ 2004;329:1317. 15. Eisenberg E, McNicol ED, Carr DB. Efﬁcacy and safety of opioid agonists in the treatment of neuropathic pain of nonmalignant origin: systematic review and meta-analysis of randomized controlled trials. JAMA 2005;293:3043–52. 16. Lee C, Straus WL, Balshaw R, Barlas S, Vogel S, Schnitzer TJ. A comparison of the efﬁcacy and safety of nonsteroidal antiinﬂammatory agents versus acetaminophen in the treatment of osteoarthritis: a meta-analysis. Arthritis Rheum 2004;51: 746 –54. 17. Higgins JP, Green S, editors. Cochrane handbook for systematic reviews of interventions 4.2.6 (updated September 2006). In: The Cochrane Library. IV. Chichester (UK): John Wiley & Sons; 2006. Warden et al 18. Cohen J. Statistical power analysis for the behavioral sciences. 2nd ed. Hillsdale (NJ): Lawrence Erlbaum; 1988. 19. Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta-analyses. BMJ 2003;327:557– 60. 20. Egger M, Davey Smith G, Schneider M, Minder C. Bias in meta-analysis detected by a simple, graphical test. BMJ 1997; 315:629 –34. 21. Sterne JA, Egger M, Smith GD. Systematic reviews in health care: investigating and dealing with publication and other biases in meta-analysis. BMJ 2001;323:101–5. 22. Christou EA. Patellar taping increases vastus medialis oblique activity in the presence of patellofemoral pain. J Electromyogr Kinesiol 2004;14:495–504. 23. Clark DI, Downing N, Mitchell J, Coulson L, Syzpryt EP, Doherty M. Physiotherapy for anterior knee pain: a randomised controlled trial. Ann Rheum Dis 2000;59:700 – 4. 24. Conway A, Malone TR, Conway P. Patellar alignment/tracking alteration: effect on force output and perceived pain. Isokinet Exerc Sci 1992;2:9 –17. 25. Cowan SM, Bennell KL, Hodges PW. Therapeutic patellar taping changes the timing of vasti muscle activation in people with patellofemoral pain syndrome. Clin J Sport Med 2002; 12:339 – 47. 26. Cushnaghan J, McCarthy C, Dieppe P. Taping the patella medially: a new treatment for osteoarthritis of the knee joint? BMJ 1994;308:753–5. 27. Finestone A, Radin EL, Lev B, Shlamkovitch N, Wiener M, Milgrom C. Treatment of overuse patellofemoral pain: prospective randomized controlled clinical trial in a military setting. Clin Orthop Relat Res 1993;293:208 –10. 28. Handﬁeld T, Kramer J. Effect of McConnell taping on perceived pain and knee extensor torques during isokinetic exercise performed by patients with patellofemoral pain syndrome. Physiother Can 2000;52:39 – 44. 29. Hinman RS, Crossley KM, McConnell J, Bennell KL. Efﬁcacy of knee tape in the management of osteoarthritis of the knee: blinded randomised controlled trial. BMJ 2003;327:135. 30. Kowall MG, Kolk G, Nuber GW, Cassisi JE, Stern SH. Patellar taping in the treatment of patellofemoral pain: a prospective randomized study. Am J Sports Med 1996;24:61– 6. 31. Lun VM, Wiley JP, Meeuwisse WH, Yanagawa TL. Effectiveness of patellar bracing for treatment of patellofemoral pain syndrome. Clin J Sport Med 2005;15:235– 40. 32. Miller MD, Hinkin DT, Wisnowski JW. The efﬁcacy of orthotics for anterior knee pain in military trainees: a preliminary report. Am J Knee Surg 1997;10:10 –3. 33. Ng GY, Cheng JM. The effects of patellar taping on pain and neuromuscular performance in subjects with patellofemoral pain syndrome. Clin Rehabil 2002;16:821–7. 34. Powers CM, Ward SR, Chan LD, Chen YJ, Terk MR. The effect of bracing on patella alignment and patellofemoral joint contact area. Med Sci Sports Exerc 2004;36:1226 –32. 35. Whittingham M, Palmer S, Macmillan F. Effects of taping on pain and function in patellofemoral pain syndrome: a randomized controlled trial. J Orthop Sports Phys Ther 2004;34: 504 –10. 36. Wilson T, Carter N, Thomas G. A multicenter, single-masked study of medial, neutral, and lateral patellar taping in individuals with patellofemoral pain syndrome. J Orthop Sports Phys Ther 2003;33:437– 43. 37. Altman R, Asch E, Bloch D, Bole G, Borenstein D, Brandt K, et al. Development of criteria for the classiﬁcation and reporting of osteoarthritis: classiﬁcation of osteoarthritis of the knee. Arthritis Rheum 1986;29:1039 – 49. 38. McConnell J. The management of chondromalacia patellae: a long-term solution. Aust J Physiother 1986;32:215–33. 39. MacIntyre NJ, Hill NA, Fellows RA, Ellis RE, Wilson DR. Patellofemoral joint kinematics in individuals with and without patellofemoral pain syndrome. J Bone Joint Surg Am 2006;88:2596 – 605. 40. Niu J, Zhang YQ, Nevitt M, Xu L, Felson DT, Zhu YY, et al. Effect of Patellar Taping and Bracing on Chronic Knee Pain 41. 42. 43. 44. 45. Patella malalignment is associated with prevalent patellofemoral osteoarthritis: the Beijing Osteoarthritis Study [abstract]. Arthritis Rheum 2005;52 Suppl 9:S456 –7. Bellamy N, Carette S, Ford PM, Kean WF, le Riche NG, Lussier A, et al. Osteoarthritis antirheumatic drug trials. III. Setting the delta for clinical trials: results of a consensus development (Delphi) exercise. J Rheumatol 1992;19:451–7. Crossley KM, Bennell KL, Cowan SM, Green S. Analysis of outcome measures for persons with patellofemoral pain: which are reliable and valid? Arch Phys Med Rehabil 2004; 85:815–22. Aminaka N, Gribble PA. A systematic review of the effects of therapeutic taping on patellofemoral pain syndrome. J Athl Train 2005;40:341–51. Bizzini M, Childs JD, Piva SR, Delitto A. Systematic review of the quality of randomized controlled trials for patellofemoral pain syndrome. J Orthop Sports Phys Ther 2003;33:4 –20. Crossley K, Bennell K, Green S, McConnell J. A systematic 83 46. 47. 48. 49. review of physical interventions for patellofemoral pain syndrome. Clin J Sport Med 2001;11:103–10. D’hondt NE, Struijs PA, Kerkhoffs GM, Verheul C, Lysens R, Aufdemkampe G, et al. Orthotic devices for treating patellofemoral pain syndrome. Cochrane Database Syst Rev 2002; 2:CD002267. Abeillon G, Dubois T, Auclair J, Calmels P, Domenach M, Minaire P. Patellar syndromes: efﬁcacy of a knee-brace with patellar window in the therapeutic programme. J Traumatol Sport 1991;8:121–7. In French. Nafstad GL, Tronstad A, Aune AK. Patellofemoral pain: a comparative study of patella stabilizing orthosis versus tape as an adjuvant to exercise-therapy. Norsk Tidsskrift For Idrettsmedisin 1996;11:26 –38. Park YS, Kim HJ. Effects of a taping method on pain and ROM of the knee joint in the elderly. Taehan Kanho Hakhoe Chi 2005;35:372– 81. In Korean.