You are here

Article Text

Article menu
Review
Is hip strength a risk factor for patellofemoral pain? A systematic review and meta-analysis
  1. M S Rathleff1,2,
  2. C R Rathleff1,
  3. K M Crossley3,
  4. C J Barton4,5,6,7
  1. 1Orthopaedic Surgery Research Unit, Research and Innovation Center, Aalborg University Hospital, Aalborg, Denmark
  2. 2Institute of Clinical Medicine, Aarhus University, Aarhus, Denmark
  3. 3Division of Physiotherapy, School of Health and Rehabilitation Sciences, University of Queensland, Brisbane, Queensland, Australia
  4. 4Complete Sports Care, Melbourne, Victoria, Australia
  5. 5Pure Sports Medicine, London, UK
  6. 6Centre for Sport and Exercise Medicine, Queen Mary University of London
  7. 7Musculoskeletal Research Centre, La Trobe University, Melbourne, Australia
  1. Correspondence to Dr Michael Skovdal Rathleff, Orthopaedic Surgery Research Unit, Research and Innovation Center, Aalborg University Hospital, Soendre Skovvej 15, Aalborg 9000, Denmark; michaelrathleff{at}gmail.com

Abstract

Objective To evaluate and synthesise the literature on hip strength among patients with patellofemoral pain (PFP) to address the following: (1) differentiate between hip strength as a risk factor and associated deficit in PFP; (2) describe hip strength in men and women with PFP across different age ranges; (3) investigate the effects of hip strengthening on biomechanical knee variables associated with PFP development.

Methods MEDLINE, CINAHL, Web of Science, SportDiscus and Google Scholar were searched in November 2013 for studies investigating hip strength among patients with PFP. Two reviewers independently assessed papers for inclusion and quality. Means and SDs were extracted from each included study to allow effect size calculations and comparisons of results.

Results Moderate-to-strong evidence from prospective studies indicates no association between isometric hip strength and risk of developing PFP. Moderate evidence from cross-sectional studies indicates that men and women with PFP have lower isometric hip musculature strength compared to pain-free individuals. Limited evidence indicates that adolescents with PFP do not have the same strength deficits as adults with PFP.

Conclusions This review highlights a possible discrepancy between prospective and cross-sectional research. Cross-sectional studies indicate that adult men and women with PFP appear to have lower hip strength compared to pain-free individuals. Contrary to this, a limited number of prospective studies indicate that there may be no association between isometric hip strength and risk of developing PFP. Therefore, reduced hip strength may be a result of PFP rather than the cause.

  • Knee
  • Strength Isometric Isokinetic

https://doi.org/10.1136/bjsports-2013-093305

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Introduction

    The incidence and prevalence of patellofemoral pain (PFP) are high in both adolescent and adult populations.1–7 Importantly, despite the implementation of conservative rehabilitation, the long-term prognosis for PFP is poor, with only one-third being pain-free, 1 year after the initial diagnosis.8–11 Identifying modifiable risk factors will facilitate targeted rehabilitation and prevention strategies for PFP, thus reducing the burden of PFP.12

    The aetiology of PFP is thought to be multifactorial.13 Local factors, such as lower knee extension strength,14 delayed onset of vastus medialis relative to vastus lateralis,4 greater knee abduction impulse during running15 and knee abduction moment during drop jump landing16 have been associated with increased risk of PFP development. While all factors can contribute to PFP and are likely to be interrelated, recently much attention has been paid to the relationship between hip function and PFP.12 It is proposed that greater hip adduction and internal rotation, especially during weight-bearing activities, may lead to altered knee and patellofemoral joint (PFJ) kinematics with subsequent increase in lateral PFJ stress.17 Recent prospective studies support this hypothesis, reporting increased peak hip internal rotation angle during landing in military recruits,6 and greater peak hip adduction angle in recreational female runners18 to be risk factors for PFP development. These altered movement patterns may result from impaired hip muscle function.6 ,18

    Lower hip muscle strength has been explored in people with PFP. The most recent systematic review of hip strength included a search completed in January 2008, and focused only on women.19 This review concluded that women demonstrated lower hip abduction, external rotation and extension strength in the affected side compared to asymptomatic individuals.19 However, no prospective research was identified to allow evaluation of a causal relationship between hip muscle function and PFP.20 Since the publication of this systematic review, many studies, including those with prospective designs, have evaluated the link between hip strength and PFP, and it is timely to conduct an updated high-quality (HQ) systematic review in this field.

    In line with the belief that impaired hip muscle function is associated with PFP, the clinical effects of hip muscle strengthening programmes have recently been evaluated.21–24 However, in order to gain a deeper understanding of the role of hip muscle strength in PFP, an evaluation of the relationship between hip strengthening interventions and changes in knee biomechanics are required.

    The gaps in the literature call for the investigation of the following: (1) differentiate between hip strength as a risk factor and associated deficit in PFP; (2) association of hip strength with PFP in men and women across different age ranges and (3) the effects of hip strengthening on biomechanical knee variables associated with PFP development.

    Methods

    Search strategy

    MEDLINE, EMBASE, CINAHL, Web of Science, Google Scholar, SportDiscus databases were searched from inception until November 2013. No data limits were used during searching. Without any modification, the search strategy from a Cochrane systematic review on exercise therapy for PFP was used for diagnostic search terms (see online supplementary appendix 1).25 The diagnostic search terms were then combined with the terms: strength, torque, force, moment, isokinetic, isometric, eccentric, concentric, power, isotonic, endurance, kinematics, moments and kinetics. Reference lists and citing articles of included papers were also screened. Additionally a cited reference search for each included paper was completed in Google Scholar for additional publications of interest. Unpublished work was not sought in this review. This could potentially lead to publication bias because significant results are more likely to be published.26 In addition, it would have been unfeasible to contact all institutions and authors around the world for unpublished work. No review protocol exists.

    Inclusion and exclusion criteria

    Data from prospective studies investigating risk factors for PFP were included to answer objective 1. Cross-sectional and prospective studies evaluating hip strength were considered for inclusion to answer objective 2 if they contained data on hip strength from a comparison between (1) participants with PFP and pain-free individuals or (2) women and men with PFP. Data from prospective and randomised trials investigating the effect of hip strengthening on knee kinematics were included to help answer objective 3. Hip strength in all six movement directions was included (flexion, extension, abduction, adduction, internal rotation and external rotation). The inclusion criteria required patients to be diagnosed with PFP, anterior knee pain or chondromalacia patellae. Studies including participants with other knee conditions such as patellar tendinopathy or osteoarthritis, where individuals with PFP could not be separately analysed, were excluded. Studies using manual assessment of hip strength without a dynamometer or similar (ie, manual muscle tests) were excluded.

    Review process

    Titles and abstracts identified in the search were downloaded into Endnote V.X4 (Thomson Reuters, Carlsbad, California, USA), cross-referenced, and duplicates deleted. All potential publications were independently assessed for inclusion by two reviewers (MSR and CRR), and full texts were obtained if necessary. Any discrepancies were resolved during a consensus meeting, and a third reviewer was available (CJB) if needed.

    Quality assessment and risk of bias

    Included papers were assessed for methodological quality by two independent raters (MSR and CRR), with any disagreements resolved by consultation with a third party. Title, journal and author details were removed to de-identify articles prior to rating. Quality ratings were performed using the Epidemiological Appraisal Instrument (EAI)27 using similar methodology as Nix et al.28 The EAI was designed as ‘a critical appraisal system rooted in epidemiological principles’ for use in systematic reviews and meta-analysis, and developed by a team of epidemiologists, physicians and biostatisticians. The robust development consisted of an eight-step process including content and criterion validation, consultation with individuals outside the research team, two associated revised versions and establishment of the scales reliability.27 Utilising the method described by Nix et al,28 the EAI was adjusted to the intended purpose of this review and 26 of the original 43 items were used. Items relating to interventions, randomisation, follow-up period or loss to follow-up that were not applicable to three individual objectives were excluded by MSR and CJB. Identical to Nix et al,29 we used a median approach to divide included studies into higher and lower quality groups, that is, describing 50% as higher quality and 50% as lower quality based on the scores from the EAI.

    In addition a PFP diagnosis checklist was used.30 The diagnosis checklist is a seven-item scale summarising the reporting of key inclusion and exclusion criteria for the diagnosis of PFP, with higher scores indicating a greater number of desired criteria is reported. Each scale was applied by two reviewers (MSR and CRR) with discrepancies resolved during a consensus meeting, and a third reviewer was available (CJB) if needed.

    Study analysis

    Sample sizes, participant demographics, population sources and pain duration were extracted by MSR and checked by CRR. Sample sizes, means and SDs of all outcomes were extracted or sought from original authors to allow calculation of the standardised mean difference (SMD) as a measure of effect size. MSR extracted all outcomes from the papers into a spreadsheet. After the content of the spreadsheet was checked by CRR and CJB for data accuracy, the SMDs were calculated using a random effects model in Review Manager V.5.2 (Copenhagen: The Nordic Cochrane Centre, The Cochrane Collaboration). Following the methodology used by Hume et al,30a the individual or pooled SMDs were categorised as small (≤0.59), medium (0.60–1.19) or large (≥1.20). These criteria were chosen to be more stringent than traditional criteria.31 Furthermore these criteria have been used in previous systematic reviews on PFP, which facilitates comparison.32 ,33 Forest plots were used to allow easy visualised comparisons between studies and outcomes. The level of statistical heterogeneity for pooled data was established using the χ2 and I2 statistics. The χ2 and I2 statistics describe heterogeneity or homogeneity of the comparisons with p<0.05 indicating a significant heterogeneity. The χ2 and I2 statistics are presented together with levels of evidence in the Result section. Definitions for ‘levels of evidence’ were guided by recommendations made by van Tulder et al34.

    Strong evidence: pooled results derived from three or more studies, including a minimum of two HQ studies, which are statistically homogenous (p>0.05)—may be associated with a statistically significant or non-significant pooled result.

    Moderate evidence: statistically significant pooled results derived from multiple studies, including at least one HQ study, which are statistically heterogeneous (p<0.05); or from multiple low-quality (LQ) studies which are statistically homogenous (p>0.05).

    Limited evidence: results from multiple LQ studies which are statistically heterogeneous (p<0.05); or from one HQ study.

    Very limited evidence: results from one LQ study.

    Conflicting evidence: pooled results insignificant and derived from multiple studies regardless of quality, of which some show statistical significance individually, which are statistically heterogeneous (p<0.05, ie, inconsistent).

    To evaluate the robustness of our meta-analysis, two sensitivity analyses were completed. One was completed with LQ studies removed, and one where studies with a score of 5 or below on the diagnosis checklist were excluded.

    Results

    The initial search produced 1947 unique citations. Following application of the selection criteria to titles and abstracts this was reduced to 77. The primary reason for exclusion of studies included lack of a control group for strength measurements. After reading full texts the number was reduced to 27 (see figure 1). An additional three citations were later excluded due to data duplication. Results of the diagnostic checklist and the EAI scale are shown in tables 1 and 2. Agreement between quality assessors, calculated as a weighted κ was 0.90 (95% CI 0.88 to 0.93). Online supplementary tables S1 and S2 summarise the main methodological details and results for the included studies. Three studies were prospective6 ,35 ,36 while the remaining 21 studies used a cross-sectional design. Four studies used concentric and/or eccentric isokinetic strength testing while the remainder used isometric strength measurements. Figure 2A–F illustrates effect sizes and meta-analysis for abduction, adduction, external rotation, internal rotation, extension and flexion strength. Figure 3 illustrates effect sizes for the effects of hip strengthening on knee kinematics and joint torques.

    View this table:
    Table 1

    Patellofemoral pain diagnosis checklist

    View this table:
    Table 2

    Quality ratings using the Epidemiological Appraisal Instrument

    Figure 1
    Figure 1

    PRISMA flowchart.

    Figure 2
    Figure 2
    Figure 2
    Figure 2
    Figure 2
    Figure 2
    Figure 2

    (A) Comparison of hip abduction, (B) comparison of hip adduction, (C) comparison of hip external rotation, (D) comparison of hip internal rotation, (E) comparison of hip extension, (F) comparison of hip flexion.

    Figure 3
    Figure 3

    Effects of hip strengthening on knee kinematics and joint torques.

    Hip strength—prospective studies

    Moderate-to-strong evidence from three HQ studies found no association between lower isometric strength in hip abduction, (SMD: 0.06, 95% CI −0.58 to 0.70, I2 = 0.75%, p=0.02), extension (SMD: −0.24, 95% CI −0.50 to 0.02; I2=0%, p=0.97), external rotation (SMD: −0.19, 95% CI −0.46 to 0.07, I2 = 0%, p=0.93) or internal rotation (SMD: −0.01 95% CI −0.40 to 0.38, I2=39%, p=0.20), and risk of developing PFP (SMD: −0.24 to 0.06, I2=0–75%, p=0.02–0.93).6 ,35 ,36 Moderate evidence from two HQ studies revealed no association between lower isometric strength in hip adduction and flexion and the risk of developing PFP (SMD: −0.06 to 0.19, I2=0–58%, p=0.12–0.41).35 ,36

    Hip strength—cross-sectional studies, mixed gender

    Moderate evidence indicates lower isometric hip abduction strength with a small effect size (ES) (4 HQ and 7 LQ studies: SMD: −0.58, 95% CI −0.85 to −0.30, I2=53%, p=0.02)24 ,37–46 and lower hip extension strength with a small ES (2 LQ studies: SMD: −0.56, 95% CI −1.04 to −0.08, I2=9%, p=0.29),37 ,43 as well as a trend for lower isometric hip external rotation (2 HQ and 5 LQ studies: SMD: −0.38, 95% CI −0.77 to 0.02, I2=60%, p=0.02)37–39 ,41 ,44 ,47 ,48 in individuals with PFP. Limited evidence from one HQ study indicates no difference in isometric hip internal rotation, (SMD: −0.06, 95% CI −0.68 to 0.56)48 or adduction strength, (SMD: 0.14, 95% CI −0.48 to 0.76).48

    Moderate evidence indicates lower eccentric hip external rotation strength with medium ES, (1 HQ and 1 LQ, SMD: −0.61 95% CI −0.98 to −0.24, I2=0%, p=0.33),49 ,50 but no difference in concentric (2 LQ, SMD: −0.29, 95% CI −0.74 to 0.15, I2=0%, p=0.78)44 ,49 or eccentric (2 LQ, SMD: −0.21, 95% CI −0.66 to 0.123, I2=0%, p=0.81)44 ,49 hip extension strength measured isokinetically. Limited evidence indicates lower eccentric isokinetic strength in hip abduction with medium ES (1 HQ and 1 LQ studies: SMD: −0.74, 95% CI −1.11 to −0.37, I2=0%, p=0.83) in individuals with PFP.49 ,50 Very limited evidence indicates no difference in concentric hip abduction (1 LQ, SMD: −0.15, 95% CI −0.77 to 0.47)49 and no difference for concentric hip external rotation strength (1 LQ study, SMD: −0.51, 95% CI −1.14 to 0.12)49 between individuals with and without PFP.

    Hip strength—cross-sectional studies, gender specific

    Men

    Limited evidence indicates reduced isometric hip abduction strength with medium ES (1 HQ study: SMD: −0.74, 95% CI −1.39 to −0.10)50 and lower isokinetic eccentric hip external rotation with medium ES (1 HQ study: SMD: −0.98, 95% CI −1.64 to −0.32)50 in men with PFP. Limited evidence indicates no difference in isokinetic eccentric hip abduction strength (1 HQ, SMD: −0.59, 95% CI −1.23 to 0.04)50 in men with PFP.

    Women

    Strong evidence indicates lower isometric hip external rotation (4 HQ and 2 LQ studies: SMD −0.83, 95% CI −1.08 to −0.57, I2=0%, p=0.46),45 ,51–55 isometric hip internal rotation (3 HQ studies: SMD: −0.44, 95% CI −0.77 to −0.12, I2=0%, p=0.49)52 ,54 ,55 and isometric hip extension (2 HQ and 1 LQ studies: SMD: −1.10, 95% CI −1.77 to −0.43, I2=58%, p=0.09)47 ,52 ,54 in women with PFP. Moderate evidence indicates lower isometric hip abduction strength (4 HQ and 3 LQ studies: SMD: −1.10, 95% CI −1.65 to −0.55, I2=74%, p<0.0001),47 ,50–55 isometric hip adduction strength (2 HQ studies: SMD: −0.71, 95% CI −1.53 to −0.35, I2=0%, p=0.86)52 ,54 and isometric hip flexion strength (2 HQ studies, SMD: −0.89, 95% CI −1.26 to −0.52, I2=0%, p=0.97)52 ,54 in women with PFP.

    Moderate evidence indicates lower eccentric hip abduction (2 HQ studies: SMD: −1.21, 95% CI −1.76 to −0.65, I2=0%, p=0.92)50 ,56 in women with PFP. Limited evidence indicates no difference in eccentric hip adduction (1 HQ study: SMD: −0.63, 95% CI −1.53 to 0.28),56 eccentric hip external rotation (2 HQ studies: SMD: −0.70, 95% CI −2.03 to 0.63, I2=82%, p=0.02)50 ,56 or eccentric hip internal rotation strength (1 HQ study: SMD: −0.45, 95% CI −1.34 to 0.44)56 in women with PFP.

    Hip strength among adolescents

    No studies have specifically investigated the association between hip strength and age. Only two HQ studies investigated hip strength among adolescents (<18 years of age).35 ,48 One used a prospective design while the other used a cross-sectional design. An unadjusted analysis from the prospective HQ study indicates higher hip abduction strength among those who developed PFP (SMD: 1.06, 95% CI 0.15 to 1.97) but no association for hip adduction, hip external or internal rotation or hip extension or flexion strength (SMD: −0.55 to 0.48).35 The cross-sectional HQ study indicates no difference in hip strength between adolescents with PFP and pain-free adolescents (SMD: −0.06 to 0.26).48

    The effect of hip strengthening on knee kinematics and joint torques

    Two LQ studies investigated the effect of hip strengthening on knee kinematics and joint torques.24 ,57 Pooling of estimates were not possible. Ferber et al24 found no significant effect on two dimensional peak knee abduction (genu valgum) angle after 6 weeks of proximal strength training (SMD: 0.47, 95% CI −0.25 to 1.20). Earl and Hoch57 showed that peak internal knee abduction moment was significantly reduced after a rehabilitation programme (SMD: −0.40, 95% CI −1.04 to 0.24), however no significant change in knee abduction range of motion (ROM) was found (SMD: 0.05, 95% CI −0.58 to 0.69).

    Sensitivity analysis

    Results from the two sensitivity analyses (ie, excluding studies scoring within the lowest 50% on the EAI or studies with a score of 5 or below on the diagnosis checklist) showed similar findings and effect sizes (see online supplementary table S3). The two studies investigating the effect of hip strengthening on knee kinematics pain were omitted in the sensitivity analyses based on the EAI but kept in the analysis using the diagnosis checklist.

    Discussion

    This review provides a number of important findings regarding the association of hip muscle function and PFP. Most importantly, there appears to be a discrepancy between prospective and cross-sectional research findings, although further research is needed to confirm this. Specifically, across women, men and mixed gender studies of adults, the meta-analysis of cross-sectional studies indicates that individuals with PFP possess weaker hip musculature compared to pain-free individuals. Contrary to this, our meta-analysis of a small number of prospective studies indicates no association between isometric hip strength and risk of developing PFP, indicating that reduced isometric hip strength may be the result, rather than one of the causes of PFP. However these findings are based on a small number of heterogeneous studies and further research is needed to confirm this. Second, the very few studies that investigated men in isolation observed similar, albeit smaller, deficits in hip abduction and external rotation muscle strength to those seen in women. Additionally, limited evidence indicates that adolescents with PFP may not possess the same hip strength deficits as adults with PFP.

    Association of hip strength deficits with risk of PFP development

    Despite the meta-analysis indicating that hip strength deficits are present in female, male and mixed gender cohorts with PFP, decreased isometric hip strength was not found to be a risk factor for PFP development. However, the meta-analysis did indicate a trend towards reduced isometric hip external rotation and hip extension, although the effect sizes were very small (from −0.25 to −0.19). Additionally, when the Finnoff et al35 data, which includes adolescents, is taken from pooled effect sizes, isometric abduction strength deficits are found to be associated with increased risk of PFP development, but again the effect size is very small (−0.29). This suggests that hip strength deficits may play a small role in the development of PFP, but become more prominent following the development of symptoms, possibly as a result of disuse and fear avoidance to movement.35 Considering favourable clinical outcomes with hip strengthening protocols in individuals with PFP,21–24 the findings of this review do not bring into question the potential benefit of such programmes for symptom reduction. However, they do indicate that hip strengthening protocols may not be sufficient to prevent the occurrence or reoccurrence of PFP.

    Limitations of previous prospective studies must be considered. First, two6 ,35 of the three36 prospective studies used a mixed-gender population, which in light of gender differences identified in cross-sectional findings may confound their results. It is plausible that hip strength deficits may be a risk factor of PFP development in women but not in men. However, Thijs et al's36 prospective study investigated women only, and did not identify isometric hip strength deficits as a risk factor in the development of PFP. Considering that this was the only study to separate genders, further prospective gender-specific studies are warranted to confirm these findings. A second consideration is that all three prospective studies investigated isometric hip strength.6 ,35 ,36 It is possible that isometric strength is not the optimal clinical measure of hip strength. However, based on this meta-analysis there is no clear trend towards isokinetic measurements showing higher effect sizes compared to isometric measurements. Importantly, all prospective studies only investigated isometric strength and further research is needed to determine whether other measures of hip muscle strength such as endurance and eccentric strength may be risk factors for the development of PFP.

    Hip strength deficits across genders

    Effect sizes for gender-specific comparisons generally indicate larger hip strength deficits (ie, larger effect sizes) for female populations compared to male and mixed gender populations. This finding supports previous assumptions that men and women represent distinct subgroups within PFP with distinct gender-specific risk factors and deficits.13 ,40 ,58 However, few studies have evaluated hip strength in men and more studies are needed to explore hip muscle function in men. The best available evidence indicates that addressing isometric hip strength deficits during rehabilitation may be more important in women than in men with PFP.

    Age and its association to hip strength

    Most studies investigated adults (ie, >18 years old), with only two studies, one prospective and one cross-sectional, investigating hip strength among adolescents.35 ,48 Overall these studies did not support the presence of hip strength deficits as a risk factor for or as a feature of PFP among adolescents, with the authors concluding that PFP may have a different aetiology in adolescents than adults. This provides further evidence that the impairments in hip strength seen in adults may result from long-standing pain associated with PFP. Supporting this, Rathleff et al46 previously found reductions in knee extension strength among adolescents aged 15–19 years but not in a younger cohort of adolescents aged 12–16,48 even though the same methodology was used and both groups were recruited from the same closed population. Importantly, the adolescents with PFP between 15 and 19 years of age were 3 years older and reported a 1-year longer symptom duration.46 These findings are consistent with our finding that reduced isometric hip strength may be a result of PFP rather than a cause.

    Association between strength and kinematics

    Previous research indicates that greater knee abduction moments,15 ,16 peak hip internal rotation6 and hip adduction motion18 (ie, mechanics associated with greater dynamic knee valgus) are risk factors for PFP development. Therefore, it is plausible that the mechanism for successful hip strengthening protocols21–24 may be the correction of these mechanics. The two studies that investigated changes in the relevant knee valgus kinematics during running, found no changes in kinematics, despite significant symptom reduction.24 ,57 These results are consistent with findings from studies of asymptomatic individuals, which also indicate that hip strengthening programmes do not alter peak hip or knee kinematics during running.59 ,60 Although hip strengthening may not alter kinematics linked to PFP, findings from Earl and Hoch57 indicate a significant reduction in peak internal knee abduction moments during running. This is consistent with previous research among pain-free individuals showing similar positive benefits of hip strengthening on reducing the knee abduction moment.59 Collectively these findings indicate that positive clinical outcomes in patients with PFP21–24 may reflect a greater ability to attenuate forces at the knee rather than changes to kinematics.

    Clinical implications

    Findings from this review suggest that hip strength is reduced following the development of PFP, but hip strength deficits may not be a risk factor. This would indicate that hip strength testing as a screening tool may be of limited value, but is still important in individuals currently experiencing PFP. Considering favourable clinical21–24 and kinetic outcomes57 following hip strengthening protocols in individuals with PFP, their implementation in clinical practice is recommended. Additionally, this review indicates the addition of hip strengthening may be particularly important in female adults, but less so in male adults and adolescents of both genders.

    When developing rehabilitation programmes, clinicians must consider hip strengthening may have no effect on knee kinematics, and current prospective findings indicate strength deficits themselves are not a risk factor for PFP development. Therefore, strength is only one aspect that should be targeted during rehabilitation. Retraining control of hip movement might be equally or more important than focusing on hip strengthening during rehabilitation, especially considering greater peak hip adduction angle and hip internal rotation angle during running and landing, respectively, have been linked to increased risk of PFP development.6 ,18 Indeed, retraining runners with excessive peak hip adduction to reduce this motion has been reported to reduce pain associated with PFP.61 ,62

    Methodological considerations and considerations for future research

    There was a large heterogeneity in the results from the included studies. Based on the study details in online supplementary table S2 some of this heterogeneity may be caused by large variations in the populations being studied. Some populations are representative of individuals active in sports,24 ,35–37 ,39 ,40 ,45 ,52 ,53 ,55 ,57 while other are representative of sedentary individuals.54 Some studies did not account for where the population being studied was recruited from or how controls were selected.38 ,41 ,42 ,47 ,49–51 ,56 This limits external validity and makes it difficult to extrapolate to whom these results are valid, and also decreases the internal validity as the populations may be representative of different backgrounds. Furthermore, the validity of the tools used for measurement of muscle strength was not assessed in this review. Statistical heterogeneity was evaluated where data pooling was completed. Heterogeneity existed for some of the comparisons, which provides some limitations to the validity of data pooling. To account for this, we incorporated the existence of heterogeneity into our predetermined levels of evidence, with lower levels allocated to the findings of comparisons with statistical heterogeneity.

    Only four studies used a blinded rater (the individual taking strength measurement).6 ,36 ,48 ,52 This may cause a detection bias towards overestimating the difference between individuals with PFP and pain-free individuals, which may inflate the effect size. To reduce the chance of detection bias, blinding of the examiner in future studies is recommended. Less than half of the studies provided a sample size calculation.24 ,39 ,40 ,45 ,50 ,52 ,53 ,57 While lack of sample size may have affected the power of individual studies to detect a between-group difference, pooling of data in the current meta-analysis counteracts this limitation. Likewise, less than half of the included studies conducted a reliability analysis of their measurements.6 ,40 ,41 ,45 ,48–51 ,54–56 Generally, most studies provided good information on inclusion and exclusion criteria with the majority scoring high on the diagnostic checklist in table 1. Furthermore, a large number of the studies reported basic demographics and measures of pain intensity and pain duration which increases description of which sample is being studied. Both sensitivity analyses showed similar results which further strengthens the robustness of our results.

    This review highlights that isometric strength testing is the most commonly used measurement to assess hip strength among individuals with PFP. However, isometric strength may not be the best descriptor of dynamic muscle function and future prospective studies should consider investigating eccentric hip strength and endurance along with more functional measurements of hip muscle function such as a single legged squat.63 In the context of rehabilitation of individuals who have already developed PFP, hip-strengthening programmes can relieve pain, but there is a need to study the effect of movement-pattern retraining with consideration to comparing these methods to more traditional strengthening protocols.

    Conclusion

    This review highlights a possible discord between prospective and cross-sectional research findings. Cross-sectional studies indicate that both adult males and adult females with PFP have lower hip strength compared to pain-free individuals. Contrary to this, there may be no association between isometric hip strength and risk of developing PFP, indicating that reduced isometric hip strength may be a result of PFP rather than a cause. This finding, combined with the absence of a link between hip strengthening programmes and changes to kinematics associated with risk of PFP development, indicates a changing focus in regard to hip rehabilitation may be needed. Specifically, targeting movement-pattern retraining separately or in combination with strengthening may improve PFP patient outcomes clinically and in research trials compared to strengthening alone.

    What are the new findings?

    • Hip strength deficits are more likely the result of rather than the cause of patellofemoral pain (PFP).

    • Both men and women with PFP possess hip strength deficits.

    • Hip strengthening programmes do not appear to change kinematics at the knee linked to PFP risk.

    References

    1. 1.↵
      1. Rathleff MS,
      2. Roos EM,
      3. Olesen JL,
      4. et al
      . Lower mechanical pressure pain thresholds in female adolescents with patellofemoral pain syndrome. J Orthop Sports Phys Ther 2013;43:414–21.
      OpenUrlCrossRefPubMed
    2. 2.↵
      1. Barber Foss KD,
      2. Myer GD,
      3. Chen SS,
      4. et al
      . Expected prevalence from the differential diagnosis of anterior knee pain in adolescent female athletes during preparticipation screening. J Athl Train 2012;47:519–24.
      OpenUrlPubMed
    3. 3.↵
      1. Taunton JE,
      2. Ryan MB,
      3. Clement DB,
      4. et al
      . A retrospective case-control analysis of 2002 running injuries. Br J Sports Med 2002;36:95–101.
      OpenUrlAbstract/FREE Full Text
    4. 4.↵
      1. Van Tiggelen D,
      2. Cowan S,
      3. Coorevits P,
      4. et al
      . Delayed vastus medialis obliquus to vastus lateralis onset timing contributes to the development of patellofemoral pain in previously healthy men: a prospective study. Am J Sports Med 2009;37:1099–105.
      OpenUrlAbstract/FREE Full Text
    5. 5.↵
      1. Duvigneaud N,
      2. Bernard E,
      3. Stevens V,
      4. et al
      . Isokinetic assessment of patellofemoral pain syndrome: a prospective study in female recruits. Isokinet Exerc Sci 2008;16:213–19.
      OpenUrl
    6. 6.↵
      1. Boling MC,
      2. Padua DA,
      3. Marshall SW,
      4. et al
      . A prospective investigation of biomechanical risk factors for patellofemoral pain syndrome: the Joint Undertaking to Monitor and Prevent ACL Injury (JUMP-ACL) cohort. Am J Sports Med 2009;37:2108–16.
      OpenUrlAbstract/FREE Full Text
    7. 7.↵
      1. Witvrouw E,
      2. Lysens R,
      3. Bellemans J,
      4. et al
      . Intrinsic risk factors for the development of anterior knee pain in an athletic population. A two-year prospective study. Am J Sports Med 2000;28:480–9.
      OpenUrlAbstract/FREE Full Text
    8. 8.↵
      1. Rathleff MS,
      2. Rasmussen S,
      3. Olesen JL
      . Unsatisfactory long-term prognosis of conservative treatment of patellofemoral pain syndrome. Ugeskr Laeger 2012;174:1008–13.
      OpenUrlPubMed
    9. 9.↵
      1. Collins NJ,
      2. Bierma-Zeinstra SM,
      3. Crossley KM,
      4. et al
      . Prognostic factors for patellofemoral pain: a multicentre observational analysis. Br J Sports Med 2013;47:227–33.
      OpenUrlAbstract/FREE Full Text
    10. 10.↵
      1. Stathopulu E,
      2. Baildam E
      . Anterior knee pain: a long-term follow-up. Rheumatology (Oxford) 2003;42:380–2.
      OpenUrlAbstract/FREE Full Text
    11. 11.↵
      1. Nimon G,
      2. Murray D,
      3. Sandow M,
      4. et al
      . Natural history of anterior knee pain: a 14- to 20-year follow-up of nonoperative management. J Pediatr Orthop 1998;18:118–22.
      OpenUrlCrossRefPubMedWeb of Science
    12. 12.↵
      1. Witvrouw E,
      2. Callaghan MJ,
      3. Stefanik JJ,
      4. et al
      . Patellofemoral pain: consensus statement from the 3rd International Patellofemoral Pain Research Retreat held in Vancouver, September 2013. Br J Sports Med 2014;48:411–14.
      OpenUrlFREE Full Text
    13. 13.↵
      1. Powers CM,
      2. Bolgla LA,
      3. Callaghan MJ,
      4. et al
      . Patellofemoral pain: proximal, distal, and local factors, 2nd International Research Retreat. J Orthop Sports Phys Ther 2012;42:A1–54.
      OpenUrlPubMed
    14. 14.↵
      1. Lankhorst NE,
      2. Bierma-Zeinstra SM,
      3. van Middelkoop M
      . Risk factors for patellofemoral pain syndrome: a systematic review. J Orthop Sports Phys Ther 2012;42:81–94.
      OpenUrlCrossRefPubMed
    15. 15.↵
      1. Stefanyshyn DJ,
      2. Stergiou P,
      3. Lun VM,
      4. et al
      . Knee angular impulse as a predictor of patellofemoral pain in runners. Am J Sports Med 2006;34:1844–51.
      OpenUrlAbstract/FREE Full Text
    16. 16.↵
      1. Myer GD,
      2. Ford KR,
      3. Barber Foss KD,
      4. et al
      . The incidence and potential pathomechanics of patellofemoral pain in female athletes. Clin Biomech 2010;25:700–7.
      OpenUrlCrossRefWeb of Science
    17. 17.↵
      1. Powers CM
      . The influence of abnormal hip mechanics on knee injury: a biomechanical perspective. J Orthop Sports Phys Ther 2010;40:42–51.
      OpenUrlCrossRefPubMedWeb of Science
    18. 18.↵
      1. Noehren B,
      2. Hamill J,
      3. Davis I
      . Prospective evidence for a hip etiology in patellofemoral pain. Med Sci Sports Exerc 2013;45:1120–4.
      OpenUrlCrossRefWeb of Science
    19. 19.↵
      1. Prins MR,
      2. van der Wurff P
      . Females with patellofemoral pain syndrome have weak hip muscles: a systematic review. Aust J Physiother 2009;55:9–15.
      OpenUrlCrossRefPubMedWeb of Science
    20. 20.↵
      1. Barton CJ,
      2. Lack S,
      3. Malliaras P,
      4. et al
      . Gluteal muscle activity and patellofemoral pain syndrome: a systematic review. Br J Sports Med 2013;47:207–14.
      OpenUrlAbstract/FREE Full Text
    21. 21.↵
      1. Fukuda TY,
      2. Melo WP,
      3. Zaffalon BM,
      4. et al
      . Hip posterolateral musculature strengthening in sedentary females with patellofemoral pain syndrome: a randomized controlled clinical trial with 1-year follow-up. J Orthop Sports Phys Ther 2012;42:823–30.
      OpenUrlCrossRefPubMed
    22. 22.↵
      1. Khayambashi K,
      2. Mohammadkhani Z,
      3. Ghaznavi K,
      4. et al
      . The effects of isolated hip abductor and external rotator muscle strengthening on pain, health status, and hip strength in females with patellofemoral pain: a randomized controlled trial. J Orthop Sports Phys Ther 2012;42:22–9.
      OpenUrlCrossRefPubMed
    23. 23.↵
      1. Dolak KL,
      2. Silkman C,
      3. Medina McKeon J,
      4. et al
      . Hip strengthening prior to functional exercises reduces pain sooner than quadriceps strengthening in females with patellofemoral pain syndrome: a randomized clinical trial. J Orthop Sports Phys Ther 2011;41:560–70.
      OpenUrlCrossRefPubMed
    24. 24.↵
      1. Ferber R,
      2. Kendall KD,
      3. Farr L
      . Changes in knee biomechanics after a hip-abductor strengthening protocol for runners with patellofemoral pain syndrome. J Athl Train 2011;46:142–9.
      OpenUrlCrossRefPubMed
    25. 25.↵
      1. Heintjes E,
      2. Berger MY,
      3. Bierma-Zeinstra SM,
      4. et al
      . Exercise therapy for patellofemoral pain syndrome. Cochrane Database Syst Rev 2003(4):CD003472.
    26. 26.↵
      1. Maher CG,
      2. Sherrington C,
      3. Elkins M,
      4. et al
      . Challenges for evidence-based physical therapy: accessing and interpreting high-quality evidence on therapy. Phys Ther 2004;84:644–54.
      OpenUrlAbstract/FREE Full Text
    27. 27.↵
      1. Genaidy AM,
      2. Lemasters GK,
      3. Lockey J,
      4. et al
      . An epidemiological appraisal instrument—a tool for evaluation of epidemiological studies. Ergonomics 2007;50:920–60.
      OpenUrlCrossRefPubMedWeb of Science
    28. 28.↵
      1. Nix SE,
      2. Vicenzino BT,
      3. Collins NJ,
      4. et al
      . Characteristics of foot structure and footwear associated with hallux valgus: a systematic review. Osteoarthritis Cartilage 2012;20:1059–74.
      OpenUrlCrossRefPubMedWeb of Science
    29. 29.↵
      1. Nix S,
      2. Smith M,
      3. Vicenzino B
      . Prevalence of hallux valgus in the general population: a systematic review and meta-analysis. J Foot Ankle Res 2010;3:21.
      OpenUrlCrossRefPubMed
    30. 30.↵
      1. Barton CJ,
      2. Munteanu SE,
      3. Menz HB,
      4. et al
      . The efficacy of foot orthoses in the treatment of individuals with patellofemoral pain syndrome: a systematic review. Sports Med 2010;40:377–95.
      OpenUrlCrossRefPubMedWeb of Science
    31. 30a.↵
      Hume P, Hopkins W, Rome K, et al. Effectiveness of foot orthoses for treatment and prevention of lower limb injuries: a review. Sports Med 2008;38:759–79.
    32. 31.↵
      1. Cohen J
      . A power primer. Psychol Bull 1992;112:155–9.
      OpenUrlCrossRefPubMedWeb of Science
    33. 32.↵
      1. Barton CJ,
      2. Levinger P,
      3. Menz HB,
      4. et al
      . Kinematic gait characteristics associated with patellofemoral pain syndrome: a systematic review. Gait Posture 2009;30:405–16.
      OpenUrlCrossRefPubMedWeb of Science
    34. 33.↵
      1. Barton CJ,
      2. Lack S,
      3. Malliaras P,
      4. et al
      . Gluteal muscle activity and patellofemoral pain syndrome: a systematic review. Br J Sports Med 2013;47:207.
      OpenUrlAbstract/FREE Full Text
    35. 34.↵
      1. van Tulder M,
      2. Furlan A,
      3. Bombardier C,
      4. et al
      . Updated method guidelines for systematic reviews in the Cochrane Collaboration Back Review Group. Spine (Phila Pa 1976) 2003;28:1290–9.
      OpenUrl
    36. 35.↵
      1. Finnoff JT,
      2. Hall MM,
      3. Kyle K,
      4. et al
      . Hip strength and knee pain in high school runners: a prospective study. PM R 2011;3:792–801.
      OpenUrlCrossRefPubMed
    37. 36.↵
      1. Thijs Y,
      2. Pattyn E,
      3. Van Tiggelen D,
      4. et al
      . Is hip muscle weakness a predisposing factor for patellofemoral pain in female novice runners? A prospective study. Am J Sports Med 2011;39:1877–82.
      OpenUrlAbstract/FREE Full Text
    38. 37.↵
      1. Bazett-Jones DM,
      2. Cobb SC,
      3. Huddleston WE,
      4. et al
      . Effect of patellofemoral pain on strength and mechanics following an exhaustive run. Med Sci Sports Exerc 2013;45:1331–9.
      OpenUrlCrossRefWeb of Science
    39. 38.↵
      1. Cowan SM,
      2. Crossley KM,
      3. Bennell KL
      . Altered hip and trunk muscle function in individuals with patellofemoral pain. Br J Sports Med 2009;43:584–8.
      OpenUrlAbstract/FREE Full Text
    40. 39.↵
      1. Dierks TA,
      2. Manal KT,
      3. Hamill J,
      4. et al
      . Proximal and distal influences on hip and knee kinematics in runners with patellofemoral pain during a prolonged run. J Orthop Sports Phys Ther 2008;38:448–56.
      OpenUrlCrossRefPubMedWeb of Science
    41. 40.↵
      1. Nakagawa TH,
      2. Moriya ET,
      3. Maciel CD,
      4. et al
      . Frontal plane biomechanics in males and females with and without patellofemoral pain. Med Sci Sports Exerc 2012;44:1747–55.
      OpenUrlCrossRefPubMedWeb of Science
    42. 41.↵
      1. Piva SR,
      2. Goodnite EA,
      3. Childs JD
      . Strength around the hip and flexibility of soft tissues in individuals with and without patellofemoral pain syndrome. J Orthop Sports Phys Ther 2005;35:793–801.
      OpenUrlCrossRefPubMedWeb of Science
    43. 42.↵
      1. Roach S,
      2. Sorenson E,
      3. Headley B,
      4. et al
      . Prevalence of myofascial trigger points in the hip in patellofemoral pain. Arch Phys Med Rehabil 2012;94:522–6.
      OpenUrlPubMed
    44. 43.↵
      1. Souza RB,
      2. Powers CM
      . Differences in hip kinematics, muscle strength, and muscle activation between subjects with and without patellofemoral pain. J Orthop Sports Phys Ther 2009;39:12–19.
      OpenUrlCrossRefPubMedWeb of Science
    45. 44.↵
      1. Souza RB,
      2. Powers CM
      . Predictors of hip internal rotation during running: an evaluation of hip strength and femoral structure in women with and without patellofemoral pain. Am J Sports Med 2009;37:579–87.
      OpenUrlAbstract/FREE Full Text
    46. 45.↵
      1. Willson JD,
      2. Binder-Macleod S,
      3. Davis IS
      . Lower extremity jumping mechanics of female athletes with and without patellofemoral pain before and after exertion. Am J Sports Med 2008;36:1587–96.
      OpenUrlAbstract/FREE Full Text
    47. 46.↵
      1. Rathleff MS,
      2. Samani A,
      3. Olesen JL,
      4. et al
      . Neuromuscular activity and knee kinematics in adolescents with patellofemoral pain. Med Sci Sports Exerc 2013;45:1730–9.
      OpenUrlCrossRefWeb of Science
    48. 47.↵
      1. Robinson RL,
      2. Nee RJ
      . Analysis of hip strength in females seeking physical therapy treatment for unilateral patellofemoral pain syndrome. J Orthop Sports Phys Ther 2007;37:232–8.
      OpenUrlCrossRefPubMedWeb of Science
    49. 48.↵
      1. Rathleff CR,
      2. Baird WN,
      3. Olesen JL,
      4. et al
      . Hip and knee strength is not affected in 12–16 year old adolescents with patellofemoral pain—a cross-sectional population-based study. PLoS ONE 2013;8:e79153.
      OpenUrlCrossRefPubMed
    50. 49.↵
      1. Boling MC,
      2. Padua DA,
      3. Alexander Creighton R
      . Concentric and eccentric torque of the hip musculature in individuals with and without patellofemoral pain. J Athl Train 2009;44:7–13.
      OpenUrlCrossRefPubMedWeb of Science
    51. 50.↵
      1. Nakagawa TH,
      2. Moriya ET,
      3. Maciel CD,
      4. et al
      . Trunk, pelvis, hip, and knee kinematics, hip strength, and gluteal muscle activation during a single-leg squat in males and females with and without patellofemoral pain syndrome. J Orthop Sports Phys Ther 2012;42:491–501.
      OpenUrlCrossRefPubMed
    52. 51.↵
      1. Bolgla LA,
      2. Malone TR,
      3. Umberger BR,
      4. et al
      . Hip strength and hip and knee kinematics during stair descent in females with and without patellofemoral pain syndrome. J Orthop Sports Phys Ther 2008;38:12–18.
      OpenUrlCrossRefPubMedWeb of Science
    53. 52.↵
      1. Cichanowski HR,
      2. Schmitt JS,
      3. Johnson RJ,
      4. et al
      . Hip strength in collegiate female athletes with patellofemoral pain. Med Sci Sports Exerc 2007;39:1227–32.
      OpenUrlCrossRefPubMedWeb of Science
    54. 53.↵
      1. Ireland ML,
      2. Willson JD,
      3. Ballantyne BT,
      4. et al
      . Hip strength in females with and without patellofemoral pain. J Orthop Sports Phys Ther 2003;33:671–6.
      OpenUrlCrossRefPubMedWeb of Science
    55. 54.↵
      1. Magalhaes E,
      2. Fukuda TY,
      3. Sacramento SN,
      4. et al
      . A comparison of hip strength between sedentary females with and without patellofemoral pain syndrome. J Orthop Sports Phys Ther 2010;40:641–7.
      OpenUrlCrossRefPubMed
    56. 55.↵
      1. McMoreland A,
      2. O'Sullivan K,
      3. Sainsbury D,
      4. et al
      . No deficit in hip isometric strength or concentric endurance in young females with mild patellofemoral pain. Isokinet Exerc Sci 2011;19:117–25.
      OpenUrl
    57. 56.↵
      1. Baldon RdeM,
      2. Nakagawa TH,
      3. Muniz TB,
      4. et al
      . Eccentric hip muscle function in females with and without patellofemoral pain syndrome. J Athl Train 2009;44:490–6.
      OpenUrlCrossRefPubMed
    58. 57.↵
      1. Earl JE,
      2. Hoch AZ
      . A proximal strengthening program improves pain, function, and biomechanics in women with patellofemoral pain syndrome. Am J Sports Med2011;39:154–63.
      OpenUrlAbstract/FREE Full Text
    59. 58.↵
      1. Boling M,
      2. Padua D,
      3. Marshall S,
      4. et al
      . Gender differences in the incidence and prevalence of patellofemoral pain syndrome. Scand J Med Sci Sports 2010;20:725–30.
      OpenUrlCrossRefPubMedWeb of Science
    60. 59.↵
      1. Snyder KR,
      2. Earl JE,
      3. O'Connor KM,
      4. et al
      . Resistance training is accompanied by increases in hip strength and changes in lower extremity biomechanics during running. Clin Biomech 2009;24:26–34.
      OpenUrlCrossRefWeb of Science
    61. 60.↵
      1. Willy RW,
      2. Davis IS
      . The effect of a hip-strengthening program on mechanics during running and during a single-leg squat. J Orthop Sports Phys Ther 2011;41:625–32.
      OpenUrlCrossRefPubMed
    62. 61.↵
      1. Noehren B,
      2. Scholz J,
      3. Davis I
      . The effect of real-time gait retraining on hip kinematics, pain and function in subjects with patellofemoral pain syndrome. Br J Sports Med 2011;45:691–6.
      OpenUrlAbstract/FREE Full Text
    63. 62.↵
      1. Willy RW,
      2. Scholz JP,
      3. Davis IS
      . Mirror gait retraining for the treatment of patellofemoral pain in female runners. Clin Biomech (Bristol, Avon) 2012;27:1045–51.
      OpenUrlCrossRefPubMed
    64. 63.↵
      1. Crossley KM,
      2. Zhang WJ,
      3. Schache AG,
      4. et al
      . Performance on the single-leg squat task indicates hip abductor muscle function. Am J Sports Med 2011;39:866–73.
      OpenUrlAbstract/FREE Full Text

    Supplementary materials

    • Supplementary Data

      This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

      Files in this Data Supplement:

    Footnotes

    • Contributors MSR, CRR, KMC and CJB were responsible for the idea of the review and discussed the results. MSR and CRR performed the literature search and critical appraisal of the included study. MSR and CJB performed the meta-analyses. MSR and CRR wrote the first draft of the manuscript while KMC and CJB helped revise the manuscript. All authors approved the final version of the manuscript.

    • Competing interests None.

    • Provenance and peer review Not commissioned; externally peer reviewed.

    • Data sharing statement All data used in the review are available upon request.

    Linked Articles