Effect of Hip Abductors and Lateral Rotators Muscle Strengthening on Pain and Functional Outcome in Patients with Patellofemoral Pain: Systematic Review and Meta-Analysis

Background: This study systematically reviews and Meta-analyses the best-published evidence on the therapeutic value of strengthening hip abductors and lateral rotators muscles for the treatment of Patellofemoral Pain (PFP) with a presumptive hypothesis that strengthening hip muscles stabilises the patellofemoral joint, relieves pain and enhances knee functions. Methodology: Electronic database searches of Medline, EMBASE, CINAHL, PEDro and PubMed Central between January 1994 and September 2019 using PICOS tool. The methodological quality of the selected studies was appraised individually using the 20-item McMaster Critical Review Form for Quantitative Studies. Supplemental quality appraisal of randomized controlled clinical trials performed using the Cochrane Collaboration's 'Risk of bias' quality criteria. Data on patient population demographics, interventions, duration of intervention and outcome measures were extracted, summarized in evidence tables and descriptive analysis made. Pooled effects size from appropriate RCTs was determined by meta-analysis under both �xed and random-effects models. Results: All included fourteen studies demonstrated that hip muscles strengthening resulted in improved pain and knee function. All RCTs, except one, demonstrated that hip muscles strengthening is superior to quadriceps strengthening. Of the �ve RCTs accessing the surplus effect of hip-quad versus quadriceps strengthening, four suggested that hip-quad strengthening is superior over the standard quadriceps strengthening alone to improve PFP and knee function. Conclusion and Implications: In patients with PFP, strengthening of hip abductors and lateral rotators have greater therapeutic signi�cance than the conventional quadriceps exercises in improving knee pain and function both in the short-and long term. However, caution is required, because a standardized hip-quad protocol is yet to be developed, which warrants further studies. Systematic


Introduction
Patellofemoral pain (PFP) is characterised by anterior knee pain emanating from the patellofemoral joint involving patella and brous tissue on the medio-lateral retinaculum [1].The aetiology is irregular patellar kinematics due to excessive pressure on the patellofemoral joint coupled with poor proximal neuromuscular control and hip muscle weakness [2][3][4].The pain in PFP is because of in ammation coupled with damage to bony, cartilaginous or the connective tissues of the patellofemoral joint [1,2,5].
The PFP incidence rate is 25-40 % of all cases of anterior knee pain which is considerably high.Hence, therapeutic interventions are imperative to reduce permanent knee disabilities and improve quality of life [5][6][7].The prevalence of PFP is higher in women and athletes than males (2:1) and is even higher (4:1) among athletes [6].
The works of literature on musculoskeletal injuries indicate a positive correlation between hip muscles weakness and PFP [8][9][10].In a case report on PFP, authors noted that excessive hip adduction coupled with the weakness of the hip extensors and abductors are predominantly musculoskeletal concerns [10].The current physiotherapy evidence strongly supports quadriceps muscle strengthening as an effective strategy to improve overall knee function in patients with PFP [11][12][13].The proximal hip muscles exercises reported to be effective in relieving patellofemoral pain and improving knee function when compared to knee exercises alone [14].Therefore, strengthening these muscles underlies the objective treatment of PFP.While quadriceps strengthening is already the standard physiotherapeutic target for PFP, it is plausible that strengthening of hip muscles will serve greater bene ts, because of its effect on greater control over the knee biomechanics [5,15] .

Relationship between Hip Muscles (abductors and external rotators) Strength and PFP
Muscles of the hip (hip abductors and external rotators) are essential for knee and pelvic stabilization during ambulation [5].The hip abductors and external rotators act synergistically to eccentrically control the hip adduction and internal rotation movements, respectively [15][16][17].The diminished strength of hip abductors and external rotators muscles may result in poor neuromuscular control during activities that necessitate loading on the patellofemoral joint [5,8,18].The weak hip abductors may cause excessive femoral adduction, thereby, augmenting lateral forces (Knee Valgum) acting on the patella [19]; while, weak hip external rotators results in unrestricted internal rotation of the femur, that augments contact pressure between the lateral facet of the patella and lateral femoral condyle [19].Hence, weak hip muscles (mainly hip abductors and external rotators) is an important aetiological factor for PFP [5,15,20,21].
Many studies compared the effectiveness of hip muscles strength in patients with PFP to matched healthy controls [16, [21][22][23].Ireland et al. reported eccentric muscle strength reduction of 26% in hip abductors and 36% in hip external rotators among females with PFP while Souza & Powers, found a reduction of 14% in hip abductors and 17% in hip external rotators eccentric muscle strength compared to healthy matched controls [16,21].Nevertheless, Piva et al. found no signi cant muscle strength differences for hip abductors and external rotators in patients with PFP compared to healthy age/gender-matched controls; however, Baldon et al. reported signi cantly reduced strength for eccentric hip abductors, but not for hip external rotators among females with PFP to healthy matched controls [22,23].
The weak hip lateral rotators cause unrestricted internal rotation of the femur about the tibia, enhancing misalignment at the knee joint that in turn leads to a biomechanical imbalance between the hip extensors and lateral rotators, that overloads the retinaculum and subchondral bone and subsequently potentiate patellofemoral pain and knee dysfunction [21].Nevertheless, Earl et al. opined that strong hip muscles (abductors and external rotators) reverses these effects over the knee joint [3].Ireland et al. and Souza & Powers noted more weakness in hip external rotators compared to hip abductors in patients with PFP[16] [21].Ferber et al. found that in patients with PPS, the three weeks of isolated hip abductors strengthening not only reduces patellofemoral pain but also increases gait-related knee-joint stability [9].Two recent randomised controlled trials found that isolated strengthening of hip abductors and external rotators effectively relieves pain and improvise knee function in females [4,17].The available pieces of evidence for PFP considered exercises to strengthen the hip muscles that reduces pain and enhances long-term knee function [3,4,6,17,24].

Outcome Measures of Pain, Knee Function and Health Status in PFP:
The available studies used self-reported Kujala Anterior Knee Pain Scale (AKPS), Visual Analogue Pain (VAS) scale, 11-point Numerical Pain Rating Scale (NPRS) and Pain Severity Scale (PSS) as an outcome measure to document patellofemoral pain in patients with PFP receiving therapeutic interventions [14,25,26].
Though, the Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) is tailored to examine the functional status of osteoarthritis, is also been used to measure health status for patients with patellofemoral pain [because patients with osteoarthritis often present with anterior knee pain which is similar to patellofemoral pain] [25].
Since, systematic reviews evaluating the effect of hip abductors and lateral rotators strengthening for patellofemoral pain, knee function and quality of life in patients with PFP are extremely limited, even though, evidence indicating the presence of weak hip abductors and external rotators.Primarily this study systematically reviews and Meta-analyses the best-published evidence on the therapeutic value of strengthening hip abductors and lateral rotators muscles for the treatment of PFP.This is being guided by the presumptive hypothesis that strengthening hip muscles stabilises the patellofemoral joint, relieves pain and enhances knee functions.

Methodology
Justi cation of the Systematic Review Approach Systematic reviews and meta-analyses are important methodologies for the qualitative and quantitative synthesis of published evidence.Shreds of evidence presented in systematic reviews are key for continuous quality and safety improvements in evidence-based clinical practice and therefore, useful for both clinicians and healthcare policymakers.The present review study used Centre for Reviews and Dissemination (CRD) guidance for undertaking reviews in health interventions to assess the value of hip muscles strengthening as therapeutic interventions in patellofemoral pain and knee function in patients with PFP [31].Additionally, the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement, recommended in CRD's guidance is used in literature searches to minimise biases in article selection and reporting [32].The review has been registered with Open Science framework with reference doi: 10.17605/OSF.IO/CWZ8V

Electronic bibliographic database searches
The controlled clinical trials (randomised and non-randomised), controlled comparative studies and cohort studies ((prospective and retrospective) published in the last 25 years (January 1994 to September 2019) in English language journals were performed across ve electronic databases [Medical Literature Analysis and Retrieval System Online (Medline); Excerpta Medica Database (EMBASE); Clinical Index of Nursing and Allied Health Literature (CINAHL); Physiotherapy Evidence Database (PEDro) and The Cochrane Central Register of Controlled Trials (CENTRAL)].Appropriate combinations of Medical Subject Headings (MeSH) or CINAHL headings with keywords (Table 1) using Boolean Operators (AND, OR & NOT) along with PICOS (target population, intervention, comparator, outcomes and study design) were used [33].
Additional electronic searches are done in the Meta register of Controlled Trials (mRCT) via the Current Controlled Trials (CCT) database to locate ongoing RCTs with potentially relevant data useful for the present systematic review.The potentially relevant clinical controlled trials and cohort studies (otherwise not indexed in any of the ve electronic bibliographic databases and mRCT), electronic searches were supplemented by searching unpublished papers from the OpenGrey (formerly SIGLE) database.The literature searches were additionally supplemented with manual bibliographic searches of relevant systematic reviews, editorials and thesis reports published by the digital libraries of the University of Manchester, University of Central Lancashire and Australian Digital Thesis programmes including ProQuest.Authors of potentially relevant unpublished reports were contacted by e-mails seeking clari cation of their respective studies with the possibility of inclusion in the present review.

Study Selection
The study selection performed by the PRISMA ow chart where returned hits for each electronic bibliographic database were screened initially based on title and abstracts.The number of potentially relevant articles were noted and citations imported into Endnote citation manager (EndNote X7), and full-text articles retrieved for further eligibility screening.Studies were included based on the following criteria:  2) [14].Knowing that biases are the main threats to the internal and external validity of RCTs, quality appraisal of RCTs performed using The Cochrane Collaboration's 'Risk of bias' tool tailored speci cally for RCTs [34].The risk of patient selection bias was examined for the selected RCTs for the su ciency of random sequence generation and concealment allocation to interventional and control groups.This helped to determine the comparability of the study groups at baseline.The risk of performance bias was evaluated based on measures (such as single blinding or double-blinding) employed to ensure study participants and personnel are blinded to interventions and outcomes.The risk of detection bias was assessed to know if the assessors were adequately blinded to patient group allocation.The risk of attrition bias and the risk of incorporation bias examined based on the rate and pattern of the dropout of participants, handling of incomplete outcome data and the indications of intention-to-treat (ITT) analysis.Finally, the risk of reporting bias evaluated based on the possibility of selective outcome reporting.
The reproducibility of exercise therapies prescribed, confounding/modifying effects of co-interventions and the levels of supervision and patient compliance to the prescribed physiotherapy during the trial were also evaluated across the RCTs studies.
Data Extraction and Qualitative Synthesis Data on effect measures were extracted for baseline patellofemoral pain levels, hip exercise interventions, including the comparator treatment, quantitative assessment of patient outcomes for patellofemoral pain and functions, follow-up duration and post-intervention practices during the follow-up periods.Statistical results (mean differences from baseline and effect measures P value at 95% con dence interval) were taken from the evidence tables for interventional studies (separately for controlled clinical trials and cohort studies).
Quantitative Synthesis (Meta-Analysis) Using MedCalc software (Version 14.10.2,MedCalc Software Ltd), data from RCTs that provided the mean difference of pain or knee function between the intervention and the comparator groups were pooled by random or xed-effect models to obtain standardised mean differences.
Separate forest plots were generated for pain and knee function outcome.

Controlled Clinical Trials
A total of 383 participants from the 10 CCTs received either hip-strengthening exercises (N=74) or quadriceps strengthening exercises (N=157) or hip/quadriceps strengthening exercises (N=108) or no exercise (N=44) (Table 2).All CCTs involved true randomization of participants except one, where participants were allocated to their respective groups alternately in a consecutive manner [40].

Intervention Protocol:
In all CCTs, the hip muscles strengthening protocol focused on hip abductors and lateral rotators.The hip exercise protocol included hip abduction against an elastic band while standing, or with weights in side-lying position coupled with hip lateral rotation against an elastic band while seated and hip extension; quadriceps strengthening involved closed kinetic chain exercise or seated knee extension, leg press, squatting and stretching of hamstrings and quadriceps; and, hip-quadriceps strengthening involved combination of the hip-quadriceps protocol.The duration of intervention ranged from 3-8 weeks, while the frequency of therapy sessions ranged from 2-4 per week (Table 3).

Follow-up duration
Post-intervention measures were immediately carried out at the end of the intervention period in all studies.However, the post-interventional follow-up period ranged from one to twelve months (Table 4).

Critical appraisal
Methodological quality assessment of the 10 CCTs based on the Cochrane Collaboration's 'Risk of bias' tool tailored for RCTs is detailed in Table 5 & 6 below [34].

I. Cohort and Case Series Studies
The three cohort studies had 88 participants [PFP (n=64); healthy controls (n=24)].The one case series involved 19 participants with PFP.
Intervention Protocol: In one cohort study, the experimental group given hip muscles exercise protocol (strengthening of hip abductors and external rotators), and the control group received knee exercises.The other two cohort studies subjects received quadriceps-strengthening [43,44].The duration of intervention ranged between three to six weeks.The case series participants completed an eight-week exercise programme focusing on hip muscles strengthening and improving dynamic misalignment (Table 7).

II. Meta-analysis (pooled effect size)
The meta-analysis was done to determine the additional effect of hip muscles strengthening as adjunctive therapy to the standard quadriceps strengthening for PFP and knee function.
A. The comparative effect size of Hip versus Quadriceps strengthening on pain and function Two RCTs [6,38] and one comparative control trial [40] provided data that compared the effect of the isolated strengthening of hip muscles (hip abductors and lateral rotators) versus the standard quadriceps strengthening on PFP and knee function.A total of 100 participants were randomly assigned to receive either hip (n=50) or quadriceps (n=50) strengthening protocols.The standardised mean difference (SMD) of PFP and functional outcomes after intervention with 95% CI under both xed and random effects models favoured hip muscles strengthening over quadriceps strengthening (p<0.001) (Fig. 2 and 3).

B. Surplus effect of hip-quad versus quadriceps strengthening on pain and function
Five RCTs contributed data assessing the surplus effect of hip muscles strengthening coupled with quadriceps strengthening compared to the standard quadriceps strengthening alone on PFP and knee function [4,24,39,41,42].For both the group (hip-quad and quadriceps alone), 16 data sets were collected from a total of 98 participants.The pooled effects of results are presented in forest plots Fig. 4 and 5 as cumulative SMD with 95% CI, under both xed and random-effects models.

Discussion
Two recent systematic reviews have demonstrated that proximal exercises targeting quadriceps and hip muscles strengthening were effective in relieving pain and improving knee function in patients with PFP, both the short-and long-term [14,36].However, this systematic review was important to delineate the effect of the isolated strengthening of hip abductors and lateral rotators on pain and knee function in patients with PFP compared to non-exercise interventions and to identify if hip muscles strengthening is superior over the quadriceps strengthening alone among them.

A. Quality of the Summarised Evidence
The methodological quality of the fourteen studies except ve i.e. [6, 9,37,43,44]  .The four studies examined only female participants, therefore, the outcome may only be generalised for the female patients with PFP, but not for the males [4,6,24,38].The three studies [39,41,42], included both males and female participants (proportion of females was higher than males), indicative of females being the greater risk of PFP than males [6].This may be attributed to the lower hip muscle mass in females compared to males [47], therefore, females exhibiting lower hip muscle strength than males [47,48].
The symptom duration is a direct measure of severity of PFP that has a signi cant in uence on therapeutic outcome [49].Therefore, patients with early diagnosis of PFP likely to respond well to therapy compared to those with late diagnosis [20].Thus, symptom duration is a key confounding variable that needs to be adjusted via the subject's strati cation.In this systematic review, the mean duration of symptoms of participants with PFP in eight studies ranged from 17 to 21 months.However, six studies [17,[37][38][39][40][41] did not report the mean duration (months) of PFP symptoms.
None of the studies performed the subject's strati cation for the PFP severity and symptom duration.This might have positively skewed pain and functional outcomes in patients with a shorter mean duration of symptoms [20].Additionally, the subject's characteristics were barely explained in three studies [38,40,41] and also not detailed in one study [37].These ndings undermine the quality of the summarised shreds of evidence.
Supervised therapeutic exercises enhance participant's compliance because unsupervised participants may refrain from pain-provoking exercises [20,50].Two previous RCTs had reported that supervised exercises for PFP result in less pain and better knee function at short-and long-term follow-up compared to usual care [20,50].In the present systematic review, all studies involved exercises administered in physiotherapy facility/ rehabilitation setting under supervision by quali ed physiotherapists, except two [6,41], where two-thirds of exercise sessions were selfadministered in patients homes (unsupervised), while one-third had at rehabilitation facility under supervision.It had an important bearing on patient compliance to intervention and the outcome.Even then, results were signi cant in these two studies suggesting that partial supervision too can yield clinically signi cant results.The hip protocol generally resulted in improved pain and knee function after three to eight weeks of training, with long-term effects observed as late as twelve months post-intervention [24].[4,24,39,41,42].All studies except one supported that hip-quadriceps strategy was superior to the standard quadriceps [42].

B. Isolated Hip Musculature Strengthening
The ndings of these ve RCTs have both internal and external validity and are, therefore, acceptable.Furthermore, meta-analysed data of these ve studies strongly indicated that quadriceps coupled with hip muscles strengthening has signi cant surplus therapeutic bene ts over the conventional quadriceps or the hip exercises in the treatment of PFP (p<0.001).Therefore, a hip-quadriceps strategy should be adopted in clinical practices for pain relief and optimal functional improvements in patients with PFP.

Limitations
The summarised pieces of evidence supported by meta-analyses indicate that strengthening of hip muscles is effective in treating PFP for pain and knee function of physically active male/female adolescents and adults.However, a few but important limitations must be noted; 1.This systematic review and meta-analysis initially were intended to review a minimum of 20 studies to examine the therapeutic outcome of hip muscles strengthening versus quadriceps alone on pain and knee functions for patients with PFP.The expanded literature search yielded only 14 studies that are adequate for systematic review limits the strength and generalisability of the summarised ndings over a wider population of patients with PFP.
2. Avraham et al. study (included in this review) used a non-exercise (electrotherapy) as a co-intervention that might have uni-directionally augmented the therapeutic effects [37].
3. Though the proportion of females to males is higher in all studies (included in this review), but this may not be considered as a limitation to generalisability for a wider group of patients with PFP, because it truly re ects the characteristics of patients with PFP that would be encountered in day-to-day clinical practice.

Implications for routine physiotherapy practice
The evidence from the present review has important implications in routine clinical practice for the patients with PFP: 1. Strong shreds of evidence favour hip muscles strengthening exercises for two to four times a week, up to three to four weeks, to have effective therapeutic outcome compared to standard quadriceps strengthening exercises alone in patients with PFP.This implies that therapists should consider hip muscles strengthening as standard therapeutic measures while treating patients with PFP.
2. Meta-analysis of the effect measures (both pain and function) has strongly supported that hip muscles coupled with quadriceps (hip-quad) strengthening has superior therapeutic effects compared to the individual isolated hip or quadriceps strengthening exercises.This evidence strongly implies that therapists should consider a combination of hip and quadriceps strengthening exercises for the treatment of patients with PFP.However, this may imply longer therapy sessions plus more sessions per week that may in uence patient's compliance to intervention, especially if prescribed as self-e cacy [51].
3. In the present review, only one study [24]out of fourteen had followed patients up to twelve months, which was a good attempt to determine the long-term therapeutic effect of hip versus quadriceps strengthening exercises on PFP and knee function.This is indicative of evidence to be generalised only for the short-term, instead of long-term pain and functional outcomes.

Future Research
Must consider strati cation of patients/results based on the symptom duration before the intervention, to eliminate the effect of time-delay modi cation on pain and functional outcomes following hip muscles strengthening in patients with PFP.

Conclusion
The results of this systematic review and meta-analysis indicate that isolated strengthening of hip abductors and lateral rotators has therapeutic bene ts compare to quadriceps strengthening alone for the treatment of PFP.It is also clear that the hip-quadriceps strategy gives a greater therapeutic outcome than isolated quadriceps or hip muscles strengthening.Therefore, we recommend developing a hip-quadriceps exercise strategy for the treatment of PFP to encourage improved compliance, even in unsupervised patients.Anterior knee pain: "pain measurement", "The Kujala Anterior Knee Pain Scale"(AKPS), "The Visual Analogue Pain Scale" (VAS),11-Point Numerical Pain Rating Scale"(NPRS), "self-reported pain", "Pain Severity Scale"(PSS).
It was coupled with knee exercise for the KE group.

sessions per week for 4 weeks
Setting: Rehabilitation facility.
For descending stairs at 3, 6 and 12 months posttreatment were -1.4 ±0.9 (95% CI: -1.7, -1.1), -0.8 ±0.9 (95% CI: -1.2, -0.4), and 0.0 ±0.9 (95% CI: -0.Hip-quad strengthening results in signi cant pain improvements compared to the standard quadriceps strengthening alone. All fourteen studies demonstrated that isolated strengthening exercises of hip abductors & lateral rotators for two to four times per week up to three to eight weeks duration is effective in relieving pain and improving knee function compared to quadriceps strengthening and non-exercise interventions.Kooiker et al. reported variations in quadriceps, hip and hip-quadriceps strengthening protocols in selected studies, and opined for the unavailability of standardized protocols for PFP[36].The common hip exercise protocol included hip abduction against an elastic band while standing, and with weights in a side-lying position coupled with hip lateral rotation against an elastic band while seated and hip extension (3 sets of 10 repetitions).Conversely, quadriceps strengthening in all studies generally involved weight-bearing and non-weight-bearing exercises such as closed kinetic chain exercises, seated knee extension, leg press, squatting and stretching of hamstrings and quadriceps (3 sets of 10 repetitions).
change in mean NPRS scores:

Figure 4
Figure 4 [14,7,39,41]40,43]esent review is excellent because it ful lled 14 of the 16-item McMaster critical review criteria.The common methodological issue observed in the majority of the selected studies was lack of sample size justi cation (sample size not determined or not achieved)[6,17,37,39,40,43].All studies with sample power inadequacy issue, achieved results with statistical signi cance, suggesting that the measured pain and functional outcomes re ect the comparative effect of the interventions.However, subject contamination in Dolak et al. was evident because hip and quadriceps groups were combined as one to receive functional strengthening exercises (as co-interventions for the last four weeks of the intervention)[6].Such subject contamination might have caused patient bias for their pain and functional outcomes, especially if they know the intervention of their cohorts in the opposite arm of the study[45].These are classical symptoms for the diagnosis of PFP[14,36].However, these symptoms may be indicative of knee osteoarthritis, but it may not be so likely because patients enrolled in RCTs were not older than 50 years of age, and therefore not likely to present with ageing-associated PFP[46].Four studies included the mixed population of both adults and adolescents aged 17 to 50 years[6,38,39,41].Since adolescents are physically active, therefore, at risk of PFP, hence, the inclusion of this age group[46].To minimise the possibility of recruiting participants with underlying knee pathologies i.e. knee osteoarthritis, no studies recruited patients with PFP who were older than 50 years of age[46] [17,40]sk of bias is a critical methodological issue in RCTs and warranted supplementary quality appraisal of all RCTs on the Cochrane Collaboration's 'Risk of bias tool[34]to highlight methodological aws (indicative of 'Risk of bias' threatening interval consistency) (Table6).All RCTs except two recruited participants with a con rmed diagnosis of PFP[17,40].However, these studies were included because they enrolled patients presented with anterior keen pain based on symptoms matching the inclusion criteria of the remaining RCTs, which recruited patients with a con rmed diagnosis of PFP.Here, 383 participants from all RCTs presented with anterior knee pain associated with prolonged sitting, climbing stairs and descending stairs in the absence of signs/symptoms of meniscal or other intra-articular pathological conditions, or history of other knee pathologies, surgeries and injuries.
37,]7,tud40] evaluated comparative therapeutic value of quadriceps versus hip muscles strengthening in treating PFP[6,37,38,40].One study by Khayambashi et al. reported superiority of hip muscles strengthening strategy over the quadriceps strengthening for both pain and functional improvement in PFP[40].The remaining three studies opined that isolated hip and quadriceps strengthening strategy have comparable therapeutic value for the PFP [6,37,38].However, a meta-analysis of the effect measures (pain 14,ugh [37]proximal strengthening exercises involving quadriceps and hip muscles are commonly effective in treating PFP, Kooiker et al., Peters &Tyson opined that a combination of hip-quadriceps strategy could add the therapeutic outcome for the patients with PFP [6,14,[36][37][38].The present systematic review included ve RCTs to examine the surplus therapeutic outcome of hip-quadriceps strengthening exercises over the standard quadriceps

Table 2 :
Methodological quality of selected studies rated on McMaster critical review form RCT= randomised controlled trial, CCT= comparative control trial RCT-p = randomised controlled pilot study, CS= cohort study, CSr= case series: Key: ü= yes (criterion ful lled), × = No (criterion not ful lled/not clear), n/a= Not applicable 1.Is the study question and aims clear?; 2. Is the background literature review adequate leading to the research questions and objectives?; 3. Is the selected study design and study setting appropriate?; 4. Is the study sample characteristic suitable?; 5. Is the sample size adequate and justi ed?; 6.Is the study ethical?; 7. Is the reliability of outcome measures reported?; 8.Is the validity of outcome measures reported?; 9. Is intervention descriptions clear and adequate?; 10.Was contamination of sample populations avoided?; 11.Is co-interventions are avoided?; 12. Are results reported in terms of statistical signi cance?; 13.Were appropriate statistical analyses were performed?; 14.Were clinical signi cance of the ndings are reported?;15.Were participants' drop-outs and withdrawals the reported?;16.Are the author's conclusions appropriate?

Table 3 :
Participants allocation in intervention and non-intervention groups with the duration of intervention and frequency of therapy in weeks

Table 4 :
Follow-up duration and interval post-intervention pain/functional outcome measures * ,Month

Table 5 :
Evidence table for controlled clinical trials

Table 6 :
Descriptions and critique of the reviewed 10 controlled clinical trials (CCTs) Inadequate sample size (pilot study) ✓ Participants were not truly randomized to the three intervention groups ✓ Allocation concealment probably not done ✓ The physiotherapist who assessed the patients was blinded to the study ✓ Blinding of outcome assessment achieved by using patient-reported outcomes on VAS for patellofemoral pain Participants recruited a/c to sample power estimation ✓ Participants were truly randomized by random sequences in a block randomization manner ✓ Allocation concealment evident ✓ Double blinding evident (participants and therapists) ✓ Blinding of outcome assessment evident because the only patient-reported pain and function outcomes collected.Participants truly randomized by random sequence or block randomization ✓ Allocation concealment evident with a random number ✓ Outcome assessors partially blinded to participants (probable detection bias) ✓ Outcome assessment blinded (the only patient-reported pain and function outcomes recorded).Incomplete outcome data managed by intention-to-treat analysis ✓ Outcome assessment blinded (the only patient-reported pain and function outcomes recorded).Incomplete outcome data managed by intention-to-treat analysis ✓ Outcome assessment blinded (the only patient-reported pain and function outcomes recorded).Therapists and assessors blinded to group allocation details ✓ Outcome assessment blinded (the only patient-reported pain and functional outcomes) Participants random allocation not followed ✓ Participants and therapists not blinded ✓ Outcome assessment blinded (the only patient-reported pain and functional outcomes) Participants and therapists blinded RCT, randomised controlled trial; CCT, comparative control trial; RCT-p, randomised controlled pilot study; CS, cohort study; CSr, case series ✓ Participants not allocated to restive groups by random allocation ✓ Participants and therapists not blinded ✓ Outcome assessment blinded (the only patient-reported pain and functional outcomes)

Table 7 :
Evidence table for follow-up studies (cohort, case-control, case series and case reports) 30guresPage 26/30