Efficacy of a home-based hip exercise programme for patients with symptomatic cam femoroacetabular impingement: a randomized controlled trial

Abstract

Aims

We aimed to determine if a home-based hip exercise programme, designed to use strength and flexibility exercises to modify pelvic positioning, is superior to usual care in reducing pain and improving function and quality of life in patients with symptomatic femoroacetabular impingement (FAI).

Methods

This was a single-blind parallel-group randomized controlled trial. Eligible participants were aged ≥ 16 years and diagnosed with symptomatic cam FAI. A sample size of 94 was needed to detect a clinically important difference. Participants were randomly allocated to: 1) an eight-week home-based exercise programme focused on posterior pelvic tilt supervised by a physiotherapist every two weeks in addition to usual care; and 2) usual care. The primary outcome measure was function (patient-reported 33-Item International Hip Outcome Tool (iHOT-33)) and secondary outcome measures were quality of life (EuroQol five-dimension five-level questionnaire (EQ-5D-5L)), and pain (visual analogue scale (VAS)) at baseline, nine weeks, and six months. The Mann-Whitney U test was used to test between-group differences in scores using a two-tailed level of significance of p < 0.05.

Results

A total of 95 participants were recruited (48 intervention, 47 control), 55 were males (57.9%), the mean age was 32 years (SD 8.5), and both groups were similar at baseline. There were no statistically significant differences in the primary outcome measure between groups at nine weeks and six months for function (nine weeks: p = 0.394, 95% CI -15.5 to 6.5; six months: p = 0.526, 95% CI -8.8 to 13.7). There were no statistically significant differences in the secondary outcome measures between groups: pain (nine weeks: p = 0.153, 95% CI -0.4 to 2.4; six months: p = 0.743, 95% CI -1.1 to 1.7), and quality of life (quality of life VAS nine weeks: p = 0.877, 95% CI -10.0 to 6.0; six months: p = 0.269, 95% CI -15.0 to 5.0; index value nine weeks: p = 0.815, 95% CI -0.067 to 0.052; six months: p = 0.217, 95% CI -0.099 to 0.020).

Conclusion

Our results demonstrated no statistical differences in pain, function, and quality of life for participants following an eight-week home-based exercised programme supervised by a physiotherapist compared to usual care.

Cite this article: Bone Jt Open 2025;6(9):1138–1145.

Introduction

Femoroacetabular impingement (FAI) is a prevalent condition where the proximal femur contacts the acetabular rim during the physiological arc of movement.^1,2 Repetitive abnormal contacts can lead to hip joint damage, limited range of motion, functional limitation, and pain significantly limiting the quality of life of the majority of affected patients.^3,4 FAI is recognized as one of the major causes of hip pain, labral tears, and early degenerative hip cartilage changes.⁵

There are two types of FAI morphology: pincer FAI characterized by acetabular over-coverage, and cam FAI, by an insufficient concavity of the femoral head/neck junction.⁶ Both forms of FAI have been linked with the development of osteoarthritis, with the evidence for cam being more conclusive.^5-8 Studies have also shown that the cam lesion forms at the time of skeletal maturity and, once formed, the lesion does not change over time.^9,10 In addition, a prospective cohort of asymptomatic volunteers with no prior history of hip pain demonstrated that those with a cam morphology were at greater risk of developing hip pain at a mean follow-up of four years.⁵ First-line treatments for FAI typically involve conservative treatments (pharmacotherapy, physiotherapy) and when they fail to alleviate symptoms, surgery is a recommended treatment option.¹¹

Although surgery has been associated with improvements in symptoms and hip mechanics,^12-14 it is still unclear what the role of non-surgical management is, as well as the definition of optimal management for cam FAI. Studies have revealed altered hip biomechanics in patients with symptomatic FAI during functional tasks such as walking and squatting.^15-18 Specifically, excessive anterior pelvic tilt, impaired hip extensor contraction time, and decreased hip extensor strength have been observed compared to matched controls.^19-21 Anterior pelvic tilt has been shown to result in earlier occurrence of FAI biomechanically during movement, whereas posterior pelvic tilt results in later occurrence of FAI.²² Evidence also suggests that cam morphology may respond differently to conservative treatments, such as physiotherapy targeting pelvic tilt.^16,23 Cam morphology has been associated with reduced sagittal pelvic motion, which could limit the effectiveness of such exercises and may benefit more from surgical interventions.¹⁶ However, robust studies evaluating the impact of targeted exercise programmes on pain and function in adults with cam FAI are lacking.

The aim of this study was to determine if a home-based hip exercise programme designed to use strength and flexibility exercises to modify pelvic positioning is superior to usual care in reducing pain, and improving function and quality of life, in patients with symptomatic FAI.

Methods

Study design

A single-blind parallel-group randomized controlled trial (RCT) was conducted and reported following the CONSORT guidelines.²⁴ This study was approved by the Ottawa Hospital Institutional Ethics Review Board (20190091-01 H) and is registered on Clinicaltrials.gov (NCT03949127). Due to the COVID-19 pandemic, biomechanical secondary outcome measures were removed from the protocol because of the nature of the assessment (in-person).

Setting and participants

Participants were recruited from three orthopaedic surgeons at a tertiary hospital during routine clinical practice from November 2019 to September 2023. Eligibility participants included patients aged 16 to 50 years with symptomatic cam FAI, at least three months of groin pain which increased with hip rotation, a positive flexion, adduction, internal rotation (FADIR) test, an α angle greater than 60° measured on multiplane imaging (either MRI, CT, or plain radiographs), presence of a labral tear on MRI, and a Tönnis grade of 0 or 1.²⁵ Patients were included even if they had tried physiotherapy previously. Patients were excluded if diagnosed with pincer FAI, health problems interfering with their capacity to accomplish the exercise programme, previously underwent hip surgery on the affected side, and were scheduled for hip surgery within six months. However, patients eligible for surgery planned in more than six months were included.

Prior to the initiation of the RCT, randomization was accomplished in sealed opaque envelopes using a simple 1:1 allocation ratio by a member of the research team not involved in data collection. After receiving the signed consent form, envelopes were opened, and participants were allocated to the intervention or control groups.

Patient characteristics

Between November 2019 and September 2023, a total of 306 patients were assessed for eligibility. Of those, 217 patients were eligible, 95 were enrolled in the study (participants recruited in majority from two of the three orthopaedic surgeons), and 48 were allocated to the eight-week home-based hip exercise programme (Figure 1). One participant in the intervention group never started the eight-week home-based hip exercise programme and two participants (n = one, intervention group; n = one, control group) were withdrawn from the study. By the six-month follow-up, seven participants (14.6%) in the intervention group and 14 participants (29.8%) in the control group were lost to follow-up. No statistically significant differences were found between respondent and non-respondent (age: p = 0.691; sex: p = 0.882; study group: p = 0.106; baseline function, p = 0.434; baseline pain: p = 0.666; baseline quality of life VAS: p = 0.358; baseline quality of life index value, p = 0.354). Most participants were males (57.9%) and the mean age was 32 years (SD 8.5) (Table I).

Fig. 1.

Flow diagram of participants.

Table I.

Characteristics of participants at baseline.

Characteristic	Total (n = 95)	Intervention (n = 48)	Control (n = 47)
Mean age, yrs (SD)	32 (8.5)	31.7 (8.6)	32.4 (8.5)
Male, n (%)	55 (57.9)	28 (58.3)	27 (57.4)
Female, n (%)	40 (42.1)	20 (41.7)	20 (42.6)
Underwent surgery, n (%)	10 (10.5)	5 (10.4)	5 (10.6)

Intervention: eight-week home-based hip exercise programme.

Control: usual care.

Intervention

Participants who were randomized to the intervention group followed an eight-week home-based hip exercise programme supervised by a physiotherapist (LPB) in addition to usual care at our institution in a tertiary hospital (see usual care under control group). The physiotherapist was trained on the exercise programme by the principal investigators (PEB, SP) of the study. The hip exercise programme was based on RCTs and a systematic review that showed feasibility for patients with cam FAI,^26-29 with a focus on improving posterior pelvic tilt. The underlying theory of the hip exercise programme was that using strength and flexibility exercises could modify pelvic positioning and in turn reduce pain, and improve function and quality of life.^30,31 The home-based exercise programme consisted of a total of five exercises: two strength exercises (abdominals and hip extensors), and three stretching exercises (two hip flexors, one hip extensors) (see Supplementary Material for the complete exercise programme).^26-29 Four levels of strengthening exercises were available (easier to harder) (Supplementary Material). At the first physiotherapy telerehabilitation appointment, the physiotherapist determined the appropriate strengthening exercise difficulty by observing: 1) participant’s capacity to biomechanically perform the exercise; and 2) participant’s perceived rate of exertion (from 10 to 15) using the Borg Rating of Perceived Exertion (RPE) scale out of 20.^32-34 Strengthening exercise difficulty was reassessed at each physiotherapy telerehabilitation follow-up by following the same criteria. The three hip flexors stretching exercises stayed the same throughout the exercise programme.

Participants performed exercises every other day for eight weeks to optimize strengthening and stretching efficacy, and participant adherence.^27-29,35 Over the eight weeks, participants met individually with a physiotherapist every two weeks, for a total of four times, using telerehabilitation to demonstrate the exercises, observe participants’ biomechanics while performing exercises, discuss adherence and problems encountered, and adjust exercises accordingly.³⁶ The duration of the first appointment was approximately 30 minutes, while the duration of follow-up appointments was around 15 minutes. The physiotherapist provided the exercise programme to participants by email following each appointment. Our hospital telerehabilitation system was used to ensure feasibility, accessibility, and adherence to the exercise programme.³⁶ A computer or a smartphone was needed to participate in the intervention group; no other equipment was needed to complete the exercise programme. Participants who were randomized to the home-based hip exercise programme were allowed to continue any other treatment regimen, including physiotherapy and exercises, that they were already following.

Control group

Participants who were randomized to the control group followed usual care at our institution in a tertiary hospital, including traditional non-specific physiotherapy (one to two sessions per week for six to eight weeks), advice to avoid or reduce physical activities provoking hip pain, and the use of non-steroidal anti-inflammatory drugs as needed for a minimum of two weeks. Participants did not receive any intervention and only completed outcome measures. Since traditional physiotherapy has been shown to have no effect on patients with symptomatic cam FAI, participants were allowed to continue any other treatment regimen, including physiotherapy and exercises that they were already following.²⁹

Outcome measures

Patient demographic characteristics (age and sex) were collected at baseline. Adherence to the exercise programme was also collected using a self-reported diary (four exercise sessions per week, over eight weeks; total of 32 exercise sessions). The diary instructed participants in the intervention group to checkmark the days the exercise programme was performed. The diary was provided to the research assistant (LPB or AP, see Acknowledgements) at the nine-week follow-up. Patient-reported outcome measures were assessed at baseline, nine weeks, and six months.

The primary outcome measure was function and was assessed with the patient-reported 33-Item International Hip Outcome Tool (iHot-33)³⁷ composed of 33 questions answered on a 10 cm visual analogue scale (VAS), with a higher score representing better function. The iHot-33 was demonstrated to be valid and reliable for a variety of hip conditions.^38,39

Secondary outcome measures were pain and quality of life. Pain was assessed using a self-reported VAS consisting of one question answered on a 10 cm scale with a higher score indicating the worst imaginable pain.⁴⁰ This self-reported approach to assessing pain has been widely used in patients with hip conditions.⁴⁰ Generic quality of life was assessed with the patient-reported EuroQol five-dimension five-level questionnaire (EQ-5D-5L) consisting of 1) self-rated health on a 100 mm VAS (0 indicating worst imaginable health and 100 indicating best imaginable health); and 2) a descriptive system of five dimensions (mobility, self-care, usual activities, pain/discomfort, anxiety/depression) answered on a five-level scale (1: no problems, 2: slight problems, 3: moderate problems, 4: severe problems, and 5: extreme problems) with results converted into in a five-digit number describing patient’s overall quality of life.^41,42 The EQ-5D-5L was demonstrated to be reliable and valid for a variety of musculoskeletal disease.⁴³

Statistical analysis

The sample size was based on the ability to detect a clinically important difference with sufficient power in the primary outcome measure iHOT-33. The minimal clinically important difference (MCID) of the iHOT-33 is 12 in pre-arthritic patients, while the SD is approximately 18.⁴⁴ Using an α level of 0.05, a power of 0.80, and a loss to follow-up of 15%, a sample size of 47 participants per group was needed, or a total of 94 participants.

The analysis was conducted on an intention-to-treat basis. As a result, data from participants who underwent surgery before the six-month follow-up were included in the analysis. Data were analyzed using descriptive statistics and non-parametric tests due to non-normalcy of variables. Baseline representativeness of the sample and comparison of respondent and non-respondent were determined using Fisher’s exact test for dichotomic variables and Mann-Whitney U for continuous variables. Mann-Whitney U test was used to test for differences in outcome scores between the intervention and control groups at the different follow-ups. A two-sided level of significance of 0.05 was used to compare outcomes between groups. Analyses were performed using BlueSky Statistics v10.3.4 (R package) (BlueSky Statistics, USA).

Results

Of the 48 participants randomized to the intervention group, 36 returned the adherence diary (75%). The adherence rate to the exercise programme by participants in the intervention group averaged 84% (SD 18%) (mean number of exercise sessions completed: n = 27 out of 32) (Figure 2).

Fig. 2.

Adherence to the exercise programme.

Outcome measures

No clinically significant baseline differences were found between participants in the intervention and control groupand with participants that underwent surgery before the six-month follow-up (underwent surgery: n = five in the intervention group and n = five in the control group) (Table I and Table II).

Table II.

Outcome measures at baseline, nine weeks, and six months.

Outcome measures	Intervention (n = 48)		Control (n = 47)		p-value	95% CI
	N	Mean (SD)	N	Mean (SD)
Primary outcome
iHOT-33 baseline	48	43.4 (20.1)	46	46.7 (20.2)	0.520	-6.2 to 11.1
9 wks	41	48.5 (21.1)	34	45.3 (23.9)	0.394	-15.5 to 6.5
6 mths	40	45.3 (20.8)	33	48.3 (26.8)	0.526	-8.8 to 13.7
Secondary outcomes
VAS baseline	48	4.7 (2.6)	45	4.4 (2.5)	0.585	-1.4 to 0.9
9 wks	40	3.8 (2.3)	33	4.7 (2.8)	0.153	-0.4 to 2.4
6 mths	41	4.2 (2.4)	32	4.4 (3.1)	0.743	-1.1 to 1.7
EQ-5D-5L VAS baseline	48	66.2 (18.2)	46	69.6 (16.6)	0.403	-5.0 to 10.0
9 wks	41	69.9 (15.3)	34	69.1 (18.4)	0.877	-10.0 to 6.0
6 mths	41	70.6 (18.9)	32	66.3 (20.6)	0.269	-15.0 to 5.0
EQ-5D-5L index value baseline	47	0.663 (0.171)	46	0.696 (0.180)	0.389	-0.025 to 0.071
9 wks	41	0.708 (0.132)	34	0.681 (0.217)	0.815	-0.067 to 0.052
6 mths	40	0.711 (0.161)	31	0.663 (0.208)	0.217	-0.099 to 0.020

EQ-5D-5L, EuroQol five-dimension five-level questionnaire; iHot-33, 33-Item International Hip Outcome Tool; VAS, visual analogue scale.

For the primary outcome, Mann-Whitney U tests revealed no statistically significant differences in function between participants in the intervention group and control group at nine weeks (n = 41 intervention group, n = 34 control group) and six months (n = 40 intervention group, n = 33 control group) (Table II).

For the secondary outcomes, Mann-Whitney U tests revealed no statistically significant differences in pain between the intervention and control groups at nine weeks (n = 40 intervention group, n = 33 control group) and six months (n = 41 intervention group, n = 32 control group). There were no statistically significant differences in quality of life between the intervention and control groups at nine weeks (n = 41 intervention group, n = 34 control group) and six months (n = 41 intervention group, n = 32 control group).

Discussion

Cam FAI as a cause of hip pain and degeneration has been a very active area of research in the last two decades.^45,46 Although other malformations of the hip joint, such as dysplasia, have long been accepted as a cause of hip arthritis leading to joint arthroplasty, one cannot assume the same for all hip deformities.^10,47 This is especially true for cam FAI, where factors such as pelvic incidence, acetabular morphology, and femoral-neck shaft angle have been shown to significantly influence risk of developing hip pain as well as affecting hip kinematics.^48,49 More specifically, Lamontagne et al⁵⁰ showed that corrective surgery for symptomatic cam FAI improved pelvic mobility and was associated with improved quality of life. Given that altered hip biomechanics, such as anterior pelvic tilt, has been identified as a potentially modifiable factor influencing the risk of impingement in patients with symptomatic cam FAI, could non-surgical management also improve quality of life?^15,16 Our trial compared patients with symptomatic cam FAI who were randomized to a control group (usual care) or an eight-week home-based hip exercise programme focusing on posterior pelvic tilt supervised by a physiotherapist, and found no statistically significant differences between groups in reducing pain, or improving function and quality of life. Although we did not control for traditional non-specific physiotherapy, evidence shows that it does not improve symptoms in patients with cam FAI.²⁹

Our study found no statistically significant improvement in function, pain, and quality of life for patients with cam FAI who completed a home-based exercise programme compared to usual care. Our findings are inconsistent with a meta-analysis of five RCTs that demonstrated improved outcomes of patients with FAI following a physiotherapy exercise programme compared to varied control groups (unsupervised physiotherapy, no physiotherapy, passive methods).⁵¹ Current evidence suggests that conservative treatments, such as physiotherapy and tailored exercise programmes, can offer some benefit, but these interventions may not be sufficient for significant symptom relief and functional improvement in most patients. Future research should focus on evaluating comprehensive treatment approaches that address underlying biomechanical dysfunctions in patients with symptomatic cam FAI, if physiotherapy plus surgery results in improved benefits compared to surgery only, and comparing the effect on patient outcomes of an in-person physiotherapy programme compared to a home-based programme.

The limited evidence supporting the effectiveness of exercise programmes compared to usual care in managing cam FAI underscores the need to consider surgical options for patients who do not respond adequately to conservative treatments. A meta-analysis of three RCTs demonstrated that patients with FAI undergoing hip arthroscopy had statistically significant and clinically meaningful reduced symptoms and improved quality of life compared to physiotherapy at one year.⁵² The meta-analysis also demonstrated clinically meaningful improvements in patients randomized to physiotherapy.⁵² These conclusions are further supported by another systematic review of four RCTs.² Given the physical malformations associated with cam FAI and the significantly better outcomes of hip arthroscopy compared to conservative management at one year,^2,52 surgical management, such as hip arthroscopy, may be warranted earlier in the treatment plan. Notably, a study demonstrated that pelvic incidence (fixed angle formed by the complementary sagittal pelvic tilt and sacral slope angles) influences the pelvic mobility of patients with symptomatic cam FAI.²⁰ These patients showed a significantly reduced hip and pelvic range of motion (98°) compared to controls (109°) during squatting.

Although the exercise programme in this RCT was based on other RCTs and a systematic review,^26-29 the programme differs from others by the emphasis on posterior pelvic tilt, tailored progression of strengthening exercises, and the biweekly telerehabilitation appointments to enhance adherence and ensure proper technique. These elements were included in our exercise programme to address specific FAI biomechanical factors and to foster a tailored and feasible home-based exercise programme. Future exercise programme protocols could benefit from incorporating these elements. However, future multicentre studies with larger simple sizes are needed to investigate the effectiveness of our exercise programme, due to the rate of loss to follow-up in the control group of our trial.

Strengths of our RCT are the use of an exercise programme based on other quality RCTs,^26-29 the total number of participants recruited meeting the sample size required to detect a clinically important difference with sufficient power in the iHOT-33, and the adherence rate to the exercise programme from participants who were randomized to the exercise group. There are four limitations to consider. First, by the six-month follow-up, up to 30% of participants in the control group were lost to follow-up, resulting in the loss of the required sample size to detect a clinically important difference with sufficient power in the iHOT-33. This higher-than-expected level of attrition in the control group may have limited our ability to detect smaller, yet clinically meaningful, differences between groups. Second, patients and the physiotherapist (LPB) were not blinded to patients’ allocation due to the nature of the study, but the researcher (SP) conducting the statistical analysis was blinded, and the treating physiotherapist in the intervention group was not involved with patents in the control group. Third, patients were recruited in one hospital setting, but across three orthopaedic surgeons (PEB, SC, GG). Recruitment from a single centre may limit the generalizability of our findings, and future studies should consider multicentre designs to enhance external validity. Finally, our RCT followed a pragmatic approach to reveal real-life results, and data collection did not include the history of previous or current conservative management regimen of participants in the intervention and control groups, which could have influenced the results of the study towards the null hypothesis.

In conclusion, our RCT revealed no statistical differences in function, pain, and quality of life for patients with symptomatic cam FAI following an eight-week home-based exercised programme focused on posterior pelvic tilt supervised by a physiotherapist, compared to usual care. The findings from our study suggest that a home-based hip exercise may have limited impact on cam FAI, supporting the view that cam FAI is primarily a structural deformity and that a surgical correction may be a necessary option to improve quality of life.

Take home message

- Our randomized controlled trial revealed no statistical differences in pain, function, and quality of life for patients with symptomatic cam femoroacetabular impingement (FAI) following an eight-week home-based exercise programme compared to usual care.

- The findings from our study suggest that a home-based hip exercise may have limited impact on cam FAI, supporting the view that cam FAI is primarily a structural deformity and that a surgical correction may be a necessary option to improve quality of life.

Data Availability

The data that support the findings for this study are available to other researchers from the corresponding author upon reasonable request.