Interportal Capsulotomy Closure Does Not Improve the Results After Hip Arthroscopy at 1-Year Follow-up: A Prospective Randomized Controlled Study

Abstract

Background:

Management of the capsulotomy at termination of hip arthroscopic procedures in the treatment of femoroacetabular impingement syndrome (FAIS) is debated. Clinical outcomes in favor of capsular closure were demonstrated in a retrospective study; nonetheless, this finding could not be confirmed in a recent randomized, controlled trial comparing capsular closure with unrepaired capsulotomy.

Purpose/Hypothesis:

This randomized, controlled multicenter trial aimed to evaluate the effect of capsular closure on subjective postoperative outcomes and revision rates in patients undergoing hip arthroscopy for FAIS. It was hypothesized was that capsular closure would result in better patient-reported outcome scores compared with no closure of the interportal capsulotomy.

Study Design:

Randomized controlled clinical trial; Level of evidence, 1.

Methods:

All eligible patients with FAIS from 4 surgical centers in Denmark, referred for hip arthroscopy (n = 200), were randomly assigned to either closure or no closure of the interportal capsulotomy at termination of the arthroscopic procedure. The capsular closure was performed with 2 to 3 absorbable sutures, using the “Quebec City Slider” knot technique. The primary outcome was the Copenhagen Hip and Groin Outcome Score (HAGOS) Sport/Recreation subscale. The secondary outcomes were the other HAGOS subscales, International Hip Outcome Tool (iHOT-12), Hip Sports Activity Scale (HSAS), pain during rest and activity measured on a numerical rating scale (NRS), and a measure of the overall well-being by the European Quality of Life–5 dimensions (EQ-5D). Furthermore, the rate of revision procedures and the rate of conversion to total hip replacement were recorded.

Results:

Baseline epidemiological and morphological characteristics were comparable between the treatment groups, except for a higher percentage of women in the capsular closure group (65% vs 48%; P < .05). Both cohorts had significantly improved scores 1 year after surgery, with no difference between the 2 groups, except for the HSAS: HAGOS-Pain (P = .19), HAGOS-Symptoms (P = .33), HAGOS-Activities of Daily Living (P = .21), HAGOS-Sport/Recreation (P = .52), HAGOS-Physical Activity (P = .59), HAGOS-Quality of Life (P = .16), iHOT-12 (P = .15), HSAS (P = .03), EQ-5D (P = .12), NRS-rest (P = .20), and NRS-activity (P = 0.23). Five patients received revision hip arthroscopy after 1 year—1 in the unrepaired versus 4 in the repaired cohort. No patients had conversion to total hip replacement.

Conclusion:

This randomized, controlled multicenter trial showed no effect on clinical outcomes (HAGOS, iHOT-12, EQ-5D, and NRS) 1 year after hip arthroscopy for FAIS of closure of the interportal capsulotomy at termination of surgery.

Registration:

NCT03158454 (ClinicalTrials.gov identifier); 1-10-72-279-16.

There is an ongoing discussion of whether to close or not to close the capsulotomy routinely at termination of arthroscopy in the treatment of patients with femoroacetabular impingement syndrome (FAIS).

The capsule can be involved to obtain good access to the hip joint, from minimal disruption, such as periportal or puncture capsulotomies, to more extensive capsular incisions, such as interportal- and T-capsulotomies. Neglecting to effectively repair the capsular defect has been described to result in iatrogenic capsular insufficiency and joint instability ¹⁹ ; however, the reported number of instability cases is small, ¹⁸ and magnetic resonance imaging (MRI) scans 24 weeks after interportal capsulotomy have demonstrated contiguous healing of the capsule. ²⁸

Significant differences in patient-reported outcome measures (PROMs) in favor of closure of interportal capsulotomy have been reported in a retrospective, comparative study describing Danish patients with FAIS based on the data from the Danish Hip Arthroscopy Registry (DHAR) ²⁶ and in other comparative series. ¹⁹ Of the 4 recently published randomized controlled trials (RCTs) investigating the effects of closing capsular slots, only 1 RCT, involving 131 patients, ¹³ demonstrated a significant advantage of capsular closure. At the 2-year follow-up, this study showed improved PROM scores after the closure of interportal capsulotomies compared with when interportal and T-capsulotomies had been left open. Two RCTs found no significant differences in outcomes at the 2-year follow-up for 84 patients ⁴ and 54 patients, ²⁹ respectively, and 1 RCT ¹ reported no significant difference in outcomes at the 1-year follow-up for 109 patients. All 4 studies utilized interportal capsulotomy, with 1 study also investigating T-capsulotomy ¹³ and another performing T-capsulotomy with the distal limb closed in all cases. ²⁹

As the evidence from these RCTs has been based on data from single-surgeon centers, the data might not be generalizable. Furthermore, the use of different PROMs makes it difficult to compare the studies. In addition, it should be considered when the results are evaluated that only 2 studies^1,4 had used a valid and adequate PROM, the Copenhagen Hip and Groin Outcome Score (HAGOS).^17,32

Therefore, the purpose of this randomized, controlled multicenter trial was to study the effect of capsular closure in patients treated with hip arthroscopic surgery for FAIS in 4 Danish surgical centers. The results were evaluated with the only validated PROM for these patients: The HAGOS. The hypothesis was that patients who had closure of the interportal capsulotomy would have better PROM scores at the 1-year follow-up compared with patients who did not.

Methods

Study Design and Participants

This was a prospective, randomized, controlled, patient-blinded multicenter trial, conducted between June 2017 and August 2022 at 4 hip arthroscopy centers across Denmark. The study protocol ¹¹ was approved by the Central Denmark Region on Biomedical Research Ethics. The study was integrated in DHAR, which was established in 2012 and approved by the Danish Health Authorities (J.nr. 2012-58-0006), as a web-based registry designed for the comprehensive documentation of hip arthroscopy procedures. ¹⁰ All patients between 18 and 50 years, examined by the 4 involved surgeons (O.K., B.L., C.D. and B.M.K.) at the 4 orthopaedic departments in Denmark for hip problems, fulfilling the predetermined inclusion criteria, were invited to participate in this trial.

The inclusion criteria consisted of radiologic and clinical signs of FAIS according to the Warwick Agreement, ¹⁴ including eliciting the patient's typical pain from the hip joint through clinical tests as the Flexion Adduction Internal Rotation test. Radiographic indications of cam morphology were characterized by an alpha angle of >55° observed in cross-table axial radiographs. In addition, the presence of a positive crossover sign and/or a center-edge angle (CE angle) surpassing 39°, suggestive of pincer-type morphology, was identified on standardized anteroposterior pelvic radiographs. ³⁰ Patients were excluded in case of previous hip surgery or previous conditions associated with the hip joints, such as Legg-Calvé-Perthes or slipped capital femoral epiphysis, malignant disease, recent hip or pelvic fractures, arthritis, Ehlers-Danlos, or Marfan disease. Patients were also excluded if they had radiological signs of osteoarthritis, defined as a lateral joint space width of <3 mm or a grade of >1 according to the Tönnis classification, or if they had radiological signs of acetabular dysplasia, defined as a CE angle of <25°. If capsular laxity was demonstrated during patient positioning and the use of mechanical traction was not necessary to have sufficient distraction of the joint, capsular closure or imbrication was performed, and the patient was excluded.

The operative parameters recorded included the number and positioning of portals, duration of surgical procedures encompassing traction intervals, evaluation of labral and cartilage injuries, and details of the surgical techniques, such as anchor selection and the number of anchors employed.

Randomization and Blinding

Before inclusion, each of the 4 centers received 50 envelopes, cluster-randomized in blocks of 4 to 6 envelopes. The content of the envelopes and the sequence of the blocks were blinded to the surgeon. After completion of the regular hip arthroscopy procedure, a randomization envelope containing information on the capsular treatment was opened. Participants were randomized into 2 groups: (1) hip arthroscopy without capsular closure (unrepaired) and (2) hip arthroscopy combined with capsular closure (repaired). The patients were completely blinded to the allocation during the first 5 years. The surgeon noted in the medical files whether patients were randomized to group 1 or 2 only. The statistical researcher (T.G.N.) who collected outcome assessments and analyzed the results was blinded to allocation by concealment of treatment.

Outcomes

The primary outcome measure to evaluate the effect of capsular repair compared with unrepaired capsulotomy was the HAGOS-Sport/Recreation subscale. The secondary outcomes were the other HAGOS subscales: Symptoms, Pain, Activities of Daily Living (ADL), participation in Physical Activities (PA) and hip and/or groin-related Quality of Life (QoL), as well as the International Hip Outcome Tool (iHOT-12), Hip Sports Activity Scale (HSAS), pain during rest and activity measured on a numerical rating scale (NRS) from 0 to 100, and a measure of the overall well-being by the European Quality of life-5 dimensions (EQ-5D). Furthermore, the rate of revision procedures and the rate of conversion to total hip replacement were recorded.

Patient-Reported Outcome Measures

The HAGOS is designed to assess the symptoms, function, and quality of life in younger individuals with hip and/or groin pain. It consists of 6 subscales, each focusing on different aspects of hip and groin function: Pain—assesses the severity and frequency of hip and/or groin pain; Symptoms—evaluates the presence and impact of other symptoms related to hip and/or groin problems; ADL—measures the difficulty experienced during daily activities due to hip and/or groin issues; Sport—assesses the ability to participate in sports and recreational activities; participation in PA—evaluates the frequency and intensity of physical activities undertaken despite hip and/or groin pain; Hip and/or groin-related Quality of Life (QoL)—assesses the overall impact of hip and/or groin problems on the individual's quality of life. Each subscale contains multiple items scored on a Likert scale, with higher scores indicating better hip and/or groin function. The HAGOS was developed with relevant inclusion of patients, and its measurement properties have been validated by item response theory (IRT) mathematics, securing content and construct validity.^17,32

The iHOT-12 is a mathematically shortened ¹⁵ version of the iHOT-33. The 33-item iHOT-33 was developed through interviews with 51 patients with a variety of hip pathologies, securing relevance and coverage (content validity) for these patients. ¹⁷ The measurement properties have not been validated by IRT methods. ¹⁷ Patients were not directly involved in the reduction from 33 to 12 items when the iHOT-12 was developed; thus, it is uncertain whether it has content validity, in particular coverage. Also, the iHOT-12 has not been validated using IRT mathematics. Therefore, the validity of this PROM is questionable. ¹⁷ However, as it is widely used in clinical trials, it was decided to include it as an outcome in the current RCT. The EQ-5D is widely used as an instrument to measure the overall well-being of groups and populations, and it is used to calculate the quality-adjusted life years. ⁶ However, patients were not involved in its development, meaning that it is not known what it measures and whether it is meaningful for people and patients. ⁸ Although not valid, it was included in the study because of its widespread use.

The HSAS was designed by modifying the Tegner activity scale to assess participation in sports activities specifically related to hip function in patients with FAIS. It measures the frequency and intensity of sports activities performed by the individual over a specified time frame, typically the past week or month. The HSAS consists of questions related to sports and physical activities associated with hip function, such as running, cycling, and swimming. Each activity is rated on a scale from 0 to 4 based on the frequency and intensity of participation, with higher scores indicating greater activity levels. The HSAS is used in clinical practice and research settings to evaluate the effect of hip conditions or interventions on sports participation and to monitor changes in activity levels over time; nonetheless, the items were developed without the involvement of patients. ²⁷

Pain levels were assessed using the NRS pain scores at rest and after 15 minutes of walking on a flat surface.

In DHAR, cartilage injuries are categorized based on the International Cartilage Repair Society (ICRS) ³⁵ classification for lesions on the femoral head and the modified Beck classification for injuries on the acetabular side.^2,3,23

The minimal clinically important difference (MCID) was established mathematically as 0.5 SD of the baseline values for the HAGOS subscales, as utilized by Thorborg et al. ³³

Patient acceptable symptom state (PASS) thresholds were determined by employing recently published cutoff values based on a Danish cohort study focusing on FAIS, ²⁰ with the following values for the 6 subscales: Pain (68.75), Symptoms (62.50), ADL (82.50), Sport/Recreation (60.94), PA (43.75), and QoL (42.50).

Surgical Procedure

Hip arthroscopy was conducted with the patient under general anesthesia and in the supine position. The surgical methodologies employed in this study exhibit variability due to differences in portal placement, types of anchors, and bone resection instruments utilized by the 4 surgeons across 4 surgical centers. The most common portals used were the anterolateral and the mid-anterior portals. An interportal capsulotomy was performed with an arthroscopic blade approximately 5 to 10 mm from the labrum, spanning the 11 to 14 o’clock position for the right hip and the 10 to 13 o’clock position for the left hip. T-capsulotomy was not performed. After addressing intra-articular pathologies, closure of the hip capsulotomy was determined based on randomization. In case of capsular closure, the hip was flexed to around 30° while employing 2 to 3 No. 2 Vicryl sutures (Ethicon) utilizing the “Quebec City Slider” knot technique. ²⁴ The sutures were tied with the hip in extension. The number of sutures used to close the capsulotomy was based on the biomechanical study by Chahla et al, ⁷ demonstrating good strength of the capsular closure using either 2 or 3 sutures. All patients received prophylactic antibiotics and prophylactic medication against heterotrophic ossification, consisting of 2 weeks’ nonsteroid anti-inflammatory drug treatment. Antithrombotic prophylaxis was used in high-risk patients. All patients were referred to a standard rehabilitation program supervised by experienced physical therapists. Weightbearing as tolerated was allowed for the first 2 weeks. External rotations beyond 30° and hyperextension were not allowed during the first 3 weeks. The number of executed training sessions, the number of completed repetitions, and the weight load were not measured.

Statistical Analysis (Randomization and Sample Size Calculation)

The planned sample size was based on clinical superiority power calculation determined on the following assumptions: SD = 20 and MCID = 10 for the HAGOS-Sport/Recreation subscale, a power of 0.9, and an alpha value of 5%. This indicated a minimum of 84 participants in each group. To accommodate potential dropouts, the projected sample size for each group was set at 100.

Dichotomous and categorical data were expressed as frequencies (n) with corresponding percentages (%). As continuous data were not normally distributed, data were presented as the median and range or median with interquartile range. The Pearson chi-square test and the Fisher exact test were employed to compare dichotomous and categorical data, contingent upon the sample sizes. The Wilcoxon rank-sum test was used to compare continuous data. Statistical significance was determined at P < .05. Statistical analyses were conducted using Stata 18 software (StataCorp).

Results

Patient Selection and Characteristics

A total of 232 patients were eligible and invited to take part in the trial. Inclusion and randomization details are illustrated in Figure 1.

Figure 1.

A CONSORT flow diagram. CONSORT, Consolidated Standards of Reporting Trials; HA, hip arthroscopy.

A total of 200 patients accepted to participate in this study, signed the consent form, and were randomly allocated to hip arthroscopy with closure (n =100) or without closure (n = 100) of the interportal capsulotomy. In the unrepaired cohort, 3 patients were excluded during surgery because of capsular laxity and the necessity for capsular duplication. In the repair cohort, 2 patients were excluded because of age <18 and age >50 years after allocation, and 2 patients were excluded due to failure of the capsular closure. Table 1 illustrates baseline characteristics in the 2 cohorts. The cohorts were well matched in terms of baseline epidemiological and morphological characteristics, except for a higher proportion of women in the repair cohort (67% vs 49%; P = .02). Furthermore, preoperative hip-related quality of life was generally comparable between the cohorts, except the unrepaired cohort, which demonstrated a higher iHOT-12 score (45.8 vs 38.2; P = .048), a lower NRS-rest pain (26.5 vs 41; P = .04), and a lower NRS-activity pain (34.5 vs 51; P = .03) scores.

Table 1.

Baseline Characteristics of the 2 Cohorts ^a

	Unrepaired	Repaired	P
n	97	96
Age, years	27 (18-48)	29 (18-47)	.28
Sex, male: female	49:48	32:64	.02
Radiological measurements
CE angle (25th-75th)	30 (27-34)	30 (26-34)	.46
Alpha angle (25th-75th)	64 (55-76)	66 (50-77)	.93
Tönnis acetabular index angle (25th-75th)	4 (2-6)	4 (2-6)	.17
Tönnis classification of hip osteoarthritis
Grade 0	80 (89)	72 (84)	.32
Grade 1	10 (11)	14 (16)
JSW, mm
3.1-4	15 (15)	16 (17%)	.82
>4	82 (85)	80 (83)
Ischial spine sign	22 (23)	22 (23)	.97
Crossover sign	44 (45)	39 (41)	.51
Posterior wall sign	14 (14)	11 (11)	.54
PROMs prerandomization (25th-75th)
HAGOS
Pain	60 (42.5-72.5)	55 (37.5-70)	.14
Symptoms	53.6 (35.7-66.1)	53 (35.7-64.3)	.45
ADL	70 (47.5-80)	55 (40-80)	.20
Sport/Recreation	37.5 (23.4-56.2)	36.4 (18.8-56.2)	.22
PA	12.5 (0-25)	12.5 (0-37.5)	0.73
QoL	30 (17.5-42.5)	30.0 (15-40)	.79
iHOT-12	45.8 (28.8-61.5)	38.4 (25.7-51.2)	.048
HSAS	2 (1-4)	2 (1-4)	.60
EQ-5D	0.72 (0.66-0.68)	0.72 (0.63-0.77)	.31
NRS-rest	26.5 (11-51)	41 (17-62)	.04
NRS activity	34.5 (16-65)	51 (25-73)	.03

^aData are presented as median (range) or n (%). Bold P values indicate statistical significance. ADL, activities of daily living; CE, center edge; EQ-5D, European Quality of Life–5 dimensions; HAGOS, Copenhagen Hip and Groin Outcome Score; HSAS: Hip Sports Activity Scale; iHOT: International Hip Outcome Tool; JSW, joint space width; NRS, numeric rating scale; PA, physical activity; PROM, patient-reported outcome measure; QoL, quality of life.

Surgical Arthroscopic Procedures

The surgical procedures are illustrated in Table 2. Capsular repair resulted in a slightly higher, although statistically insignificant, surgical time. Most patients (94% in both cohorts) received a combined correction of both cam and pincer morphologies. Labral repair was performed in almost all cases (97 vs 96 %; P ≥.999). The proportion of high-grade femoral head cartilage injury (ICRS 2-4) was 22% versus 26% (P = .55), and the proportion of patients who demonstrated a high-grade of acetabular cartilage injury (modified Becks 2-4) was 89% versus 92% (P = .47), which was similar between groups. The treatment of cartilage injuries in the acetabulum was also similar and consisted mainly of superficial cartilage debridement (79% vs 81 %; P = .85) with a radiofrequency wand or deeper cartilage resection (19% vs 22 %; P = .82).

Table 2.

Surgical Procedure Data ^a

	Unrepaired	Repaired	P
Mean operation time, min (25th-75th)	66 (42-84)	72 (42-93)	.14
Mean traction time, min (25th-75th)	42 (28-60)	40 (28-55)	.94
Intra-articular surgical procedures
Bony procedures
Isolated pincer resection	5 (5)	5 (5)	≥.99
Isolated cam resection	1 (1)	1 (1)
Combined cam & pincer resection	91 (94)	89 (94)
Labrum procedures
Labrum resection	4 (4)	3 (3)	≥.99
Labrum repair	92 (96)	92 (97)
Labrum reconstruction	0 (0)	0 (0)
Cartilage procedures
Cartilage debridement	77 (79)	80 (81)	.85
Cartilage resection femoral head	2 (2)	3 (3)
Cartilage resection acetabulum	18 (19)	22 (22)
Microfracture femoral head	1 (1)	0 (0)
Microfracture acetabulum	3 (3)	2 (2)

^aData are presented as median (range) or n (%).

Subjective Outcomes

Table 3 summarizes the subjective outcome scores in the cohorts at the 1-year follow-up.

Table 3.

Subjective Outcomes at the 1-Year Follow-up ^a

	Unrepaired	Repaired	P
n	(n = 83)	(n = 82)
PROMs 1-year Follow-up (25th-75th)
HAGOS
Pain	87.5 (70-95)	83.8 (67.5-92.5)	.19
Symptoms	75 (64.3-85.7)	75 (57.1-85.7)	.33
ADL	95 (80-100)	90 (70-100)	.21
Sport/Recreation	71.9 (59.4-87.5)	71.9 (46.9-90.6)	.52
PA	50 (25-75)	50 (12.5-75)	.59
QoL	65 (45-80)	60 (35-75)	.16
iHOT-12	76.5 (60.1-93.2)	73 (50.7-86.8)	.15
HSAS	4 (2-5)	4 (2-4)	.03
EQ-5D	0.82 (0.72-1)	0.78 (0.72-1)	.12
NRS-rest	4 (0-12)	5 (1-17)	.20
NRS-activity	5 (0-15)	5.5 (0-30)	.23
HAGOS PASS
Pain	78 (80.4)	74 (76.3)	.48
Symptoms	78 (80.4)	71 (73.2)	.23
ADL	74 (76.3)	66 (68)	.20
Sport	74 (76.3)	68 (70.1)	.33
PA	62 (63.9)	60 (61.9)	.77
QoL	82 (84.5)	67 (69.1)	.01
HAGOS MCID
Pain	61 (77.2)	58 (74.4)	.68
Symptoms	56 (70.9)	54 (69.2)	.82
ADL	53 (67.1)	50 (64.1)	.69
Sport	57 (72.2)	58 (74.4)	.76
PA	57 (72.2)	50 (64.1)	.28
QoL	65 (82.3)	58 (74.4)	.23

^aData are presented as median (range) or n (%). The bold P values indicate significance. ADL, activities of daily living; EQ-5D, European Quality of Life–5 dimensions; HAGOS, Copenhagen Hip and Groin Outcome Score; HSAS: Hip Sports Activity Scale; iHOT: International Hip Outcome Tool; MCID, minimal clinically important difference; NRS, numeric rating scale; PA, physical activity; PASS, patient acceptable symptom state; PROM, patient-reported outcome measure; QoL, quality of life.

Because of 5 reoperations consisting of revision hip arthroscopy (1 patient in the unrepaired group and 4 patients in the repair group) and not showing up at the 1-year follow-up (n = 23), there were complete outcome data for 83 patients in the unrepaired group and 82 patients in the repaired group. The power calculation indicated 84 participants in each group.

Both cohorts demonstrated significant improvements in all patient-reported outcomes (P < .05). There were no differences between the 2 cohorts after 1 year in HAGOS, iHOT-12, EQ-5D, and NRS scores. A significant difference in HSAS median values was observed because of differences in observations within the groups, but this might not reflect a clinically relevant difference. Because a higher proportion of women were randomized to the repair group, a subgroup analysis of the influence of sex on the HAGOS-Sport/Recreation outcome was performed and demonstrated no significant difference (P = .43).

The percentages of patients achieving the PASS and the MCID in the different HAGOS subscales are summarized in Table 3. For all subscales, the PASS was achieved for more patients in the unrepaired group; however, only significantly for the HAGOS-QoL.

Discussion

The primary finding of this randomized controlled multicenter study in patients with FAIS was that capsular closure of the interportal capsulotomy at termination of hip arthroscopy did not improve hip-related subjective outcomes after 1 year, and the research hypothesis could therefore not be confirmed. Unfortunately, 3 patients were excluded during the operation, but after randomization in the unrepaired cohort and 4 in the repair cohort. Five patients were reoperated on and excluded from the follow-up. Also, 10% and 13% of the patients did not attend the 1-year follow-up in the repaired and unrepaired cohorts, respectively. This was very close to the dropout rate that was considered when planning the study. As a result, the originally estimated statistical power was not achieved. Even though only marginally different and only significant for the 1 HAGOS domain, the PASS was reached slightly more often in the unrepaired group, which supports the conclusion that there is no effect of capsular repair on hip arthroscopy. If a power of 0.8 was used in the calculation above with an SD of 20, an MCID of 10, and an alpha value of 5%, the number of patients needed in each group would have been 64.

The results of this multicenter study involving a large cohort are in line with the conclusions from 3 of 4 previously published RCT studies^1,4,13,29 on capsular closure. They are from single-surgeon centers, including fewer patients than the present study, and there were other differences. In 1 study ²⁹ T-capsulotomy was allowed, but the distal limb was closed in all cases before randomization to repaired or unrepaired interportal capsulotomy. In 1 study, ¹³ unrepaired T-capsulotomies were compared with unrepaired or repaired interportal capsulotomy. Two studies^1,4 used the HAGOS as the primary PROM instrument. Bech et al ¹ demonstrated no differences in HAGOS subscales after 1 year; however, there was a significant difference in the HAGOS-Sport in favor of the unrepaired group at the 3-month follow-up. However, this difference might not be relevant, because many patients are most likely not sports active so early after hip arthroscopic surgery. Bonin et al ⁴ evaluated outcomes after 2 years and found no difference in any of the HAGOS subscales. Bech et al ¹ did not report on the joint space width or the degree of arthritis at baseline, which is relevant, as osteoarthritis negatively affects the outcome. Bonin et al ⁴ excluded the pincer type of FAIS, and as there are sex differences in the occurrence of cam and pincer type FAIS, ³¹ most of their participants were men (88%). One study ²⁹ only used patient-reported outcomes of doubtful validity: the Hip Outcome Score, the modified Harry Hip Score, and ¹⁷ the iHOT-12, which is of particular importance in cases when there is no difference between groups, as the use of inadequate PROMs carries the risk that no difference is a type 2 error. ¹⁶

A positive effect of capsular closure has been reported in comparative series of patients; for instance, by the Danish retrospective study ²⁶ from DHAR, comparing the outcome after capsular closure of an interportal capsulotomy with results in a matched control group who did not have capsular closure. There were superior outcomes after capsular closure in all HAGOS subscales, HSAS, NRS pain, and EQ-5D scores after 1 and 2 years. However, the results were based on historical data, and all patients who had capsular closure had been operated on by the same surgeon, which may be a source of bias. A systematic review—including 8 studies with levels of evidence varying from 2 to 4—found ²¹ similar clinical outcomes as the majority of the RCTs and identical arthroplasty conversion rates between capsular closure and unrepaired capsulotomy groups at a minimum of 5 years after operation for FAIS. However, there was a lower revision rate in favor of capsular closure. It is uncertain whether capsular repair prevents the need for subsequent surgery and total hip replacement, but recently, a multicenter analysis demonstrated positive associations between several patient factors (older age, higher Tönnis grade) and potentially modifiable surgical factors such as labral debridement and capsular nonrepair and the rate of conversion to THA. ⁵

An argument for capsular repair at termination of hip arthroscopy is the risk of microinstability postoperatively. This has been defined as a persistent excessive hip motion that cannot be classified as a dislocation or subluxation. ²² The global expert panel participating in a recent Delphi consensus project defined the major and minor diagnostic criteria for hip microinstability based on patient history, patient examination, and radiological imaging. Some of the major criteria were hip pain, sensation of instability (giving away), connective tissue disorder, positive dial or log roll test, positive anterior apprehension test, excessive internal or external rotation (>60^°), hypermobility (Beighton >5 of 9), radiological signs of dysplasia, and ease of distraction and vacuum sign during hip arthroscopy. Minor factors were female sex, symptoms related to activities, positive response of diagnostic intra-articular injection, acetabular retroversion, and thin anterior capsule measured by preoperative MRI. However, it was concluded that the relative complexity of the diagnostic tool that was suggested for microinstability of the hip highlights the difficulty of establishing the diagnosis in clinical practice. ²² In terms of female sex and Beighton score, Turner et al ³⁴ demonstrated that all female patients, regardless of the Beighton score (BTS), and all patients with a BTS of ≥4 planned for primary hip arthroscopy for FAIS were more likely to have thinner superior hip capsules on MRI. Furthermore, an in vivo study by Metz et al ²⁵ demonstrated, based on MRI measurements, that female sex and decreased thickness of the superior hip capsule were predictors of increased joint distraction at 0 lb and 100 lb of axial traction force. The present study does not answer the question of whether the capsulotomy in patients with laxity should be handled with repair or plication at the end of a hip arthroscopic procedure, because these patients were excluded.

How is the natural course when leaving the capsulotomy unrepaired? This is a relevant question, as the interportal capsulotomy is interrupting one of the main stabilizing structures of the hip joint, the iliofemoral ligament, which restricts both external rotation and anterior translation. ⁹ Imaging-based investigations utilizing MRI on the healing of a small-to-medium-sized interportal capsulotomy, 6 and 24 weeks after surgery ²⁸ demonstrated an insignificant increase in the percentage of healed capsulotomies after capsular repair 6 weeks postoperatively compared with unrepaired capsulotomies. However, at the 24-week follow-up, all capsules appeared continuous with no disparity in capsular thickness between groups. Although clinical or functional outcomes had not been assessed postoperatively, this short-term follow-up study suggested that routine capsular repair may be unnecessary; however, long-term data are not available.

In a long-term follow-up of retrospective data from a group who had capsular plication compared with a group who had no repair of the hip capsulotomy, ¹² the unrepaired cohort demonstrated a decline in PROM scores between the 2- and 5-year follow-up, indicating an advantageous outcome after capsular repair. Hopefully, data on this will be available from the existing RCTs reporting results of longer-term follow-ups.

Strengths and Limitations

A strength of this study is its multicenter design, as these studies include participants from various geographic locations, cultural backgrounds, and healthcare settings, making the findings more representative of the broader population. Furthermore, the results are less likely to be influenced by site-specific practices or regional biases, increasing the relevance of findings across different settings. Pooling of data from various centers minimizes the influence of single-center idiosyncrasies, such as specific surgeon expertise, equipment, or local practices, providing a more balanced perspective. A valid PROM, specifically developed for this type of patient, has been used, minimizing the risk of a type 2 error.

There are some limitations to the study besides the fact that the dropout rate was higher than anticipated, resulting in an analysis of patient data just below the calculated power analysis. First, although the patients were provided with standardized oral and written instructions as per the agreed-upon rehabilitation protocol, the ideal rehabilitation regimen might not have been completed in all patients; however, it is expected that the adherence to rehabilitation is the same in the 2 groups. Second, despite randomization, a significant difference in sex distribution was demonstrated between cohorts, and this can serve as a confounding factor. The repair group included more women, and this might raise a concern, because women are suggested to show microinstability or hip joint laxity more often than men, as discussed above. Beighton scores were not measured in the present study.

Third, the present study only investigated the interportal capsulotomy, and the results are not necessarily representative of other types of capsulotomies, such as T-capsulotomy and periportal capsulotomy. Fourth, the present study could only be blinded to the participants and not the surgeons, and this might potentially have created some bias.

Conclusion

This randomized, controlled multicenter trial showed no effect on clinical outcome 1 year after hip arthroscopy for FAIS of closure of the interportal capsulotomy at termination of surgery.