Patient data were prospectively collected from patients who underwent hip arthroscopy from February 2008 to December 2018 ( Fig 1 ). Patients were excluded if they had previous ipsilateral hip conditions besides IJD, a lateral center-edge angle (LCEA) <18°, previous ipsilateral hip surgery, or preoperative Tönnis osteoarthritis grade >1. The inclusion criteria for this study were patients undergoing primary hip arthroscopy with a rheumatologist-confirmed diagnosis of IJD (rheumatoid arthritis, systemic lupus erythematosus) at the time of surgery. Additional inclusion criteria included completion of preoperative and a minimum 5-year follow-up of patient-reported outcomes (PROs). The PROs assessed in this study included the modified Harris Hip Score (mHHS), Non-Arthritic Hip Score (NAHS), Hip Outcome Score–Sports-Specific Subscale (HOS-SSS), and visual analog scale (VAS) for pain. ^, Patient satisfaction was rated on a 0 to 10 scale, with 0 being completely dissatisfied with their surgical outcomes and 10 being completely satisfied with their surgical outcome. ^,, Rates of meeting previously defined patient acceptable symptom state (PASS), substantial clinical benefit (SCB) thresholds, and maximal outcome improvement were compared. ^,,,, Minimal clinically important difference (MCID) was calculated via the distribution method using preoperative PROs. Complications were postoperatively documented in clinical visits and follow-up questionnaires.

Patient selection flow diagram.

All surgeries were performed under general anesthesia in the supine position on a traction table by the senior author (B.G.D.). For each arthroscopy, a minimum of 2 portals were used (the anterolateral and mid-anterior portals). Other portals were used on a case-by-case basis. Interportal capsulotomy with a blade was used to access the joint. While in the joint, diagnostic arthroscopy was used to initially assess the labrum, intra-articular cartilage, and ligamentum teres. Labral tears were classified using the Seldes classification system. Femoral head chondral pathologies were noted with the acetabular labrum articular disruption and Outerbridge classifications. ^, Tear of the ligamentum teres were defined using the Domb and Villar classifications. ^,

With the use of a burr under fluoroscopic guidance, acetabular osseous deformities and femoral head-neck osseous deformities were corrected. Considering labral pathology, labral tears were selectively debrided or repaired to maximize stability based on arthroscopic findings. If the labral tear was unrepairable, hamstring allografts were used to perform a labral reconstruction. If full-thickness chondral defects were present, microfractures were performed. Patients who had a presurgical history of painful, internal snapping hip syndrome received an iliopsoas fractional lengthening. Iliopsoas fractional lengthening was performed by cutting the iliopsoas tendon at the muscle-tendon junction, leaving the muscular region intact. Ligamentum teres tears were treated with debridement. With a multifactorial consideration of age, range of motion, generalized ligamentous laxity, acetabular coverage, and arthritic changes, the capsule was repaired or plicated.

Using R (version 4.3.2), propensity matching to a cohort of patients without inflammatory joint disease was used to control for potentially confounding variables. IJD hips were matched to control hips in a 1:3 ratio based on age at the time of surgery, sex, preoperative Tönnis grade, and body mass index (BMI) as matching parameters with an optimal pair matching algorithm. Matching was performed until no further matches could be made.

An a priori power analysis was used to determine the sample size needed in each matched group to achieve 80% power with an α of 0.05. A treatment sample size of 17 was determined to achieve 80% power. For continuous variables, the Shapiro-Wilk test was performed to test for normal distribution, with P >.05 indicating normality. The paired 2-tailed t test or Wilcoxon signed-rank test was used to compare continuous data between groups. To detect significant differences between categorical variables, the χ ² and Fisher exact tests were used. Descriptive statistics, including means, standard deviations, and proportions, were reported when relevant.

Using the distribution method, the MCID for the IJD and control groups was 8.97 and 6.71 for the mHHS, 10.02 and 6.68 for the NAHS, and 12.07 and 9.95 for the HOS-SSS. SCB and PASS values for the mHHS, NAHS, and HOS-SSS were compared to previously defined values from the literature. ^,,

All 27 IJD hips (23 patients) were matched to 81 control hips (79 patients). Table 1 summarizes demographic characteristics of the study groups. Between the IJD group and the control group, there was no significant difference in age at surgery, BMI, and sex ( P >.05). The control group was diagnosed with no IJDs ( P <.01). There were no significant differences in radiographic findings. All patients in each group underwent labral treatment, addressing the cause of FAIS and the presence of ligamentum teres tears ( P <.05) ( Table 2 ).

Preoperatively, the IJD group had 5 (18.5%) patients with Tönnis grade 1, an average LCEA of 32.91° ± 8.53 ° , an average α angle of 59.15° ± 17.60 ° , and an average Tönnis angle of 5.50° ± 4.96°. The control group had 9 (11.1%) patients with Tönnis grade 1, an average LCEA of 31.69° ± 6.17 ° , an average α angle of 58.69° ± 10.85°, and an average Tönnis angle of 4.30° ± 5.11 °. There were no significant differences between the 2 cohorts for preoperative Tönnis grade ( P =.51), LCEA ( P =.42), α angle ( P =.88), and Tönnis angle ( P =.62).

Postoperatively, the IJD group had 4 (15.4%) patients with Tönnis grade 1, an average LCEA of 30.47° ± 6.59 ° , an average α angle of 44.04° ± 5.63 ° , and an average Tönnis angle of 4.88° ± 3.60°. The control group had 7 (10.4%) patients with Tönnis grade 1, an average LCEA of 30.46° ± 5.91 ° , an average α angle of 43.67° ± 5.16°, and an average Tönnis angle of 4.70° ± 4.30 °. There were no significant differences between the 2 cohorts for preoperative Tönnis grade ( P =.76), LCEA ( P =.99), α angle ( P =.77), and Tönnis angle ( P =.84).