Acetabular Morphologic Characteristics Predict Early Conversion to Arthroplasty After Isolated Hip Arthroscopy for Femoroacetabular Impingement

Abstract

Background: Hip arthroscopy in the setting of dysplasia and borderline dysplasia is controversial. Dysplasia severity is most often defined by the lateral center edge angle (LCEA) but can also be evaluated radiographically by the acetabular inclination (AI). Purpose/Hypothesis: The purpose was to determine the effect of AI on outcomes after isolated hip arthroscopy for femoroacetabular impingement (FAI). We hypothesized that patients with dysplasia would have higher rates of arthroplasty as well as inferior clinical and functional outcomes compared with patients who did not have dysplasia. Study Design: Cohort study; Level of evidence, 3. Methods: A hip arthroscopy registry was reviewed for participants undergoing arthroscopic correction of FAI from February 28, 2008, to June 10, 2013. Participants required a clinical diagnosis and isolated arthroscopic correction of FAI with preoperative imaging and intraoperative cartilage status recorded. AI dysplasia was defined as an AI greater than 10°, LCEA dysplasia as LCEA less than 18°, and borderline LCEA dysplasia as LCEA 18° to 25°. Patients without an acetabular deformity (LCEA 25°-40°; AI <10°) served as a control population. Postoperative variables included patient-reported outcome surveys with conversion to arthroplasty as the primary endpoint. Minimum 5-year outcome scores were obtained for 337 of 419 patients (80.4%) with an average follow-up of 75.2 ± 12.7 months. Results: This study included 419 patients: 9 (2%) with LCEA dysplasia, 42 (10%) with AI dysplasia, and 51 (12%) with borderline dysplasia. The AI but not LCEA was significantly correlated with lower outcome scores on the modified Harris Hip Score ( r = 0.13; P = .01), Non-Arthritic Hip Score ( r = 0.10; P = .04), and Hip Outcome Score–Sports Subscale ( r = 0.11; P = .04). A total of 58 patients (14%) underwent arthroplasty at 31 ± 20 months postoperatively. Patients with LCEA dysplasia had an arthroplasty rate of 56% (odds ratio, 8.4), whereas patients with AI dysplasia had an arthroplasty rate of 31% (odds ratio, 3.3), which was significantly greater than the rate for the nondysplastic cohort (13.5%; P < .0001). Patients with borderline LCEA dysplasia did not have increased rates of arthroplasty. A multivariate analysis found increasing age, increasing AI, Tönnis grade higher than 1, and femoral Outerbridge grade higher than 2 to be most predictive of conversion to arthroplasty. Conclusion: We found that an elevated AI, along with increasing age, Tönnis grade, and femoral Outerbridge grade significantly predict early conversion to arthroplasty after isolated hip arthroscopy. We recommend using the AI, in addition to the LCEA, in evaluating hip dysplasia before hip arthroscopy.