Exploring the Relationship Between Combined, Acetabular, and Femoral Version on Postoperative Outcomes 2 Years After Hip Arthroscopy for Femoroacetabular Impingement Syndrome

Background: Many studies have examined the prevalence of acetabular version (AV) and femoral version (FV) abnormalities and their effect on patient-reported outcomes (PROs) after hip arthroscopy for femoroacetabular impingement syndrome (FAIS), but few have explored the prevalence and influence of combined version (CV) abnormalities. Purpose: To (1) describe the distribution of AV, FV, and CV in the largest cohort to date and (2) determine the relationship between AV, FV, and CV and PROs after hip arthroscopy for FAIS. Study Design: Cohort study; Level of evidence, 3. Methods: Patients were identified who underwent primary hip arthroscopy for FAIS between 2012 and 2018 and received computed tomography containing the pelvis and transcondylar knee slices. AV and FV were assessed on computed tomography, with CV calculated as their sum. PROs were collected preoperatively and 2 years postoperatively. The distributions of AV, FV, CV, and combinations thereof were described based on published ranges. The relationships between version measurements and PROs were analyzed based on a given measurement's distance from a normative value, defined as the mean within the study population (eg, relative acetabular anteversion [AAr]). These relationships were first assessed using univariate natural (restricted) cubic regression splines to account for nonlinearity. The relationship between each relative version group (to the mean) and PROs was then assessed using multiple linear regression, with the other 2 version measurements held constant. Results: In total, 566 patients were included (66.4% female; mean age, 32.6 +/- 11.9 years; mean body mass index, 25.2 +/- 5.1). The mean follow-up was 28.0 months. The mean AV, FV, and CV were 17.1 degrees +/- 5.2 degrees, 12.1 degrees +/- 9.2 degrees, and 29.3 degrees +/- 11 degrees, respectively. Univariate natural cubic regression splines demonstrated that AAr had a significant negative association with 4 of 5 PROs preoperatively and 3 of 5 PROs 2 years postoperatively (P <= .018). All other relative version groups did not have a significant relationship with any PRO at either time point (P > .05). The association between AAr and relatively worse PROs was maintained after controlling for relative FV and CV via multiple linear regression, particularly in patients with relative femoral retroversion. Conclusion: FV and CV, as distinct measures, are not associated with outcomes after hip arthroscopy for FAIS. AAr is associated with worse preoperative status and less improvement at 2 years postoperatively, particularly in patients with relative femoral retroversion.