Hip arthroscopy for femoroacetabular impingement syndrome: preoperative radiographic findings and risk factors for reoperation at 3–6 years postoperatively

Abstract

Introduction: Hip arthroscopy (HA) for femoroacetabular impingement syndrome (FAIS) has been demonstrated to improve outcomes but carries inherent risks of complications and the need for revision arthroscopy or conversion to total hip arthroplasty (THA). The objective of this study is to evaluate preoperative radiographs, patient outcomes, and risk factors for reoperation in patients undergoing HA for FAIS. Methods: A retrospective review of 351 consecutive patients undergoing HA for FAIS from July 2015 to June 2018 was conducted. The primary endpoints were revision hip arthroscopy (RHA) and THA. Univariate and multivariate analysis was performed to evaluate the risk factors for these endpoints. Results: By the end of the study period, 21 (6.0%) patients had undergone an RHA and 27 (7.7%) patients had undergone a THA. When comparing patients who had an RHA to those that did not, there were significant differences in preoperative radiographic characteristics, including smaller sourcil angle (P = 0.012) and lower Tönnis grade (P = 0.038) between groups. These patients were also more likely to have had a postoperative injection in the 1st year (P < 0.001). No factor was independently predictive of revision HA in multivariate analysis. Patients who underwent a subsequent THA were significantly older (P < 0.001) and required more suture anchors for repair (P < 0.001) than those that did not. On preoperative radiographs, Tönnis grade (P = 0.43) skewed higher. The multivariate regression models generated area under the curve of 0.740 and 0.864 for RHA and THA, respectively, indicating acceptable to strong predictive performance. Conclusion: Individual preoperative patient characteristics and radiographic findings are of limited value in assessing risk for RHA or THA after HA for FAIS. However, predictive models incorporating demographic and radiographic findings hold promise for identifying patients at risk for these outcomes. Further study is needed to refine models and assess their value in preoperative patient selection and counseling.