"Two Strikes, You're Out": Contemporary Revision Patterns Following Instability after Posterior and Anterior Approach Total Hip Arthroplasty.

Abstract

Dislocation remains one of the most challenging complications after total hip arthroplasty (THA). Previous studies suggested that three or more dislocations were a threshold for revision; however, these data predate contemporary implants and modern surgical approaches. We used a contemporary cohort of primary THA patients to evaluate the relationship between dislocation recurrence, timing of instability, and subsequent revision for instability. There were 40,315 patients who underwent primary THA at our institution from 2016 to 2024. Among these patients, those who sustained a dislocation event were included. Keyword and diagnosis code search was performed to identify patients, and charts were reviewed to confirm dislocation. A total of 285 patients (0.7%) were included at a mean 5.7-year follow-up, of whom 242 (84.9%) ultimately underwent revision for instability. Instability was categorized by recurrence (one, two, or ≥ three dislocations) and timing (early ≤ 90) versus late (greater than 90 days)). Multivariable logistic regressions were performed to identify factors independently associated with risk of revision. Revision risk increased sharply with recurrence, rising from 69.1% after one dislocation to 93.9% after two and 98.5% after three or more. Patients who had two dislocations had significantly greater odds of revision than those who had one (odds ratio (OR) 7.70, 95% confidence interval (CI) 3.28 to 20.49, P < 0.001), while risk did not significantly increase beyond the second dislocation. Late dislocations were more likely to result in revision than early events (97.0 versus 74.9%, P < 0.001) and remained independently associated with revision (OR 7.76, 95% CI 2.46 to 35.10, P = 0.002). The association between recurrence, timing, and revision risk was consistent across both approaches. In contemporary THA practice, revision frequently occurred after a second dislocation, and late dislocations were more likely to require revision than early events. Future multicenter studies are needed to further refine instability management strategies.