Repeat Two-Stage Revision Arthroplasty for Periprosthetic Joint Infection of the Hip: A Multicenter Study with a Mean 10-Year Follow-Up.

INTRODUCTION Repeat two-stage revision remains a salvage option for managing recurrent periprosthetic joint infection (PJI) of the hip after a failed initial revision. However, the evidence base for this approach is limited. This study aimed to evaluate treatment outcomes and identify risk factors for failure following repeat two-stage revision arthroplasty for hip PJI.

METHODS We conducted a retrospective review of prospectively maintained databases from two tertiary centers to identify patients who underwent repeat two-stage revision for chronic hip PJI between 1991 and 2022. Treatment success was defined using the 2019 Musculoskeletal Infection Society (MSIS) Tier 1 criteria. Kaplan-Meier survivorship analysis and Cox regressions were used to assess outcomes and potential predictors of reinfection. There were 52 patients who met the inclusion criteria. There were 16 patients who did not proceed to second-stage implantation, leaving 36 cases who completed repeat two-stage exchange. Mean follow-up of 10.7 years (range, 0.3 to 33.4).

RESULTS Using MSIS Tier-1 criteria, treatment success was achieved in 21 of 36 patients (58.3%). All-cause implant survival was 66.2% at five years, while survival free from PJI was 76.4%. At final follow-up, recurrent PJI occurred in 11 cases, with further surgery or long-term suppressive antibiotics required in 15 patients. There were no variables, including body mass index, American Society of Anesthesiologists, or McPherson host grade, predictive of failure. A change in causative organism was observed in 47.2% of culture-positive cases.

CONCLUSION Repeat two-stage revision for hip PJI is associated with low success rates, even in carefully selected patients. These findings highlight the complexity of managing recurrent infection and underscore the need for realistic counselling, multidisciplinary care, and consideration of alternative strategies.