Reinfection and Complication Rates After One- Versus Two-Stage Revision Arthroplasty for Hip and Knee Periprosthetic Joint Infection: A Referral Cohort

Introduction & Aim because of the great morbidity and even mortality associated with periprosthetic joint infection (PJI), adequate treatment is imperative. Both one- and two-stage revision arthroplasty are used in clinical practice. Their relative reinfection and complication rates remain debated, especially in an unselected heterogeneous referral population. The aim of our study is to investigate the reinfection and complication rates after one- and two-stage revision arthroplasty for hip and knee PJI in a real-life tertiary referral cohort. Methods a retrospective cohort study was performed including all patients treated with either one- or two-stage revision arthroplasty for a first PJI of the hip or knee, between January 2014 and December 2025. Patients with prior debridement, antibiotics and implant retention, prior revision surgery and immunocompromised patients were included. Patient-related, diagnostic, surgical, microbiological and antibiotic treatment-related parameters were collected. Complications were classified as minor, major and total complications. Results 246 patients were included (150 hip and 96 knee PJIs). No significant differences in reinfection rates were observed between one- and two-stage revision arthroplasty in either the hip or knee cohort at one year, four years or at full follow-up. Two-stage revision arthroplasty of the hip was associated with significantly higher minor and overall complication rates compared with one-stage revision (p=0.002 and p=0.003, OR 4.02, 95% CI [1.57–10.26] and OR 3.61, 95% CI [1.51-8.34]). However, the two-stage hip group comprised a higher proportion of revision arthroplasties as the index procedure. In the knee cohort, complication rates did not differ significantly. Prior revision surgery of the hip was associated with a significantly higher reinfection risk over time (log-rank p = 0.011) and at one year initial polymicrobial infections had the poorest survival (log-rank p=0.045). The most common microorganisms in our cohort were gram-positive, particularly S. aureus and S. epidermidis, followed by gram-negative microorganisms. Conclusion in this real-life tertiary referral cohort, including patients with known PJI risk factors, one- and two-stage revision showed similar reinfection rates in both hip and knee PJI, while two-stage hip revision was associated with more complications. Hip and knee periprosthetic joint infections should be analysed separately, as they differ substantially in patient population and microbiological patterns.