Two-Stage Exchange Protocol for Periprosthetic Joint Infection Following Total Knee Arthroplasty in 245 Knees without Prior Treatment for Infection

Background: For patients undergoing 2-stage exchange for the treatment of periprosthetic joint infection (PJI) following total knee arthroplasty, the long-term risk of reinfection and mechanical failure and long-term clinical outcomes are not well known. The purpose of our study was to determine the long-term clinical results of 2-stage exchange for PJI following total knee arthroplasty. Methods: We identified 245 knees that had undergone total knee arthroplasty and were subsequently treated with 2-stage exchange due to infection during the period of 1991 to 2006; the cohort had no prior treatment for PJI. Major, or 4 of 6 minor, Musculoskeletal Infection Society (MSIS) diagnostic criteria were fulfilled by 179 (73%) of the knees. The cumulative incidence of reinfection and of aseptic revision, accounting for the competing risk of death, were calculated. Risk factors for reinfection were evaluated using Cox proportional hazards regression. Knee Society Score (KSS) values were calculated. The mean age at spacer insertion was 68 years; 50% of the patients were female. The mean follow-up was 14 years (range, 2 to 25 years) following reimplantation. Results: The cumulative incidence of reinfection was 4% at 1 year, 14% at 5 years, 16% at 10 years, and 17% at 15 years. Factors that were predictive of reinfection included a body mass index of ≥30 kg/m2 (hazard ratio [HR], 3.1; p < 0.01), previous revision surgery (HR, 2.8; p < 0.01), and a McPherson host grade of C (HR, 2.5; p = 0.04). The cumulative incidence of aseptic revision for loosening was 2% at 5 years, 5% at 10 years, and 7% at 15 years. Femoral (HR, 5.0; p = 0.04) and tibial (HR, 6.7; p < 0.01) bone-grafting at reimplantation were predictive of aseptic failure. The most common complications were wound-healing issues, requiring reoperation in 12 (5%) of the knees. The rate of death at 2 years following reimplantation was 11%. The mean KSS improved from 45 at PJI diagnosis to 76 at 10 years following reimplantation (p < 0.01). Conclusions: Long-term reinfection rates following 2-stage exchange for PJI after total knee arthroplasty were similar to those of shorter-term reports and were maintained out to 15 years. Mechanical failure rates were low if bone loss was addressed at the time of reimplantation. Improvements in clinical outcomes were maintained at long-term follow-up.