Comorbidities Affect the Racial Disparities in the Incidence of Periprosthetic Joint Infection after Total Knee Arthroplasty

BACKGROUND: Racial disparity exists in arthroplasty-related outcomes. However, there is little known about racial disparity in the incidence and management of periprosthetic joint infections (PJIs). This study aimed to evaluate racial differences in the incidence of PJI after total knee arthroplasty (TKA) and in the incidence of above-knee amputation (AKA) after PJI. METHODS: Patients who had a PJI were identified using International Classification of Diseases, 10th edition, and Current Procedural Terminology codes from a national database. The primary outcomes were incidence of PJI and incidence of AKA after PJI stratified by race. There were 175,205 patients who underwent a TKA and were included in the cohort, of which 152,270 (86.9%) were non-Hispanic White, 4,259 (2.4%) were Hispanic, 10,712 (6.1%) were non-Hispanic Black, and 7,964 (4.5%) were non-Hispanic other. High-dimensional propensity score-matched analysis of non-Hispanic White patients was used to evaluate the effect of race on PJI and AKA incidence, controlling for confounding factors with alpha error set to 0,05. RESULTS: Compared to non-Hispanic White patients, non-Hispanic Black patients had a higher incidence of PJI after TKA (hazard ratio [HR] 1.29, 95% confidence interval [CI] 1.18 to 1.42, P < 0.001) in the unadjusted model, but there was no difference in incidence after propensity score matching (HR 1.11, CI 0.97 to 1.27, P = 0.119). There were no differences in incidence of PJI for non-Hispanic other (HR 1.08, CI 0.92 to 1.28, P = 0.350) and Hispanic patients (HR 0.99, CI 0.80 to 1.24, P = 0.950) after propensity score matching. In the subgroup of patients who had a PJI, there was no difference in the incidence of AKA across race, but men were associated with a higher incidence of AKA (HR 1.09, CI 1.09 to 3.43, P = 0.023) after propensity score matching. CONCLUSIONS: Racial disparity exists in the incidence of PJI. However, this observed difference was lost after propensity score matching, suggesting that comorbidities drive the observed difference, not race as an independent factor.