In-hospital Mortality after Septic Revision TKA: Analysis of the New York and Florida State Inpatient Databases

The aims of this study were to investigate (1) in-hospital mortality rates following septic revision total knee arthroplasty (rTKA); (2) compare septic rTKA mortality rates between differing knee revision volume (KRV) hospitals; and (3) identify independent risk factors associated with in-hospital mortality after septic rTKA (up to 2-year follow-up). The Healthcare Cost and Utilization Project State Inpatient Databases of New York and Florida were used to identify septic rTKA, and control groups of aseptic rTKA and primary TKA between 2007 and 2012 via International Classification of Diseases, Ninth Revision codes. Mortality was compared between septic rTKA and aseptic rTKA/primary TKA control groups. Hospital KRV was stratified, and independent risk factors of in-hospital mortality were identified and analyzed using unadjusted and adjusted logistic regression analyses. In this study, 3,531 septic rTKA patients were identified; 105 (3%) patients suffered in-hospital mortality, compared with the control aseptic rTKA (n = 178; 1.7%; p < 0.0001) and primary TKA groups (n = 930; 0.6%; p < 0.0001). Being an octogenarian (adjusted odds ratio [AOR]: 2.361; 95% confidence interval [CI]: 1.514-3.683; p < 0.0002) and having a medium- or high-Elixhauser comorbidity score was associated with in-hospital mortality (AOR: 2.073; 95% CI: 1.334-3.223; p = 0.0012, and AOR: 4.127; 95% CI: 2.268-7.512, p < 0.0001). There were no significant in-hospital mortality rate differences in high- versus medium- versus low-KRV hospitals (1.9 vs. 3.6 vs. 2.9%, respectively, p = 0.0558). Age >81 years and higher comorbidity burden were found to contribute to increased risk of 2-year postoperative mortality after septic rTKA. This association could not be established for hospital KRV.