Differences in Risk of Revision and Mortality Between Total and Unicompartmental Knee Arthroplasty. The Influence of Hospital Volume

Abstract

Background: The volume of arthroplasties performed in a hospital by year has an influence on the outcomes of total knee arthroplasty (TKA) and unicompartmental knee arthroplasty (UKA). The aims of this study are (1) to evaluate and compare the risk of revision and mortality of TKA and UKA and (2) to assess whether hospital volume is related to differences in revision risk and mortality. Methods: All individuals recorded in the Catalan Arthroplasty Register between 1/1/2005 and 31/12/2016, diagnosed with osteoarthritis, undergoing cemented TKA and UKA were included. A propensity score matching method was used to obtain comparable cohorts, including 2374 matched prostheses overall. Hospital volume was considered as a dichotomous variable (lower/higher). Descriptive analyses were done before and after matching. Risks of revision and mortality at 30 days, 90 days, 1, 3, and 5 years were calculated and competing risks models and Cox models were fitted. Results: For the population as a whole, higher risk of revision (sub-hazard ratio, 1.98; 95% confidence interval, 1.25-3.17) was found in UKA than in TKA but higher mortality was not. Considering the volume groups, significantly higher risk of revision in UKA than TKA was found in the lower-volume group only (sub-hazard ratio, 1.95; 95% confidence interval, 1.11-3.44). No differences in mortality between TKA and UKA were found in either group. Conclusion: Mortality and revision rates after TKA and UKA at higher-volume hospitals are similar. UKAs performed at lower-volume hospitals have higher revision rates. Volume-dependent specialization thus might help to reduce revision and mortality after surgery.