Are We Treating Similar Patients? Hospital Volume and the Difference in Patient Populations for Total Knee Arthroplasty

Background: Early findings of superior total knee arthroplasty (TICS) outcomes at high volume centers have been thought to have led to distinct referral patterns. However, the effect of these selective referral processes has not been well assessed. Therefore, this study compared the characteristics of primary TICS patients at high, intermediate, and low volume hospitals. Methods: A total of 12,541 primary TICS patients were stratified into risk groups based on age (>65 years), body mass index (>40), and Charlson Comorbidity Index (>= 4). Hospitals were classified as low, intermediate, or high volume based on mean annual TICS volumes (<250, 250-499, and >500). Multivariate logistic regression models evaluated the relationship between baseline patient characteristics and hospital volume. Results: There was a greater percentage of high risk patients at high volume (19%, n = 853) compared to those at intermediate (16%, n = 899) or low volume (17%, n = 444) hospitals (P < .001). Patients with a body mass index >40 were more likely to be treated at high compared to intermediate (odds ratio [OR] 1.4, 95% confidence interval [CI] 1.2-1.6, P < .001) and low volume centers (OR 1.4, 95% CI 1.2-1.7, P < .001). Patients with Charlson Comorbidity Index scores >= 4 were also more likely be treated at high compared to intermediate (OR 1.5, 95% CI 1.3-1.6, P < .001) or low (OR 1.2, 95% CI 1.0-1.4, P = .002) volume centers. Conclusion: This study found that TICS patients at high volume centers have significantly different baseline characteristics compared to those at lower volume centers. This study highlights the importance of considering hospital volume status and the associated disparity in the preoperative risk of patients when comparing primary TICS outcomes between centers. (C) 2020 Elsevier Inc. All rights reserved.