Abstract
Background: Models for risk stratification and prediction of outcome, such as the Charlson Comorbidity Index (CCI), the Elixhauser Comorbidity Method (ECM), the 5-factor modified Frailty Index (mFI-5), and the Hospital Frailty Risk Score (HFRS) have been validated in orthopedic surgery. The aim of this study is to compare the predictive power of these models in total hip and knee replacement. ...
Abstract
Background: Models for risk stratification and prediction of outcome, such as the Charlson Comorbidity Index (CCI), the Elixhauser Comorbidity Method (ECM), the 5-factor modified Frailty Index (mFI-5), and the Hospital Frailty Risk Score (HFRS) have been validated in orthopedic surgery. The aim of this study is to compare the predictive power of these models in total hip and knee replacement. Methods: In a retrospective analysis of 8250 patients who had undergone total joint replacement between 2011 and 2019, CCI, ECM, mFI-5, and HFRS were calculated for each patient. Receiver operating characteristic curve plots were generated and the area under the curve (AUC) was compared between each score with regard to adverse events such as transfusion, surgical, medical, and other complications. Multivariate logistic regression models were used to assess the relationship among risk stratification models, demographic factors, and postoperative adverse events. Results: In prediction of surgical complications, HFRS performed best (AUC: 0.719, P < .001), followed by ECM (AUC: 0.578, P < .001), mFI-5 (AUC: 0.564, P = .003), and CCI (AUC: 0.555, P = .012). With regard to medical complications, other complications, and transfusion, HFRS also was superior to ECM, mFI-5, and CCI. Multivariate logistic regression analyses revealed HFRS as an independent risk stratification model associated with all captured adverse events (P <= .001). Conclusion: The HFRS is superior to current risk stratification models in the context of total joint replacement. As the HRFS derives from routinely collected administrative data, healthcare providers can identify at-risk patients without additional effort or expense. (C) 2020 Elsevier Inc. All rights reserved.
Altmetric