Comparison of transcatheter aortic valve replacement risk score against currently accepted surgical risk models as predictors of 30-day mortality in transcatheter aortic valve replacement

Abstract

Objectives: Aim of the study was to assess the predictive capability of Transcatheter Aortic Valve Replacement Risk Score (TAVR-RS) in comparison with Society of Thoracic Surgeon-Predicted Risk of Mortality (STS PROM) and European System for Cardiac Operative Risk Evaluation II (EuroSCORE II) on 30-day mortality following TAVR. Background: With exponentially increasing use of TAVR, a risk stratification model to accurately predict mortality risks in patients undergoing TAVR is urgently warranted. Methods: Retrospective analysis of 182 TAVRs between 2014 and 2017, 179 by transfemoral, 3 by subclavian approach. Clinical, laboratory and echocardiography variables were collected. The performance of risk models was evaluated using equivalence tests, receiver operating characteristic (ROC) and area under the ROC curve (AUC). Outcome was 30-day mortality prediction. Results: Observed 30-day mortality was 5.49%. TAVR-RS underestimated (4.0%) while surgical models (STS PROM and EuroSCORE II) overestimated mortality, 7.24% and 8.14%, respectively. The TAVR-RS was found to have statistically significant correlation with both individual and group mortalities. AUC was highest for TAVR-RS 0.66 (95%CI: 0.31–0.96), but no difference in 30-day mortality prediction was found in comparison with STS PROM (P = 0.06) or EuroSCORE II (P = 0.2161). Conclusions: The TAVR-RS was a better predictor of both group and individual mortality at 30 days. The outcomes were comparable on pairwise testing against surgical risk models, although TAVR-RS was on verge of significance when compared to STS PROM. This study supports the current dogma that a risk model specifically tailored for TAVR population should be implemented to obtain a better patient selection.