A Novel Model to Predict 1-Year Mortality in Elective Transfemoral Aortic Valve Replacement: The TAVR-Risk Score.

Abstract

We aimed to develop and validate an effective prediction model for 1-year mortality risk in elective transfemoral transcatheter aortic valve replacement (TAVR), ie, the TAVR-Risk (TARI) model. TAVR is the primary treatment for patients with symptomatic severe aortic valve stenosis; however, risk assessment tools for longer-term outcomes after TAVR remain scarce. This retrospective cohort study used logistic regression to test univariate and multivariate associations. The German Aortic Valve Registry (GARY) was the derivation (n = 20,704) and the Swedish SWEDEHEART TAVR Registry (SWENTRY) was the validation cohort (n = 3982). The main outcome was the area under the curve (AUC) in the prediction of 1-year mortality. The final model included 12 parameters that were associated with 1-year mortality in a multivariate analysis. The TARI model (AUC, 0.66; 95% confidence interval [CI] 0.65-0.67) performed better as compared with the Society of Thoracic Surgeons (STS) score (AUC, 0.63; 95% CI, 0.62-0.64; P<.001) and logistic EuroSCORE I (AUC, 0.60; 95% CI, 0.59-0.61; P<.001) in the GARY derivation cohort, and discriminated the risk for 1-year mortality better than logistic EuroSCORE I in the SWENTRY validation cohort (AUC, 0.62; 95% CI, 0.60-0.64 vs AUC, 0.59; 95% CI, 0.57-0.61; P=.04). This novel TARI score provides a relatively easy-to-use risk model and offers a superior prediction for 1-year mortality in European TAVR patients.