AHA Scientific Sessions 2026 Abstract · Nov 2, 2026
Prognostic Utility of Contemporary Risk Scores for 1-Year Mortality After Transcatheter Aortic Valve Replacement
By Hoyek K, Haroun E, Arockiam AD et al.
Background: Risk scores are widely used for prognostication in cardiac surgery and are increasingly applied to patients undergoing transcatheter aortic valve replacement (TAVR). However, contemporary surgical and TAVR-specific scores remain incompletely validated in external TAVR cohorts. We evaluated the performance of contemporary risk scores for predicting 1-year mortality after TAVR. Methods: Consecutive patients with native aortic stenosis undergoing TAVR at Cleveland Clinic from 2019-2020 were included. Fifteen contemporary surgical and TAVR-specific risk scores were retrospectively calculated, including EuroSCORE I/II, STS-AVR PROM, STS/ACC TVT, France-2, FTS, German Aortic Valve Score I/II, Observant, TARI, TARIS, TAVI 2, TAVI futility risk models 1 and 3, and UK TAVI. Discrimination for 1-year all-cause mortality was assessed using area under the receiver-operating characteristic curve. Calibration was assessed using observed/expected ratios when applicable. Results: Among 1,035 patients, mean age was 78.9±8.9 years, 430 were women (41.5%), and 998 (96.4%) underwent transfemoral TAVR. Mortality occurred in 5 patients (0.5%) at 30 days, 71 (6.9%) at 1 year, and 240 (23.2%) during a mean follow-up of 2.1±0.9 years. Patients with 1-year mortality had greater symptomatic burden and higher prevalence of cardiovascular comorbidities, including prior cardiac surgery, coronary heart disease, atrial fibrillation/flutter, heart failure, cardiac implantable electronic devices, smoking history, and chronic liver disease. Among surgical scores, STS-AVR PROM had the highest discrimination for 1-year mortality (AUC 0.616, 95% CI 0.546-0.687). Among TAVR-specific scores, TARIS, FTS, and TAVI futility risk model 3 performed best, with AUCs of 0.701, 0.696, and 0.668, respectively. Calibration for 1-year mortality was assessable for FTS, TARI, and TARIS, with observed/expected ratios of 1.35, 0.68, and 0.35, respectively. Conclusion: In this contemporary TAVR cohort, available surgical and TAVR-specific risk scores demonstrated only modest discrimination for 1-year mortality. TAVR-specific models, particularly TARIS, FTS, and TAVI futility risk model 3, performed better than most surgical scores, although calibration was suboptimal. Further refinement and external validation are needed before these scores can be routinely used as stand-alone tools for risk stratification and clinical decision-making after TAVR. Clinical Implication: Improved TAVR-specific risk models incorporating frailty, right-heart function, pulmonary hypertension, anatomic risk features, and non-cardiovascular comorbidities may better identify high-risk patients and support individualized treatment planning.