Development and Validation of a Prognostic Nomogram for Post-Transcatheter Aortic Valve Replacem

This multicenter cohort study developed and validated a prognostic nomogram to predict heart failure hospitalizations after transcatheter aortic valve replacement (TAVR) in patients with severe aortic stenosis and heart failure with preserved ejection fraction (HFpEF). The cohort included 321 patients treated between 2017 and 2024, split into derivation (n=191) and external validation (n=130) groups, with a median follow-up of 20 months.

Feature selection using Boruta and LASSO, plus variance inflation factor analysis, identified six independent predictors. Eight predictive models were tested; the Cox regression–based nomogram yielded superior external validation performance, with time-dependent AUCs of 0.824 at 12 months and 0.818 at 20 months.

Calibration was strong, and discrimination and reclassification analyses showed the nomogram outperformed the European System for Cardiac Operative Risk Evaluation. An interactive web-based decision support tool was developed to aid point-of-care use.

Uncertainty remains regarding specific predictor details and the generalizability beyond the study populations; data on exact variables and thresholds were not provided in the abstract. Overall, the model demonstrated high predictive accuracy for HF hospitalization in this high-risk post-TAVR HFpEF cohort.