Epicardial adipose tissue volume and the Naples Prognostic Score add incremental value to EuroSCORE II for predicting MACE after TAVR: a retrospective cohort study

Abstract Background Calcific aortic valve disease (CAVD) is a common degenerative valvular disease in the elderly, and transcatheter aortic valve replacement (TAVR) is now its mainstream treatment. Conventional risk scores based on demographic and comorbidity data do not capture local cardiac inflammation or systemic nutritional status, limiting prediction of major adverse cardiovascular events (MACE) after TAVR. Epicardial adipose tissue (EAT) volume and the Naples Prognostic Score (NPS) reflect these dimensions, but their combined incremental value is unknown. We assessed their independent and incremental value for MACE after TAVR, using EuroSCORE II as reference. Methods In this single-center retrospective cohort study, 427 consecutive patients with severe CAVD undergoing TAVR (January 2020 to July 2025) were enrolled. EAT volume was quantified on computed tomography angiography and dichotomized at the cohort median; NPS was derived from routine laboratory indices and categorized as low (0–2) or high (3–4). The endpoint was MACE after TAVR, defined by Valve Academic Research Consortium 3 criteria. Multivariable Cox regression assessed independent prognostic value; nested models were compared by Harrell’s C-index, time-dependent ROC, net reclassification index (NRI), integrated discrimination improvement (IDI) and decision curve analysis. Proportional hazards were assessed. Results During a median follow-up of 24.51 months, 89 patients (20.8%) experienced MACE. After adjustment, both high EAT and high NPS were independent predictors, with average hazard ratios (HRs) of 3.01 (95% confidence interval (CI) 1.80–5.03; P < 0.001) and 1.96 (95% CI 1.27–3.04; P = 0.003). Adding both markers raised the C-index from 0.716 to 0.751 (likelihood-ratio P < 0.001); the 24-month area under the curve was 0.757. Integrated discrimination improvement was 0.068 (95% CI 0.031–0.134; P = 0.011) and net reclassification index was 0.075 (95% CI − 0.081–0.239; P = 0.345). Calibration and sensitivity analyses supported the model; time-varying analysis showed that the effect of high EAT increased over follow-up whereas that of high NPS was confined to the early period. Conclusions Preoperative EAT volume and NPS are independent, complementary predictors of MACE after TAVR. Their combination improves on conventional risk scores and requires only routine preoperative examinations.