PVCuRe: a machine-learning based tool to predict left ventricular systolic function recovery in patients undergoing ablation for premature ventricular contraction.

BACKGROUND: Ablation of frequent premature ventricular complexes (PVCs) can improve left ventricular ejection fraction (LVEF) in patients with systolic dysfunction, especially in suspected PVC-induced cardiomyopathy. However, many patients fail to normalize LVEF despite successful ablation, and current tools do not reliably distinguish true PVC-induced cardiomyopathy from underlying cardiomyopathy exacerbated by PVCs. OBJECTIVE: To develop and externally validate a machine learning (ML) model using routinely available clinical, echocardiographic, and electrocardiographic variables to predict LVEF recovery after PVC ablation. METHODS: In this retrospective multicenter study, 256 patients with LVEF <50% undergoing successful PVC ablation at three international referral centers were included. Predictors were selected using the Boruta algorithm, and five ML models were trained. Performance was assessed with 10-fold cross-validation and ROC curve analysis. The best-performing model underwent calibration and threshold analysis and was externally validated in an independent cohort from three additional centers. RESULTS: The Random Forest model showed the best performance, with an AUC of 0.88 (95% CI 0.79-0.98) in the internal test set and good calibration (Hosmer-Lemeshow p=0.562). External validation confirmed consistent discrimination (AUC 0.83, 95% CI 0.72-0.95). Key predictors included baseline PVC burden, QRS duration in sinus rhythm, and preprocedural LVEF. CONCLUSION: This ML-based tool, built on widely available variables, accurately estimates the probability of LVEF recovery after PVC ablation and may support clinical decision-making and patient counselling.