A Novel Predictive Model for Left Atrial Low-Voltage Areas in Paroxysmal Atrial Fibrillation

BACKGROUND Left atrial low-voltage areas (LVAs), indicative of atrial fibrosis and structural remodeling, are present in a subset of patients with paroxysmal atrial fibrillation (AF). OBJECTIVE This study aimed to develop and validate a novel predictive model for identifying LVAs in paroxysmal AF patients. METHODS Paroxysmal AF patients receiving their initial radiofrequency ablation in the Department of Cardiology at Shanghai East Hospital were enrolled. LVAs were defined as regions with a bipolar voltage of less than 0.5 mV during left atrial voltage mapping. Logistic regression analysis was used to identify independent predictors and construct the prediction model. An independent prospective cohort and a multicenter cohort of paroxysmal AF patients were used for validation. RESULTS A total of 383 patients with paroxysmal AF were enrolled respectively, among whom 104 patients (27.2%) had left atrial LVAs. Multivariate logistic regression analysis identified female, prior stroke, left atrial diameter, PR interval, hemoglobin level, and serum creatinine level were independent predictors of LVAs. The HeSLeF-PC score was developed based on these factors, and could predict the presence of left atrial LVAs in paroxysmal AF patients (AUC = 0.810, 95% CI: 0.762-0.859), which was further validated in the prospective cohort (AUC = 0.826, 95% CI: 0.757-0.896) and the multicenter cohort (AUC = 0.767, 95% CI 0.678-0.857). Decision curve analysis confirmed its clinical utility. CONCLUSIONS The HeSLeF-PC score could effectively predict the presence of left atrial LVAs in paroxysmal AF patients and may assist in preprocedural risk stratification and ablation planning.