Abstract WE517: Epicardial Fat Predicts Heart Failure in Atrial Fibrillation Through Associations with Cardiac Remodeling

Abstract

Background: Atrial fibrillation (AF) frequently coexists with cardiometabolic conditions such as diabetes and heart failure (HF), which increase morbidity and mortality. While body mass index (BMI) is widely used for risk stratification, epicardial adipose tissue (EAT) may better reflect local cardiac fat burden and contribution to adverse remodeling. Objective: To assess whether EAT and EAT indexed to body surface area (iEAT) provide incremental predictive value over BMI for diabetes and heart failure in patients with AF, and to investigate associations between BMI and epicardial fat and echocardiographic markers of cardiac remodeling. Methods: We identified 348 participants from an AF registry with EAT area quantified from a pre-ablation CT scan. Echocardiographic parameters and clinical covariates were collected from the electronic medical record. Multivariable logistic regression models adjusted for age, sex, and race assessed associations between adiposity and outcomes. ROC curves were used to compare predictive performance. Multivariable linear regression analysis assessed echocardiographic parameters, including left ventricular ejection fraction, LV dimensions, LV mass, LA AP diameter, and pulmonary artery systolic pressure (PASP). Results: Among 348 participants, higher BMI, EAT and iEAT were each associated with older age and increased prevalence of cardiometabolic comorbidity. In multivariable models, the highest tertiles of EAT and iEAT were each independently associated with heart failure [EAT OR 4.83 (95% CI 1.82–12.84); iEAT OR 4.40 (95% CI 1.70–11.43)], while BMI was not significant after adjusting for iEAT. ROC analysis showed that iEAT best discriminated heart failure (AUC = 0.74), while BMI best discriminated diabetes (AUC = 0.69). In adjusted models, all three adiposity metrics were significantly associated with higher LV mass; EAT and iEAT were additionally linked to higher pulmonary artery systolic pressure. Conclusion: In patients with atrial fibrillation, epicardial fat is a superior predictor of heart failure compared to BMI, possibly due to its association with adverse cardiac remodeling. In contrast, BMI remains a stronger predictor of diabetes. Incorporating epicardial fat enhances risk stratification for heart failure in patients with atrial fibrillation.