Abstract 16865: Epicardial Adipose Tissue Density Predicts the Presence of Atrial Fibrillation and Its Recurrence After Catheter Ablation: Three-Dimensional Reconstructed Image Analysis

Abstract

Introduction: Epicardial adipose tissue (EAT) induces inflammation in the atria and is associated with atrial fibrillation (AF). Several studies have examined the relationship between EAT volume (EAT-V) and density (EAT-D) and the recurrence of AF after catheter ablation. However, the conclusions of these studies have not always been consistent, which might be due to the differences in the location and method of EAT measurement. Methods: This study included 43 consecutive patients who underwent catheter ablation for AF and 30 control patients with no documented history of AF who were matched for baseline characteristics. EAT-V and EAT-D around the entire heart, both atria, left atrium (LA), and right atrium (RA) were measured in detail using reconstructed three-dimensional (3D) EAT images from dual-source computed tomography before ablation. Results: None of the measurements of EAT-V differed significantly between patients with AF and controls or between patients with recurrent AF and those without. On the other hand, all measurements of EAT-D were higher in patients with AF than in controls (both atria, p<0.001; RA, p<0.001; LA, p=0.002). All EAT-D measurements were associated with the presence of AF. Among patients with AF who underwent ablation, all EAT-D measurements were higher in patients with recurrent AF than in those without. The difference was significant for EATRA-D (p=0.032). All atrial EAT-D values predicted recurrent AF (EATRA-D: hazard ratio [HR], 1.208; 95% confidence interval [95%CI], 1.053-1.387; p=0.007; EATLA-D: HR, 1.108; 95% CI,1.001-1.225; p=0.047; EATatrial-D: HR, 1.174; 95% CI,1.040-1.325; p=0.010). According to the ROC curve analysis, EATatrial-D> –80.9 HU is associated with AF recurrence with 100% sensitivity and 58% specificity (AUC, 0.75; p<0.05), and EATRA-D > –80.9 HU could distinguish patients with AF recurrence with 86% sensitivity and 61% specificity (AUC, 0.76; p<0.01). However, EATLA-D > –85.3 HU had low accuracy (sensitivity, 100%; specificity, 33%; AUC, 0.65; p=0.209). Conclusion: For predicting the presence of AF, 3D analysis of EAT-D, rather than EAT-V, is useful. Peri-atrial EAT-D is responsible for recurrent AF after ablation and useful for predicting recurrent AF after catheter ablation.