Association of epicardial and intramyocardial fat with ventricular arrhythmias

Abstract

Background: Among patients with ischemic cardiomyopathy (ICM) and nonischemic cardiomyopathy (NICM), myocardial fibrosis is associated with an increased risk for ventricular arrhythmia (VA). Growing evidence suggests that myocardial fat contributes to ventricular arrhythmogenesis. However, little is known about the volume and distribution of epicardial adipose tissue and intramyocardial fat and their relationship with VAs.

Objective: The purpose of this study was to assess the association of contrast-enhanced computed tomography (CE-CT)-derived left ventricular (LV) tissue heterogeneity, epicardial adipose tissue volume, and intramyocardial fat volume with the risk of VA in ICM and NICM patients.

Methods: Patients enrolled in the PROSE-ICD registry who underwent CE-CT were included. Intramyocardial fat volume (voxels between -180 and -5 Hounsfield units [HU]), epicardial adipose tissue volume (between -200 and -50 HU), and LV tissue heterogeneity were calculated. The primary endpoint was appropriate ICD shocks or sudden arrhythmic death.

Results: Among 98 patients (47 ICM, 51 NICM), LV tissue heterogeneity was associated with VA (odds ratio [OR] 1.10; P = .01), particularly in the ICM cohort. In the NICM subgroup, epicardial adipose tissue and intramyocardial fat volume were associated with VA (OR 1.11, P = .01; and OR = 1.21, P = .01, respectively) but not in the ICM patients (OR 0.92, P =.22; and OR = 0.96, P =.19, respectively).

Conclusion: In ICM patients, increased fat distribution heterogeneity is associated with VA. In NICM patients, an increased volume of intramyocardial fat and epicardial adipose tissue is associated with a higher risk for VA. Our findings suggest that fat's contribution to VAs depends on the underlying substrate.

Keywords: Contrast-enhanced computed tomography; Epicardial fat; Intramyocardial fat; Ischemic cardiomyopathy; Left ventricular tissue heterogeneity; Nonischemic cardiomyopathy; Ventricular arrhythmia.

Figure 1.

CE-CT derived EAT and intramyocardial fat distributions across patients hearts. Representative image of intramyocardial and epicardial fat distribution in a patient with an ICM (top) and a NICM (bottom). The EAT was measured at a distance of 2.5, 5, 7.5, and 10 mm from the epicardium, and the yellow region indicates the presence of intramyocardial fat.

Figure 2.

CE-CT derived LV tissue heterogeneity assessment in A) ICM patient with high tissue heterogeneity, B) NICM patient with high tissue heterogeneity, C) NICM patient with low tissue heterogeneity, D) ICM patient with low tissue heterogeneity. Image gradients were calculated across the entire myocardium.

Figure 3.

ROC curve comparison. The area under the curve (AUC) for the NICM patients in EAT alone was 0.78 (95% CI 0.62–0.95, cutoff = 0.39), the combination of EAT and intramyocardial fat was 0.81 (95% CI 0.65–0.97, cutoff = 0.33), and for the combination of EAT and tissue heterogeneity was 0.78 (95% CI 0.61–0.96, cutoff = 0.33).