BACKGROUND: Although epicardial adipose tissue (EAT) volume is associated with increased incidence of coronary artery disease (CAD), its role in myocardial systolic dysfunction is unclear. The present study aimed to identify independent determinants of EAT volume in patients without obstructive CAD, and to evaluate the association between EAT volume (vs other measures of obesity) and myocardial systolic strain analysis. METHODS: We prospectively recruited 130 patients without obstructive CAD on contrast-enhanced cardiac computed tomography imaging and normal left ventricular ejection fraction on 3-dimensional (3D) echocardiography. EAT volume was quantified from cardiac computed tomography imaging, and 3D multidirectional (longitudinal, circumferential, radial, and area) strain were measured. RESULTS: The mean EAT volume was 97.5 ± 43.7 cm. In multivariable analysis, measures of obesity (body mass index [P = 0.007] and waist/hip ratio [P = 0.001]) were independently associated with larger EAT volume. EAT volume was correlated with 3D global longitudinal (r = 0.601; P < 0.001), circumferential (r = 0.375; P < 0.001), radial (r = -0.546; P < 0.001), and area (r = 0.558; P < 0.001) strain. In multivariable analyses, epicardial fat volume was the strongest predictor of 3D global longitudinal (standardized β = 0.512; P < 0.001), circumferential (standardized β = 0.242; P = 0.006), radial (standardized β = -0.422; P < 0.001), and area (standardized β = 0.428; P < 0.001) strain. In contrast, other measures of obesity including body mass index and waist/hip ratio were not independent determinants of 3D multidirectional global strain (all P > 0.05). CONCLUSIONS: EAT volume is independently associated with impaired myocardial systolic function despite preserved 3D left ventricular ejection fraction and absence of obstructive CAD, and might play a significant role in the pathophysiology of diabetic, obesity, and metabolic heart disease.