Objective. Hypertensive response at peak exercise and blunted blood pressure (BP) recovery, altered BP responses obtained from exercise stress testing, have been suggested as risk factors for future onset of hypertension in previous studies. Epicardial fat, a new cardiometabolic risk factor, has been linked to hypertension in some recent studies. In this study, we tested the primary hypothesis suggesting that the epicardial fat thickness (EFT) is related to altered BP responses to treadmill exercise testing. We also evaluated the sensitivity and specificity of the EFT as a predictor of hypertensive response to peak exercise. Methods. Normotensive subjects underwent to treadmill stress testing and transthoracic echocardiography. Hypertensive response to peak treadmill exercise testing was defined as >= 210/105 mmHg and >= 190/105 mmHg at peak exercise in males and females, respectively. BP recovery index (BPRI) was defined as the ratio of the BP at the 3rd minute of the recovery phase to BP at peak exercise. EFT was measured by echocardiography. Thirty-two subjects with hypertensive response to peak exercise constituted Group 1 and 48 subjects with normal response constituted Group 2. Results. The mean EFT of subjects in Group 1 was significantly higher (8.2 +/- 1.1 mm vs 5.1 +/- 1.5 mm; p = 0.0001) than subjects in Group 2. In correlation analysis performed in Group 1, EFT was found to be significantly correlated with BPRI (r = 0.51, p < 0.003). An EFT of >= 6.5 mm predicted the hypertensive response to peak exercise test with 68.8% sensitivity and 87.5% specificity (receiving operator characteristic area under curve: 0.879, 95% CI 0.793-0.965, p < 0.001). Patients with EFT >= 6.5 mm showed a significantly increased BPRI (0.89 +/- 0.07 vs 0.74 +/- 0.09, p < 0.0001) and peak systolic BP (198.4 +/- 15.3 mmHg vs 169.4 +/- 19.8 mmHg, p < 0.0001). There were significant differences in metabolic equivalents, maximum heart rate, homeostatic model assessment of insulin resistance, high-density lipoprotein-cholesterol, waist circumference and age values between two patients groups dichotomized according to the cut-off value of EFT. BPRI was the only independent variable related to EFT in the multivariate analysis (odds ratio = 1.4, 95% CI 2.75-7.16, p = 0.001). Conclusions. EFT was found to be related to altered BP responses to exercise stress testing. The echocardiographic measurement of EFT may serve as a useful non-invasive indicator of heightened risk of future hypertension.