Adipose tissue surrounding the coronary arteries has been suggested to induce development of atherosclerosis. We explored the relation between epicardial adipose tissue (EAT) volume and coronary atherosclerosis using multislice computed tomography. The study population consisted of 190 patients who had undergone multislice computed tomographic coronary angiography. Coronary artery calcium score was assessed. In addition, patients were classified as having (1) no atherosclerosis, (2) nonobstructive atherosclerosis (luminal narrowing <50%), (3) obstructive atherosclerosis (luminal narrowing >or=50%) in a single vessel, or (4) obstructive atherosclerosis in the left main coronary artery and/or multiple vessels. Cross-sectional tomographic cardiac slices (3.00-mm thickness, range 35 to 40 slices per heart) were traced semiautomatically from the border of EAT below the apex to a point at the center of the left atrium. Tissue with values from -250 to -30 HU was assigned as EAT. EAT volume within the traced area was then automatically quantified. Mean EAT volume was 84 +/- 41 ml. Patients with a coronary artery calcium score >10 had significantly larger average EAT volume (100 +/- 40 ml) compared with patients with calcium scores <or=10 (59 +/- 27 ml, p <0.001). Sensitivity and specificity for prediction of a calcium score >10 were 77% and 70% with a cut-off EAT value of 73 ml. In patients with normal coronaries mean EAT volume (63 +/- 31 ml) was significantly smaller than in patients with atherosclerosis (99 +/- 40 ml, p <0.001). Using a cut-off EAT volume of 75 ml, the sensitivity and specificity for presence of atherosclerosis were 72% and 70%. Interestingly, quantity of EAT did not significantly increase with increasing extent or severity of atherosclerosis. After adjustments for risk factors EAT volume remained a significant predictor of coronary atherosclerosis (p = 0.001). In conclusion, a significant relation was shown between EAT volume and presence of coronary atherosclerosis. Quantification of EAT may be useful to identify patients at risk for coronary artery disease.