The purpose of this study was to examine the relation between adiponectin levels and subclinical carotid atherosclerosis in patients undergoing coronary artery bypass grafting (CABG). Serum concentrations of adiponectin and carotid intima/media thickness (IMT) were measured in 84 consecutive patients who underwent CABG. Carotid IMT both at the common carotid artery and carotid bulb level was correlated negatively and significantly (r = -0.581 and r = -0.415, respectively, P < 0.01) with the serum concentrations of adiponectin. Linear regression modeling identified adiponectin as the strongest predictive variable for carotid IMT both at the common carotid artery and carotid bulb level (P < 0.001). Stepwise regression analyses also showed that adiponectin was the strongest independent determinant of the carotid IMT both at the common carotid artery and the carotid bulb level (F = 20.215 and F = 19.565, respectively, P < 0.001). The mean number of diseased coronary arteries, mean number of distal anastomoses, cardiopulmonary bypass time, and aortic cross-clamping time did not significantly correlate with the serum concentrations of adiponectin. The findings indicate the presence of an inverse relationship between serum concentrations of adiponectin and subclinical carotid atherosclerosis in patients undergoing CABG. In these patients, the absence of a significant correlation between severity of coronary atherosclerosis and adiponectin might suggest that adiponectin levels may predict the early stages rather than further progression of atherosclerosis.