Noninvasive ultrasonographic assessment of carotid artery intima-media thickness (IMT) can improve risk stratification for coronary artery disease (CAD) in certain patients. Several measurements have been used to evaluate carotid atherosclerosis by ultrasonography. Although it has been reported that angiographic arterial irregularities correlate with pathologic changes of atherosclerosis and the occurrence of cardiovascular events, only a few studies have assessed carotid arterial wall irregularity by ultrasonography. The purpose of this study was to evaluate the irregularity of IMT quantitatively, and its association with the presence or absence of CAD. The correlation of maximum and mean IMT values, and IMT irregularity with the presence or absence of CAD, was investigated in 90 patients who had undergone coronary angiography. IMT was measured by manual tracking of the far wall of the common carotid arteries, carotid bulbs, and internal carotid arteries. The IMT irregularity was defined as the root mean square (RMS) difference between each IMT and averaged IMT. Multiple logistic regression analysis, after adjustment for coronary risk factors, indicated that the RMS difference was a more accurate predictor of CAD than were the mean or maximum IMT values. These results indicate that the evaluation of IMT irregularity by ultrasonography is a useful predictor for the presence of coronary atherosclerosis.