Atherosclerotic lesions in the thoracic aorta detected by transesophageal echocardiography (TEE) have been correlated with coronary artery disease (CAD). We determined whether simple or complex aortic plaques seen on transesophageal echocardiogram correlated with extent, location, and severity of CAD. The study population consisted of 188 patients who underwent TEE and coronary angiography. Atherosclerotic plaques seen on transesophageal echocardiogram were defined as (1) complex plaques in the presence of protruding atheroma ≥4-mm thickness, mobile debris, or plaque ulceration or (2) simple plaques in the absence of findings consistent with complex plaques. Extent of CAD was grouped into 4 groups according to number of coronary vessels with ≥70% stenosis. Numbers of patients with CAD with 0-, 1-, 2-, and 3-vessel disease were 99, 31, 28, and 30 respectively. Compared to patients without CAD, patients with CAD (n = 89) had a significantly greater prevalence of aortic atherosclerotic plaques irrespective of degree of plaque complexity or location (p <0.05). Multivariate analysis found that hypertension (odds ratio 3.0, 95% confidence interval 1.3 to 7.0, p = 0.013), diabetes mellitus (odds ratio 2.4, 95% confidence interval 1.1 to 4.9, p = 0.022), and aortic plaque (odds ratio 3.8, 95% confidence interval 1.8 to 8.2, p = 0.001) were significantly associated with CAD. There was a significant relation between simple and complex aortic plaques with increasing severity of CAD (p <0.001). Multivariate logistic regression analysis showed that complex plaque in the descending aorta (odds ratio 5.4, 95% confidence interval 1.8 to 16.4, p = 0.003) was the strongest predictor of CAD. In conclusion, simple and complex thoracic atherosclerotic plaques detected by TEE are associated with increasing severity of CAD. Complex plaque in the descending aorta was the strongest association with presence of CAD.