Background/Aims: Our aim was to examine the significance of thoracic aortic plaque detected by transesophageal echocardiography (TEE) in the prediction of coronary artery disease (CAD) in patients with chronic kidney disease (CKD). Methods: We examined 118 patients (mean age 52 ± 12 years) with CKD and followed them for a mean of 3.4 ± 0.8 years. The study group included 52 predialysis patients with moderate to severe CKD (plasma creatinine ≧200 µmol/l), 32 patients on dialysis treatment, and 34 renal transplant recipients. At baseline, TEE was performed to evaluate thoracic aortic atherosclerosis. CAD was defined by a history of a documented myocardial infarction, a coronary angiogram or a post-mortem autopsy finding showing significant occlusive CAD by the end of the follow-up period. Results: CAD was documented in 31 (26%) of the 118 study patients. The presence of thoracic aortic plaque had a sensitivity of 100% and a specificity of 37% for CAD and the positive and negative predictive values were 36 and 100%, respectively. In the subset of 36 patients with morphological findings of coronary arteries by angiogram or autopsy, the presence of large thoracic aortic plaques (≧3 mm in diameter) had a 73% sensitivity and 90% specificity for significant coronary artery stenosis. The positive and negative predictive values were 95 and 56%, respectively. Conclusion: TEE may be used for detecting high-risk patients with CKD; the absence of thoracic aortic plaque predicted the absence of CAD, and the presence of large aortic plaques predicted significant coronary artery stenosis.