We studied the diagnostic accuracy of 64-slice computed tomography for the diagnosis of significant coronary artery disease (CAD) compared with conventional coronary angiography (CA) in patients with chronic aortic regurgitation (AR) referred for elective aortic valve surgery. Fifty consecutive patients with chronic AR (38 men, mean age 54 +/- 14 years) scheduled for valve surgery underwent 64-slice computed tomographic (CT) coronary angiography and CA. Significant stenosis was defined as a luminal diameter decrease >50%. Mean heart rate during CT scanning was 65.5 +/- 7.4 beats/min. Mean Agatston score was 136 +/- 278 (range 0 to 1207); prevalence of significant CAD in the study population was 26% (13 of 50 patients). Thirteen of 742 segments (1.8%) in 3 patients were considered nondiagnostic with computed tomography because of motion artifacts (n = 9) or calcium (n = 4). In a patient-based analysis taking nonevaluative segments as falsely positive, sensitivity, specificity, and positive and negative predictive values of computed tomography were 100%, 95%, 87%, and 100%, respectively. Preoperative CA could have been avoided in 70% of patients (35 of 50), CA would have been performed to confirm the CT diagnosis in 26% (13 of 50), and unnecessary CA would have been performed in 4% (2 of 50) on the basis of false-positive CT ratings. In conclusion, 64-slice CT coronary angiography provides high diagnostic accuracy for diagnosing significant CAD in patients with chronic AR and may be used as a filter test before valve surgery to decide whether CA should be performed.