Dipyridamole thallium 201 scintigraphy (DTS) was used to screen 116 consecutive patients referred for aortic reconstructive surgery for coronary artery disease (CAD). Thallium redistribution was found to have the best statistical correlation with postoperative myocardial infarction (MI). No MIs occurred after aortic operations among 60 patients with normal scans compared with 8 of 31 patients with abnormal ones. The odds of a patient with abnormalities found by DTS having a postoperative MI were 12 times greater than for those with a normal scan. No symptom or combination of symptoms of CAD was as good as an abnormal DTS in identifying patients at risk. The incidence of MI was 7.0% for patients with symptomatic CAD and 8.5% for those who were asymptomatic. Cardiac complications gradually declined as surgeons gained confidence in the use of DTS, to the point where no postoperative MIs occurred during the final year of the study. DTS approaches the ideal preoperative test for CAD in patients with peripheral vascular disease. The test does not require exercise, is minimally invasive, safe, and of sufficient sensitivity to detect myocardial ischemia in the absence of symptoms. Furthermore, it is cost-effective; only those patients with an abnormal scan and an imperative need for aortic surgical treatment need further cardiac evaluation.