The aim of this study was to assess the relative prognostic use of clinical risk stratification and intravenous dipyridamole thallium-201 scintigraphy in patients with an intermediate to high prevalence of coronary artery disease (CAD) who have undergone major noncardiovascular surgery, and to assess the effects of medical therapy or coronary revascularization based on the result of this clinical scintigraphic screening on perioperative cardiac morbidity and mortality. Patients (n = 161) with an intermediate to high likelihood of CAD had clinical assessment and intravenous dipyridamole planar thallium-201 testing which was analyzed semiquantitatively. Cardiac events were cardiac death (n = 9), nonfatal myocardial infarction (n = 6), acute pulmonary edema (n = 6), and unstable angina (n = 4). Multiple (> or = 2) clinical risk variables predicted any cardiac event (p = 0.04). Presence of multiple (> or = 2) abnormal thallium-201 segments was the only independent predictor of cardiac death or nonfatal myocardial infarction (p < 0.001), and was the most powerful multivariate predictor of any cardiac event (p < 0.002). Patients with an abnormal dipyridamole thallium-201 scan had a higher risk of perioperative cardiac death, myocardial infarction (18% vs 2%; p < 0.001), or any perioperative cardiac event (27% vs 6%; p < 0.001) when compared with those with a normal scan. Preoperative changes in anti-ischemic therapy or coronary revascularization in 36 of 72 patients with abnormal dipyridamole thallium-201 studies reduced perioperative death or myocardial infarction from 31% to 6% (p < 0.01), and all cardiac events from 47% to 8% (p < 0.001) compared with those in patients without intervention.(ABSTRACT TRUNCATED AT 250 WORDS)