The ability of dipyridamole-thallium-201 imaging to predict in-hospital and late cardiac events when performed very early (62 +/- 21 hours, range 23 to 102) after acute myocardial infarction (AMI) was tested in 50 patients. During hospitalization, 1 patient developed recurrent AMI and 8 patients developed recurrent angina after MI associated with ST-segment depression at 60 +/- 42 hours after the dipyridamole-thallium-201 imaging; of these, 6 required urgent coronary revascularization. No patient died in-hospital. There were no serious adverse effects during the dipyridamole protocol. Using stepwise multivariate logistic regression analysis, the best and only statistically significant predictor of in-hospital ischemic cardiac events was the presence of thallium-201 redistribution within the infarct zone (p = 0.0001). Of 20 patients with infarct zone thallium-201 redistribution, 9 (45%) developed in-hospital ischemic cardiac events compared to 0 of 30 patients without infarct zone thallium-201 redistribution (p less than 0.0001). During a follow-up 12 +/- 7 months after discharge, 3 additional patients with infarct zone thallium-201 redistribution developed recurrent AMI or unstable angina, whereas no patient without infarct zone thallium-201 redistribution developed ischemic cardiac events. These data suggest that dipyridamole-thallium-201 imaging performed very early after AMI may identify a subgroup of patients at high risk for in-hospital and late ischemic cardiac events. Such patients may benefit from early cardiac catheterization and revascularization. Patients without infarct zone thallium-201 redistribution appear to be at very low risk for in-hospital and late ischemic cardiac events and may be candidates for early discharge.