The feasibility, safety and usefulness of dipyridamole echocardiography (two-dimensional echocardiography and 12 lead electrocardiographic monitoring during dipyridamole infusion, up to 0.84 mg/kg over 10 min) were evaluated in 94 asymptomatic patients 8 to 10 days after uncomplicated acute myocardial infarction. The results were compared with those of symptom-limited treadmill exercise testing and correlated with coronary angiography. Two mechanical patterns of positivity of dipyridamole echocardiography could be identified: 1) a new wall motion abnormality confined to the infarct zone or to the adjacent segments (24 patients), and 2) transient remote asynergy (33 patients). The success rate in recording adequate images during dipyridamole infusion was 100%. Interobserver agreement concerning diagnosis occurred in 89 (93%) of 94 patients. Dipyridamole echocardiography was well tolerated; no complication was observed during or after the test. Seventy-three patients underwent coronary angiography within 6 weeks after acute myocardial infarction. Transient remote asynergy on echocardiography was present in 27 of 40 patients with multivessel disease and in none of 33 patients without multivessel disease. Results of treadmill exercise testing were positive in 28 patients with multivessel disease and 8 patients without multivessel disease. Thus, the sensitivity of dipyridamole-induced transient remote asynergy was 68% compared with 52% for treadmill testing (p less than 0.05); specificity was 100% and 72%, respectively (p less than 0.005). The overall accuracy of dipyridamole echocardiography (81%) was higher than that of dipyridamole stress electrocardiography (63%) or exercise electrocardiography (60%) (p less than 0.02). It is concluded that dipyridamole echocardiography is a useful, feasible and inexpensive nonexercise-dependent test for detecting the extent of coronary artery disease early after acute myocardial infarction.