To compare the value of dobutamine and dipyridamole stress echocardiography with exercise stress testing for the diagnosis of coronary artery disease (CAD), 80 patients with chest pain of suspected myocardial ischemic origin (57 with CAD and 23 without significant CAD) underwent dobutamine stress echocardiography (5 to 40 micrograms/kg/min), dipyridamole echocardiography (0.84 mg/kg over 10 minutes) and bicycle exercise electrocardiography after discontinuation of antianginal treatment. Dobutamine echocardiography and exercise testing revealed a higher overall sensitivity than dipyridamole echocardiography (79 vs 60%, p < 0.005; 77 vs 60%, p < 0.05, respectively); this finding was due to a higher dobutamine and exercise sensitivity in 1-vessel CAD (62 vs 33%, p < 0.05 for both tests), whereas sensitivity of the 3 tests was similar in multivessel CAD. Dobutamine and dipyridamole showed a higher specificity than exercise (83 vs 43%, p < 0.01; 96 vs 43%, p < 0.005, respectively). Diagnostic accuracy of dobutamine echocardiography was higher than that of exercise (80 vs 67%, p < 0.05), whereas the difference with dipyridamole (80 vs 70%) was not significant. In the tests that yielded positive results, double product during exercise was significantly higher than that during dobutamine and dipyridamole echocardiography. No major complications occurred during the tests, but adverse effects were more frequent during dobutamine testing. Thus, dobutamine echocardiography may be superior to dipyridamole echocardiography and exercise electrocardiography for the diagnosis of CAD.