Both exercise echocardiography and rubidium-82 positron emission tomography are used in the detection and characterization of coronary artery disease. This study compared results of both in 74 patients with known coronary anatomy, by use of exercise echocardiography before and after treadmill exercise and positron emission tomography with intravenous dipyridamole-handgrip stress. Significant (greater than 50%) coronary stenoses were present in 70 patients; exercise echocardiography and positron emission tomography each identified 63 patients (sensitivity 90%). Significant stenoses without previous myocardial infarction were present in 34 patients; 29 (85%) were identified by exercise echocardiography and 28 by positron emission tomography (82%, p = NS). Four patients had no significant coronary disease, and were all identified as normal by both methods. Segments were classified as either normal or showing stress or resting abnormalities, and the diagnoses were compared in the territories of the three major coronary arteries. Results were concordant with respect to the presence or absence of coronary disease in 185 of 222 territories (83%). The remaining 37 regions had abnormalities by exercise echocardiography or positron emission tomography but not both. Stress defects were identified by only one of the tests in 24 areas (in 12 [50%], angiographic findings correlated with positron emission tomography). Resting defects were diagnosed by only one modality in 13 regions (angiographic findings correlated with the results of positron emission tomography in 9 [69%] of these). Both exercise echocardiography and positron emission tomography are sensitive for the identification of coronary artery disease, although on a regional basis, positron emission tomography appears to be more specific for the diagnosis of resting perfusion defects.