The distinction between fibrotic and viable myocardium is a key issue in patients with coronary artery disease and left ventricular dysfunction. Metabolic imaging with positron emission tomography (PET) and labeled tracers, along with the study of myocardial perfusion, is now available to identify hibernating myocardium. However, PET imaging of myocardial metabolism is a high-cost and time-consuming technique, and requires an on-site cyclotron. The aim of this study is to test the reliability of dobutamine echocardiography (DE) compared with PET imaging, for the identification of hibernating myocardium. In 16 patients, scheduled for myocardial revascularization, left ventricular shapes were divided in eight segments both for echocardiographic and nuclear study evaluation. All patients underwent a technetium 99m MIBI single-photon emission tomography stress-rest study of perfusion, a fluorine-18-labeled deoxyglucose (FDG(/PET study of metabolism, and a DE test (baseline, at a 5 micrograms/kg/min infusion of dobutamine for 8 minutes and at a 10 micrograms/kg/min dose for additional 8 minutes). Neither myocardial ischemia nor arrhythmia occurred during the DE test. Baseline echocardiograms showed 90 segments with wall motion abnormalities: wall motion impairment was decreased or reversed in 33 of 90 segments; it remained unchanged in 57 of 90 segments. In 32 of 33 segments considered viable on the basis of DE and in 21 of 57 segments with unchanged kinesis, some degree of FDG was detected. Thus, sensitivity and specificity of DE compared with nuclear studies was 60% and 97% respectively. Moreover, a good correlation and agreement (kappa = 0.51) between DE and the presence of FDG were found. We conclude that DE is a safe and reliable test for the screening of hibernating myocardium in patients with chronic coronary artery disease and left ventricular dysfunction.