We examined 17 angina-free patients with left ventricular dysfunction, referred for surgical decision-making, who presented with no or few signs and symptoms of myocardial ischemia according to treadmill stress test. On cardiac catheterization they were affected by severe multi-vessel coronary artery disease; the mean left end-diastolic pressure of this population was 26.3 +/- 5.5 mm Hg (mean +/- SD) and their mean ejection fraction was 27.6 +/- 4.9% (mean +/- SD). They all were investigated for the presence of viable myocardium by the combined assessment of cardiac perfusion and metabolism using single photon emission tomography with [99mTc] labelled hexakis-2-methoxy-isobutyl-isonitrile [99mTc]MIBI/SPET) and positron emission tomography with [18F]-2-fluoro-2-deoxy-D-glucose ([18F]FDG/PET), respectively. Patients were considered for coronary surgery when [18F]FDG was detectable in at least two cardiac segments with wall motion abnormalities and perfusion defects. Nine patients were operated on, six were medically treated and two were scheduled for heart transplantation. We recorded no in-hospital mortality. At a mean follow-up of 28.4 +/- 9.8 (mean +/- SD) months all surgical patients were alive and their NYHA functional classes have improved, except in one case. Among the patients refused for bypass surgery, three are in stable conditions, three have worsened clinical statuses and two died while waiting for heart transplantation. In conclusion, for patients with bypassable coronaries, left ventricular dysfunction and lack of angina, successful coronary revascularization may be predicted by the presence of viable myocardium demonstrated with positron emission tomography.