Regional myocardial glucose utilization was measured with [18F]fluorodeoxyglucose (FDG) and positron emission tomography in normal and postischemic tissue after 3 h of intracoronary balloon occlusion in closed-chest chronically instrumented anesthetized dogs. Estimates of glucose metabolic rates were made using the Sokoloff model, assuming the lumped constant to be unchanged in reperfused tissue. Myocardial sectors were classified as normal, reversibly injured, or infarct containing based on occlusion blood flow images and postmortem histology. Occlusion flow, measured by microspheres, was reduced by 38% in reversibly injured and 74% in infarct-containing sectors, recovering to 91 and 66%, respectively, 1 h postreperfusion. One month postreperfusion, flow was normal in reversibly injured sectors but remained depressed at 60% in infarct-containing sectors. Glucose utilization at baseline was homogeneous, averaging 0.8 mumol.g-1.min-1. After 3 h of reperfusion following occlusion of the left anterior descending coronary artery, regional glucose metabolic rate was increased 60% relative to baseline in normal myocardium but not in postischemic sectors, leading to an enhancement of FDG uptake in normal relative to postischemic myocardium. At 24 h postreperfusion, the glucose metabolic rate decreased in normal remote tissue to 46% of baseline levels, probably reflecting increased plasma free fatty acid levels, but was not significantly altered in reversibly injured myocardium, leading to enhanced FDG uptake in reversibly injured relative to normal myocardium. Subsequently, glucose metabolism in normal and postischemic sectors was not significantly different. Prolonged relative enhancement of glucose metabolic rate in postischemic tissue was found when the glucose metabolic rate in normal myocardium was low. Myocardial glucose utilization correlated with hg, the rate constant for FDG phosphorylation under all conditions (r = 0.88).