Anesthetized dogs were studied in two protocols to determine the effect of isoflurane on the extent of myocardial injury resulting from left anterior descending coronary artery (LAD) occlusion. In 22 dogs (11 treated with isoflurane 1% inspired, beginning 1 hr after LAD occlusion, and 11 control) myocardial infarct size measured postmortem after 6 hr of LAD occlusion was significantly less with isoflurane than without it, 23.4 +/- 3.8% vs 36.2 +/- 2.4% of left ventricle; regional myocardial blood flow (RMBF) did not differ between groups and hemodynamic differences were slight. Fifty-two other dogs underwent two 15-min periods of LAD occlusion separated by 1 hr of reperfusion. Without isoflurane (n = 12), hemodynamic, RMBF, and regional metabolic data did not differ between the two occlusion periods. When isoflurane 1.3% inspired was administered during one of the two occlusion periods by random assignment, coronary perfusion pressure, left ventricular stroke work index, and systolic left ventricular pressure decreased more than when isoflurane was not administered. Both oxygen (O2) consumption and supply in ischemic myocardium decreased proportionately during LAD occlusion, but more so with isoflurane. Neither lactate production, potassium release, glucose extraction, nor coronary venous carbon dioxide (CO2) or O2 content differed between LAD occlusion periods with and without isoflurane. Thus, isoflurane decreased the extent of myocardial necrosis produced by LAD occlusion but neither RMBF nor metabolic indications were improved during transitory ischemia.