The efficacy of superoxide dismutase administered using synchronized coronary venous retroperfusion (SRP) on the extent of myocardial reperfusion injury was studied. Eighteen mongrel dogs were divided into three groups. A control group (group A) consisting of six dogs was subjected to 90 minutes of acute myocardial ischemia via balloon occlusion of the left anterior descending coronary artery (LAD) followed by 6 hours of reperfusion following abrupt deflation of the balloon. In the second group (group B) consisting of six dogs, the LAD was occluded for 2 hours followed by 5.5 hours of reperfusion. In this group, SRP was applied for 30 minutes prior to full reperfusion. In the third group (group C) consisting of six dogs, balloon inflation and deflation was performed in the same manner as group B except the administration of 10 mg/kg of superoxide dismutase (SOD) using SRP. During the occlusion of LAD, severe ischemia was detected by blood flow measurement using color microsphere in all groups. After reperfusion regional blood flow expressed as the percent of preocclusion value in the subendocardial area in three groups were 25% (group A), 38% (group B) and 76% (group C), respectively. There were significant differences between the groups (p < 0.05). Left ventricular function was assessed as global ejection fraction. Although occlusion of the LAD resulted in a reduction of the left ventricular ejection fraction with a similar magnitude in all groups (A; 37 +/- 5%, B; 32 +/- 7%, C; 58 +/- 10%), there was an improvement in groups B and C during reperfusion (p < 0.05). Infarct size was assessed by triphenyl tetrazolium chloride staining.(ABSTRACT TRUNCATED AT 250 WORDS)