This study was designed to assess the effects of blood-free reperfusion with oxygenated or unoxygenated intracoronary perfluorochemical (Fluosol-DA 20%) on myocardial perfusion and to determine its mechanism or mechanisms of limiting no reflow. Twenty-four dogs underwent 90 min of coronary occlusion followed by 210 min of reperfusion and were randomized to either: 1) blood-free reperfusion with intracoronary oxygenated perfluorochemical (20 ml/kg per min) for 20 min followed by blood reperfusion (n = 8); 2) intracoronary unoxygenated perfluorochemical administered as in those treated with oxygenated perfluorochemical (n = 8); and 3) blood reperfusion alone (control) (n = 8). Regional myocardial blood flow was serially determined and global myocardial perfusion was assessed by an intravenous injection of the fluorescent dye (thioflavin-S). Quantitative studies were performed to determine neutrophil infiltration and extent of endothelial injury. Hemodynamic variables were similar in all groups. The zone of impaired perfusion (thioflavin negative), expressed as a percent of the left ventricle, averaged 10 +/- 2%, 6 +/- 2% and 3 +/- 1%, in control and unoxygenated and oxygenated perfluorochemical groups, respectively (control versus oxygenated perfluorochemical p less than 0.004). The reduction in thioflavin-negative area with oxygenated perfluorochemical was associated with a notable recovery of endocardial blood flow (0.97 +/- 0.22 vs. control 0.39 +/- 0.08 ml/min per g; p less than 0.04) at 210 min of reperfusion. The number of capillaries plugged by neutrophils (per 200 capillaries) in thioflavin-negative areas was similar with both oxygenated (5.9 +/- 1.4) and unoxygenated perfluorochemical (5.4 +/- 0.8) treatment and was significantly less than that with the control group (18.9 +/- 3.2, p less than 0.003).(ABSTRACT TRUNCATED AT 250 WORDS)