Determinants of myocardial hemorrhage after coronary reperfusion in the anesthetized dog.

Abstract

Intramyocardial hemorrhage often occurs with reperfusion in experimental acute myocardial infarction and is thought to be associated with extension of necrosis. To determine if hemorrhage was associated with extension of necrosis, 20 anesthetized dogs were reperfused after 6 hours of circumflex coronary artery occlusion and 10 others had control occlusion with no reperfusion. Fifteen of the 20 reperfused dogs had gross hemorrhage and none of the control dogs did. In 12 reperfused and 10 control dogs, radioactive microspheres were injected after coronary occlusion to quantitate collateral flow and in the reperfusion group microspheres were injected to quantitative reflow. Complete flow data were available in eight reperfused and 10 control dogs. Twenty-four hours after coronary occlusion, 1-g segments of infarct and control regions were analyzed for hemorrhage, collateral flow and creatine kinase activity. Serial microscopic examination was performed in eight additional dogs reperfused after 6 hours to determine if hemorrhage occurs into otherwise microscopically normal myocardium. Pathologic examination indicated that hemorrhage did not occur into otherwise microscopically normal myocardium. In dogs with hemorrhage, the extent of hemorrhage was inversely related to myocardial creatine kinase concentration and collateral flow. Mean collateral flow in 47 hemorrhagic segments was 4.5 ml/100 g (4.2% of control). Mean creatine kinase in 36 hemorrhagic segments was 233 mIU/g (21% of control). No hemorrhage was found in areas with collateral flow more than 21% of control or creatine kinase more than 37% of control. Mean reflow in hemorrhagic segments was 78.5% of control flow. These studies indicate that hemorrhage on reperfusion is associated with severe myocardial necrosis and markedly depressed flow before reperfusion and thus occurs only into myocardium already markedly compromised at the time of reperfusion. There is no evidence for hemorrhage into areas that had normal or even moderately depressed flows before reperfusion.