Intravenous adenosine decreases infarct size in experimental models of myocardial ischaemia/reperfusion. Ischaemia/reperfusion is associated with a significant increase in cardiac release of endothelin. The effect of cardioprotective doses of adenosine on endothelin release was explored in dogs undergoing 90 min coronary occlusion and 210 min reperfusion. Dogs were assigned to intravenous adenosine in a dose of 0.15 mg.kg-1.min-1 (n = 12) or control (n = 11) during the first 150 min reperfusion. Serial endothelin levels were obtained from the coronary sinus and aortic blood and measured by radioimmunoassay. Adenosine significantly reduced infarct size expressed as a percent of the risk region (28.8 6% v 14.4 2%; p = 0.03). A similar increase in aortic and coronary sinus blood endothelin was observed in both groups during temporary occlusion. A significant transcardiac increase in endothelin levels was present in the control group 60 min after reperfusion whereas no increase occurred in the adenosine treated group control 5.6(SEM 1.9) v adenosine -0.2(1.4) pg.ml-1; p = 0.02. Similarly, intravenous adenosine tended to prevent the increase in myocardial endothelin production seen in control animals during the early reperfusion period control 280(146) v adenosine -57(55) pg.min-1; p = 0.05. Endocardial blood flow in the ischaemic zone 210 min after reperfusion was significantly higher in the adenosine group, at 0.60(0.02) v 0.38(0.02) ml.min-1.g-1; p < 0.05. A significant correlation between endothelin levels, endocardial flow and infarct size was observed in the control group 3 h after reperfusion: r = 0.73, p = 0.02; r = 0.62, p = 0.03 respectively. This relationship was absent in animals treated with adenosine. Intravenous adenosine suppresses the release of endothelin from the previously ischaemic myocardium during the early reperfusion period. This effect may in part contribute to the improvement by adenosine in postischaemic microcirculatory flow resulting in attenuation of the "no reflow" phenomenon.