Recent evidence suggests a cardioprotective effect of adenosine in myocardial ischemia and reperfusion. The present study was undertaken to determine (1) whether adenosine attenuates myocardial stunning, (2) if so, whether the beneficial effect of adenosine takes place during ischemia or after reperfusion, and (3) whether adenosine preconditions against myocardial stunning. A total of 93 dogs were used. In phase A of the study, open-chest dogs undergoing a 15-minute occlusion of the left anterior descending coronary artery followed by 4 hours of reperfusion received an intracoronary infusion of either saline (group I [control], n = 14), 2 mg/min adenosine from 30 minutes before occlusion until 1 hour after reperfusion (group II, n = 10), or 2 mg/min adenosine from 2 minutes before reperfusion until 1 hour after reperfusion (group III, n = 11). Regional myocardial function (assessed as systolic wall thickening) was similar in the three groups at baseline and during ischemia. After reperfusion, dogs treated with adenosine before, during, and after ischemia (group II) demonstrated a significant improvement in the recovery of function that persisted throughout the 4 hours of reperfusion. In contrast, in dogs treated only during the reperfusion period (group III), the recovery of function was not statistically different from that in control dogs. The enhanced recovery effected by adenosine in group II could not be ascribed to differences in ischemic zone size, collateral flow during occlusion, coronary flow after reperfusion, arterial pressure, heart rate, or other hemodynamic variables. In phase B of the study, dogs received an intracoronary infusion of either saline (group IV [control], n = 6) or adenosine (4 mg/min from 40 to 10 minutes before occlusion [group V, n = 6]). Despite pretreatment with adenosine, the recovery of function in group V was indistinguishable from that in the control group. This study demonstrates that (1) continuous administration of adenosine before, during, and after ischemia results in a significant and sustained attenuation of myocardial stunning; (2) this improved recovery of function cannot be attributed to nonspecific variables, such as collateral flow during coronary occlusion, coronary flow after reperfusion, or other hemodynamic factors, and therefore reflects a direct cardioprotective action of adenosine; (3) the protection against stunning is lost or markedly diminished if adenosine is given only at reperfusion; and (4) administration of adenosine before ischemia does not precondition the myocardium against the stunning induced by a 15-minute occlusion.(ABSTRACT TRUNCATED AT 400 WORDS)