We tested the hypothesis that intracoronary administration of L-arginine (L-Arg), the physiological nitric oxide (NO) precursor, during reperfusion would attenuate postischemic damage by L-Arg NO-pathway mechanisms. Open-chest, anesthetized dogs underwent 60 min of left anterior descending coronary arterial (LAD) occlusion followed by 270 min of reperfusion. Dogs received intracoronary 10 mM L-Arg (n = 9 dogs), intracoronary 10 mM D-arginine (D-Arg, n = 7), or saline vehicle (Veh, n = 10) in the LAD during the first 120 min of reperfusion using an extracorporeal system. After 270 min of reperfusion, segmental systolic and diastolic function were comparably impaired in all three groups. Infarct size (triphenyltetrazolium chloride) expressed as a percentage of the area at risk (An/Ar) was significantly (P < 0.05) reduced in the L-Arg group (17.7 +/- 3.2%) compared with the Veh group (34.8 +/- 2.4%); D-Arg reversed this cardioprotection (48.8 +/- 5.2%, P < 0.05 vs. L-Arg, Veh). Cardiac myeloperoxidase activity, an index of neutrophil accumulation (U/100 mg tissue), was significantly (P < 0.05) lower in the necrotic tissue of the L-Arg group (0.88 +/- 0.26) than in the Veh group (2.46 +/- 0.38). Furthermore, responses to endothelium-dependent vasodilators acetylcholine and A23187 in isolated ischemic-reperfused LAD rings were significantly (P < 0.05) greater in the L-Arg group than in the other two groups. We conclude that intracoronary infusion of L-Arg during the early phase of reperfusion reduced neutrophil accumulation and infarct size and the infusion preserved endothelial function, possibly by increasing NO release or production by the endothelium.