This study tests the hypothesis that nitric oxide, which is endothelial-derived relaxing factor, produces reoxygenation injury via the L-arginine–nitric oxide pathway in hypoxemic immature hearts when they are placed on cardiopulmonary bypass. Twenty 3-week-old piglets undergoing 2 hours of hypoxemia (oxygen tension about 25 mm Hg) on a ventilator were reoxygenated by initiating cardiopulmonary bypass (oxygen tension about 400 mm Hg). Five animals were not treated, whereas the pump circuit was primed with the nitric oxide–synthase inhibitor N G -nitro- L-arginine methyl ester( L-NAME, 4 mg/kg) in five piglets. L-Arginine, the substrate for nitric oxide, was administered in a fivefold excess (20 mg/kg), together with L-NAME in five piglets ( L-NAME and L-arginine), and given alone in five other piglets ( L-arginine). Five normoxemic, instrumented piglets served as a control group, and five others underwent 30 minutes of cardiopulmonary bypass without preceding hypoxemia. Left ventricular contractility was determined as end-systolic elastance by pressure-dimension loops. Myocardial conjugated dienes were measured as a marker of lipid peroxidation, and the antioxidant reserve capacity (malondialdehyde production in tissue incubated with t-butylhydroperoxide) was measured. Nitric oxide level was determined in coronary sinus plasma as its spontaneous oxidation product, nitrite. Cardiopulmonary bypass per se did not alter left ventricular contractility, cause lipid peroxidation, or lower antioxidant capacity. Reoxygenation without treatment depressed cardiac contractility (end-systolic elastance 38% ± 12% of control*), raised nitric oxide (127% above hypoxemic values), increased conjugated dienes (1.3 ± 0.2 vs 0.7 ± 0.1, control*), and reduced antioxidant reserve capacity (910 ± 59 vs 471 ± 30, control*). Inhibition of nitric oxide production by L-NAME improved end-systolic elastance to 84% ±12%,** limited conjugated diene elution (0.8 ± 0.1 vs 1.3 ± 0.2, no treatment**), and improved antioxidant reserve capacity (679 ± 69 vs 910 ± 59, no treatment**). Conversely, L-arginine counteracted these beneficial effects of L-NAME, because left ventricular function recovered only 24% ± 6%,* conjugated dienes were 1.2 ± 0.1,* and antioxidant reserve capacity was 826 ± 70.* L-Arginine alone caused the same deleterious biochemical changes as L-NAME/ L-arginine and resulted in 60% mortality. The close relationship between postbypass left ventricular dysfunction (percent end-systolic elastance) and myocardial conjugated diene production ( r = 0.752) provides in vivo evidence that lipid peroxidation contributes to myocardial dysfunction after reoxygenation. These results suggest the L-arginine–nitric oxide pathway, which is normally of physiologic benefit by producing endothelial-derived relaxing factor, is involved in the pathogenesis of myocardial reoxygenation injury (* p < 0.05 vs control; ** p < 0.05 vs no treatment, analysis of variance). (J T HORAC C ARDIOVASC S URG 1995;110:1200-11)