Objectives: The ideal cardioplegic calcium (Ca +2 ) concentration in newborns continues to be debated. Most studies examining cardioplegia calcium concentrations have been done with a nonclinical model (i.e., isolated heart preparation), the results of which may not be clinically applicable, and they have not examined the effect of calcium concentration in a clinically relevant stressed (hypoxic) heart. Methods: Twenty neonatal piglets 5 to 18 days old were placed on cardiopulmonary bypass, and their aortas were crossclamped for 70 minutes with hypocalcemic or normocalcemic multidose blood cardioplegic infusions. Group 1 ( n = 5; low Ca +2, 0.2 to 0.4 mmol/L) and group 2 ( n = 5; normal Ca +2, 1.0 to 1.3 mmol/L) were nonhypoxic (uninjured) hearts. Ten other piglets were first ventilated at an Fio 2 of 8% to 10% (O 2 saturation 65% to 70%) for 60 minutes (i.e., causing hypoxia) and then reoxygenated at an Fio 2 of 100% with cardiopulmonary bypass, which produces a clinically relevant stress injury. They then underwent cardioplegic arrest (as described above) with a hypocalcemic ( n = 5, group 3) or normocalcemic ( n = 5, group 4) blood cardioplegic solution. Myocardial function was assessed with pressure volume loops and expressed as a percentage of control values. Coronary vascular resistance was measured during each cardioplegic infusion. All values were reported as the mean ± standard error. Results: In nonhypoxic hearts (groups 1 and 2), good myocardial protection was achieved at either concentration of cardioplegia calcium, as demonstrated by preservation of postbypass systolic function (104% vs 99% end-systolic elastance), minimally increased diastolic stiffness (152% vs 162%), no difference in myocardial water (78.9% vs 78.9%), and no change in adenosine triphosphate levels or coronary vascular resistance. Low-calcium blood cardioplegia solution repaired the hypoxic reoxygenation injury in stressed hearts (group 3), resulting in no statistical difference in myocardial function, coronary vascular resistance, or adenosine triphosphate levels compared with nonhypoxic hearts (groups 1 and 2). Conversely, when a normocalcemic cardioplegia solution was used in hypoxic hearts (group 4), there was marked reduction in postbypass systolic function (49% ± 4% end-systolic elastance; p < 0.05), increased diastolic stiffness (276% ± 9%; p < 0.05), increased myocardial water (80.1% ± 0.2%; p < 0.05), rise in coronary vascular resistance ( p < 0.05), and lower adenosine triphosphate levels compared with groups 1, 2, and 3. Conclusions: This study demonstrates that, in the clinically relevant, intact animal model, good myocardial protection is independent of cardioplegia calcium concentration in nonhypoxic (noninjured) hearts; hypoxic (stressed) hearts are extremely sensitive to the cardioplegic calcium concentration; and normocalcemic cardioplegia is detrimental to neonatal myocardium subjected to a preoperative hypoxic stress. (J T HORAC C ARDIOVASC S URG 1996;112:1193-201)