Background: This study compares oxyhemoglobin dissociation during the nonperfused periods of hypothermic cardioplegic arrest in two blood cardioplegic solutions with different hemoglobin concentrations. The hypothesis is that more oxygen will dissociate from hemoglobin in a blood cardioplegic solution with a higher hemoglobin content than from a cardioplegic solution with a lower hemoglobin content. However, the increment in the volume of oxygen that dissociates from hemoglobin will be less than anticipated by a ratio of hemoglobin concentrations in the cardioplegic solution. Methods and results: Pigs (n = 22) were supported by bypass and subjected to 60 minutes of hypothermic cardioplegic arrest with either a high-hemoglobin (n = 10) or low-hemoglobin (n = 12) blood cardioplegic solution. Aortic root and coronary sinus blood samples were obtained before bypass and 5 seconds after the start of cardioplegic infusions at 20, 40, and 60 minutes of cardioplegic arrest. Oxyhemoglobin dissociation occurred in both experimental groups during the ischemic intervals of cardioplegic arrest. However, there were no significant differences between the high- and low-hemoglobin groups in the arterial-venous oxygen content differences for samples taken after each of the three ischemic intervals (p values: control = 0.78; cardioplegia interval 1 = 0.95; interval 2 = 0.56; and interval 3 = 0.12). Conclusions: The present study emphasizes the inherent limitations of unmodified erythrocyte hemoglobin as an oxygen source in hypothermic alkalotic cardioplegic solutions. These limitations may be obviated by methods that increase the dissolved oxygen content of the cardioplegic solution or methods that decrease the affinity of hemoglobin for oxygen under conditions of hypothermia and alkalosis.