Objectives: We have assessed clinically systemic tissue oxygenation by monitoring mixed venous oxygen saturation (SvO2) in addition to hematocrit (Hct) during cardiopulmonary bypass. Based on results of experimental studies together with clinical experience, we previously defined the lower limits of the critical range as an Hct of 12% and an SvO2 of 46%. However, these values do not provide direct information about cerebral oxygenation. This study was performed to identify critical values for these variables that would be able to ensure sufficient jugular venous oxygen saturation (SjO2), which reflects global cerebral oxygenation.

Methods: Normovolemic hemodilution was performed in ten rabbits. Hct, SvO2 and SjO2 were measured every 7 minutes. The safety limit for cerebral oxygenation was defined as an SjO2 of 50% based on studies of Croughwell et al. and Cook et al. The limit point was defined as 7 minutes before the time that the SjO2 decreased below 50% for the first time.

Results: Minimal values for Hct and SvO2 to maintain SjO2 at 50% or more during normovolemic normothermic hemodilution, expressed as the 95% confidence interval, were Hct of 7.4% to 10.0% and SvO2 of 41.8% to 51.4%.

Conclusion: Adopting the higher values of these pairs, safety limits for cerebral oxygenation would be an Hct of 10.0% and an SvO2 of 51.4%. In conclusion, our experiments in rabbits suggest new safety limits during normovolemic normothermic hemodilution of Hct of 12% and SvO2 of 52%, taking both whole-body and cerebral oxygenation into consideration.