Relationship between anesthetic depth and venous oxygen saturation during cardiopulmonary bypass

During cardiopulmonary bypass, mixed venous oxygen saturation (Svo2) is frequently measured to assess circulatory adequacy. Fluctuations in Svo2 not related to patient movement or inadequate oxygen delivery have been attributed clinically to increased cerebral oxygen consumption due to "light" anesthesia. To evaluate the relationship between anesthetic depth and Svo2, we prospectively measured bispectral index (BIS) and Svo2 values in patients undergoing cardiac surgery with cardiopulmonary bypass.

Adults scheduled for cardiac surgery with cardiopulmonary bypass were recruited for this prospective observational study. During bypass, BIS and Svo2 values were recorded every 5 min. To control for confounding effects of changes in other variables known to affect Svo2, temperature, hematocrit, bypass pump flow, muscle relaxant use, and intravenous and inhaled anesthetic doses were also recorded. Only periods with limited variation in other variables affecting Svo2 were analyzed. The relationship between BIS and Svo2 was evaluated using mixed linear regression.

One thousand thirty-four data points were obtained in 41 patients. No overall association between BIS and Svo2 was observed, either in unadjusted analysis or adjusted for covariates. In data points with temperatures less than the median (T < 34.1 degrees C), a significant association between BIS and Svo2 was observed both in unadjusted (beta = -0.32, P = 0.01) and adjusted (beta = -0.27, P = 0.04) analyses.

In patients undergoing cardiopulmonary bypass, we found no overall association between BIS and Svo2. A weak but statistically significant association between BIS and Svo2 was observed in patients with temperatures less than 34.1 degrees C. These data suggest that low Svo2 values on bypass are unlikely to be due to light or inadequate anesthesia. The relationship among temperature, BIS and Svo2 deserves further study.