Accuracy of in-line venous saturation and hematocrit monitors in pediatric perfusion

Cardiopulmonary bypass (CPB) in neonates and infants routinely employs lower blood flow rates (BFR), temperatures, and hematocrits (Hct) than those typically observed in adult CPB. The purpose of this study was to evaluate the accuracy of three devices available for continuous in-line measurement of venous oxygen saturation (SvO2) and Hct during pediatric CPB. Venous blood samples were obtained over a range of BFR, temperatures, and Hct and analyzed on a Corning 2500 Co-Oximeter and HematoStat C-70 centrifuge. These values were then compared to those measured by the Bentley OxySat SM-0200, the Gish StatSat, and the 3M CDI 100 in-line monitors. Seventy samples were obtained and analyzed using linear regression, paired t-test and residual analysis to establish the reliability and accuracy of each device. The results demonstrate the CDI to be more statistically accurate (p > 0.05) than the Gish and Bentley devices for measuring SvO2 in nearly all circumstances, though all correlated well with the control (r > 0.70). When comparing spun Hct to the CDI and Gish values, the CDI in-line monitor demonstrated a greater reliability to predict actual patient Hct (r > 0.90) than the Gish StatSat (r > 0.60). Residual analysis revealed that even though the Gish StatSat had higher calculated p values (p > 0.05) than the CDI 100 for interpreting Hct, it was shown to display more inconsistent and sporadic values over the ranges of BFR and temperature studied. It is concluded that the CDI 100 proved to be more accurate, reliable, and consistent than the Gish StatSat and the Bentley Oxysat devices in determining SvO2 and Hct over all evaluated parameters in this study.