Noninvasive cerebral oximeter as a surrogate for mixed venous saturation in children

We evaluated the relationship between regional cerebral oxygen saturation (rSO(2)) measured by near-infrared spectroscopy (NIRS) cerebral oximeter with superior vena cava (SVC), inferior vena cava (IVC), right atrium (RA), and pulmonary artery (PA) saturation measured on room air and 100% inspired oxygen administered via a non-rebreather mask (NRB) in children. Twenty nine pediatric post-orthotopic heart transplant patients undergoing an annual myocardial biopsy were studied. We found a statistically significant correlation between rSO(2) and SVC saturations at room air and 100% inspired oxygen concentration via NRB (r = 0.67, p = 0.0002 on room air; r = 0.44, p = 0.02 on NRB), RA saturation (r = 0.56, p = 0.002; r = 0.56, p = 0.002), and PA saturation ( r = 0.67, p < 0.001; r = 0.4, p = 0.03). A significant correlation also existed between rSO(2) and measured cardiac index (r = 0.45, p = 0.01) and hemoglobin levels (r = 0.41, p = 0.02). The concordance correlations were fair to moderate. Bias and precision of rSO(2) compared to PA saturations on room air were - 0.8 and 13.9%, and they were 2.1 and 15.6% on NRB. A stepwise linear regression analysis showed that rSO(2) saturations were the best predictor of PA saturations on both room air (p = 0.0001) and NRB (p = 0.012). In children with biventricular anatomy, rSO(2) readings do correlate with mixed venous saturation.