Comparison of different near-infrared spectroscopic cerebral oxygenation indices with central venous and jugular venous oxygenation saturation in children

Background: We compared two different near-infrared spectrophotometers: cerebral tissue oxygenation index (TOI) measured by NIRO 200 and regional cerebral oxygenation index (rSO(2)) measured by INVOS 5100 with venous oxygen saturation in the jugular bulb (SjO(2)) and central SvO(2) from the superior caval vein (SVC) during elective cardiac catheterization in children. Methods: A prospective observational clinical study in 31 children with congenital heart defects in a catheterization laboratory was undertaken. TOI was compared with SjO(2) in the left jugular bulb and with SvO(2). rSO(2) was compared with SjO(2) from the right jugular bulb and SvO(2). Linear regression analysis and Pearson's correlation coefficient were calculated and Bland-Altman analyses were performed. Results: Cerebral TOI and SjO(2) were significantly correlated (r = 0.56, P < 0.0001), as well as TOI and SvO(2) with r = 0.74 (P < 0.0001). Bland-Altman plots showed a mean bias of -4.3% with limits of agreement of 15.7% and -24.3% for TOI and SjO(2) and a mean bias of -4.9% with limits of agreement of 10.3% and -20.1% for TOI and SvO(2). Cerebral rSO(2) and SjO(2) showed a significant correlation (r = 0.83, P < 0.0001) and rSO(2) and SvO(2) showed excellent correlation with r = 0.93 (P < 0.0001). Bland-Altman plots showed a mean bias of -5.2% with limits of agreement of between 8.4% and -18.8% for rSO(2) and SjO(2) and a mean bias of 5.6% with limits of agreement of 13.4% and -2.2% for rSO(2) and SvO(2). Conclusions: Both near-infrared spectroscopy devices demonstrate a significant correlation with SjO(2) and SvO(2) values; nevertheless both devices demonstrate a substantial bias of the measurements to both SjO(2) and SvO(2).