Near‐infrared spectroscopic cerebral oxygenation reading in neonates and infants is associated with central venous oxygen saturation

BACKGROUND: The aim of the study was to elucidate easily determinable laboratory and vital parameters in clinical practice to explain variability of near-infrared spectroscopic cerebral oxygenation readings in critically ill newborns and infants using the NIRO 300 spectrometer. METHODS: Near-infrared spectroscopy (NIRS) cerebral tissue oxygenation index (cTOI) was measured on the forehead of critically ill neonates and infants with existing arterial and/or central venous access. We recorded patient characteristics and simultaneously determined sedation state, hemodynamic, respiratory and laboratory data, such as arterial blood gas analysis, electrolytes, hemoglobin and arterial lactate concentration, blood glucose and central venous oxygen saturation. Data were compared using linear, multiple and forward stepwise regression analysis (P < 0.05). RESULTS: A total of 155 neonates and infants aged from 0 to 365 days (median 12 days) were studied. cerebral tissue oxygenation index (cTOI) values ranged from 32.1 to 91.0% (60.5 +/- 11.5%). Simple linear regression analysis revealed significant associations between cTOI and arterial oxygen saturation (r = 0.254, P = 0.001), transcutaneously measured arterial oxygen saturation (r = 0.320, P < or = 0.0001), central venous oxygen saturation (r = 0.489, P < 0.0001), arteriovenous oxygen extraction (r = 0.445, P < 0.0001) and presence of a cardiac shunt (r = 0.250, P = 0.024). Multiple regression analysis and forward stepwise regression revealed two independent, significant predictors for cTOI, namely SvO2 (P < 0.0001) and presence or absence of a cardiac shunt (P = 0.003). SvO2 alone explained 23.9% of the variability of cTOI. The addition of the variable 'cardiac shunt' improved the model to 33%. CONCLUSIONS: Based on our study results cerebral tissue oxygenation readings by the NIRO 300 near-infrared spectrometer is influenced by central venous oxygen saturation, which partially explains intersubject variability of NIRS cerebral oxygenation readings.