Objective: To correlate regional brain saturations (RSO(2)) measured by cerebral Near-infrared spectroscopy (cNIRS) with serological markers indicative of neurological injury (neuron-specific enolase (NSE) and S100beta).
Methods: Children with at least one organ failure who were undergoing cNIRS monitoring were eligible for enrollment, while children with hyperbilirubinemia and cyanotic heart disease were excluded. Children were further analyzed based on the presence of an acute neurological injury (defined as hypoxic/ischemic injury after cardiac arrest, status epilepticus, meningitis, encephalopathy) as well as survival. RSO(2) was measured continuously (every 30 s) and averages were obtained at 6 h and 24 h epochs prior to serum collection (E6 and E24, respectively). Serum was collected for NSE and S100beta, which were both determined by ELISA. Serum from children undergoing evaluation for fever in the Emergency department served as serological controls. Correlations were determined using the Pearson Product Moment Correlations.
Results: A total of 26 children underwent cNIRS monitoring for a total of 47 days. Overall NSE was greater in critically ill children compared to controls, as well as in all subsets of children analyzed (acute CNS injuries, no acute CNS injuries, survivors and non-survivors). S100beta tended to be greater in critically ill children, but this did not reach statistical significance. Average RSO(2) in E6 and E24 was 68.0% +/- 1.5 and 68.6% +/- 1.6, respectively, in a total of 131,036 measurements and E6 RSO(2) was strongly, negatively correlated with S100beta in children with acute neurological injuries.
Conclusions: This is the first study to correlate averaged RSO(2) measured by cNIRS with neurological injury markers in critically ill children. We believe that this data can be used to establish thresholds for RSO(2) that can be tested in future trials to determine if this technology is predictive of long-term neurological outcome.