Evaluation of Cerebral Oxygenation During Procedural Sedation in Children Using Near Infrared Spectroscopy

Study objectiveWe evaluate the utility of near infrared spectroscopy monitoring and its correlation to conventional respiratory monitors during changes in cardiorespiratory characteristics during pediatric procedural sedation.MethodsIn this prospective observational study of 100 children, cerebral oxygenation (rSO2), pulse oximetry (SpO2), and end-tidal carbon dioxide (etco2) were monitored continuously. Values were manually recorded at least every 3 minutes from baseline until 30 minutes after sedative administration, resulting in 1,515 triplicate (simultaneous near infrared spectroscopy/etco2/SpO2) measurements. Correlations between conventional monitoring characteristics (SpO2 and etco2) and rSO2 were determined, with focus during adverse cardiorespiratory events.ResultsCerebral oxygenation remained normal in 1,483 of 1,515 measurements (97.9%). rSO2 decreased significantly during 3 of 13 hypoxic events occurring in 13 patients and during 5 of 17 hypercarbic events occurring in 8 patients, with 15 measurements of greater than 20% decrease from baseline. Cerebral oxygenation increased transiently in 88% of children. During 31 cerebral desaturation recordings, 3 hypoxic recordings (9.3%, always in combination with hypercarbia) and 5 hypercarbic recordings (15.6%) were observed, whereas in 23 (74.2%), cardiorespiratory characteristics were unchanged. There was poor correlation between rSO2 and both SpO2 and etco2, with correlation coefficients of 0.05 (95% confidence interval 0.04 to 0.07) and 0.01 (95% confidence interval –0.01 to 0.02), respectively.ConclusionCerebral oxygenation as measured by near infrared spectroscopy demonstrated few significant negative changes during pediatric procedural sedation. Transient cardiorespiratory events seldom altered rSO2, with hypercarbia having a greater effect than hypoxemia. However, cerebral desaturations frequently occurred without associated cardiorespiratory changes.