(749) Spectral Electroencephalographic Markers of Sedation Depth in the Pediatric Intensive Care Unit

Introduction: Optimal sedation is critical to the care of PICU patients. Scores for sedation depth rely on subjective criteria, however, and cannot be used for muscle-relaxed patients. Electroencephalogram (EEG)-derived metrics, such as spectral edge frequency (SEF), currently used for objectively assessing sedation depth in the operating room, may have a role in sedation monitoring and titration in the PICU. We aimed to assess the association between EEG spectral characteristics and traditional metrics of sedation for patients in the PICU.

Methods: We enrolled patients (1-18 years of age) receiving at least one continuous sedative medication. We excluded patients with an abnormal baseline EEG or frontal cerebral abnormalities. A four-lead frontal EEG was recorded with the Sedline Brain Function Monitor (Masimo Corporation) for 1-24 hours per subject. The timing of sedation boluses was extracted from the electronic health record, and nurses were asked to assess their clinical response. Spectral analysis was performed using the Chronux toolbox in MATLAB (MathWorks).

Results: A total of 16 subjects were enrolled. Using linear mixed-effects modeling, the mean (+/- 95% confidence interval) SEF was significantly lower in patients receiving dexmedetomidine with an opiate or benzodiazepine (8.2 +/- 1.9 Hz) compared to patients on analgosedatives without dexmedetomidine (17.7 +/- 1.9 Hz, p< 0.001). When stratified by nursing assessment, boluses associated with an observed clinical response had a greater mean decrease in SEF (-10.6 +/- 23.5%) than those without a response (-1.0 +/- 7.8%), however this was not statistically significant (p=0.86). Further modeling revealed that age had a statistically significant association with SEF (beta 0.65 +/- 0.20, p< 0.001), but state behavioral scale did not (beta -0.048 +/- 0.623, p=0.89).

Conclusions: We have described significantly lower SEF in PICU patients on analgosedatives that include dexmedetomidine compared to those without dexmedetomidine. This finding is consistent with adult studies. Furthermore, the relationship between the decrease in SEF and clinical response after sedation boluses should be explored. Our observations suggest that SEF may have a role in objective monitoring of sedation depth in the PICU, and is influenced by age and sedating agent class.