The Glasgow Coma Scale (GCS) is an essential coma scale in critical care for determining the neurological status of patients and for estimating their long-term prognosis. Similarly, cerebral autoregulation (CA) monitoring has shown to be an accurate technique for predicting clinical outcomes. However, little is known about the relationship between CA measurements and GCS scores among neurological critically ill patients. This study aimed to explore the association between noninvasive CA multimodal monitoring measurements and GCS scores.
Acutely comatose patients with a variety of neurological injuries admitted to a neurocritical care unit were monitored using near-infrared spectroscopy–based multimodal monitoring for up to 72 hours. Regional cerebral oxygen saturation (rScO2), cerebral oximetry index (COx), GCS, and GCS motor data were measured hourly. COx was calculated as a Pearson correlation coefficient between low-frequency changes in rScO2 and mean arterial pressure. Mixed random effects models with random intercept was used to determine the relationship between hourly near-infrared spectroscopy–based measurements and GCS or GCS motor scores.
A total of 871 observations (h) were analyzed from 57 patients with a variety of neurological conditions. Mean age was 58.7±14.2 years and the male to female ratio was 1:1.3. After adjusting for hemoglobin and partial pressure of carbon dioxide in arterial blood, COx was inversely associated with GCS (β=−1.12, 95% confidence interval [CI], −1.94 to −0.31, P=0.007) and GCS motor score (β=−1.06, 95% CI, −2.10 to −0.04, P=0.04). In contrast rScO2 was not associated with GCS (β=−0.002, 95% CI, −0.01 to 0.01, P=0.76) or GCS motor score (β=−0.001, 95% CI, −0.01 to 0.01, P=0.84).
This study showed that fluctuations in GCS scores are inversely associated with fluctuations in COx; as COx increases (impaired autoregulation), more severe neurological impairment is observed. However, the difference in COx between high and low GCS is small and warrants further studies investigating this association. CA multimodal monitoring with COx may have the potential to be used as a surrogate of neurological status when the neurological examination is not reliable (ie, sedation and paralytic drug administration).
Departments of *Anesthesiology and Critical Care Medicine
†Neurology, Johns Hopkins University School of Medicine
§Department of Anesthesiology, Northwestern University Feinberg, School of Medicine, Chicago, IL
‡School of Medicine, Faculty of Health, University of Valle, Cali, Colombia
∥Department of Anesthesia and Critical Care Medicine, The Johns Hopkins Hospital, Baltimore, MD
C.W.H. receives research funding from Medtronic/Covidien, Dublin, IR.
C.W.H. is the PI on an NIH-sponsored clinical study (R01 HL 92259) and he serves as a consultant to Medtronic/Covidien and Ornim Medical Inc., Foxborough, MA. L.R.-L. is the PI on an American Academy of Neurology/American Brain Foundation, Covidien/Metronic, and Ornim grant. The remaining authors have no conflicts of interest to disclose.
Address correspondence to: Ryan J. Healy, BSc. E-mail: firstname.lastname@example.org.
Received November 21, 2017
Accepted April 19, 2018