Abstract
Introduction
Disruption of sleep may have significant implications in acute brain injury, functional recovery, and critical illness. Few data exist characterizing sleep architecture in patients admitted to an intensive care unit (ICU). We aim to describe sleep and clinical characteristics in patients with acute brain injury and critical illness.
Methods
Retrospective analysis was performed in ICU patients who underwent continuous electroencephalographic (EEG) monitoring from 2018-2019. Sleep was scored based on AASM-defined EEG criteria. Clinical variables, EEG characteristics, and modified Ranking Scale (mRS) were collected. Good outcome was defined as mRS<3. Differences were assessed using chi-square analysis and t-test.
Results
205 patients were reviewed with a mean age of 57 years (range 18-91) and a majority (57%) were male. Patients carried a primary neurologic/neurosurgical (61%) or medical/surgical (39%) diagnosis. Status epilepticus, subdural hemorrhage, traumatic brain injury, encephalopathy and cardiac arrest accounted for the majority of diagnoses encountered. Only 58 patients (28%) achieved N1 sleep; of these 76.4% achieved N2, 2.8% N3, and none achieved REM. Of those achieving any sleep, 43% had good outcomes versus only 23% in those who did not (t=-7.45, p<0.001). Neurological patients were more likely to attain sleep compared to those with other primary diagnoses (χ 2 (1)=7.08, p=0.008). Centrally acting anesthetics did not account for sleep differences between neurologic and non-neurologic patients (χ² (1)=2.01, p=0.16). However, those with primary brain injury reached sleep more often in the absence of anesthetic use (χ 2 (1)=4.82, p=0.03). The overall mortality was 32% in this cohort.
Conclusion
Most critically ill patients do not achieve electrophysiologic sleep. Of those who do, N1/N2 stages are seen most often. Neurological patients were more likely to sleep, and achieving any sleep was associated with improved functional outcome. Further studies are needed to determine whether sleep augmentation in the critically ill impacts functional outcome.
Support
N/A.