Sleep in the neurological intensive care unit: feasibility of quantifying sleep after melatonin supplementation with environmental light and noise reduction.

Abstract

Sleep deprivation may be particularly detrimental to intensive care unit (ICU) patients. Polysomnography has demonstrated abnormal sleep in medical and surgical ICU populations. Both environmental factors and circadian disruption have been implicated. We hypothesized that patients in a neurologic ICU would demonstrate similar sleep disturbances and that a combination of sleep-promoting interventions would increase sleep time. Twelve patients were enrolled in this pilot-randomized, controlled, study in a neurologic ICU. For adult patients undergoing continuous EEG for clinical purposes, noise-cancelling headphones and eye masks were worn, and an oral dose of melatonin was administered for 3 days, or until EEG was stopped. Sleep was scored according to standard criteria; EEG was characterized and analyzed quantitatively. Sixty-five percent of the patients' recordings were unscorable based on accepted standardized criteria; therefore, sleep measures could not be compared. For those with sleep that could be scored, total sleep time was normal, although sleep was fragmented and time spent in slow-wave or rapid eye movement sleep was notably decreased. Patients with unscorable recordings had worse injury severity measures, absent or significantly slower posterior dominant rhythm, and less coherence of posterior faster frequencies. Clinical outcomes were similar between intervention and control groups. Although sleep-promoting interventions were feasible, sleep quantification based on currently accepted criteria limited the ability to score sleep. Similar to other ICUs, sleep in the neurologic ICU is abnormal; patients with unscorable sleep-like states have greater injury severity. This study was limited by strict enrollment criteria. A reliable method to quantify sleep and sleep-like states in the ICU is needed.