Study objectives: To objectively measure sleep in critically ill patients requiring mechanical ventilation and to define selection criteria for future studies of sleep continuity in this population.
Design: Prospective cohort analysis.
Setting: University teaching hospital medical-surgical ICU.
Patients: Twenty critically ill (APACHE II [acute physiology and chronic health evaluation II] acute physiology score [APS], 10 +/- 5), mechanically ventilated adults (male 12, female 8, age 62 +/- 15 years) with mild to moderate acute lung injury (lung injury score, 1.8 +/- 0.9) 10 +/- 7 days after admission to the ICU.
Measurements and results: Patients were divided into three groups based on 24-h polysomnography (PSG) findings. No patient demonstrated normal sleep. In the "disrupted sleep" group (n = 8), electrophysiologic sleep was identified and was distributed throughout the day (6:00 AM to 10:00 PM; 4.0 +/- 2.9 h) and night (10:00 PM to 6:00 AM; 3.0 +/- 1.9 h) with equivalent proportions of non-rapid eye movement (NREM) and rapid eye movement (REM) sleep. Nocturnal sleep efficiency was severely reduced (38 +/- 24%) with an increased proportion of stage 1 NREM sleep (40 +/- 28% total sleep time [TST]) and a reduced proportion of REM sleep (10 +/- 14% TST). Severe sleep fragmentation was reflected by a high frequency of arousals (20 +/- 17/h) and awakenings (22 +/- 25/h). Electrophysiologic sleep was not identifiable in the PSG recordings of the remaining patients. These were classified either as "atypical sleep" (n = 5), characterized by transitions from stage 1 NREM to slow wave sleep with a virtual absence of stage 2 NREM and reduced stage REM sleep, or "coma" (n = 7), characterized by > 50% delta or theta EEG activity with (n = 5) and without (n = 2) evidence of EEG activation either spontaneously or in response to deep painful stimuli. The combined atypical sleep and coma groups had a higher APS (13 +/- 4 vs 6 +/- 4) and higher doses of sedative medications than the disrupted sleep group.
Conclusion: Sleep, as it is conventionally measured, was identified only in a subgroup of critically ill patients requiring mechanical ventilation and was severely disrupted. We have proposed specific criteria to select patients for future studies to evaluate potential causes of sleep disruption in this population.