Sleep Pattern Altered During Mechanical Ventilation
Abstract & Commentary
Because sleep is believed to be essential for health and little is known about sleep patterns in the critically ill, Cooper and colleagues prospectively studied sleep in 26 ICU patients requiring mechanical ventilation. Patients were hemodynamically stable and not being treated for depressant drug overdose. Patients were studied for a 24-hour period with continuous polysomnography (PSG) including at least two electroencephalographic (EEG) leads, an electro-oculogram, a submental and anterior tibial myogram, an electrocardiographic lead I, and pulse oximetry. Total sleep time throughout the day was calculated, as was sleep efficiency, the percentage of time asleep of the monitoring time. Movement and EEG arousals were identified as well as periods of rapid eye movement (REM) sleep. A neurologist evaluated the EEG data in patients who demonstrated atypical sleep. Scoring was performed manually using standard definitions.
Six of the studies could not be classified due to technical problems producing artifacts. In the 20 remaining patients, lung injury was the most common cause for ICU admission. Three patients were unresponsive to commands. None was experiencing multiple organ failure. Their average age was 61 years, average APACHE II score was 17 ± 8, with a lung injury score of 1.7. Only the eight patients with "disordered" (D) sleep demonstrated typical periods of REM and non-REM sleep; however, periods were equally spread throughout the 24-hour period. Sleep efficiency was also decreased, with fewer minutes spent sleeping. Patients with "atypical" (A) sleep had no stage 2 non-REM and rare REM periods. They had frequent arousals and often made the transition to stage 3 or 4 non-REM sleep without progressing through stage 2. The patients classified as "coma" (C) (n = 7) had no identifiable neurological insult and most ultimately regained consciousness. Patients in the D and C groups were more severely ill by APACHE scores than those in the A group (13 vs 6).
The A pattern seen in these patients on mechanical ventilation is similar to the sleep seen in other acutely ill patients. This degree of disruption is associated with impairment of daytime function in noncritically ill patients. Sleep was only identifiable in relatively healthy patients with normal levels of consciousness. (Cooper AB, et al. Chest 2000;117:809-818.)
COMMENT BY CHARLES G. DURBIN, Jr., MD, FCCM
This study is important in that it has identified a severe reduction in sleep quantity and quality even in the "healthy" patients undergoing mechanical ventilation. While the causes and consequences of the identified abnormalities are not known, this is a fertile area for research. Although suggested in this study, the effect of specific medications (i.e., narcotics and benzodiazepines) was not predictive of a particular abnormal pattern. Whether the efforts of the caregivers in this study toward establishing "normal" sleep were influential in improving the sleep pattern was not investigated. In the future, the effects of particular agents ("sleep" medications) and care patterns (quiet times at night or daytime naps) should be evaluated.
This paper is helpful because it has identified the magnitude of the problem and has suggested ways to evaluate and compare the PSG records to get these answers. One of the difficulties in taking this work to the next level is the need for manual interpretation of PSG. It is also the reason that PSGs will not soon be a clinical tool useful in managing ICU patients.
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content