It is well recognized that sleep is abnormal in mechanically ventilated patients in the intensive care unit (ICU). Although this has been described for decades, there is still no consensus on the underlying pathogenesis and the best way to manage it. Moreover, the assumption that abnormal sleep is not good for patients who are critically ill is based primarily on extrapolation from models of sleep loss and sleep disruption in other patient populations and not on evidence that abnormal sleep affects the clinical outcomes of patients in the ICU. Nevertheless, there is growing interest in this topic as the technology to measure sleep evolves and new ways are sought to improve patients’ ability to recover from their critical illness. This chapter outlines the current understanding of the causes and potential consequences of sleep disruption in ventilator-supported patients and how this may be further researched and treated.
Sleep is objectively assessed by means of polysomnography, the simultaneous recording of several electroencephalographic and physiologic parameters.1 Sleep periods are classified as non–rapid eye movement (NREM) and rapid eye movement (REM) sleep. NREM sleep is further subdivided into four stages, with stages 3 and 4 also referred to as slow-wave sleep (SWS). Each sleep stage is recognized by characteristic changes on the electroencephalogram and, in addition, REM sleep has distinctive, intermittent rapid eye movements. During normal sleep, periods of NREM and REM alternate throughout the night in a recognizable pattern, so that most SWS occurs during the first half of the night and most REM sleep occurs during the second half (Fig. 57-1). The “normal” duration of sleep required and the proportion of time spent in each stage of sleep depends on many factors including age and genotype.2 In healthy, middle-aged individuals, however, nocturnal sleep lasts 7 to 8 hours, and 5% to 10% of that time is spent in stage 1 NREM sleep, 50% in stage 2 NREM sleep, 15% to 20% in SWS, and 25% in REM sleep.1 There are also standardized electroencephalographic criteria for identifying arousals and awakenings on the polysomnograph;3,4 up to 10 arousals per hour of sleep is considered to be within normal limits.5 The term sleep architecture refers to the amount of time spent in each sleep stage and sleep disruption is reflected by an increased frequency of arousals and awakenings.
Normal sleep (nocturnal hypnogram). Vertical axis: REM, rapid eye movement sleep; NREM, non–rapid eye movement sleep stages 1, 2, 3, and 4. Horizontal axis: Time in hours.
When questioned after discharge from the ICU, patients consistently report sleep disruption during their stay.6–11 In one study,10 200 patients from four different ICUs received questionnaires that evaluated the quality of their sleep at home and in the ICU and the effect of noise ...