Effects of deep sedation on sleep in critically ill medical patients on mechanical ventilation

Atypical EEG patterns not consistent with standard sleep staging criteria have been observed in medical intensive care unit (ICU) patients. Our aim was to examine the relationship between sleep architecture and sedation in critically ill mechanically ventilated patients pre- and post-extubation. We performed a prospective observational repeated measures study where 50 mechanically ventilated patients with 31 paired analyses were examined at an academic medical centre. The sleep efficiency was 58.3 ± 25.4% for intubated patients and 45.6 ± 25.4% for extubated patients (p = .02). Intubated patients spent 76.33 ± 3.34% of time in non-rapid eye movement (NREM) sleep compared to 64.66 ± 4.06% of time for extubated patients (p = .02). REM sleep constituted 1.36 ± 0.67% of total sleep time in intubated patients and 2.06 ± 1.09% in extubated patients (p = .58). Relative sleep atypia was higher in intubated patients compared to extubated patients (3.38 ± 0.87 versus 2.79 ± 0.42; p < .001). Eleven patients were sedated with propofol only, 18 patients with fentanyl only, 11 patients with fentanyl and propofol, and 10 patients had no sedation. The mean sleep times on "propofol", "fentanyl", "propofol and fentanyl," and "no sedation" were 6.54 ± 0.64, 4.88 ± 0.75, 6.20 ± 0.75 and 4.02 ± 0.62 hr, respectively. The sigma/alpha values for patients on "propofol", "fentanyl", "propofol and fentanyl" and "no sedation" were 0.69 ± 0.04, 0.54 ± 0.01, 0.62 ± 0.02 and 0.57 ± 0.02, respectively. Sedated patients on mechanical ventilation had higher sleep efficiency and more atypia compared to the same patients following extubation. Propofol was associated with higher sleep duration and less disrupted sleep architecture compared to fentanyl, propofol and fentanyl, or no sedation.