Nonsedation of mechanically ventilated patients does not improve outcomes compared with light interrupted sedation

Référence

Olsen HT, Nedergaard HK, Storm T, et al. Nonsedation or light sedation in critically ill, mechanically ventilated patients. N Engl J Med 2020;382(12):1103-1111.

Sommaire

Previous studies have shown that providing a daily interruption of light sedation in mechanically ventilated patients improves outcomes compared with continuous light sedation. This trial identified 710 adults intubated within 24 hours of arrival in the intensive care unit who were expected to remain intubated for at least 24 hours. Patients who were comatose, brain dead, who had been transferred from another intensive care unit, or who were being treated in the prone position were excluded. Analysis was by modified intention to treat, with imputation of missing data and multivariate adjustment for the primary outcome. It's not clear why that should be necessary as the groups were relatively balanced at baseline. Patients were randomized to receive light sedation with daily interruption or no sedation. The sedation group received propofol for the first 48 hours, followed by midazolam, with the goal of achieving a score of -2 to -3 on the Richmond Agitation Scale ("arousable"). The nonsedated group could receive morphine if needed, but if sedation was believed to be required those patients were treated with the same protocol as the sedation group. Patients in the sedation group had a mean Richmond score of -2.3 that increased to -1.8 on day 7 (increased alertness), while the nonsedated group had a mean Richmond score of -1.3 that increased to -0.8 on day 7. The primary outcome of all-cause mortality was more common in the nonsedated group (42.4% vs 37.0%), but this difference was not statistically signfiicant. Patients in the sedated group were more likely to experience a major venous thromboembolic event within 90 days (2.8% vs 0.3%; P < 0.05; number needed to treat to harm = 40). There were no other differences between groups. There was less difference in sedation between groups than expected, likely because almost 40% of the nonsedated group received sedation at some point.