A meta-analysis of sleep-promoting interventions during critical illness

Background: Sleep quality and quantity are severely reduced in critically ill patients receiving mechanical ventilation with a potential for adverse consequences. Our objective was to synthesize the randomized controlled trials (RCTs) that measured the efficacy of sleep-promoting interventions on sleep quality and quantity in critically ill patients.

Methods: We included RCTs that objectively measured sleep with electroencephalography or its derivatives and excluded observational studies and those that measured sleep by subjective reports. The research was performed according to Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines.

Results: Of 6022 studies identified, 13 met eligibility criteria involving 296 critically ill patients. Eight trials looked at different modes of mechanical ventilation as sleep interventions, and the remaining 5 involved pharmacologic, nonpharmacologic, or environmental interventions. Meta-analysis of the studies revealed that sleep-promoting interventions improved sleep quantity (pooled standardized mean difference [SMD], 0.37; 95% confidence interval [CI], 0.05-0.69; P = .02) and sleep quality through reduction in sleep fragmentation (SMD, -0.31; 95% CI, -0.60 to -0.01; P = .04). Subgroup analysis revealed that timed modes of ventilation improved sleep quantity when compared with spontaneous modes of ventilation (SMD, 0.45; 95% CI, 0.10-0.81; P = .01). Nonmechanical ventilation interventions tended to improve sleep quantity (SMD, 0.65; 95% CI, -0.03 to 1.33; P = .06) and to reduce sleep fragmentation (SMD, -0.29; 95% CI, -0.61 to 0.03; P = .07).

Conclusions: The synthesized evidence suggests that both mechanical ventilation- and nonmechanical ventilation-based therapies improve sleep quantity and quality in critically ill patients, but the clinical significance is unclear. In the future, adequately powered multicenter RCTs involving pharmacologic interventions to promote sleep in critically ill patients are warranted.