Sleep and Delirium in the ICU: Rest for the Weary?

Aaron B. Holley, MD


October 15, 2018

Quality sleep is critical for recovery from illness. Unfortunately, the hospital tends to be a difficult place to sleep. The intensive care unit (ICU), in particular, poses challenges to optimal sleep. Causes for interruptions and sleep fragmentation in the ICU are well documented and include nursing checks, alarms, shared rooms, excessive light exposure, and effects from underlying illness.[1] Delays in recovery directly attributable to inadequate sleep are difficult to quantify, but likely substantial.

Delirium is also common in the ICU, with prevalence rates approaching 80% in some studies.[2] There is an association between sleep and delirium, but the interactions are complex and incompletely understood.[3,4] Still, interventions to reduce delirium typically target sleep continuity/architecture, sleep duration, and circadian rhythm.

Recently, two randomized controlled trials designed to reduce ICU delirium by improving sleep were published. One used nocturnal infusions of dexmedetomidine, a selective alpha2-adrenergic agonist with sedative properties,[5] and the other used nightly administration of ramelteon, a melatonin receptor agonist.[6] Both claimed reductions in delirium. Interestingly, the dexmedetomidine study found no change in sleep per patient-reported questionnaire[5] and the ramelteon study showed an improvement in sleep per nursing assessment.[6]

Previously published studies have shown both dexmedetomidine[7] and ramelteon[8] impact on sleep and delirium. Dexmedetomidine had never been studied for delirium prophylaxis and ramelteon was studied only for ward patients. Dexmedetomidine has proven to be an effective treatment for delirium once it occurs,[7] and ramelteon reduced delirium for hospitalized ward patients in one, single-center study.[8]

These recent studies are exciting because having an impact on sleep and delirium in the ICU should result in improvement in outcomes. In fact, the ramelteon study showed a trend toward a reduced ICU stay for the intervention group.[6]

However, enthusiasm should be tempered by several realities. First, positive ramelteon studies[8,9] have been small, single-center trials. Second, the recently published ramelteon trial[6] relied on statistical manipulations (log transformation of variables) that are valid in certain settings but lack sufficient explanation in their methods section and should give us pause. Third, sleep was poorly tracked in all trials, and the absence of differences between groups in several trials[5,8] denies us the biologic plausibility needed for reassurance that the improvements in delirium are due to the intervention. Last, both ramelteon and dexmedetomidine are expensive, which will limit widespread use.

It is time that we start focusing intently on sleep and delirium in the ICU. I am not entirely convinced that ramelteon and dexmedetomidine are the solution, but both deserve more study. I am also not ready to implement either on a regular basis, but I am closer than I was before these recent trials were published.

Follow Medscape on Facebook, Twitter, and Instagram


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.