REVIEW OF: Hatta K et al, J Clin Psychiatry 2017;78(8):e970–e979; Hatta K et al, J Clin Psychiatry 2019;81(1):19m12865

Delirium is an acute confusional state that occurs frequently in older hospitalized patients, especially those with serious illness in the ICU. We try our best to prevent it because it is associated with worse medical outcomes, including mortality. Since one of the hallmarks of delirium is interrupted sleep, sedative-hypnotics are a natural treatment option. Unfortunately, benzodiazepines can worsen delirium. Non-benzodiazepine hypnotics such as ramelteon (a melatonin receptor agonist) and melatonin have shown promise at preventing delirium in small clinical trials. Recently, two trials examined whether a novel hypnotic, suvorexant (an orexin receptor antagonist), prevents delirium in patients at risk.

In the first study, 72 acute hospitalized patients, ages 65–89, were randomized to either suvorexant (15 mg given at 9 pm) or placebo for 3 days. The patients were followed for 7 days, starting the day after the drug was first given. Six placebo-receiving patients became delirious, while no suvorexant patients became delirious (17% vs 0%; p = 0.025). There was no difference in subjective or clinician-rated sleep parameters (such as total minutes of sleep) between the two groups.

The second study was a non-randomized “real-world” follow-up study examining delirium incidence in hospitalized patients given ramelteon (8 mg/day), suvorexant (15 mg/day), or both, compared to patients who chose to take neither. Those who took ramelteon and/or suvorexant developed delirium less frequently than those who took neither (63/401 = 16% vs 30/125 = 24% [RR 1.5]).

Researchers also studied patients who had been delirious the night before but were no longer confused the day the study started. Of this group, those who chose to take ramelteon and/or suvorexant also were less likely to develop delirium (133/333 = 40% vs 59/89 = 66% [RR 1.7]) and also showed sleep/wake cycle improvements at the end of the intervention.

CHPR’s Take
While suvorexant appears effective at preventing delirium, it’s unlikely to be a first-line agent in your hospitalized patients. Given its cost, it’s usually not on hospital formularies, and it can cause significant next-day sedation and even sleep paralysis and cataplexy, especially if used at higher doses (30–40 mg/day). For these reasons, if a sleeping agent is needed in your hospitalized patients, we recommend starting with ramelteon and only using suvorexant as a second- or third-line agent. Continue recommending common interventions to prevent delirium, including reorientation, uninterrupted sleep hours, hydration, avoidance of delirium-causing medications, and active treatment of underlying medical ­conditions.