A double-blind placebo-controlled randomised pilot study of nocturnal melatonin in tracheostomised patients

Patients in the intensive care unit often suffer from lack of sleep at night. We hypothesised that nocturnal melatonin may increase observed nocturnal sleep in tracheostomised patients.Double-blind, randomised, placebo-controlled pilot study.ICU of a tertiary hospital.Thirty-two ICU patients with tracheostomy who were not receiving continuous sedation.We administered either oral melatonin (3mg) or placebo at 20:00. We collected pre- and post-dosage blood samples on Days 1 and 3 to confirm drug delivery. Primary outcome measure was number of hours of observed sleep at night, assessed by the bedside nurse. Secondary outcome measures included comparison of the incidence of agitation, assessed by score on the Riker Sedation-Agitation Scale, and requirement for sedatives or haloperidol to settle agitation.Pre-treatment melatonin levels in the two groups were similarly low: 4.8 pg/mL (95% CI, 2.4-7.5) for melatonin versus 2.4 (95% CI, 1.6-3.2) for placebo (P=0.13). Post-treatment, melatonin levels increased significantly in the melatonin group compared with the placebo group (3543 pg/mL versus 3 pg/mL; P<0.0001). However, subsequent observed nocturnal sleep was similar in the two groups: 240 minutes (range, 75-331.3) for melatonin v 243.4 minutes (range, 0-344.1) for placebo (P=0.98). Observed diurnal sleep was also similar: 138.7 minutes (range, 50-230) with melatonin v 104 minutes (range, 0-485) for placebo (P=0.42). The incidence of agitation was non-significantly higher in the melatonin group (31% v 7%; P=0.11), while the requirement for extra sedation or use of haloperidol was slightly higher in the placebo group (57% versus 46%; P=0.56).Melatonin is well absorbed, and a standard dose increases blood levels approximately 1000-fold. However, in this pilot assessment, these high levels failed to increase observed nocturnal sleep or induce other observable benefits in tracheostomised ICU patients.