Melatonin is Effective for Sleep Disorders in Young People with Developmental Delays, but has Variable Effects on Seizures

Abstract & Commentary

Synopsis: It seems prudent to strongly consider use of melatonin in our cognitively delayed patients with disordered sleep, but to monitor those with epilepsy very closely for exacerbation, and withdraw the drug.

Source: Coppola G, et al. Melatonin in Wake-Sleep Disorders in Children, Adolescents and Young Adults with Mental Retardation with or without Epilepsy: A Double-Blind, Cross-Over, Placebo-Controlled Trial. Brain Dev. 2004;26:373-376.

Sleep problems are among the most common recurrent complaints brought to neurologists by caretakers of those with cognitive delays. Poor sleep is a significant family stressor, produces increased daytime somnolence, behavioral problems, school/work difficulties, and frequently exacerbates epilepsy. The prevalence of sleep difficulties in intellectually impaired populations has been found to range from 15%-67%.

This paper presents a small randomized, double-blind, cross-over, placebo-controlled trial of melatonin in mentally retarded young people with sleep disturbances. Initial screening was by questionnaire regarding sleep latency, awakenings, total sleep time by day and night, and early arousals. Thirty-two patients were enrolled. Patients had cognitive impairment from mild to severe, a broad range of seizure types (18/25 were epileptic), as well as a variety of related pathologies, including genetic syndromes (20%), cerebral palsy (32%), and congenital visual impairment (16%).

Caregivers kept sleep logs during constant nocturnal patient observation. Polysomnograms were performed when seizures were suspected to be causing, or mimicking, a sleep abnormality. Seizure diaries were also kept, with frequency, type, and duration, before and during the trial. No changes were made to any antiepileptic drugs (AEDs).

Patients were randomized to either oral synthetic fast-release melatonin (3 mg q.h.s., increasing 3 mg weekly as tolerated, if needed, to 12 mg q.h.s. max) or placebo for 4 weeks, underwent a 1 week cross-over, then a second 4 week phase. 25/32 patients completed both phases (16 male; 9 female; aged 3.6-26 years). Responders then entered a 2-month, open-label phase.

Melatonin had a significant effect on sleep latency (P = 0.019) at doses of 3 mg/d (29.2%), 6 mg/d (45.8%), 9 mg (20.8%), and 12 mg (4.2%; 1 patient). One patient did not respond. No effect on nocturnal awakenings (P = 0.768) or daytime sleep (P = 1) was found. Interestingly, total nocturnal sleep time increased with either melatonin or placebo vs baseline (7.9, 7, and 4.4 h respectively), though Coppola and colleagues state that the data are skewed by persistence of melatonin effect into the placebo phase, in those randomized first to melatonin. There was a trend toward fewer early arousals (P=0.123), with melatonin as well. Improved behavior and daytime alertness was noted (though not quantified) in half of the treated individuals, and families reported subjective improvement in overall quality of life. No side-effects were reported by any participants, including the 7 drop-outs.

Of note, 2 of 11 previously seizure-free epilepsy patients, relapsed after 1 month, and remitted after discontinuing melatonin.

Of the 7 patients whose epilepsy was uncontrolled from the start, 1 became seizure-free, 2 improved, 2 worsened, and 2 were unchanged.

Further improvements in sleep latency (mean, 0.2 h), duration of night sleep (mean, 8.4 h), and early arousal (mean, 2.0) were seen in the final 2 month, open-label phase of the study.


Melatonin, a.k.a. N-acetyl-5-methoxytryptamine, is synthesized from serotonin in the pineal gland. It is regulated by the light-dark cycle, and plays a role in immunity, endocrine regulation, the stress response, as an antioxidant, and in the regulation of circadian rhythms.

Given melatonin’s well-known ability to reset the sleep cycle, and induce and improve sleep, one would expect that epileptics might experience decreased seizure frequency from melatonin treatment, via a reduction in sleep deprivation alone. Additionally, abundant data, in a vast array of animal seizure models, demonstrate a direct anticonvulsant effect. In humans, reduced melatonin levels have been found in patients with intractable epilepsy, as well as in non-epileptics with mental retardation. Lastly, several small studies have shown clinical improvement in epileptic children who took melatonin. On the other hand, there have been occasional reports of melatonin having proconvulsant effects.

In the study summarized here, however, there was no reliable effect of melatonin on epilepsy, despite a clear improvement in sleep latency, and a likely improvement in total sleep and fewer arousals. The effects seen here on sleep and behavior, similar to those seen in a number of other small studies of melatonin in children with cognitive disabilities, imply that there is clear utility of melatonin for sleep disturbance in mentally disabled populations. However, several studies, including this one, imply that despite the experimental evidence suggesting that melatonin has inherent anticonvulsant properties, it may exacerbate seizures in some children and young people with epilepsy, while helping many others.

Given melatonin’s obvious benefit to many, but a significant possible risk to some epileptics, we eagerly await a larger, prospective, double-blind, placebo-controlled study with sufficient power to allow for multivariate analysis by seizure type, sex, age, mental and/or visual disability, specific concomitant AED use, and other factors. Only then, might we determine whether there are definable subgroups that might particularly benefit from melatonin therapy, or others in whom it might be a relatively contraindicated. Until then, it seems prudent to strongly consider use of melatonin in our cognitively delayed patients with disordered sleep, but to monitor those with epilepsy very closely for exacerbation, and withdraw the drug. — Susan E. Snyder

Dr. Snyder is a resident in pediatric neurology, New York Presbyterian Hospital, Weill Cornell Medical Center.