BMJ  2006;332:373-374 (18 February), doi:10.1136/bmj.332.7538.373

Editorial

Does melatonin help people sleep?

It's a misapplied but probably safe miracle drug

In North America melatonin is a popular wonder drug which has the legal status of a "nutritional supplement," although that is a legal fiction. As a result it is not regulated as a medicine and is advertised and sold widely—in pharmacies, drug stores, health food shops, and on the internet. Many millions of people use it, mostly because they believe it will help them sleep. However, the claims made for melatonin products and their pharmaceutical quality are not controlled, and their safety has not been systematically studied.

The systematic review in this issue by Buscemi and colleagues at the University of Alberta (p 385) examines the value of melatonin in sleep disorders.¹ Such a review is much needed, as the support for this work from the US National Center for Complementary and Alternative Medicine recognises.

A thorough search of the literature yielded 12 mostly small randomised controlled trials of melatonin in secondary sleep disorders associated with medical and neurological disorders and those related to substance misuse. Another set of 13 randomised controlled trials were conducted with people who had sleep disorders arising from "sleep restriction," in which imposed or self imposed lifestyles or work patterns lead to inadequate sleep. The two sets of randomised controlled trials were analysed separately. Nine trials in each group met the criteria for inclusion in the efficacy review.

In the trials in secondary sleep disorder, melatonin had no significant effect on the time taken to fall asleep and caused a small but unimportant increase in the proportion of time in bed spent asleep ("sleep efficiency"). The trials were very heterogeneous—in adults with dementia, schizophrenia, and major depression, and in children with developmental disability, Rett syndrome, and tuberous sclerosis; the duration of trials ranged from one to eight weeks, and doses of melatonin varied. Evidently melatonin does not help such patients to sleep better.

The trials in people with sleep restriction fall into two quite different groups. Jet lag occurs in people who cross five or more time zones in a day or two and then mostly remain at their destination for many days or weeks. Their functions and habits adapt to the new time zone and environment within a few days of travel, so that their own melatonin secretion soon occurs mainly during the hours of darkness. Only people who repeat such trips in quick succession, typically long distance pilots and cabin crew, experience more complicated and intractable jet lag.

Shiftwork disorder differs, in that the time zone and environment remain the same while people are subjected to new rhythms of sleep and wakefulness. These altered rhythms sometimes continue for long periods and often occur in repeated cycles separated by periods of normal working times. In these circumstances melatonin secretion does not adapt in the same way and is much less predictable. To lump jet lag and shiftwork disorder together in a meta-analysis thus makes no sense. The conventional statistical test for heterogeneity (noted by Busecmi and colleagues in figure 3¹) is too crude and no substitute for considering the trials in detail. For example, jet lag is worst during the first two days after arrival and steadily lessens, so the time course of symptoms must be tracked accurately and compared at several points. Trials differ in this respect, and in the size and timing of doses of melatonin.

The systematic review takes no account of what melatonin does and how it works.¹ The popular misconception underlying the widespread use of this drug is that it induces sleep pharmacologically. It doesn't. Melatonin acts as a regulating switch, pushing the body's circadian phase forward or backward, depending on when the drug is taken. If exogenous melatonin is taken at or after the onset of darkness it substitutes for the endogenous melatonin secretion which normally starts then, and the phase shifts forward, towards the sleep phase. The effect is greater because the doses used are vastly greater than the amount naturally secreted. If exogenous melatonin is taken on waking, phase change is delayed—in physiological terms, the nocturnal period of lowered alertness and performance tends to be prolonged. So, taken in the morning on arrival after a long flight eastwards, melatonin delays circadian adaptation.

The symptoms of jet lag and shiftwork disorder are due to desynchronisation between various body rhythms and environmental rhythms.² Sleep disturbance is merely the most prominent symptom and so gets most attention. It is not surprising that melatonin hardly affects it.

The systematic review also summarises the adverse events reported from the 25 trials included, and concludes from these that short term use of melatonin is safe.¹ The trials seem to have met Downs and Black's quality criterion for reporting adverse events.³ But most randomised controlled trials report adverse events in a cursory and uninformative way,⁴ and few reports describe at what time points the adverse events are detected or elicited and how, factors which greatly influence what is found. Databases of adverse reactions and types of publications that were excluded from the review should also be considered, as the Handbook for Systematic Reviews of Interventions now recommends. The Cochrane review of melatonin for jet lag did this.⁵ It found hints of a possible interaction with warfarin and a suggestion of harm in children with severe epilepsy⁶; both these problems remain to be investigated.

9 Park Crescent, London N3 2NL
(a{at}herxheimer.net)

---

Research p 385

Competing interests: None declared.

References

1. Buscemi N, Vandermeer B, Hooton N, Pandya R, Tjosvold L, Hartling L, et al. Efficacy and safety of exogenous melatonin for secondary sleep disorders and sleep disorders accompanying sleep restriction: meta-analysis. BMJ 2006;332: 385-8.[Abstract/Free Full Text]
2. Herxheimer A, Waterhouse J. The prevention and treatment of jet lag. BMJ 2003; 326: 296-7.[Free Full Text]
3. Downs SH, Black N. The feasibility of creating a checklist for the assessment of the methodological quality both of randomised and non-randomised studies of healthcare interventions. J Epidemiol Commun Health 1998;52; 377-84.[Abstract]
4. Ioannidis JP, Lau J. Completeness of safety reporting in randomized trials: an evaluation of 7 medical areas. JAMA 2001;285: 437-43.[Abstract/Free Full Text]
5. Loke YK, Price D, Herxheimer A. Including adverse effects. In: Cochrane Handbook for systematic reviews of interventions. 2005. Appendix 6b. www.cochrane.dk/cochrane/handbook/hbook.htm (accessed 13 Feb 2006).
6. Herxheimer A, Petrie KJ. Melatonin for the prevention and treatment of jet lag. Cochrane Database Syst Rev 2002(2): CD001520.

CiteULike    Complore    Connotea    Del.icio.us    Digg    Reddit    Technorati    What's this?

Relevant Articles

Efficacy and safety of exogenous melatonin for secondary sleep disorders and sleep disorders accompanying sleep restriction: meta-analysis

Nina Buscemi, Ben Vandermeer, Nicola Hooton, Rena Pandya, Lisa Tjosvold, Lisa Hartling, Sunita Vohra, Terry P Klassen, and Glen Baker
BMJ 2006 332: 385-393. [Abstract] [Full Text] [PDF]

The prevention and treatment of jet lag

Andrew Herxheimer and Jim Waterhouse
BMJ 2003 326: 296-297. [Extract] [Full Text] [PDF]

This article has been cited by other articles:

• Sury, M. R. J., Fairweather, K. (2006). The effect of melatonin on sedation of children undergoing magnetic resonance imaging. Br J Anaesth 97: 220-225 [Abstract] [Full text]  
• (2006). Melatonin: Safe, but Ineffective for Some Sleep Disorders. JWatch Gastroenterology 2006: 10-10 [Full text]  
• (2006). Melatonin: Safe, but Ineffective for Some Sleep Disorders. JWatch General 2006: 4-4 [Full text]  

Rapid Responses:

Read all Rapid Responses

It seems to work in ITU patients

Gopinath Ramaswamy, et al.
bmj.com, 18 Feb 2006 [Full text]