- Top of page
Rapid eye movement (REM) sleep behaviour disorder (RBD) has been suggested to predict the development of neurodegenerative disorders. Patients with RBD are acting out dream behaviour associated with loss of normal muscle atonia of REM sleep. The aim of the present study was to confirm that exogenous melatonin improves RBD. Eight consecutively recruited males (mean age 54 years) with a polysomnographically (PSG) confirmed diagnosis of RBD were included in a two-part, randomized, double-blind, placebo-controlled cross-over study. Patients received placebo and 3 mg of melatonin daily in a cross-over design, administered between 22:00 h and 23:00 h over a period of 4 weeks. PSG recordings were performed in all patients at baseline, at the end of Part I of the trial and at the end of Part II of the trial. Compared to baseline, melatonin significantly reduced the number of 30-s REM sleep epochs without muscle atonia (39% versus 27%; P = 0.012), and led to a significant improvement in clinical global impression (CGI: 6.1 versus 4.6; P = 0.024). Interestingly, the number of REM sleep epochs without muscle atonia remained lower in patients who took placebo during Part II after having received melatonin in Part I (–16% compared to baseline; P = 0.043). In contrast, patients who took placebo during Part I showed improvements in REM sleep muscle atonia only during Part II (i.e. during melatonin treatment). The data suggest that melatonin might be a second useful agent besides clonazepam in the treatment of RBD.
- Top of page
Rapid eye movement (REM) sleep behaviour disorder (RBD) is clinically impressive by virtue of its vigorous sleep behaviours, which are usually accompanied by vivid dreams and often result in injuries to the patients themselves or to bed partners (AASM, 2006; Gagnon et al., 2006b; Schenck and Mahowald, 2002). Even though the neurophysiological correlate of RBD, loss of muscle atonia during REM sleep, was described in the mid-1970s, the complete clinical syndrome was not defined until 1986 (Schenck and Mahowald, 2002).
Although its aetiology is virtually unknown, RBD is far from rare (Gagnon et al., 2006b; Schenck and Mahowald, 2002). Approximately one-third of patients with Parkinson’s disease experience RBD prior to the onset of Parkinson’s disease symptoms (Olson et al., 2000; Schenck et al., 1996) and more than 90% of patients with multiple system atrophy (MSA) have RBD (Iranzo et al., 2006; Plazzi et al., 1997). There is also a strong association between RBD and neurodegenerative diseases, especially Lewy body dementia (Boeve et al., 1998; Iranzo et al., 2006). An increasing number of psychoactive drugs are being recognized as inducing a loss of muscle atonia – the neurophysiological basis of RBD – during REM sleep (Gagnon et al., 2006c). Recently, RBD was even proposed to be a predictor and early marker of the development of neurodegenerative diseases (Gagnon et al., 2006b; Iranzo et al., 2006; Turek and Dugovic, 2005).
Because the origin of RBD is unknown, therapeutic approaches to date have been limited to symptomatic treatment. Clonazepam was reported to be highly effective in two cohort studies and is widely accepted among clinicians as a treatment for RBD (Gagnon et al., 2006b; Schenck and Mahowald, 2002). Nevertheless, there are patients who do not respond to clonazepam, and the drug can cause side effects such as daytime somnolence, muscle relaxation in the elderly, reduced sleep quality, cognitive impairment and worsening of sleep apnoea.
- Top of page
Because of administrative changes in our psychiatric department, the study was terminated before all 12 patients had been randomized. At the time of study termination, a total of eight patients had been randomized and had completed all study procedures. Three patients had been assigned randomly to placebo in Part I, and five patients had been assigned randomly to placebo in Part II. Data are given in Table 1 for baseline, melatonin and placebo for all subjects. To evaluate the enduring effects of melatonin (i.e. beyond the period of melatonin administration itself), changes from baseline to placebo are given separately for the placebo group in Part I (i.e. prior to melatonin) and the placebo group in Part II (i.e. after melatonin).
Table 1. Demographics and clinical results of eight RBD patients with melatonin
|Patient||Age||Gender||Duration of RBD (years)||Clinical improvement of RBD||Concomittant disorder|
|1||59||Male||20||Much||Narcolepsy, PLMD (30 years)|
|2||26||Male||6||No dream outacting||Narcolepsy, PLMD (9 years)|
|3||37||Male||6||Much||Idiopathic insomnia (20 years)|
|4||55||Male||10||Much||Idiopathic insomnia (30 years)|
|6||63||Male||5||No dream outacting||Parkinson’s (4 years)|
|7||64||Male||5||No dream outacting||None|
|8||67||Male||8||No dream outacting||None|
None of the eight patients evaluated in our study failed to take their medication more than once during any one treatment period. Patients maintained good sleep hygiene as indicated by the low SD in bedtimes recorded by means of individual sleep logs and actigraphy. The 24-h excretion of aMT6s in urine at baseline was 28 μg on average (range 15–42 μg), which is similar to the 24-h excretion observed in patient groups in our earlier studies (Kunz and Bes, 1999, 2001; Kunz et al. 1999, 2004).
Except for a reduction in sleep-onset latency, placebo in Part I did not affect sleep significantly (Table 2: baseline versus placebo). Compared to baseline (Table 2: baseline versus melatonin), melatonin reduced sleep-onset latency significantly and the percentage of REM sleep epochs without muscle atonia. We were unable to observe significant changes in any of the other sleep variables evaluated in our study. Melatonin did not change REM density or phasic muscle activity during REM sleep, but did improve the CGI score significantly. Melatonin did not change REM latency consistently or time of temperature minimum [baseline: 04:30 h (SD = 2 h 40 min); melatonin: 04:32 h (SD = 2 h 44 min); placebo: 04:39 h (SD = 2 h 45 min)], indicating that the circadian phase also remained unchanged. When the results of placebo in Part II are compared to baseline (Table 2: placebo II), the effects on REM sleep epochs without muscle atonia still remained significantly different (P = 0.043). This would seem to indicate that melatonin has an enduring effect that outlasts the actual period of melatonin administration.
Table 2. Effects of melatonin on sleep in patients with RBD
| ||Baseline||Melatonin||Placebo||Placebo I – Baseline*||Placebo II – Baseline*||Baseline vs. Melatonin||Baseline vs. Placebo||Melatonin vs. Placebo|
|n = 8||n = 8||n = 8||n = 3||n = 5||n = 8||n = 8||n = 8|
|Mean SD||Z (Wilcoxon test) P|
|Phasic muscle twitches (%)||48.6||48.3||49.8||1.89||0.83||−0.42||−0.28||−0.98|
A total of seven of eight patients reported clear improvements in symptoms during melatonin treatment, and four of these seven were free of clinical RBD symptoms at the end of the treatment period according to their own and their bed partner’s reports. One patient reported slight improvements in RBD symptoms, but felt unchanged with regard to daytime symptoms. Improvements started during the first week of active treatment and increased gradually during the 4-week treatment period. None of the responders reported having frightening dreams, nor did any of them fall or jump out of bed after 4 days of treatment had passed. No adverse events or side effects were reported. All patients were able to distinguish placebo from verum based on a reduction in dream mentation and/or the fact that they felt more refreshed in the morning. The most frequently cited subjective changes during melatonin treatment were a reduction in daytime fatigue (n = 3), a stronger sense of feeling refreshed after awakening (n = 5) and increased sleepiness in the evening (n = 3).
- Top of page
Our data represent the first placebo-controlled clinical treatment study of RBD. Clonazepam, the most widely accepted therapy in the treatment of RBD today, reduces phasic muscle activity (Lapierre and Montplaisir, 1992). In contrast, melatonin increases the number of epochs scored as REM sleep without muscle atonia. Thus, melatonin seems to have a different mode of action in patients suffering from RBD than clonazepam.
Although generated in an ultradian manner, expressions of REM sleep such as REM sleep latency, REM sleep episode length and REM continuity are under strong circadian control (Bes et al., 1996; Dijk and Czeisler, 1995; Wurts and Edgar, 2000). However, melatonin clearly influences REM sleep. Melatonin administered during the early evening hours to healthy subjects changes REM latency and lengthens the first REM sleep episode (Cajochen et al., 1997). Low endogenous melatonin is accompanied with reduced REM sleep duration (Mahlberg et al., 2008). Exogenous melatonin also increases REM sleep percentage to normal levels in patients with reduced REM sleep duration and re-organizes REM sleep episode length during night-time sleep (Kunz et al. 2004). Beta-blockers decrease melatonin excretion and reduce the number of REM sleep epochs simultaneously, which can be reversed by exogenous melatonin (van den Heuvel et al., 1997).
One striking finding in patients treated with melatonin in comparison to those treated with clonazepam is the enduring effect of melatonin. The symptoms of RBD re-occur immediately after clonazepam is discontinued (Schenck and Mahowald, 2002). In the present study, the positive clinical and neurophysiological effects in the five patients treated with melatonin in Part I extended into Part II (i.e. the placebo part) of the trial; in other words, the effects of melatonin lasted for at least 5 weeks. We and others have reported a resolution of clinical RBD symptoms lasting for up to 3 years after discontinuation of melatonin treatment (Boeve, 2001; Kunz and Bes, 1999).
Some caveats need to be considered: only a small sample size was included, thus questioning generalizability. Nevertheless, the study was confirmatory and patients were included unselected in a consecutive manner. Only one night of polysomnography was included in the evaluation for each phase. However, no reports exist on a high night-to-night variation of REM sleep muscle atonia.
It should be emphasized that our RBD patients differ partly from others reported in the literature in terms of the severity of their clinical symptoms. In most of our patients, RBD was identified by chance at an early stage. Of our eight subjects, only one was suffering from Parkinson’s disease and five had idiopathic RBD. Except for the two narcolepsy patients, most patients had contacted the sleep clinic primarily for unrestorative sleep. Most RBD patients in other studies contacted the sleep clinic because they had been acting out their dreams, which indicates more severe disease. Accordingly, the percentage of REM sleep epochs without muscle atonia in these patients was higher than in our sample. In our experience, melatonin improves RBD symptoms in neurodegenerative disorders, but we observed a complete resolution of symptoms only in idiopathic RBD patients. Thus, it may well be that combining melatonin and clonazepam will produce superior results, especially in neurodegenerative disorders.
Melatonin did not cure RBD in our patients. Four of our patients exhibited complete clinical resolution of RBD behaviour, two showed marked improvement, one showed little improvement, and one remained unchanged. REM sleep without muscle atonia did not resolve completely in any of the subjects. Elsewhere we have published a case report on a patient who showed gradual improvements over the course of 3 months (Kunz and Bes, 1997). Thus, it needs to be determined whether a longer period of melatonin treatment might result in even better results.