Reference
Whittom S, Dumont M, Petit D, Desautels A, Adam B, Lavigne G, Montplaisir J. Effects of melatonin and bright light administration on motor and sensory symptoms of RLS. Sleep Med. 2010;11(4):351-355.
Design
Simple open trial design
Participants
Eight subjects with restless leg syndrome (RLS)
Study Medication and Dosage
The subjects were studied at night under 3 conditions: at baseline, after administration of melatonin, and during bright light exposure.
Outcome Measures
The severity of RLS symptoms was assessed by the suggested immobilization test (SIT), which allows quantification of both sensory and motor manifestations (SIT-PLM) of RLS.
Key Findings
There was a significant increase of SIT-PLM index when subjects took melatonin compared to both baseline and bright light conditions. Bright light exposure had no effect on leg movements but did produce a significant decrease in sensory symptoms. Exogenous melatonin may have a detrimental effect on motor symptoms, and bright light exposure produced small but significant improvement in leg discomfort.
Practice Implications
This small study may in time be seen as one of the first clinical trials that eventually led to a change in the way we approach restless leg syndrome (RLS) treatment in clinical practice.
Though it is rare for patients to present with RLS as their chief complaint, the condition is common, affecting 10–11% of the general population. Being female increases risk and so does the number of children a woman has had: Having 1 child nearly doubles the risk of having RLS, 2 children triples the risk, and 3 or more children increase the odds ratio to 3.57.1 It is surprising that more patients don’t complain about RLS.
RLS symptoms also follow a circadian pattern, being worse in the late evening at a time when presumably melatonin levels are either increasing or peaking.
Sir Thomas Willis first described RLS symptoms in 1685.2 Karl Axel Ekbom, a Swedish neurologist, is credited with writing the first modern clinical description and is also credited with naming the condition.3 The disease is actually being renamed; what we call restless leg syndrome will soon be known as Ekbom disease. “The main characteristics are the strong urge to move, accompanied or caused by uncomfortable, sometimes even distressing, paresthesia of the legs, described as a “creeping, tugging, pulling” feeling. The symptoms often become worse as the day progresses, leading to sleep disturbances or sleep deprivation, which leads to decreased alertness and daytime functions.”4
To comprehend the potential significance of this current paper by Whittom et al, we must review the current understanding of this disease.
Almost from the start, research on RLS has focused on iron deficiency, with Ekbom reporting an association in 1960.5 In 2005 Thorpy described RLS as an iron-deficiency symptom. This model explained the higher incidence of RLS in women.6
Over the last 7 years, several clinical trials have suggested that IV iron infusions will improve RLS symptoms. In an open label series, Earley started patients on single 1,000 mg doses of infused iron initially and then gave 450 mg if the patients' ferritin levels dropped below 300 mcg/l. These patients experienced benefit.7,8,9 An interim report on a placebo-controlled trial by Earley’s group at Johns Hopkins published in 2009 had not yet found measurable benefit at the time of writing.10 A different RCT published in 2009 by Swedish researchers did report that iron transfusions reduced RLS symptoms both in both short term and long term.11 We can probably say that iron infusions help some patients at least some of the time.
It is interesting—and counterintuitive—to note that blood donation does not worsen RLS symptoms.12
The relationship between iron and RLS is more complicated than a simple iron deficiency. In RLS the regulation of iron absorption and transport differs from that of healthy people. Earley reported that after the initial infusion, ferritin levels in his RLS subjects declined an average of 6.6 mcg/l/week, faster than the expected rate of <1 mcg/l per week. The faster the ferritin level dropped, the less likely the subsequent iron treatment was to provide benefit.13
Theories to explain RLS now center on iron levels in the brain. Conner et al have published a series of papers reporting on their analysis of brain samples obtained from RLS patients on autopsy. In 2003 their early data suggested that RLS results “from impaired iron acquisition by the neuromelanin cells. … The underlying mechanism may be a defect in regulation of the transferrin receptors.” In 2004 Connor et al reported in comparing further autopsy results that iron transport to the substantia nigra (SN) in RLS is different due to a defect in iron regulatory protein 1.14
The current data continue to suggest that RLS is a neurological disorder involving abnormal levels of iron in various parts of the brain. Mitochondrial ferritin (FtMt) levels correlate between iron levels and mitochondrial function in the SN. Autopsy samples from RLS patients have more FtMt than control samples. Neuromelanin-containing neurons in the SN were the predominant cell type expressing FtMt. These results suggest that increased numbers of mitochondria in neurons in RLS and increased FtMt might contribute to insufficient cytosolic iron levels in RLS SN neurons; these results are consistent with the hypothesis that energy insufficiency in these neurons may be involved in the pathogenesis of RLS.15
In 2011 Connor et al reported finding an iron regulatory protein that is decreased in RLS. They suggest that this leads to an alteration in iron management at the blood-brain interface. Thus there are “fundamental differences in brain iron acquisition in individuals with restless legs syndrome.” These researchers now identify a decrease in transferrin receptor expression in the microvasculature of the brains of RLS patients as that leads to demylination of neurons in the brain as possibly the underlying problem.16
Let us return to the current Whittom paper under analysis and see how it may relate to this understanding of how iron dysregulation underlies RLS.
Serum iron levels have long been thought to fluctuate according to a circadian rhythm, and the general consensus has been that they are higher in the morning, although not all data support this.17
RLS symptoms also follow a circadian pattern, being worse in the late evening at a time when presumably melatonin levels are either increasing or peaking.18 Working a rotating night shift, a practice that will likely disrupt melatonin production and circadian cycles, increases RLS symptom intensity.19 Micahud et al, who had looked for biochemical markers correlating with RLS symptoms, wrote in a 2004 report that “changes in melatonin secretion were the only ones that preceded the increase in sensory and motor symptoms in RLS patients. This result and those of other studies showing that melatonin exerts an inhibitory effect on central dopamine secretion suggest that melatonin might be implicated in the worsening of RLS symptoms in the evening and during the night.”20
A 2007 paper reported that melatonin changes the expression of transferrin receptors in the pineal glands of rats. These are the “membrane bound glycoproteins which function to mediate cellular uptake of iron from transferrin." Melatonin reduced transferring receptors.21 If melatonin can affect iron movement in the brains of rats, it would be a fair assumption that it may do the same in humans.
Two earlier papers looking at melatonin and RLS should be mentioned. Austrian researchers made the correlation that people with RLS often complain of insomnia and that melatonin may help insomnia, so they tested melatonin levels in individuals who have RLS hoping to find a deficiency. They did not. “Insomnia in RLS does not seem to be correlated with a deficit of melatonin.”22
An earlier paper published in 2001 will confuse the matter. Assuming that there was a deficiency of melatonin in RLS, German researchers gave 3 mg of melatonin to RLS patients for 6 weeks. They reported improved well-being and a reduction in limb movements in 7 of the 9 patients in their trial.23 Is there an explanation that will reconcile these contradictory results, or are we simply looking at an example of why the results of small open trials without control groups should be viewed with caution?
Let us assume for the moment that the results of both trials are true and that some unrevealed difference in protocol or patient selection accounts for the contrasting results. Then we might conclude that melatonin sometimes improves RLS symptoms and sometimes worsens them. In either event, moderating melatonin should be considered in patients with RLS.
It would make sense to find out how a particular patient will respond to melatonin supplementation as a first step in treating RLS. Depending on their reaction, it may be useful to encourage or discourage melatonin production. For example appropriate frequency lighting can be used to either increase or decrease melatonin production.
The relationship between ferritin and RLS is now well established, and levels should be monitored and corrected in all patients complaining of RLS.
One last bit of information is worth mentioning. An extract of the herb Salvia miltiorrhiza called Tanhinone II was recently reported to prevent brain iron dyshomeostasis. Though no published data as of yet appear in PubMed that suggest using these extracts to treat RLS, a number of proprietary herbal formulas do contain significant percentages of the herb and are sold specifically to treat RLS.24
