Sleep Medicine: Restless Legs Syndrome

Abstract

Restless Legs Syndrome is a highly prevalent sensorimotor disorder characterized by urge to move the legs due to discomfort that primarily happens in the evening or at nights. Although the exact pathophysiology remains unclear, brain iron deficiency and altered dopaminergic function appears to play an important role in the pathogenesis of this condition. This disorder affects women more frequently and is associated with significant morbidity.

Introduction

Restless legs syndrome/Willis Ekbom disease (RLS/WED) is a complex sensorimotor disorder with a high degree of clinical variability. This disease is associated with an uncontrollable urge to move due to painful and uncomfortable sensations in the legs. Rarely, the symptoms can be experienced in the arms and in the trunk. RLS is very commonly associated with periodic limb movements in sleep. The principal abnormality in RLS appears to be dopamine deficiency and brain iron dysregulation. In the last few years, multiple candidate genes have been discovered that increase the risk of RLS. Secondary causes of RLS should always be excluded when evaluating patients with symptoms of RLS. Dopaminergic therapies are the mainstay treatment for RLS though their prolonged use is associated with augmentation. This article provides a clinically useful overview of the disease, pathophysiology, the differential diagnoses and treatment strategies.

History of the disease

The disease was first described by the English physician Sir Thomas Willis in 1672 where he emphasized that people with RLS experience sleep disruption and limb movements. The term ‘fidgets in the legs’ has also been used as early as the early nineteenth century. Wittmaack, a German neurologist labelled it ‘anxietas tibiarum’ which was also used by Beard, Bing and Oppenheim in their writings. They considered this a form of hysteria or a sign of neurosis¹. Almost three centuries after its initial description by Willis, a Swedish neurologist, Karl- Axel Ekbom coined the term “Restless Legs” and provided a comprehensive report of this condition in his doctoral thesis, Restless legs: Clinical study of hitherto overlooked disease². Ekbom based his thesis on 34 personal observations and favored vascular cause and even recommended vasodilators as prime therapy¹. He observed that the disease has proclivity to affect women and described the familial tendency seen in some cases. Ekbom’s work was largely ignored until it was rediscovered by Arthur S. Walters and Wayne A. Hening in the 1980s. Periodic limb movements in sleep (PLMS) were first recognized by Symonds in 1953 and were recorded on a polysomnogram by Lugaresi et al in 1965. This polysomnographic documentation of PLMS was a crucial step in supporting the organic nature of RLS. The first RLS diagnostic criteria were established in 1995 and revised in 2003. In 2013, the Restless Legs Syndrome foundation renamed itself the ‘Willis-Ekbom Disease Foundation’ and encouraged the use of the name Willis Ekbom disease.

Epidemiology

Once thought to be a rare disorder, RLS is now considered a common condition, with considerable clinical variability. The prevalence of RLS varies between different regions. Over 50 population-based prevalence studies have shown that RLS affects approximately 5–10% of European and North American adults with about 2–3% experiencing moderate to severe symptoms. However, they indicate a lower prevalence of the disease in Africa and Asia and among Hispanics. The overall pediatric prevalence is 2–4% in the United States. Based on a population-based survey, the prevalence of RLS was 3% between ages 18 and 29, 10% between ages 30 and 79 and 19% in people older than 80 years of age. The mean age of onset is during the third or the fourth decade³.

Clinical Features of Restless Legs syndrome

The core feature of this disorder is uncomfortable and irresistible need or urge to move the legs with accompanying unpleasant and sometimes, painful sensation in the legs. These symptoms may extend beyond the legs and involve arms or the trunk. Rest typically brings on the symptoms while movement relieves them. In children, these symptoms are frequently misdiagnosed as growing pains. Early onset RLS (<45 years at onset) usually starts with intermittent symptoms and progresses slowly. Late onset RLS is usually stable at onset or rapidly progresses over five years to a stable pattern. The symptoms in RLS can be unilateral or bilateral. Up to half the people with RLS complain about restlessness in the arms as well. Typically, this happens as the disease worsens in severity. RLS is typically associated with semi rhythmic leg movements during sleep that are referred to as ‘Periodic Limb Movements in Sleep’ (PLMS). These symptoms of RLS can profoundly affect sleep resulting in insomnia, daytime fatigue, reduced concentration and mood problems (Figure 1).

Figure 1.

Core Clinical symptoms of Restless legs syndrome

Pathophysiology of RLS

Ekbom, in 1945 postulated that microvascular alterations in the leg might be the cause the symptoms of RLS. He proposed chronic venous insufficiency as a potential mechanism in his original disease description. However, with the advent of dopaminergic drugs, the research focus shifted towards analyzing the regulation of brain dopamine. Based on the current research, we discuss the pathophysiological mechanisms that might be operating in the pathogenesis of the disease.

Iron Dysregulation

Clinically, iron deficiency is based on the bone marrow iron concentration with hemoglobin and other serum measures developed to correlate with bone marrow iron levels. Iron homeostasis is organ specific. In the brain, the iron homeostasis is regionally regulated, based on the local cellular energy demand and its accessibility from the blood brain barrier. Low brain iron, despite having normal levels of peripheral iron is the best-established neurobiological abnormality in RLS. Cerebrospinal fluid (CSF) ferritin is lower in patients with RLS than controls. Multiple studies have shown that lower ferritin is associated with increased severity of the disease. Two studies found a reduction in CSF ferritin, along with an increase in transferrin, in RLS patients even though the serum ferritin and transferrin levels did not differ significantly from healthy controls. Diet induced brain iron deficiency in rodents represents a well-accepted pathophysiological model of RLS.

Dopaminergic Dysfunction

Dopaminergic dysfunction has been thought to be strongly associated with RLS since the symptoms of RLS respond well to dopaminergic medications. Similarly, dopamine antagonists worsen RLS symptoms. Brains of RLS patients show an increased dopamine production along with increased dopaminergic activity. However, there is a reduction in the dopamine D2 receptors and dopamine transporters in the striatum suggesting inadequate dopaminergic stimulation.

Role of Glutamate and Adenosine

When the brain iron is experimentally reduced, an increase in glutaminergic activity can be seen. Based on an MR spectroscopic study, a similar increase in glutaminergic activity has been noted in RLS patients. Drugs that decrease glutaminergic activity like NMDA receptor antagonists and α2δ ligands reduce the sensor y symptoms of RLS.

Adenosine Hypothesis

Recent studies have shown that brain iron deficiency in rodents is associated with a hypoadenosinergic state with downregulation of adenosine A1 receptors in the striatum and cortex⁴. This hypo-adenosinergic state may provide a mechanism to explain PLMS secondary to hypersensitivity of cortico-striate terminals⁵. Similarly, the hyperarousal in RLS may also be due to the downregulation of the A1 receptors in the cortex, basal forebrain or hypothalamus^4,6.

Genetics of RLS

RLS is a classical complex genetic disorder in which common and rare genetic variants contribute to the phenotype. A population based survey in Canada reported concordance rates of RLS to be 54% in monozygotic twins and 19% in dizygotic twins demonstrating a substantial genetic contribution in RLS⁷. Genome Wide Association Studies (GWAS) made it easier to identify genetic variants associated with increased risk of RLS. For RLS, GWAS performed between 2007 and 2011, resulted in the identification of common risk variants in six genomic regions. MEIS1 locus is by far the strongest risk factor for RLS with an odds ratio around 2.7^8–10. In cases of Periodic limb movements in RLS, variants in BTBD9 and MEIS1 are associated with a higher risk. The currently known risk variants still account for only a small proportion of the genetic component of RLS. Most of the genetic risk is contributed by rare variants, which need to be clearly characterized.

Neuropathology of RLS

Autopsy studies show that RLS is not associated with neurodegeneration, gliosis, or formation of pathological entities like plaques and inclusions in the brain or spinal cord. RLS brains show a decreased staining for iron with a robust immunostaining of dopaminergic neurons for tyrosine hydroxylase. Quantitative analysis actually reveals a higher concentration of tyrosine hydroxylase in the substantia nigra¹¹. While iron deficient cells elsewhere typically show increased expression of transferrin receptors, the neuromelanin cells in RLS are deficient in transferrin receptors. There is also decreased staining for ferritin in the neuromelanin cells and an increase in transferrin immunostaining¹². A T-cell protein called Thy-1 that is expressed on neuronal cell membranes is also deficient in RLS brains. This protein is thought to be associated with adherence of the pre- and postsynaptic membranes and promotion of the attachment of vesicles to the presynaptic membrane¹³.

Pediatric RLS

RLS is first described in the pediatric literature in 1994. It has been estimated that more than 1 million children in the US are affected by RLS with a prevalence of 2–4%. The symptoms of RLS in children can be nonspecific that include insomnia, sleepiness during the day, restless sleep, growing pains etc. Family history of RLS is very commonly seen in pediatric RLS and is commonly associated with pediatric limb movements in sleep. However, due to the non-specific nature of the symptoms and due to the kids’ difficulty in describing the symptoms, the disease is often overlooked, and the diagnosis is delayed. A sleep disturbances and the diagnosis of definite RLS is around 11.6 years¹⁴.

RLS in Pregnancy

The overall prevalence of RLS in pregnant women is as high as 21%. This is 10-fold higher than in general population. The prevalence of RLS increases with each trimester. RLS in pregnancy is associated with a lower quality of life and can increase the risk of complications including hypertension, cardiovascular disease and preeclampsia. Some studies suggest that RLS can negatively affect the neonatal outcomes. Women with RLS often report depressive symptoms, increased daytime sleepiness and insomnia¹⁵.Multiple factors including iron deficiency, hormonal changes and ethnic differences probably contribute to RLS in pregnancy.

RLS and Other Disease Associations

RLS is a complex, heterogeneous disorder with genetic and environmental contribute to its phenotype. Systematic reviews of the association between RLS and other disorders suggest that RLS is highly prevalent in iron deficiency and in renal disorders while there is also some association between RLS and other disorders. Other conditions most consistently associated with RLS in epidemiological studies are major depressive disorder, generalized anxiety disorder, panic disorder, attention-deficit hyperactivity disorder. Below, we briefly discuss the association of RLS with a few select disorders.

RLS and Attention Deficit Hyperactivity Disorder

Evidence from the clinical studies demonstrates an association between RLS and Attention Deficit Hyperactivity Disorder (ADHD). The degree of association between these two conditions differ from one study to the other¹⁶. The RLS symptoms are usually underdiagnosed in children¹⁷ and both these conditions might share a common dopamine dysfunction.

RLS and Renal disease

RLS is one of the most troublesome conditions experienced by end-stage renal disease patients treated with hemodialysis. The recorded prevalence of RLS in renal disease ranges from 15% to 68%¹⁸. Nearly all the studies investigating RLS in renal disease patients employed an observational, cross-sectional design, evaluating dialysis patients for symptoms of RLS. Factors associated with an increased risk for RLS included increase intake of calcium antagonists, lower baseline levels of intact parathyroid hormone, duration of dialysis, lower serum transferrin saturation, type II Diabetes Mellitus, the number of associated co-morbidities, lower education levels and Caucasian ethnicity¹⁸. Successful renal transplant in end-stage renal failure patients can substantially improve the course of RLS¹⁹.

RLS and Cardiac Disease

Two cross-sectional studies detected higher numbers of patients with RLS among individuals with lower coronary flow (27%) compared with those with normal coronary flow (17%). In women with duration of physician diagnosed RLS > 3 years, the Odds Ratio of non-fatal myocardial infarction and fatal cardiovascular disease (CVD) were 1.80 and 1.49 respectively¹⁸. However, large epidemiological cohorts assessing RLS using three well-known questionnaires could not detect any increased risk of CVD in patients with RLS. Using interviews for diagnosis of RLS, the Wisconsin Sleep cohort found an association between RLS and CVD²⁰. Similarly, a significantly increased risk of cardiovascular disease (22.7%) was reported in Taiwanese patients assessed by validated phone interviews¹⁸.

RLS and Other Neurological Diseases

In migraine patients, the prevalence of RLS as a comorbid disorder ranges from 8.7% to 39%²¹. Studies investigating the risk of stroke in RLS patients did not find any higher incidence of stroke in RLS patients. However, poststroke patients may develop an increased prevalence of RLS, which might be related to cardiovascular risk factors. Patients with de novo, untreated Parkinson disease (PD) did not differ from controls in terms of RLS prevalence. However, RLS prevalence in PD patients who are treated is higher than expected. Majority of the studies report an increased prevalence of RLS in patients with polyneuropathy²². Several studies reported a higher frequency of RLS in patients with MS with the reported frequencies varying from 13.3% to 65.1%²³.

Diagnosis and Differential Diagnosis of RLS

The diagnosis of RLS is clinical. There is no conclusive diagnostic test. It is important for the clinician to identify any secondary cause for the symptoms of RLS. A polysomnogram is not required in the diagnosis though the presence of periodic limb movements during the sleep study would strongly suggest the existence of RLS. A positive family history of RLS supports the diagnosis (Table 1).

Table 1.

Differential diagnoses of RLS. The diagnosis of idiopathic RLS should be when the symptoms are not explained by other etiologies.

Nocturnal Sleep related Leg Cramps Peripheral Neuropathy Habitual/Anxious foot tapping Myalgias Painful leg/feet and moving toes Neuroleptic induced akathisia

Laboratory testing should screen for iron deficiency and renal/hepatic dysfunction. Routine imaging studies in idiopathic RLS are normal and are not indicated. However, most of the MRI studies using iron-sensitive sequences suggest a state of reduced brain iron content in RLS patients. PET and SPECT studies consistently support dysfunction of dopaminergic pathways involving both nigrostriatal and mesolimbic pathways²⁴.

Treatment of Restless Legs Syndrome

Dopaminergic medications have been the mainstay treatment of Willis-Ekbom disease. Prior to the advent of L-dopa and dopamine agonists, benzodiazepines were the treatment of RLS symptoms for a long time. Some cases were tried on baclofen and antiepileptic drugs (gabapentin, carbamazepine, valproic acid) with some success¹. In the United States, three dopamine receptor agonists (DAs) have been approved based on numerous large placebo controlled studies that showed therapeutic efficacy with relatively minor side effects²⁵. Table 2 shows the various DAs used in the treatment of RLS symptoms along with the initial starting doses and the maximum recommended doses. Levodopa should be avoided due to its high incidence of Augmentation.

Table 2.

Dopamine agonists used in the treatment of RLS.

	Initial Dose	Maximum recommended dose	Half-lives
Premipexole	0.125mg/day	0.75mg/day	8–12h
Ropinrole	0.25mg/day	4mg/day	6h
Rotigotine	1mg/day	3mg/day	24hr transdermal system

Along with DAs, α2δ ligands also are used in the treatment of RLS. They can be used along with DAs or alone. They can be especially helpful in patients with RLS who also have peripheral neuropathy as Gabapentin and pregabalin help with the positive symptoms of symptomatic neuropathy. Table 3 shows the α2δ ligands used in the treatment of RLS along with the recommended starting doses.

Table 3.

α2δ ligands used in the treatment of RLS symptoms

	Starting dose < 65 yrs	Starting dose < 65 yrs	Usual effective dose
Gabapentin	300 mg	100mg	900–2400mg
Pregavalin	75mg	50mg	150–450mg
Gabapentin Enacarbil	600mg	300mg	600mg

In pregnant women, non-pharmacological options like exercise, yoga, pneumatic compression devices are the first line of therapy. Co-existing sleep disorder and other aggravating factors should also be addressed. Pharmacological options in pregnancy include iron therapy, clonazepam, gabapentin and Levodopa. DAs are to be avoided in pregnant women.

Perampanel acts as a non-competitive selective antagonist at the postsynaptic ionotropic AMPA glutamate receptor and was found to be effective in a two-month open-label trial in patients with idiopathic RLS²⁶. Future controlled studies might open the possibility of Perampanel becoming a promising alternative to dopaminergic therapies.

Dipyridamole is a non-selective inhibitor of equilibrative nucleoside transporters ENT1 and ENT2 causing an increase in extracellular adenosine and was found to be effective in a small open label clinical trial to improve both sensory and motor symptoms of RLS along with sleep²⁷. This study suggests that adenosine can also be a target for therapeutic inter vention in RLS. Direct targeting of A1 receptors, however, does not seem to be a therapeutic option due to significant peripheral effects of bradycardia and hypotension.

High frequency deep brain stimulation (DBS) is frequently used to treat a variety of movement disorders including PD, essential tremor, and dystonia. The most commonly used targets for DBS are Globus Pallidus Interna (GPi), Subthalamic Nucleus and Ventro- Intermediate Nucleus of Thalamus. Improvement in the symptoms of RLS was reported in a cases of PD and dystonia^28,29 treated with DBS with GPi as the target. Few cases of refractory RLS treated with DBS, with GPi as target, showed improvement in the symptoms of RLS³⁰.

Non-Pharmacological Treatments for RLS

Multiple nonpharmacological treatment options exist to help with the symptoms of RLS. They include exercise, counter stimulation therapy devices (hot/cold baths, warm blankets, limb massagers, compression stockings, counter pulsation devices), near infrared light therapy and acupuncture. Each has varying levels of evidence of therapeutic benefit and can be used in conjunction with the standard treatment.

Augmentation

Augmentation is a feature of RLS patients on treatment with dopaminergic medications where the severity of the symptoms is worse than before the initiation of medication. Discontinuation of the medication results in amelioration of the symptoms. This phenomenon was first described by Allen and Earley³¹ in 1996 in patients receiving levodopa/carbidopa for the treatment of RLS. In their study, about 3/4^th of the patients with RLS who are on levodopa/carbidopa developed worsening of symptoms. Current evidence suggests that the prevalence of augmentation is close to 50% in patients using DA receptor agonists for approximately 10 years³². (Table 4).

Table 4.

Comparison between Augmentation, end of the dose rebound of RLS symptoms, Tolerance to treatments, Natural progression of the disease and symptom profile in the setting of exacerbating factors.

	Augmentation	End of the dose Rebound	Tolerance	Natural Progression of the disease	Exacerbating factors
Worse before Rx	Yes	Yes	No	Yes	Yes
Spread to arms	Yes	No	No	Yes	Yes
Earlier Onset	Yes	Yes	No	Yes	Yes
Breakthrough at night	Yes	Yes	No	Yes	Yes
Worsening with dose escalation	Yes	No	No	No	No
Improvement with dose reduction	Yes	No	No	No	No

The following strategies can help minimize the risk of augmentation in patients with RLS:

• Use dopaminergic medications intermittently if the symptoms are not present all the time.

• Start with the smallest recommended dose and avoid going beyond the recommended daily maximum dosing.

• Iron deficiency increases the risk of augmentation. Thus, as a standard of care, iron deficiency should always be examined before pharmacological treatment.

• If augmentation develops with symptoms happening earlier, the same dose of dopaminergic medication can be split.

• Consider longer acting dopamine agonist like Rotigotine.

• Consider α2δ ligand and lower the dose or discontinue DAs. Similarly, a low dose opioid medication can be used instead of DAs. Between discontinuing the DAs and initiating alternative treatment, a 10-day washout of DAs may be considered.

Conclusion

Restless legs syndrome is one of the most common movement disorders characterized by an irresistible, often uncomfortable, urge to move the legs. The symptoms have a circadian rhythmicity occurring mainly in the evening and night times. The diagnosis of this condition is purely clinical and secondar y causes of restless legs syndrome should be recognized. Brain iron dysregulation underlies the pathophysiology of this condition. Although dopaminergic treatments have been the mainstay therapy of RLS, they are complicated with the occurrence of augmentation. RLS is associated with many other systemic illnesses and untreated RLS can have significant morbidity.

Figure 2.

Genes implicated in RLS pathogenesis. The most common genetic variants account for only 10% while 90% are rare or unknown.

Biography

Pradeep C. Bollu, MD, Anudeep Yelam, MD and Mahesh Thakkar, PhD are in the Department of Neurology, University of Missouri-Columbia.

Contact: BolluP@health.missouri.edu