CHAPTER 37 RESTLESS LEGS SYNDROME
Restless legs syndrome (RLS) is the most common neurological sleep disorder. It was first described in 1672 and was rediscovered in 1945 by K. A. Ekbom,1 who extensively studied this disorder and contributed important findings that are still relevant today. Major advances in the clinical definition of RLS, in understanding the basic mechanisms, the discovery of genetic factors, and the successful treatments of RLS have increased medical attention in the field of neurology. However, awareness of the disorder among general physicians is still low, despite its high prevalence and significant morbidity. RLS is often misdiagnosed, underdiagnosed, or undiagnosed. One survey showed that the rate of correct diagnosis of RLS by general physicians is less than 7% of that of correct diagnosis by specialists. Even when diagnosed, RLS is often not appropriately treated.2 However, the prevalence of individuals whose symptoms of RLS are sufficiently severe for them to seek medical advice is estimated at approximately 3% of general practice populations.2 Increased awareness will improve the rate of recognition and treatment of RLS, which most often dramatically improves patients’ lives.
CLINICAL FEATURES AND DEFINITION OF RESTLESS LEGS SYNDROME
RLS is a distressing neurological sensorimotor disorder characterized by an almost irresistible urge to move the limbs that is most often but not necessarily accompanied by uncomfortable sensations in the legs. RLS symptoms are evoked by rest, such as lying down or sitting, and are worse in the evening or night. Patients describe the sensations, most commonly deep in the legs, as “creepy-crawly,” “like an electric current,” “pulling,” “tearing,” “itching bones,” “aching,” or “throbbing.” Because the symptoms are unfamiliar, patients often find them extremely difficult to describe. The regions between the knees and the ankles are especially affected, whereas the feet are frequently spared. In response to the discomfort, patients often rub or stretch and flex their legs, turn in bed, or pace the floor, because passive or active movement characteristically provides temporary relief. The arms may also be involved, particularly in patients with severe RLS. Because the symptoms of RLS occur predominantly in the evening or during the night, RLS has a significant effect on sleep, and patients often present with a sleep problem. In addition to difficulty initiating sleep, many patients with RLS have problems maintaining sleep, with frequent awakenings or short arousals that result in poor sleep efficiency. Patients with moderate to severe RLS may sleep on average less than 5 hours per night and may chronically have less sleep time than do patients with almost any other persistent disorder of sleep.3 In many patients with RLS, quality of life is poor, and these patients are at increased risk for depressive and anxiety disorders.4,5 Approximately 80% to 90% of the patients with RLS have associated nocturnal, involuntary periodic limb movements (PLMs)—during sleep (PLMSs) or during wakefulness (PLMWs)—which are usually present in the legs.6 PLMs are mostly rhythmic extensions of the big toe and dorsiflexion of the ankle that resemble the Babinski reflex, with occasional flexion of the knee and hip.7 Movements are often bilateral, involving both legs, but may be predominant in one leg or alternate between legs. The quantification of PLMs is routinely performed in the sleep laboratory by recording both anterior tibialis muscles with surface electrodes. Polysomnographic recordings show that they tend to occur every 20 to 40 seconds and are frequently associated with arousals or complete awakenings. Thus, PLMs further contribute to sleep fragmentation, especially in advanced cases in which the PLMS arousal index (number of PLMS-associated arousals per hour total sleep time) may be very high. PLMs may be assumed if bed partners of patients with RLS report nocturnal kicking of the leg or sometimes if patients themselves experience involuntary jerks of the legs while lying in bed.
EPIDEMIOLOGY
Well-performed epidemiological studies have concordantly shown that the prevalence of RLS in the general population is between 5% and 10%8–13 and that RLS prevalence is higher in women with more births.2,9,12–15 In addition, one study showed that the frequency of RLS in women increases by the number of births.13 The time of onset varies between early childhood and 80 years of age or even older. In retrospective surveys, 23%16 and 13%6 of adult patients with RLS reported the onset of RLS before the age of 10 and 25% between 10 and 20 years.6,16 Although valid data about the prevalence of RLS in adults exist, respective data for the adolescent population or children are lacking. However, existing data suggest that RLS and sleep disturbances caused by RLS are a frequent but undiagnosed problem in this younger age group. RLS varies in severity, and the prevalence of adults in a general practice population whose RLS is severe enough to seek medical advice has been shown to be about 3.4%.2 Despite the prevalence, there is a lack of awareness of RLS among health care providers. One epidemiological study showed that only 12.9% of cases were accurately diagnosed by the general practitioner, although patients explicitly reported their symptoms. The rate of correct diagnosis of RLS by general practitioners is less than 7% that by specialists. Even when diagnosed, RLS is often not appropriately treated.2
Primary Restless Legs Syndrome
Most affected individuals suffer from primary RLS, which shows a familial association in more than 50% of cases.17–19 Patients with familial RLS have an earlier onset of symptoms than do those suffering from the sporadic type, and the progression is slower. The relevant age at onset of the disease at which to subdivide patients with RLS into early- and late-onset groups remains to be investigated. One study revealed that patients who were younger than 45 years at onset have a greater probability of having an affected first-degree relative than do those who were older than 45 years at onset.20 In another study, early-onset RLS (with onset at younger than 30 years) has been found to be genetically different from late-onset RLS: In early-onset RLS, it was suggested that a major gene is segregating in an autosomal dominant mode of inheritance with an additional multifactorial component.19 Genomewide studies have been conducted to map genes that play a role in the vulnerability to RLS. So far, linkage was found to a chromosomal locus on 12q,21 which was not confirmed in four other families; on 14q,22 which was replicated in one family23; and on 9p.24 Inspection of huge pedigrees indicates anticipation and variable penetrance.25,26 Investigations of candidate gene coding for receptors and enzymes related to dopaminergic transmission27,28 revealed no differences in the genotypic or allelic distributions between RLS and control subjects.
Secondary Restless Legs Syndrome
Although most RLS cases may be idiopathic, RLS is often linked to other medical or neurological disorders. The most important associations of RLS are with end-stage renal disease29,30 and with iron deficiency.31,32 RLS may also develop during pregnancy33,34 or may intensify after treatment with various drugs. In many reports, RLS has been linked to dopamine antagonists, such as typical and atypical neuroleptics and metoclopramide, or to antidepressants, such as tricyclic and tetracyclic antidepressants, serotonin reuptake inhibitors, and lithium. Although supporting data are limited, peripheral neuropathies, such as axonal neuropathy,35 cryoglobulinemic neuropathy,36 familial amyloid polyneuropathy,37 Charcot-Marie-Tooth disease type 2,38 and small-fiber neuropathies39 may be associated with RLS. Because the prevalence of RLS among patients with peripheral polyneuropathies has been shown to be only 5.2%40 and thus does not exceed that in the general population, this association is of unclear significance. Other neurological diseases that have been reported to be linked with RLS are radiculopathies41 and myelopathies, such as multiple sclerosis or syringomyelia.42 It has also been shown that some children who were diagnosed with attention deficit/hyperactivity disorder may have PLMS or in fact suffer from RLS.43–45
PATHOPHYSIOLOGY
Although RLS pathophysiology is still unknown, knowledge about it has developed significantly, which has led to diverse hypotheses.46 The medication responses indicate that RLS results from abnormal functioning in the nervous system.47 Overall, the central nervous system areas involved in RLS pathophysiology are unknown, and the role of the peripheral nervous system, if any, is uncertain.48 Several studies have provided evidence that RLS is associated with peripheral neuropathy; therefore, it is possible that peripheral nerve lesions are a trigger for RLS.
Dopaminergic Function in Restless Legs Syndrome
Dopaminergic mechanisms are supposed to play an important role, particularly inasmuch as dopaminergic drugs have shown to be especially beneficial in RLS. In addition, the circadian manifestation of RLS symptoms coincides with lower levels of central dopamine activity.49 However, several investigations, including positron emission tomography and single photon emission computed tomography studies of the dopaminergic nigrostriatal system,50 cerebrospinal fluid analysis of dopaminergic metabolites,51,52 and even histopathological studies, have not provided evidence of a primary dopaminergic deficit or neurodegeneration in the basal ganglia53 that favors a functional impairment or a modulating effect of the dopaminergic system. The fact that melatonin secretion, which is increased at night, is correlated with the severity of RLS symptoms54 supports the dopamine hypothesis, inasmuch as melatonin exerts an inhibitory effect on central dopamine secretion. Daytime and nighttime melatonin excretion per se is normal in patients with RLS.55
Iron Metabolism in Restless Legs Syndrome
A strong negative correlation between serum ferritin levels and RLS severity,31,32 the therapeutic effect of iron in RLS,56 the reduction of iron in the substantia nigra as shown by magnetic resonance imaging,57 and reduced cerebrospinal fluid ferritin levels and increased cerebrospinal fluid transferrin levels58 in patients with RLS indicate that an impaired iron metabolism may be another important factor in the pathophysiology of RLS. Neuropathological investigations have shown that iron and heavy-ferritin staining was markedly decreased (and light-ferritin staining was strong) in the substantia nigra. In addition, transferrin receptor staining on neuromelanin-containing cells was decreased in brains of patients with RLS, whereas transferrin staining in these cells was increased.59 These data suggest that iron acquisition is compromised in neuromelanin cells, which may interfere with dopaminergic mechanisms.
Role of the Nociceptive System
Stiasny-Kolster and colleagues (2004)60 showed that the nociceptive system is involved in RLS pathophysiology. Patients with RLS exhibited a profound static mechanical hyperalgesia in response to punctate stimuli. This type of hyperalgesia improved significantly after long-term treatment with dopaminergic drugs. These findings suggests that RLS may be associated with central sensitization of spinal neurons, such as that in chronic neuropathic pain resulting from abnormal peripheral input and/or from altered descending inhibition (i.e., involving the supraspinal dopaminergic system). Thus, in addition to being a motor and sleep disorder, RLS may also be a pain disorder. These findings may also explain why substances that are well accepted for the treatment of neuropathic pain, such as opioids or anticonvulsants, alleviate RLS symptoms.
Localization of the Dysfunction
Regardless of the neurotransmitter systems and neurophysiological mechanisms involved in RLS pathophysiology, the site of the underlying neurological alteration is still unclear. Cortical involvement was previously considered unlikely because of the absence of cortical activity in functional magnetic resonance imaging (fMRI) studies and the absence of a Bereitschaftspotential61 associated with the involuntary movements in RLS. Spectral power analysis of electroencephalographic studies, however, provides evidence that characteristic involuntary leg movements in RLS are preceded by a preparatory cortical activation. In addition, several neurophysiological studies with single or paired pulse stimulation showed a reduced intracortical inhibition62 and a shortened cortical silent period in patients with RLS.63,64 This cortical disinhibition may result from a shift in the balance of excitatory and inhibitory mechanisms toward either more excitability or less inhibition (e.g., as a result of increased sensory input or impaired subcortical mechanisms). Interestingly, levodopa (L-dopa) leads to a prolongation of the silent period and thus to an improvement of impaired inhibitory mechanisms.64 Some studies have suggested the involvement of subcortical areas in RLS pathophysiology. Lesions in the dopaminergic diencephalospinal tract (A11 neurons) have been proposed as an animal model for RLS65 and are discussed as potential underlying cause of RLS in humans.66 The effects of acute dopaminergic treatment on RLS symptoms and long-term dopaminergic treatment on static mechanical hyperalgesia support this suggestion. Thus, the dopaminergic system is either directly or indirectly involved in the central sensitization in RLS.60
The only fMRI investigations in patients with RLS conducted so far67 revealed a bilateral activation of the cerebellum and a contralateral activation of the thalamus during sensory symptoms. During the occurrence of PLMs and sensory symptoms, the patients showed additional activation in the red nuclei and brainstem close to the reticular formation. These results support aforementioned neurophysiological studies that suggested that subcortical cerebral generators are involved in the pathogenesis of RLS. During the combined condition (PLMs and sensory symptoms), the activation was still larger in the cerebellum than during each condition alone, which points to a neuronal participation of the cerebellum in the motor and sensory process as well. This finding is in agreement that in healthy subjects, the cerebellum is involved in sensory processing. When the distribution of the fMRI-activated cerebellar areas (subcortical in both hemispheres) were compared with histological autoradiographical studies of the human cerebellum, a similar distribution of opioid μ receptors could be detected. Thus, the cerebellum could be another possible site of interaction of opioidergic agents. The spinal cord may also play a major role in the pathophysiology of RLS. Patients with spinal cord lesions, particularly those with transsection of the spinal cord, often have PLMs, which suggests that PLMs are generated in the spinal cord.68–71 In these cases, however, the number of PLMs is lower and the circadian rhythmicity and the response to dopaminergic agents are less pronounced than in patients with RLS.69,72,73 Flexor reflex studies showed an increased spinal cord excitability during sleep in patients with RLS, which indicates that signal processing in the spinal cord is altered. In addition, the presence of mechanical hyperalgesia in RLS indicates central sensitization of spinal neurons. However, the central sensitization in RLS may also be based on afferent input-induced plasticity of spinal nociceptive transmission,74 and long-standing abnormal peripheral input may explain various kinds of secondary RLS: for example, RLS triggered by polyneuropathies. Thus, it is conceivable that at least in a subgroup of patients with RLS, peripheral lesions may represent a pathophysiological factor for the development or aggravation of RLS.
DIAGNOSTIC CRITERIA FOR RESTLESS LEGS SYNDROME
RLS is a clinical diagnosis that relies entirely on the patient’s symptoms. According to international diagnostic criteria75 (Walters, 1995) and a consensus established by the International Restless Legs Syndrome Study Group76 (Allen et al, 2003) RLS is characterized by four essential criteria (Table 37-1). To make a definite diagnosis of RLS, all four diagnostic criteria must be established. In some cases, these criteria may be inadequate for ruling out conditions mimicking RLS, such as peripheral neuropathy, leg cramps, positional discomfort, or parkinsonian symptoms, inasmuch as these conditions may sometimes lead to positive answers referring to different problems. If these features are comorbid with RLS, it can, in turn, also be difficult to recognize RLS.
TABLE 37-1 Essential Diagnostic Criteria of the International Restless Legs Syndrome Study Group
From Allen R, Picchietti D, Hening WA, et al: Restless legs syndrome: diagnostic criteria, special considerations, and epidemiology. A report from The Restless Legs Syndrome Diagnosis and Epidemiology Workshop at the National Institutes of Health. Sleep Med 2003; 4:101-119.
The following supportive features that are not necessary to make the diagnosis RLS but may, especially in doubtful cases, help to diagnose or exclude RLS have been established.76
Periodic Limb Movements in Sleep
PLMSs are reported to occur in 80% to 90% of patients with RLS. However, PLMS also commonly occur in other disorders and in the elderly. A PLMS index (number of PLMSs per hour of sleep) of greater than 5 is considered pathological, although data supporting this feature are very limited. The occurrence of PLMs during nocturnal periods of wakefulness (PLMW) is considered to be more specific for RLS.77 Thus, the presence of a high number of PLMs is supportive for RLS, but the absence of PLMs does not exclude RLS.
In addition to the essential and supportive criteria, the natural clinical course, the presence and character of sleep disturbances, and the physical examination findings are other features that may be helpful for diagnosis of RLS.76
Natural Clinical Course
RLS is usually a chronic disease, and symptoms typically increase over time. As mentioned, the age at onset of RLS varies widely, from childhood to 80 years of age and even older. Patients with early-onset RLS are more likely to have affected family members than are those with late-onset RLS.19 When RLS symptoms appear at a young age, the onset is more often insidious. RLS symptoms may be present intermittently for many years with fluctuating severity. Initially, mild RLS may occur in single nights, sometimes with or without sleep disturbances, or in aggravating situations during the day, such as during a long overseas flight. There may be long periods of remission, but the severity and frequency typically increase over time. Many patients with RLS experience their RLS as nightly events with difficulty falling asleep and several awakenings with walking around or other actions that may lead to temporary relief, such as taking a cold foot bath. Other patients progress to daily symptoms within several years. Many of these patients have disease onset in late adult life and have secondary RLS. The natural clinical course may be influenced by the development of augmentation as a result of medication side effects (see following discussion).
