Clinical manifestations

Presentation and course

Diagnostic criteria for the diagnosis of restless legs syndrome include (1) an urge to move the legs usually accompanied or caused by uncomfortable and unpleasant sensations in the legs, (2) the urge to move or unpleasant sensations begin or worsen during periods of rest or inactivity, such as lying or sitting, (3) the urge to move or unpleasant sensations are partially or totally relieved by movement, such as walking or stretching, at least as long as the activity continues, and (4) the urge to move or unpleasant sensations are worse in the evening or night than during the day or only occur in the evening or night. The occurrence of the above features is not solely accounted for as symptoms primary to another medical or behavioral condition (eg, myalgia, venous stasis, leg edema, arthritis, leg cramps, positional discomfort, habitual foot tapping) (18; 333).

Supportive clinical features include a positive family history, response to dopaminergic therapy, and periodic limb movements during wakefulness or sleep. Additional criteria have been defined for cognitively impaired elderly persons and for children (19). Restless legs syndrome is often considered to have a mean age of onset in the sixth decade; however, in a review of 133 patients studied with surveys and polysomnography, mean age at onset was 27.2 years, whereas onset was before 20 years of age in 38.3% of cases (214). In another study, as many as 33% of patients had onset before 10 years of age (328). The syndrome is rarely diagnosed before 20 years of age, although in some patients, the onset of restless legs symptoms may be traced to infancy. In one study of 23 children, 20 (87%) had a family history of restless legs compatible with an autosomal dominant inheritance pattern (220).

The sensory symptoms of restless legs syndrome are paresthesias and dysesthesias in the calves and legs. Descriptions of these phenomena include words such as aching, pulling, drawing, numbness, tingling, prickling, creeping, or crawling. The unpleasant sensations occur during rest and inactivity and are worse in the late evening and when attempting to sleep. They are relieved by moving the legs or walking; patients pace the room, toss and turn in bed, and rub their feet and legs (105).

Body rocking and marching in place may also be seen, particularly in elderly subjects with longstanding severe disease (326). Occasionally, rhythmic movement disorders consisting of stereotypic movements in sleep, such as body rocking, head banging, and head rolling, may be noted in severe restless legs syndrome (323). Although it would seem that the motor phenomena occur solely in response to the unpleasant sensations, recordings during forced immobilization suggest that muscle activity may occur independent of the unpleasant sensation (248). About one-half of patients with restless legs syndrome also complain of restlessness in the arms (203). Restless genitalia has been described in one patient with Parkinson disease and responds well to low-dose dopamine agonists (22).

Most patients with restless legs syndrome have other symptoms of sleep disorder. Most common are periodic leg movements of sleep, present in 80% of patients (218). In addition, over 55% of patients report daytime crises frequently associated with reduced daytime activity (318). Periodic leg movements of sleep are stereotyped movements of the legs occurring in long trains. The movements always include dorsiflexion of the big toe or foot. More intense movements include flexion of the knee and hip, and when fully developed, resemble the triple flexion withdrawal response of pyramidal disease. Periodic arm movements are also common (50). Although sometimes called nocturnal myoclonus, the movements do not look like other forms of myoclonus. Movements tend to recur every 20 seconds to 40 seconds and are often associated with a K-complex on EEG and increases in heart rate, blood pressure, and depth of respiration (190). The frequency of leg movements is related to sleep stage. Movement frequency and duration decrease with progression from wakefulness through stage 1 and stage 2 of non-REM sleep, and movements are dramatically suppressed during REM and slow wave sleep (223). The occurrence of periodic leg movements of sleep may be related to an underlying circadian rhythm, the sleep stages, and body position (309). For example, peak intensity of subjective discomfort and motor restlessness is on the falling phase of the core temperature cycle, suggesting that restless legs syndrome is partially controlled by a process that varies with the normal circadian rhythm (124). The circadian pattern of periodic limb movements in sleep was very similar to the circadian pattern previously reported in restless legs syndrome (79). Suppression of periodic limb movements of sleep with dopamine agonists such as pramipexole did not affect arousal or EEG instability in sleep, whereas the use of benzodiazepines lowered EEG instability and arousal in sleep without affecting periodic limb movements, suggesting that periodic limb movement disorder may not be directly related to arousal from sleep or sleep fragmentation (195). Glutamate/creatine ratio in the thalamus on MR spectroscopy was higher in patients with restless legs syndrome than in controls and correlated with wake time during sleep period but not with periodic limb movements of sleep, adding credence to the hypothesis that periodic limb movements themselves do not contribute to arousals in restless legs syndrome (11).

Despite the clinical similarity, some authors have termed this movement "resting dyskinesias while awake" (253). Variable levels of sleep disruption are reported by patients and especially by their bed partners. Patients with restless legs syndrome have increased sleep latency, decreased sleep efficiency, and frequent awakenings. Nocturnal eating and sleep-related eating disorders can also be seen in restless legs syndrome and may respond to dopaminergic therapy. Amnestic sleep-related eating disorders can occur in restless legs syndrome in the setting of sedative use (132). Headache, anxiety, social isolation, reduced libido, and mild depression have been reported. Some reports suggest significant deficits in cognitive tests sensitive to prefrontal cortex dysfunction (245). Subsequent studies have shown no effect on cognition but have found a strong correlation between untreated restless legs syndrome and depression (175). However, studies have shown that decision-making, problem-solving, and working memory performance seem to improve with dopaminergic therapy in restless legs syndrome (92; 147).

Reflection impulsivity, a function of cognitive impulsivity, is common in restless legs syndrome in drug naïve patients and those on dopaminergic treatment (121). Jumping to conclusions and irrational decision-making are more common in patients with restless legs syndrome on dopaminergic therapy with augmentation (122).

Patients with restless leg syndrome are more likely to show symptoms of depression and higher rates of neuroticism than controls (300). Restless legs syndrome was associated with a high risk of suicide and self-harm, and the risk was independent of associated conditions and comorbidities (348).

Primary restless legs syndrome sufferers had a mean productivity loss of 1 day/week, resulting in increased financial disease burden (13). In the workplace, restless legs syndrome has been noted as a highly prevalent sleep disorder contributing to worker absenteeism, decline in productivity, and employer costs (200).

The impact on partners of patients with restless leg syndrome has been noted concerning sleep, leisure time activity, and quality of relationship (233).

Physical and neurologic examinations of most sufferers of restless legs syndrome are normal (primary restless legs syndrome). However, in some cases, examination or diagnostic studies uncover evidence of systemic or neurologic disease (secondary restless legs syndrome) (236).

Prognosis and complications

Restless legs syndrome is generally a lifelong condition. Symptoms tend to progress with age. Brief remissions and variability over time are common. Restless legs syndrome and periodic limb movements in sleep are reported to be associated with an increased risk of hypertension, cardiovascular disease, and stroke (329). Men with restless legs syndrome have been reported to have a higher likelihood of concurrent erectile dysfunction because dopaminergic hypofunction contributes to both conditions (93).

Clinical vignette

A 73-year-old woman presented for evaluation of gait disorder. Two years before the present evaluation, she began to notice "slapping" of her feet. Evaluations by a podiatrist and a neurosurgeon revealed peripheral neuropathy. Shortly after this diagnosis, she noticed "twitching" in both feet and a sense of restlessness in the feet when resting in bed at night. The weight of the bedclothes seemed unbearable, and she repeatedly got out of bed and walked the floor of her room. The discomfort was relieved by walking, only to recur when she returned to bed. Her medical history was remarkable only for osteoarthritis, hypothyroidism treated with levothyroxine, and remote kidney stones. She was an ex-smoker and moderate alcohol drinker. The family history was negative. Examination revealed signs suggestive of a mild peripheral neuropathy. No involuntary movements were seen. Investigation revealed mild axonal neuropathy and cerebellar atrophy. She was prescribed carbidopa at 50 mg daily, carbidopa-levodopa at 50 mg carbidopa, and 200 mg levodopa at bedtime with good control of the symptoms.

Biological basis

Etiology and pathogenesis

Primary restless legs syndrome is hereditary in at least one third of cases; some authors suggest that 50% to 92% of such cases may be familial (236; 328). In these families, the inheritance pattern is autosomal dominant; however, variable penetrance and evidence of genetic anticipation in some families with restless legs syndrome suggest the disorder’s genetic basis may be heterogeneous (173). There was high concordance in a small study of monozygotic twin pairs in which one pair member had restless legs syndrome (237). However, due to variability in clinical manifestations, particularly symptom severity, family history may be falsely negative. In other cases, the disorder is idiopathic.

A review proposed a continuous spectrum in restless legs syndrome with major genetic contribution on one end and comorbid or environmental contributions on the other end rather than an idiopathic versus secondary classification (301).

Phenotypes.

(1) Early onset. Starts before 36 years of age; familial history is evident; symptoms are more severe; syndrome is highly genetically determined and highly dependent on brain iron levels.

(2) Delayed onset. Starts after 36 years of age; family history is not evident; symptoms are mostly secondary; syndrome is rapidly progressive over 2 to 3 years.

Secondary restless legs syndrome may be due to systemic or neurologic disease. Changes in iron status may be etiologic in at least some patients. Restless legs syndrome is associated with iron deficiency anemia, including frequent blood donation (281). The prevalence of clinically significant restless legs syndrome in iron deficiency anemia is nine times higher than in the general population (10). Another study interviewing 15 blood donors showed no increase in restless legs syndrome (24). Pathologic studies suggest decreased iron and ferritin staining in the substantia nigra of patients with restless legs syndrome (60; 103). Serum and cerebrospinal fluid ferritin levels are lower in patients with restless legs syndrome than controls, and symptoms may respond to ferrous sulfate in those patients with low ferritin (207). Iron deficiency in rodent models has been associated with increased extracellular striatal dopamine, reduced striatal dopamine uptake, and blunted D2 receptor feedback of dopaminergic uptake. An iron-deficient diet in mouse models with brain iron deficiency mutations has shown earlier onset of restless legs syndrome-like symptoms, supporting the possibility of a continuous spectrum in restless legs syndrome with genetic and environmental contributors (16). Iron infusion has been shown to reverse some of these changes, providing a causal link between iron deficiency and restless legs syndrome (320). Levels of serum hepcidin, a key regulator of iron metabolism, as well as hepcidin/ferritin ratio, were found to correlate with periodic limb movements in sleep severity and restless legs syndrome symptoms better than ferritin alone (67). Low ferritin has been associated with developing restless leg syndrome in patients with Parkinson disease (178).

Another common association is between restless legs syndrome and uremia associated with renal failure. Uremic patients appear to have a more rapidly progressive course with a less robust response to dopaminergic treatment (83). As many as 58.3% of dialysis patients have restless legs syndrome, and as many as 70.8% have periodic limb movement disorder (83). The presence of restless legs symptoms correlates to higher levels of urea nitrogen and creatinine and lower levels of intact parathyroid hormone and is associated with increased morbidity and mortality. Positive family history, low or absent residual urine, and presence of peripheral neuropathy were predictors of risk for restless legs syndrome in patients with end-stage renal disease undergoing hemodialysis (252). Renal transplantation appears to decrease the prevalence of symptoms in this population (210). Other patients have frank vitamin deficiency, diabetes mellitus (192), rheumatoid arthritis, gastric or pulmonary disease (225), peripheral arterial occlusive disease, or congestive heart failure (115). Longer duration of dialysis and greater number of dialysis sessions were associated with restless legs syndrome, but not adequacy of dialysis (319).

Restless legs syndrome may complicate singleton or multiple gestation pregnancy (231). Predictors for restless legs syndrome in pregnancy include a family history of restless legs syndrome and a personal history of restless legs syndrome in previous pregnancies (133). The prevalence of restless legs syndrome during pregnancy is two to three times the normal population, the mechanism is thought to be related to hormonal changes, and the iron and folate levels and incidence are influenced by the trimester and the number of parity (286). Pregnancies complicated by restless legs syndrome have a higher risk for preterm birth, suggesting a need for improved sleep health in pregnant women (201). Conversely, a history of pregnancy-induced hypertension was found to be twice as likely to be associated with restless legs syndrome in women (137). An increased prevalence of preeclampsia and cardiovascular disease has also been noted in pregnant women with restless legs syndrome (186). High estradiol, low iron, ferritin, and vitamin D deficiency may play a role in developing restless legs syndrome in pregnancy (110). Additionally, it has been found that magnesium and zinc levels are lower in pregnant women with restless legs syndrome, and the lower levels are correlated with higher rates of adverse perinatal outcomes, such as miscarriage, which suggests that zinc and magnesium may play a role in the etiology of restless legs syndrome in pregnancy (345).

A prospective study suggested transient restless legs syndrome developed in nearly 9% of patients following spinal anesthesia. Those with low mean corpuscular volume and mean corpuscular hemoglobin were most at risk. Restless legs syndrome in these patients was transient, with a mean duration of 33 days (127). Symptoms of restless legs syndrome have been said to complicate the use of certain medications, including second-generation antidepressants, especially mirtazapine (03; 261), zonisamide (54), venlafaxine (160), and olanzapine (154). On the other hand, bupropion may help with restless legs syndrome (160).

Acute restless leg has been described after pontine infarcts, often in the paralyzed limb contralateral to the stroke (315). In a study, the incidence of restless leg syndrome after acute lacunar infarction was 5.33%, with pons, centrum semiovale, and basal ganglia being the common locations. Poststroke restless leg syndrome appeared to be more often unilateral and involved the arm (316). Diagnosis of restless leg syndrome after acute ischemic stroke was found to be associated with increased arterial stiffness and poorer clinical outcome 3 months after stroke (113).

Peripheral neuropathy is the most commonly reported neurologic cause of secondary restless legs syndrome (236). Although most cases are apparent by history or examination, the condition is subclinical in some cases. In eight consecutive patients with restless legs syndrome and no symptoms of peripheral neuropathy, six had clinical or electrophysiological evidence of denervation. Sural nerve biopsies showed endoneurial fibrosis, acute axonal degeneration, and thinly myelinated fibers (135). In some patients with painful polyneuropathy, nociceptive deafferentation may trigger overactivity of spinal structures implicated in restless legs syndrome (99).

Several authors have suggested that the prevalence of restless legs syndrome is increased in patients with Parkinson disease. A large cohort of Parkinson disease cases revealed a similar prevalence of restless legs syndrome in the general population, but restless legs syndrome symptoms may have been masked in patients with Parkinson disease due to dopaminergic treatment for Parkinson disease. In this study, severity of restless legs syndrome was related to the severity of nondopaminergic Parkinson disease symptoms (322). Lang has suggested that restless legs syndrome can be confused with akathisia in this population (168). Restless legs syndrome in Parkinson disease appears to be influenced by a younger onset of Parkinson disease, male gender, higher Mini-Mental State Examination score, and less advanced Hoehn and Yahr stage (27).

A study demonstrated that patients experiencing restless legs syndrome symptoms more than 15 days a month were more likely to be diagnosed with Parkinson disease within 4 years of follow-up, suggesting that restless legs syndrome symptoms may be an early sign of Parkinson disease (340).

Subsequent studies have demonstrated a higher prevalence of restless legs syndrome in patients with Parkinson disease, with a greater impact on mood, nutritional status, and quality of life (87; 344). Patients with Parkinson disease and restless legs syndrome have an older age of onset of restless legs syndrome, shorter duration of symptoms unilaterality in accordance with Parkinson disease symptoms, and lower dosage as well as shorter duration of dopaminergic medication use compared to patients with Parkinson disease without leg restlessness (347). They have also been found to exhibit variants, including restlessness without the urge to move and leg motor restlessness, and the severity of restless legs syndrome has been associated with autonomic symptoms (199).

Leg motor restlessness, defined as an urge to move the legs but not meeting the four essential criteria for restless legs syndrome, was found to be more prevalent than restless legs syndrome in Parkinson disease than controls; however, it is unclear if leg motor restlessness is a distinct entity or a part of restless legs syndrome spectrum (291). Another condition that may resemble restless legs syndrome is leg stereotypy syndrome, which is manifested by repetitive, rhythmical, stereotypic leg movement, especially when sitting, but without the urge to move or diurnal variation, which is typical of restless legs syndrome (143). Similar to restless legs syndrome, leg stereotypy syndrome is almost always familiar, but it is even more common, occurring in 7% of the general population and even more frequently in patients with other movement disorders (189).

On the other hand, there may be evidence to suggest that idiopathic restless legs syndrome may be associated with delayed onset of Parkinson disease, less dyskinesias, and maybe even slower progression (76).

Moccia and colleagues conducted a longitudinal study on restless legs syndrome in Parkinson disease and demonstrated increasing prevalence over time, later onset of Parkinson disease, and worse sleep and cardiovascular measures (208). A prospective study of restless legs syndrome in de novo patients with Parkinson disease showed a higher cumulative incidence of restless legs syndrome than in the general population. Interestingly, at the end of 3-year follow-up, current restless legs syndrome was significantly more prevalent than previously seen in the drug naïve patients with Parkinson disease and controls, suggesting that restless legs syndrome emerging in the course of chronic dopaminergic therapy is a key determinant of the comorbid association with Parkinson disease (196). Restless leg syndrome has also been found to be an independent predictor of impulse control disorders in Parkinson disease (197).

A review raised questions on whether dopaminergic therapy in Parkinson disease may mask the true prevalence of restless legs syndrome, whether restlessness in Parkinson is different from idiopathic restless legs syndrome, and if restless leg symptoms may be an early manifestation rather than a risk factor for Parkinson disease (88).

Restless legs syndrome is more common in siblings of Parkinson disease with an onset age of less than 60 than in the general population, suggesting common pathogenic genetic factors and early living environment for neurodegeneration (187).

Restless legs syndrome appears to be more common than expected in essential tremor, and many of these patients have a positive family history of restless legs syndrome, suggesting some shared genetic factors (234).

Associations have also been reported between restless legs syndrome and startle disease (hyperekplexia), stiff person syndrome, Huntington chorea, spinocerebellar ataxia 3, amyotrophic lateral sclerosis, myasthenia gravis (279), polio and postpolio syndrome (164), brainstem strokes (264), ankylosing spondylitis (293), Tourette syndrome, multiple sclerosis (69; 38), neuromyelitis optica (134), attention deficit hyperactivity disorder (314; 262), migraine (276), Isaacs syndrome (25), obesity (183), chronic obstructive pulmonary disease (188; 23), fibromyalgia (288), diabetes mellitus (108), irritable bowel syndrome (32; 346), inflammatory bowel disease (08), epilepsy (100), endometriosis (294), and cardiovascular events (172).

Bidirectional association has been noted between migraines and restless legs syndrome, with the association being stronger and longer lasting for migraine-triggering restless legs syndrome (55).

A large prospective study did not reveal an association between restless legs syndrome and increased risk of cardiovascular events (338). A prospective study of restless legs syndrome and coronary heart disease in women suggested a marginal increase in risk for cardiovascular events with symptom duration greater than 3 years (180). Men with restless legs syndrome have been found to have higher mortality, even after excluding chronic conditions and adjusting for common risk factors. This association was more frequently seen in respiratory, endocrine, metabolic, and immunological disorders, suggesting the need for further studies to understand the pathophysiology of these associations (181). Diabetes and stroke appear to increase the odds of restless legs in males, whereas alcohol consumption and frequent exercise seemed to lower the odds (337). However, a United States veterans’ study showed a four-fold increase in coronary heart disease and stroke with incident restless legs syndrome compared to controls (211). Then a subsequent case-control study showed no significant association between primary restless legs syndrome and cardiovascular disease or hypertension after controlling for confounders (58). A systemic review also showed that after adjusting for confounders, restless legs syndrome was associated with increased all-cause mortality alone and not cerebrovascular events (150).

In a large, prospective study, obesity and high cholesterol, but not high blood pressure, were significantly associated with an increased risk of developing restless legs syndrome (72). Higher comorbidity of influencing conditions, such as diabetes, hypertension, myocardial infarction, obesity, stroke, cancer, renal disease, anemia, depression, thyroid disease, and migraine, was associated with a greater risk for restless legs syndrome (292).

Patients with restless legs syndrome exhibit a tendency toward hypertension and reduced amplitude of both sympathetic and parasympathetic responses as well as blunted parasympathetic drive to blood pressure changes in autonomic testing, suggesting autonomic nervous system involvement during wakefulness and consequently an enhanced cardiovascular risk in restless legs syndrome (142). Compromised cardiovagal tone and increased peripheral vascular resistance may mediate some of the cardiovascular effects (39). Significant correlations have been noted between periodic limb movements of sleep and nocturnal systolic, diastolic blood pressure elevations (48), restless legs syndrome, and nocturnal systolic blood pressure (280). Frequent restless leg syndrome, along with PLMS index scores of 26/hour or more, was associated with a higher risk of resistant hypertension (123).

Flow-mediated dilation, a measure of endothelial vascular function, has been noted to be affected in restless legs syndrome, suggesting a putative mechanism for vascular effects from restless legs syndrome (158). Poorer endothelial function has also been demonstrated on transcranial Doppler studies of cerebral blood flow in patients with restless legs syndrome (156).

A systematic review found limited evidence for restless legs syndrome as a prognostic factor for cardiovascular events or all-cause mortality; however, periodic limb movements in sleep may be an independent prognostic factor (152). A prospective study that included 57,147 women without cancer, cardiovascular disease, or renal failure at baseline found that restless leg syndrome was not associated with a higher risk of all-cause mortality, but it was significantly associated with cardiovascular disease mortality (182). However, a case-control longitudinal study that followed 235 patients, both men and women over 11 years, showed that age-standardized mortality rates and cardiovascular risk factors were not elevated in patients with restless leg syndrome (64).

Patients with restless legs syndrome exhibited an impairment of the long-term depression-like mechanisms induced by inhibitory repetitive transcranial magnetic stimulation, thus supporting the involvement of GABA in restless legs syndrome pathophysiology and impairment of short-term neural plasticity (170).

Pre-stroke restless legs syndrome was found to be a predictor for subcortical stroke (111), and silent small vessel disease was noted to be more common in patients with long-standing restless legs syndrome (more than 10 years) (90).

Multiple mechanisms may be involved regarding cardiovascular risk in restless legs syndrome, ranging from premenstrual syndrome-mediated vascular changes to the effect of iron deficiency on cardiovascular function and sleep deprivation or fragmentation and its effects on the oxidative, metabolic, and vascular systems (107).

The pathogenesis and pathophysiology of restless legs syndrome are not clear, although the association with peripheral neuropathy suggests that peripheral nervous system dysfunction may be important in the generation of the syndrome. Increased risk of restless legs syndrome in familial amyloid polyneuropathy supports the association between restless legs syndrome and polyneuropathy (295). The circadian rhythm of the movements and the periodicity of associated leg movements of sleep suggest the movements are organized at a higher level of the nervous system.

Electrophysiologic studies suggest that the movements are involuntary and are organized at the brainstem or spinal level (304; 57). Patients with periodic leg movements of sleep, with or without associated restless legs syndrome, may have abnormal blink reflexes, somatosensory-evoked responses, long latency responses, or H-reflexes, suggesting pontine or more rostral dysfunction (198). Functional magnetic resonance imaging studies suggest that the red nucleus and brainstem are generators for the abnormal movements of restless legs syndrome (44). Jerk-locked back averaging reveals no cortical precedent to the movements occurring during sleep (190). During wakefulness, no Bereitschaftspotential (cortical pre-movement) precedes the movements (305). Research subjects with restless legs syndrome have shortened cortical silent period compared to controls as assessed by transcranial magnetic stimulation with recording from the anterior tibialis muscle. Treatment with the dopamine agonist cabergoline normalized the cortical silent period in these subjects (106). Functional imaging studies have shown significantly increased connectivities in the sensory thalamic, ventral, and dorsal attention, basal ganglia-thalamic, and cingulate networks in patients with restless legs syndrome, suggesting that the pathophysiology of restless legs syndrome is beyond the sensorimotor regions (104).

MRI voxel-based morphometry showed increased gray matter volume in the pulvinar bilaterally in 51 subjects with idiopathic restless legs syndrome compared to 51 age-matched control subjects (84). Similarly, a decrease in gray matter volumes has been noted in the left hippocampal gyrus, bilateral parietal lobe, medial frontal areas, and cerebellum (51). Alterations in subcortical gray matter volume in putamen have been noted to correlate with disease duration in restless leg syndrome (179).

Average cortical thickness in the bilateral post-central gyrus and the corpus callosum posterior midbody was noted to be lower in patients with restless legs syndrome than in controls, suggesting altered white matter properties in the somatosensory pathway (174). White matter changes were detected in the left corticospinal tract and cingulum and in the right anterior thalamic radiation and inferior fronto-occipital fasciculus in patients with restless leg syndrome. Also, the number of periodic leg movements and movement arousal index were correlated with alterations in white matter in the corticospinal tract (241).

Neuropathological studies suggest that restless legs syndrome may result from impaired iron acquisition by neuromelanin cells, possibly because of impaired transferrin receptor regulation (60). Subjects with restless legs syndrome have lower serum and CSF iron and ferritin. There is a correlation between these measures in serum and cerebrospinal fluid, but the slope is lower in restless legs syndrome subjects than in normal controls, suggesting disordered transport of iron from the periphery to the central nervous system (207). A postmortem and imaging-based study revealed a decrease in myelin in patients with restless legs syndrome similar to that reported in animal models of iron deficiency (61). The presence of less myelin and loss of myelin integrity in the brains of patients with restless legs syndrome in combination with decreased ferritin and transferrin in the myelin fractions, presents a strong case of brain iron insufficiency in restless legs syndrome. A study also noted a continuum of substantia nigra echogenicity between patients with Parkinson disease (with and without restless legs) and those with SCA-3 (with and without restless legs) as well as controls, and restless legs syndrome severity inversely correlated with substantia nigra echogenicity, further supporting the role of brain iron metabolism in the pathogenesis of restless legs syndrome (246).

Phase analysis MRI has shown higher phase values in patients with restless legs syndrome at the level of the substantia nigra, thalamus, putamen, and pallidum, indicating reduced iron content in several regions of the brain of the patients and supporting the hypothesis of reduced brain iron content in patients with restless legs syndrome (259).

Patients with restless legs syndrome showed reduced thalamic connectivity with the right parahippocampal gyrus, right precuneus, right precentral gyrus, and bilateral lingual gyri, and restless legs syndrome severity was negatively correlated with connectivity between the thalamus and right parahippocampal gyrus, suggesting that patients with restless legs syndrome may have deficits in controlling and managing sensory information (161). Additionally, patients with restless legs syndrome showed alterations in salience network functional connectivity in the right pyramis, bilateral orbitofrontal gyri, and right postcentral gyrus, suggesting that sensory processing and inhibitory mechanisms could be disrupted (162).

A review on restless legs syndrome synthesizing the results from PET studies, functional MRI of white matter pathway, and brain iron content in restless legs syndrome hypothesized that the primary change could be the reduction of brain iron content, which leads to dysfunction of mesolimbic and nigrostriatal dopaminergic pathways, and in turn to dysregulation of limbic and sensorimotor networks (258).

A study involving quantification of advanced oxidation protein products and total thiol levels revealed that increased advanced oxidation protein products, malondialdehyde levels, and decreased thiol and nitric oxide levels may be indicative of patients with restless legs syndrome under oxidative stress (30). These factors may cause restless legs syndrome pathogenesis.

Dysregulation, instead of upregulation or downregulation, of central dopaminergic neurotransmission could be the underlying cause of the restless legs syndrome pathogenesis, as revealed by the assessment of dopamine transporter density using [(123) I]2β-carbomethoxy-3β-(4-iodophenyl)tropane single-photon emission computed tomography (SPECT) and [(123) I]iodobenzamide SPECT. A decreased dopaminergic neurotransmission may cause moderate to moderately severe restless legs syndrome in elderly patients (155).

Symptoms of restless legs syndrome improve when treated with dopaminergic medications and worsen with gamma-hydroxybutyrate, which blocks dopamine release (35), and pimozide (a dopamine receptor blocker) (06). In a single patient with familial restless legs syndrome, CSF-free dopamine and homovanillic acid levels were increased (217). Some authors have shown evidence of reduced 18-F-dopa uptake or reduced dopamine receptor binding in the caudate and putamen of patients with restless legs syndrome, suggesting a central dopaminergic deficit (263), but others have failed to confirm this (311). Single-photon emission computed tomography with the striatal dopamine transporter ligands [(123)I] N-(3-iodopropen-2-y1)-2beta-carbomethoxy-3beta-(chloro-phenyl) tropane and [(123)I]-2-beta-carboxymethoxy-3 beta-[4-iodophenyl] tropane showed no difference between subjects with restless legs syndrome and controls, compared to subjects with Parkinson disease, suggesting no presynaptic dopamine deficit in restless legs syndrome (184; 219). However, studies of striatal D2 receptor binding using iodobenzamide SPECT were suggestive of a post-synaptic dopaminergic disorder at this level, either a reduced number of D2 receptors or decreased affinity of the receptors for the ligand (205). The alterations in mesolimbic D2/3 receptor function reflect the pathophysiology of idiopathic restless legs syndrome, and the baseline availability of these receptors may predict the clinical outcome when patients are treated with D2/3 agonists (224). Preferential targeting of the dopamine D3 receptor subtype by agonists shows higher efficacy on restless legs syndrome than preferential targeting of the D2 receptor subtype (194). CSF studies showed large diurnal changes in tetrahydrobiopterin and serotonin and dopamine metabolites in subjects with restless legs syndrome compared to control subjects, suggesting greater than normal fluctuations in dopaminergic circadian rhythms (80). An animal model of restless legs syndrome produced by injecting the neurotoxin 6-hydroxy dopamine into A11 dopaminergic neurons in rats remains under development (235). The thymocyte differentiation antigen-1 (Thy-1) knockout mouse is a potential rodent model of restless legs syndrome based on clinical observations and the role of dopamine in the disease (46). Drosophila fly models with mutations in the homologue of human gene BTBD9 showed disruption in sleep architecture and motor restlessness, thus recapitulating key features of restless legs syndrome (277). They also showed lower brain dopamine levels. The abnormal sleep fragmentation was rescued by D2 receptor agonist pramipexole, thus supporting the evidence for BTBD9 as a risk factor for restless legs syndrome. Genetic studies have suggested that excess high-activity monoamine oxidase alleles contribute to the susceptibility to restless legs syndrome in women; however, this association is not particularly strong (70).

There has been evidence for adenosine pathway involvement and a complex interplay between adenosine, dopamine receptors, and cell adhesion molecules in the neural plasticity in restless legs syndrome and development of augmentation, suggesting possible new targets for treatment (81). Brain iron deficiency in rat models has been shown to downregulate adenosine A1 receptors, leading to hypersensitive striatal glutaminergic terminals and facilitation of striatal dopamine release (89). Striatal H3 receptor levels in mouse models were also found to positively correlate with periodic limb movements of sleep and negatively correlate with hematocrit, suggesting a role for H3 receptor blockers in restless leg syndrome (167).

Response of symptoms to opioids, and interference with this response by the opiate antagonist naloxone, suggests that restless legs syndrome may also relate to dysfunction of the endogenous opiate system (330). A theory that attempts to unite these pharmacologic observations with studies of abnormal iron metabolism in restless legs syndrome suggests that altered iron status affects both the opiate and dopaminergic systems because both the dopamine D2 receptor and the mu-opiate receptor are iron-dependent (287). Triple knockout mice lacking mu, kappa, and delta receptors showed hyperactivity and increased probability of waking during the rest period as well as decreased serum iron concentration and a significant dysfunction in dopamine metabolism, suggesting a role for the endogenous opioid system in iron metabolism (191). A small MRI study suggested reduced putaminal iron in patients with restless legs syndrome (12).

A study of hypocretin-1 (orexin A) levels in CSF obtained in the late evening in 16 patients with restless legs syndrome and eight control subjects suggested increased hypocretin that was especially marked in early-onset cases (17). It has been proposed that elevations in hypocretin might promote insomnia and increased motor activity.

Family history is positive in up to 92% of cases, which means that genetics plays a key role in the pathogenesis of restless legs syndrome. Most studies suggest autosomal dominant transmission with nearly complete penetrance (310). The phenomenon of genetic anticipation may be present, although ascertainment bias cannot be excluded in such studies (310). Family studies have suggested linkage to a number of genetic loci, including chromosome 12q (71), 13q (28), 14q (43), 19p13 (151; 283) and 20p13 (177). Genome-wide association studies have identified positive associations with protein tyrosine phosphatase receptor type delta (PTPRD) at 9p23-24 (274) and loci on chromosomes 9p22-24, 2q, 6p, and 15q (56; 251), loci on 2p14 and 16p12.1 (335) and 16p12.1 in a French Canadian pedigree (176). Restless legs syndrome and sensorimotor axonal polyneuropathy with fasciculations have been reported in a few patients with intermediate repeat lengths in the spinocerebellar ataxia-3 (Machado-Joseph disease) gene (321).

A genome-wide association study identified a significant association of SNP rs1026732 in MAP2K5/SKOR1, in addition to variants in the MEIS1 and BTBD2 genes, and restless legs syndrome (343). The MEIS1 variant influences iron homeostasis (49) as well as altered sympathovagal balance in homozygotes for risk allele (297). MEIS1 variants have been confirmed to be the strongest genetic risk factor for restless legs syndrome (275); others include PTPRD and BTBD9 variants. MEIS1 ortholog in animal models has been linked to iron homeostasis and reduced MEIS1 expression in mouse models resulted in circadian hyperactivity phenotype similar to restless leg syndrome, suggesting a link to the restless leg biological pathway (270).

PCDHA3 has been suggested as a candidate gene for restless legs syndrome based on genome-wide linkage analysis and exome sequencing in a German family with autosomal dominant restless legs syndrome (332).

A study suggested an association between GABRR3rs832032 polymorphism and the risk for restless legs syndrome and a modifier effect of GABRA4 rs2229940 on the age of onset of restless legs syndrome (145).

Mutational load analysis identified 14 genes of significance with restless leg syndrome; nine of them (AAGAB, ATP2C1, CNTN4, COL6A6, CRBN, GLO1, NTNG1, STEAP4, VAV3) are in the vicinity of known restless legs syndrome loci, and 5 (BBS7, CADM1, CREB5, NRG3, SUN1) were not previously associated with restless legs syndrome (299).

A large genome-wide association identified three novel associations with restless leg syndrome: rs112716420-G, rs10068599-T, and rs10769894-A, which are associated with MICALL2 and UNCX, RANBP17 (Ran-binding protein 17), and LMO1 proteins, respectively. MICALL2 and UNCX are associated with hematocrit, hemoglobin, and glomerular filtration rates, but their role in iron homeostasis is unknown. RANBP17 (Ran-binding protein 17) is mainly expressed in the basal ganglia and cerebral cortex, and variants have been associated with visceral fat, body mass index, and high-density lipoprotein cholesterol. LMO1 encodes the protein rhombotin-1, and variants have been associated with neuroblastoma and T-cell leukemia because MEIS1 is also associated with leukemias and neuroblastomas (73). No animal model replicates all of the clinical features of resltess legs syndrome, but BTBD9 knockout mice showed motor restlessness, and striatal MEIS knockout mice demonstrated hyperactivity with a circadian trend (153).

Upregulation of the hypoxia pathway is demonstrated in multiple cell types in restless legs syndrome individuals. Increased nitric oxide synthase and nitrotyrosine suggest that nitric oxide is involved in the hypoxia activation (243).

Reduced expression of mitochondrial iron genes as well as iron regulatory protein has been noted in restless leg patients, with improvement in mitochondrial respiratory capacity in patients on dopaminergic medications (118).

Epidemiology

No interventions have been shown to prevent the occurrence of restless legs syndrome.

Differential diagnosis

Akathisia is defined as an inner sense of restlessness relieved by voluntary leg movement. Although both restless legs syndrome and akathisia are typified by motor restlessness and sleep disorder, akathisia sufferers describe an inner sense of restlessness rather than the leg discomfort seen in restless legs syndrome. Also absent in akathisia is the striking predominance of evening and bedtime symptoms characteristic of restless legs syndrome. Neuroleptic use or exposure is a common feature in the history of those with akathisia. Polysomnography suggests that although both syndromes decrease sleep efficiency, the abnormalities are milder with akathisia (327).

Hypnic myoclonus, or sleep starts, occurs during the transition from wakefulness to sleep. Brief, massive body movements are synchronous on both sides and lack the periodicity of periodic leg movements of sleep. They are associated with sudden arousal and may be accompanied by a sensation of sinking or falling. No localized discomfort exists in the legs, nor is there inner restlessness (216).

"Vesper's curse" comprises lumbosacral and leg pain with calf cramps and fasciculations that arouse the patient from sleep. The pain and paresthesias are often accompanied by an inner urge to move the legs, similar to that seen in restless legs syndrome. The etiology of "Vesper's curse" is lumbar spinal stenosis with congestive heart failure. Increased right atrial filling pressure, secondary to the failure, is reflected in increased paraspinal venous volumes within the reduced confines of a stenotic lumbar spine. The symptoms may be relieved by assuming an erect or semi-reclining sleep position and are improved with adequate treatment of congestive heart failure (166).

Painful legs and moving toes syndrome may develop in the context of chronic or postsurgical back pain. Severe pain in one or both feet is associated with involuntary irregular wriggling and writhing movements of the toes, foot, or proximal muscles.

The patient cannot imitate the movements, which do not show preference for evening hours (284).

Diagnostic workup

Physical and neurologic examinations are usually normal in restless legs syndrome. The diagnosis rests on clinical criteria. The International Restless Legs Study Group (IRLS) rating scale has been shown to correlate significantly with polysomnographic measures of periodic leg movements during sleep and sleep efficiency as well as with results of the suggested immobilization test (07). Stiasny-Kolster has reported that a reduction in the severity of restless legs symptoms in the 2 hours following a single dose of carbidopa/levodopa 25/100 can diagnose the disorder with 88% sensitivity and 100% specificity (290).

No laboratory study reliably identifies restless legs syndrome. However, because of the association with systemic illness, especially with iron deficiency, it is wise to do screening serum chemistries and hematologic tests (including iron and ferritin) for systemic causes of the syndrome. Peripheral neuropathy may be ruled out using EMG and nerve conduction velocity studies. When clinically indicated, additional cardiac or pulmonary studies may be helpful to assess for the presence of underlying systemic illness.

Although all-night polysomnography may help quantify leg movements of wakefulness and confirm the presence of periodic leg movements of sleep, its utility in the diagnosis is debatable. One study suggested that two polysomnographic measures, the index of periodic leg movements during nocturnal wakefulness and the mean subjective leg discomfort score during the suggested immobilization test, correctly classified 88% of subjects, with a diagnostic sensitivity of 82% and specificity of 100% (204).

Management

There are many therapeutic strategies to achieve optimal outcomes in patients with restless legs syndrome (126). Several ways exist to assess treatment outcomes in restless legs syndrome. Many studies have relied on subjective reports of the presence and severity of paresthesia, leg movements, and sleep disruption. Polysomnographic recordings allow quantitation of leg movements when awake and demonstrate the number and amplitude of periodic leg movements of sleep (218).

In the forced immobilization test, the subject sits with the legs restrained and signals awareness of abnormal leg sensations when concurrent EMG recordings are made from the tibialis anterior muscles (218). The need for objective measures and placebo-controlled trials in this condition is underscored by a robust response to placebo in many trials (42). The magnitude of placebo response in restless legs syndrome is over the threshold for minimal clinically important difference, making it hard to interpret the response (282).

The International Restless Legs Scale correlates with objective signs of restless legs syndrome as measured by the frequency of and arousals associated with periodic limb movements of sleep and shows good psychometric properties for studying restless legs syndrome (342). The Restless Legs Syndrome Quality of Life questionnaire is a reliable, valid, and responsive tool in clinic-based populations and large-scale clinical trials (01).

The first aim of medical management of restless legs syndrome is to appropriately correct iron deficiency. Iron supplementation in patients with restless legs syndrome and low iron and ferritin levels improves motor and sensory symptoms and also may improve sleep, daytime sleepiness, depression, fatigue, and quality of life (65). Iron treatment could be useful in pediatric restless legs syndrome, as demonstrated in Japanese children (209). Studies have noted poor iron status in regular blood donors, and one study suggests iron supplementation with intravenous iron sucrose as more efficient than oral iron sulfate; this was shown as especially helpful in women to prevent the development of restless legs syndrome (40). Evaluating a new iron formulation, ferric carboxymaltose, with molecular properties that may make iron more available to the brain than iron sucrose, significantly improved primary and secondary restless legs syndrome symptoms (09). A single dose of intravenous ferric carboxymaltose led to significant improvement in restless legs symptoms in patients with iron deficiency or low normal ferritin, with responses noted within 1 week (130), though a randomized study suggested more delayed response by 12 weeks (312).

In one study, vitamin D supplementation was associated with improved restless legs syndrome severity in patients with vitamin D levels lower than 50 nmol/L; median levels in the study were around 21.7 nmol/l (325). Another study supplemented vitamin D-deficient patients with 50,000 units per week for 2 months and showed a positive impact on symptom severity, disease measures, and sleep (317).

The focus of management should then turn to symptom control. Symptomatic management of restless legs syndrome includes four major pharmacologic categories: (1) benzodiazepines, (2) opioids, (3) dopaminergics, and (4) some antiepileptics (112). Other medications used include clonidine, baclofen, carbamazepine, quinine, vitamin B12 and vitamin E, and folic acid (218). Vitamins C and E and their combination were shown to be effective in reducing restless legs syndrome severity in hemodialysis patients (265). According to an evidence-based review, the following drugs were considered efficacious for the treatment of restless legs syndrome: levodopa, ropinirole, pramipexole, cabergoline, pergolide, gabapentin, and a prodrug, gabapentin enacarbil; drugs considered likely efficacious are rotigotine, bromocriptine, oxycodone, carbamazepine, valproic acid, clonidine (308), and pregabalin (45).

Benzodiazepines. Benzodiazepines have been used in uncontrolled and a few controlled studies. Clonazepam is the most studied member of this class. A single nighttime dose of 0.5 to 2.0 mg has been shown to improve subjective measures of sleep efficiency as well as unpleasant leg sensations in short- and long-term uncontrolled trials (272; 218). In a controlled trial in six patients, clonazepam was found to be superior to placebo in improving sleep quality and leg discomfort (213). In contrast, Boghen found that clonazepam was not superior to placebo in improving subjective measures in six patients with restless legs syndrome (42). Studies of the effect of clonazepam on the movements of periodic leg movements of sleep are contradictory. In a double-blind, placebo-controlled parallel study of clonazepam in 20 patients with periodic leg movements of sleep, clonazepam decreased the number of leg movements and improved sleep parameters (247). Others have found that benzodiazepines improve the sleep of patients with periodic leg movements of sleep, but without significantly altering the number of leg jerks (268). Long-term (more than 6 months) therapy with clonazepam remains efficacious and is not associated with a high risk of dose tolerance, adverse effects, or abuse (273). Diazepam, alprazolam, nitrazepam, temazepam, and triazolam have also been used to treat restless legs syndrome and periodic leg movements of sleep. Short-acting benzodiazepines may be less effective than clonazepam in suppressing leg movement in these conditions, but they do effectively decrease nighttime arousals (75). A Cochrane review concluded that the effects of benzodiazepines on restless legs syndrome are unknown at this point (47). Effects of benzodiazepine therapy include excessive sedation, decreased libido, and nocturnal confusion or wandering.

Opioids. The history of opioid treatment of restless legs syndrome dates to Ekbom's 1945 manuscript. The effectiveness of opioids may relate to decreasing arousals without altering the frequency of leg movements (149). A retrospective review of 113 opioid-treated patients with restless legs syndrome suggested good long-term efficacy using clinical and polysomnographic measures (331). Concerns about the addictive potential of these treatments have led many authors to recommend them as treatments of last resort (216). Tramadol, a centrally acting analgesic that is better tolerated and has a lower addictive potential than classical opioids, has been shown effective in a small open-label trial (171). Methadone (15.5 ± 7.7 mg/day) reduced symptoms in an open trial in patients with refractory restless legs syndrome (232). An acute challenge with intravenous apomorphine (a combined opioidergic and dopaminergic agonist) was effective in a small trial (313). Two patients whose restless legs syndrome symptoms proved refractory to other medical interventions were safely and effectively treated with nocturnal subcutaneous apomorphine infusion (257). Refractory symptoms have been reported to improve with intrathecal morphine (260).

Prolonged release oxycodone/naloxone improved restless legs syndrome symptom severity and sleep quantity and adequacy, resulting in greater sleep satisfaction, less daytime tiredness, and improved quality of life, suggesting an alternative treatment option for severe restless legs syndrome that cannot be controlled by first-line dopaminergic medications (228).

Dopaminergics. Levodopa administered with a peripheral dopa decarboxylase inhibitor (carbidopa or benserazide) has been shown to be effective in treating restless legs syndrome and periodic leg movements of sleep in many studies (62). Although there is considerable evidence of short-term efficacy, long-term course has not been adequately studied. The drug’s short half-life is the main limitation of levodopa therapy for restless legs syndrome. Symptoms and polysomnographic abnormalities generally recur 4 to 6 hours following a bedtime dose of standard levodopa, which is consistent with the drug’s half-life. This phenomenon, called “augmentation,” has been approached by administration of a second levodopa dose 3 hours after bedtime or by combining controlled release levodopa with the standard preparation. The combination of immediate release and sustained release levodopa preparations leads to a significant improvement in measures of restless legs syndrome. Improvements in sleep efficiency and subjective sleep quality also improve, though the onset of improvement in these parameters may be somewhat delayed (267). However, long-term studies suggest that only 40% of patients continue to have good efficacy for 1 year or more (307). An evidence-based review found good evidence to support the use of short-term treatment with levodopa (particularly for treating periodic leg movements of sleep) but insufficient evidence of long-term safety and efficacy (62). In patients who have developed augmentation with more than one dopaminergic drug, treatment should be switched to nondopaminergics and opiates (98).

Direct-acting dopamine agonists have the advantage of longer half-lives than levodopa and have emerged as the preferred treatment for the disorder. Symptoms of restless legs syndrome can improve after the first dose of agonist. There have been two large randomized, double-blind, placebo-controlled clinical trials of ropinirole in restless legs syndrome. Both studies showed that ropinirole was safe, well-tolerated, and improved symptoms of restless legs syndrome, sleep quality, quality of life, and anxiety (41). A 6-week, randomized, double-blind study of pramipexole (median dose 0.35 mg/day) in subjects with severe restless legs syndrome showed significant improvement in the pramipexole arm. Overall, 63% of pramipexole-treated versus 32.5% of placebo-treated patients improved (229). In a long-term open-label study, 81% of pramipexole-treated subjects had a 50% or more reduction in International RLS Study Group Scale score at 26 weeks, and 90% rated themselves as much or very much improved. Headache and fatigue were the most common adverse effects (242). Long-term benefit and good long-term tolerability have been reported with pramipexole (215). A meta-analysis of published clinical trials of ropinirole and pramipexole found both treatments superior to placebo for treating restless legs syndrome. Pramipexole was felt to show superior efficacy and tolerability compared to ropinirole in this study (255). Both pramipexole and ropinirole have been FDA-approved for treating restless legs syndrome. In a Japanese study, primary restless legs syndrome sufferers of older age (55 years or older) and low IRLS score at baseline (less than 21.5 points) were significantly associated with early response to low-dose pramipexole therapy (139; 140). Pramipexole has also been reported to improve restless legs syndrome-related mood disturbances (212). Similar benefits have been demonstrated with 1 to 4 mg of cabergoline daily (226). A randomized, blinded clinical trial comparing cabergoline (2 to 3 mg daily) and levodopa and benserazide (200 to 300 mg levodopa) for 30 weeks in 361 research subjects with restless legs syndrome showed cabergoline was superior to, but somewhat less well tolerated than, levodopa (302). Randomized, double-blind, placebo-controlled studies of rotigotine transdermal patch in moderate to severe idiopathic restless legs syndrome showed good efficacy and tolerability for doses up to 3 mg/24 hours (229; 227; 230; 303; 125). A 2-year safety follow-up trial using a rotigotine patch in doses between 0.5 to 4 mg/24 hours demonstrated that it is efficacious and well-tolerated in moderate to severe restless legs syndrome (128). The rate of typical dopaminergic side effects, nausea, and fatigue was low; application site reactions declined over time (53). The safety profile of rotigotine is comparable to other non-ergolinic dopamine agonists. Nevertheless, the transdermal delivery may result in local, mild skin reactions leading to treatment discontinuation in a minority of patients (285). A rotigotine patch is now approved for use in moderate to severe restless legs syndrome, with the highest recommended dose being 3 mg per 24 hours. Extended-release pramipexole has been shown to treat augmentation induced by treatment with l-dopa, rotigotine, ropinirole, and immediate-release pramipexole, with treatment results sustained over mean follow-up periods of 13 months or more (193). Rotigotine has also been shown to reduce augmentation in severe restless legs syndrome on oral dopaminergic therapy (306). Additionally, rotigotine has been shown to reduce periodic limb movement-associated nocturnal systolic blood pressure elevations in moderate to severe restless legs syndrome (33).

Other dopamine agonists suggested to be useful in open-label or small controlled trials include piribedil (85), alpha-dihydroergocryptine (296), and oral and transdermal lisuride (36). Because they are well-tolerated, highly effective, and associated with little, if any, symptom rebound, dopamine agonists have largely replaced levodopa in the treatment of restless legs syndrome (59).

Treatment with dopaminergic agents usually promotes wakefulness in patients with restless legs syndrome with daytime sleepiness but may infrequently cause daytime sleepiness (148). Tolerance and augmentation may occur with long-term use of dopamine agonists, but these problems tend to improve with dose adjustment (336). At baseline, patients with restless legs syndrome were found to have similar incidences of impulsivity, impulse control disorders, and addictive behavior as age- and sex-matched controls (34). Some agonist-treated patients with restless legs syndrome will develop impulse control disorders, such as pathological gambling, shopping, or hypersexuality. Patients with restless legs syndrome with mood and stress states may be at greater risk of impulse control disorders with dopamine agonists (254). Patients with restless legs syndrome and augmentation may have up to 6-fold increased risk for exhibiting impulse control disorder symptoms, suggesting a need to screen such patients for impulse control disorder symptoms (120).

Typically, stopping the agonists or lowering the dose results in improvement but, first, may lead to severe personal and financial distress (159). A survey of 261 patients with idiopathic restless legs syndrome revealed increased gambling behavior in about 7% and increased sexual desire in 5% (78). The mean duration of treatment before onset of impulse control disorders was 9.5 months (63).

Dopamine agonists frequently cause gastrointestinal side effects and are associated with excessive daytime sleepiness and peripheral edema. These agents have also infrequently been associated with serious complications such as pleuropulmonary or retroperitoneal fibrosis. Most such cases have been associated with longstanding treatment with ergot-derived agonists (pergolide, bromocriptine, cabergoline) in Parkinson disease, but there have been anecdotal reports in restless legs syndrome (341; 66). These effects are generally believed to be related to their ergot properties. There is accumulating evidence that fibrotic valvulopathy, such as is seen in carcinoid syndrome, may occur in association with pergolide or bromocriptine therapy in Parkinson disease (131). Pergolide has been withdrawn from the market due to the risk of cardiac valvulopathy. Although the risks of dopamine agonists in restless legs syndrome treatment need further study, the treating physician should exercise caution in their use. It would seem prudent to avoid ergot-derived dopamine agonists and to clinically monitor patients treated with non-ergoline agents for toxicity.

Gabapentin. A double-blind placebo-controlled crossover study of gabapentin in 24 patients with restless legs syndrome (22 idiopathic, two related to iron deficiency) showed improvements in clinical rating scales, reduced periodic leg movements during sleep and improved sleep architecture during gabapentin treatment. The mean effective dose was just under 2000 mg daily. The duration of the study was short, with just 6 weeks of treatment in each arm of the study. A double-blind placebo-controlled crossover study in 16 hemodialysis patients with restless legs syndrome found gabapentin superior to placebo (298). A brief open trial in which treatment to ropinirole (0.25 to 1.5 mg) or gabapentin (300 to 1200 mg) was randomized showed the two agents were similarly effective and well-tolerated (116). Gabapentin has been shown to be safe and efficacious in low doses of 100 mg for uremic restless legs syndrome (52). Because gabapentin, in contrast to dopaminergic drugs, has a low risk of augmentation and gabapentin enacarbil is often not covered by third-party payers, generic gabapentin is now considered the drug of choice in restless legs syndrome (333).

Pregabalin. Pregabalin was found to be more effective than placebo for moderate to severe restless legs syndrome, with mild adverse effects of daytime somnolence, headache, and unsteadiness (45). A dose of 123.9 mg/day provided 90% efficacy at 6 weeks (15). It also improves sleep architecture and periodic limb movements in placebo-unresponsive patients (97).

Pregabalin provided significantly improved treatment outcomes compared to placebo, and augmentation rates were significantly lower with pregabalin than with 0.5 mg of pramipexole (14).

Gabapentin enacarbil. Gabapentin enacarbil is a GABA analogue, a precursor to gabapentin, and is better absorbed with sustained dose-dependent bioavailability. It has been approved by the U.S. Food and Drug Administration for the management of restless legs syndrome in adults. In clinical trials, it has demonstrated superiority over placebo for restless legs syndrome and is well tolerated and safe (165; 04; 136; 119). Doses from 1200 to 1800 mg/day appeared effective in treating restless legs syndrome with mild side effects of dizziness and somnolence (202). It has been shown that gabapentin enacarbil at 600 mg/day or lower doses may treat subjective restless legs syndrome symptoms, and 1200 mg/day or higher doses may help both subjective restless legs syndrome symptoms and associated problems, such as severe sleep disturbances (163).

Long-term treatment, up to 52 weeks, with gabapentin enacarbil improved restless legs syndrome symptoms in Japanese patients with an acceptable safety profile (141).

Interestingly, at least one study has demonstrated that prior dopaminergic use reduced the effectiveness of gabapentin enacarbil, suggesting that nondopaminergic options may be better as first-line treatment for restless legs syndrome (94).

Clonidine. In three small uncontrolled studies and two small controlled studies, clonidine was found effective in treating restless legs syndrome. Overall, 37 of 51 patients included in these studies benefited from clonidine treatment (114; 31; 289; 324). However, another study failed to confirm the usefulness of clonidine in this condition (29).

Baclofen. Baclofen has been shown in a single study to reduce sleep disturbance in patients with restless legs syndrome. Although the force and amplitude of periodic leg movements were reduced, their number was increased. The reported dose range was 20 to 40 mg at bedtime (109).

Other drugs. Amantadine (100 to 300 mg daily), a drug with putative dopaminergic and antiglutamatergic effects, improved symptoms of restless legs syndrome in 21 patients in an open-label trial (86). Two double-blind placebo trials using carbamazepine 200 to 1000 mg daily have shown improvements in subjective measures of restless legs symptoms. The effects of carbamazepine have not been studied using polysomnography. A few case reports outlined the benefit of oxcarbazepine, the keto derivative of carbamazepine (146). A double-blind placebo-controlled trial of iron in 27 patients with restless legs syndrome showed no benefit (68).

Preliminary results suggest significant effects on sensory and motor symptoms in restless legs syndrome with the AMPA antagonist perampanel, suggesting an alternative to dopaminergic agents (95).

A small study testing the adenosine hypothesis in restless legs syndrome has shown that nonselective equilibrative nucleoside transporter (ENT) 1/2 Inhibitor, dipyridamole, has significant therapeutic effects on both sensory and motor symptoms, as well as sleep with minimal side effects (96).

Two independent pilot trials of botulinum toxin A in restless legs syndrome showed no benefit (221; 101). However, in another pilot trial treatment with onabotulinum toxin A (Botox®) demonstrated a statistically significant improvement in restless legs syndrome during the first 4 weeks after the treatment (05). There has been some debate about intramuscular versus intradermal botulinum toxin administration, with the latter possibly enhancing the nociceptive effects of botulinum toxin on the sensory symptoms in restless legs syndrome (144). A trial of incobotulinumtoxin A injected into the tibialis anterior, gastrocnemius, and biceps femoris muscles showed a reduction in the severity of restless legs syndrome symptoms, pain score, and quality of life without any adverse effects (206).

An isolated case report mentions the use of intravenous physostigmine for perioperative restless leg exacerbation with good response (244).

Some patients with restless legs syndrome rapidly develop tolerance to medications. Current practice is to prescribe medication cycles, changing between medication classes every several weeks.

Nonpharmacologic treatments. Nonpharmacologic treatments may be useful in patients who rapidly become tolerant to medications. Transcutaneous electric nerve stimulation may be useful in some patients.

Some suggest that high-frequency repetitive transcranial magnetic stimulation delivered to the leg area of the primary motor cortex may raise functional activity in the sensorimotor regions, leading to improved restless legs syndrome symptoms (185). Low-frequency, repetitive transcranial magnetic stimulation to the primary somatosensory and motor areas has also been shown to alleviate sensory and motor symptoms in restless legs syndrome (169).

Parathyroidectomy can relieve some symptoms associated with restless legs syndrome in hemodialysis patients with severe secondary hyperparathyroidism and elevated serum phosphate (269).

A randomized trial in 28 subjects with restless legs syndrome showed a significant effect of aerobic and lower-body resistance training over 12 weeks (26). Intradialytic aerobic exercise has shown improvement in restless legs syndrome symptoms in hemodialysis patients (266).

An exploratory trial looking at 12 weeks of yoga in patients with moderate to severe restless legs syndrome showed improvements in the severity of restless legs syndrome, perceived stress, mood, and sleep quality (138).

A proof of concept study suggested a specific cognitive behavioral therapy (the RELEGS coping therapy program) may improve quality of life and mental health status in patients with restless legs syndrome, but this requires further study (129).

Acupuncture point injection markedly improved sensory symptoms and motor signs in two patients with restless legs syndrome (91). A randomized trial also showed acupuncture with gabapentin had an additive therapeutic effect over gabapentin alone (256).

A systemic review of 11 randomized controlled trials of nonpharmacological interventions for restless leg syndrome showed that repetitive transcranial magnetic stimulation, exercise, compression devices, counterstrain manipulation, infrared therapy, and standard acupuncture effectively controlled symptoms. Although vibration pads, cryotherapy, yoga, compression devices, and acupuncture significantly improved sleep-related outcomes, the quality of the evidence was not high and posed the question about the placebo effect in these studies (117).

Experimental therapies. In a group of six patients with Parkinson disease who underwent bilateral subthalamic nucleus deep brain stimulation for Parkinson disease, restless legs syndrome had excellent improvement, with IRLS scores improving by a mean of 84% even though their dopaminergic medications were lowered by 56% postoperatively (77). Subthalamic nucleus deep brain stimulation decreased restless legs syndrome symptoms in patients with Parkinson disease despite a decrease in dopaminergic therapy, and this was sustained for at least 2 years after implant (157).

Another group implanted bilateral ventralis intermedius nucleus deep brain stimulation stimulators in nine patients with essential tremor with coexistent restless legs syndrome; the patients reported no benefit in their restless legs syndrome symptoms (234). A case report suggested the potential role for deep brain stimulation to the posteroventral lateral globus pallidus in a patient with medication refractory restless legs syndrome with sustained benefits at 2 years postprocedure (238). The globus pallidus was chosen for stimulator implantation due to the hyperkinetic restless leg phenotype seen in the patient, which resembled dyskinesias seen in Parkinson disease.

A small case series looking into spinal cord stimulation for chronic low back or leg pain showed benefits with concomitant restless legs syndrome (02).

Special considerations

Pregnancy

Reported rates of restless legs symptoms in pregnancy range between 17% and 30% (271). In a series of 500 pregnant women, 97 had restless legs syndrome. In 16 patients, symptoms were present before pregnancy; in one-third, symptoms were exacerbated during the third trimester of pregnancy but returned to baseline following delivery. Patients without symptoms before pregnancy were nearly all normal by 10 days postpartum. In a French population study, one third of pregnant women in their third trimester were reported to have restless legs syndrome, which usually improved after delivery. There is no allowed treatment; typically, only counseling and iron status assessment should be provided (222). The teratogenic risks of dopamine agonists are unknown, so they are currently not recommended in pregnancy. In refractory cases after correction of iron status, some medications may be considered, such as opioids, antiepileptics (eg, carbamazepine and gabapentin), and some benzodiazepines, which have a more extensive safety record in pregnancy (74).

Anesthesia

Spinal anesthesia may be poorly tolerated in restless legs syndrome if dysesthetic sensations persist and leg movement is not possible. Shin reported excessive agitation due to an inability to move restless legs following epidural anesthesia in a woman about to undergo a Cesarean section (278). No other known anesthetic considerations exist.