UK
Europe
USA&Canada
Aust.&NZ
Asia
Editorials Online™
Medical
Law News
E-texts
Links
Classifieds
Home
|
|
|
NEJM
Christopher J. Earley, M.B., B.Ch., Ph.D.
This Journal feature begins with a case vignette highlighting a common clinical problem. Evidence supporting various strategies is then presented, followed by a review of formal guidelines, when they exist. The article ends with the author's clinical recommendations. A 45-year-old woman reports having had nightly insomnia for years. On further questioning, she reports having uncomfortable sensations in her legs when she lies down at night. She describes a feeling of needing to move her legs, which is relieved only by getting up and walking around. How should this patient be evaluated and treated? The Clinical Problem The Diagnosis The restless legs syndrome (RLS), also known as Ekbom's syndrome,1 is a neurologic disorder with a prevalence in the general population of between 2.5 and 15 percent.2 The prevalence of RLS increases with age and appears to be higher among women than among men.2 A family history of RLS is particularly common among patients whose symptoms appear before 40 years of age.3 Patients with an onset of symptoms after 50 years of age have less familial occurrence of RLS and are more likely to have symptoms or signs of neuropathy.3,4 Studies of large kindreds suggest an autosomal dominant mode of inheritance,2 and a major susceptibility locus for RLS has been identified on chromosome 12q.5 The core feature of this syndrome is a distressing, irresistible need or urge to move the legs (akathisia). The akathisia often coexists with an uncomfortable, though not usually painful, sensation deep within the legs.1,6 The sensation may be described as a muscle ache or tension; other patients describe a "creepy-crawly" sensation or a feeling "like ants marching in my legs" or "like soda water in the veins." The symptoms may extend beyond the legs to involve the arms or trunk. The symptoms are brought on with rest (sitting or lying down). The more comfortable the patient becomes, the more likely it is that the symptoms will occur. The reverse is also true — the less comfortable the patient is, the less likely it is that he or she will have symptoms; some patients may therefore find it easier to sleep on a hard floor than in a comfortable bed. Movement immediately relieves the symptoms, and continued movement (such as walking) provides ongoing relief. However, if patients stop moving their legs, the symptoms may return. Ignoring the urge to move the legs may lead to progressive intensification of the akathisia, until patients either move their legs or the legs jerk involuntarily. Although initially the symptoms occur at bedtime or during the night, as the syndrome progresses, symptoms start to occur earlier in the day and become more intense at night. Even when they occur throughout the day, the symptoms are always worst in the evening or at night. RLS is usually associated with semirhythmic leg movements during sleep that are referred to as periodic limb movements of sleep.7 Sometimes the patient's bed partner reports having seen frequent leg or limb movements during sleep years before the patient recognized sensory symptoms. Not all patients with RLS have periodic limb movements of sleep, and not all patients with such movements have RLS, making the identification of these movements of questionable value in diagnosis.6 Whether because of the sensory symptoms or associated periodic limb movements of sleep, RLS may have profound negative effects on sleep. Symptoms of insomnia or fatigue may be the problem that is initially reported; in addition, reduced concentration and memory, decreased motivation and drive, and depression and anxiety may be reported. Since rest often brings on manifestations of RLS, activities that require prolonged sitting, such as traveling long distances or going to movies, are often avoided. RLS has been reported to occur in 20 percent of women during pregnancy8,9 and in 20 to 62 percent of patients undergoing dialysis.10,11 There is an association between neuropathy and RLS, and small-fiber neuropathy may be a contributing factor in patients with late-onset RLS.2,4 Despite this association, the reported 5.2 percent prevalence of RLS in a clinical population of patients with polyneuropathy does not suggest a prominent causal relation between neuropathy and RLS.12 Finally, iron deficiency has been reported to occur at a high frequency among patients with RLS,1,13 and brain iron deficiency may be a critical factor in the pathology of RLS.14 Differential Diagnosis Leg cramps are a common sleep-related problem. Although some patients with RLS describe the symptoms as "cramps," in most cases cramps are easily distinguished from RLS by the presence of a clear knotting of a muscle and the presence of severe local muscle pain in the former. Paresthesias may occur in a foot or leg after the patient has been sitting for a long time and may be relieved by a change of position or by crossing and uncrossing of the legs. A less common condition called hypotensive akathisia has been reported in patients with autonomic failure. These patients have vague symptoms of discomfort in their legs when they sit for long periods, which are relieved by crossing and uncrossing of their legs.15 Both of these conditions manifest themselves only when the patient is sitting (not when he or she is lying down), do not disrupt sleep, and can occur at any time of day. Although patients with these conditions frequently change position to make themselves more comfortable, there is no associated "irresistible" urge to move the legs, as is characteristic of RLS. Periodic leg movements that are apparent on polysomnography, although associated with RLS, are also associated with other conditions, such as sleep apnea, neurodegenerative diseases, spinal cord lesions, stroke, and narcolepsy, as well as with treatment with antidepressant or neuroleptic agents. In the absence of the core clinical features of RLS (Table 1), the diagnosis of RLS cannot be made, and other causes of periodic limb movements of sleep or isolated periodic limb movement disorder16 must be considered. The relation between periodic limb movement disorder and RLS is unclear, but treatments used for RLS may be effective in this disorder as well.17
RLS may coexist with conditions that cause leg pain, such as neuropathy or arthritis, and these conditions may aggravate each other. Thus, patients with pain from another known cause who report having insomnia should be asked specifically about characteristic symptoms of RLS, as described above. Strategies and Evidence Evaluation The diagnosis of RLS is based on the clinical history. Polysomnography should be performed if there is clinical suspicion of sleep apnea or if, after treatment of the symptoms of RLS, sleep is still disrupted. Since RLS may be a symptom of iron deficiency1,18,19 and since iron deficiency is frequently present in the absence of an anemia,20 the iron status (the serum ferritin level and iron saturation21) should be assessed. No other laboratory tests are routinely indicated. Treatment Drug therapy for RLS varies according to the frequency of symptoms and the presence or absence of associated pain (Table 2).
Dopaminergic Agents The dopaminergic agents are considered the first line of treatment in idiopathic RLS.17,22 Levodopa, bromocriptine, pergolide, and pramipexole have been studied in randomized, placebo-controlled trials and have been shown to provide clinically and statistically significant improvement in symptoms of RLS, periodic limb movements of sleep, or both.17,23,24 These agents have provided 90 to 100 percent relief of symptoms in RLS and have reduced the frequency of periodic limb movements of sleep by 70 to 100 percent. Common side effects of these agents include insomnia, nasal congestion, swelling of the hands or feet, bloating, chest pain, and nausea or vomiting.17,23,24,25 Hypersomnia has been reported with long-term use of dopamine agonists in patients with Parkinson's disease,26 although this has not appeared to be a problem among patients with RLS. A very common problem with the long-term use of dopaminergic drugs is augmentation of the symptoms of RLS, which takes the form of an onset of symptoms progressively earlier in the day.25,27 Increases in the dose of medication to cover the expanded symptomatic period relieve the symptoms in the short term, but ultimately, symptoms begin to appear even earlier, accompanied by a noticeable decrease in the duration of action and the effectiveness of the drug. Some patients have an augmentation in the form of an extension of the symptoms from the legs to the arms or trunk. Patients may present with total-body restlessness that is unrelenting even when they are walking. The symptoms are indistinguishable from those of acute akathisia induced by neuroleptic drugs. Augmentation has been reported less frequently with dopamine agonists (in 20 to 30 percent of patients)28,29,30 than with levodopa (in 80 percent of patients).17,25 However, this difference may reflect the pattern of use of dopamine agonists in clinical studies, which have involved relatively low doses and short durations of treatment. It is possible that longer-term use and higher doses of dopamine agonists may result in rates of augmentation similar to those reported with levodopa. All patients will have some withdrawal symptoms when treatment with any of the dopaminergic agents is discontinued. Withdrawal manifests itself as an intensification of the symptoms of RLS that is often quite severe for the first 48 hours; the symptoms then slowly return to their base-line level after four to seven days. In general, the longer the drug has been used and the higher the dose, the worse the withdrawal syndrome. Other Agents Opiates (propoxyphene, oxycodone, and methadone), benzodiazepines (clonazepam), and anticonvulsant agents (carbamazepine and gabapentin) have been shown, mostly in open-label studies, to improve subjective ratings of the symptoms of RLS. They have also reduced the occurrence of periodic limb movements of sleep on polysomnography.17,22 Side effects of these agents in patients with RLS, including the risk of tolerance or dependence, are no different from those reported for the general population (Table 3). Early-morning sedation is an issue of particular concern with the bedtime use of clonazepam, which has a long half-life. None of these agents have been linked to an augmentation of symptoms.
Iron Supplementation A high prevalence of iron deficiency has been found among patients with RLS, and treatment of the deficiency has been reported to improve or resolve symptoms.1,18,19 A low serum ferritin level (less than 50 µg per liter) has been associated with greater severity of RLS and with a reduction in the quantity of sleep as determined by polysomnography.13,31 In patients with a serum ferritin level of less than 18 µg per liter, treatment with oral iron supplements resulted in improvements in the severity of the symptoms of RLS and, in some patients, the complete resolution of the symptoms.13 The use of oral iron therapy in subjects with normal-to-high ferritin levels, in contrast, was ineffective but similarly had little effect on the serum ferritin level.32 In one report, 21 of 22 subjects who had normal serum iron levels had a resolution of symptoms of RLS after high-dose intravenous iron supplementation.33 The findings support the treatment of iron deficiency as a potential therapy for symptoms of RLS. Patients Undergoing Hemodialysis Although patients undergoing hemodialysis have a high prevalence of RLS, there are only a few reported studies of therapy in this population. Levodopa (100 to 250 mg) and pergolide (0.25 mg) were less effective against RLS among patients undergoing hemodialysis than among patients with idiopathic RLS.34,35,36,37 Several case series with small numbers of subjects have reported improvement in symptoms of RLS with the use of clonidine, clonazepam, or epoetin alfa, or after renal transplantation.17 Pregnancy Treatment of iron deficiency in pregnant women with RLS had been advocated,1 but therapeutic trials are lacking in this population. Some studies have found that lower folate levels are associated with RLS during pregnancy and have led to recommendations for treating affected women with folate,38 although this approach has not been systematically evaluated. Most drugs used in treating idiopathic RLS fall into pregnancy category C or higher risk (agents in category C have had toxic effects in studies of animals, but the results of studies in humans are inadequate); the exception is pergolide, which is a category B agent (a category for which there is no evidence of risk in humans). Areas of Uncertainty Despite the distinctive clinical features of RLS, there remains substantial variability in responses to treatment and in clinical progression and outcome. Better understanding is needed of potential modifiers of the response to treatment, such as family history, age at onset of symptoms, sex, iron status, and the presence or absence of small-fiber neuropathy. There is currently inadequate information on the efficacy of medications other than the dopaminergic drugs and on the management of RLS in patients who are pregnant or undergoing dialysis. There are no published data on the use of combinations of agents for treatment. The role of iron supplementation, especially among patients who do not have frank iron deficiency, also remains uncertain. Guidelines The American Academy of Sleep Medicine has published practice recommendations regarding the treatment of RLS that are based solely on a review of published, peer-reviewed articles.17,22 The report recommends that treatment of RLS be limited to persons meeting specified diagnostic criteria and emphasizes the distinction between RLS and periodic limb movement disorder and the fact that treatments for the two disorders may differ. An associated guideline from this organization (a strategy for patient care that reflects a moderate degree of clinical certainty) recommends the use of levodopa, pergolide, oxycodone, propoxyphene, or carbamazepine. Other agents (clonazepam, gabapentin, clonidine, and iron supplementation) were considered as "options," although their benefits are uncertain. Despite the higher recommendations given for carbamazepine, gabapentin has become the preferred anticonvulsant agent because it is associated with lower risk. (Carbamazepine has a black-box warning regarding the risk of aplastic anemia and agranulocytosis.) On the basis of available data, no medication achieved the highest level of recommendation ("standard of care"). The Restless Legs Syndrome Foundation, a national organization focusing on this disorder (http://www.rls.org), is a reliable source of information about the disorder and about current research and supportive services. Summary and Recommendations The primary symptoms of RLS are a distressing urge to move the legs, associated with an uncomfortable feeling in the legs that is often hard to describe. The sensation is brought on with rest and relieved with movement and occurs predominantly, if not exclusively, at night. Although there have been limited clinical trials whose results support the use of various therapies, several recommendations may be made on the basis of a combination of clinical experience and available research. For patients with nightly symptoms, such as the woman described in the vignette, a dopamine agonist (pergolide, pramipexole, or ropinirole) is the treatment of choice. Reasonable alternatives, if problems arise with a dopamine agonist, are opiates, sedative–hypnotic agents, or gabapentin. Although there are no data from direct comparisons of these alternative choices, the opiates oxycodone and propoxyphene have been more strongly recommended by the American Academy of Sleep Medicine, and in my own practice have appeared more effective, than clonazepam or gabapentin. In practice, tolerance often develops quickly to sedative–hypnotic agents, whereas a majority of patients may continue to receive the same dose of opiates for years. When symptoms occur occasionally (e.g., only on some nights or when the patient is sitting through a meeting), it is reasonable to treat the patient with levodopa (50 to 100 mg). Levodopa should be taken before symptoms are likely to develop. It is fast-acting (takes 15 to 20 minutes to become effective when taken on an empty stomach), and augmentation of symptoms should not be a problem with occasional use. For symptoms that are frequent (occurring three to four times a week, for example) but not nightly, it is reasonable to use medication on an as-needed basis. A fast-acting drug is preferred. The use of levodopa is not contraindicated but carries the risk of developing augmentation. Because the dopamine agonists take at least two hours to reach their peak effect, they are not the optimal therapy in this setting. Opiates and sedative–hypnotic agents can provide rapid treatment benefits. Clonazepam is sedating and, in my experience, no more effective for the management of symptoms than the short-acting or intermediate-acting sedative–hypnotic agents, although these other agents have not been studied specifically for use in RLS. If there is associated pain, then it would be reasonable to treat first with an opiate or gabapentin, although data to support the use of these drugs are lacking. Some combination of the four classes of drugs used in RLS may be required to treat symptoms adequately, although there are no good data regarding combinations of these drugs, and side effects are more likely to occur with combination therapy. If symptoms are more than just a nighttime problem but therapy is limited by side effects, the primary goal should be a reduction in nighttime symptoms to improve the quality of sleep. Testing for iron deficiency (a serum ferritin level of less than 18 µg per liter or an iron-saturation value of less than 16 percent) should be performed routinely, and such a deficiency should be treated. A reasonable approach is to provide 65 mg of elemental iron (e.g., 325 mg of ferrous sulfate) along with 100 mg of vitamin C on an empty stomach one to three times daily and to recheck the ferritin level and iron saturation after three months. It is suggested that oral iron supplementation be continued until ferritin levels are at or above 50 µg per liter and the iron saturation is above 20 percent. Many patients will be able to discontinue dopamine-agonist therapy after the supplementation of deficient iron stores.
From the Department of Neurology, Johns Hopkins School of Medicine, Baltimore. Address reprint requests to Dr. Earley at the Johns Hopkins Bayview Medical Center, 5501 Hopkins Bayview Circle, AAC-1B-82, Baltimore, MD 21224. References
|