Restless legs syndrome and periodic limb movement disorder in children

INTRODUCTION — Restless legs syndrome (RLS), also known as Willis-Ekbom disease, is a common, complex, and treatable neurologic condition. A substantial body of literature about pediatric RLS has emphasized its impact on sleep, as well as frequent comorbidity with attention deficit hyperactivity disorder (ADHD), mood disorders, and anxiety disorders [1-4]. Consensus diagnostic criteria for children were updated in 2013 [5,6], and pediatric-specific treatment data are emerging.

Periodic limb movement disorder (PLMD) is characterized by clinical sleep disturbance and by repetitive limb jerking in sleep (known as periodic limb movements of sleep [PLMS]) that is not better explained by another condition, medication use, or substance use [7]. In children, PLMD appears to be closely related to RLS, although they are distinct diagnostic entities [1,8,9].

Because of the similarities of pediatric RLS and PLMD [9], especially the overlap in the approach to treatment, they are presented together in this topic review, with important differences noted when applicable. The following topic reviews have related content:

●(See "Assessment of sleep disorders in children".)

●(See "Clinical features and diagnosis of restless legs syndrome and periodic limb movement disorder in adults".)

●(See "Management of restless legs syndrome and periodic limb movement disorder in adults".)

●(See "Nocturnal leg cramps".)

EPIDEMIOLOGY — The prevalence of RLS in adults is approximately 5 to 10 percent, based upon population-based studies in Europe and North America [10,11]. However, the prevalence of moderate to severe or clinically significant RLS is lower, estimated at 2 to 3 percent [12-16]. The severity of symptoms is directly relevant to clinical care and is an important component of treatment decisions. The prevalence of RLS in most Asian populations is less, with most studies finding overall rates lower than 5 percent, except in Korea, where prevalence is similar to Europe and North America [10,17-19]. In all ethnic groups, women are affected approximately twice as often as men [10]. Onset of symptoms prior to age 20 years is reported in approximately 40 percent of adults with RLS [20-22].

The prevalence of RLS in children and adolescents is 1 to 4 percent in population-based studies from the United Kingdom, the United States, Turkey, China, Brazil, and Poland, and the prevalence of moderate to severe RLS is 0.5 to 1 percent [23-30]. This approaches the prevalence of obstructive sleep apnea in children but exceeds the prevalence of both pediatric epilepsy and diabetes. Boys are affected as often as girls in younger age groups, except in China, where there are more females affected [28]; the female predominance noted in adults consistently emerges in the mid- to late teens or early 20s [10,24-27].

Population-based studies of RLS in adults report high incidence rates (annual new cases) of 0.8 to 1.7 percent, as well as high remission rates (approximately 50 percent of cases) [31-34]. A single pediatric study found similarly high incidence and remission rates [35]. These findings are in contrast to clinical cohorts, in which remission is much less common [20,36-38]. As an example, a large study in a clinical cohort reported that RLS was progressive in 36 percent, stable in 41 percent, diminished in 15 percent, and remitted in 8 percent [39]. Together, these observations suggest that mild, nonclinical cases that are captured in population-based studies are more likely to remit spontaneously than are those cases that come to medical attention and that are captured in clinical cohorts. Nonetheless, the potential for remission should be considered when a treatment plan is discussed.

The epidemiology of PLMD is less understood because no well-constructed studies are available using the more current definition for PLMD [7]. However, large case series suggest that this diagnosis is not uncommon in pediatric sleep clinics in which children are routinely evaluated for PLMD [9,40-42].

PATHOPHYSIOLOGY — Genetics, brain iron deficiency, and brain dopamine have been implicated in the pathophysiology of RLS and PLMD [43-47]:

●A genetic mechanism is suggested by the high familial prevalence (40 to 92 percent) among individuals with RLS onset before age 40 [37,39,48,49]. Similarly, in a pediatric population-based study, a family history of RLS was reported in 71 percent of affected children between 8 and 11 years and in 80 percent of those between 12 and 17 years [24]. Both parents were affected in 16 percent of these cases. In addition, several genome-wide studies have found an association of RLS with genetic variants of BTBD9, MEIS1, PTPRD, SKOR1, TOX3, and MAP2K5 [44,50-53]. More recently, 13 new risk loci have been identified [54]. A single study of 23 children with RLS found an association with variants in MEIS1 and MAP2K/LBXCOR1 but not in BTBD9 [55]. Thus far, two of the variants, BTBD9 and MEIS1, appear to influence expression of periodic limb movements of sleep (PLMS), as well as iron homeostasis [50,56].

●Iron deficiency is implicated in the pathogenesis of RLS and PLMS, as suggested by the association between these disorders and low iron stores found on autopsy data [57], magnetic resonance imaging (MRI) [58,59], brain sonography [60,61], and cerebrospinal fluid analysis [62]. Iron is important in brain dopamine production and synaptic density, as well as in myelin synthesis, in energy production, and probably in norepinephrine and serotonin neurotransmitter systems [63-65].

●A role for brain dopaminergic signaling in RLS and PLMS is supported by the efficacy of dopaminergic drugs in treating these disorders, demonstrated in multiple randomized clinical trials [66]. (See 'Medication' below.)

CLINICAL FEATURES — Among children presenting to our pediatric sleep and neurology clinic who were ultimately diagnosed with restless legs syndrome (RLS) or periodic limb movement disorder (PLMD), we have observed the following characteristics [9]:

Chief complaint:

●Sleep concerns – 46 percent

●Behavior concerns – 46 percent

Sleep problems:

●Restless sleep – 86 percent

●Sleep onset problems – 81 percent. Sleep onset problems are somewhat more common in children diagnosed with RLS compared with those with PLMD.

●Sleep maintenance problems – 65 percent. Sleep maintenance problems are somewhat more common in children diagnosed with PLMD compared with those with RLS.

Surprisingly, leg symptoms rarely are the chief complaint. We think this is because caregivers are more likely to be aware of and concerned about the behavioral impact of these disorders rather than the more specific leg or limb symptoms. These observations underscore the importance of asking specifically about leg and arm symptoms, if the family does not spontaneously relate concerns. (See 'Diagnosis' below.)

We also noted a high prevalence of attention deficit hyperactivity disorder (ADHD) in this sample (78 percent), which reflects our clinical focus on ADHD. The prevalence of other comorbid conditions probably varies among clinical centers that see children with these disorders. Leg discomfort may be a more common presentation of RLS in a pediatric rheumatology clinic, and limb jerking during sleep may be more common as a presentation of PLMD in an epilepsy clinic.

History — To evaluate for RLS, the key clinical characteristics that should be elicited by the history are (table 1):

●An urge to move the legs, usually accompanied by uncomfortable or unpleasant sensations in the legs

●These symptoms:

•Begin or worsen during rest or inactivity (eg, lying down or sitting)

•Are relieved by movement

•Occur exclusively or predominantly in the evening or night

●These symptoms are not solely accounted for by another medical or behavioral condition

These symptoms are subjective but follow typical patterns that are described in different ways by children. Therefore, it is important to have the child describe any leg-related symptoms in his or her own words rather than by the caregiver and to make these inquiries with open-ended questions, using age-appropriate vocabulary (table 2). Descriptive responses should be encouraged instead of yes/no responses. Sufficient prompts should be used to determine whether RLS diagnostic criteria are met, but once the child starts describing symptoms, "tell me more" is often the best way to elicit further diagnostic information.

The interview may start with nonspecific prompts, such as "How did you sleep last night?". More specific prompts to assess the urge to move the legs are "Do your legs bother you?"; "Do your legs bother you at night?"; "Do your legs bother you at school?"; and "Do your legs ever hurt?". It is very uncommon for young children to understand or use the word "urge." If given the opportunity, most children with RLS will use more than one descriptor for the RLS sensations. Common descriptors used by children are "want to move"; "have to move"; "got to kick"; "bugs"; "ants"; "tingling"; "wiggly"; "itchy"; "pain"; and "hurts" [8,24,67-69]. If "pain" or "hurts" is the initial descriptor, it is important to probe for additional descriptors since many of the conditions in the differential diagnosis are described as painful but lack additional RLS characteristics. Another helpful prompt to identify discomfort is "Is that a good or a bad feeling?".

To identify the temporal characteristics of the symptoms, useful prompts are "What are you doing when you get those feelings?" or "Do you get those feelings more during the day or at night?". To probe for relief with movement, a useful prompt is "When you move your legs are the feelings better, worse, or about the same?".

An additional technique that we find useful diagnostically is having the child make a drawing of symptoms (figure 1 and figure 2). We have found that children will often talk about their RLS symptoms in detail if they are asked to draw a picture of the feelings [68].

Finally, the clinical significance of RLS or PLMD is determined by assessing for significant distress or impairment in social, occupational, educational, or other important areas of functioning. Caregiver input is helpful in this assessment. The impact of the symptoms on sleep, mood, and schoolwork are typically the focus of concerns [24,29,30,70,71].

PLMD is characterized by repetitive limb jerking in sleep, in association with clinical disturbance of sleep or impaired daytime function that cannot be accounted for by another sleep disorder or etiology [7]. The possibility of PLMD may be raised by the history but must be confirmed by polysomnography because caregiver or child observations for periodic limb movements of sleep (PLMS) are not reliable for diagnosis [8,9,72-75]. Caregivers report children with PLMD as "restless in sleep" in 70 to 90 percent of cases [9,40,72]. However, this descriptor is not specific for PLMD and can be seen in a similar percentage of children with obstructive sleep apnea [40]. Observed "leg kicking in sleep" is reported in 40 to 50 percent but also has a low specificity. Conversely, absence of restless sleep and of leg kicking in sleep has a high negative predictive value. Other findings that can raise the index of suspicion for PLMD include a family history of RLS [9,40], a personal history of some but not full diagnostic criteria for RLS [8], hypertension [76], and a history of frequent parasomnias, such as confusional arousals, sleep terrors, or sleepwalking [9,40,77].

Polysomnography and accelerometry — For PLMD, polysomnography is essential for diagnosis and helps to rule out sleep apnea as the etiology for leg movements. Polysomnography should be considered for children with clinical sleep disturbance (for which another cause is not evident) and with a history suggestive of PLMD [78].

Night-to-night variability of PLMS can complicate the assessment by polysomnography [79-83]. Therefore, in cases in which we have a high index of suspicion for PLMD and in which polysomnography is negative for PLMS, we monitor for PLMS by five nights of ankle accelerometry [80,84,85]. In the scoring of pediatric PLMS, it should be appreciated that the intermovement intervals may be quite variable, resulting in irregular-appearing clusters, as shown in the bottom panel of the figure (waveform 1), rather than the highly stereotyped "runs" seen in adult RLS [86-89]. Of note, PLMS greater than 5/hour are usually abnormal in children [86,90-94]. An exception is normal three- to five-year-olds at high altitude, who were found to have a mean PLMS index of 10.1 [95]. Also, new evidence suggests slightly higher normative values for children one to five years of age, with PLMS during rapid eye movement (REM) sleep contributing to the increase [96]. In contrast, PLMS greater than 5/hour are common in apparently asymptomatic adults over age 40 years [97,98]. Interestingly, the cardiovascular lability associated with PLMS in adults [99,100] has also been found in children [101]. PLMS typically occur during stage N1 and N2 sleep, decrease in frequency in stage N3, and are usually absent during REM (also known as stage R) sleep [7].

Iron deficiency — Iron deficiency is associated with both RLS and PLMD, and iron supplementation improves symptoms in a subset of patients. Therefore, laboratory evidence of iron deficiency supports the diagnosis of RLS or PLMD. However, iron deficiency is neither necessary nor sufficient to make the diagnosis of RLS or PLMD, because each of these disorders is of a multifactorial etiology. (See 'Pathophysiology' above and 'Iron supplementation' below.)

DIAGNOSIS — In children, RLS is closely related to PLMD, although they are distinct diagnostic entities [1,8,9]. Periodic limb movements of sleep (PLMS) are central to the diagnosis of PLMD but also are a common and supportive finding in RLS. A diagnosis of RLS supersedes a diagnosis of PLMD. Thus, a child can have RLS with PLMS but cannot have RLS and PLMD [7]. Some children (especially those who are young or who are developmentally delayed) are diagnosed with PLMD but cannot be adequately assessed for the possibility of RLS, because they lack the language skills to describe the sensory component of RLS.

Restless legs syndrome — The diagnosis of RLS is based primarily upon the clinical history. In atypical or equivocal cases, such as those in which RLS sensations are worse during the day, polysomnography is useful [78]. PLMS determined polysomnographically are supportive features of RLS [5,7,102]. In RLS, PLMS >5/hour have been found in 63 to 74 percent of cases [9,55,103]. However, the overall frequency and periodicity of PLMS in children and adolescents with RLS are less than those of adults with RLS [104].

Updated criteria for the diagnosis of RLS in children were published in 2013 [5]. These were developed by a pediatric RLS expert panel in collaboration with the International Restless Legs Syndrome Study Group [5,6]. The criteria previously used for diagnosis of pediatric RLS were simplified and integrated with the newly revised adult RLS criteria (table 1). There are specific recommendations for pediatric application of the criteria, including consideration of typical words used by children to describe their symptoms (table 2). (See 'History' above.)

The five diagnostic criteria for RLS are:

●An urge to move the legs, usually accompanied by uncomfortable or unpleasant sensations in the legs

●The symptoms begin or worsen during rest or inactivity (eg, lying down or sitting)

●The symptoms are relieved by movement

●The symptoms occur exclusively or predominantly in the evening or night

●These symptoms are not solely accounted for by another medical or behavioral condition

This final criterion is important to the specificity of RLS, which ensures that "mimics" are not misdiagnosed as RLS [105]. (See 'Differential diagnosis' below.)

Additionally, the clinical significance of RLS is assessed by determining whether the symptoms cause significant distress or impairment of function. The domains of impact have been defined for both adult and pediatric populations [24,70,106,107]. For children, the impact of RLS symptoms on sleep, awake activities, emotions, and quality of life is prominent [23,30,70,71,108].

Diagnosis of RLS in children with neurodevelopmental conditions can be very difficult. Structured behavioral observations, including a Suggested Clinical Immobilization Test, are being developed as a way to better assess this population [109].

Periodic limb movement disorder — The diagnosis of PLMD in children is based primarily upon findings on polysomnography [5,7,78] and requires each of the following (see 'Polysomnography and accelerometry' above):

●PLMS greater than or equal to five episodes per hour, as measured during polysomnography

●Clinical sleep disturbance or impaired daytime function (eg, difficulty sleeping or daytime fatigue)

●Exclusion of other causes of the PLMS

Specific criteria for PLMD are described in the table (table 3). The presence of clinical sleep disturbance or daytime fatigue is essential to differentiate PLMS as a polysomnographic finding from PLMD as a disorder [7,41]. In addition, other conditions associated with PLMS, such as obstructive sleep apnea, must be ruled out.

Laboratory tests — We suggest evaluating the iron status of all children with suspected or established RLS or PLMD because of the association between these disorders and iron deficiency. (See 'Pathophysiology' above and 'Iron deficiency' above.)

Serum ferritin is the best single measure of iron stores. However, serum ferritin is an acute phase reactant, and false elevation can occur for up to four weeks after febrile illness [110]. As a result, we wait four weeks after acute illness before performing this test. If there is suspicion for chronic inflammation, then this can be investigated by the measurement of C-reactive protein (CRP). Less sensitive iron measures such as percent iron saturation and total iron-binding capacity (TIBC) also may be helpful.

Differential diagnosis

Restless legs symptoms — Common and uncommon conditions that may mimic pediatric RLS are listed in the table (table 4). Important distinctions are:

●Positional discomfort is easily relieved by simple position change rather than the recurrent movements needed for RLS.

●Leg pain is a characteristic of several disorders mimicking RLS, but in most of these conditions, the pain is not associated with a need to move the legs. In a large pediatric case series of RLS, pain was a descriptor for RLS in almost one-half of cases, but it was never the only descriptor [68]. Thus, the other descriptors, including those indicating a need to move, "bugs," "ants," and "weird feelings," are more likely to be selective for RLS and should be the focus in diagnosis. (See 'History' above.)

●Numbness is not a typical descriptor of RLS, helping in the differentiation from positional ischemia and peripheral neuropathy.

●Nocturnal leg cramps are characterized by sudden painful tightness and by palpable hardening of the muscle. Initially, movement worsens the pain. Nocturnal leg cramps are uncommon in children eight years of age or younger but occur infrequently in approximately 7 percent of older children [111]. (See "Nocturnal leg cramps".)

●"Growing pains" is a diagnosis of exclusion, and it should not be the final diagnosis if RLS criteria are met [24,112-114]. (See "Growing pains".)

●A newly described pediatric sleep disorder, restless sleep disorder (RSD), is characterized by large muscle group movements of the arms, legs, trunk, or head [115-119]. By definition, the movements are exclusively sleep-related, occur most nights, and cause clinically significant impairment. Video-polysomnography demonstrates movements that exceed 5/hour of sleep but PLMS <5/hour and an absence of significant sleep apnea. The condition is not better explained by another disorder, such as RLS, atopic dermatitis, or asthma, or the effect of a substance such as caffeine [120]. Like RLS, RSD may respond to iron supplementation [121]. (See "Sleep-related movement disorders in childhood", section on 'Restless sleep disorder'.)

While is it very important that RLS symptoms should not be solely accounted for as symptoms primary to another condition, it should also be appreciated that any of these conditions can co-occur in an individual who has RLS, including leg cramps or even sickle cell disease. This necessitates a separate focus on each possible condition in the diagnostic process and in the assessment of impact [105].

Leg movements during sleep — The diagnosis of PLMD depends upon documentation of abnormal PLMS on polysomnography (table 3). However, PLMD is also a diagnosis of exclusion, so it is important to determine whether the PLMS are attributable to some other cause, including RLS, sleep apnea, narcolepsy, and some medications.

●If there is evidence of RLS or obstructive sleep apnea, these diagnoses take precedence over PLMD because they are probably the cause of the abnormal PLMS. (See 'Restless legs syndrome' above and "Evaluation of suspected obstructive sleep apnea in children".)

●Testing for narcolepsy is recommended when PLMS are accompanied by moderate to severe daytime sleepiness because this level of sleepiness is uncommon in children with PLMD [8,9,73,78]. The diagnostic evaluation for narcolepsy includes a focused history, physical examination, polysomnography, actigraphy, and multiple sleep latency test (MSLT). (See "Clinical features and diagnosis of narcolepsy in adults", section on 'Diagnostic evaluation' and "Clinical features and diagnosis of narcolepsy in children", section on 'Diagnosis'.)

●Some medications can induce or exacerbate PLMS, particularly serotonergic antidepressants [122-125]. (See "Medical disorders resulting in problem sleeplessness in children", section on 'Medication-induced sleep disturbance'.)

●Rapid eye movement (REM) sleep behavior disorder (RBD) is rare but can occur in children, and it is often associated with PLMS [126]. RBD is characterized by aggressive motor behavior as part of dream enactment, resulting from loss of muscle atonia during REM sleep. (See "Parasomnias of childhood, including sleepwalking", section on 'Rapid eye movement sleep behavior disorder'.)

●Myoclonic epilepsy can be associated with nocturnal limb movements and myoclonic jerks. Myoclonic epilepsy is usually distinguished by daytime seizures, myoclonic movements upon awakening, and/or abnormalities on the electroencephalogram (EEG) recorded during a sleep study. However, distinguishing epilepsy from PLMS occasionally is difficult in a child with only nocturnal symptoms. (See "Epilepsy syndromes in children", section on 'Juvenile myoclonic epilepsy'.)

●Some normal movements during sleep may be confused with PLMS:

•Hypnic jerks (sleep starts) are typically limited to the transition from wakefulness to sleep, are not periodic, and are briefer than PLMS. (See "Sleep-related movement disorders in childhood", section on 'Sleep starts (hypnic jerks)'.)

•Movements associated with normal phasic REM sleep are limited to REM sleep and typically occur in irregular bursts.

COMORBIDITY — Multiple neuropsychiatric and somatic comorbidities have been described in association with pediatric RLS and periodic limb movements of sleep (PLMS) [127]. Attention deficit hyperactivity disorder (ADHD), depression, and anxiety occur more frequently in children and adults with RLS than in the general population. Approximately one-quarter of individuals with RLS meet criteria for ADHD [24,128]. Conversely, 7 to 35 percent of children with ADHD meet criteria for RLS [129-133]. The co-occurrence of ADHD with PLMS may be a clinically significant phenotype, characterized by greater executive function deficits and internalizing behaviors [75].

RLS frequently is associated with depression and anxiety in adults [134-137], as demonstrated in a large population-based study that reported odds ratios of 4.7 for major depressive disorder and 12.9 for panic disorder in individuals with RLS [138]. Similarly, children with RLS have increased risk for depression and anxiety [8,24,139]. A study at a pediatric sleep center found that two-thirds of children with RLS have comorbid psychiatric conditions [139].

Commonly, children with RLS and PLMD have comorbid parasomnias, including sleepwalking, night terrors, nightmares, and enuresis [9,25,73,140].

Children with chronic kidney disease have an increased prevalence of RLS as compared with healthy children [141-144], similar to findings in adults with chronic kidney disease [145,146] (see "Clinical features and diagnosis of restless legs syndrome and periodic limb movement disorder in adults", section on 'Uremia'). In addition, children with sickle cell disease have an increased prevalence of symptoms of RLS, sleep disruption, and frequent PLMS [147,148]. There is also evidence for an increased prevalence of RLS or PLMS in children with iron deficiency [149,150], obesity [29,151], nocturnal enuresis [152-154], hypertension [76], liver transplant [127], Ehlers-Danlos syndrome [155], migraine [156-158], and tension-type headache [157]. A history of prematurity is associated with increased risk for pediatric RLS and PLMD [159]. This is an association for which iron deficiency may play a role. However, data for an increased prevalence of RLS or PLMS in autism are discordant [160-163].

TREATMENT — Multiple treatment reviews for adult RLS and PLMD have been published [66,164-166] but only a few for children [2,4,167]. Because there is limited direct information for children, the pediatric recommendations discussed below are based upon empirical data from adults, pediatric case series, pediatric case reports, and our clinical experience. When considering treatment options, the following should be taken into account: certainty of diagnosis, severity and impact of symptoms, comorbidities, and potential risks of individual treatments. For both pediatric RLS and PLMD, the focus of treatment is typically improved sleep rather than leg symptoms.

The major points of the recommendations are summarized in an algorithm for the evaluation and management of pediatric RLS (algorithm 1).

Nonpharmacologic interventions — Establishing healthy sleep habits is an important aspect of a comprehensive treatment plan. In milder cases, these interventions alone can be sufficient (table 5A-B). (See "Behavioral sleep problems in children", section on 'Young children with behavioral insomnia'.)

Families and caregivers may need specific advice on recommended hours of sleep for age (figure 3):

●2 years of age – Approximately 12 hours (with one nap included)

●5 years of age – Approximately 11 hours

●10 years of age – Approximately 10 hours

●Adolescence – Approximately 9 hours

Physical exercise should be encouraged. It has been shown to increase deep sleep in children, improve RLS symptoms, and be of benefit for mental health, especially depression [168-170]. On nights when RLS sensations are particularly bothersome, the symptoms may be improved by leg massage, which acts as a counter-stimulus.

Exacerbating factors — Assessing for factors that may aggravate RLS or periodic limb movements of sleep (PLMS) and working to reduce these are important. Known or suspected exacerbating factors include insufficient sleep, irregular sleep, obstructive sleep apnea, pain, caffeine, and nicotine [66,171,172]. Some medications, including sedating antihistamines, serotonergic antidepressants, and dopamine blockers, can aggravate RLS and/or PLMS [66,122,124,171,173].

Iron supplementation — For children with evidence of iron deficiency and RLS or PLMD, we suggest a trial of oral iron supplementation (algorithm 1). This suggestion is based upon evidence for a role of iron deficiency in the pathophysiology of RLS and PLMD, upon evidence of benefit with iron supplementation and consensus guidelines [174].

●Indications – We suggest oral iron supplementation for children with RLS or PLMD whose serum ferritin is <50 mcg/L. We also consider oral iron treatment for those with serum ferritin in the 50 to 75 mcg/L range, consistent with newer evidence in adults and children [165,175-179]. Consensus guidelines recommend oral iron for adults with serum ferritin <75 mcg/L [165,174]. Intravenous iron may be considered for moderate to severe RLS when ferritin is <50 mcg/L and is not improving with oral iron, or when there is a condition that blocks oral iron absorption or makes a response unlikely [174]. Intravenous iron should be administered at an infusion center with pediatric experience and with care taken to avoid extravasation of iron subcutaneously. (See "Iron deficiency in infants and children <12 years: Treatment", section on 'Intravenous iron therapy'.)

●Dose and monitoring – We suggest the following approach for oral iron:

•Adolescents 12 years and older – Ferrous sulfate 65 mg of elemental iron per tablet, 1 to 2 tablets once daily.

•Children younger than 12 years – Ferrous sulfate 3 mg/kg/day of elemental iron (to a maximum of 130 mg elemental iron daily), administered once daily.

The rationale for these doses and strategies to avoid side effects and optimize adherence are discussed separately. (See "Iron requirements and iron deficiency in adolescents", section on 'Oral iron therapy' and "Iron deficiency in infants and children <12 years: Treatment", section on 'Oral iron therapy'.)

•Iron should not be taken with milk or other dairy and calcium-containing products, as this will lessen absorption.

•We also advise parents and caregivers about safe storage of iron supplements to prevent accidental ingestion, which can be lethal. (See "Acute iron poisoning".)

We recheck serum ferritin in two to three months and adjust treatment as needed to target serum ferritin levels between 50 and 100 mcg/L. This range generally reflects optimal body iron stores, unless the value is falsely elevated by an inflammatory state. For treatment of RLS and PLMD, it appears to be important to target optimal iron status rather than typically defined "normal" ranges for tests of iron sufficiency. (See 'Laboratory tests' above.)

Once iron stores have been restored, we continue to monitor the child's iron status by checking serum ferritin periodically. In many children, it is difficult to attain and maintain a serum ferritin level >50 mcg/L due to growth, decreased iron absorption once levels get to the low-normal range, and due to blood loss with menses in adolescent girls. In some children, we continue therapeutic doses of iron in anticipation of growth spurts. It is important to monitor children during treatment with therapeutic doses of iron to avoid the rare but serious complication of iron overload, which can occur in individuals who carry hemochromatosis genes [180].

●Efficacy – Iron supplementation is supported by pediatric case series and results from trials of oral iron in children with RLS or PLMS, with or without attention deficit hyperactivity disorder (ADHD) [4]. In four pediatric studies, treatment with oral iron supplements was associated with symptomatic improvement, accompanied by a rise in mean serum ferritin. Oral iron supplements achieved a rise in serum ferritin from 26.6 to 83.5 mcg/L and 21 to 59 mcg/L, respectively, in two uncontrolled studies in children with RLS [175,181]; from 40.8 to 74.1 mcg/L in an uncontrolled study in children with PLMS [149]; and from 29.1 to 55.7 mcg/L in a randomized study of children with RLS or PLMD symptoms and with ADHD [182]. Two additional uncontrolled studies, including one with structured outcomes measures, focused on children with RLS and reported high rates of symptomatic improvement with oral iron [71,183]. In the only study of long-term oral iron supplementation, children with RLS or PLMD had improved clinical symptoms and PLMS at three to six months, one to two years, and >2 years, coinciding with average serum ferritin values increasing from 27.4 at baseline to 45.6, 52.0, and 54.7 mcg/L, respectively [176]. Interestingly, oral iron reduced PLMS for children who were on serotonergic medications as much as those who were not. Parasomnias comorbid with RLS and PLMD resolved with oral iron therapy in 40 percent of children in an uncontrolled study [140].

Intravenous iron has been studied in children who fail or do not tolerate oral iron. A retrospective study of intravenous iron in 16 children with RLS reported improved sleep in 62.5 percent, coinciding with a rise in the mean serum ferritin level from 16.4 to 45.7 mcg/L [184]. A larger study of 39 children with RLS or PLMD treated with a single dose of intravenous iron found that clinical symptoms were much or very much improved in all but one patient, with improved laboratory parameters and few adverse events [185].

Medication — No medications have been approved by the US Food and Drug Administration (FDA) for pediatric RLS or PLMD. This is also true for other pediatric sleep disorders, despite evidence of long-term negative impact when sleep problems in children are not treated [186]. Therefore, the interventions discussed below are off-label uses of medications based upon evidence extrapolated from studies in adults and from limited experience in children.

For adult RLS, multiple, large, randomized controlled trials support use of dopaminergic medications and gabapentin, with lower levels of evidence for other treatments [164]. For pediatric RLS, we generally emphasize nonpharmacologic interventions for mild cases and reserve medication for chronic moderate to severe cases in a setting of long-term specialty care.

The following recommendations apply to children 6 to 18 years of age:

Gabapentin — We suggest gabapentin as a first-line medication for children six years and older with RLS who require pharmacotherapy. Gabapentin has been shown to improve sleep quality and to reduce the sensory symptoms of RLS [187-189]. It is approved by the FDA as an anticonvulsant for children as young as three years. Gabapentin enacarbil is a prodrug with more linear pharmacokinetics that is FDA-approved for adult RLS [190-192] (see "Management of restless legs syndrome and periodic limb movement disorder in adults", section on 'Gabapentinoids'). We prefer gabapentin to dopamine agonists because of concerns about impulse control issues with dopamine agonists. (See 'Dopamine agonists' below.)

●For children 6 to 12 years of age – Begin gabapentin at 100 mg, one-half to one hour prior to bedtime, and increase every one to two weeks by 100 mg until symptoms are suppressed or to a maximum of 600 mg.

●For children older than 12 years of age – Begin gabapentin at 200 mg, one-half to one hour prior to bedtime, and increase every one to two weeks by 100 mg until symptoms are suppressed or to a maximum of 900 mg.

Benzodiazepines — In adult treatment algorithms, benzodiazepines are considered second- or third-line agents for treatment of RLS [66,165]. Several pediatric case series have reported benefit with benzodiazepines for RLS and PLMD [8,9,40,71,73,103,175]. A paradoxical alerting reaction is the most common adverse effect [40]. Although the possibility for next-morning "hangover" sedation should be monitored, we have found this to be very uncommon in RLS and PLMD, probably due to the hyperarousal seen in these disorders. Benzodiazepines are particularly useful for children with disturbed sleep or comorbid anxiety. (See 'Children with anxiety and/or depression' below.)

●Clonazepam – 0.25 to 0.5 mg, one-half hour prior to bedtime.

●Temazepam – The initial dose is 7.5 mg, one-half hour prior to bedtime. For children 6 to 12 years of age, the dose may be increased to 15 mg if needed. For children older than 12 years of age, the dose may be increased sequentially by 7.5 mg every five to seven days, to a maximum of 22.5 mg, if needed.

Clonidine — Clonidine was developed for the treatment of hypertension but was subsequently found to help ADHD and sleep onset in children. It is the most commonly used prescription medication for sleep in children [193-195] and is particularly useful when there are severe sleep onset problems in pediatric RLS and PLMD. However, because clonidine has a relatively short duration of effect, it may not be effective for problems with sleep maintenance. Adverse effects include vivid dreams or nightmares, which are dose-related and which occur in approximately 5 percent of children treated with clonidine. Because of a risk for rebound hypertension, clonidine should not be stopped abruptly.

●For children 6 to 12 years of age – An initial dose is 0.05 mg, one-half hour prior to bedtime. The dose can be sequentially increased every three to seven days in 0.05 mg/day increments until sleep onset is improved or until a maximum of 0.3 mg per dose is reached.

●For children older than 12 years of age – An initial dose is 0.1 mg, one-half hour prior to bedtime. The dose can be sequentially increased every three to seven days in 0.1 mg/day increments until sleep onset is improved or until a maximum of 0.4 mg per dose is reached.

Dopamine agonists — We suggest using the dopamine agonists pramipexole or ropinirole for children with moderate to severe RLS who have failed other treatments. These drugs are considered first-line medications for adults with RLS, and multiple published studies have shown efficacy for pediatric RLS [8,9,40,55,71,103,196-199]. However, in adults, dopamine agonists have been associated with augmentation (paradoxical worsening of RLS sensations over time) and with impulse control problems. In addition, dopaminergics have a limited effect on disturbed sleep [200-202]. Therefore, we do not use these as first-line agents for treatment of RLS in children, and when we do use them, we monitor the children carefully for augmentation and changes in behavior [66]. (See "Management of restless legs syndrome and periodic limb movement disorder in adults", section on 'Dopamine agonists'.)

●Pramipexole – We initiate treatment at 0.125 mg once daily. We advance, if needed, to a maximum of 0.25 mg daily in children 6 to 12 years of age or to a maximum of 0.375 mg daily in adolescents.

●Ropinirole – We initiate treatment at 0.25 mg once daily. We advance, if needed, to a maximum of 0.5 mg daily in children 6 to 12 years of age or to a maximum of 0.75 mg daily in adolescents.

For children with RLS, these medications should be given two hours before the RLS symptoms start, which is usually in the evening. For those with PLMD, they should be given one to two hours prior to bedtime.

Combination drug therapy — In refractory cases of RLS or PLMD, it is important to reassess for possible alternate diagnoses or comorbidities, for iron status, and for compliance with nonpharmacologic and pharmacologic therapies. If this reassessment does not lead to an alternate strategy, then we suggest a trial of the combination therapy of a dopamine agonist and a medication that will consolidate sleep, such as gabapentin or temazepam. An alternative combination is clonidine (for sleep onset) with gabapentin, clonazepam, or temazepam (for sleep maintenance).

Other — A small open-label study of the dopamine agonist rotigotine, administered to adolescents with RLS via a transdermal patch, found similar pharmacokinetics as in adults, improvement of symptoms, and good tolerance [203]. A single open-label study has reported benefit of levetiracetam for children with RLS and comorbid ADHD with interictal epileptic discharges on electroencephalogram (EEG) [204]. Melatonin was found to have a detrimental effect on motor symptoms in a small study of adults with RLS [205]. Levodopa/carbidopa was of benefit for children with RLS and PLMS in a small double-blind, placebo-controlled study [196]. However, for clinical use in RLS, this medication has been supplanted by dopamine agonists, which have fewer adverse effects. For severe refractory RLS in adults, opioids are considered [66,164,206], but evidence is lacking for children.

Special populations — Special considerations apply to pharmacotherapy in the following populations:

Young children — For children younger than six years of age, we emphasize nonpharmacologic interventions and optimization of iron status. However, for those who have a clear diagnosis of RLS or PLMD and who continue to have substantial impairment despite optimal iron supplementation and nonpharmacologic interventions, we sometimes treat with low-dose clonazepam (not clonidine) or gabapentin. (See 'Benzodiazepines' above and 'Gabapentin' above.)

Children with attention deficit hyperactivity disorder — As long as the effect has worn off by bedtime, the stimulant medications used for treatment of ADHD do not adversely affect RLS, PLMS, or sleep [86,207-209]. Conversely, treatment of the sleep disorder may improve ADHD symptoms to some extent [199]. Interestingly, iron supplementation may benefit RLS, PLMD, and ADHD, suggesting the possibility of a common mechanism underlying these disorders [4,210].

Children with anxiety and/or depression — In children with comorbid anxiety or depression, treatment of RLS or PLMD may improve the mood symptoms [135,211,212]. However, if the depressive and anxiety symptoms are prominent, specific treatment is typically needed. In this case, family or individual counseling should be considered.

If pharmacotherapy is needed for depression, we suggest using a non-serotonergic medication such as bupropion because serotonergic antidepressants can worsen RLS or PLMS [66,124,135,211,213-216] (see 'Exacerbating factors' above). Bupropion also has significant anti-anxiety properties [217-219]. Bupropion sustained-release or extended-release is given in the morning.

Other options include:

●Low-dose benzodiazepines, such as clonazepam or temazepam, given at bedtime – For children with comorbid anxiety. Potential adverse effects include paradoxical hyperactivity or irritability. (See 'Benzodiazepines' above.)

●Low-dose fluoxetine (10 mg) or duloxetine (20 mg) – For children with comorbid depression, if the response to bupropion or a benzodiazepine is incomplete. (See "Pediatric unipolar depression and pharmacotherapy: Choosing a medication".)

●Dopamine agonists, such as pramipexole or ropinirole – These drugs are effective for RLS or PLMD and also have some antidepressant properties [220-224]. (See 'Dopamine agonists' above.)

Psychosocial considerations — Family support is very important for children who are affected by chronic medical problems such as RLS and PLMD. Because RLS is highly familial, the entire family may benefit from effective treatment for an affected parent. Furthermore, parents and caregivers often need specific guidance to help the child establish and maintain a good sleep schedule to minimize RLS and PLMD symptoms. As adolescence approaches, it is important to foster self-awareness and self-management skills. (See 'Nonpharmacologic interventions' above.)

Advocacy for appropriate school accommodations may be needed when the disorder interferes with a child's ability to function in the classroom. Accommodations may include allowing physical activity during recess, avoiding long stationary positions in the classroom, and giving permission to move around if leg sensations are bothersome.

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Restless legs syndrome".)

SUMMARY AND RECOMMENDATIONS

●Restless legs syndrome (RLS), also known as Willis-Ekbom disease, is a common and treatable neurologic condition. It is characterized by an urge to move the legs, usually accompanied by uncomfortable or unpleasant sensations in the legs. The symptoms begin or worsen during rest or inactivity, are relieved by movement, and occur exclusively or predominantly in the evening or night (table 1). (See 'Restless legs syndrome' above.)

●The diagnosis of RLS is based primarily upon the history, which must be carefully targeted to capture relevant information from the patient (rather than just the caregiver). Other causes of similar symptoms should be considered and excluded (table 4). (See 'History' above and 'Restless legs symptoms' above.)

●Periodic limb movement disorder (PLMD) is characterized by repetitive limb movements during sleep (known as periodic limb movements of sleep [PLMS]), in association with clinical disturbance of sleep or impaired daytime function that cannot be accounted for by another sleep disorder or etiology. The diagnosis is established by documenting abnormal numbers of PLMS during polysomnography and by ruling out other causes for the symptoms, including RLS and obstructive sleep apnea (table 3). (See 'Polysomnography and accelerometry' above and 'Leg movements during sleep' above.)

●In children, RLS is closely related to PLMD, although they are distinct diagnostic entities. PLMS are central to the diagnosis of PLMD but also are a common and supportive finding in RLS. Both conditions have been associated with relative iron deficiency and have a familial tendency. (See 'Pathophysiology' above.)

●Conditions associated with RLS and PLMD include attention deficit hyperactivity disorder (ADHD), depression, and anxiety. Recognition of these comorbid conditions is important to treatment decisions. (See 'Comorbidity' above and 'Special populations' above.)

●Management of pediatric RLS is summarized in the algorithm (algorithm 1) and includes the following:

•Iron status should be evaluated for all children with suspected or established RLS or PLMD because of the association between these disorders and iron deficiency. A serum ferritin level of 50 to 100 mcg/L (50 to 100 ng/mL) is optimal. For children with ferritin levels below this range, we suggest iron supplementation (Grade 2B). (See 'Laboratory tests' above and 'Iron supplementation' above.)

•Factors that may exacerbate RLS or PLMD include insufficient sleep, irregular sleep, obstructive sleep apnea, pain, caffeine, nicotine, and some medications, including sedating antihistamines, serotonergic antidepressants, and dopamine blockers. (See 'Exacerbating factors' above.)

•Establishing healthy sleep habits is an important aspect of a comprehensive treatment plan for RLS or PLMD (table 5A-B). Regular physical activity also may be helpful. In milder cases, these interventions alone can be sufficient. (See 'Nonpharmacologic interventions' above.)

•For pediatric RLS, we generally emphasize nonpharmacologic interventions for mild cases and reserve medication for chronic moderate to severe cases in a setting of long-term specialty care. For children six years and older with RLS who require pharmacotherapy, we suggest gabapentin as a first-line medication rather than dopamine agonists or benzodiazepines (Grade 2C). Other considerations that affect choice of treatment include the severity of symptoms, the type of sleep disturbance (sleep onset or sleep maintenance), and the presence of comorbid depression, anxiety, or ADHD. (See 'Medication' above.)