Restless legs syndrome in paediatric patients with epilepsy

Abstract

Introduction

Restless legs syndrome (RLS) is a neurological condition with defined clinical criteria. The prevalence of RLS in children is around 2%-4%; as patients approach adulthood, prevalence increases. The prevalence of RLS in paediatric patients with epilepsy is unknown.

Material and methods

Four Madrid hospitals participated in the study (Fundación Jiménez Díaz, Hospital Universitario Niño Jesús, Hospital Universitario Villalba, and Hospital Universitario Infanta Sofía). The researchers developed a prospective observational study of a cohort of patients with epilepsy under follow-up at paediatric neurology consultations. The data were collected in the department of paediatric neurology. The data collection schedule was established between December 2020 and March 2023.

Results

Data were obtained from 117 patients (56% male). The age range was from 4 to 17 years (mean age, 10.3 years). Twenty patients (17.1%) met criteria for RLS, 5 (4.3%) met clinical criteria for possible RLS, and 39 (33.9%) reported insomnia. Sleep onset insomnia was present in 21 patients (18.1%), sleep maintenance insomnia in 8 patients (6.9%), and both types in 6 (5.2%). Insomnia was associated with RLS (P = .041)

Conclusions

In our sample, we observed a higher percentage of RLS among paediatric patients with epilepsy than in the general paediatric population. More studies are needed to confirm this finding.

Resumen

Introducción

El síndrome de piernas inquietas (SPI) es una enfermedad neurológica con criterios clínicos definidos. La prevalencia del SPI en niños es de alrededor del 2-4 %; a medida que el paciente se acerca a la edad adulta, la prevalencia aumenta. Se desconoce la prevalencia del SPI en pacientes pediátricos con epilepsia.

Material y métodos

Estudio observacional prospectivo de una cohorte de pacientes con epilepsia en seguimiento en consultas de Neurología Pediátrica. Participaron cuatro hospitales madrileños (Fundación Jiménez Díaz, Hospital Universitario Niño Jesús, Hospital Universitario Villalba y Hospital Universitario Infanta Sofía). Los datos fueron recogidos en los Servicio de Neurología Pediátrica. El calendario de recogida de datos se ha establecido entre diciembre de 2020 y marzo de 2023.

Resultados

Se obtuvieron datos de 117 pacientes (56% varones). El rango de edad fue de 4 a 17 años (edad media 10.3 años). 20 pacientes (17.1%) cumplieron los criterios del SPI y 5 pacientes (4.3%) cumplieron los criterios clínicos para posible RLS. 39 pacientes (33.9%) reportaron insomnio. El insomnio de conciliación estuvo presente en 21 pacientes (18.1%), el insomnio de mantenimiento estuvo presente en 8 pacientes (6.9%) y 6 pacientes (5.2%) reportaron ambos componentes. La presencia de insomnio se relacionó con SPI de manera significativa (p = 0.041).

Conclusiones

En nuestra muestra hemos obtenido un mayor porcentaje de SPI que en la población pediátrica epiléptica en comparación con la población pediátrica general. Son necesarios más estudios para confirmar este resultado.

Introduction

Restless legs syndrome (RLS; also known as Willis-Ekbom disease) is a neurological disease with defined clinical criteria. Its pathogenesis involves genetic factors, alterations to dopamine metabolism, and an interaction with iron metabolism. Some drugs, such as sedating antihistamines, serotonergic antidepressants, and dopamine antagonists, may exacerbate symptoms. These effects are not described with antiseizure medications (ASM). In fact, gabapentin is an ASM that is used to treat RLS.^1–6

Although the disease was initially described in adults, it is now known also to affect children. Symptoms may even appear at preschool ages. RLS is more common among patients with such conditions as chronic kidney disease, sickle cell anaemia, obesity, iron deficiency, and/or attention-deficit/hyperactivity disorder. RLS affects 5%-10% of adults in the Unites States and Western Europe. Its prevalence is approximately 2%-4% among paediatric patients, increasing closer to adulthood. The importance of identifying and treating RLS lies in its relationship with sleep quality, as well as its impact on the social, work, behavioural, and emotional spheres. RLS is sometimes associated with periodic limb movement disorder (PLMD), though not in all cases.^7–10

Epilepsy is a disease that causes a predisposition to recurrent seizures. It is relatively frequent, with 3%-4% of the general population being affected at some point in their lives. Among children, the prevalence of epilepsy is estimated at 0.6% in developed countries and 4.4% in developing countries. Sleep disorders are estimated to affect 24%-66% of children with epilepsy. Multiple causes explain this increase in sleep disorders among patients with epilepsy. These include manifestations of brain activity due to epileptiform abnormalities, seizures themselves, or even the use of ASMs. Sleep problems may exacerbate somnolence and behavioural alterations, and disrupt seizure control in these patients.^11–16

Studies analysing the prevalence of RLS in adults with epilepsy report differing results. A search of paediatric publications on medical databases (PubMed, MEDLINE, Fisterra), using the terms “epilepsy,” “RLS,” “sleep disorders,” and “Willis-Ekbom disease,” identified no studies of children with epilepsy and RLS. We decided to perform an observational study to establish the percentage of children meeting criteria for RLS among all patients diagnosed with epilepsy and under follow-up at paediatric neurology consultations. As a secondary objective, we gathered data on such other variables as type of seizure, type of epilepsy, and/or the use of concomitant ASMs.^17–20

Material and methods

We conducted a prospective, observational, multi-centre study with the participation of 4 hospitals: Fundación Jiménez Díaz, Hospital Universitario Niño Jesús, Hospital Universitario Villalba, and Hospital Universitario Infanta Sofía. One of these, Hospital Universitario Niño Jesús, is a reference centre for paediatric epilepsy. We developed a prospective, observational study of a cohort of patients with epilepsy under follow-up by the paediatric neurology department. Data were collected from patients meeting the study inclusion criteria. Data were gathered at paediatrics departments from December 2020 to March 2023, and analysed in May 2023. Clinical and demographic data were obtained from patient clinical records. Data were processed with a Microsoft Excel database used exclusively by the research team. The study was approved by the hospital ethics committee. Written informed consent was obtained from the parents of all patients; patients older than 12 years of age were also asked to sign an informed consent form adapted to their age.

Definition of restless legs syndrome

Several sets of diagnostic criteria for RLS have been developed by the American Academy of Sleep Medicine (International Classification of Sleep Disorders-third edition [ICSD-3]), the American Psychiatric Association (Diagnostic and Statistical Manual of Mental Disorders-fifth edition [DSM-5]), and the International Restless Legs Syndrome Study Group (IRLSSG), with subtle differences between the 3. However, all 3 require the presence of the same 5 essential criteria: urge to move the legs, usually due to abnormal sensations; beginning or worsening of symptoms during rest; partially or temporarily relieved by movement, walking, or stretching; worsening of symptoms later in the day or at night; and exclusion of symptoms mimicking RLS. Physicians must tailor their language to the age of the child (Appendix 1).^2,21,23

Inclusion criteria

Inclusion criteria were diagnosis of epilepsy and age between 4 and 17 years. RLS was diagnosed according to the IRLSSG criteria. We excluded patients younger than 4 years and those whose speech or cognitive development prevented them from communicating the presence of RLS symptoms. Data were also collected on the type of seizure, current pharmacological treatment, onset of epilepsy, aetiology of epilepsy, and the associated sleep problems and movement disorders. All patients described symptoms in their own words.

We excluded patients meeting criteria for RLS mimics (positional discomfort, muscle pain in the legs, ligament/tendon distension, dermatitis, haematomas, growing pains, muscle cramps in the legs, arthritis, peripheral neuropathy, radiculopathy, myelopathy, myopathy, fibromyalgia, complex regional pain syndrome, drug-induced akathisia, sickle-cell anaemia).

Epilepsy was considered to have resolved in individuals who had age-dependent epilepsy syndromes but were now past the applicable age, and in those who had been seizure-free for the last 10 years and had not received ASMs in the last 5.

Statistical analysis

The Jamovi software, version 2.3.25, was used for all statistical analyses. Differences between groups were analysed using the chi-square test. P-values < .05 were considered statistically significant.

Results

Data were obtained from a total of 117 patients (56% boys). Age ranged from 4 to 17 years (mean, 10.3). Epilepsy onset occurred before 5 years of age in over 70% of patients. Data on seizure type, epilepsy aetiology, and ASMs used are presented in Table 1. A total of 7 patients met criteria for resolved epilepsy. Twenty patients (17.1%) in our sample met the IRLSSG criteria for RLS. Five met the criteria for possible RLS. Of these 5, 4 reported that the symptoms interfered with their lives. Twenty-two patients (19.8%) reported symptoms of RLS in their parents (this diagnosis was confirmed by an adult care neurologist in 3 families). The association between diagnosis of RLS and family history of the disorder was statistically significant (P < .001; odds ratio [OR]: 39; 95% confidence interval [CI], 10.9-139). Thirty-nine patients (33.9%) reported insomnia: sleep onset insomnia in 21 (18.1%), sleep maintenance insomnia in 8 (6.9%), and both in 6 (5.2%). Patients diagnosed with RLS reported insomnia more frequently than those without RLS (P < .041; OR: 2.73; 95% CI, 1.02-7.31). The sleep maintenance insomnia subtype showed a significant association with RLS (P = .013; OR: 5.56; 95% CI, 1.26-24.5). Three patients with probable RLS presented insomnia. Eight patients reported snoring, with none meeting clinical criteria for sleep apnoea-hypopnoea syndrome. None of these patients met criteria for RLS. No association was found between insomnia and any ASM in monotherapy or polytherapy, seizure type, refractory epilepsy, or epilepsy aetiology.

Table 1.

Clinical characteristics of patients in the study sample.

		Patients meeting IRLSSG criteria for RLS
Aetiology of epilepsy
Idiopathic	71/118 (60.7%)
Genetic	14/118 (12%)	P = .77
Structural	32/118 (27.3%)	P = .75
Seizure type
Focal	67/118 (57.8%)	P = .19
Generalised	39/118 (33.6%)	P = .68
Both	10/118 (8.6%)	P = .13
Refractory epilepsy	48/118 (41.2%)	P = .54
ASM
None	12/118 (10.3%)	P = .54
Monotherapy	60/118 (51.3%)	P = .41
	Levetiracetam 15/118 (13%)	P = .63
	Valproic acid 11/118 (9.4%)	P = .43
	Sodium channel blockers 19/118 (16.2%)	P = .88
	Clobazam 2/118 (1.7 %)	P = .51
	Multichannel blockers other than valproic acid 10/118 (8.5%)	P = .42
Polytherapy	46/118 (39.3%)	P = .41

ASM: antiseizure medications; IRLSSG: International Restless Legs Syndrome Study Group; RLS: restless legs syndrome.

Discussion

RLS is a treatable neurological disease. It is a sensorimotor disorder characterised by uncontrollable sensations in the legs, accompanied by an irresistible urge to move them, generally resulting immediately in partial or complete, though transient, relief of the sensation. It was first described in 1672 by T. Willis, in adult patients. The first paediatric cases were reported by Brenning in 1960. The syndrome is diagnosed according to the clinical criteria of the DSM-5, IRLSSG, and ICSD-3. RLS is caused by predisposing genetic factors and alterations in the metabolism of dopamine and iron.^21–27

The prevalence of RLS in children and adolescents is reported at 2%-4% in population studies conducted in the United Kingdom, United States, Turkey, China, and Brazil, with the prevalence of moderate to severe cases amounting to 0.5%-1%. RLS is known to be more common in patients with coeliac disease, kidney disease, and iron deficiency. Some studies report prevalence of approximately 38% in patients with coeliac disease and 15%-30% in those with kidney disease. It is unclear whether RLS prevalence also increases in patients with such other disorders as hypertension, Parkinson’s disease, or migraine.^28–31

The relationship between RLS and epilepsy has been studied in the adult population. The prevalence of active epilepsy is estimated at 1.2% in adults and 0.62% in children. Some research groups have observed increased risk of RLS among patients with epilepsy, whereas others report no such relationship. According to some authors, patients with temporal lobe epilepsy present higher rates of RLS than the general population. Other studies have analysed the relationship between epilepsy aetiology, type of seizure, and RLS, but conclude than there is no clear relationship between them. In our study, the percentage of children with epilepsy who presented symptoms compatible with RLS was higher than that reported in the general paediatric population.^18,32–34

While epilepsy is more frequent in the male sex, no sex difference was observed in our sample. In the adult population, women are more likely than men to present RLS; higher parity is associated with increased risk of the condition. This female predominance is also reported in children and adolescents. In our sample, however, no difference was observed between boys and girls. RLS is characterised by strong familial clustering, with up to 63% of patients having at least one first-degree relative who is also affected by the condition, particularly in the early-onset form. Risk loci have been described in 5 genomic regions: MEIS1, BTBD9, PTPRD, MAP2k/SKOR1, and TOX3/BC034767. Twenty-two patients (19.8%) reported symptoms of RLS in their parents (this diagnosis was confirmed by an adult care neurologist in 3 families). The association between (confirmed or reported) family history and diagnosis with RLS according to the IRLSSG criteria was also observed in our paediatric sample (Fig. 1).^28,35,36

Figure 1.

Association between restless legs syndrome and family history of restless legs syndrome.

RLS: restless legs syndrome.

Numerous studies report higher rates of sleep disorders among patients with epilepsy than in the general population. Some studies report sleep disorders in 24%-55% of adult patients with epilepsy, and up to 73% among paediatric patients. Compared to healthy children, children with epilepsy present reduced sleep time and greater difficulty sleeping, with particular involvement in the domains night-time wakefulness, parasomnias, and sleep-related breathing disorders. Epilepsy and sleep present a clear bidirectional relationship. Sleep itself, but especially sleep deprivation and sleep disorders, may exacerbate epilepsy. Epileptic anomalies and seizures may vary throughout the night in line with sleep phases. ASMs also influence this situation. One-third of patients in our study met criteria for insomnia (sleep onset insomnia, maintenance insomnia, or both). Patients diagnosed with RLS reported these sleep disorders more frequently than patients without RLS. In our sample, none of the patients reporting snoring met criteria for RLS.^{12,15,33,35,37–40}

Some drugs have been associated with the appearance of RLS symptoms. This is frequently observed with dopamine antagonists (eg, metoclopramide and antipsychotics). Some authors report a similar phenomenon with antidepressants, antihistamines, alcohol, nicotine, and caffeine; however, the methodology of these studies is not sufficiently robust to confirm a causal association. Some ASMs are reported to cause symptoms of RLS; examples include phenytoin, topiramate, and zonisamide. We analysed the use of ASMs in monotherapy and polytherapy. No relationship was observed between RLS and pharmacological treatment.^{21,22,36,41,42}

Epilepsy may result from multiple causes. Seizure type and frequency, and association with wakefulness or sleep vary greatly between patients. Epilepsy presents a clear association with sleep. In some forms of epilepsy (eg, nocturnal frontal lobe epilepsy), seizures occur almost exclusively during sleep. Epileptiform abnormalities are also known to increase during sleep. Epileptiform discharges may be activated or inhibited depending on the phase of sleep. In general, epileptiform discharges probably spread during non–REM sleep and in sleep-wake transitions, as these states are characterised by greater synchronisation between sleep spindles and delta waves. On the other hand, REM sleep is characterised by asynchronous cellular discharge patterns, which decrease the likelihood of propagation of epileptic potentials. In our study, RLS showed no association with epilepsy aetiology or seizure type. Similarly, no association was found between insomnia and seizure type, refractory epilepsy, or epilepsy aetiology (Fig. 2).^12,15,43,44

Figure 2.

Association between restless legs syndrome and insomnia.

RLS: restless legs syndrome.

Our study presents certain limitations. Firstly, it includes no control group, and we did not use polysomnography to objectively study sleep. Furthermore, a high number of patients present refractory epilepsy, as one of the participating centres is a reference hospital for paediatric epilepsy. Finally, we did not collect standardised data on iron metabolism.

Conclusions

We believe that this study contributes new evidence on sleep problems in children with epilepsy. Higher prevalence of RLS was observed in paediatric patients with epilepsy than in the general paediatric population. Further studies are needed to confirm these findings. Given the importance of adequate sleep regulation for seizure control in patients with epilepsy, physicians should seek to identify symptoms compatible with RLS in paediatric patients with epilepsy, and start treatment if these are detected.