Sleep disturbances in SCN8A-related disorders

Abstract

SCN8A encodes the voltage-gated sodium channel subunit Nav1.6, which is expressed in the brain.¹ Neuronal hyperexcitability, seizures, and neurocognitive problems are the result of impaired Nav1.6 channel inactivation.^{2, 3} Pathogenic variants in the SCN8A gene are frequently related to epilepsy, ranging from self-limiting epilepsies^{4, 5} to severe developmental and epileptic encephalopathies (DEE)⁶ often refractory to anti-seizure medications (ASM).^7-9 Gardella and Møller described for the first time the phenotypic spectrum of SCN8A-related disorders detailing the distinguishing features of different sub-phenotypes.^{4-6, 10}

Epileptic seizures and sleep quality have a complex bidirectional relationship; in people with DEE, comorbidities such as intellectual disability, attention deficit, and movement disorder add complexity to this interaction.^{11, 12} Significant sleep disturbances are often observed in patients with DEE, causing major disruption to their quality of life.¹³ Although sleep disturbances are frequently reported in patients with genetic epilepsies, only few studies exploring this issue have been performed.^14-17

Studies in SCN8A as well as SCN1A mice models showed sleep disturbances, such as increased NREM and decreased REM sleep.^{18, 19} Additionally, the mice displayed altered circadian rhythm of corticosterone secretion, with lowered and flattened diurnal level, indicating hypofunctioning hypothalamic–pituitary–adrenal (HPA) axis, and suggesting a sodium channels' role in sleep regulation.^{18, 19}

Our study aims to characterize the prevalence and nature of sleep disturbance in patients with different SCN8A-related disorders.

We enrolled patients with SCN8A-related disorders through a network of physicians and caregivers in Europe and in the USA (14 centers). We included all patients with pathogenic SCN8A variants, with available electro-clinical data, and excluded patients with SCN8A variants of uncertain significance and the ones who did not accept to participate in the study. We reviewed their medical history including demographic and genetic data, epilepsy features, cognitive and motor development, and relevant comorbidities. Information about seizures (types, frequency, and timing—specifically wakefulness versus sleep predominance), seizure control, and medications (anti-seizure and sleep medications) were obtained through a semi-structured spreadsheet.

Since the SCN8A phenotypic spectrum is extremely heterogeneous, using phenotypic subgroups is of pivotal importance. As we previously described based on large cohort studies,^{6, 10} the patients were divided into five phenotypic sub-groups, consisting of (i) Severe DEE, (ii) Focal epilepsy of intermediate severity, (iii) Generalized epilepsy, (iv) Self-limited familial infantile epilepsy, and (v) Neurodevelopmental disorder without epilepsy.

We defined the seizure types according to ILAE classification.²⁰ We defined as “frequent seizures” when occurring from several times daily to weekly, while “rare seizures” if recurring monthly to yearly.

We distributed to caregivers the Sleep Disturbance Scale for Children (SDSC),²¹ the Children's Sleep Habits Questionnaire (22-item version),²² the Pediatric Daytime Sleepiness Scale (PDSS),²³ a sleep diary (adapted from the Sleep Council diary)²⁴ (see supplementary material) to collect a description of their sleep. The SDSC has been validated in patients younger than 20 years, however, we used it also for older patients with intellectual disability, as in previous studies.²⁵

When possible, we also performed 24-h video-EEG-polysomnographic recordings, which have been reviewed and analyzed by two neurologists with expertise in epilepsy (EG and FF) and scored according to the AASM manual by a trained neurologist with expertise in sleep medicine (FF). The setting for video-EEG-polysomnographic recordings consisted of 19 EEG channels, ECG, and four EMG derivations from splenius capitis, mylohyoideus, and left and right anterior tibialis muscles. Video was recorded and reviewed for seizures and sleep movements. We analyzed the EEG and the sleep architecture including: (i) total sleep time (TST) in minutes, (ii) NREM sleep, dividedinto NREM 1 (N1), NREM 2 (N2), NREM 3 (N3) stages (total time in minutes and percentage), (iii) REM (R) stage (total time in minutes and percentage), (iv) wakefulness after sleep onset (WASO; total time in minutes), and (v) arousals rate.

According to the American Academy of Sleep Medicine manual (AASM manual),²⁶ we classified modifications of the EEG frequency (with increased chin tone if in REM stage) as “arousal” if lasting 3–15 s, and as “awakening” if lasting more than 15 s. The representation of NREM3 sleep and REM sleep was considered reduced when accounting for less than 20% and 25% of TST, respectively. Conversely, the representation of NREM1 sleep was considered increased when accounting for more than 5% of TST.²⁶

We enrolled 47 unrelated patients (24 males and 23 females) in age range 2–39 years (median age: 7 years), including 24 novel patients and 23 previously published. They harbored 36 different pathogenic variants (32 missense and 4 truncating), 35 recurring de novo, 3 parental inherited, and 9 unknown inheritance. An overview of the clinical features including epilepsy, neurological, behavioral, and sleep disturbances is reported in Tables 1 and 2 and Table S1.

Our study showed that the majority of patients with SCN8A-related disorders experience sleep disturbances, mainly consisting of difficulties in initiating and maintaining sleep.

Sleep disturbances were more often reported in patients with ongoing seizures and severe/profound motor and cognitive impairment. There is a complex bi-directional relationship between sleep, epilepsy, and developmental disorders. Many types of epilepsy have sleep-activated seizures and interictal epileptiform discharges, with the highest preponderance reported in NREM sleep.^{27, 28} On the other hand, people with epilepsy can have poorer sleep quality and impaired sleep micro- and macro-structure.²⁸ The influence of epilepsy on sleep can be related to shared (patho)physiological mechanisms, to the effect of seizures on sleep architecture, to ongoing ASM, or a combination of all these factors.^{27, 28} Genetic factors can also negatively impair sleep. A number of developmental and epileptic encephalopathies with genetic etiology have been associated with specific sleep disturbances.^{14, 16, 17, 29, 30}

In our cohort, we observed very frequent sleep disturbances in general, with the highest rate in patients with ongoing seizures (independently from seizure frequency), but with greater severity of the most common sleep disturbance (DIMS) in patients with sleep-related motor seizures. This suggests a multifactorial origin of the sleep disorders, which likely results from a combined effect of epilepsy (and sleep-related motor seizures), the developmental encephalopathy, and possibly also of the gene defect itself.

Although parents of patients with sleep disturbances may have been more willing to complete the questionnaire than those without, our overall questionnaire return rate was 65%, comparable to other studies (36%–79%).^{31, 32} In our cohort, 82% of patients with SCN8A-related disorders were reported with sleep disturbances, which is far greater than the sleep disturbances reported in young children in the general population (31%)³³ or in patients with epilepsy (66%).³⁴ The rate and features of sleep disturbances reported in our cohort are similar to those observed in patients with Dravet syndrome (74%), mainly consisting in night awakenings (77.3%) and daytime sleepiness (40.9%).¹⁴ In fact, in our study 82% of patients had sleep disturbances, represented by DIMS (64%), followed by SBD (43%), SWTD and DOES (34% each).

Sleep disturbances were observed also in other genetic neurodevelopmental disorders, such as Angelman syndrome (range 20–80%),¹⁵ Rett syndrome (range: 80–94%),¹⁷ and SYNGAP1-related disorders³⁰ (62%). Each syndrome was characterized by specific sleep disturbances such as bedtime resistance and night awakenings in Angelman and SYNGAP1 diseases, laughing, teeth grinding, and screaming in Rett syndrome and parasomnias and daytime sleepiness in SYNGAP1-related disorders.^{15-17, 30}

Sleep disturbances are reported as one of the major comorbidities that families coping with DEE struggle to negotiate.³⁵ They could increase the likelihood of seizures due to sleep deprivation, could impact on the learning performances, and could affect the family's overall quality of life.³⁴

DIMS seemed to be the most recurring sleep disturbance in patients with SCN8A disorders, and it was the most frequently reported problem at all ages and in all phenotypes, even if more frequent in the severe DEE. The increased propensity to wake up throughout the night represents a marker of sleep instability and might be due to altered sleep architecture in patients with SCN8A-DEE, independently from the presence of seizures during the night. We documented a high WASO (mean WASO: 100 min), with an arousal index within normal ranges (mean arousal index: 1.26/h),^{36, 37} both in patients with and without sleep-related seizures, suggesting sleep instability related to the SCN8A related disorder “per se” and not only to seizure-related sleep disruption. However, WASO was higher during the nights when sleep-related seizures were recorded (Figure 3), highlighting a combined influence of epilepsy on sleep. A retrospective polysomnographic study of children with SCN1A-Dravet syndrome also found increased sleep instability with an increase in cyclic alternating pattern although normal arousal index.³⁸

In the severe DEE group, we observed a higher percentage of all the SDSC items score, except for SHY which resulted in almost the same in the severe and intermediate phenotypes. The high prevalence of sleep breathing disorders was mainly found in individuals with severe motor impairment and might be more likely related to the neurological condition (e.g., hypotonia) than to a specific role of the gene defect on respiration. Sleep breathing disorders in our cohort were expressed especially in adults; this result is in line with the literature, where a higher prevalence is described in patients older than 30 years of age.³⁹

The high percentage of SWTD in our cohort, mainly characterized by the presence of movements in the transition from wakefulness to sleep and during the night, could be overestimated and confused by the caregivers of patients with seizures and physiological hypnagogic myoclonus. A sleep PSG, unfortunately not available for the patients with reported SWTD problems, could help with the differential diagnosis.

Moderate daytime sleepiness does not appear related in our cohort to specific ASM but could be influenced by therapy (73% of the patients in poly-therapy with 3 or more ASMs versus 27% with 1–2 ASMs).

The percentage of DA in our cohort (7%) is slightly lower than in other neurological disorders (23%),⁴⁰ possibly related to the cognitive inability of patients with severe SCN8A-DEE (62% in our cohort) to report nightmares.

The percentage of SHY in our cohort is low (14%), as reported also in other neurological disorders (7.6%).⁴⁰

Looking at the specific sleep scales that we used for this study, we found that the SDSC was the most informative and sensitive supported by the anamnestic report from the caregivers that also added important data, followed by the CSHQ that confirmed the main features, while the PDSS did not look very sensitive.

Polysomnographic recordings on other genetic DEE showed an alteration in sleep architecture consisting of significant reduction in total sleep time, and sleep percentage, as well as significantly higher REM latency, and number of awakenings/h⁴¹; in Angelmann syndrome a significantly lower percentage and duration of REM sleep, and significantly higher percentage of slow waves sleep (SWS) was observed.²⁹ The polysomnographic analysis in our cohort showed increase of WASO, sleep fragmentation due to arousals, and increase of light sleep (NREM1) representation, with representation of the physiological sleep figures. The high rate of awakenings and arousals was a common feature in all recordings, with and without concomitant seizures. During nights with recorded seizures, 90% of the awakenings/arousals were not seizure-related. These results suggest an intrinsic sleep instability in SCN8A-DEE. On the other hand, the fact that the WASO was more prolonged in patients with recorded seizures suggests an influence of epilepsy on sleep. Melatonin was the most used sleep medication for DIMS, effective in 71% of our cohort. The superiority of Melatonin for DIMS in DEEs have been similarly reported also in other DEEs in our experience and in the literature, followed by benzodiazepines,¹⁴ trazodone,³⁰ and clonidine.³⁰

Mouse models of SCN1A-related disorders have shown that the Nav1.1 channel encoded by SCN1A is expressed in cells important for sleep regulation, including the GABAergic neurons in the hypothalamus, thalamic reticular nucleus, and the cortex.¹⁹ A drug-naive SCN1A Dravet syndrome mouse model demonstrated impaired sleep homeostasis secondary to the loss of Nav1.1 channels in the inhibitory forebrain GABAergic neurons, implicating the gene's involvement in sleep disruption.⁴²

Likewise, the SCN8A dysfunction may lead to sleep disruption by dysregulation of neurological sleep networks. In fact, studies on SCN8A mice models showed that the dysfunction of the SCN8A voltage gated sodium channel Nav1.6 alters sleep architecture by reducing diurnal corticosterone levels. This ends in a relative increase in the amount of NREM sleep and a decrease in REM sleep.¹⁸ Similarly, in our patients, we observed a reduction in REM sleep and an increase in the first stage of NREM sleep (NREM1) which represents a transition period between wakefulness and sleep. This observation, together with the high prevalence of prolonged WASO duration, confirms the hypothesis of sleep instability in subjects with SCN8A-related disorders.

Several possible factors may contribute to the poor sleep quality in SCN8A-related disorders. For example, refractory seizures, and multiple ASMs can increase the frequency of sleep disturbance in patients with DEE.¹² However, the majority of patients (79%) were treated with sodium channel blockers. We did not find a significant direct effect of sodium channel blockers on sleep.

On the other hand, different comorbidities, such as autistic features, behavioral problems, and developmental delay, frequently seen in patients with DEE, may also contribute to the high rate of sleep disturbances in this population.⁴³ We did not observe a significant influence of these factors in our cohort, with the exception of the presence of persistent seizures which was associated with a higher percentage of sleep disorders (87% in patients with persistent seizures vs. 29% in seizure-free patients). This suggests that good seizure control can contribute to obtaining a better sleep quality. Melatonin, Clonazepam, and Chloral hydrate were also reported to effectively facilitate sleep initiation and/or maintenance in 89% of our patients, suggesting the importance of undertaking sleep medications in these cases.

Even if this is the first study on sleep in SCN8A-related disorders including a cohort of 47 patients with a detailed description of their sleep, this is a preliminary study, limited by the small number of patients in some of the groups analyzed. Unfortunately, due to the relative rarity of this phenotype, we were able to investigate the sleep features only in seven patients seizure-free and in one patient with SCN8A disorder without epilepsy. Further studies with larger cohorts, such as polysomnographic studies in SCN8A patients with and without epilepsy of different age groups, are needed to better define the sleep pattern in SCN8A-related disorders. Regarding the impact of patients sleep disturbances on caregivers, we did not have the possibility to ask the parents/caregivers to fill in sleep questionnaires; this can be a future step of our research.

Sleep disturbances are a common feature of patients with SCN8A-related disorders. Given the high frequency and impact of sleep disturbances in patients with SCN8A-DEE, it is important to ask specifically about sleep quality and habits. Clinical evaluation, appropriate investigation, and active management are recommended especially for all patients with SCN8A-related disorders who report symptoms of poor sleep. Effective management of sleep disorders and sleep related seizures is likely to improve the quality of life of the patient and the family and has the potential to optimize developmental outcome and improve seizure control.

FF analyzed the data and wrote the manuscript, EG conceived and designed the study, collected and analyzed data, and wrote the manuscript, KMJ CMB, RB, AAS, MM, JF, SM, VDM, JP, PV, GR, GC, KO, RSM collected the data and reviewed the manuscript.

None of the authors has any conflict of interest to disclose.

Patients gave written informed consent. All human and animal studies have been approved by the appropriate ethics committee and have therefore been performed in accordance with the ethical standards laid down in the 1964 Declaration of Helsinki and its later amendments.

We confirm that we have read the Journal's position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.

Written informed consent was obtained from all patients in this study.