Circadian Rhythm Disorders in Childhood


Time of administration in children

If used as chronobiotic, administer 2–3 h before dim light melatonin onset or administer melatonin 3–4 h before actual sleep-onset time

Dosage

Start with a low dose of 0.2–0.5-mg fast-release melatonin; increase by 0.2–0.5 mg every week until effect appears; if there is no response after 1 week, increase dose by 1 mg every week until effect appears. When 1 mg is effective: try lower dose; if there are sleep maintenance problems, start after melatonin treatment; melatonin dose is probably too high

Maximum dose: <40 Kg, 3 mg; >40 Kg, 5 mg

Treatment duration

It should be no less than 1 month. It can be withdrawn just before puberty or shortly after puberty. Stop melatonin treatment once a year for 1 week (preferably in summer) after a normal sleep cycle has been established

When melatonin treatment is no longer effective:

Check timing of administration. In some cases dose reduction is warranted instead of dose escalation, because loss of efficacy of melatonin treatment is most likely caused by slow melatonin metabolism. Metabolism slower, oral contraceptives, cimetidine, fluvoxamine; metabolism faster, carbamazepine, esomeprazole, omeprazole

Reconsider diagnosis: look for neuropsychiatric comorbidity. For very severe delayed sleep-wake rhythm, consider chronotherapy


Reproduced from Bruni et al. [51], with permission of Elsevier



The timing of treatment is crucial because unless therapy is started at the appropriate circadian time, the patients are likely to get worse. Morning exposure to light will facilitate entrainment in humans who have an intrinsic period that exceeds 24 h, whereas evening light exposure will entrain individuals with an intrinsic period that is shorter than 24 h [19]. The treatment options in clinical practice for circadian rhythm sleep disorders comprise bright light treatment and exogenous MLT administration. Although chronotherapy has been used, the data available documenting its efficacy are still insufficient [49]. Chronotherapy, which has proved to some extent successful in DSWPD, consists in delaying the sleep period by 2–3 h every day until the preferred target sleep time is achieved [40]. In order to administer the treatment correctly, it is essential to identify the circadian phase, i.e., the nadir of the core body temperature rhythm or the endogenous MLT rhythm. The core body temperature usually peaks in the late afternoon or evening and reaches its lowest point, i.e., nadir, in the early morning, with sleep normally ending approximately 2 h following the nadir. MLT secretion increases soon after the onset of darkness, peaks in the middle of the night, and gradually falls during the second half of the night [49]. The protocol followed to estimate the nadir of the core body temperature requires that the patient be placed in a semirecumbent position in a laboratory environment for a number of consecutive hours (often 26), with a light intensity of less than 50 lx; the patient receives a 100-kcal meal every hour [49]. The DLMO measurement is based on several samples of MLT (normally with a 30- or 60-min interval) of saliva, urine, or plasma. The level of illumination currently recommended for sampling is 10 lx [50]. While the samples are being collected, subjects should avoid drinks with artificial colorants, alcohol, or caffeine as well as teeth-brushing, lipstick/lip gloss, chewing gum, lemons, and bananas, as well as avoid eating, drinking, or using tobacco for 30 min prior to sampling [46]. In cases in which the sleep phase disorder is severely delayed or advanced, saliva should, if possible, be collected at later or earlier times or hourly for 24 h [51]. A simple way to assess the nadir is to instruct the patient to sleep until he/she wakes up spontaneously (i.e., without an alarm clock) – it is reasonable to assume that the nadir will be approximately 2 h before the subject awakens. The body temperature nadir usually coincides with the moment in which the greatest difficulty in staying awake is encountered, which is worth bearing in mind when the nadir in jet lag disorder and night workers needs to be calculated. Short wavelengths (blue light) have a stronger MLT-suppressing effect and a stronger phase-shifting effect on the human circadian rhythm. Light exposure before the nadir of the core body temperature rhythm causes a phase delay, whereas light administered after the nadir causes a phase advance. Bright light is typically administered by portable units yielding about 10,000 lx, with an exposure time of approximately 30–45 min per day being required to advance sleep, and units yielding about 4000 lx for 2 h being required to delay sleep. The patient is instructed to keep his gaze directed at the light source, but not continuously. Whenever outdoor light of a sufficient intensity is available, it is preferable to be outdoors than sit in front of a light box [49]. Some rare cases of mania as a side effect of phototherapy have been reported [52]. Compliance may be increased in some patients by using a light visor. Blue light-blocking glasses may be useful in adolescents as a countermeasure for alerting effects induced by light exposure through light-emitting diode screens [53], while a prototype light mask using narrow-band “green” light to deliver light through closed eyelids suppresses MLT by 40 % through the closed eyelid without disrupting sleep [54]. In ASWPD, the subject should be exposed to bright light as close to bedtime as possible. The effects of bright light may not be as clear in ASWPD as in DSWPD because exposure to light in the former occurs many hours before the nadir, and waking up the patients in the middle of the night (i.e., just before the nadir) is normally not considered acceptable. In sighted individuals with a free-running disorder, exposure to bright light may be tried before administration of MLT, when the rhythm is in phase with the environment [49].

Exogenously administered MLT has phase-shifting properties, with the effect following a phase response curve (PRC) that is about 12 h out of phase with the PRC of light. MLT administered in the afternoon or early evening will phase advance the circadian rhythm, whereas when it is administered in the morning, it will phase delay the circadian rhythm, with maximal phase shifts occurring when melatonin is scheduled around dusk or dawn. The doses used in most studies range from 0.5 to 5 mg, though it is not clear whether the effects of MLT are dose-related [49]. The recommendations of a European consensus conference held in Rome in 2014 that was aimed at assessing the current role of melatonin in childhood sleep disturbances were recently published [51] (see Table 12.1, modified version of these recommendations). MLT displays its maximum phase-advancing effect 3–5 h before the DLMO, whereas when it is administered 2–3 h after the DLMO, it may have either no effect or a reversal effect in DSWPD [51]. There is no evidence that slow-release MLT is preferable to the fast-acting MLT [51]. Possible side effects include increased blood pressure, headache, dizziness, nausea, and drowsiness [55]. Bright light has been recommended as the first treatment approach to SDWPD; if the response is not satisfactory, MLT is added, usually 12 h before exposure to light [49]. MLT is administered 12 h after the last awakening in blind patients with free-running disorders. In conclusion, the most important practical points to bear in mind when treating CRSWDs are do not start bright light treatment for SDWPD in the early morning but wait until the patient wakes up spontaneously (without an alarm clock); bright light is subsequently administered 1 h earlier every day until entrainment. Appropriate timing of melatonin administration is approximately 12 h after bright light treatment [49].



Subjects with Neurodevelopmental Disorders Have Increased Risk of CRSWDs



Autism Spectrum Disorders (ASD)


Sleep problems are particularly common in children with ASD, with prevalence rates ranging from 50 to 80 % compared with 9–50 % in age-matched, normally developing children. Such problems tend to increase with age, rather than disappear [56, 57]. The most common sleep problem reported by caregivers is insomnia, whose pathogenesis is multifactorial and includes disruption of circadian rhythms and MLT dysregulation. It has recently been hypothesized that the increased use of media combined with the bright screen of the media devices may contribute to the alterations in MLT secretion in ASD [58]. An overresponse to sensory input at bedtime associated with increased precognitive arousal has been observed in such children [59]. Furthermore, the relationship between sleep problems and ASD is complex because circadian abnormalities and epilepsy are both strong, bidirectional contributors [60, 61]. The multifactorial factor that is implicated in ASD may explain the marked night-to-night variability of the sleep-wake cycle and the fragmented irregular sleep-wake patterns, including the inconsistent sleep onset and rise time, free-running sleep-wake rhythm, sleep onset delay, and early morning awakening [62, 63]. Significantly lower levels of nocturnal and daytime blood melatonin levels, as well as lower levels of its primary metabolite 6-sulfoxymelatonin, have been observed in many individuals with ASD when compared with normally developing controls [64, 65]. A disruption of the serotonin pathway, associated with high whole-blood serotonin levels and reduced plasma MLT, was also reported in one study on 278 ASD subjects [61]. Most of the studies that have investigated MLT-related genes in ASD focused on the ASMT gene, with reports indicating that a decreased expression of the ASMT transcript is correlated with decreased MLT blood levels in ASD patients and their relatives [65]. The few studies that analyzed genome-wide gene expression found 15 circadian rhythm regulatory or responsive genes that are differentially expressed in the most severe ASD subgroup though not in the mild or savant subgroups, thereby pointing to a link between circadian rhythm dysregulation and the severity of language impairment. In particular, the presence of the gene encoding AANAT was reported to be significantly reduced [66, 67]. Many polymorphisms within the CYP1A2 gene that alter MLT degradation and are predictive of slow metabolizing alleles have been implicated in the pathogenesis of ASD and sleep problems, accompanied by a loss of efficacy of MLT supplementation after 1 or 2 months [68]. It has been hypothesized that normal nocturnal blood values of MLT in some children with ASD may be caused by a combination of lower MLT production levels and the slow metabolic activity of CZP1A2 [68]. CRSWDs in children with ASD have been investigated above all by means of sleep questionnaires and actigraphic recordings (usually associated with sleep logs), while the detection of plasma, urine, or salivary MLT secretions continues to be used as a diagnostic tool for research purposes [56]. Basic principles of sleep hygiene, including the selection of an appropriate bedtime and establishment of a positive bedtime routine aimed at reducing emotional and/or behavioral stimulation at night and thus at minimizing television viewing and playing computer or video games, may represent an important means of improving sleep [58]. A parent group program of intervention may help to manage insomnia in ASD [69]. Supplemental MLT in ASD may go beyond the treatment of a deficiency state alone; for example, melatonin, which acts as a hypnotic when used independently of a deficiency state (in ASD children with normal endogenous MLT levels), also has antianxiolytic effects that may mitigate hyperarousal-related insomnia [68]. Moreover, some trials and meta-analysis studies in which MLT has been used to treat ASD highlighted the efficacy of MLT in this disorder (with varying dosages of up to 6 mg administered under both immediate- and controlled-release conditions, 5–6 h before the desired bedtime) [70, 71]. In one randomized placebo-controlled study, controlled-release melatonin treatment combined with behavioral interventions in 134 children with autism and long-lasting sleep problems proved to be highly effective [72].


Neurodevelopmental Disabilities (NDD)


The most common sleep disturbances in NDD are delayed sleep onset, frequent awakenings during sleep, lasting minutes or hours, and early morning awakenings. Day-night reversals, advanced sleep onset, and free-running sleep-wake rhythm disorders are much less common. Since individuals affected by NDD cannot sleep when sleep is desired, needed, or expected, most of the sleep disturbances belong to the diagnostic category of CRSWDs [73]. CRSWDs in children with NDD tend to be misunderstood and underdiagnosed. Light/darkness influences brain functions, including cognition, via pathways other than the monosynaptic retinal-hypothalamic tracts [74]. The prevalence of CRSWDs is increasing among normally developing and healthy children owing to a combination of lifestyle changes and increased exposure to light. This change also applies to children with NDD, 70 % of whom are affected by CRSWDs, especially those with moderate-severe NDD, i.e., with bilateral and extensive brain lesions. Persistent early morning awakenings are common in children with NDD and fulfill the criteria of CRSWDs because such children are unable to sleep when sleep is desired, needed, or expected. Delayed sleep-onset disorders in children with severe NDD are associated with a marked variability in the timing of sleep onset and frequency of a delay, with the total sleep time per 24 h not being age appropriate, but frequently reduced [75]. A good response to melatonin administered at bedtime has been reported in such cases [76]. It is important to bear in mind that restless legs syndrome, with or without periodic limb movement, may also cause delayed sleep onset. The prevalence of this syndrome is likely to be underestimated because of the difficulties encountered in diagnosing it in children with NDD [73]. Free-running sleep-wake rhythms with no other sleep disturbances are generally observed in neurologically healthy children who have total ocular blindness, though this condition is rare because total ocular visual loss is now uncommon as a result of improved ophthalmological care. By contrast, children with loss of visual acuity due to occipital lobe visual impairment do not exhibit free-running sleep-wake disturbances because the monosynaptic retinal-hypothalamic pathways to the SCN remain intact. Since children with NDD have altered cortical connectivity, they may not be able to properly perceive the environmental cues required to develop the sleep-wake rhythm, which results in inadequate thalamic signals to the hypothalamus and, ultimately, in CRSWDs [73]. A recent EEG study on children with extensive brain damage and profound developmental disabilities showed that up to 100 % of them had persistent, severely impaired sleep-wake patterns, though very few had ocular lesions, thus demonstrating that the cerebral structures play a major role in sleep-wake regulation [77]. Cerebral palsy (CP) is defined as a group of nonprogressive disorders of movement and posture resulting in activity limitations that occur in the developing fetal or infant brain and affect 1.5–2.5 children per 1000 live births. About 20–50 % of children with CP have a cortical visual impairment, resulting in a free-running circadian rhythm [78]. Smith-Magenis syndrome (SMS) is a rare multisystemic disorder that occurs in 1:25,000 births. It is caused by a mutation or small deletion in a transcriptional regulator gene of the mammalian circadian clock, i.e., RAI1 (retinoic acid induced) on chromosome 17p11.258, and is characterized by intellectual disability of varying degrees, short stature, a deep hoarse voice, obesity, scoliosis, distinctive facies (deep, close-set eyes, midfacial hypoplasia, and broad, square-shaped face), and peripheral neuropathy [79]. Maladaptive and disruptive behavior is typical of this syndrome and is associated with 24-h sleep disorder. Actigraphic data have revealed a total sleep time that is 1 or 2 h shorter than that in normal healthy controls and fragmented sleep starting from as early as 6 months and persisting throughout school age [80]. Many studies have found that an alteration in the circadian clock gene action results in inverted endogenous melatonin secretion, which thus peaks during daytime, in the vast majority of children with SMS [81]. Oral acebutolol administered alone in the early morning and combined with MLT in the evening has been used in an attempt to improve the sleep-wake rhythm in patients with SMS [82]. Angelman syndrome (AS) is a neurodevelopmental disorder that is characterized by mental retardation, seizures, gait ataxia, speech impairment, epilepsy, craniofacial abnormalities, and easily provoked laughter and is due to abnormalities in chromosome 15q11–q13. The prevalence of sleep problems associated with this syndrome is usually very high, with up to 90 % subjects being affected [83]. Sleep problems mainly consist of difficulties in falling asleep and multiple awakenings. A lower level of MLT and a high prevalence of CRSWDs (irregular, free-running, and sleep phase delayed disorders) have been reported in AS, though open-label and placebo-controlled trials have shown that treatment with MLT supplementation yields a good response [8486]. A possible explanation for this positive response is the lack of ubiquitin protein ligase E3A gene expression on the maternal chromosome 15q11–q13, which is reported in AS and is known to be implicated in the control and development of circadian rhythm [87]. This hypothesis is supported by another study in which a patient with Rett syndrome, an abnormality that affects the ubiquitin protein ligase E3A gene, N24SWD, markedly improved following MLT oral supplementation [88].

Although genetic and/or epigenetic abnormalities in sleep-wake circadian regulation may predispose children with NDD to CRSWDs, poor sleep hygiene, negative associations, and the lack of restrictions all contribute to the maintenance of sleep problems. The active collaboration of caregivers is essential to be able to adopt behavioral treatment strategies, such as creating a dark, quiet, non-stimulating environment and reducing the number of stimuli (such as electronic devices) [71]. A recent large clinical trial confirmed the efficacy of MLT as a means of treating sleep problems in children with NDDs, using doses ranging from 0.5 to 12 mg, which were found to reduce sleep latency and increase total sleep time [89].


Attention-Deficit Hyperactivity Disorders (ADHD)


About 25–50 % of children and adolescents with attention-deficit hyperactivity disorder (ADHD) experience sleep problems, with objective data based on actigraphic recordings demonstrating an increase in sleep-onset latency associated with a decreased amount of time spent asleep in such subjects [90]. According to the data in the literature, five sleep phenotypes may be identified in ADHD: a sleep phenotype characterized mainly by a hypoarousal state, resembling narcolepsy, which may be considered a “primary” form of ADHD; a second phenotype associated with a delayed sleep-onset latency and a higher risk of bipolar disorder; a third phenotype associated with sleep disordered breathing; a fourth phenotype related to restless legs syndrome and/or periodic limb movements, which may further extend the delays in the sleep phase disorder; and, lastly, a fifth phenotype related to epilepsy/or EEG interictal discharges [90]. We will discuss the second phenotype here. Sleep-onset delayed insomnia is the most common sleep disorder in children with ADHD. The onset of sleep delayed phase disorders may occur at as early as 3 years of age, with an accumulation of sleep deprivation over time. Children in this subgroup have a delayed DLMO associated with a significant delay in sleep latency when compared with ADHD children without insomnia [91]. Preliminary evidence from severe mood dysregulation-related disorders indicates that morning light therapy has a positive effect on depressive symptoms, circadian rhythms, inattention, and irritability [92]. It has been suggested that the core endophenotypic characteristic of pediatric bipolar sleep is a phase delayed circadian sleep-wake cycle rather than a reduced need for sleep per se (see below) [93]. Many studies have demonstrated the efficacy and safety of MLT in the treatment of insomnia in children with ADHD, with doses ranging between 3 and 6 mg [94]. MLT may, if required, be combined with light therapy, particularly in children that are at risk of developing bipolar disorder as the use of stimulants remains controversial in such subjects [90]. Stimulant medication does not appear to affect the core symptoms related to a lower vigilance state in children with sleep delayed insomnia. Furthermore, the use of stimulant medications may exacerbate insomnia in children with ADHD, thereby affecting circadian motor activity levels, as has been demonstrated by actigraphic analyses [95]. The authors of this review believe that the administration of long-acting medications may increase the risk of developing or worsening sleep-onset insomnia in children with ADHD [95]. A large placebo-controlled trial on ADHD studied the effects of 4-week MLT therapy on the sleep-onset latency and circadian phase, as assessed by means of the DLMO [96]. The results of that trial did not detect any improvement in ADHD symptoms or cognition at the end of the 4 weeks [96]. A follow-up study revealed that improvements in behavior and mood after long-term treatment (2–3 years) only occurred in those children still using melatonin, while discontinuation of MLT resulted in a relapse of sleep-onset insomnia [97]. In one study on adult ADHD patients [98], treatment with early morning bright light therapy improved ADHD symptoms after 3 weeks, with the positive effects appearing to occur more rapidly than following administration of MLT. Interestingly, the effects of both sensorimotor rhythm (SMR) and slow cortical potential (SCP) neurofeedback treatment of ADHD symptoms last longer than those induced by medication, possibly because they act by increasing sleep spindle density and normalizing sleep-onset insomnia, thereby resulting in vigilance stabilization. Although neurofeedback does not target the circadian phase delay directly, this effect is mediated by subcortical and cortical circuits that regulate sleep spindle production and sleep onset [99].


Mood Disorders


In view of the significant changes in sleep and circadian rhythms that occur during a person’s lifespan, age may contribute to the heterogeneity in sleep-wake profiles linked to mood disorders. The severity of depressive symptoms is expected to be associated with a more pronounced phase delay during youth and later diagnosis of bipolar disorder, while a reduced sleep duration and consolidation and disorganization of circadian rhythms is expected in older age [100]. Indeed, it has been suggested that the restoration of normal circadian rhythms contributes to the remission of depression and prevention of relapses in young people with depressive symptoms. Actigraphic monitoring has been used to show that poor sleep is a hallmark of major depression during a stable depressive phase among young people (13–35 years old) [101]. A reduced need for sleep, together with elation, grandiosity, and racing thoughts, distinguishes mania and bipolar disorder from attention-deficit hyperactivity and other childhood psychiatric disorders. Children with manic bipolar I disorder typically experience a decreased need for sleep resembling that of adults, whereas many children who are bipolar, who exhibit part-day manic episodes (pediatric bipolar type IIA and type IIB), or who have chronic mixed conditions (pediatric bipolar type IIIA) do not [93]. Children with bipolar type IIA exhibit prominent diurnal cycles on most days (pediatric bipolar type IIA): initial morning depression and subsequent (typically late afternoon and/or evening) mania. They display disturbed sleep patterns, characterized by an evening acceleration and a significant delay in sleep onset, which may, or may not, be accompanied by a decreased need for sleep and difficulty in awakening for school; moreover, a decreased need for sleep has been observed in subjects with manic cycles lasting days (pediatric bipolar type I) or chronic mania [93]. It has been suggested that the main bipolar sleep defect is a heritable phase delay in the sleep-wake cycle resulting from mutations in SCN circadian clock genes, which interact with, but are independent of, evening or ongoing manic psychomotor accelerations [93]. Several clock genes, such as CRY1 and NPAS2, have been associated with affective disorders, with CLOCK and VIP being specifically linked to the mania-hypomania phenotype [102]. This hypothesis predicts (i) that most bipolar children and adolescents, whose afternoon and/or evening manic acceleration typically terminates overnight, with ultradian cycling (pediatric bipolar types IIA and IIB), will display delayed sleep onset but a low prevalence of decreased need for sleep; (ii) that the intrinsic sleep-onset phase delay, when coupled with bedtime and early morning manic psychomotor acceleration (hedonic or dysphoric), reduces the need for sleep; and (iii) that the reduced need for sleep is greatest among individuals whose manic cycles last longer than 1 day (pediatric bipolar type I) or among those with chronic mania (pediatric bipolar type IIIA). An increase in tobacco use was recently found among depressed young people with a delayed sleep phase and short sleep duration [103]. To sum up, sleep-onset phase delays and delayed sleep phase syndromes that occur during euthymic or depressed states may be trait markers of bipolar spectrum illness [93].


Case Reports



Case Report 1: SDWPD


A female adolescent aged 17.5 years came to the sleep disorder outpatient service of the Neurocenter of Italian Switzerland because she had been suffering from sleep-onset difficulties since she was 3 years old. Her bedtime had become increasingly delayed in the last 3 years (sleep time: from 01.30 am to 7 am), and she complained of excessive daytime sleepiness, requiring a 3-h nap, and of school difficulties. She also reported fear and strong nausea at bedtime and usually fell asleep at around 5 am. At weekends she tended to sleep for more than 12 h. She suffered from lypothymic attacks and dizziness during daytime, which made her feel more irritable. DSWPD was confirmed by means of the MLT salivary test (DLMO after 00.30 am) and actigraphic recording (see Fig. 12.1). She was placed on therapy with long-acting MLT 2 mg at 8 pm, which led to the complete disappearance of the anxiety, lipothymic attacks, and dizziness; restored healthy sleep, from 9:30 pm to 9 am; and eliminated the diurnal hypersomnolence or napping. She now feels happy.

A331183_1_En_12_Fig1_HTML.gif


Fig. 12.1
Actigraphic recording of case report 2 (weekend fourth and fifth days)


Case Report 2: Sleep-Onset Insomnia (First Suspected Diagnosis of SDWPD)


A male 14-year-old adolescent came to the sleep disorder outpatient service of the Neurocenter of Italian Switzerland because he had been suffering from sleep-onset difficulties for many years. His sleep problems had got worse in the last months after he had started therapy with long-acting methylphenidate following a diagnosis of ADHD, learning disabilities, anxiety disorder, and suspected mood disorders (pediatric bipolar disorders type IIA). His anxiety increased in the afternoon and evening after the onset of treatment. He reported that his bedtime was 10.00 pm. Sleep onset occurred 30 min later, though sometimes even after midnight, and he woke up at 7.00 am. He said he had difficulties in waking up in the morning. He had a history of motor tics, tonsillitis, and respiratory allergy, as well as familiarity for somnambulism and ADHD. A video-polysomnographic recoding revealed a very mild mixed-sleep apnea disorder, some periodic leg movements during sleep, and increased sleep latency. A 1-week actigraphic recording demonstrated a tendency to fall asleep late (around 23:00 pm, with a mean sleep latency of one and a half hours), with no sleep fragmentations or diurnal nap (see Fig. 12.2). Blood examinations revealed a ferritin level of 54 mcgr/l. He was placed on therapy with long-acting MLT at 19.30, 2 mg, for 3 months, though with no benefit. The final diagnosis was sleep-onset insomnia with anxiety disorder exacerbated by stimulant therapy, which was subsequently replaced by a more appropriate therapy containing a mood stabilizer.
Aug 15, 2017 | Posted by in NEUROLOGY | Comments Off on Circadian Rhythm Disorders in Childhood

Full access? Get Clinical Tree

Get Clinical Tree app for offline access