Michael Kohrman

Introduction

Excessive daytime sleepiness in the adolescent is epidemic in our society. In the adolescent population, hypersomnolence manifests primarily as excessive daytime sleepiness. In part, symptoms are the result of school and lifestyle pressures. This is in combination with changes in circadian phase and total sleep time. In contrast, excessive daytime sleepiness manifesting as hypersomnolence is uncommon in young children. The primary manifestation of sleepiness in young children is hyperactivity. Young children respond to the symptom of sleepiness by increasing their motor activity to maintain alertness. However, even in this group, careful questioning does reveal symptoms of daytime sleepiness. The clinical definition of idiopathic hypersomnolence in childhood must take into account this difference in the clinical response of children to the same ‘sleep pressure’ manifested as excessive daytime sleepiness in adults.

Idiopathic hypersomnia (IH) is characterized by excessive daytime sleepiness and normal or long nocturnal sleep times, along with frequent and often prolonged naps. Both major sleep period and naps are typically unrefreshing.¹ PSG and MSLT demonstrate the absence of frequent periods of sleep onset with REM (SOREMP). The second edition of The International Classification of Sleep Disorders: Diagnostic and Coding Manual divides IH based on nocturnal sleep time: idiopathic hypersomnia with long sleep time (ICD-10-CM G47.11) and idiopathic hypersomnia without long sleep time (ICD-10-CM G47.11). Table 19.1 contrasts the clinical criteria between these two diagnoses and narcolepsy without cataplexy (ICD-10-CM G47.419).¹ The key objective difference between narcolepsy and IH is SOREMP in two or more MSLT naps associated with narcolepsy.

Both disorders of IH are associated with difficult awakenings, sleep drunkenness, and unrefreshing primary sleep periods/naps.1,2 Guilleminault and Pelayo have divided idiopathic hypersomnia into three types based on etiology: those patients with a positive family history and positive HLA Cw2 antigen, a second type with history of viral infection, and a third group of those patients not included in the first two groups.³ Bassetti and Aldrich divided their patients into three groups based on clinical symptoms.² The ‘classic idiopathic hypersomnia’ group tended to have sleepiness that was not overwhelming, take unrefreshing naps of up to 4 hours, have prolonged nighttime sleep, and difficulty awakening. The second group they termed a ‘narcoleptic type’ presented with overwhelming excessive daytime sleepiness, took short refreshing naps, and awakens without sleepiness. Their third group was a ‘mixed group’ with features of both syndromes.

Idiopathic hypersomnolence is a diagnosis of exclusion. A complete evaluation of other causes of hypersomnolence must be undertaken. The differential diagnosis of hypersomnolence is quite broad, ranging from mood disorders, to narcolepsy, to circadian rhythm disturbances or primary sleep disorders that cause excessive daytime sleepiness. Table 19.1 provides a framework based on etiology for further discussion of the differential diagnosis below.

Epidemiology

The incidence of diagnosis at major sleep centers varies from approximately 10% to 60% that of narcolepsy.4,5 Anderson et al., in their series of 77 patients, noted that hypersomnia began at an age of 16.6 ± 9.4 years. Approximately two thirds of patients developed symptoms before the age of 18. The mean age of diagnosis was 30 years.⁴ In addition, Anderson et al. found 34% had a family history of similar symptoms, and, in eight cases, more than one family member was affected. Ali et al. in their series of 85 patients, found a familial incidence of 58%.⁶

Anderson found that 18% of IH patients were positive for the HLA DQB1*0602 antigen compared to 98% of patients with narcolepsy.⁴ The Cw2 antigen was observed in 10% of IH patients. A case report of three adolescent-onset cases of IH in a two-generation family raises the question of autosomal dominant inheritance.⁷

Pathophysiology

The underlying cause of the sleepiness in IH continues to be an active area of inquiry. Kanbayashi et al.⁸ examined CSF levels of histamine in patients with IH, narcolepsy, OSA and normal neurologic controls. They found that low CSF histamine levels were mostly observed in non-medicated patients, and significant reductions compared to controls in histamine levels were observed only in non-medicated patients with hypocretin deficiency and with IH. In contrast, the levels in the medicated subjects with hypocretin deficiency and with IH did not differ significantly from those in control subjects. The authors also note that hypocretin-1 levels did not differ between medicated and non-medicated subjects in any of the hypersomnia categories.

Scammell and Mochizuki⁹ note in an editorial in Sleep that reduced histamine signaling is an attractive explanation for the sleepiness of idiopathic hypersomnia. Parallel behavior between individuals with IH and mice with central histamine knockout (humans have great difficulty rousing from sleep in the morning, and mice lacking histamine exhibit a reduction in wake at the beginning of the usual active period).¹⁰ Histamine H₃ receptors are inhibitory, exclusively located in the CNS.⁸ H₃ receptor antagonists enhance central histaminergic neurotransmission specifically and are reported to enhance alertness in mouse models of narcolepsy. Kanbayashi’s finding that low CSF histamine levels are observed in hypersomnia of central origin, regardless of hypocretin-1 status suggests a wider role for this mechanism in the treatment of hypersomnia.⁸

Evaluation of Sleepiness in Children

Objective evaluation of sleepiness begins with a complete sleep history and physical examination to eliminate other causes of excessive daytime sleepiness (see Box 19-1). Sleep logs should be obtained to document sleep times, and actigraphy may be utilized to confirm the sleep log data. On the night prior to laboratory testing for daytime sleepiness, a polysomnogram should be obtained to confirm the absence of a primary sleep disorder. Hypersomnolence can be characterized quantitatively via the multiple sleep latency test (MSLT)¹¹ or maintenance of wakefulness test (MWT).¹² Both tests measure the time to fall asleep during the daytime. The MSLT requires the patient to fall asleep while the MWT requires the patient to stay awake. The mean adult sleep latency is 18.7 minutes. It is generally accepted that in the adult population the pathologic average sleep latency is 8 minutes or less as the result of the MSLT with an average sleep latency of 13.4 minutes.¹³ Results from the MSLT and MWT often differ. Bonnet and Arand suggest that the MWT measures the sleep propensity and also the arousal system secondary to motivation and posture, while the MSLT is done supine and only measures sleep propensity.¹⁴ Carskadon and Dement have demonstrated that sleep latency in children is age-dependent.¹⁵ Using Tanner stages to group children during adolescence, they observed a drop in average sleep latency on the MSLT of 20% between Tanner stages 1 and 3. This decreased sleep latency persisted in later adolescence (Tanner 4 and 5). In the same subjects, total sleep time and total REM time remained constant but total slow wave sleep time deceased by 70%. Mean sleep latency of Tanner 1 and 2 children (mean age 12 years sleeping their habitual average of 9.1 h) is 18 min. Compared to the college student (mean age 19 y, habitual sleep 7.1 h) of less than 5.5 minutes. The definition of pathological sleepiness in children must account for this age difference.

A number of pediatric sleepiness scales have been developed over the past 10 years. Lewandowski et al. reviewed 21 measures and found that only six met ‘well-established’ evidence-based assessment criteria.¹⁶ Of these, only two: the Pediatric Daytime Sleepiness Scale (PDSS)¹⁷ and the Sleep Disturbance Scale for Children (SDSC),¹⁸ assessed EDS.

Clinical Presentation

Bassetti and Aldrich reviewed 42 cases of idiopathic hypersomnia.² Onset of hypersomnolence was a mean age of 19 ± 8 years (range 6–43). Onset was associated with insomnia in five, weight gain in two, viral illness in four, and minor head trauma in three. Forty-five percent of patients snored. Sixty percent took one or more involuntary naps during the day. Over half the subjects took naps of 30 minutes or longer, and ¾ reported the naps were unrefreshing. Over half of the patients had psychiatric problems. Polysomnographic recordings demonstrated short sleep latency of 6.6 ± 5.7 min. Mean latency on the MSLT was 4.3 ± 2.1 minutes. In 5 of 12 patients who underwent esophageal pressure monitoring were found to have upper airway resistance syndrome. None of these patients reported improved sleep with CPAP.

In their series of 77 patients with IH, Ali et al.⁶ found that hypersomnia began at a mean ± SD age of 16.6 ± 9.4 years (range 0–46 years). Symptoms began before 18 years of age in 49 of 77 patients. Some patients and their families believed that the symptoms had been present from the first year of life, and, in 7 cases, worsening occurred over a period of several years. The mean age of diagnosis in this group was 30 years.⁶ Precipitants were described by three subjects who reported a transient viral illness at the time of symptom onset, and one subject reported onset of symptoms over a day. The mean ESS score at initial presentation prior to treatment was 16.3 ± 3.3 (range 11–24). In their study, Ali et al. noted that daytime somnolence interfered with work and social activities. The mean length of nighttime sleep reported was 9.2 ± 1.8 hours, and this did not correlate with EES score.⁶ There was no significant difference in MSLT between those with a long sleep time and those with a sleep time shorter than 10 hours. Seventy-six patients took nighttime sleep and sleep drunkenness was not observed.⁶

Typically, symptoms begin during childhood including prolonged nighttime sleep, and awakening difficulties often precede the onset of daytime sleepiness. Roth reported continuous non-imperative sleepiness prolonged unrefreshing naps without dreaming and difficult arousal.¹⁹ The risk of automobile accident or near-miss event was reported in 50% of 50 adult patients with IH over a 5-year period in Japan.¹⁰

Polysomnographic studies from Baseetti and Aldrich’s 42 patients demonstrated a sleep efficiency of 93%, mean of 20 total awakenings greater than 1 minute, 8% slow wave sleep, 18% REM sleep, automatic behaviors in 61%, and sleep paralysis in 40%.² Hours of sleep per day were 8.4 ± 1.9 and time to activity in the morning 42 min. Forza et al. studied 10 patients with IH.²⁰ All had onset before age 21. There was no statistical difference between IH patients and controls for TST or sleep latency. Mean sleep latency was 9.1 minutes. IH patients demonstrated decreased SWS and increased REM sleep percent with a sleep latency of 5.  ± 0.7 min.

Anderson et al.,⁴ in their study of 77 IH patients, found a mean sleep latency of 11.5 minutes ± 8.2, increased mean slow wave sleep of 22.9% ± 8.7, and mean sleep efficiency of 94.3%. Ten patients had a sleep efficiency of less than 89%, and none. REM sleep latency and percentages of light sleep and REM sleep were normal. Results from MSLT in this group of IH patients demonstrated that a mean sleep latency in patients with idiopathic hypersomnia was 8.3 ± 3.1 minutes, versus the narcolepsy group (4.1 ± 2.6; P <0.001). Comparison of the IH grouped by MSLT with sleep latency ≥8 min or <8 min demonstrated no differences between these two groups in the duration of light sleep, slow-wave sleep, REM sleep, sleep latency, sleep efficiency, REM sleep latency, body mass index, or untreated ESS.

Ali et al.⁶ found the median nocturnal sleep latency was similar for males and females, but on MSLT, the mean sleep latency for males was lower than that for females. Thirteen patients had one sleep-onset REM period (SOREMP) on either overnight polysomnography or MSLT, but none had two or more SOREMPs.

In a study of 75 patients with idiopathic hypersomnia with and without long sleep time, Vernet and Arnulf²¹ found that hypersomniacs had more fatigue, higher anxiety and depression scores, and more frequent hypnagogic hallucinations (24%), sleep paralysis (28%), sleep drunkenness (36%), and unrefreshing naps (46%) than controls. DQB1*0602 genotype was similar observed in 19% of controls and 24% of IH patients. Comparing the IH patients with long versus normal sleep time, those with long sleep time were younger, slimmer and more evening type characteristics and higher sleep efficiencies than those without long sleep time. MSLT latencies were normal (>8 min) in 71% of IH patients with long sleep time.²¹