Clinical manifestations

Presentation and course

The onset of idiopathic hypersomnia is often insidious and occurs in the second decade of life with a key feature of near constant daytime sleepiness and complaints that the patient almost never feels alert. Along with this background of excessive sleepiness, patients often have long episodes of daytime sleep, with naps lasting over 1 hour. Compared to the daytime sleep episodes that occur in patients with narcolepsy, those associated with idiopathic hypersomnia tend to be more resistible, longer in duration, and not as refreshing. However, this difference from narcolepsy is not absolute and some patients with idiopathic hypersomnia have irresistible daytime sleep episodes and refreshing naps (02; 15). The most important clinical distinction is the absence of cataplexy in patients with idiopathic hypersomnia.

Night sleep in idiopathic hypersomnia is typically long and consists of an average of 10 hours or more (up to 19 hours) (136). The nocturnal sleep is generally undisturbed in contrast to patients with narcolepsy, a condition in which fragmented nocturnal sleep is common (76). Nocturnal sleep is usually consolidated with sleep efficiencies exceeding 85%, short sleep-onset latency, higher proportion of REM sleep (commensurate with the prolonged sleep durations), and a lower proportion of N3 sleep (101; 76). Awakening is often difficult, even with the use of sophisticated alarm clocks, and patients may describe morning disorientation in time and space, slowing of thought and speech, and inappropriate “automatic” behaviors lasting from several minutes to an hour or more, which is a state referred to as “sleep drunkenness”. The presence of sleep drunkenness has been reported in 36% to 52% of patients with idiopathic hypersomnia (08; 133). Of significance, patients may report hypnagogic hallucinations and sleep paralysis, suggesting that these symptoms are not specific to narcolepsy (02; 15). An interesting association in some patients with idiopathic hypersomnia is the presence of what appears to be autonomic nervous system dysregulation, which includes orthostatic hypotension, headaches, perceived temperature dysregulation, and Raynaud-type phenomena (06). Like narcolepsy, idiopathic hypersomnia is often associated with depressive symptoms (35).

According to a retrospective study, idiopathic hypersomnia has a prevalence of 2 to 5 per 100,000, which is just 40% of the clinical prevalence of narcolepsy (15; 19; 08). However, in the absence of systematic studies, the prevalence of idiopathic hypersomnia remains to be determined. Features that distinguished idiopathic hypersomnia from a comparable group of 63 narcoleptics were prolonged and nonrefreshing daytime naps, a positive family history of daytime hypersomnolence, increased slow-wave sleep, and longer sleep latency on the multiple sleep latency test (08). Patients with idiopathic hypersomnia have a reduced quality of life as evidenced by trouble maintaining employment and academic schedules and increased risk for divorce (95).

Patients with idiopathic hypersomnia have 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 (133). In addition, they depend on others to awaken them, have mental fatigability, and have difficulty maintaining alertness during the day (134).

Prognosis and complications

Idiopathic hypersomnia is usually a chronic, persistent disorder (15; 125). However, a study suggested that symptom remission rates of nearly 33% were noted on 5-year follow up (58), which is in accordance with the results of repeated MSLT assessments in this patient population (130; 73). The complications of idiopathic hypersomnia are mostly related to impairment in daytime functioning due to the excessive sleepiness with significant impact on social, professional, and employment-related activities (125). The greatest mortality risk with idiopathic hypersomnia remains falling asleep while driving or fatigue-associated driving impairment (06).

Clinical vignette

A 23-year-old woman presented with a 2-year history of excessive daytime sleepiness. She reported that she “never” felt refreshed and had significant difficulty awakening from sleep. This caused difficulty with her ability to finish college. Her usual bedtime was 10 PM, and she was awoken by her mother by 7 AM. If not awoken by her parent, she would sleep until 10 AM or longer. She reported having a regular sleep schedule and no difficulty with sleep onset or sleeping through the night. After she was awoken, she reported feeling excessively sleepy, including feeling “sleep drunk,” which typically lasted 2 to 3 hours. She denied any nightmares, restless legs symptoms, cataplexy, hypnagogic hallucinations, or sleep paralysis. She drank six to eight double espressos in the morning and coffee throughout the day without a favorable effect. The patient was unable to keep a job due to the sleepiness. She had been exposed to hexane and had developed a neuropathy that had been improving. Her family history was unremarkable, and review of systems was normal.

With a weight of 142 pounds and a height of 5 feet 2 inches, her BMI was calculated to be 26 kg/m². Her vital signs revealed an arterial blood pressure of 113/64 mmHg, temperature of 98.0° F, pulse of 84 beats per minute, respiration rate of 16/min. On physical exam, she was awake and appropriate without neurologic deficit or other abnormalities noted.

An MRI of the head was unremarkable. Laboratory testing showed normal values for electrolytes, renal, hepatic, and thyroid function tests and hemoglobin A1c. She had 2 weeks of actigraphy preceding an overnight polysomnogram (PSG), which demonstrated an average daily sleep duration of 9.5 hours. The overnight polysomnogram showed a sleep onset latency of 13 minutes and a REM sleep latency (from the first 30-second epoch scored as sleep) of 43 minutes. Despite a total sleep time of 9 hours and 37 minutes, the sleep efficiency (total sleep time divided by time in bed) was 92%. She snored lightly and had an apnea-hypopnea-index of three per hour. The five-nap multiple sleep latency test revealed a mean sleep latency of 5.5 minutes without sleep-onset REM periods (SOREMPs) on any of the five naps.

She was diagnosed with idiopathic hypersomnia and was started on a low-sodium oxybate. Because she had undisturbed nocturnal sleep, she was titrated up 6 grams at bedtime (instead of the twice nightly dosing frequently used in narcolepsy), resulting in progressive improvement in her idiopathic hypersomnia-related symptoms. Once her excessive daytime sleepiness improved, she was able to complete school and find a new job.

Biological basis

Etiology and pathogenesis

The exact cause of idiopathic hypersomnia remains unknown. There are limited data regarding the neurobiology and pathogenesis/pathophysiology of idiopathic hypersomnia, and there is no existing animal model for more detailed study. Many neurochemical studies regarding the disorder have been inconclusive. In a study comparing patients with idiopathic hypersomnia and narcolepsy type 1, using sleep diary and self-report measures, two subgroups emerged: one with family history and one with a postinfectious onset (23). These findings suggest the possibility of gene-environment interactions that can contribute to disease in a fashion similar to narcolepsy with hypocretin deficiency.

Neurotransmitters. From a structural perspective, the destruction of noradrenergic neurons of the rostral third of the locus coeruleus complex or of the noradrenergic bundle at the level of the isthmus in the cat leads to hypersomnia with a proportional increase of non-REM sleep and REM sleep, resembling idiopathic hypersomnia (100). This state is accompanied by a decrease of diencephalic norepinephrine (100). These findings are concordant with the observations of von Economo during the encephalitis lethargica epidemic, which highlighted that patients suffering from excessive sleepiness often had lesions at the junction of the posterior hypothalamus and midbrain (137). Additionally, therapeutic measures for idiopathic hypersomnia, such as stimulants and modafinil/armodafinil, typically act on catecholamine (dopamine, norepinephrine) signaling mechanisms, indicating that there may be some contribution of these neurochemical pathways in the pathogenesis of idiopathic hypersomnia. Other aspects that have been evaluated as possibly contributing to the pathogenesis of idiopathic hypersomnia are histamine signaling, melatonin secretion abnormalities, immunologic and inflammatory processes, somnogens, and genetic factors.

In the 1980s, studies in patients with less well-characterized clinical idiopathic hypersomnia found some differences in monoamine metabolites and suggested that idiopathic hypersomnia may have a component of disrupted dopamine signaling. One study found decreased dopamine and indoleacetic acid (tryptophan metabolites) compared to controls in the CSF of patients with narcolepsy and idiopathic hypersomnia (86). However, an analysis of the CSF levels of 11 biogenic amines (including the monoaminergic neurotransmitters and their metabolites) and five trace amines among 39 hypocretin-deficient patients with type 1 narcolepsy, 31 patients with objective sleepiness non-orexin-deficient (type 2 narcolepsy and idiopathic hypersomnia), and 24 patients without objective sleepiness demonstrated no significant differences among the three groups (13).

Histamine has also been suggested as a biomarker in the pathophysiology of idiopathic hypersomnia, though the evidence is inconclusive (108). Histamine is a wakefulness producing neurotransmitter. Antihistamines (H1 antagonists) are well known to increase somnolence across many species including humans. Two studies evaluated CSF histamine in patients with hypersomnia of various origins. One small, uncontrolled study looked at CSF histamine in patients with multiple causes of excessive daytime sleepiness and found an inverse correlation of CSF histamine with a severity of reported sleepiness based on the Epworth Sleepiness Scale (16). Kanbayashi and colleagues found that CSF histamine was reduced (compared to controls) in patients with narcolepsy (type 1 and type 2), as well as idiopathic hypersomnia, but not in patients with obstructive sleep apnea (57). However, a larger study looking at histamine, t-methylhistamine, and hypocretin in various conditions that lead to hypersomnia failed to show differences in CSF histamine or t-methylhistamine levels in patients with idiopathic hypersomnia compared to controls (30; 13). From a more clinical perspective, histaminergic compounds like pitolisant, an inverse H3 agonist, have been useful in the management of drug resistant conditions associated with hypersomnia, including idiopathic hypersomnia, through their resultant increases in brain histamine and other monoamine levels (67).

Investigators have reported on a potential 500-to-3000-dalton somnogen that potentiates γ-hydroxybutyrate (GABA) through activity at the benzodiazepine site on the α2 subunit of the GABAA receptor (108). However, attempts to replicate this finding in well-curated idiopathic hypersomnia cohorts have been unsuccessful (33). Nonetheless, blockade of the activity with a benzodiazepine antagonist, flumazenil, has yielded modest results in a small proportion of a cohort of mixed hypersomnia patients, suggesting that such a mechanism may play a role in certain individuals with otherwise unexplainable hypersomnolence (129).

Although hypocretin-1 is implicated in narcolepsy type 1, this neurotransmitter does not appear to be pathogenic in idiopathic hypersomnia (30).

Inflammatory cytokines. Some studies indicate that idiopathic hypersomnia may have an immunological or inflammatory component, similar to narcolepsy type 1. Among patients with idiopathic hypersomnia, human leukocyte antigen-Cw2 was found in 22.2% compared to 5.7% in controls (p < 0.05) (86). Also, HLA-DR5, which is in linkage equilibrium with HLA-Cw2, was present in 38.8% of patients versus 14.6% of controls (86). Other studies, however, did not find any similar association with idiopathic hypersomnia and HLA haplotypes (15; 20). However, the poor associations with HLA regions may be explained by the variability of subjects exposed to a triggering factor (eg, infection, as in narcolepsy type 1) (72; 34), as well as the fact that idiopathic hypersomnia represents a heterogenous population. Nonetheless, additional data that indicate there may be an immunological component are the serum total IgG levels in 28 Japanese patients with idiopathic hypersomnia and long sleep times. Those patients with idiopathic hypersomnia had high IgG3 and IgG4 levels, low IgG2 level, and IgG1/IgG2 imbalance, which was different than healthy individuals (121).

Related to the immune-mediated-disease hypothesis is the observation that proinflammatory cytokines interleukin-6 (IL-6) and tumor necrosis factor-alpha (TNFα) are elevated in disorders associated with excessive daytime sleepiness, such as sleep apnea, narcolepsy, and idiopathic hypersomnia. Although these cytokines are part of the innate immune system, their activity meets the criteria of a somnogen/sleep regulatory substance (60). Most notably, sleep deprivation also leads to sleepiness and daytime hypersecretion of IL-6. These findings suggest that IL-6 and TNFα are possible mediators of excessive daytime sleepiness in humans (135). Prostaglandin D2, an inflammatory molecule, may be another possible contender, given the finding that serum and CSF levels of lipocalin-type prostaglandin D synthase (LPGDS) were noted to be higher in NT1, NT2, and idiopathic hypersomnia patients, but this finding needs replication (139).

Genetics. A genetic basis for some cases of idiopathic hypersomnia is suggested by several investigations. Early studies indicated that a large minority of patients with idiopathic hypersomnia had a similarly affected first degree relative (91; 20). Although these data strongly suggest a genetic component, possibly autosomal dominant, the limited number of families studied has not permitted definitive determination of the mode of inheritance. A number of studies have sought to explore the genetic basis of sleep duration using genome-wide association studies. Unfortunately, sample size limitations and phenotype heterogeneity have limited the findings thus far. To date, the largest analysis of genetic contributions to sleep duration has come from a study of individuals in the UK Biobank (http://www.ukbiobank.ac.uk) (56). The findings confirmed a prior association (47) with the thyroid-specific transcription factor, PAX8, and identified a novel association with a multicascade serine/threonine kinase, VRK2 (04). However, the variants associated with the identified genes conferred insubstantial effects on sleep duration (2.6 min per PAX8 allele and 1.6-2.0 min per VRK2 associated variant), highlighting the difficulties in identifying the genetic contributors to such multifactorial phenotypes. Future study of extreme, well-phenotyped individuals and multiplex families may reveal more meaningful associations.

Additionally, circadian rhythm gene-expression analyses have revealed some abnormalities in individuals with idiopathic hypersomnia; however, causality has yet to be established (64). In 2014, a study by Lippert and colleagues evaluated gene expression of certain circadian clock genes in cultured dermal fibroblasts of patients with a diagnosis of idiopathic hypersomnia compared to those from normal healthy controls (70). This study revealed statistically significant reductions in the amplitude of the expected oscillation in the transcription of BMAL1, PER1, and PER2 (70). The authors postulated that the dampening of oscillations of these transcriptional factors in the SCN, which modulate pineal melatonin secretion, may explain the decreased melatonin secretion seen in idiopathic hypersomnia (122; 89). Another study by Materna and colleagues demonstrated an approximately 0.82 hour (95%, CI 0.44 to 1.20 hours) longer mean period length in patients with idiopathic hypersomnia (relative to controls), based on a BMAL1 promoter activity assay in fibroblasts (79). In concordance with this is the finding that melatonin and cortisol secretion are significantly delayed in individuals with idiopathic hypersomnia (89); these findings may suggest a genetically based abnormality in entrainment as a contributor to the pathophysiology of idiopathic hypersomnia.

In sum, despite several intriguing findings spanning the domains of CNS neurotransmitters, immunology, and genetics, the pathophysiology underlying idiopathic hypersomnia remains nebulous. A number of factors contribute to this elusive disorder, not the least of which include the poor objective characterization of this phenotype with the currently available diagnostic methods (see management section below), the genetic complexity, the unclear risk factors, and environmental exposures.

Epidemiology

Evaluation of the prevalence of idiopathic hypersomnia in the general population is difficult due to the limited number of patients with this condition and the difficulty of making a definitive diagnosis. For example, after excluding patients with symptoms and medication that might confound a diagnosis in patients with daytime sleepiness, Laffont and colleagues comprehensively evaluated 128 patients (16 to 77 years of age) in whom no clear diagnosis had been established for their daytime sleepiness and found that only 12% (15 patients) fit criteria for idiopathic hypersomnia. Using their own nomenclature, they characterized the other subjects as having various disorders such as: mild hypersomnia type 1, hypersomnia associated with HLA type DR2-DQw1, mild hypersomnia type 2, patients with morning recovery from disrupted sleep, young “long sleepers” with difficulty waking up, poor short sleepers since childhood, and older poor sleepers with late onset of symptoms (62).

Despite the limited nature of epidemiological data, the prevalence of patients with idiopathic hypersomnia compared to narcoleptic patients provides some indication of overall prevalence. For comparison, improvements in recognition and diagnosis of type 1 narcolepsy have pointed to an estimated prevalence of 0.02% to 0.03% (81). Based on review of a large cohort of over 6000 patients with sleep disorders in the United Kingdom, idiopathic hypersomnia was found to be 40% to 60% as prevalent as narcolepsy, depending upon what MSLT criteria are used (08). These findings are lower than those in earlier studies, probably because of improved identification of sleep disorders that were formerly diagnosed as idiopathic hypersomnia. Integrating these findings into a meta-analysis exploring the epidemiology, diagnosis, pathophysiology, and treatment of idiopathic hypersomnia suggested that the prevalence of idiopathic hypersomnia in the adult population is between 1 in 5000 and 1 in 50,000.

Clinically, approximately 10% to 29% of individuals referred to a sleep center for excessive daytime sleepiness ultimately receive a diagnosis of idiopathic hypersomnia (118; 110). For comparison, there is a population prevalence of about 1.6% for individuals who sleep more than 9 hours per day and are distressed or impaired by their sleep duration (92).

Prevention

Prevention and risk factors are unknown.

Differential diagnosis

Confusing conditions

Idiopathic hypersomnia is probably one of the most frequently misdiagnosed sleep disorders. Many other conditions produce such sleepiness and can mimic or coexist with a hypersomnia of central origin. All of these possible confounders need to be considered in the differential diagnosis as possibly causing or contributing to the excessive sleepiness in a patient with hypersomnia of central origin (87).

An exploration of the causes of excessive daytime sleepiness (EDS) in 16,583 randomly sampled Pennsylvanians identified 1742 sleepy individuals who were then recruited for laboratory evaluation. Regression analysis revealed that a report of being treated for depression is the most significant risk factor for the complaint of excessive daytime sleepiness (with increasing effects in younger individuals), followed by BMI, age, subjective estimate of typical sleep duration, diabetes, smoking, and finally sleep apnea (21). This emphasizes that potentially confounding medical/psychiatric disorders and their treatments (96; 46) precludes the diagnosis of a primary CNS hypersomnia, such as idiopathic hypersomnia, and their presence must be assessed and corrected (where possible) before a diagnosis can be made (06). Additionally, primary sleep disorders that affect the quality and quantity of sleep must be ruled out, including upper airway resistance syndrome (UARS, a milder form of obstructive sleep apnea), other primary CNS hypersomnias (eg, Kleine-Levin syndrome and narcolepsy type 1 or 2), circadian sleep phase disorder, and behaviorally induced insufficient sleep syndrome.

Hypersomnia due to a medical disorder includes an irresistible urge to sleep and excessive daytime sleepiness for at least 3 months due to underlying medical or neurologic condition, and it does not meet criteria for narcolepsy. Associated conditions can include neurodegenerative diseases (eg, Parkinson disease or multiple system atrophy), posttraumatic brain injury, stroke or brain tumors (potentially involving the hypothalamus or midbrain), hypothyroidism, or metabolic derangement (chronic renal insufficiency, hepatic encephalopathy, neuroglycopenia) (29; 74; 06). Posttraumatic hypersomnia may closely mimic idiopathic hypersomnia. Hypersomnia usually develops 6 to 18 months after head trauma (49). Excessive sleepiness may be the first symptom of progressive hydrocephalus in the absence of other features of hydrocephalus. Systemic exertional intolerance disease (formerly known as chronic fatigue syndrome, or myalgic encephalomyelitis, despite a lack of evidence of CNS inflammation) is characterized by persistent or relapsing fatigue that does not resolve with bedrest. It is important to differentiate sleepiness (a propensity for dozing off) from fatigue (a lack of physical energy, or body “tiredness”), in any such patients (61). Additionally, there are a number of genetic syndromes that result in hypersomnia including autosomal dominant cerebellar ataxia, deafness, narcolepsy (141), Moebius syndrome (97; 131; 59), Coffin-Lowry syndrome (88), Niemann-Pick disease type C1 (132; 94; 99), Norrie disease (138; 97; 116), myotonic dystrophy (75), and Prader-Willi syndrome (97; 82).

Hypersomnia due to medication or substance is based on excessive daytime sleepiness and an irresistible urge to sleep due to substance/sedative use or withdrawal from stimulant medications. Substances or sedating medications can include alcohol, opiates, marijuana, benzodiazepines, nonbenzodiazepine hypnotics, barbiturates, antiepileptic medications, antipsychotic medications, anticholinergics, antihistamines, antidepressants, or combinations of medications. Beyond the obvious psychotropic sedatives, a plethora of medications have drowsiness listed as a side effect, either through primary or off-target effects (46). Withdrawal from or abrupt cessation of stimulants such as amphetamines or caffeine can also lead to symptoms of hypersomnia. This can be distinguished from idiopathic hypersomnia with a careful and detailed history, and polysomnography is generally unnecessary.

Hypersomnia associated with psychiatric disorders (dysthymia and related mood disorders) can be distinguished from idiopathic hypersomnia based upon its onset later in life and findings of low-grade, chronic depression revealed through clinical interviews and psychometric testing. In fact, as noted above, depression conferred the highest risk of excessive daytime sleepiness (OR 6.85, p < 0.001) in a large multiple logistic regression model (21). The multiple sleep latency test often does not demonstrate a short mean sleep latency in depression associated hypersomnia, and continuous 24-hour polysomnography shows normal daytime wakefulness despite patients having relative motor quiescence. Thus, they may not move but they are awake (38). Treatment of hypersomnia may or may not improve depression scores, indicating that the relationship between mood disorders and hypersomnia is more complex than simple cause and effect (52), particularly in light of the fact that patients with idiopathic hypersomnia often develop depression as a result of diagnostic delays and the psychosocial burden of their disease (35; 134). Thus, a concerted approach focused on treating both disorders is essential.

After addressing the primary medical and psychiatric conditions that may result in sleepiness, the most obvious disorders that contribute to excessive daytime sleepiness are the sleep disorders that impair sleep quality and quantity: sleep-disordered breathing (eg, obstructive sleep apnea), restless legs syndrome, periodic limb movement disorder, and circadian phase disorders (115). A less-commonly-known sleep breathing disorder is the upper airway resistance syndrome (UARS). Subjects with UARS may complain of excessive daytime sleepiness (51). Polysomnography discloses short alpha EEG arousals lasting 3 to 14 seconds that regularly interrupt snoring periods, which are polysomnographically scored as respiratory-event-related arousals (RERA). UARS is classified within obstructive sleep apnea by the American Academy of Sleep Medicine based on the most recent edition of the International Classification of Sleep Disorders (ICSD) (06). To make this diagnosis in adults, the calculated respiratory disturbance index (RDI) should be greater than five events per hour (with AHI less than 5 per hour) (50). Furthermore, circadian misalignment (most commonly problematic in those with delayed sleep phase syndrome) is a diagnostic consideration in patients whose main complaint is extreme difficulty awakening at the desired time and excessive morning sleepiness. However, these patients are not sleepy in the evening and go to bed extremely late in the night (140). The diagnosis can be confirmed with review of sleep diary and actigraphy.

Consideration of other primary CNS hypersomnias must also be given. In particular, type 1 narcolepsy is often differentiated through the telltale presence of cataplexy, which is a sudden loss of muscle tone in response to (usually positive) emotion (111). Differentiation of idiopathic hypersomnia from the non-hypocretin-deficient, type 2 narcolepsy is far more challenging, and clinical symptoms alone cannot provide a clear differentiation of the disorders (117), as they are nonspecific to any of the primary hypersomnias (110). Kleine-Levin syndrome is a rare disorder characterized by recurrent episodes of excessive sleepiness (episodes typically lasting 10 days, but possibly up to a few weeks) associated with cognitive dysfunction, altered perception of the environment, eating disorder (commonly hyperphagia), or disinhibited behavior (such as hypersexuality), with onset usually in adolescence, and gradual remission often over many years. Most remarkable is that these patients appear to return to a completely normal baseline in the months between the episodes of hypersomnia and may suffer from a degree of amnesia or report derealization of the event (06).

Given the current epidemic of sleep deprivation (Centers for Disease Control and Prevention 2015), behaviorally induced insufficient sleep syndrome must be considered in all individuals presenting with excessive daytime sleepiness, impaired concentration, and lowered energy level. A detailed history of the current sleep schedule is revealing, and actigraphy can also be useful in establishing a wake-sleep pattern (105).

Finally, normal variations in the amount of sleep needed per night can include adults who require longer sleep times, greater than 10 hours (per 24-hour period). They may be misdiagnosed with idiopathic hypersomnia because of extremely long sleep episodes at night. However, these patients are normally alert once they have slept their needed amount of sleep.

Associated or underlying disorders

Autonomic dysfunction is often present in patients with idiopathic hypersomnia, most notably Raynaud phenomenon as well as symptoms of cold extremities (46%), orthostatic intolerance (32%), temperature intolerance (25%), palpitations (23%), and digestive issues (22%) (15; 134). Additionally, disturbed heart rate variability measures pointed to increased parasympathetic tone during sleep and wake in patients with idiopathic hypersomnia (114). In a survey of patients with idiopathic hypersomnia from a mixed clinical cohort and hypersomnia registry, the prevalence of autonomic disorders was higher than that of the general population (15% and 17%, respectively), the most common of which was postural orthostatic tachycardia syndrome (POTS), which was present in a higher proportion than expected for the general population (7% and 13%, respectively) (80). In this study, the severity of autonomic symptoms in patients with idiopathic hypersomnia was significantly higher than in controls and was correlated with the severity of daytime sleepiness, fatigue, and quality of life (80). Although it’s not clear why autonomic dysfunction is so prevalent in the idiopathic hypersomnia population, a number of possibilities have been proposed, such as deconditioning due to excessive sleeping and recumbency, overlapping (possibly autoimmune) pathophysiology, and shared genetic predisposition.

Diagnostic workup

The diagnosis of idiopathic hypersomnia is mainly based upon findings discovered during a clinical history. However, actigraphy, polysomnography, and multiple sleep latency testing (MSLT) are necessary to confirm the diagnosis and to rule out other sleep disorders.

Establishing an initial diagnostic impression can be gathered through the clinical history, supplemented by questionnaires to establish the severity of the patient’s sleepiness. The Epworth sleepiness scale is quite possibly the most extensively validated metric to assess subjectively reported sleepiness over the past 4 weeks, with an accepted cutoff of greater than 10 points defining excessive daytime sleepiness (55) and 2 points indicating a clinically meaningful difference (77). However, idiopathic hypersomnia is a multifaceted condition that extends far beyond hypersomnolence. To address this, a new patient-report questionnaire—the Idiopathic Hypersomnia Severity Scale (IHSS)—was recently devised to account for the constellation of symptoms characteristic of this disorder (32). Psychometric validation of the IHSS supports a 3-component nature of the scale: daytime functioning; sleep duration and inertia; and napping, as well as meaningful clinical cutoffs of 22 (differentiating untreated idiopathic hypersomnia from controls) and 26 (differentiating untreated idiopathic hypersomnia from treated idiopathic hypersomnia in cross-sectional and longitudinal analyses) (32; 103). Finally, a change of 4 points in the scale was suggested to be clinically meaningful (103). Additionally, quantification of sleep inertia with the extensively validated psychomotor vigilance task (PVT) was studied in a diversity of clinical populations with hypersomnolence, demonstrating the strongest associations between sleep inertia and psychomotor vigilance task metrics--lapse number increase and slowest 10% 1/RT decrease–during the early morning hours (7:00 AM and 7:30 AM) (41). These scales and measures can serve to support the clinical impression in addition to providing a semiquantitative means of monitoring therapeutic response.

The most widely used test is overnight polysomnography, followed by the MSLT. This sequence of sleep studies allows sleep-disordered breathing, narcolepsy, periodic leg movements, and sleep fragmentation to be ruled out as causes of excessive daytime sleepiness.

One of the most common mistakes made performing the PSG-MSLT is to wake the patient up in the morning instead of letting them sleep in as long as they can. This leads an individual who may have a prolonged sleep requirement or a circadian rhythm delay to have shortened sleep latencies and risk a false positive result.

The MSLT was originally devised for the diagnosis of narcolepsy type 1 (24), where it is best validated (42). However, outside of narcolepsy type 1, its diagnostic utility is more limited (83; 130). A mean sleep latency cutoff of 8 minutes (with less than 8 minutes being suggestive of hypersomnia) has poor reliability, as upwards of 40% of patients with idiopathic hypersomnia have mean sleep latencies of more than 8 minutes (leading to a false negative result) and anywhere from 10% to 25% of the general public have a mean sleep latency of less than 8 minutes (leading to a false positive result) (06; 39). Moreover, attempts to verify a diagnosis of idiopathic hypersomnia with a second MSLT revealed that 75% of those initially diagnosed as idiopathic hypersomnia were normal on repeat testing (73). Further corroborating evidence included the finding that 71% of individuals with long sleep time were noted to have a normal MSLT (133), drawing into question the value of the MSLT as a diagnostic test in individuals with a clinical history suggestive of idiopathic hypersomnia. Maintenance of wakefulness testing (MWT) has also been used as a tool to determine treatment response and to assess a patient’s ability to remain awake in those whose hypersomnia may constitute a safety issue. However, like the MSLT, the MWT suffers from similar problems of reliability, and its uses remain controversial (48; 71).

The most recent version of the International Classification of Sleep Disorders has allowed for the usage of prolonged polysomnography in the objective documentation of excessive sleep (more than 660 minutes) (06). With prolonged polysomnographic monitoring, without provoked morning awakening, recordings are performed for up to 24 hours, and the subject is asked not to fight sleep. This test will typically show prolonged night sleep and one or two prolonged naps during daytime. Contrasting with narcolepsy, the macroarchitecture of sleep has generally been believed to demonstrate few or no awakenings, normal percentage of NREM sleep stages and REM sleep, and no sleep-onset rapid eye movement periods (SOREMPs) either at night or during daytime. However, a meta-analysis of polysomnographic studies of patients with idiopathic hypersomnia revealed a mix of expected and surprising findings, suggesting that, relative to controls, individuals with idiopathic hypersomnia have: (1) significantly greater total sleep time, (2) shorter nocturnal sleep onset latency, (3) relatively normal sleep efficiency (though this is in the context of much longer total sleep time), (4) reduced percentage of slow wave sleep, and (5) an increase in percentage of nocturnal REM sleep (101). The author of the meta-analysis noted that there was high heterogeneity among the studies analyzed, possibly explaining some of the findings that ran counter to conventional wisdom (101). In contrast to these findings, a meta-analysis of polysomnographic characteristics that may differentiate among the central disorders of hypersomnolence suggested that narcolepsy type 1 demonstrated more wake after sleep onset (WASO), arousals, N1 sleep (as a percentage of total sleep time), and periodic limb movements as well as lower sleep-onset latency, sleep efficiency, and N2 sleep (as a percentage of total sleep time) (142). Comparatively, there were few differences between narcolepsy type 2 and idiopathic hypersomnia, with only higher percentage of REM sleep and lower REM sleep latency in individuals with narcolepsy type 2 when compared to patients with idiopathic hypersomnia without long sleep time (142).

A group has proposed a 32-hour bed-rest protocol that follows a time-limited nocturnal polysomnogram (from 11 PM to 7 AM) and “modified MSLT,” (5 nap opportunities every 2 hours, between 9 AM and 5 PM, but where participants are awakened after 1 minute of sleep) (39). Although recognized as impractical for clinical study, the authors found that a cutoff of 19 hours was the optimal cutoff to discriminate between patients with idiopathic hypersomnia, those without idiopathic hypersomnia, and controls, suggesting a much more substantial sleep need in individuals with idiopathic hypersomnia (39). Importantly, the individuals with total sleep time greater than 19 hours were more frequently overweight, with higher frequency of sleep inertia and shorter sleep latencies on MSLT as well as undisturbed nocturnal sleep with high sleep efficiency and short sleep latency, in contrast to the meta-analysis by Plante (discussed above) and in keeping with conventional wisdom (39). As an aside, the authors also explored the diagnostic value of the current ICSD-3 criterion using at least 11 hours of sleep in a polysomnographically monitored, 24-hour period, revealing only 57% specificity for a diagnosis of idiopathic hypersomnia, suggesting that the current diagnostic strategy employing polysomnography may be lacking (39). A more extensive validation of the 32-hour protocol in 266 drug-free individuals (201 women; median age: 26.50 years [16.08; 60.87]) with complaints of hypersomnolence was subsequently published, providing additional support for the 19-hour cutoff and an algorithm for defining clear-cut idiopathic hypersomnia (40). Interestingly, in the group of individuals defined as clear-cut idiopathic hypersomnia based on objectively verified hypersomnia (EQS; excessive quantity of sleep) or hypersomnolence (EDS; excessive daytime sleepiness), there was no association with sex, body mass index, Epworth sleepiness scale, and depressive symptoms (40). Previously, other groups have developed similar prolonged sleep opportunity protocols, but despite similarly validated cutoffs, they have not gained traction in the clinical domain (11). As an alternative to polysomnographic recording, an average sleep duration of greater than 11 hours on at least 7 days of conventional actigraphy can be used to confirm long sleep durations (06).

HLA typing is not sufficiently specific or sensitive to use for diagnosis of idiopathic hypersomnia. The HLA-DQB1*06:02 allele is highly prevalent (97%) in narcolepsy type 1 but is also found frequently (approximately 40% to 50%) in narcolepsy type 2 and (approximately 12% to 38%) in the general population (69; 84; 09). A CT scan or MRI scan of the brain may be required if a structural brain lesion is a consideration, but such a decision is usually guided by clinical history evincing other symptoms or abnormalities on neurologic examination (124). Although not currently validated for clinical practice, functional neuroimaging may prove promising in helping to diagnose idiopathic hypersomnia. One study using single-photon emission computed tomography (SPECT) suggested decreased regional cerebral blood flow in areas associated with the default mode network (22). Studies using positron emission tomography (PET) have pointed to possible hypermetabolism in the salience network (31), as well as patterns of hypermetabolism in various brain regions (somewhat overlapping with hypermetabolic profiles of individuals with type 1 narcolepsy) when compared to non-sleepy controls (127). These findings have yet to be replicated, and their application to diagnostic strategies has yet to be determined. Using functional MRI, a group found anatomic and functional differences in the brains of a small number of idiopathic hypersomnia patients, such that there was lower connectivity in the anterior (but not posterior) default mode network, a finding possibly analogous to the diminished connective seen in sleep-deprived individuals (102). Psychiatric evaluation may be necessary if a mood disorder is suspected. Thyroid panel, iron studies (iron, total iron binding capacity, percent saturation, and ferritin), complete blood count, vitamin B12, or other relevant blood work may be useful for ruling out other medical conditions that are associated with hypersomnia and fatigue.

Management

Treatment of idiopathic hypersomnia is similar to that of narcolepsy, however, with substantially less evidence to support the various therapies (77; 78). In 2021, the FDA approved low-sodium oxybate for the treatment of idiopathic hypersomnia, based on demonstrated improvements in the primary (Epworth sleepiness scale scores) and secondary (patient global impression of change and idiopathic hypersomnia severity scale score) endpoints in a double-blind, placebo-controlled, randomized withdrawal study (28). All other medications are used off-label and, thus, must be used cautiously with clearly expressed and documented patient understanding of the risks and benefits of such usage.

Nonmedication treatment can include lifestyle modification, attention to sleep hygiene, or scheduled naps (01). A formal program of cognitive behavioral therapy for hypersomnia (CBT-H) was developed, demonstrating improvements in depressive symptoms (though not meeting the prespecified improvement rate of 50%) as well as global self-efficacy (93). Subjective sleepiness, as measured by the Epworth sleepiness scale, was also improved in the post-hoc subset analysis of the 12 patients with idiopathic hypersomnia (93). Education about the impact of the condition on academic development, work, driving, and social life can assist patients as they attempt to live a normal life.

Sodium oxybate, an agonist at the γ-hydroxybutyrate and GABAB receptors that promotes slow-wave sleep, was studied in 46 idiopathic hypersomnia patients, revealing a good response in 65% of the 39 patients that chose to fill the prescription (66). Additionally, a phase 3, double-blind, placebo-controlled, randomized withdrawal trial of low-sodium oxybate (once or twice nightly) in patients with idiopathic hypersomnia demonstrated efficacy in all pre-specified endpoints (28), prompting FDA approval of the medication for this indication in August of 2021. More information may be accessed at the following website: https://www.fda.gov/news-events/press-announcements/fda-grants-first-its-kind-indication-chronic-sleep-disorder-treatment. Following improvements in Epworth sleepiness scale and Idiopathic Hypersomnia Severity Scale from baseline to the stable-dosing period, there was clinically meaningful and statistically significant symptom recurrence in individuals randomly withdrawn to placebo, relative to those maintained on their low-sodium oxybate dose: least squares mean difference between groups in change in Epworth sleepiness scale score was 6.5 points worse (95% CI 5.0 to 8.0; p< 0·0001) in those withdrawn to placebo; estimated median difference in change in Idiopathic Hypersomnia Severity Scale score was 12.0 points worse (95% CI 8.0 to 15.0; p< 0·0001) in those withdrawn to placebo (28). Subgroup analyses indicated similar trends regardless of dosing regimen (once or twice nightly) or sleep duration (w/wo long sleep time) (28). The side effects and adverse events were consistent with prior studies and post-market surveillance of the therapeutic use of oxybates (28).

Modafinil and armodafinil (the R racemate of modafinil) have a benefit/risk profile in idiopathic hypersomnia similar to their effect on narcolepsy type 1 (17; 65). Modafinil has been reported to have a modest but significant effect on hypersomnia in children (53; 08). With response rates as high as 72% (after accounting for dropouts due to factors such as cost) (03) and a growing body of clinical evidence, the American Academy of Sleep Medicine has provided a “strong” recommendation for the use of modafinil as an effective therapy for idiopathic hypersomnia (77; 78). All other mediations reported on in the 2021 guideline (clarithromycin, methylphenidate, pitolisant, and sodium oxybate) received a “conditional” recommendation (77; 78). It is important to note, however, that the guideline was published prior to the publication of the results of the pivotal phase 3 trial of low-sodium oxybate for the treatment of idiopathic hypersomnia that resulted in the subsequent FDA approval. Moreover, in two survey-based assessments of treatment efficacy, the vast majority of patients reported persistence of symptoms and adverse psychosocial impacts despite the usage of off-label therapeutics (125; 113).

In light of variable response rates and intolerance of side effects to first-line monotherapy, a majority of patients with residual symptoms on monotherapy have been shown to respond to combination or alternate therapies (123). In these circumstances, other psychostimulant drugs such as dextroamphetamine and methylphenidate are the mainstays of treatment; however, due to their high potential for abuse and undesirable side effects, these medications are often reserved for second-line therapy, with a preference for long-acting agents (87). For unclear reasons, these medications do not often result in sustained benefits for idiopathic hypersomnia as they do in narcolepsy. This may result in attempts to achieve better clinical response by exceeding recommended maximum guideline dosages. Exceeding standard dosages often precipitates psychosis, headache, substance misuse, tachyarrhythmia, anorexia, or other significant side effects (12). In a small case series, an atypical dosing schedule, which consisted of evening dosing of long-acting methylphenidate or bupropion, demonstrated efficacy in combatting one of the more bothersome symptoms in idiopathic hypersomnia: morning sleep inertia (112).

Additional therapies have been employed in treatment-refractory idiopathic hypersomnia patients. The antibiotic clarithromycin may be a negative allosteric inhibitor of the GABA_A receptor, which prompted study of its effects in a retrospective chart review and small clinical trial, both of which showed modest benefits (128; 126). Similarly, the benzodiazepine receptor antagonist, flumazenil, was studied by the same group, with minor benefits in psychomotor vigilance task measures and subjective alertness (108). An inverse agonist targeting the autoinhibitory H3 histamine receptor, pitolisant, has been studied in idiopathic hypersomnia and narcolepsy with clinically significant improvements in excessive daytime sleepiness (67; 120). Another potential therapy, the GABAA receptor modulating steroid antagonist (GAMSA) GR3027, began phase 2a clinical trial enrollment of patients with idiopathic hypersomnia in 2017. In 2022, the FDA granted an orphan drug designation for mazindol ER, a partial hypocretin 2 receptor agonist and a reuptake inhibitor of dopamine, norepinephrine, and serotonin that has also been studied for use in narcolepsy and ADHD. Finally, a trial of transcranial direct current stimulation demonstrated an improvement in sleepiness and attention compared to baseline evaluation (45). Hopefully, further investigations of these and other therapies targeting pathophysiologic mechanisms will result in the discovery of therapies with sustained efficacy for idiopathic hypersomnia.

Outcomes

Even if treated, most patients with idiopathic hypersomnia suffer from psychosocial impairments in their work and personal lives as well as score lower than healthy individuals on assessments of general health (35; 95; 125; 113). Nonetheless, outcomes can be favorable with stimulant medication, and are usually based on patient subjective reports. In a survey of patients from the Hypersomnia Foundation Registry, among 379 patients with idiopathic hypersomnia on active therapy, 64.1% still reported excessive daytime sleepiness, 60.2% still required multiple alarms to awaken, and 61.1% had impaired alertness upon waking, and cognitive issues of brain fog and memory impairment persisted in 54.0% and 51.8%, respectively (125). Comparatively, fewer than 15% continued to require excessive sleep durations while on therapy (125). Consistent with reports of unremitting symptoms, patients rated their global satisfaction with treatment as low in a survey of 75 individuals with idiopathic hypersomnia (37 with long sleep time), 89% of whom were on at least one off-label therapy (predominantly stimulants and wake-promoting agents) (113).

Clinical monitoring of therapeutic response can be performed with the maintenance of wakefulness test (MWT) or actigraphy; however, this is not common unless there is an indication (eg, commercial truck drivers). A few objective tests are employed in research (eg, the sustained attention to response task) (44) and the psychomotor vigilance test (37; 14) and can provide objective outcome measures, though they tend to not correlate with symptom reports/subjective measures (110).

Complications of treatment are medication specific. In general, the psychostimulants used to treat idiopathic hypersomnia result in side effects through their action on the monoaminergic neurotransmitter systems. Medications that work primarily through reuptake inhibition (eg, modafinil) are generally better tolerated than those that also result in dopamine release or norepinephrine reuptake inhibition at higher doses (eg, amphetamine salts). Postmarketing monitoring revealed a risk for Stevens-Johnson syndrome in modafinil (27), and there is diminished hormonal contraceptive efficacy while taking this medication class (26; 27). The traditional psychostimulants (eg, amphetamine salts and methylphenidate) are more sympathomimetic, resulting in a higher risk of toxicity, in addition to the higher risk of abuse (119; 68; 85). Oxybates have a heightened risk of anxiety and depression with suicidal ideation, and they carry a boxed warning regarding the risk of central nervous system and respiratory depression, as well as the risk for abuse/misuse. More information can be accessed at the following site:https://pp.jazzpharma.com/pi/xywav.en.USPI.pdf. As a result of the potential risks of this schedule III medication, patients and providers must both be enrolled in the risk evaluation and mitigation strategy (REMS) program, which can be found at the following site:https://www.xywavxyremrems.com/ (54).

Special considerations

Pregnancy

No data are available on the effect of pregnancy on idiopathic hypersomnia or of idiopathic hypersomnia on pregnancy. Although FDA pregnancy categories no longer apply, given that all of the medications used to treat idiopathic hypersomnia are psychotropic and have the potential to cause harm to an unborn fetus, most of the relevant medications were previously in category C (ie, amphetamines, methylphenidate, modafinil, and armodafinil).

Anesthesia

In patients with idiopathic hypersomnia, stimulant drugs should be discontinued in anticipation of surgery. The effects of anesthesia on idiopathic hypersomnia are unknown.