Michel Billiard MD

In comparison with narcolepsy, which is characterized by well-defined clinical, polysomnographic, and immunogenetic features, idiopathic hypersomnia is not well delineated and its history is much more recent.

In 1966, Dement et al [10] proposed that subjects with excessive daytime sleepiness but no cataplexy, sleep paralysis, or sleep-onset rapid eye movement (REM) periods should not be considered narcoleptic. Subsequently, Berti-Ceroni et al [6] and Passouant et al [22] described narcolepsy with non-rapid eye movement (NREM) sleep episodes. The clinical picture was still vague, however, and it is very likely that among the subjects under consideration were subjects with other sleep disorders, including respiratory disorders associated with sleep. In 1972, Roth et al [28] circumscribed the limits of the condition in describing a type of "hypersomnia with sleep drunkenness," which consisted of difficulty in coming to complete wakefulness accompanied by confusion, disorientation, poor motor coordination, and slowness associated with deep and prolonged sleep. Four years later, the same group proposed a classification of hypersomnolent conditions, which included narcolepsy, hypersomnia, and the subwakefulness syndrome. [25] Hypersomnia was considered either symptomatic or functional, with the functional form subdivided into a functional hypersomnia with a short cycle and a functional hypersomnia with a long cycle (periodic hypersomnia). Functional hypersomnia with a short cycle was further subdivided into idiopathic hypersomnia and neurotic hypersomnia. Finally, idiopathic hypersomnia was considered either monosymptomatic, manifested only by excessive daytime sleepiness and not by abnormal awakening, or


polysymptomatic, characterized by excessive daytime sleepiness, nocturnal sleep of abnormally long duration, and signs of "sleep drunkenness" on awakening.

According to the International Classification of Sleep Disorders (ICSD), [3] idiopathic hypersomnia is "a disorder of presumed central nervous system cause that is associated with a normal or prolonged major episode and excessive sleepiness consisting of prolonged (1-2 hours) sleep episodes of NREM sleep." Clearly, the ICSD definition merges Roth's monosymptomatic and polysymptomatic forms by indicating that some patients report great difficulty waking up and experience disorientation after awakening.

Finally, based on a retrospective review of clinical and polygraphic findings and questionnaire results in groups of subjects with narcolepsy, narcolepsy without cataplexy, idiopathic hypersomnia, insufficient sleep syndrome, mild sleep apnea, and excessive daytime sleepiness not otherwise specified, Aldrich [2] suggests that the syndrome of idiopathic hypersomnia is a heterogeneous syndrome. He also reports that relatively few patients have the polysymptomatic form of idiopathic hypersomnia described by Roth [25] and that other subjects experience prolonged sleep without sleep drunkenness, excessive daytime sleepiness without prolonged sleep hours, a clinical syndrome with sleep paralysis, or sleep-related hallucinations clinically indistinguishable from narcolepsy without cataplexy but without REM sleep abnormalities.

From the Neurology B Department, Gui-de-Chauliac Hospital, Montpellier, France

Address reprint requests to
Michel Billiard, MD
Neurology B Department
Gui-de-Chauliac Hospital
34295 Montpellier, Cedex 05


Because of the somewhat vague limits of idiopathic hypersomnia and its relative rarity, no prevalence study has ever been conducted. The only available data are the ratios of idiopathic hypersomnia to narcolepsy in different sleep disorders populations (Table 1) . The ratios



(no. patients)

(no. patients)

(no. patients)

Ratio of
to Narcolepsy

Roth [26]





Coleman et al [9]





Baker et al [4]





Aldrich [2]





Billiard et al
  (unpublished data)






Figure 1. Age of onset of idiopathic hypersomnia in the author's 32-patient population. Onset after 20 years of age is relatively rare.

tend to decrease in the successive reports, probably due to the use of more stringent criteria, with the notable exception, however, of Aldrich's [2] recent series.


In contrast with narcolepsy, a precise onset of idiopathic hypersomnia is often difficult to determine because of the insidious beginning of the condition and the difficulty, in young persons, of settling a posteriori a limit between long sleep and abnormally long sleep or normal wakefulness and impaired wakefulness. As is the case for narcolepsy and the Kleine-Levin syndrome, however, onset of this condition is most often during adolescence and rarely after the age of 30 years (Fig. 1) .


The symptoms of idiopathic hypersomnia may vary. [2] The most typical form, which Roth et al [25] referred to as polysymptomatic, is characterized by excessive daytime sleepiness leading to unrefreshing prolonged naps, nocturnal sleep of long duration (as much as 12 hours or more), and sleep drunkenness. Subjects affected with this type of idiopathic hypersomnia often refrain from naps because of their spontaneous


long duration and their unrefreshing nature. They use sophisticated alarm clocks and often report drinking a cup of strong coffee before being able to take a shower. In the author's own series of patients, 19 out of 32 disclosed this set of symptoms.

In other cases, subjects complain of excessive daytime sleepiness, have naps of shorter duration that may be refreshing or unrefreshing, and do not report abnormally long sleep or sleep drunkenness. This set of symptoms represents a more heterogeneous form of the condition. It would not be a suprise if conditions other than idiopathic hypersomnia are identified in these subjects with more sophisticated methods of evaluation. In addition, some subjects report headaches, which may be migrainous in quality, fainting episodes, orthostatic hypotension, and peripheral vascular complaints of the Raynaud type. Hypnagogic hallucinations and sleep paralysis may be encountered occasionally.


Idiopathic hypersomnia is a life-long disorder with no tendency to remit spontaneously. Complications are mostly social and professional. [1] [8] Whether subjects' symptoms can be temporarily alleviated by having their sleep extended has not been investigated.


The diagnosis of idiopathic hypersomnia is mainly based on clinical features and the absence of associated symptoms such as cataplexy, snoring at night, periodic leg movements, or depression. Polysomnography and some other tests are necessary to rule out other sleep disorders, however. Indeed, in the author's experience, idiopathic hypersomnia is one of the most overdiagnosed sleep disorders.

The most widely used polysomnographic procedure is nocturnal sleep recording followed by a multiple sleep latency test (MSLT). This procedure shows sleep of normal quality with few awakenings and a normal proportion of the different sleep stages. Sleep apneas, restless legs, and periodic movements in sleep are absent. MSLT usually demonstrates a mean sleep latency of less than 10 minutes. It may be, however, that subjects with a very long nocturnal sleep episode and sleep drunkenness do not show a very short sleep latency on the MSLT. Sleep-onset REM periods (SOREMPs) are uncommon.

The diagnostic value of the MSLT is somewhat questionable in subjects with the polysymptomatic form of idiopathic hypersomnia, however. In these cases, awakening the subject early in the morning for


the MSLT precludes documentation of the prolonged nighttime sleep, and the MSLT protocol precludes the recording and observation of prolonged, unrefreshing daytime sleep episodes. Thus, it is more useful to perform an all-night sleep polygraphic recording followed by an MSLT and then, from 7:00 PM onward, a 24-hour continuous polysomnography, either at home with an ambulatory system or in the laboratory on an ad lib protocol. Using this approach in a group of 19 patients with polysymptomatic idiopathic hypersomnia, the author found that total sleep time was 819.50 minutes ± 23.10 during the 24-hour recording after the MSLT (Fig. 2) .

Despite some reports that suggest an increased frequency of human leukocyte antigen (HLA) Cw2 and HLA DR5 in idiopathic hypersomnia, [19] HLA typing is of no help in the positive diagnosis of idiopathic hypersomnia.

Psychological interview and tests are appropriate to exclude a neurotic hypersomnia and a CT scan or MR imaging of the brain is indicated when a structural brain lesion is a consideration.

Differential Diagnosis

Idiopathic hypersomnia is probably one of the most frequently overdiagnosed sleep disorders because there is a tendency to classify in

Figure 2. Twenty-four-hour continuous polysomnogram in a 44-year-old man. Night total sleep time was 739 minutes. During the following day, there was a unique, unrefreshing nap of 156-minute duration. Twenty-four-hour total sleep time was 895 minutes. Mean sleep latency on the multiple sleep latency test performed the previous day was 8 minutes. S = sleep stage; MT = movement time.


this category all hypersomnias that do not fit the criteria of either narcolepsy or the sleep apnea syndrome. Indeed, the difficulty does not stem from disorders of excessive daytime sleepiness, such as narcolepsy or the obstructive sleep apnea syndrome, which are identifiied easily by their clinical and polysomnographic features; instead, it comes from other disorders associated with excessive daytime sleepiness that require more sophisticated investigation or that still are delineated insufficiently both clinically and polysomnographically.

The first diagnosis to consider is the upper airway resistance syndrome. [16] Excessive daytime sleepiness is a frequent feature. Other symptoms include heavy snoring--more so in men than in women--and disrupted nocturnal sleep. Bruxism is present frequently during sleep. Men and women are affected equally. Obesity is uncommon in this condition. Anatomic abnormalities include a triangular face, a high and narrow arched palate, a long uvula, a class II malocclusion of the mouth (hypoplastic mandible), and, at times, a retroposition of the mandible. Polysomnographic recording shows short alpha electroencephalographic (EEG) arousals lasting 3 to 4 seconds. The presence of increased airway resistance during sleep requires the monitoring of esophageal pressure (Pes) to measure transpleural pressure. Quantification of airflow using a face mask with a pneumotachometer for the calculation of tidal volume is also helpful, but the tight-fitting mask may affect sleep continuity. The most typical pattern is an EEG arousal occurring after a sequence of breaths, with progressively increasing inspiratory efforts indicated by progressively more negative peak end inspiratory pressure at Pes measurements (a pattern called a crescendo) and a simultaneous decrease in tidal volume compared with preceding breaths not accompanied by increased effort. (For a more thorough discussion of this topic, refer to the article by Chervin and Guilleminault elsewhere in this issue.)

Narcolepsy without cataplexy, also referred to as monosymptomatic, independent, ambiguous, or atypical narcolepsy, or more recently as hypersomnia with sleep-onset REM periods, [13] is a clinical variant of narcolepsy in which cataplexy is not yet manifest or will not ever manifest. Positive diagnosis requires the presence of two or more SOREMPs on the MSLT and an association with the HLA DR15-DQ6 haplotype.

Hypersomnia associated with dysthymia and related mood disorders is a frequent alternative diagnosis. The complaints of such patients are similar to those of patients with idiopathic hypersomnia; however, its onset is usually later in life, and interview and personality tests point to low-grade, chronic depression or an inability to cope with stressful situations. The MSLT often does not demonstrate a short mean sleep latency. A continuous 24-hour polysomnographic recording without provoked morning awakening may show repeated night awakenings and


little sleep during the daytime, despite the maintenance by the patient of a reclining position. [7]

Chronic fatigue syndrome is characterized by persistent or relapsing fatigue that does not resolve with bed rest. Several minor symptoms often are associated. Polysomnographic recording shows reduced sleep efficiency and alpha intrusion into sleep EEG. [18]

Hypersomnia that develops after a viral infection, such as atypical viral pneumonia, mononucleosis, or Guillain-Barre syndrome, usually develops within weeks or months after the infection. Individuals complain of fatigue and somnolence and may sleep most of the 24-hour day. The outcome is favorable, but the resolution may take months or years. [14]

Post-traumatic hypersomnia may mimic idiopathic hypersomnia closely. Hypersomnia usually develops 6 to 18 months after head trauma. [15] Excessive daytime sleepiness may be the first symptom of progressive hydrocephalus in the absence of other features of hydrocephalus.

Delayed sleep phase syndrome is a diagnostic consideration in some patients whose main complaints are extreme difficulty awakening at the desired time and excessive morning sleepiness. These patients are not sleepy throughout the day, however, and go to bed or fall asleep extremely late at night. [29] (For more information on this topic, refer to the article by Wagner elsewhere in this issue.)

Insufficient sleep syndrome is associated with excessive daytime sleepiness, impaired concentration, and lowered energy level. A detailed history of the subject's current sleep schedule is revealing. [24]

Pain or other medical symptoms responsible for fragmented sleep at night frequently result in excessive daytime sleepiness.

Long sleepers, also called "healthy hypersomniacs," are persons who require more sleep at night than the norm. They may be misdiagnosed with idiopathic hypersomnia because of extremely long sleep episodes at night. These subjects are normally alert, however, once they have obtained their needed amount of sleep.


In contrast with narcolepsy, no animal model of idiopathic hypersomnia is available; thus, experimental approaches are limited. 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, however, leads to hypersomnia with a proportional increase of NREM sleep and REM sleep, resembling idiopathic hypersomnia. This state is accompanied by a decrease of diencephalic norepinephrine. [23]

Not much progress has been made concerning the neurochemical


basis of idiopathic hypersomnia since the early 1980s. At that time, Montplaisir et al [20] found that dopamine and indoleacetic acid, a tryptamine metabolite, were decreased significantly in the cerebrospinal fluid (CSF) of both idiopathic hypersomniac and narcoleptic subjects in comparison with control subjects. On the other hand, Faull et al [11] did not find any difference in the mean CSF concentrations of monoamine metabolites (DOPAC, MHPG, HVA, and 5-HIAA) in narcoleptic, idiopathic hypersomniac, and control subjects. Later reanalysis of the results of this second study using principal component analysis as the multivariate statistical model, however, brought new perspective to this study. This reanalysis showed that all four monoamine metabolites were intercorrelated highly in the normal volunteers, whereas HVA and DOPAC did not correlate with the other two metabolites in the narcoleptic subjects and MHPG did not correlate with the other three metabolites in the idiopathic hypersomniac subjects. [12] These data suggest that a malfunction of the dopamine system could exist in narcolepsy with a similar malfunction of the norepinephrine system in idiopathic hypersomnia.

A genetic basis for idiopathic hypersomnia is suggested by several works. In one study, 9 subjects in a group of 23 probands had one or several family members with excessive sleepiness. [21] In a second study, 45 subjects in a group of 167 probands had one or several relatives affected. [27] In the author's population of 32 subjects selected according to stringent criteria, 11 subjects indicated that one or more first-degree relatives were affected with excessive daytime sleepiness. Although these data strongly suggest a genetic component to idiopathic hypersomnia, the limited number of families studied to date does not permit determination of the mode of inheritance.

Because of the association of narcolepsy with specific HLA markers, similar immunogenetic studies have been conducted with idiopathic hypersomnia. Harada et al [17] found no significant difference in the distribution of HLA antigens in 41 subjects with "essential hypersomnia" in comparison with control subjects. On the other hand, Montplaisir and Poirier, [19] in a group of 18 subjects with idiopathic hypersomnia, found an association with HLA-Cw2 in 22.2% of patients, which was greater than the 5.7% found in controls ( P < 0.05), and an association with HLA-DR11, a subtype of DR5, in 42.1% of patients but in only 26% of controls. In the author's population of 32 probands, an increased frequency of HLA-Cw2 and DR11 was not present, but there was a significant increase of DQ3 ( P < 0.04).


Treatment of idiopathic hypersomnia is far from satisfactory. Stimulant drugs, such as dextroamphetamine, methylphenidate, mazindol, or


pemoline, are the most commonly prescribed medications. These drugs often are less effective in idiopathic hypersomnia than in narcolepsy, however, and sometimes not as well tolerated. Recently, a new compound, modafinil, the mechanism of which is still unclear, produced good results in subjects with idiopathic hypersomnia. [5] A double-blind controlled study has not yet been done.

In contrast with narcolepsy, naps may not be refreshing for patients with idiopathic hypersomnia; consequently, subjects avoid napping. Yet, "sleep therapy," consisting of sleeping as much as possible for several days and nights, has been proposed. [27]


Idiopathic hypersomnia is not as well delineated as narcolepsy and its history is much more recent. There are at least two forms of the disorder: (1) a polysymptomatic form, characterized by excessive daytime sleepiness, nocturnal sleep of abnormally long duration, and signs of sleep drunkenness on awakening, and (2) a monosymptomatic form that manifests only by excessive daytime sleepiness. The most widely used laboratory procedures are nocturnal polysomnographic recording followed by an MSLT demonstrating a mean sleep latency of less than 10 minutes. At least in the polysymptomatic form, however, continuous polysomnography on an ad lib protocol deserves to be performed to catch the abnormally long major sleep episode and the long unrefreshing naps. Idiopathic hypersomnia is probably one of the most overdiagnosed sleep disorders. Several other disorders must be excluded before the diagnosis can be considered conclusive. Treatment of idiopathic hypersomnia relies on stimulants, which are frequently less effective and less well tolerated than in narcolepsy.


1. Aldrich MS: Automobile accidents in patients with sleep disorders. Sleep 12:487-494, 1989

2. Aldrich MS: The clinical spectrum of narcolepsy and idiopathic hypersomnia. Neurology 46:393-401,1996

3. American Sleep Disorders Association: ICSD--International Classification of Sleep Disorders: Diagnostic and Coding Manual. Diagnostic Classification Steering Committee, Thorpy MJ (Chairman). Rochester, MN, American Sleep Disorders Association, 1990, p 285

4. Baker TL, Guilleminault C, Nino-Murcia G, et al: Comparative polysomnographic study of narcolepsy and idiopathic central nervous system hypersomnia. Sleep 9:232-242, 1986

5. Bastuji H, Jouvet M: Successful treatment of idiopathic hypersomnia and narcolepsy with Modafinil. Prog Neuropsychopharmacol Biol Psychiatry 12:695-700, 1988

6. Berti-Ceroni G, Coccagna G, Gambi D, et al: Considerationi clinico poligrafiche sull narcolessia essenziole a somno lento Sist Nerv 19:81-89, 1967


7. Billiard M, Dolenc L, Aldaz C, et al: Hypersomnia associated with mood disorders: A new perspective. J Psychosom Res 38(suppl 1):41-47, 1994

8. Broughton R, Nevsimalova S, Roth B: The socioeconomic effects (including work, education, recreation and accidents) of idiopathic hypersomnia. Sleep Research 7:217, 1978

9. Coleman RM, Roffwarg HP, Kennedy SJ, et al: Sleep-wake disorders based on a polysomnographic diagnosis: A national cooperative study. JAMA 247:997-1003, 1982

10. Dement WC, Rechtschaffen A, Gulevich G: The nature of the narcoleptic sleep attack. Neurology 16:18-33, 1966

11. Faull KF, Guilleminault C, Berger PA, et al: Cerebrospinal fluid monoamine metabolites in narcolepsy and hypersomnia. Ann Neurol 13:258-263, 1983

12. Faull KF, Thiemann S, King RJ, et al: Monoamine interactions in narcolepsy and hypersomnia: A preliminary report. Sleep 9:246-249, 1986

13. Guilleminault C: Narcolepsy syndrome. In Krieger MH, Roth T, Dement WC (eds): Principles and Practice of Sleep Medicine, ed 2. Philadelphia, WB Saunders, 1994, p 549

14. Guilleminault C, Mondini S: Infectious mononucleosis and excesive daytime sleepiness: A long term follow-up study. Arch Intern Med 146:1333-1335, 1986

15. Guilleminault C, Faull KF, Miles L, et al: Posttraumatic excessive daytime sleepiness: A review of 20 patients. Neurology 33:1584-1589, 1980

16. Guilleminault C, Stoohs R, Clerk A, et al: A cause of excessive daytime sleepiness: The upper airway resistance syndrome. Chest 104:781-787, 1993

17. Harada S, Matsuki K, Honda Y, et al: Disorders of excessive daytime sleepiness without cataplexy, and their relationship with HLA in Japan. In Honda Y, Juji T (eds): HLA in Narcolepsy. Berlin, Springer-Verlag, 1988, p 172

18. Moldofsky H: Non-restorative sleep and symptoms after a febrile illness in patients with fibrositis and chronic fatigue syndrome. J Rheumatol 16(suppl 19):150-153, 1989

19. Montplaisir J, Poirier G: HLA in disorders of excessive daytime sleepiness without cataplexy in Canada. In Honda Y, Juji T (eds): HLA in Narcolepsy. Berlin, Springer-Verlag, 1988, p 186

20. Montplaisir J, de Champlain J, Young SN, et al: Narcolepsy and idiopathic hypersomnia: Biogenic amines and related compounds in CSF. Neurology 32:1299-1302, 1982

21. Nevsimalova-Bruhova S, Roth B: Heredofamilial aspects of narcolepsy and hypersomnia. Schweiz Arch Neurol Neurochirg Psychiat 110:45-54, 1972

22. Passouant P, Popoviciu L, Velok G, et al: Etude polygraphique des narcolepsies au cours du nycthemere. Rev Neurol (Paris) 118:431-441, 1968

23. Petitjean F, Jouvet M: Hypersomnie et augmentation de lacide 5-hydroxy-indolacetique cerebral par lesion isthmique chez le chat. Comptes Rendus des Seances de la Societe de Biologie (Paris) 164:2288-2293, 1970

24. Roehrs T, Zorick E, Sickelsteel R, et al: Excessive daytime sleepiness associated with insufficient sleep. Sleep 6:319-325, 1983

25. Roth B: Functional hypersomnia. In Guilleminault C, Dement WC, Passouant P (eds): Narcolepsy. New York, Spectrum, 1976, p 333

26. Roth B: Narcolepsy and hypersomnia. Schweiz Arch Neurol Psychiatr 119:31-41, 1976

27. Roth B: Narcolepsy and Hypersomnia. Basel, Karger, 1980

28. Roth B, Nevsimalova S, Rechtschaffen A: Hypersomnia with sleep drunkenness. Arch Gen Psychiatry 26:456-462, 1972

29. Weitzman ED, Czeisler CA, Coleman RM, et al: Delayed sleep phase syndrome: A chronobiological disorder with sleep onset insomnia. Arch Gen Psychiatry 38:737-746, 1981