Primary Psychiatry. 2004;11(4):25-27
• Periodic hypersomnia, accompanied by behavioral changes, excessive-compulsive eating, and hypersexual behavior in adolescent males, is strongly suggestive of Kleine-Levin syndrome (KLS).
• Diagnostic tools, such as polysomnography and the multiple sleep latency test, may be helpful in establishing daily hypersomnia and KLS.
• Prodromal symptoms, such as fever and stress, may imply a possible viral etiology for KLS.
Kleine-Levine syndrome (KLS) is a rare and relatively benign adolescent disorder affecting mainly males. The clinical picture of KLS is distinct and enables a clinical diagnosis without the need for additional laboratory work-up in most cases, although diagnostic tools, such as polysomnography and the multiple sleep latency test, might be helpful. KLS is identified by the presence of repeated attacks of daily hypersomnolence lasting for days to weeks, accompanied by behavioral changes, excessive-compulsive eating, and hypersexual behavior. However, a particular attack might consist of hypersomnolence accompanied only by behavioral changes. In such cases, the diagnosis cannot be confirmed without first considering drug misuse or abuse, encephalitis, and a number of psychiatric aberrations. KLS is also transitory and eventually all patients completely recover—a fact that should be disclosed to parents and patients as soon as a diagnosis is confirmed. A viral etiology for KLS has been postulated recently, and the possibility that Borna disease virus may be the causative factor should be further studied.
Paroxysmal aberrant behavior associated with drowsiness is not uncommon in adolescents and is frequently attributed to drugs and/or alcohol abuse. Occasionally, such adolescents may find themselves admitted to a psychiatric ward where they may receive a diagnosis of acute psychosis and are treated accordingly. Others may be placed in custody for alleged “drug abuse.” If the initial referral is to a general hospital, the diagnosis of encephalitis is frequently entertained, followed by an inevitable lumbar puncture, electroencephalography, neuroimaging, and not rarely, a course of intravenous antiviral or antibiotic agent.
A small number of these patients have neither a psychotic disorder nor a central nervous system (CNS) infection, but are affected with the relatively rare and underdiagnosed condition known as Kleine-Levin syndrome (KLS). This review summarizes the known clinical features, laboratory work-up, and long-term prognosis of this peculiar syndrome.
Critchley and Hoffman1 coined the term KLS in 1942, giving credit to Willi Kleine,2 who first reported periodic hypersomnia in 1925, and to Max Levin,3 who had written on narcolepsy and other morbid hypersomnias in 1929. In 1962, Critchley4 reviewed the case histories of 15 adolescent males, drew upon 9 personal cases, and precisely outlined the main diagnostic criteria and the long-term prognosis of KLS, which is still valid today.
KLS is relatively rare. From 1925–1998, there were only approximately 119 articles published on KLS or periodic hypersomnia.5 Of these reports, 108 contained data from 163 cases; 142 of these cases fulfilled the current required diagnostic criteria for KLS.5
The single largest published series of KLS cases5 diagnosed 34 Israeli patients over the last 25 years. Follow-up data for periods <15 years from onset of the first attack of hypersomnia were obtained in 25 patients. This relatively large number of patients diagnosed was due to the fact that the sleep laboratory at the Technion, the Israel Institute of Technology in Haifa, where the sleep studies of those patients were performed, was for most of time the only sleep laboratory in Israel; practically every young patient with periodic hypersomnia studied by that laboratory was offered a clinical work-up.
The most striking clinical feature of KLS is the repeated attack of progressive and deepening drowsiness, leading to a period of sleep lasting from a few days to several weeks. This period of hypersomnia may be accompanied by behavioral and cognitive impairment in the form of mood and perception changes, confusion with temporal disorganization, irritability, subdued psychomotor activity, and autistic behavior. Recurrent hypersomnia, compulsive eating, and hypersexuality are also generally associated with KLS, though their actual prevalence is less common. In the above mentioned study,5 65.5% of patients had compulsive eating in addition to hypersomnolence, and 43.7% had increased sexual drive and behavioral aberrations during the attack. This may explain in part the significant diagnostic delay in some patients, which can be as long as 14 years.5
The mean onset of KLS occurs at 15 years of age (specifically between 9 and 21 years of age).5 The male-female ratio for KLS is 3:1, but the clinical presentation and course of illness is identical for both genders.6 KLS is sporadic; however, in one family, a mother and son were affected and in another, a brother and sister.5 The long-term prognosis for KLS is excellent and all Israeli patients were free of attacks after 30 years of age.5
Though clinical presentation of KLS generally remains constant, the nature of acute attacks can vary, even in the same patient. At onset, the main disabling symptom is the hypersomnolent attack. Patients and families frequently recall the first attack as a dramatic and frightening event. Later in the course of illness, the duration of sleep gradually gets shorter; during the attenuated attacks, patients usually complain of fatigue and describe a feeling of being in another world or in a state of twilight (between sleeping and waking). A new stressful experience frequently can precipitate an attack. Brief febrile illness, strenuous climatic conditions, sleeping outside for the first time, and menstruation, for example, were reported by almost 50% of patients in the Israeli study.5 One female patient who was free of attacks for 10 years had an attack while hiking in New Zealand immediately after leaving her companions to travel “on her own.” The premorbid history of KLS is usually unremarkable: in particular; there is no history of head trauma, endocrinopathy, drug abuse, psychopathology, or previous neurological impairment.
Acute attacks may lead to compulsive eating and hypersexuality. Uncontrolled eating may take the form of an irresistible urge for certain foods, especially chocolate. One patient in the Israeli study5 even entered the ward’s kitchen and broke into the locked refrigerator looking for chocolate. Hypersexual behavior, on the other hand, is not always evident to the inexperienced observer, as it can be expressed by aberrant sexual ideation; it can go unnoticed if not expected by the examiner. One female KLS patient5 mentioned in her diary that during the last days of an attack when she was awake but still not fully alert, she took a bus drive and felt that the driver was having sex with her by “deliberately shaking the bus.” The patient gave the diary to the clinician only after the subject of “thinking sexual” was raised during a follow-up visit. Another 11-year-old patient5 was taken to his mother’s bed during an attack; when she suddenly realized that he was mounting her, she immediately rushed him to a psychiatric hospital. This was his fourth attack, although the parents were not concerned until then because a diagnosis of viral infection had been given for the previous attacks.
Between attacks, the patients behave normally and have normal wake-sleep patterns, eating habits, and sexual behavior. Years later, however, during follow-up interviews,5 many patients realize that even between attacks, their state of alertness had been much different from their present fully recovered state. They perceive that period as “hazy,” “in twilight,” and “unreal.”
Both during and between attacks, the neurological examination is normal except for the cognitive and behavioral changes mentioned above. Extensive laboratory work-up performed in several patients was unremarkable,5 as were the cerebrospinal fluid, electroencephalography, brain computed tomography (CT) scan, and magnetic resonance imaging (MRI). The most common diagnosis at the initial presentation of hypersomnia was encephalitis, followed by epilepsy and drug abuse. Hypothalamic tumors, which are relatively frequent in this age group, may mimic some of the features of KLS. However, “dysautonomic” features, such as brady-tachy arrhythmia, spiky fever, labile blood pressure, anorexia, failure to thrive, and endocrinopathy, should alert the physician to the possibility of hypothalamic structural lesions.
Drug therapy in the form of central stimulants, such as amphetamines, nonamphetamine wake-promoting agents (ie, modafinil), antidepressants, anticonvulsants, neuroleptics, melatonin, and antiviral agents, were considered ineffective in the Israeli study5; however, some of the above mentioned medications were considered effective in anecdotal case reports.7
The cause of this transient and peculiar neuropsychological syndrome is obscure. The combination of episodic hypersomnia, hyperphagia, and hypersexuality points toward the hypothalamus as an attractive site for investigation. However, there is no sound endocrinological, radiological, or pathological support for such an assumption. Studies8-11 of diurnal secretion of hypothalamic hormones during and between attacks have indicated the presence of hypothalamic dysfunction, but the methods applied by the various investigators were diverse and no data were obtained for patients who completely recovered. Brain CT in 21 patients and brain MRI in 26 patients reported in the literature were normal,5,12 and the only four published neuropathological studies13-16 of patients with atypical presentation disclosed changes compatible with focal encephalitis. In three of those studies,13-15 the hypothalmus was entirely spared, while in the fourth case,16 inflammatory changes were confined to the hypothalamus, temporal lobe, and amygdala.
The psychiatric literature is sparse on KLS. During attacks, irritability, depression, euphoria, illusions, hallucinations, amnesia, disorientation, and incoherent speech have been observed.17 This spectrum of behavioral aberrations mentioned in the early literature and based on anecdotal case reports may represent the inaccuracy of the diagnosis, as well as an attempt to lump all cases of hypersomnolence or transient alteration in alertness accompanied by behavioral changes into the “waste basket” diagnosis of KLS. Conditions, such as psychotic, affective, or dissociative disorders, narcolepsy, phase-shift of daily sleep-wake cycle, and bipolar disorders, should be part of the differential diagnosis.
The young age of onset and the relatively frequent occurrence of antecedent febrile illness precipitating attacks, together with pathological findings compatible with “viral encephalitis,” raises the possibility of a viral etiology. This has prompted clinicians to administer antiviral drugs, mainly amantadine, to several patients during an attack and as prophylaxis. Unfortunately, this seems to be ineffective.
Another possible candidate in KLS is Borna disease virus (BDV). This is a neurotropic, nonsegmented, negative-stranded, ribonucleic acid virus that persistently infects warm-blooded animals and can induce progressive neurological disorders associated with diverse pathological manifestations.18 In horses and other natural animal hosts, infection with BDV can persist in the CNS and can induce alterations in brain cell functions, neurodevelopmental abnormalities, and behavioral disturbances.19 Altered sleep-wake cycles and preference for salt-containing solutions observed in the rat model20 may have some relevance to the complex symptoms of patients with KLS. Moreover, an association between BDV and psychiatric disorders (essentially schizophrenia and affective disorders) has been suggested by some serologic and molecular studies.21
With this in mind, Gadoth and colleagues (unpublished data, 2000) studied the presence of BDV antigen and BDV antibodies in plasma and mononuclear cells obtained from 12 patients with KLS (8 were evaluated between attacks and 4 during an attack). The preliminary results suggest a possibility of BDV infection in some of the patients.
Polysomnography during and between attacks may be helpful in ruling out narcolepsy and sleep-phase shift. Normal sleep structure with decreased sleep efficiency due to frequent awakenings from stage-2 sleep is the characteristic finding during an attack. Similar but milder changes were recorded during the asymptomatic period of KLS.5 The multiple sleep latency test may be helpful in objectively demonstrating hypersomnia.22
Recently, an autoimmune mechanism was postulated based on the fact that among 30 patients with KLS, there were 3 homozygotes (2 siblings and an additional sporadic case) for the human leukocyte antigen (HLA) DQB1*0201. However, the results of this report were only marginally significant due to small sample size.12
Although KLS is rare, the familiarity of psychiatrists with the spectrum of signs and symptoms of the disorder will increase the number of properly diagnosed cases, as many of them are initially seen by child or adolescent psychiatrists. It will also save patients and families unnecessary and sometimes invasive diagnostic tests; furthermore, alarmed patients and parents will be greatly relieved to learn about the favorable prognosis. These parents will be better able to cope with the attack and may not rush their child to the hospital.
KLS patients can lead productive lives. Several patients have been drafted into the Israeli army and have served successfully during their years of compulsory service. Military medical authorities have been advised to let the patients sleep during an attack at their home and then get back to their army duties, which was eventually successful.
The future goal of research in KLS should be directed toward precise diagnosis, thus providing researchers with a uniform group of patients in whom the etiological factors, such as BDV and HLA susceptibility, could be systematically studied. PP
1. Critchley M, Hoffman HL. The syndrome of periodic somnolence and morbid hunger (Kleine-Levin). Br Med J. 1942;1(4230):137-139.
2. Kleine W. Periodische Schlafsucht. Monatsschr Psychiatr Neurol. 1925;57:285-320.
3. Levin M. Narcolepsy (Gelineau’s syndrome) and other varieties of morbid somnolence. Arch Neurol Psychiatr. 1929;22:1172-1200.
4. Critchley M. Periodic hypersomnia and megaphagia in adolescent males. Brain. 1962;85:627-656.
5. Gadoth N, Kesler A, Vainstein G, Peled R, Lavie P. Clinical and polysomnographic characteristics of 34 patients with Kleine-Levin syndrome. J Sleep Res. 2001;10(4):337-341.
6. Kesler A, Gadoth N, Vainstein G, Peled R, Lavie P. Kleine-Levin syndrome (KLS) in young females. Sleep. 2000;23(4):263-267.
7. Muratori F, Bertini N, Masi G. Efficacy of lithium treatment in Kleine-Levin syndrome. Eur Psychiatry. 2002;17(4):232-233.
8. Chesson AL Jr, Levine SN, Kong LS, Lee SC. Neuroendocrine evaluation in Kleine-Levin syndrome: evidence of reduced dopaminergic tone during periods of hypersomnolence. Sleep. 1991;14(3):226-232.
9. Fernandez JM, Lara I, Gila L, O’Neill of Tyrone A, Tovar J, Gimeno A. Disturbed hypothalamic-pituitary axis in idiopathic recurring hypersomnia syndrome. Acta Neurol Scand. 1990;82(6):361-363.
10. Gadoth N, Dickerman Z, Bechar M, Laron Z, Lavie P. Episodic hormone secretion during sleep in Kleine-Levin syndrome: evidence for hypothalamic dysfunction. Brain Dev. 1987;9(3):309-315.
11. Malhotra S, Das MK, Gupta N, Muralidharan R. A clinical study of Kleine-Levin syndrome with evidence for hypothalamic pituitary axis dysfunction. Biol Psychiatry. 1997;42(4):299-301.
12. Dauvilliers Y, Mayer G, Lecendreux M, et al. Kleine-Levin syndrome: an autoimmune hypothesis based on clinical and genetic analysis. Neurology. 2002;59(11):1739-1745.
13. Carpenter S, Yassa R, Ochs R. A pathologic basis for Kleine-Levin syndrome. Arch Neurol. 1982;39(1):25-28.
14. Fenzi F, Simonati A, Crosato F, Ghersini L, Rizzuto N. Clinical features of Kleine-Levin syndrome with localized encephalitis. Neuropediatrics. 1993;24(5):292-295.
15. Koerber RK, Torkelson R, Haven G, Donaldson J, Cohen SM, Case M. Increased cerebrospinal fluid 5-hydroxytryptamine and 5-hydroxyindoleacetic acid in Kleine-Levin syndrome. Neurology. 1984;34(12):1597-1600.
16. Takrani LB, Cronin D. Kleine-Levin syndrome in a female patient. Can Psychiatr Assoc J. 1976;21(5):315-318.
17. Orlosky MJ. The Kleine-Levin syndrome: a review. Psychosomatics. 1982;23(6):609-621.
18. de la Torre JC. Bornavirus and the brain. J Infect Dis. 2002;186(suppl 2):S241-S247.
19. Chen CH, Chiu YL, Wei FC, et al. High seroprevalence of Borna virus infection in schizophrenic patients, family members and mental health workers in Taiwan. Mol Psychiatry. 1999;4(1):33-38.
20. Bautista JR, Schwartz GJ, de La Torre JC, Moran TH, Carbone KM. Early and persistent abnormalities in rats with neonatally acquired Borna disease virus infection. Brain Res Bull. 1994;34(1):31-40.
21. Taieb O, Baleyte JM, Mazet P, Fillet AM. Borna disease virus and psychiatry. Eur Psychiatry. 2001;16(1):3-10.
22. Rosenow F, Kotagal P, Cohen BH, Green C, Wyllie E. Multiple sleep latency test and polysomnography in diagnosing Kleine-Levin syndrome and periodic hypersomnia. J Clin Neurophysiol. 2000;17(5):519-522.
Dr. Gadoth is professor and chair of the Department of Neurology at Meir General Hospital in Kfar-Saba, and is professor of neurology at the Sackler Faculty of Medicine at Tel-Aviv University in Tel-Aviv, both in Israel.
Disclosure: The author reports no financial, academic, or other support of this work.
Please direct all correspondence to: Natan Gadoth, MD, Department of Neurology, Meir General Hospital, Kfar-Saba 44281, Israel; Tel: 972-9-747-2828; Fax: 972-9-747-1317; E-mail: firstname.lastname@example.org.