J7ournal ofNeurology, Neurosurgery, and Psychiatry 1995;59:221-224 221 J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.59.3.221 on 1 September 1995. Downloaded from

gOU7ouZI of NEUROLOGY NEUROSURGERY & PSYCHIATRY

Editorial

The narcoleptic syndrome

The two main clinical features of the narcoleptic syn- is a disorder of unknown aetiology, drome, excessive daytime sleepiness and collapse with which is characterised by excessive sleepiness laughter or emotion, were recognised by the French that typically is associated with and neuropsychiatrist Gelineau in 1880.1 He described a other REM phenomena such as sleep Parisian wine merchant who had many sleep attacks each paralysis and hypnagogic hallucinations. day and episodic muscle weakness at the sight of a grotesque person, or with the excitement of being dealt a In this classification or hypnagogic good hand at cards. Four characteristic features of this hallucinations (as well as automatic behaviour and a disorder are now recognised: day sleep attacks, weakness disrupted major sleep episode) may replace cataplexy for with laughter, sleep paralysis, and socalled "hypnagogic diagnostic purposes. The classification outlines severity hallucinations".2 These symptoms can occur in different criteria (mild, moderate, severe) but not level of diagnos- combinations, and the presence of all is not essential for tic certainty (possible, probable, definite). In practice, the diagnosis. clinical diagnosis of the narcoleptic syndrome depends on The narcoleptic syndrome is life long. Remissions are the recognition of the pathological, not physiological, exceptional but the condition does not affect life span. nature of daytime sleepiness and weakness with laughter Excessive daytime sleepiness may slowly worsen and or intense emotion, as shown by the frequency, severity, cataplexy improve with increasing age, but this is contro- and persistence of these symptoms. versial. The first symptoms usually present at the age of Laboratory investigations have been used in the past 15 to 25, although cases under 5 have been described. decade both to establish and to support diagnosis. These The narcoleptic syndrome results in considerable include the multiple sleep latency test to show a mean disability, as great or greater than that caused by sleep latency of less than seven minutes. Facilitation of epilepsy. Problems at school, work, and in the home are rapid eye movement (REM) sleep is shown by the occur-

frequent and there is a high road traffic accident rate. rence of two or more REM periods in a five http://jnnp.bmj.com/ Interpersonal problems, separation, and divorce are com- protocol. will show frequent awak- mon.3 4 Prevalence estimates vary from 26 cases per enings and fragmented REM periods during night 100 000 in Finland to 590 cases per 100 000 in Japan.5 6 sleep.'0 Human leucocyte antigen testing shows HIA There may be a total of 20 000 cases in the United DR2 positivity in 98% of patients with the narcoleptic Kingdom. At present many of these go unrecognised. A syndrome. The most specific marker in Caucasian and simple effective screening programme needs to be developed. black patients is DQB1 0602.1" These investigations are Several problems in determining the different causes of sometimes considered essential for diagnosis and used excessive daytime sleepiness are due to inexact terminol- as a gold standard to separate the narcoleptic syndrome on September 24, 2021 by guest. Protected copyright. ogy. In the past the term narcolepsy has been used to from other sleep disorders. As with many if not all cover several different forms of sleepiness. This usage has of the clinical features, however, sleep laboratory findings resulted in occasional diagnostic confusion. Narcolepsy- are non-specific. They may be seen in other conditions, that is, sleep attack-is not an exact synonym for the nar- and HLA DR2 is found in 20-30% of normal subjects. coleptic syndrome and the term narcolepsy should not be In particular, routine electroencephalograms are normal used to refer, for example, to the condition of idiopathic other than showing considerable drowsiness. . The term cataplexy has proved difficult to The main difficulty in the diagnosis of the narcoleptic define. Descriptions such as "to strike down with terror", syndrome occurs when cataplexy is not present, or when "the hypnotic state in animals while shamming dead", the history of this is equivocal. The gap between the and "stupefaction",8 do not match the condition reported onset of daytime sleepiness and cataplexy is usually short, by Gelineau and do not stress the importance around two years. In most cases, both symptoms are pre- of the specific trigger factors of muscular weakness sent at the time of initial presentation. A gap of 40 years that he described. The International Classification of between the presentation of daytime sleepiness, and that Sleep Disorders more correctly characterises cataplexy by of cataplexy, has been recorded. Early diagnosis may be sudden loss of bilateral muscle tone provoked by strong impossible here. The accessory symptoms of sleep paraly- emotion9. sis and "hypnagogic hallucinations" can be difficult to The 1990 International Classification of Sleep evaluate and may have little or no diagnostic value (see Disorders defines the essential features of "narcolepsy" later). The situation is complicated when features of as follows: other sleep-wake disorders such as obstructive sleep 222 Parkes, Clift, Dahlitz, Chen, Dunn J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.59.3.221 on 1 September 1995. Downloaded from apnoea are present. In our experience, and that of others, one with a high carbohydrate content. This trough is this occurs in about 10-15% of elderly male patients with found in narcoleptic patients as well as normal subjects the narcoleptic syndrome.'2 Alternatively, in those cases although the overall level of daytime sleepiness, as deter- where only a single symptom is present-and all authors mined by the Epworth sleepiness scale, is five times who have studied many "narcoleptic" patients agree that greater in the first than the second group. Drowsiness some patients have nothing but sleep attacks-the diag- levels are overall slightly higher in patients with the nar- nostic state may be uncertain. The term"monosympto- coleptic syndrome than those with other causes of exces- matic" narcolepsy-that is, excessive daytime sleepiness, sive daytime sleepiness, including symptomatic or cataplexy, but not both symptoms-has been used to obstructive sleep apnoea and "idiopathic" hypersomnia. describe a possible forme fruste of the narcoleptic syn- Cataplexy is difficult to define, to measure, and to drome. This concept may be correct, particularly in cases separate clearly from weakness with laughter. As stressed where excessive daytime sleepiness is accompanied by earlier, a postural atonia rating scale, based on clinical REM sleep at sleep onset. In practice, however, some observations about cataplexy, is useful to determine the degree of diagnostic uncertainty often remains. The propensity to loss of muscle tone with sudden intense problem of diagnosis is highlighted in familial cases. emotion including anger and surprise as well as laughter. Between 10-25% of patients with both excessive daytime In our experience such a rating scale will differentiate sleepiness and weakness with laughter report a parent, patients with the narcoleptic syndrome from normal sub- brother, sister, or offspring with one symptom (usually jects and, in the narcoleptic syndrome, the propensity to excessive daytime sleepiness) alone. Here it may be loss of muscle tone with laughter (high) and noise (low) difficult to separate normal from pathological findings. is opposite to that found in normal subjects. Patients Subjectively reported sleep-wake habits need to be with the narcoleptic syndrome often report that cataplexy documented and quantified. Several simple methods is most severe when they are tired, and patients with the have been developed, with sleep-wake diaries, sleep logs, highest degree of sleepiness also have the highest propen- and rating scales. 13,14 These include the Stanford and the sity to cataplexy. Epworth sleepiness scales."5 16 Sleep patterns as shown by What is cataplexy? Cataplexy has some resemblance to many of these subjective scales conform quite well with drop attacks, sudden brief falls that occur without any objective measures of sleepiness as shown by the mutiple warning. By contrast with cataplexy, drop atacks are not sleep latency test, or with rest-activitiy cycles as measured triggered and are often associated with cardiac or by activity monitor recording.'5 17 Also, in many of these cerebrovascular disease. Other distinctive clincal features scales, test-retest reliability after more than a year is include autonomic symptoms and signs, dizziness, or ver- good. Sleepiness scales can measure different areas such tigo. These do not occur in cataplexy unless triggered by as attack duration, ("How long do your sleep attacks the emotional stimulus that provokes the attack. As well last?"), and attack propensity ("How likely are you to fall as a superficial resemblance to drop attacks, cataplexy asleep while driving?"). The recently developed has some relation with physiological weakness with Ullanlinna narcolepsy scale includes items about sleep laughter. The progression of motor atonia and paralysis latency ("How fast do you usually fall asleep in the in cataplexy, starting with weakness of the face, mouth, evening?"), and sleep attack frequency ("Do you sleep and neck, spreading to the trunk and limbs, and some- during the day, take , do you fall asleep unintention- times ending in a fall, rarely, if ever, however, accompa- ally during the day?")).18 Using this scale and combining nies laughter in normal subjects. The range of emotional subjective measures of abnormal sleep tendency with fre- factors that triggers cataplexy is wide, although pleasure quency estimates of loss of muscle tone with emotion, is more potent than anger. Some emotional factors, such patients with the narcoleptic syndrome can be reliably as anger and fear, commonly result in an increase in http://jnnp.bmj.com/ distinguished from normal subjects and those with other muscle tone in normal subjects, but a decrease in the forms of excessive daytime sleepiness. narcoleptic syndrome. The behavioural and physiological The use of a simple cataplexy rating scale, based, correlates of cataplexy are different from those of normal for example, on attack propensity ("How likely are you laughter. In cataplexy, behavioural features of REM sleep to lose muscle tone or fall down when you laugh?"), can be seen. These include phasic muscle jerking around as well as on attack characteristics ("Do the muscles the face and, particularly, around the mouth. This is

around your mouth jerk when you laugh?") may help sometimes accompanied by limb jerking and eye move- on September 24, 2021 by guest. Protected copyright. diagnosis. All these scales are useful in evaluation of ments. These features are reported by, or can be seen in, treatment. 70% of our patients during an attack of cataplexy. In Excessive daytime sleepiness is found in 0 3-4% of the cataplexy, but not in physiological weakness with laugh- population. Most of these cases are not due to the nar- ter, atonia may be accompanied by laboratory evidence coleptic syndrome but to more common conditions of REM sleep, occasional mentation, and rapid including chronic and obstructive sleep transition to sleep accompanied by abolition of the ten- apnoea. don jerks and evidence of strong postsynaptic inhibition The Epworth sleepiness scale grades sleep propensity of motor neurons as shown electrophysiologically by in eight common situations (sitting reading; watching diminished H reflexes and F responses. We conclude that television; sitting in a public place; as a car passenger; cataplexy is different not only in degree, but also in lying down for afternoon rest; sitting talking; sitting after nature, from weakness with laughter. lunch; in a car stopped in traffic). Most patients with the Many patients with the narcoleptic syndrome report narcoleptic syndrome report a very high propensity to fall that the anticipation of laughter, rather than laughter asleep in each of these situations, the highest propensity itself, is the essential factor that provokes cataplexy. being when lying down for afternoon rest, the lowest in a Cataplexy may be defined as laughter induced loss of car stopped in traffic. This distribution corresponds well facial and jaw control, spreading to involve the trunk and with the monotony level of each different situation and is other body areas, accompanied by muscle jerking around similar to that found in normal subjects. the mouth. These episodes are sometimes portrayed as Sleep propensity varies with the time of day. There is a "jelly" attacks. This description may help patients to trough of vigilance at afternoon siesta time, and sleep identify their somewhat unusual symptoms, and also to tendency may increase after a heavy meal, particularly differentiate these from epilepsy. Cataplexy is probably Editorial 223 J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.59.3.221 on 1 September 1995. Downloaded from unique to the narcoleptic syndrome although a, closely reported occasionally by 5% of normal subjects and, in analogous condition has been reported in children with our experience, by about the same number of patients the Prader-Willi syndrome and with Niemann-Pick with the narcoleptic syndrome. By contrast, reports of disease type C. Here the presentation of apparent cata- frequent, egocentric, scene shifting, and often frightening plexy may be subtly different from that in the narcoleptic vivid are much more common in patients with syndrome, with different trigger factors, and without defi- the narcoleptic syndrome than in normal subjects. The nite evidence of REM sleep related activity during usual estimate of dream frequency in normal subjects is attacks. 19 20 around three to four times a week, whereas patients with Sleep paralysis, the inability to move during the transi- the narcoleptic syndrome may report dreaming 20 or tion between sleep and wakefulness, is often considered a more times each week. Dreams occur both in the day, rare phenomenon that only occurs as part of the narco- bordering sleep episodes, and at night. In the narcoleptic leptic syndrome and examples of which in other con- syndrome dreams occur at sleep onset and this unusual ditions are not convincing. This notion is false and sleep dream timing, not the dream content, is the important paralysis has been increasingly recognised to be a diagnostic determinant, reflecting sleep onset REM common isolated disorder. This may occur after tired- activity. ness or irregular sleep, sleep deprivation, or psychological How often is the diagnosis of the narcoleptic syndrome stress. The correct view of sleep paralysis is that the missed? We have noted that diagnosis takers more than narcoleptic syndrome is only one of very many causes. five years to achieve in at least one third of patients Sleep paralysis has a wide prevalence rate of between 5% eventually referred to a clinic. In a survey and 62%, often with a strong familial tendency.2' of 132 patients with definite cataplexy and excessive day- Despite this high prevalence only a few subjects seek time sleepiness attending the Maudsley Hospital sleep medical advice and the condition often goes unreported disorders clinic, the period from symptom onset to or unrecognised despite the severe terror associated with disease diagnosis was reported as follows: less than one some if not most attacks. We have found that occasional year, 43 (33%); 1-5 years, 42 (33%); 6-10 years, 16 sleep paralysis is described by two thirds of patients with (12%); 11-20 years, 12 (9%); 21-40 years, 16 (12%); excessive daytime sleepiness and cataplexy as compared 41-60 years, two (2%); and over 60 years, one (1%). with 5-10% of patients with symptomatic obstructive Before diagnosis, many of these patients were considered sleep apnoea. By contrast, sleep paralysis is very un- school failures, lazy, bored, or work shy. The effect of common in . Sleep paralysis, cata- this can be devastating for a patient who misses out on plexy, and REM sleep have many features in common work training or university, cannot hold a job, and and it is surprising that, whereas cataplexy may be develops a low level of self esteem. Clinical neurologists unique to the narcoleptic syndrome, sleep paralysis in the report an average total practice occurs in many different circumstances. of between 10 and 20 "narcoleptics". If most patients The major complaint in the narcoleptic syndrome is with a diagnosis of the narcoleptic syndrome are reviewed usually one of daytime sleepiness, not night time insom- at some stage by a neurologist, only one quarter of the nia. Mitler and his colleagues have, however, stressed the total estimated 20 000 patients in the United Kingdom importance of disturbed night sleep, adding this symp- may be recognised. tom to the key diagnostic features of the syndrome.22 The How accurate is the diagnosis of the narcoleptic syn- usual complaints are of light sleep, non-refreshing sleep, drome? At present, the gold standard for diagnosis, as and frequent arousals. Indeed it has been claimed that discussed earlier, is based on the presence of characteris- some patients with "narcolepsy" wake over 300 times tic clinical features, supported by specific sleep laboratory each night.2' is thus a common but much findings. http://jnnp.bmj.com/ neglected symptom. The time taken to fall asleep at When all the typical features are present, the level of night, however, is only half that of normal subjects. We diagnostic accuracy is likely to be high despite the have found that this short sleep latency is a useful diag- absence of any definite disease marker such as a unique nostic feature. Total night sleep time is about 20 to 40 narcolepsy "gene", or any distinctive pathological or bio- minutes less in patients with the narcoleptic syndrome chemical finding. That diagnosis can be very accurate is than in normal subjects, with three to four times more shown by the many studies confirming Honda's original recalled arousals. The complaint of insomnia increases finding of the tight association between the narcoleptic with both disease duration and with age. In some cases it syndrome phenotype and HLA DR2.26 This finding on September 24, 2021 by guest. Protected copyright. is possible that insomnia is increased by daytime central would not have been expected had the disease pheontype stimulant drug treatment. A few patients report entirely not been exact. normal night sleep habits. We have seen an interesting By contrast with research patients, many, if not most, variant of the narcoleptic syndrome where the primary of the total population of patients with the narcoleptic complaint was of night time, not daytime, symptoms, syndrome do not attend a sleep disorder clinic and do despite the presence of daytime sleepiness and weakness not have laboratory investigations. To study the level of with laughter. diagnostic accuracy without specialised investigations, we Motor dyscontrol is a characteristic feature of the have compared the pattern of disease presentation and narcoleptic syndrome with sleep talking, sleep walking, severity in three groups of patients presenting with the and muscle jerking. All these features are several times complaint of excessive daytime sleepiness accompanied more common in patients with the narcoleptic syndrome by excessive weakness with laughter. These groups com- that in normal subjects. We interpret this finding as a prised those with a definite diagnosis of the narcoleptic consequence of fragmented night sleep. Interestingly this syndrome, confirmed by multiple sleep latency testing, motordorrdr may start in childhood, before the onset polysomnography, and HLA typing, those with a of daytime sleepiness or cataplexy.24 probable diagnosis, as shown by sleep clinic review but The term hypnagogic hallucinations was introduced by without laboratory investigation, and those with a Maury to describe a specific form of mental experience, possible diagnosis, in cases presenting before sleep clinic with visual decorative or fanciful patterns, distinct from review or investigation. The clinical characteristics were true dream mentation, and occurring at sleep onset.25 identical in these three groups. We interpret this finding This type of sleep onset experience is uncommon. It is as suggesting that sleep clinic review and laboratory 224 Parkes, Clift, Dahlitz, Chen, Dunn J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.59.3.221 on 1 September 1995. Downloaded from

may not be essential for clinical diagnosis J D PARKES investigations S J CLIFT in the presence of unequivocal cataplexy. Such a review M J DAHLITZ is essential in uncertain cases, as well as in research University Department ofNeurology, studies. King's College School ofMedicine and Dentistry, London, UK S Y CHEN can be made A diagnosis of the narcoleptic syndrome Department ofPsychiatry, China Medical College Hospital, on clinical grounds'alone, without laboratory support in Taichung, Taiwan most cases, but despite well defined symptoms, the G DUNN Institute ofPsychiatry, or wrongly diagnosed. The Department ofBiostatistics and Computing, condition is often missed London, UK most characteristic symptom, on which diagnosis primarily depends, is cataplexy, not excessive daytime Correspondence to: Professor J D Parkes, Department of Neurology, Institute of Psychiatry, De Crespigny Park, London SE5 8AF, UK. sleepiness. The syndrome can only be diagnosed with certainty if cataplexy is present. Cataplexy can be defined as "jelly" attacks with laughter provoked loss of muscle 1 Gelineau JB. De la narcolepsie. Gaz H6p (Paris) 1880;53:626-8; 635-7. tone, often accompanied by facial muscle jerking. The 2 Daniels LE. Narcolepsy. Medicine (Baltimore) 1934;13:1-12. 3 Broughton WA, Broughton RJ. Psycholosocial impact of narcolepsy. Sleep idea of a narcoleptic tetrad, cataplexy and daytime sleepi- 1994;17:S45-59. ness accompanied by sleep paralysis and hypnagogic 4 Broughton R, Guberman A, Roberts J. Comparison of psycho-social effects of epilepsy and of narcolepsy/cataplexy: a controlled study. halluciations, needs revision. Sleep paralysis is common Epilepsia 1984;25:423-33. in the narcoleptic syndrome but is found also in many 5 Hublin C, Kaprio HJ, Partinen M, Koskenvuo M, Heikkila K, Koskimies S, Guilleminault C. The prevalence of narcolepsy: an epi- other sleep-wake disorders and is not a diagnostic equiva- demiological study of the Finnish twin cohort. Ann Neurol 1994; lent of cataplexy. True hypnagogic hallucinations are not 35:709-16. 6 Tashiro T, Konbayashi T, lijima S, Hishikawa Y. An epidemiological a specific feature of the narcoleptic syndrome. Here the study of narcolepsy in Japanese. _7 Sleep Res 1 993;S228. complaint is one of very frequent dream recall, with sleep 7 Parkes JD. Sleep and its disorders. London: WB Saunders, 1985;281. 8 The Concise Oxford English Dictioniary. 6th ed. Oxford: Clevedon, 1976: onset dream timing. Additional features include dis- 155. turbed night sleep, with a short sleep latency, multiple 9 International Classification of Sleep Disorders. Diagnostic antd coding manual. Diagnostic Classification Steering Committee. Thorpy MJ, arousals, reduced total night sleep time, and the com- Chairman, Rochester, Minnesota: American Sleep Disorders plaint of insomnia. Motor dyscontrol during night sleep Association, 1990:38-43. 10 Mitler MM, van den Hoed J, Carskadon MA, et al. REM sleep episodes is shown by phasic muscle jerking, sleep talking, and during the multiple sleep latency test in narcoleptic patients. sleep walking. These features may antedate other symp- Electoencephalogr Clin Neurophysiol 1979;46:479-8 1. 11 Mignot E, Lin X, Arrigoni J, et al. DQBI*0602 and DQA1*0102(DQ1) toms. The narcoleptic syndrome is a common and very are better markers than DR2 for narcolepsy in Caucasian and black disabling condition. An effective health screening pro- Americans. Sleep 1994;17:S60-67. 12 Guilleminault C, van den Hoed J, Mitler MM. Clinical overview of the gramme is needed. This could be based on a simple syndromes. In: Guilleminault C, Dement WC, eds. Sleep rating scale to allow clinical diagnosis, and aid in the apnea syndromes. New York: Alan R Liss, 1978:1-12. 13 McGhie A, Russell SM. The subjective assessment of normal sleep identification of cases that at present are missed. The patterns. _7 Metnt Sci 1962;108:642-54. fundamental requirement of any such scale is the certain 14 Monk TH, Reynolds CF, Kupfer DJ, et al. The Pittsburgh sleep diary. 7 Sleep Res 1994;3:111-20. recognition of cataplexy. 15 Hoddes E, Zarcone V, Smythe H, Dement W. Quantification of sleepi- Treatment of excessive daytime sleepiness in the ness; a new approach. Psychophysiology 1973;1O:431-6. 16 Johns MW. A new method for measuring daytime sleepiness: The narcoleptic syndrome depends at present on central stim- Epworths Sleepiness scale. Sleep 199 1;14:540-5. ulant drugs such as dexamphetamine, methyphenidate, 17 Johns MW. Reliability and factor analysis of the Epworth sleepiness scale. Sleep 1992;15:376-81. and mazindol. These all have limited efficacy, cause 18 Hublin C, Kaprio J, Partinen M, Koskenvuo M, Heikkila K. The frequent sympathomimetic side effects, and can lead to Ullanlinna narcolepsy scale: validation of a measure of symptoms in the narcoleptic syndrome. 7 Sleep Res 1994;3:52-9. tolerance. Cataplexy usually responds to serotonin re- 19 Helbing-Zwanenburg B, Mourtazaev MS, d'Amaro J, et al. HLA types in

uptake inhibitors including clomipramine and fluoxetine. the Prader-Willi syndrome. _7 Sleep Res 1993;2: 115. http://jnnp.bmj.com/ 20 Kandt RS, Emerson RG, Singer HS, Valle DI, Moser HW. Cataplexy in A combination of two different types of drug may be variant forms of Niemann-Pick disease. Ann Neurol 1982;12:284-8. needed for treatment but can lead to sweating, irritabil- 21 Dahlitz M, Parkes JD. Sleep paralysis. Lancet 1993;341:406-7. 22 Mitler MM, Hajdukovic R, Erman MK. Treatment of narcolepsy with ity, increase in appetite and weight, and impaired sexual methamphetamine. Sleep 1993;16:306-17. function. Functional neuroimaging studies indicate that 23 Jovanovic UJ. Der normale, abnorme und pathologische Schlaf Polygraphische Untersuchungen Vehr Dtsch Ges Inn Med. Munich: JH Bergmann, 1965: amphetamine causes a major increase in regional blood 807-19. flow in cortical areas after both visual and auditory stim- 24 Mayer G, Meier-Ewert K. Motor dyscontrol in sleep in narcoleptic patients (a life-long development?). _7 Sleep Res 1993;2:143-8. ulation. The advent of new central stimulant drugs such Les hallucinations ou les erreurs des sens dans 25 Maury A. hypnagogiques on September 24, 2021 by guest. Protected copyright. as modafinil, and other novel drugs which inhibit both l'etat intermediare entre la veille et le sommeil. Ann Mid Psychol 1848;11:26-40. noradrenaline and serotonin re-uptake should eventually 26 Honda Y, Juji T, eds. HLA and tnarcolepsy. Berlin: Springer-Verlag, improve the treatment of the narcoleptic syndrome. 1988:208.