neurological progress the neurobiology, diagnosis, and...

The Neurobiology, Diagnosis, andTreatment of Narcolepsy

Thomas E. Scammell, MD

Narcolepsy is a common cause of chronic sleepiness distinguished by intrusions into wakefulness of physiological aspectsof rapid eye movement sleep such as cataplexy and hallucinations. Recent advances provide compelling evidence thatnarcolepsy may be a neurodegenerative or autoimmune disorder resulting in a loss of hypothalamic neurons containingthe neuropeptide orexin (also known as hypocretin). Because orexin promotes wakefulness and inhibits rapid eye move-ment sleep, its absence may permit inappropriate transitions between wakefulness and sleep. These discoveries haveconsiderably improved our understanding of the neurobiology of sleep and should foster the development of rationaltreatments for a variety of sleep disorders.

Ann Neurol 2003;53:154–166

Narcolepsy was first described more than 100 yearsago, but until recently, the cause of this common sleepdisorder was largely unknown. Rare lesions of the pos-terior hypothalamus provided some hints, spurring vonEconomo to predict that, “It is very probable thoughnot yet proved, that the narcolepsy of Gelineau, West-phal, and Redlich has its primary cause in a yet un-known disease of that region.”1 This review touches onthe compelling, recent evidence that narcolepsy is a dis-order of neurons in the posterior hypothalamus thatproduce the neuropeptide orexin.

Clinical Manifestations of NarcolepsyNarcolepsy is characterized by chronic sleepiness and amarked disorganization of sleep/wake behavior. It usu-ally begins with excessive daytime sleepiness and unin-tentional naps in the teens and early twenties, althoughsymptoms can begin in young childhood or after theage of 40 years.2,3 All narcoleptic subjects have chronicsleepiness, but the intensity varies across the day andbetween individuals. This sleepiness is most trouble-some during periods of inactivity and is often im-proved temporarily by a brief nap. As a consequence ofsleepiness, patients may report inattention, poor mem-ory, blurry vision, diplopia, and automatic behaviorssuch as driving without awareness.

Researchers have debated whether this sleepiness iscaused by increased sleep drive or an impaired arousal

system. The latter appears more likely because narco-leptic subjects have normal amounts of sleep over 24hours,4,5 but disrupted nighttime sleep is common innarcolepsy and may partially contribute to this chronicsleepiness. Narcoleptic patients with the greatest sleepdisturbance have more severe daytime sleepiness,6 buteven those with good nighttime sleep still have sub-stantial daytime sleepiness. Thus, the quality of night-time sleep is probably just a minor contributor to thechronic sleepiness of most patients.6,7

Narcoleptic subjects often have abnormal manifesta-tions of rapid eye movement (REM) sleep that intrudeinto wakefulness, including cataplexy, hypnagogic hal-lucinations, and sleep paralysis. Hypnagogic hallucina-tions are dream-like, often frightening hallucinationsthat typically occur with drowsiness or the onset ofsleep. These hallucinations are usually visual, with re-ports of seeing people or animals, but tactile, auditory,or even vestibular hallucinations such as a sense of sud-den falling are not uncommon. Sleep paralysis is pro-found weakness that occurs at the onset of sleep or onawakening. This is probably an intrusion of REM sleepparalysis into wakefulness, and can be associated with asensation of fear or suffocation. Hypnagogic hallucina-tions and sleep paralysis are not specific to narcolepsyand can be seen with other conditions of increasedsleep pressure such as chronic sleep deprivation or ob-

From the Department of Neurology, Beth Israel Deaconess MedicalCenter, Boston, MA.

Received May 14, 2002, and in revised form Aug 5 and Oct 7.Accepted for publication Oct 8, 2002.

Address correspondence to Dr Scammell, Department of Neurol-ogy, Beth Israel Deaconess Medical Center, Harvard Institutes ofMedicine, 77 Avenue Louis Pasteur, Boston, MA 02115.E-mail: [email protected]

NEUROLOGICAL PROGRESS

154 © 2003 Wiley-Liss, Inc.

structive sleep apnea, and occasionally in normal indi-viduals.8,9

Cataplexy is sudden muscle weakness brought on bystrong emotions, particularly joking, laughter, or an-ger.2,10,11 Most likely, this also is an intrusion of REMsleep atonia into wakefulness, but in contrast withsleep paralysis, cataplexy occurs almost exclusively innarcolepsy (see below). Sixty percent of narcolepticsubjects develop cataplexy, usually around the onset ofsleepiness or within 3 to 5 years.2,3 Severe episodesproduce bilateral, generalized weakness sufficient tocause a fall, although usually without injury. Other ep-isodes may be partial, affecting only the face, voice, ora limb. Consciousness is never impaired unless the pa-tient subsequently falls asleep or begins having hypna-gogic hallucinations. Episodes of cataplexy may beginwith clonic inhibitory movements leading to a fall, fol-lowed by a period of atonia and areflexia usually lastingless than 2 minutes12 (a video of cataplexy is avail-able at http://www-med.stanford.edu/school/Psychiatry/narcolepsy/). Many normal individuals report slightmuscle weakness with intense emotion, and duringlaughter the H-reflex is markedly reduced in controlsand in narcoleptic subjects.13,14 Cataplexy may resultfrom excessive activation of these same descending mo-tor inhibitory pathways that are active during strongemotions or REM sleep.15

Laboratory FindingsThe history alone is often very suggestive of narcolepsy,but a complete evaluation includes an overnight poly-somnogram followed by a Multiple Sleep Latency Test(MSLT).16 The polysomnogram helps evaluate sleepquality and excludes other causes of sleepiness such asobstructive sleep apnea, periodic limb movements ofsleep, or REM behavior disorder that are common innarcolepsy and may warrant specific treatment.6,17,18

This overnight study is followed the next day by aMSLT in which a patient is given four or five oppor-tunities to nap every 2 hours.19 Normal subjects fallasleep in approximately 10 to 15 minutes, but narco-leptic subjects often fall asleep in less than 5 minutes,providing objective evidence of their sleep propen-sity.20,21 The naps of narcoleptic subjects often includeREM sleep,22 and the occurrence of these sleep-onsetREM periods (SOREMs) in two or more naps ishighly suggestive of narcolepsy. However, SOREMscan be seen with other disorders that increase REMsleep pressure such as sleep deprivation, sleep apnea,or withdrawal from REM sleep–suppressing medica-tions.23

Differential DiagnosisIn the absence of other neurological deficits, the com-bination of chronic sleepiness with cataplexy or morethan two SOREMs is almost always caused by narco-

lepsy. Sleepiness, cataplexy, and SOREMs can occur inuncommon diseases such as Prader–Willi syndrome,Niemann–Pick disease type C, and Norrie disease,24–26

but all these individuals have mental retardation andother obvious neurological deficits. In patients withoutcataplexy, diagnosing narcolepsy can be difficult, andone should consider other causes of sleepiness such assleep apnea, periodic limb movements of sleep, insuf-ficient sleep, or the effects of sedating medications.21

Narcolepsy without cataplexy may overlap with idio-pathic hypersomnia, a heterogeneous disorder ofchronic sleepiness.27 By definition, patients with idio-pathic hypersomnia lack cataplexy and have less thantwo SOREMs on the MSLT. Some of these individualshave deep, excessively long periods of sleep, difficultywaking from sleep, and long unrefreshing naps, butmany have symptoms similar to narcolepsy.8,27

Secondary NarcolepsyOn infrequent occasions, narcolepsy occurs as a conse-quence of a focal central nervous system lesion. Theselesions almost always involve the posterior hypothala-mus, and tumors are the most common cause.28–31

Strokes or arteriovenous malformations of the hypo-thalamus are rare but can cause narcolepsy.30,32,33 Hy-pothalamic sarcoidosis has caused narcolepsy in at leasttwo cases, although neither had cataplexy.29,30 Posten-cephalitic sleepiness resembling narcolepsy was com-mon in the 1920s,34–37 but we know of no similarepidemic since then. A recent report described patientswith paraneoplastic anti–Ma antibodies who have hy-pothalamic inflammation, sleepiness, and cataplexy,but polysomnograms were not reported.38 Melberg andcolleagues have described a Swedish family with auto-somal dominant cerebellar ataxia, deafness, and narco-lepsy.39 Affected members gradually develop chronicsleepiness and cataplexy in young adulthood along withenlargement of the third ventricle suggestive of hypo-thalamic atrophy.

Strokes or tumors of the brainstem can produce iso-lated cataplexy but rarely produce the full narcolepsysyndrome.29,40–42 Most likely, these lesions affect de-scending pathways from the hypothalamus that regu-late motor tone but not arousal.

The association of narcolepsy with head injury andmultiple sclerosis is more controversial. Most peoplewith hypersomnolence after closed head injury do nothave narcolepsy,43 but some patients with narcolepsyreport that their symptoms began after a head injury.On several occasions, narcolepsy has occurred withmultiple sclerosis,30,44–47 but it is usually unclearwhether the narcolepsy was caused by a focal plaque.Although these associations are rare, it remains possible

Scammell: The Neurobiology of Narcolepsy 155

that central nervous system inflammation or injurycould increase the risk of developing narcolepsy.

Epidemiology and Genetics of IdiopathicNarcolepsyNarcolepsy occurs in approximately 1 in 2,000 individ-uals, and most cases are sporadic.3,48 Genetic factorsplay an important role because first-degree relativeshave a 40-fold increased risk of developing narco-lepsy.49,50 Nevertheless, genetics are only a partial in-fluence because even among monozygotic twins inwhich one has narcolepsy, the second twin is affectedonly approximately 25% of the time.50

The most robust of these genetic factors are specifichuman leukocyte antigens (HLAs). Early studies usinglow-resolution serological techniques found that narco-lepsy is associated with two HLA class II antigens,DR2 and DQ1. This association varies with ethnicity;DR2 is found in 100% of Japanese, 90 to 95% ofwhites, and 60% of African American narcoleptic sub-jects with cataplexy.49,51,52 High-resolution mappingof these regions demonstrated the existence of numer-ous alleles, and DQB1*0602, a DQ1 subtype allele, isstrongly linked with narcolepsy. DQB1*0602 is foundin 88 to 98% of narcoleptic subjects with unambigu-ous cataplexy, whereas it is found in only 12% of whiteAmerican and 38% of African American controls.53,54

Testing for DQB1*0602 can be helpful, especially inpatients with questionable cataplexy, but this allele ishardly predictive of narcolepsy because greater than99% of DQB1*0602-positive individuals do not havenarcolepsy.

DQB1*0602 may influence the severity of narco-lepsy. In part, this may be a direct effect on the ex-pression of REM sleep because, even among normaladults, DQB1*0602-positive subjects enter REM sleepmore quickly.55 DQB1*0602 also may affect the sever-ity of narcolepsy because DQB1*0602-positive narco-leptic subjects have longer episodes of cataplexy, morenocturnal sleep disruption, and more daytime sleepi-ness.2 Whether this more severe phenotype is causedby more extensive neuropathology or altered regulationof REM sleep remains unknown.

As in some other HLA-linked diseases, tumor necro-sis factor-� (TNF-�) has been implicated in narco-lepsy. Polymorphisms in the TNF-� and TNF receptor2 genes are associated with narcolepsy in Japan,56–58

with especially high susceptibility when these alleles arepresent in both genes.58 Plasma TNF-� levels are nor-mal in narcolepsy,59 but these polymorphisms may stillaffect TNF-� signaling, thus contributing to an in-flammatory or degenerative process. Because TNF-�promotes sleep in laboratory animals,60 altered TNFsignaling also might directly influence the activity ofsleep/wake regulatory pathways in the brain.

Neurobiological Control of Sleep andWakefulnessSubstantial new evidence indicates that the neuropep-tide orexin plays a critical role in the neurobiology ofnarcolepsy. To understand this role, it is helpful to re-view some of the other pathways involved in the con-trol of wakefulness, REM sleep, and non-REM sleep(for detailed reviews, see Saper and colleagues61 andJones62). Normal wakefulness requires the coordinatedactivity of several ascending arousal systems located inthe rostral brainstem and posterior hypothalamus (Fig1). Magoun and colleagues demonstrated that brain-stem transections below the midpons preserved wake-fulness whereas those between the midbrain and ponsproduced coma.63 These and other studies helpedidentify a diverse collection of projections within thebrainstem tegmentum that innervate the thalamus andhypothalamus. Some of these fibers originate in thedorsal pons in cholinergic neurons of the laterodorsaland pedunculopontine tegmental nuclei (LDT/PPT).These neurons project to several thalamic nuclei in-cluding the reticular nucleus, a critical regulator ofthalamocortical transmission. During wakefulness, thischolinergic input switches thalamic neurons from aninherent oscillatory pattern seen during non-REMsleep to a more active, “transmission mode,” facilitating

Fig 1. Major nuclei involved in the control of sleep andwakefulness. Cell groups that promote wakefulness are foundbetween the pons and posterior hypothalamus. Cholinergic(ACh) neurons in the laterodorsal and pedunculopontine teg-mental nuclei (LDT/PPT) innervate the thalamus, promotingthe flow of information to and from the cortex. Aminergicneurons in the locus coeruleus (LC), raphe nuclei (Raphe),and tuberomamillary nucleus (TMN) project diffusely through-out the forebrain increasing neuronal activity. These arousalregions are inhibited by sleep-promoting neurons of the ventro-lateral preoptic area (VLPO). NE � norepinephrine; 5HT �serotonin; HIST � histamine.

156 Annals of Neurology Vol 53 No 2 February 2003

the flow of information to and from the cortex.64,65

During deep non-REM sleep, these cholinergic neu-rons are much less active, thus decreasing thalamic andcortical activity. Attention and full cortical activationdepend on a separate population of cholinergic neuronsin the basal forebrain, but it is unclear whether thesecells are required for the production of wakefulness it-self.66

Wake-promoting aminergic neurons in the rostralbrainstem also project through the pontine and mid-brain tegmentum. These include noradrenergic neu-rons of the locus coeruleus (LC) and serotonergic neu-rons of the dorsal raphe (DR) nuclei. Histaminergicneurons of the tuberomamillary nucleus (TMN) in theventral posterior hypothalamus also contribute to thispathway. All three of these aminergic nuclei diffuselyinnervate and activate much of the forebrain, firingmost rapidly during wakefulness, slower in non-REMsleep, and then becoming silent in REM sleep.67–69

Dopaminergic signaling also may promote wakeful-ness because dopamine antagonists can producesleep,70 and amphetamines primarily increase wakeful-ness by increasing the extracellular concentrations ofdopamine.71,72 The source of this dopaminergic signal-ing is unclear because the mean firing rates of neuronsin the substantia nigra and ventral tegmental area donot vary with behavioral state.73,74 However, lesions ofdopaminergic cells in the ventral midbrain can reducewakefulness,75–77 and one population of dopaminergicneurons in the ventral periaqueductal gray is activeduring wakefulness,77 perhaps serving as a critical sitefor dopaminergic effects on wakefulness.

These cholinergic and aminergic regions interact toproduce REM sleep.78 A distinct population of cholin-ergic neurons in the LDT/PPT activates the thalamusduring REM sleep, producing cortical desynchrony.These REM-active cells also generate profound atoniathrough a polysynaptic, descending pathway that usesglycine to hyperpolarize � motor neurons. Noradren-ergic and serotonergic afferents from the LC and DRinhibit the REM sleep–producing neurons duringwakefulness and non-REM sleep,79,80 but during REMsleep, the LC and DR become inactive, thus disinhib-iting the REM-generating neurons. This aminergic in-hibition of REM sleep is commonly encountered in thesleep laboratory; patients taking selective serotonin re-uptake inhibitors usually have a marked decrease inREM sleep.

Non-REM sleep is produced by neurons in the pre-optic area and brainstem. The dorsal vagal complex,raphe nuclei, and ventral midbrain contain some non-REM–generating cells,81,82 but the ventrolateral preop-tic area (VLPO) is the best characterized sleep-producing region. VLPO neurons are active duringnon-REM sleep83,84 and are necessary for sleep becauselesions of the VLPO produce marked insomnia.85

These GABA-containing neurons innervate and proba-bly inhibit neurons in the TMN, LC, and DR.86 Inturn, VLPO neurons are inhibited by amines,87 thuscreating a reciprocally inhibitory circuit in which wake-and sleep-promoting regions inhibit one another.61

This type of mutually inhibitory circuit should dem-onstrate bistability, being most stable during full wake-fulness or sleep. However, these elements alone couldproduce a system with low thresholds to transition be-tween behavioral states. This may be the fundamentalproblem in narcolepsy in which irresistible sleep, intru-sions of REM-like phenomena, and nocturnal awaken-ings result from abnormally low thresholds to transi-tion between wakefulness, REM, and non-REM sleep.The importance of these thresholds is evident duringperiods of prolonged wakefulness when an individualmust remain awake despite a gradually increasing sleepdrive. An additional stabilizing influence is needed tohelp maintain wakefulness and other states.

Neurobiology of Orexin/HypocretinJust a few years ago, two research groups simulta-neously discovered a new neuropeptide that may pro-vide this stabilizing influence (for review, see Willieand colleagues88). One group named this peptideorexin for its presumed role in appetite whereas an-other group named it hypocretin for its possible resem-blance to secretin.89,90 This peptide exists in twoforms, orexin-A and orexin-B, which are derived fromthe same precursor by proteolytic processing and arecolocalized in the same neurons. Orexin-containingneurons are found only in the posterior and lateral hy-pothalamus,89–91 but they project widely throughoutthe central nervous system, innervating the aminergicand cholinergic regions that promote wakefulness92,93

(Fig 2). These targets contain high levels of the twoorexin receptors, ox1 and ox2 (also known ashypocretin-1 and -2 receptors).94 The ox1 and ox2 re-ceptors are coupled to G proteins, and ligand bindinggenerally has excitatory effects.

Recent animal studies provide considerable evidencethat orexin promotes wakefulness and inhibits REMsleep. The orexin neurons are active during wakefulnessas indicated by the expression of Fos,95 and extracellu-lar concentrations of orexin are higher during periodsof wakefulness.96 Electrophysiological recordings fromthe orexin neuron region identified many wake-activeneurons, with particularly high firing rates when an an-imal is physically active.97 In vitro, orexin excites LCand other aminergic neurons.98–102 In vivo, injectionsof orexin into the lateral ventricles or near specificarousal regions such as the LC increase wakefulnessand markedly suppress REM sleep.98,103 Currently, itis only partially known which regions are necessary forthe wake-promoting effects of orexin, but the histamin-ergic neurons of the TMN may play an essential role.


In normal mice, infusion of orexin near the TMN in-creases wakefulness, but orexin does not increase wake-fulness in mice lacking the histamine H1 receptor.104

In addition, narcoleptic dogs have low cortical concen-trations of histamine,105 and a histamine antagonistpartially blocks the wake-promoting effects of orexin.100

Suppression of REM sleep by orexin also may occurthrough activation of these aminergic areas that theninhibit the REM-producing neurons of the LDT/PPT.

Impaired Orexin Signaling Causes NarcolepsyFour years ago, two laboratories simultaneously discov-ered that the orexin system plays a critical role in nar-colepsy. After a extensive linkage analysis of dogs withautosomal recessive narcolepsy, Lin and colleaguesfound exon-skipping mutations in the ox2 receptorgene that probably result in a nonfunctional recep-tor.106 Meanwhile, Yanagisawa’s group found thatorexin knockout mice or mice lacking the orexin re-ceptors have a phenotype strongly resembling humannarcolepsy, with brief periods of wakefulness, frequenttransitions into REM sleep, and behavior resemblingcataplexy.93,107,108 Mice lacking the ox2 receptor havea more severe phenotype than those lacking the ox1receptor, providing additional evidence that the ox2 re-ceptor may play a more prominent role in the controlof sleep/wake behavior.

These studies demonstrated that congenitally im-paired orexin signaling can produce symptoms of nar-colepsy, but human narcolepsy is usually an acquireddisease. To create a mouse model similar to humannarcolepsy, Hara and colleagues produced mice inwhich the orexin promoter drives the expression ofataxin-3, a truncated Machado–Joseph disease geneproduct that induces apoptosis in neurons.109 Theseorexin/ataxin-3 mice have a normal number of orexinneurons at birth, but as they approach adulthood, theorexin neurons degenerate. This acquired loss of orexinneurons roughly parallels the typical timing of humannarcolepsy, and, just as in humans, these mice cannotremain awake for long periods, rapidly enter REMsleep, and have behavioral arrests resembling cataplexy.

Mignot, Nishino, and colleagues rapidly confirmedthat most narcoleptic subjects with cataplexy have im-paired orexin signaling.110 Approximately 90% of nar-coleptic subjects with cataplexy have no detectableorexin-A in their cerebrospinal fluid (CSF) even withina few months of symptom onset.110–115 In contrast,nearly all narcoleptic subjects without cataplexy havenormal CSF concentrations of orexin.113,115 Undetect-ably low levels of orexin are not seen in other neuro-logical diseases, except for some uncommon casesof severe Guillain–Barre syndrome or myxedemacoma.111,115 Orexin levels may be low though not ab-sent in some cases of secondary narcolepsy probablybecause of partial lesions of the orexin neurons andtheir projections to arousal regions.31,32,115,116 In somecases of familial narcolepsy, orexin levels are normal inyounger patients but low in older patients, suggesting agradual failure to produce orexin.115

Neuropathology of NarcolepsyTwo landmark studies have shown that this markeddecrease in CSF orexin most likely results from a lossof the orexin-producing neurons. Using in situ hybrid-ization, Peyron and colleagues found no detectable

Fig 2. (A) Orexin-containing neurons of the posterior andlateral hypothalamus heavily innervate and excite aminergicregions that promote wakefulness such as the tuberomamillarynucleus (TMN), raphe, and locus coeruleus (LC). These amin-ergic regions also inhibit rapid eye movement (REM)–produc-ing neurons of the LDT/PPT. Decreases in orexin and aminesignaling allow the occurrence of REM sleep, with thalamicactivation and dreaming, and paralysis through intense inhibi-tion of � motor neurons. (B) Innervation of the human locuscoeruleus by orexin-containing fibers. Numerous black, orexin-immunoreactive boutons innervate a proximal dendrite of abrown, tyrosine hydroxylase–immunoreactive locus coeruleusneuron. Scale bar � 50�m.


prepro-orexin mRNA in the hypothalamus of narcolep-tic subjects with cataplexy.117 Thannickal and col-leagues found a 90% reduction in the number oforexin immunoreactive neurons.118 This elimination oforexin-producing neurons is highly selective becauseboth studies showed no reduction in the number ofadjacent neurons containing melanin-concentratinghormone. These two studies included a total of fivebrains from narcoleptic subjects with cataplexy and onewithout, and the case without cataplexy had the great-est number of remaining orexin neurons. Further stud-ies may determine whether narcoleptic subjects withcataplexy simply have a greater loss of orexin neuronsthan those without cataplexy.

The process that causes this loss of orexin is un-known. In general, narcoleptic subjects lack mutationsin the genes coding for orexin or its receptors,117,119,120

although one individual with severe symptoms in in-fancy was found to have a mutation in the prepro-orexin gene that probably produces inappropriate traf-ficking of the peptide and possible degeneration of theorexin neurons.117 Alternatively, impaired transcriptionof the prepro-orexin gene or anti-orexin antibodiescould produce a loss of orexin without a loss of theorexin neurons.

Many researchers have proposed that a selective lossof the orexin neurons could be caused by an autoim-mune or neurodegenerative process, but, thus far, thereis little evidence to support these ideas. Despite the as-sociation of narcolepsy with specific HLA and TNFalleles, no changes in immune function have beenidentified.59,117,121 Neuroimaging studies of idiopathicnarcolepsy generally have found no abnormalities.122–125

The CSF of narcoleptic subjects lacks increased proteinor oligoclonal bands, and their serum lacks autoanti-bodies including any against orexin126–128 (E. Mignot,personal communication). Microglia and TNF-�mRNA are not increased in the hypothalami of narco-leptic subjects examined decades after the onset ofsymptoms,117 suggesting that there is no ongoing in-flammation. However, the number of astrocytes maybe moderately increased in the orexin neuron re-gion118,129 (but see Peyron and colleagues117). If sup-ported by additional studies, this gliosis would beconsistent with a neurodegenerative or inflammatoryprocess.

Timing and Coordination of Sleep/WakeBehavior with Other Hypothalamic FunctionsUnstable sleep/wake behavior is the hallmark of narco-lepsy, but narcoleptic subjects also have poor circadiantiming of sleep and wakefulness.5,130 In normal indi-viduals, wakefulness is strongly promoted throughmuch of the day, and REM sleep occurs mainly be-tween 2 and 8 AM.131 Narcoleptic subjects have diffi-culty maintaining wakefulness, and their naps often in-

clude bouts of REM sleep, regardless of the time ofday.130,132 Most likely, this is not caused by a simpledisinhibition of sleep because narcoleptic subjects havenormal amounts of sleep over 24 hours.4,5,130 In addi-tion, this marked attenuation of the normal sleep/wakerhythm is not caused by an underlying defect in thegeneration of circadian rhythms because the rhythms ofbody temperature, cortisol, and melatonin are essen-tially normal.133–135

The orexin neurons may play a central role in thistiming of sleep and wakefulness. The suprachiasmaticnucleus is the main generator of circadian rhythms.136

This nucleus sends timing information to many areas,but lesions dorsal and posterior to the suprachiasmaticnucleus can disrupt the timing of sleep/wake behaviorwithout affecting other rhythms.137,138 This timing in-formation then may be relayed to the orexin neuronregion. Lesions of this area also upset the timing ofsleep,139 and orexin knockout mice show a similar lackof rhythmicity, especially in the timing of REMsleep.93 By promoting wakefulness and inhibitingREM sleep during the day, orexin may fundamentallyinfluence the circadian timing of sleep/wake behavior.

The orexin neurons may do more than just regulatesleep and wakefulness. Orexin also increases feeding,sympathetic tone, and motor activity.109,140–142 Theimportance of these effects becomes apparent in narco-leptic humans and mice. Around the time that theirorexin neurons are lost, orexin/ataxin-3 mice becomeobese despite eating less than normal mice.109 Narco-leptic subjects with cataplexy are approximately 10 to15% overweight,2,143–145 even though they may eatless than average.146 Thus, this mild obesity may becaused by a decrease in metabolic rate. Considered to-gether, these findings suggest that the orexin neuronsmay help ensure that one is alert, hungry, and meta-bolically active at the correct time of day.

Current Approaches to Treating NarcolepsyNarcolepsy is a chronic disorder, requiring long-termtreatment. Symptoms sometimes worsen during thefirst few years, but then generally remain stable, al-though cataplexy may partially improve with age.147

For most patients with narcolepsy, chronic sleepiness isthe most disabling symptom. Two 15-minute daytimenaps can temporarily reduce sleepiness, especially in pa-tients with more severe sleepiness.148,149 However,naps are rarely adequate as the sole therapy, and mostpatients require treatment with a wake-promotingdrug. None of these drugs completely eliminates sleep-iness, but these medications usually reduce sleepinessenough for substantial improvements in performanceand quality of life.


Pharmacological Treatment of Chronic SleepinessTraditionally, daytime sleepiness has been treated withamphetamine-like drugs including methylphenidate,dextroamphetamine, or methamphetamine, with thelatter two possibly being the most potent150 (Table 1).Although it is moderately effective, pemoline produceshepatic failure on rare occasions and should be usedonly when other drugs have failed. The wake-promoting activity of these drugs is proportionate totheir ability to inhibit the reuptake of amines by thedopamine transporter.71,72,151 By blocking this trans-porter, these drugs increase the release and inhibit thereuptake of dopamine, norepinephrine, and serotonin,resulting in higher synaptic concentrations. This in-creased aminergic signaling may promote wakefulnessthrough direct effects on the cortex or via activation ofsubcortical pathways. Side effects include nervousness,headaches, insomnia, loss of appetite, and palpitations,and all have potential for abuse.150 After a morningdose, many patients may experience some sleepiness inthe afternoon or evening, and a combination ofsustained-release methylphenidate or dextroamphet-amine in the morning with additional short actingdrugs in the afternoon is often effective.

Modafinil is a new, wake-promoting drug that effec-tively treats sleepiness with a minimum of side effects.Modafinil reduces subjective sleepiness and clearly im-proves the ability of narcoleptic subjects to stay awake,although their latency to sleep is still shorter than nor-mal.152,153 Unlike high doses of amphetamines,modafinil does not reduce cataplexy, and an additionalanticataplexy medication may be needed. Modafinil has

a remarkably low incidence of side effects. Headachecan occur in 13%, and nervousness, nausea, or drymouth occurs in less than 10%,154,155 with fewer sideeffects if treatment is begun with a low dose.153 Incontrast with amphetamines, typical doses of modafinildo not produce hypertension, arrhythmias, restlessness,anorexia, or disruption of nighttime sleep.153 Toler-ance or difficulties during withdrawal are very rare,153

and modafinil appears to have little potential for ad-diction.156,157 Some patients find that modafinil is notas effective as amphetamines, but the low incidence ofside effects and low potential for abuse have made it avery popular treatment.

The mechanism through which modafinil promoteswakefulness is only partially understood. Early worksuggested that modafinil might act through �-adrenergicreceptors, but modafinil does not bind to � receptors,or other receptors or reuptake site for noradrenaline,GABA, benzodiazepines, or adenosine.158 Modafinil canreduce the extracellular concentrations of GABA,159,160

and this decrease in GABA may disinhibit specific wake-promoting systems such as the tuberomamillary neurons.Compelling, recent evidence suggests that modafinil mayincrease dopaminergic signaling. All brain regions acti-vated by modafinil receive dopaminergic innervation,161

and modafinil binds weakly to the dopamine reuptaketransporter.158 In mice lacking the dopamine trans-porter, modafinil fails to increase wakefulness.72

Whether this dopaminergic mechanism is the entire ex-planation awaits further characterization of new deriv-atives of modafinil that do not bind to the dopaminetransporter.162

Table 1. Drugs Commonly Used in the Treatment of Narcolepsy

Drug Dosage Side Effects and Comments

Excessive daytime sleepinessMethylphenidate 10–30mg bid, or 20mg SR qam with

10–20mg qpmIrritability, headaches, insomnia

Dextroamphetamine 5–30mg bid, or 10mg SR qam with10–20mg qpm

Same, reduced appetite

Methamphetamine 5–20mg bid SameModafinil 100–400mg qam, or 200mg bid Few side effects, headache, nervousness

Cataplexya

Venlafaxine 75–150mg bid, or 75–150 mg XR qam Few side effects, nauseaFluoxetine 20–80mg qam Same, dry mouthClomipramine 10–150mg qam or qhs Dry mouth, sweatingProtriptyline 5–60mg qam or qhs Same, anxiety, disturbed nighttime sleep�-Hydroxybutyrate 1.5–4.5gm qhs and 2–3 hours later Morning sedation, nausea, dizziness, urinary

incontinenceDisrupted nighttime sleepb

Zolpidem 5–10mg qhs Few side effects, daytime sleepiness, dizziness�-Hydroxybutyrate 1.5–4.5gm qhs and 2–4 hours later Sedation, nausea, dizziness, urinary incontinence

aAnti-cataplexy medications are also useful for problematic sleep paralysis or hypnagogic hallucinations.bSome sleep disruption in narcolepsy may be caused by additional sleep disorders that require specific treatment (for example, obstructive sleepapnea or periodic limb movements of sleep).bid � twice a day; qam � every morning; qpm � every evening; qhs � every bedtime.


Pharmacological Treatment of CataplexyMost patients with narcolepsy do not require treatmentof their cataplexy because it is mild or infrequent, butin some patients, cataplexy occurs several times eachday, producing embarrassment or even injury. Nearlyall anticataplexy medications are potent inhibitors ofREM sleep, a phenomenon consistent with the hypoth-esis that cataplexy is the inappropriate occurrence ofREM sleep paralysis. Noradrenaline and serotonin in-hibit REM-producing neurons in the dorsal pons, andcataplexy is substantially reduced by drugs that increaseaminergic transmission, especially those that increasenoradrenergic signaling by blocking the norepinephrinetransporter.71,163,164 The norepinephrine/serotonin re-uptake inhibitor venlafaxine has become the firstchoice of many clinicians because it is very effectiveand well tolerated. By blocking the reuptake of amines,tricyclics are highly effective, but their use often is lim-ited by their anticholinergic side effects. Abrupt with-drawal from these drugs can markedly worsen cata-plexy, producing status cataplecticus.

�-Hydroxybutyrate (GHB, sodium oxybate) can im-prove cataplexy and daytime sleepiness. GHB is a re-cently approved, sedating drug with a short half-lifethat is given at bedtime with another dose part waythrough the night. GHB occurs naturally in the brainas a metabolite of GABA and binds strongly to GHB-specific receptors with weak binding to GABA-B recep-tors, although how it influences sleep and the symp-toms of narcolepsy is unknown.165–167 In uncontrolledstudies, patients report clear reductions in the fre-quency and severity of cataplexy, and many feel morealert during the day.168 Some hypothesize that these

improvements may be because of better nighttime sleepbecause GHB increases sleep continuity, non-REMsleep, and possibly REM sleep.169 In a small, earlystudy, GHB at bedtime reduced subjective daytimesleepiness and the number of hypnagogic hallucina-tions and daytime naps.170 In a recent, larger study,136 patients took 3 to 9gm GHB, with half of thedose at bedtime and the remainder 2.5 to 4 hourslater.171 After 1 month of treatment the highest dosesubstantially reduced the frequency of cataplexy attacksand produced a moderate reduction in subjective sleep-iness but lower doses were less effective. This high dosealso produced nausea and dizziness in one third of thesubjects, and 14% had urinary incontinence. Food andDrug Administration approval of GHB was difficultbecause of concerns about its safety and potential forabuse. GHB and related drugs can produce vomiting,incontinence, coma, respiratory depression, anddeath,172 and withdrawal can cause anxiety, insomnia,and delirium.172,173 GHB has achieved notoriety as adate-rape drug that is water soluble and tasteless, pro-ducing sedation, disinhibition, and amnesia. Althoughthe clinical trials suggest clear benefits of GHB, thepotential for harmful side effects and abuse warrantscautious use.

Future DirectionsThese recent advances in narcolepsy research havespurred a new look at the nosology of narcolepsy. Asemphasized by Matsuki, Honda,175 and others174 nar-colepsy with cataplexy may be a somewhat differentdisease from narcolepsy without cataplexy or the re-lated syndrome of idiopathic hypersomnia (Table 2).

Table 2. Features of Narcolepsy Syndromes

Feature Narcolepsy with Cataplexy Narcolepsy without Cataplexy Idiopathic Hypersomnia

Nighttime sleepStage 1 (%) 24 16 NDAwakenings (n) 13.5 8.5 20Sleep efficiency (%) 81 91 93

Daytime sleepinessDaily nap (%) 85 45 10Accidents due to sleepiness (%) 48 20 ND

Hypnagogic hallucinations (%) 70–86 15–60 40Sleep paralysis (%) 50–70 25–60 40–50MSLT

Average sleep latency (min) 2.7 3–5 4.3Sleep-onset REM periods (n) 3–3.7 2–3.3 0.2

DQB1*0602 (%) 90–100 40–60 52CSF orexin Undetectable in 90% Normal NormalNeuropathology Markedly reduced number

of orexin neuronsPossibly a partial loss of

orexin neurons or injury tocritical orexin pathways

Unknown

Body mass index Increased 5–15% Normal Normal

Data are from articles cited in the text.6,8,112,113,115,176,177

ND � not determined; MSLT � Multiple Sleep Latency Test; REM � rapid eye movement; CSF � cerebrospinal fluid.


In contrast with these latter groups, nearly all narco-leptic subjects with cataplexy have very low CSF orexinconcentrations and HLA DQB1*0602112,113,115,176

Narcoleptic subjects with cataplexy also are sleepier,have more SOREMs on the MSLT, and generally haveworse nighttime sleep.6,177

Do narcoleptic subjects without cataplexy simplyhave milder or more focal neuropathology? Narcolepticsubjects without cataplexy may have sufficient orexinproduction to maintain normal CSF levels and staveoff cataplexy, but the loss may still be great enough toproduce sleepiness. Alternatively, patients with narco-lepsy and normal CSF orexin concentrations may havesleepiness from a loss of orexin projections to essentialwake-promoting regions, but other projections couldbe intact, resulting in normal CSF levels.178

Perhaps narcolepsy is a disease similar to diabetes,with a severe phenotype due to ligand deficiency and aless severe phenotype without cataplexy when the li-gand is present but somehow less effective. Whentreated with orexin, narcoleptic subjects with cataplexymight have a complete improvement in symptoms, butnarcoleptic subjects without cataplexy might have ablunted or absent response because of abnormalities inthe orexin receptors or their signaling pathways.

With the discovery that orexin deficiency causes nar-colepsy with cataplexy researchers are now seeking toidentify the cause. Might the orexin neurons be presentbut failing to produce orexin? Is there a focal hypotha-lamic encephalitis at the onset of narcolepsy? Doorexin neurons express a particular antigen that makesthem susceptible to an autoimmune attack? If narco-leptic subjects can be identified just as symptoms aredeveloping, would immune suppression alter the courseof disease? These and other questions will undoubtedlybe the focus of research in the years to come.

This work was supported by grants from the NIH (MH62589,MH01507).

We thank C. Saper, J. Matheson, J. Mullington, and B. Murray forthoughtful comments on this manuscript, and M. Papadopoulou forexcellent technical assistance.

References1. von Economo C. Sleep as a problem of localization. J Nervous

Mental Disord 1930;71:249–259.2. Okun ML, Lin L, Pelin Z, et al. Clinical aspects of

narcolepsy-cataplexy across ethnic groups. Sleep 2002;25:27–35.

3. Silber MH, Krahn LE, Olson EJ, et al. The epidemiology ofnarcolepsy in Olmsted County, Minnesota: a population-based study. Sleep 2002;25:197–202.

4. Broughton R, Dunham W, Newman J, et al. Ambulatory 24hour sleep-wake monitoring in narcolepsy-cataplexy comparedto matched controls. Electroencephalogr Clin Neurophysiol1988;70:473–481.

5. Broughton R, Krupa S, Boucher B, et al. Impaired circadianwaking arousal in narcolepsy-cataplexy. Sleep Res Online1998;1:159–165.

6. Harsh J, Peszka J, Hartwig G, et al. Night-time sleep anddaytime sleepiness in narcolepsy. J Sleep Res 2000;9:309–316.

7. Broughton R, Dunham W, Weisskopf M, et al. Night sleepdoes not predict day sleep in narcolepsy. ElectroencephalogrClin Neurophysiol 1994;91:67–70.

8. Aldrich MS. The clinical spectrum of narcolepsy and idio-pathic hypersomnia. Neurology 1996;46:393–401.

9. Dahlitz M, Parkes JD. Sleep paralysis. Lancet 1993;341:406–407.

10. Guilleminault C, Gelb M. Clinical aspects and features of cat-aplexy. Adv Neurol 1995;67:65–77.

11. Anic-Labat S, Guilleminault C, Kraemer HC, et al. Validationof a cataplexy questionnaire in 983 sleep-disorders patients.Sleep 1999;22:77–87.

12. Rubboli G, d’Orsi G, Zaniboni A, et al. A video-polygraphicanalysis of the cataplectic attack. Clin Neurophysiol 2000;111(suppl 2):S120–S128.

13. Overeem S, Lammers GJ, van Dijk JG. Weak with laughter.Lancet 1999;354:838.

14. Lammers GJ, Overeem S, Tijssen MA, et al. Effects of startleand laughter in cataplectic subjects: a neurophysiological studybetween attacks. Clin Neurophysiol 2000;111:1276–1281.

15. Siegel JM, Nienhuis R, Fahringer HM, et al. Neuronal activ-ity in narcolepsy: identification of cataplexy-related cells in themedial medulla. Science 1991;252:1315–1318.

16. Littner M, Johnson SF, McCall WV, et al. Practice parametersfor the treatment of narcolepsy: an update for 2000. Sleep2001;24:451–466.

17. Mayer G, Meier-Ewert K. Motor dyscontrol in sleep of nar-coleptic patients (a lifelong development?). J Sleep Res 1993;2:143–148.

18. Schenck CH, Mahowald MW. Motor dyscontrol in narco-lepsy: rapid-eye-movement (REM) sleep without atonia andREM sleep behavior disorder. Ann Neurol 1992;32:3–10.

19. Carskadon MA, Dement WC, Mitler MM, et al. Guidelinesfor the multiple sleep latency test (MSLT): a standard measureof sleepiness. Sleep 1986;9:519–524.

20. Roehrs T, Roth T. Multiple Sleep Latency Test: technical as-pects and normal values. J Clin Neurophysiol 1992;9:63–67.

21. Moscovitch A, Partinen M, Guilleminault C. The positive di-agnosis of narcolepsy and narcolepsy’s borderland. Neurology1993;43:55–60.

22. Mitler MM, Van den Hoed J, Carskadon MA, et al. REMsleep episodes during the Multiple Sleep Latency Test in nar-coleptic patients. Electroencephalogr Clin Neurophysiol 1979;46:479–481.

23. Chervin RD, Aldrich MS. Sleep onset REM periods duringmultiple sleep latency tests in patients evaluated for sleep ap-nea. Am J Respir Crit Care Med 2000;161:426–431.

24. Manni R, Politini L, Nobili L, et al. Hypersomnia in thePrader Willi syndrome: clinical-electrophysiological featuresand underlying factors. Clin Neurophysiol 2001;112:800–805.

25. Kandt RS, Emerson RG, Singer HS, et al. Cataplexy in vari-ant forms of Niemann-Pick disease. Ann Neurol 1982;12:284–288.

26. Vossler DG, Wyler AR, Wilkus RJ, et al. Cataplexy andmonoamine oxidase deficiency in Norrie disease. Neurology1996;46:1258–1261.

27. Bassetti C, Aldrich MS. Idiopathic hypersomnia. A series of42 patients. Brain 1997;120:1423–1435.

28. Schwartz WJ, Stakes JW, Hobson JA. Transient cataplexy af-ter removal of a craniopharyngioma. Neurology 1984;34:1372–1375.


29. Aldrich MS, Naylor MW. Narcolepsy associated with lesionsof the diencephalon. Neurology 1989;39:1505–1508.

30. Malik S, Boeve BF, Krahn LE, et al. Narcolepsy associatedwith other central nervous system disorders. Neurology 2001;57:539–541.

31. Arii J, Kanbayashi T, Tanabe Y, et al. A hypersomnolent girlwith decreased CSF hypocretin level after removal of a hypo-thalamic tumor. Neurology 2001;56:1775–1776.

32. Scammell TE, Nishino S, Mignot E, et al. Narcolepsy and lowCSF orexin (hypocretin) concentration after a diencephalicstroke. Neurology 2001;56:1751–1753.

33. Clavelou P, Tournilhac M, Vidal C, et al. Narcolepsy associ-ated with arteriovenous malformation of the diencephalon.Sleep 1995;18:202–205.

34. von Economo C. Encephalitis lethargica: its sequela and treat-ment. London: Oxford University Press, 1931.

35. Stiefler G. Narkolepsie nach Enzephalitis Lethargica. WienKlin 1924;37:1044–1046.

36. Adie J. Idiopathic narcolepsy: a disease sui generis, with re-marks on the mechanism of sleep. Brain Res 1926;49:257–307.

37. Lhermitte J, Rouques L. Narcolepsie en apparence id-iopathique, en realite sequelle d’encephalite frustre. Rev Neu-rol (Paris) 1927;1:751–822.

38. Rosenfeld MR, Eichen JG, Wade DF, et al. Molecular andclinical diversity in paraneoplastic immunity to Ma proteins.Ann Neurol 2001;50:339–348.

39. Melberg A, Dahl N, Hetta J, et al. Neuroimaging study inautosomal dominant cerebellar ataxia, deafness, and narco-lepsy. Neurology 1999;53:2190–2192.

40. D’Cruz OF, Vaughn BV, Gold SH, et al. Symptomatic cata-plexy in pontomedullary lesions. Neurology 1994;44:2189–2191.

41. Ma TK, Ang LC, Mamelak M, et al. Narcolepsy secondary tofourth ventricular subependymoma. Can J Neurol Sci 1996;23:59–62.

42. Rivera VM, Meyer JS, Hata T, et al. Narcolepsy followingcerebral hypoxic ischemia. Ann Neurol 1986;19:505–508.

43. Guilleminault C, Faull K, Miles L, et al. Post-traumatic day-time sleepiness: a review of 20 patients. Neurology 1983;33:1584–1589.

44. Berg O, Hanley J. Narcolepsy in two cases of multiple sclero-sis. Acta Neurol Scand 1963;39:252–257.

45. Schrader H, Gotlibson O. Multiple sclerosis and narcolepsy/cataplexy in a monozygotic twin. Neurology 1980;30:105–108.

46. Younger DS, Pedley TA, Thorpy MJ. Multiple sclerosis andnarcolepsy: possible similar genetic susceptibility. Neurology1991;41:447–448.

47. Autret A, Lucas B, Mondon K, et al. Sleep and brain lesions:a critical review of the literature and additional new cases.Neurophysiol Clin 2001;31:356–375.

48. Hublin C, Kaprio J, Partinen M, et al. The prevalence ofnarcolepsy: an epidemiological study of the Finnish Twin Co-hort. Ann Neurol 1994;35:709–716.

49. Guilleminault C, Mignot E, Grumet FC. Familial patterns ofnarcolepsy. Lancet 1989;2:1376–1379.

50. Mignot E. Genetic and familial aspects of narcolepsy. Neurol-ogy 1998;50:S16–S22.

51. Honda Y, Asaka A, Tanimura M, et al. A genetic study ofnarcolepsy and excessive daytime sleepiness in 308 familieswith a narcolepsy or hypersomnia proband. In: GuilleminaultC, Lugaresi E, eds. Sleep/wake disorders: natural history, epi-demiology, and long-term evolution. New York: Raven Press,1983:187–199.

52. Neely S, Rosenberg R, Spire JP, et al. HLA antigens in nar-colepsy. Neurology 1987;37:1858–1860.

53. Rogers AE, Meehan J, Guilleminault C, et al. HLA DR15(DR2) and DQB1*0602 typing studies in 188 narcoleptic pa-tients with cataplexy. Neurology 1997;48:1550–1556.

54. Mignot E, Lin L, Rogers W, et al. Complex HLA-DR and-DQ interactions confer risk of narcolepsy-cataplexy in threeethnic groups. Am J Hum Genet 2001;68:686–699.

55. Mignot E, Young T, Lin L, et al. Nocturnal sleep and daytimesleepiness in normal subjects with HLA-DQB1*0602. Sleep1999;22:347–352.

56. Hohjoh H, Terada N, Nakayama T, et al. Case-control studywith narcoleptic patients and healthy controls who, like thepatients, possess both HLA-DRB1*1501 and -DQB1*0602.Tissue Antigens 2001;57:230–235.

57. Hohjoh H, Nakayama T, Ohashi J, et al. Significant associa-tion of a single nucleotide polymorphism in the tumor necro-sis factor-alpha (TNF-alpha) gene promoter with human nar-colepsy. Tissue Antigens 1999;54:138–145.

58. Hohjoh H, Terada N, Kawashima M, et al. Significant asso-ciation of the tumor necrosis factor receptor 2 (TNFR2) genewith human narcolepsy. Tissue Antigens 2000;56:446–448.

59. Hinze-Selch D, Wetter TC, Zhang Y, et al. In vivo and invitro immune variables in patients with narcolepsy and HLA-DR2 matched controls. Neurology 1998;50:1149–1152.

60. Kubota T, Li N, Guan Z, et al. Intrapreoptic microinjectionof TNF-alpha enhances non-REM sleep in rats. Brain Res2002;932:37–44.

61. Saper CB, Chou TC, Scammell TE. The sleep switch: hypo-thalamic control of sleep and wakefulness. Trends Neurosci2001;24:726–731.

62. Jones B. Basic mechanisms of sleep-wake states. In: Kryger M,Roth T, Dement W, eds. Principles and practice of sleep med-icine. Philadelphia: WB Saunders, 1994:145–162.

63. Lindsley D, Schreiner L, Knowles W, et al. Behavioral andEEG changes following chronic brain stem lesions in the cat.Electroencephalogr Clin Neurophysiol 1950;2:483.

64. Steriade M, Llinas RR. The functional states of the thalamusand the associated neuronal interplay. Physiol Rev 1988;68:649–742.

65. Steriade M, McCormick DA, Sejnowski TJ. Thalamocorticaloscillations in the sleeping and aroused brain. Science 1993;262:679–685.

66. Kapas L, Obal F Jr, Book AA, et al. The effects of immunole-sions of nerve growth factor-receptive neurons by 192 IgG-saporin on sleep. Brain Res 1996;712:53–59.

67. Aston-Jones G, Chiang C, Alexinsky T. Discharge of norad-renergic locus coeruleus neurons in behaving rats and monkeyssuggests a role in vigilance. Prog Brain Res 1991;88:501–520.

68. McGinty D, Harper R. Dorsal raphe neurons: depression offiring during sleep in cats. Brain Res 1976;101:569–575.

69. Vanni-Mercier G, Sakai K, Jouvet M. Specific neurons forwakefulness in the posterior hypothalamus in the cat. C RAcad Sci III 1984;298:195–200.

70. Rye D, Jankovic J. Emerging views of dopamine in modulat-ing sleep/wake state from an unlikely source: PD. Neurology2002;58:341–346.

71. Nishino S, Mignot E. Pharmacological aspects of human andcanine narcolepsy. Prog Neurobiol 1997;52:27–78.

72. Wisor JP, Nishino S, Sora I, et al. Dopaminergic role instimulant-induced wakefulness. J Neurosci 2001;21:1787–1794.

73. Trulson ME, Preussler DW. Dopamine-containing ventraltegmental area neurons in freely moving cats: activity duringthe sleep-waking cycle and effects of stress. Exp Neurol 1984;83:367–377.

74. Miller JD, Farber J, Gatz P, et al. Activity of mesencephalicdopamine and non-dopamine neurons across stages of sleepand walking in the rat. Brain Res 1983;273:133–141.


75. Jones B, Bobillier P, Pin C, et al. The effect of lesions ofcatecholamine-containing neurons upon monoamine contentof the brain and EEG and behavioral waking in the cat. BrainRes 1973;58:157–177.

76. Sakata M, Sei H, Toida K, et al. Mesolimbic dopaminergicsystem is involved in diurnal blood pressure regulation. BrainRes 2002;928:194–201.

77. Lu J, Xu M, Saper C. Identification of wake-active dopami-nergic neurons in the ventral periaqueductal gray (PAG). Sleep2002;25:A290.

78. McCarley RW, Hobson JA. Neuronal excitability modulationover the sleep cycle: a structural and mathematical model. Sci-ence 1975;189:58–60.

79. Williams JA, Reiner PB. Noradrenaline hyperpolarizes identi-fied rat mesopontine cholinergic neurons in vitro. J Neurosci1993;13:3878–3883.

80. Leonard CS, Llinas R. Serotonergic and cholinergic inhibitionof mesopontine cholinergic neurons controlling REM sleep:an in vitro electrophysiological study. Neuroscience 1994;59:309–330.

81. Steriade M, McCarley RW. Brainstem control of wakefulnessand sleep. New York: Plenum, 1990.

82. Lai YY, Shalita T, Hajnik T, et al. Neurotoxic N-methyl-D-aspartate lesion of the ventral midbrain and mesopontine junc-tion alters sleep-wake organization. Neuroscience 1999;90:469–483.

83. Sherin JE, Shiromani PJ, McCarley RW, et al. Activation ofventrolateral preoptic neurons during sleep. Science 1996;271:216–219.

84. Szymusiak R, Alam N, Steininger TL, et al. Sleep-waking dis-charge patterns of ventrolateral preoptic/anterior hypothalamicneurons in rats. Brain Res 1998;803:178–188.

85. Lu J, Greco MA, Shiromani P, et al. Effect of lesions of theventrolateral preoptic nucleus on NREM and REM sleep.J Neurosci 2000;20:3830–3842.

86. Sherin JE, Elmquist JK, Torrealba F, et al. Innervation of his-taminergic tuberomammillary neurons by GABAergic and ga-laninergic neurons in the ventrolateral preoptic nucleus of therat. J Neurosci 1998;18:4705–4721.

87. Gallopin T, Fort P, Eggermann E, et al. Identification ofsleep-promoting neurons in vitro. Nature 2000;404:992–995.

88. Willie JT, Chemelli RM, Sinton CM, et al. To eat or to sleep?Orexin in the regulation of feeding and wakefulness. AnnuRev Neurosci 2001;24:429–458.

89. Sakurai T, Amemiya A, Ishii M, et al. Orexins and orexinreceptors: a family of hypothalamic neuropeptides and Gprotein-coupled receptors that regulate feeding behavior. Cell1998;92:573–585.

90. de Lecea L, Kilduff TS, Peyron C, et al. The hypocretins:hypothalamus-specific peptides with neuroexcitatory activity.Proc Natl Acad Sci USA 1998;95:322–327.

91. Elias CF, Saper CB, Maratos-Flier E, et al. Chemically definedprojections linking the mediobasal hypothalamus and the lat-eral hypothalamic area. J Comp Neurol 1998;402:442–459.

92. Peyron C, Tighe DK, van den Pol AN, et al. Neurons con-taining hypocretin (orexin) project to multiple neuronal sys-tems. J Neurosci 1998;18:9996–10015.

93. Chemelli RM, Willie JT, Sinton CM, et al. Narcolepsy inorexin knockout mice: molecular genetics of sleep regulation.Cell 1999;98:437–451.

94. Marcus JN, Aschkenasi CJ, Lee CE, et al. Differential expres-sion of orexin receptors 1 and 2 in the rat brain. J CompNeurol 2001;435:6–25.

95. Estabrooke IV, McCarthy MT, Ko E, et al. Fos expression inorexin neurons varies with behavioral state. J Neurosci 2001;21:1656–1662.

96. Yoshida Y, Fujiki N, Nakajima T, et al. Fluctuation of extra-cellular hypocretin-1 (orexin A) levels in the rat in relation tothe light-dark cycle and sleep-wake activities. Eur J Neurosci2001;14:1075–1081.

97. Alam MN, Gong H, Alam T, et al. Sleep-waking dischargepatterns of neurons recorded in the rat perifornical lateral hy-pothalamic area. J Physiol 2002;538:619–631.

98. Hagan J, Leslie R, Patel S, et al. Orexin A activates locus co-eruleus cell firing and increases arousal in the rat. Proc NatlAcad Sci USA 1999;96:10911–10916.

99. Bayer L, Eggermann E, Serafin M, et al. Orexins (hypocretins)directly excite tuberomammillary neurons. Eur J Neurosci2001;14:1571–1575.

100. Yamanaka A, Tsujino N, Funahashi H, et al. Orexins activatehistaminergic neurons via the orexin 2 receptor. Biochem Bio-phys Res Commun 2002;290:1237–1245.

101. Eriksson KS, Sergeeva O, Brown RE, et al. Orexin/hypocretinexcites the histaminergic neurons of the tuberomammillarynucleus. J Neurosci 2001;21:9273–9279.

102. Brown RE, Sergeeva O, Eriksson KS, et al. Orexin A excitesserotonergic neurons in the dorsal raphe nucleus of the rat.Neuropharmacology 2001;40:457–459.

103. Bourgin P, Huitron-Resendiz S, Spier AD, et al. Hypocretin-1modulates rapid eye movement sleep through activation of lo-cus coeruleus neurons. J Neurosci 2000;20:7760–7765.

104. Huang ZL, Qu WM, Li WD, et al. Arousal effect of orexin Adepends on activation of the histaminergic system. Proc NatlAcad Sci USA 2001;98:9965–9970.

105. Nishino S, Fujiki N, Ripley B, et al. Decreased brain hista-mine content in hypocretin/orexin receptor-2 mutated narco-leptic dogs. Neurosci Lett 2001;313:125–128.

106. Lin L, Faraco J, Li R, et al. The sleep disorder canine narco-lepsy is caused by a mutation in the hypocretin (orexin) re-ceptor 2 gene. Cell 1999;98:365–376.

107. Tokita S, Chemelli R, Willie J, et al. Behavioral characteriza-tion of orexin-2 receptor (OX2R) knockout mice. Sleep 2001;24:A20–A21.

108. Kisanuki Y, Chemelli R, Tokita S, et al. Behavioral and poly-somnographic characterization of orexin-1 receptor and orexin-2receptor double knockout mice. Sleep 2001;24:A22.

109. Hara J, Beuckmann CT, Nambu T, et al. Genetic ablation oforexin neurons in mice results in narcolepsy, hypophagia, andobesity. Neuron 2001;30:345–354.

110. Nishino S, Ripley B, Overeem S, et al. Hypocretin (orexin)deficiency in human narcolepsy. Lancet 2000;355:39–40.

111. Ripley B, Overeem S, Fujiki N, et al. CSF hypocretin/orexinlevels in narcolepsy and other neurological conditions. Neu-rology 2001;57:2253–2258.

112. Nishino S, Ripley B, Overeem S, et al. Low cerebrospinalfluid hypocretin (Orexin) and altered energy homeostasis inhuman narcolepsy. Ann Neurol 2001;50:381–388.

113. Kanbayashi T, Inoue Y, Chiba S, et al. CSF hypocretin-1(orexin-A) concentrations in narcolepsy with and without cat-aplexy and idiopathic hypersomnia. J Sleep Res 2002;11:91–93.

114. Dalal MA, Schuld A, Haack M, et al. Normal plasma levels oforexin A (hypocretin-1) in narcoleptic patients. Neurology2001;56:1749–1751.

115. Mignot E, Lammers G, Ripley B, et al. The role of CSF hypo-cretin measurement in the diagnosis of narcolepsy and otherhypersomnias. Arch Neurol 2002;59:1553–1562.

116. Melberg A, Ripley B, Lin L, et al. Hypocretin deficiency infamilial symptomatic narcolepsy. Ann Neurol 2001;49:136–137.


117. Peyron C, Faraco J, Rogers W, et al. A mutation in a case ofearly onset narcolepsy and a generalized absence of hypocretinpeptides in human narcoleptic brains. Nat Med 2000;6:991–997.

118. Thannickal T, Moore RY, Nienhuis R, et al. Reduced numberof hypocretin neurons in human narcolepsy. Neuron 2000;27:469–474.

119. Olafsdottir BR, Rye DB, Scammell TE, et al. Polymorphismsin hypocretin/orexin pathway genes and narcolepsy. Neurol-ogy 2001;57:1896–1899.

120. Hungs M, Lin L, Okun M, et al. Polymorphisms in the vi-cinity of the hypocretin/orexin are not associated with humannarcolepsy. Neurology 2001;57:1893–1895.

121. Matsuki K, Juji T, Honda Y. Immunological features in Ja-pan. In: Honda Y, Juji T, eds. HLA and narcolepsy. NewYork: Springer Verlag, 1988:58–76.

122. Frey JL, Heiserman JE. Absence of pontine lesions in narco-lepsy. Neurology 1997;48:1097–1099.

123. Bassetti C, Aldrich MS, Quint DJ. MRI findings in narco-lepsy. Sleep 1997;20:630–631.

124. Han F, Chen E, Wei H, et al. Childhood narcolepsy in NorthChina. Sleep 2001;24:321–324.

125. Kwen PL, Pullicino P. Pontine lesions in idiopathic narco-lepsy. Neurology 1998;50:577–578.

126. Rubin RL, Hajdukovich RM, Mitler MM. HLA-DR2 associ-ation with excessive somnolence in narcolepsy does not gen-eralize to sleep apnea and is not accompanied by systemic au-toimmune abnormalities. Clin Immunol Immunopathol 1988;49:149–158.

127. Plazzi G, Montagna P, Provini F, et al. Pontine lesions inidiopathic narcolepsy. Neurology 1996;46:1250–1254.

128. Fredrikson S, Carlander B, Billiard M, et al. CSF immunevariables in patients with narcolepsy. Acta Neurol Scand 1990;81:253–254.

129. Erlich SS, Itabashi HH. Narcolepsy: a neuropathologic study.Sleep 1986;9:126–132.

130. Dantz B, Edgar DM, Dement WC. Circadian rhythms innarcolepsy: studies on a 90 minute day. ElectroencephalogrClin Neurophysiol 1994;90:24–35.

131. Dijk DJ, Czeisler CA. Contribution of the circadian pace-maker and the sleep homeostat to sleep propensity, sleep struc-ture, electroencephalographic slow waves, and sleep spindle ac-tivity in humans. J Neurosci 1995;15:3526–3538.

132. Lavie P. REM periodicity under ultrashort sleep/wake cycle innarcoleptic patients. Can J Psychol 1991;45:185–193.

133. Mayer G, Hellmann F, Leonhard E, et al. Circadian temper-ature and activity rhythms in unmedicated narcoleptic pa-tients. Pharmacol Biochem Behav 1997;58:395–402.

134. Bourgine N, Claustrat B, Besset A, et al. Melatonin plasmaconcentration during daytime and nighttime in narcolepticsubjects and controls. Sleep Res 1986;15:46.

135. Weitzman E. Twenty-four hour neuroendocrine secretorypatterns: observations on patients with narcolepsy. In: Guil-leminault C, Dement W, Passouant P, eds. Narcolepsy. NewYork: Spectrum, 1976:521–542.

136. Herzog ED, Schwartz WJ. Invited review: a neural clock-work for encoding circadian time. J Appl Physiol 2002;92:401– 408.

137. Lu J, Zhang Y, Chou T, et al. Contrasting effects of ibotenatelesions of the paraventricular nucleus and subparaventricularzone on sleep-wake cycle and temperature regulation. J Neu-rosci 2001;21:4864–4874.

138. Chou T, Scammell T, Lu J, et al. Excitotoxic lesions of thedorsomedial hypothalamic nucleus markedly attenuate circa-dian rhythms of sleep and body temperature. Paper presentedat: Annual Meeting of the Society for Neuroscience; 2000.

139. Gerashchenko D, Kohls MD, Greco M, et al. Hypocretin-2-saporin lesions of the lateral hypothalamus produce narco-leptic-like sleep behavior in the rat. J Neurosci 2001;21:7273–7283.

140. Date Y, Ueta Y, Yamashita H, et al. Orexins, orexigenic hy-pothalamic peptides, interact with autonomic, neuroendocrineand neuroregulatory systems. Proc Natl Acad Sci USA 1999;96:748–753.

141. Chen CT, Hwang LL, Chang JK, et al. Pressor effects of orex-ins injected intracisternally and to rostral ventrolateral medullaof anesthetized rats. Am J Physiol Regul Integr Comp Physiol2000;278:R692–R697.

142. Matsumura K, Tsuchihashi T, Abe I. Central orexin-A aug-ments sympathoadrenal outflow in conscious rabbits. Hyper-tension 2001;37:1382–1387.

143. Schuld A, Hebebrand J, Geller F, et al. Increased body-massindex in patients with narcolepsy. Lancet 2000;355:1274–1275.

144. Schuld A, Blum WF, Uhr M, et al. Reduced leptin levels inhuman narcolepsy. Neuroendocrinology 2000;72:195–198.

145. Dahmen N, Bierbrauer J, Kasten M. Increased prevalence ofobesity in narcoleptic patients and relatives. Eur Arch Psychi-atry Clin Neurosci 2001;251:85–89.

146. Lammers GJ, Pijl H, Iestra J, et al. Spontaneous food choicein narcolepsy. Sleep 1996;19:75–76.

147. Furuta H, Thorpy MJ, Temple HM. Comparison in symp-toms between aged and younger patients with narcolepsy. Psy-chiatry Clin Neurosci 2001;55:241–242.

148. Mullington J, Broughton R. Scheduled naps in the manage-ment of daytime sleepiness in narcolepsy-cataplexy. Sleep1993;16:444–456.

149. Rogers AE, Aldrich MS, Lin X. A comparison of three differ-ent sleep schedules for reducing daytime sleepiness in narco-lepsy. Sleep 2001;24:385–391.

150. Mitler MM, Aldrich MS, Koob GF, et al. Narcolepsy and itstreatment with stimulants. ASDA standards of practice. Sleep1994;17:352–3571.

151. Nishino S, Mao J, Sampathkumaran R, et al. Increased dopa-minergic transmission mediates the wake-promoting effects ofCNS stimulants. Sleep Res Online 1998;1:49–61.

152. US Modafinil in Narcolepsy Multicenter Study Group. Ran-domized trial of modafinil for the treatment of pathologicalsomnolence in narcolepsy. Ann Neurol 1998;43:88–97.

153. US Modafinil in Narcolepsy Multicenter Study Group. Ran-domized trial of modafinil as a treatment for the excessive day-time somnolence of narcolepsy. Neurology 2000;54:1166–1175.

154. Fry J. A new alternative in the pharmacologic management ofsomnolence: a phase III study of modafinil in narcolepsy. AnnNeurol 1996;40:493.

155. Mitler M, Harsh J, Hirshkowitz M, et al. Long-term efficacyand safety of modafinil (PROVIGIL�) for the treatment ofexcessive daytime sleepiness associated with narcolepsy. SleepMed 2000;1:231–243.

156. Jasinski DR. An evaluation of the abuse potential of modafinilusing methylphenidate as a reference. J Psychopharmacol2000;14:53–60.

157. Gold LH, Balster RL. Evaluation of the cocaine-like discrim-inative stimulus effects and reinforcing effects of modafinil.Psychopharmacol (Berl) 1996;126:286–292.

158. Mignot E, Nishino S, Guilleminault C, et al. Modafinil bindsto the dopamine uptake carrier site with low affinity. Sleep1994;17:436–437.


159. Tanganelli S, Fuxe K, Ferraro L, et al. Inhibitory effects of thepsychoactive drug modafinil on gamma-aminobutyric acidoutflow from the cerebral cortex of the awake freely movingguinea-pig. Possible involvement of 5-hydroxytryptaminemechanisms. Naunyn Schmiedebergs Arch Pharmacol 1992;345:461–465.

160. Ferraro L, Tanganelli S, O’Connor WT, et al. The vigilancepromoting drug modafinil decreases GABA release in the me-dial preoptic area and in the posterior hypothalamus of theawake rat: possible involvement of the serotonergic 5-HT3 re-ceptor. Neurosci Lett 1996;220:5–8.

161. Scammell TE, Estabrooke IV, McCarthy MT, et al. Hypotha-lamic arousal regions are activated during modafinil-inducedwakefulness. J Neurosci 2000;20:8620–8628.

162. Miller M, Chatterjee S, Aimone L, et al. Novel non-dopaminergic wake promoting and cortical activating agents.Paper presented at: Annual Meeting of the Society forNeuroscience; 2001; San Diego.

163. Larrosa O, de la Llave Y, Bario S, et al. Stimulant and anti-cataplectic effects of reboxetine in patients with narcolepsy: apilot study. Sleep 2001;24:282–285.

164. Frey J, Darbonne C. Fluoxetine suppresses human cataplexy: apilot study. Neurology 1994;44:707–709.

165. Hechler V, Weissmann D, Mach E, et al. Regional distribu-tion of high-affinity gamma-[3H]hydroxybutyrate bindingsites as determined by quantitative autoradiography. J Neuro-chem 1987;49:1025–1032.

166. Tunnicliff G. Sites of action of gamma-hydroxybutyrate(GHB)—a neuroactive drug with abuse potential. J ToxicolClin Toxicol 1997;35:581–590.

167. Xie X, Smart TG. Gamma-hydroxybutyrate hyperpolarizeshippocampal neurones by activating GABAB receptors. EurJ Pharmacol 1992;212:291–294.

168. Mamelak M, Scharf MB, Woods M. Treatment of narcolepsywith gamma-hydroxybutyrate. A review of clinical and sleeplaboratory findings. Sleep 1986;9:285–289.

169. Scrima L, Hartman PG, Johnson FH Jr, et al. The effects ofgamma-hydroxybutyrate on the sleep of narcolepsy patients: adouble-blind study. Sleep 1990;13:479–490.

170. Lammers GJ, Arends J, Declerck AC, et al. Gamma-hydroxybutyrate and narcolepsy: a double-blind placebo-controlled study. Sleep 1993;16:216–220.

171. Cook H, for the US Xyrem Multicenter Study Group. A ran-domized, double blind, placebo-controlled multicenter trialcomparing the effects of three doses of orally administered so-dium oxybate with placebo for the treatment of narcolepsy.Sleep 2002;25:42–49.

172. Zvosec DL, Smith SW, McCutcheon JR, et al. Adverse events,including death, associated with the use of 1,4-butanediol.N Engl J Med 2001;344:87–94.

173. Li J, Stokes SA, Woeckener A. A tale of novel intoxication:seven cases of gamma-hydroxybutyric acid overdose. AnnEmerg Med 1998;31:723–728.

174. Matsuki K, Honda Y, Juji T. Diagnostic criteria for narcolepsyand HLA-DR2 frequencies. Tissue Antigens 1987;30:155–160.

175. Honda Y. Clinical features of narcolepsy: Japanese experience.In: Honda Y, Juji T, eds. HLA in narcolepsy. Berlin:Springer-Verlag, 1988:24–57.

176. Mignot E, Hayduk R, Black J, et al. HLA DQB1*0602 isassociated with cataplexy in 509 narcoleptic patients. Sleep1997;20:1012–1020.

177. Rosenthal LD, Zorick FJ, Merlotti L, et al. Signs and symp-toms associated with cataplexy in narcolepsy patients. BiolPsychiatry 1990;27:1057–1060.

178. Mignot E. A commentary on the neurobiology of thehypocretin/orexin system. Neuropsychopharmacology 2001;25:S5–S13.


neurological progress the neurobiology, diagnosis, and...

Documents