The n e w e ng l a n d j o u r na l of m e dic i n e

Review Article
Edward W. Campion, M.D., Editor

Narcolepsy
Thomas E. Scammell, M.D.​​

N
From the Department of Neurology, Beth arcolepsy, one of the most common causes of chronic sleepi-
Israel Deaconess Medical Center and ness, affects about 1 in 2000 people. Despite the frequency of narcolepsy,
Boston Children’s Hospital, Boston. Ad-
dress reprint requests to Dr. Scammell at the average time from the onset of symptoms to diagnosis is 5 to 15 years,
Beth Israel Deaconess Medical Center, and narcolepsy may remain undiagnosed in as many as half of all affected people
330 Brookline Ave., Boston, MA 02215, or with narcolepsy, since many clinicians are unfamiliar with this disorder.1 Fortu-
at ­tscammel@​­bidmc​.­harvard​.­edu.
nately, awareness of narcolepsy and other sleep disorders is increasing, and over
N Engl J Med 2015;373:2654-62. the past several years researchers have made great progress in understanding
DOI: 10.1056/NEJMra1500587
Copyright © 2015 Massachusetts Medical Society.
narcolepsy. Clinicians now recognize two types of narcolepsy. Type 1 is caused by
extensive loss of hypothalamic neurons that produce the neuropeptides orexin-A
and -B (also referred to as hypocretin-1 and -2); type 2 includes most of the same
symptoms, but its cause is unknown. This review focuses on the symptoms and
pathologic and neurobiologic features of narcolepsy and provides a framework for
diagnosis and effective treatment.

S ymp t oms
Narcolepsy usually begins between the ages of 10 and 20 years with the sudden
onset of persistent daytime sleepiness, although it can also develop gradually. In
many persons with narcolepsy, the sleepiness is severe, resulting in difficulty focus-
ing and staying awake at school, at work, and during periods of inactivity (e.g.,
when watching a movie). Quite often, the diagnosis is made only after serious
problems have arisen, such as declining grades at school, poor performance at
work, or a motor vehicle accident. Although it may appear to be challenging to
distinguish daytime sleepiness due to narcolepsy from that caused by insufficient
sleep, especially in teenagers, people with narcolepsy are sleepy every day, even
with adequate nighttime sleep. In contrast to people with disorders such as ob-
structive sleep apnea who have poor-quality sleep, those with narcolepsy usually
feel refreshed after a full night’s sleep or a brief nap, but their sleepiness returns
1 to 2 hours later, especially when they are sedentary.
Narcolepsy is also characterized by disordered regulation of rapid-eye-move-
ment (REM) sleep. REM sleep normally occurs only during the usual sleep period
and includes vivid, storylike dreams, rapid (saccadic) eye movements, and paralysis
of nearly all skeletal muscles, except the muscle of respiration. REM sleep can occur
in persons with narcolepsy at any time of day, and the classic elements of REM
sleep often intrude into wakefulness, creating peculiar intermediate states.
The most dramatic of these REM sleep–like states is cataplexy — sudden episodes
of partial or complete paralysis of voluntary muscles. These episodes are triggered
by strong emotions (Fig. 1), most often by positive emotions such as those associ-
ated with laughing at a joke or unexpectedly encountering a friend. In some peo-
ple, however, cataplexy can be triggered by intense negative emotions, such as
frustration or anger. The paralysis usually evolves over many seconds, first affect-
ing the face and neck and then causing weakness in the trunk and limbs, although
the muscles associated with breathing are spared. With partial cataplexy, slurred

2654 n engl j med 373;27 nejm.org  December 31, 2015

The New England Journal of Medicine

Downloaded from nejm.org at UNIVERSITY OF WOLLONGONG on December 30, 2015. For personal use only. No other uses without permission.
Copyright © 2015 Massachusetts Medical Society. All rights reserved.
 Narcolepsy

Strong
emotional Partial cataplexy Complete cataplexy
trigger

Weakness in the
face and neck

Weakness in the
limbs and trunk

Figure 1. Cataplexy.
Cataplexy is characterized by sudden, emotionally triggered episodes of muscle weakness with preserved consciousness. These episodes
typically begin with weakness of the muscles of the face and neck that then spreads to involve the muscles of the limbs and trunk.

speech and a sagging face are common; with person acts on them (e.g., calling the police to
complete episodes, the person may slump to the report that there is a burglar in the house).3 Like
ground, fully conscious but immobile for as cataplexy, sleep paralysis and hypnagogic hallu-
long as 1 or 2 minutes. Children with cataplexy cinations rarely last more than 1 or 2 minutes.
can have long-lasting periods of low muscle Because hypnagogic hallucinations and sleep
tone, with a wobbly gait and perioral movements paralysis occur occasionally in about 20% of the
such as grimacing and tongue protrusion.2 A general population, they are less informative
patient’s report of a history of cataplexy is diag- diagnostically than a history of cataplexy.
nostically very helpful to the physician, since Narcolepsy often includes additional problems
cataplexy occurs almost exclusively in type 1 that may require independent treatment. Although
narcolepsy. people with narcolepsy are sleepy much the day,
The paralysis and dreams typical of REM they often have fragmented sleep at night 4 and
sleep can also occur at the borders of sleep. sometimes require treatment with a sleep-pro-
Sleep paralysis is much like cataplexy, but it can moting medication. They also have a tendency
occur spontaneously on awakening from sleep to gain excess weight; at the onset of narco-
and occasionally as the person is falling asleep. lepsy, children can gain 20 to 40 lb (9 to 18 kg),
Dreamlike and often disturbing hallucinations and the body-mass index in adults is approxi-
are common in narcolepsy. Those that occur at mately 15% above average, possibly because of a
the onset of sleep are referred to as hypnagogic low metabolic rate. Other sleep disorders that
hallucinations, and those that occur on awaken- are more prevalent among persons with narco-
ing as hypnopompic hallucinations. Typical hallu- lepsy than in the general population include
cinations might include the sense that a threat- obstructive sleep apnea, periodic limb movement
ening stranger is in the bedroom or that one is disorder (nocturnal myoclonus), sleepwalking,
being attacked by animals. In contrast to people and REM sleep behavior disorder.5 In addition,
with psychotic disorders, those with narcolepsy depression is common in persons with narco-
rarely have complex auditory hallucinations or lepsy,6 although it remains unclear whether this
fixed delusions, although hypnopompic hallu- is due to the effect of narcolepsy on the person’s
cinations can occasionally be so vivid that the life or the underlying neuropathologic state.

n engl j med 373;27 nejm.org December 31, 2015 2655

The New England Journal of Medicine
Downloaded from nejm.org at UNIVERSITY OF WOLLONGONG on December 30, 2015. For personal use only. No other uses without permission.
Copyright © 2015 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

Table 1. Differential Diagnosis of Chronic Daytime Sleepiness.*

Diagnosis Distinguishing Characteristics

Insufficient sleep Sleepiness decreases with more sleep on weekends and holidays
Obstructive sleep apnea Snoring, witnessed episodes of apnea, large tonsils, large tongue, long uvula, obesity
Narcolepsy Cataplexy, hypnagogic and hypnopompic hallucinations, sleep paralysis, fragmented
sleep
Delayed sleep phase disorder Sleepiness in the morning, alertness at night
Periodic limb movement Sleep disrupted by kicking movements; often occurs with the restless legs syndrome,
disorder iron deficiency, uremia, and neuropathy
Shift-work sleep disorder Sleepiness when working at night, insufficient sleep during the day
Use of sedating medications Insomnia medications, opiates, anxiolytics, anticonvulsants, antipsychotics, anti­
depressants, antihistamines, among others
Idiopathic hypersomnia Lengthy nighttime sleep and long naps, difficulty waking from sleep
Depression Increased time in bed but little functional sleepiness on the multiple sleep latency test
Other medical disorders Symptoms of hypothyroidism, Parkinson’s disease, the Prader–Willi syndrome, myo-
tonic dystrophy, among others

* With most of these disorders, people do not feel rested when arising in the morning; however, most people with narco-
lepsy feel alert on awakening.

Di agnosis correct conditions.9 Medications that suppress

REM sleep should be discontinued well in ad-
The diagnosis of narcolepsy is often apparent vance of the test (e.g., 3 weeks for antidepres-
from the clinical history, but it is essential to sants with a long half-life), and any other psy-
confirm the diagnosis with overnight polysom- choactive medications, especially stimulants,
nography followed by a multiple sleep latency should be discontinued 1 week in advance. The
test the next day. The overnight sleep study helps patient should obtain and document with a sleep
rule out other potential causes of daytime sleep- log an adequate amount of sleep each night in
iness; in people with narcolepsy, it may show the week before the multiple sleep latency test,
fragmented, light sleep and an early transition since inadequate nighttime sleep may result in
into REM sleep (<15 minutes after the onset of short daytime sleep latencies. During the over-
sleep).7,8 During the multiple sleep latency test, night sleep study preceding the test, adults
the patient is encouraged every 2 hours to fall should get at least 6 hours of sleep, and children
asleep for 20 minutes; the test usually begins at should get more. Adherence to these conditions
8 a.m. and ends at approximately 5 to 6 p.m. is important, because people with sleep depriva-
Given the opportunity to nap, people with narco- tion or shift-work schedules and those who
lepsy usually fall asleep in less than 8 minutes, receive psychoactive medications can have test
whereas healthy people generally fall asleep in results that are similar to those seen in people
15 minutes or more. In addition, people with with narcolepsy.10,11
narcolepsy usually have REM sleep during at least Because persistent sleepiness can occur with
two of these daytime naps (known as sleep- many conditions, it is important in considering
onset REM sleep periods), whereas people with- a diagnosis of narcolepsy to rule out other sleep
out narcolepsy rarely have any daytime REM disorders12 (Table 1). Insufficient sleep is very
sleep. A positive multiple sleep latency test common in teenagers and adults, and extending
(defined as a short time to fall asleep plus REM the sleep period to 8 or 9 hours each night for a
sleep in at least two of the naps) provides strong, week can be diagnostically helpful. The clinical
objective evidence of excessive sleepiness and history and results of the overnight sleep study
poorly regulated REM sleep. can usually rule out sleep apnea, delayed sleep
Because the diagnosis of narcolepsy relies phase disorder, and periodic limb movement
heavily on the multiple sleep latency test, it is disorder. The effects of shift work or sedating
essential that the test be performed under the medications should also be considered. Idio-

2656 n engl j med 373;27 nejm.org  December 31, 2015

The New England Journal of Medicine

Downloaded from nejm.org at UNIVERSITY OF WOLLONGONG on December 30, 2015. For personal use only. No other uses without permission.
Copyright © 2015 Massachusetts Medical Society. All rights reserved.
 Narcolepsy

pathic hypersomnia is an uncommon cause of DQB1*06:02 is also detected in roughly 50% of

chronic sleepiness; in people with this condi- people with type 2 narcolepsy, but only in 12 to
tion, as in those with narcolepsy, the mean sleep 30% of the general population in the United
latency on the multiple sleep latency test is short, States, Europe, and Japan. Overall, this gene is
but there is only one or no sleep-onset REM predicted to increase the risk of narcolepsy by a
sleep periods. Some people with idiopathic hyper- factor of about 200.23 In persons who are positive
somnia can sleep for long periods (>10 hours) for DQB1*06:02 and in whom it is difficult to
every day and have long, unrefreshing naps, but make a diagnosis, measurement of orexin-A levels
people with narcolepsy typically sleep for nor- in cerebrospinal fluid can be very helpful, al-
mal amounts of time over a 24-hour period and though the test is not yet commercially available.24
feel refreshed after a 15-to-20-minute nap. Type 1 narcolepsy has also been linked to poly-
morphisms in other genes. DQB1*06:02 is almost
always accompanied by the linked DQA1*01:02
Pathol o gic a nd Gene t ic
Fe at ur e s gene, and their gene products form a heterodi-
mer that presents antigens to T-cell receptors on
Clinicians now recognize two types of narco- CD4 T cells.25 Variations in other HLA-DQ al-
lepsy.13 In both disorders, people have chronic leles, HLA-DP, and HLA class I also contribute to
daytime sleepiness and positive results on the genetic susceptibility, as do polymorphisms in
multiple sleep latency test. Type 1 narcolepsy is other genes that affect immune function, such as
characterized by cataplexy and very low levels of TCRA, TCRB, P2RY11, EIF3G, ZNF365, IL10RB-IFNAR1,
orexin-A in cerebrospinal fluid. In contrast, peo- CTSH, and TNFSF4.25-27 These genetic factors are
ple with type 2 narcolepsy do not have cataplexy clearly important, but they are not the whole
and have normal orexin-A levels. The difference story. Narcolepsy is usually sporadic, and the
in orexin levels suggests that type 1 and type 2 risk that an affected parent will have an affected
narcolepsy may be phenotypically similar disor- child is only 1%. Even if one monozygotic twin
ders with different underlying causes. Sleepiness has narcolepsy, there is only about a 30% chance
and other symptoms are generally more severe that narcolepsy will develop in the other twin.
in type 1 narcolepsy, and the diagnosis is usu- The onset of symptoms of narcolepsy most
ally straightforward. In contrast, the diagnosis commonly occurs in the late spring; this sug-
of type 2 narcolepsy can be a challenge, since gests that the disease may be triggered by winter
other potential causes of sleepiness must be infections.28 For example, high titers of antibod-
ruled out and the multiple sleep latency test can ies against antistreptolysin O are common soon
be falsely positive or falsely negative.10-12 after the onset of narcolepsy, which suggests that
Type 1 narcolepsy is caused by a severe loss streptococcal infections may trigger the dis-
of neurons that produce the orexin neuropep- ease.29 In the winter of 2009–2010, a dramatic
tides. Orexin-A and orexin-B are small peptide spike in new cases of narcolepsy provided the
neurotransmitters made only by a cluster of neu- clearest evidence so far that the disease can
rons in the lateral hypothalamus.14,15 In 1999, be caused by an autoimmune process. During
researchers discovered that mice and dogs with the H1N1 influenza pandemic that winter, a spe-
disrupted orexin signaling have symptoms simi- cific brand of vaccine against H1N1 (Pandemrix)
lar to those of narcolepsy in humans,16,17 and that contained a potent adjuvant was used widely
shortly thereafter, two groups of investigators in Scandinavia and some other parts of Europe.
found that type 1 narcolepsy is caused by a severe In these regions, the number of new cases of
but highly selective loss of the orexin-producing narcolepsy increased by as much as a factor of 12,
neurons.18-20 Type 2 narcolepsy may be caused by with onset of symptoms 1 to 2 months after vac-
less extensive injury to these neurons,21 but data cination, but only in children and teenagers car-
on the disease process in type 2 are quite limited. rying the DQB1*06:02 gene.30,31 In China, H1N1
The process that destroys the orexin-produc- influenza infection was common in the winter
ing neurons remains a mystery, but genetic factors of 2009–2010, and new cases of narcolepsy tripled
are clearly important. More than 98% of people in the subsequent year.28 This surge in new cases
with type 1 narcolepsy carry HLA-DQB1*06:02, suggests that the combination of DQB1*06:02,
a finding that makes this the strongest asso- young age, and particular immune stimuli strong-
ciation between HLA and disease known.22,23 ly increases the risk of narcolepsy.

n engl j med 373;27 nejm.org  December 31, 2015 2657

The New England Journal of Medicine
Downloaded from nejm.org at UNIVERSITY OF WOLLONGONG on December 30, 2015. For personal use only. No other uses without permission.
Copyright © 2015 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

These findings have spurred many research- of REM sleep — such as paralysis or dreamlike
ers to hypothesize that narcolepsy is caused by hallucinations — to occur during wakefulness.38
an autoimmune process. It is probable that an Cataplexy is triggered by signals related to strong,
immune stimulus such as influenza or strepto- positive emotions that may be relayed through
coccus infection triggers a T-cell response and the medial prefrontal cortex and amygdala to
that, in genetically susceptible persons, this in- activate circuits in the pons that cause muscle
flammatory reaction damages the orexin-produc- paralysis.39-41 Orexin signaling also increases me-
ing neurons, perhaps through a process of mo- tabolism, sympathetic tone, and rewarding be-
lecular mimicry.32 Countering this idea, however, haviors such as drug seeking. It is possible that
are the facts that no signs of inflammation are dysfunction in these pathways contributes to the
detected in cerebrospinal fluid or seen on mag- obesity and depression that occur in many peo-
netic resonance imaging in people with narco- ple with narcolepsy.6,42
lepsy and that there is no strong evidence of an
increased prevalence of other autoimmune dis- T r e atmen t
eases among such persons. Furthermore, small
trials of immunomodulating agents have pro- Narcolepsy is treated with a combination of be-
duced little reduction in symptoms. havioral and pharmacologic approaches.43,44 Day-
On rare occasions, narcolepsy occurs as part time sleepiness often partially decreases with
of a broader injury to the hypothalamus or to the sufficient good-quality sleep at night and a
projections of the orexin-producing neurons as a 15-to-20-minute nap in the afternoon. Any ad-
consequence of sarcoidosis, demyelination, or a ditional sleep disorders in the patient, such as
stroke, tumor, or paraneoplastic disorder.33 Peo- sleep apnea, should also be addressed.
ple with narcolepsy caused by a structural lesion Most people with narcolepsy also require treat-
are easily distinguished from those with typical ment with wakefulness-promoting medications
narcolepsy, since they have increases in total (Table 2).45 For mild-to-moderate daytime sleepi-
amounts of sleep and additional signs of hypo- ness, modafinil is often a good choice, since it
thalamic or brain-stem injury, such as pituitary has an acceptable side-effect profile and has a
dysfunction, abnormal eye movements, or upper- low potential for abuse.46 It is probable that
motor-neuron weakness. modafinil improves wakefulness by reducing the
reuptake of dopamine. Methylphenidate, as well as
dextroamphetamine and similar amphetamines,
Neurobiol o gic Fe at ur e s
can be more potent than modafinil, but side ef-
The discovery that type 1 narcolepsy is caused by fects are more common with these drugs. Diver-
the loss of orexin-producing neurons has greatly sion and abuse of these medications can also be
advanced understanding of the underlying neuro- a source of concern, but the risk may be lower
biologic features of this disorder. The orexin with slow-release formulations. These drugs
neurons are active during wakefulness,34-36 and probably promote wakefulness by blocking the
the orexin neuropeptides stimulate target neu- reuptake and increasing the release of dopamine;
rons that promote wakefulness, including those they have similar effects on serotonin and nor-
in the cortex and basal forebrain and those in epinephrine signaling, though to a lesser extent.47
the brain stem and hypothalamus that produce The Epworth Sleepiness Scale is a helpful tool
norepinephrine, serotonin, dopamine, and his- for assessing subjective sleepiness and the re-
tamine (Fig. 2). Orexins have long-lasting effects sponse to medications.48 The multiple sleep la-
on target neurons, and this sustained activity tency test and the maintenance of wakefulness
may help maintain wakefulness throughout the test, which measures alertness during the day,
day. Conversely, loss of orexin signaling in nar- can provide complementary, objective measures
colepsy may result in inconsistent activity in these of the degree of sleepiness.9
wakefulness-promoting brain regions, resulting Cataplexy often is reduced with a low dose of
in frequent lapses into sleep.37 an antidepressant. Extended-release venlafaxine
Orexins increase activity in brain regions that generally has an acceptable side-effect profile,
suppress REM sleep, and reduced orexin signal- and it is effective for most of the day. The anti-
ing in narcolepsy may therefore enable elements cholinergic effects of clomipramine (e.g., seda-

2658 n engl j med 373;27 nejm.org  December 31, 2015

The New England Journal of Medicine

Downloaded from nejm.org at UNIVERSITY OF WOLLONGONG on December 30, 2015. For personal use only. No other uses without permission.
Copyright © 2015 Massachusetts Medical Society. All rights reserved.
 Narcolepsy

tion and dry mouth) can be bothersome, but it A Mechanisms of Sleepiness in Narcolepsy
potently suppresses cataplexy and can be quite
helpful when taken before an event that is likely
to trigger cataplexy, such as a party or wedding.
Norepinephrine and serotonin suppress REM
sleep, and by blocking reuptake of these neuro-
transmitters, antidepressants reduce REM sleep
and substantially reduce cataplexy. If withdrawal Orexin loss

is necessary, they should be discontinued gradu-

ally, since sudden withdrawal can produce severe Acetylcholine
rebound cataplexy. Hypnagogic hallucinations Histamine
Dopamine
and sleep paralysis are usually managed by pa- Inconsistent
Serotonin

tient education; when these symptoms are se- neurotransmitter Norepinephrine

release
vere, however, the same antidepressants may lead
Sleepiness,
to a reduction in symptoms. lapses into sleep, poor
Unlike the medications discussed above, which regulation of REM sleep

are taken in the morning or several times during B Mechanisms of Cataplexy in Narcolepsy
the day, sodium oxybate (the sodium salt of
γ-hydroxybutyrate) is a highly sedating liquid
Strong positive
given at bedtime and 2.5 to 4 hours later. It pro- emotions
duces very deep non-REM sleep, probably by ac-
tivating γ-aminobutyric acid type B receptors.49
After several weeks of treatment, sodium oxy- MEDIAL
PREFRONTAL
bate usually reduces daytime sleepiness and CORTEX Orexin loss Reduced activity in
brain regions that
cataplexy50 through an unknown mechanism. inhibit REM sleep
Because sodium oxybate can produce nausea, Increased activity in
dizziness, and other side effects at typical doses, Amygdala brain regions that
and coma with overdose, it is generally used only promote REM sleep

in persons with moderate-to-severe sleepiness or

Motor
cataplexy. neurons inhibited
Even with good nighttime sleep, daytime naps,
and appropriate medications, many people with Cataplexy
narcolepsy still have some lingering daytime
Figure 2. Neurobiologic Mechanisms of Narcolepsy.
sleepiness and inattentiveness. Thus, it is impor-
The neurons that normally produce orexin help sustain wakefulness by
tant for people with narcolepsy to have honest
stimulating a variety of wakefulness-promoting neurons in the cortex, brain
discussions with their family members and med- stem, and basal forebrain (Panel A). With loss of the orexin-producing neu-
ical providers about lifestyle choices. For exam- rons in people with narcolepsy, neurons in these target regions may fire
ple, the risk of motor vehicle accidents among less consistently, leading to sleepiness and lapses into sleep. These neu-
people with narcolepsy is increased by a factor rons also excite brain-stem regions that suppress rapid-eye-movement
(REM) sleep (Panel B). Strong positive emotions are thought to activate
of three to five; some people may therefore
neurons in the medial prefrontal cortex that excite the orexin-producing
choose to take a stimulant before driving, drive neurons and the amygdala. With loss of the orexin neurons in narcolepsy,
only for short periods, or not drive at all. With imbalance in this pathway inhibits brain regions that suppress REM sleep,
regard to work, persons with narcolepsy can thus permitting the activation of descending pathways that strongly inhibit
thrive in stimulating environments such as teach- motor neurons; the result is partial or complete cataplexy. Pathways shown
in green are excitatory; those shown in red are inhibitory. Open circles indi-
ing, but sedentary jobs that require sustained
cate neuron-cell bodies, and dotted lines indicate reduced activity.
attention may be a poor fit.

therapies for this disorder, but many questions

F u t ur e Dir ec t ions
remain unanswered. Investigators are now work-
Over the past several years, researchers and cli- ing to clarify the pathologic process that de-
nicians have made great advances in under- stroys the orexin-producing neurons, because
standing narcolepsy and in developing effective this information might provide an opportunity

n engl j med 373;27 nejm.org December 31, 2015 2659

The New England Journal of Medicine
Downloaded from nejm.org at UNIVERSITY OF WOLLONGONG on December 30, 2015. For personal use only. No other uses without permission.
Copyright © 2015 Massachusetts Medical Society. All rights reserved.
 2660
Table 2. Drugs Used in the Treatment of Narcolepsy.

Drug Dose in Adults* Common Side Effects Serious Risks and Side Effects Relative Cost†
For excessive daytime sleepiness
Modafinil 100–400 mg every morning, Headache, anxiety, nausea, insomnia Severe rash Moderate
or 200 mg twice daily
Armodafinil 150–250 mg every morning, Headache, anxiety, nausea, insomnia Severe rash Moderate
or 125 mg (half of a 250-mg tablet)
twice daily
Methylphenidate 10–30 mg twice daily, or 20 mg sus- Reduced appetite, nausea, headache, Potential for abuse; psychosis, mania, Low
The

tained-release formulation every ­insomnia seizures, cardiovascular effects

morning with an additional
10–20 mg every afternoon
Dextroamphetamine 5–30 mg twice daily, or 10 mg of a sus- Reduced appetite, nausea, headache, High potential for abuse; psychosis, Low
tained-release formulation every ­insomnia mania, seizures, cardiovascular
morning with an additional ­effects
10–20 mg every afternoon
Amphetamine–dextroamphetamine 10–30 mg twice daily, or 20 mg of a sus- Reduced appetite, nausea, headache, High potential for abuse; psychosis, Low
tained-release formulation twice ­insomnia mania, seizures, cardiovascular
daily ­effects
Sodium oxybate (sodium salt 2.25–4.5 g at bedtime and an additional Nausea, dizziness, urinary incontinence, Potential for abuse; confusion, psy- High
of γ-hydroxybutyrate) 2.25–4.5 g given 2.5 to 4 hr later sleepwalking, morning sedation, chosis, severe sedation or coma
n e w e ng l a n d j o u r na l

anxiety with overdose

The New England Journal of Medicine

of

For cataplexy
Venlafaxine 37.5–75 mg twice daily, or 37.5–150 mg Transient nausea, headache, insomnia; in- None Low
of an extended-release formulation crease in blood pressure when adminis-
every morning tered in higher doses

n engl j med 373;27 nejm.org  December 31, 2015

Copyright © 2015 Massachusetts Medical Society. All rights reserved.

Fluoxetine 20–80 mg every morning Nausea, dry mouth, insomnia None Low
m e dic i n e

Clomipramine 10–150 mg at bedtime or each Dry mouth, constipation, sweating, dizzi- Cardiotoxicity, hypotension, seizures Low
morning ness, somnolence, weight gain, ortho-
static hypotension
Sodium oxybate (sodium salt 2.25–4.5 g at bedtime and an additional Nausea, dizziness, urinary incontinence, Potential for abuse; confusion, psy- High
of γ-hydroxybutyrate) 2.25–4.5 g given 2.5–4 hr later sleepwalking, morning sedation, chosis, severe sedation or coma
anxiety with overdose

* Wakefulness-promoting medications are usually taken in the morning; if necessary, additional doses can be taken at midday or in the early afternoon, 30 to 60 minutes before the morn-
ing dose wears off. The second dose should not be given late in the afternoon, because such late doses can cause insomnia. Sodium oxybate should not be combined with alcohol or
sedatives. Adapted from Scammell.45
† Relative costs range from low to high.

Downloaded from nejm.org at UNIVERSITY OF WOLLONGONG on December 30, 2015. For personal use only. No other uses without permission.
 Narcolepsy

to prevent or cure narcolepsy. Others are study- some early progress in developing small-molecule
ing how the loss of these neurons results in orexin agonists. Although narcolepsy affects
sleepiness, cataplexy, and obesity; understand- many facets of patients’ daily lives, current treat-
ing the neurobiologic process involved may lead ments can be quite effective, and even better
to new therapeutic opportunities. The restora- treatments may soon be available.
tion of orexin signaling in the brain may be an Dr. Scammell reports receiving consulting fees from Jazz
Pharmaceuticals, Eisai, Vertex Pharmaceuticals, Prexa Pharma-
ideal therapy, akin to administering insulin to ceuticals, and Merck, fees for providing expert testimony on
persons with type 1 diabetes. Currently, such behalf of Jazz Pharmaceuticals in a case involving a patent for
treatment would be difficult, since the orexin sodium oxybate, and grant support from Eisai. No other poten-
tial conflict of interest relevant to this article was reported.
neuropeptides cannot easily cross the blood– Disclosure forms provided by the author are available with the
brain barrier; however, researchers have made full text of this article at NEJM.org.

References
1. Thorpy MJ, Krieger AC. Delayed diag- 13. International classification of sleep HLA-DPB1 and HLA class I confer risk of
nosis of narcolepsy: characterization and disorders. 3rd ed. Darien, IL:​American and protection from narcolepsy. Am J Hum
impact. Sleep Med 2014;​15:​502-7. Academy of Sleep Medicine, 2014. Genet 2015;​96:​136-46.
2. Plazzi G, Pizza F, Palaia V, et al. Com- 14. Sakurai T, Amemiya A, Ishii M, et al. 26. Han F, Faraco J, Dong XS, et al. Ge-
plex movement disorders at disease onset Orexins and orexin receptors: a family of nome wide analysis of narcolepsy in China
in childhood narcolepsy with cataplexy. hypothalamic neuropeptides and G pro- implicates novel immune loci and reveals
Brain 2011;​134:​3477-89. tein-coupled receptors that regulate feed- changes in association prior to versus
3. Wamsley E, Donjacour CE, Scammell ing behavior. Cell 1998;​92:​573-85. after the 2009 H1N1 influenza pandemic.
TE, Lammers GJ, Stickgold R. Delusional 15. de Lecea L, Kilduff TS, Peyron C, et al. PLoS Genet 2013;​9(10):​e1003880.
confusion of dreaming and reality in nar- The hypocretins: hypothalamus-specific 27. Faraco J, Lin L, Kornum BR, et al.
colepsy. Sleep 2014;​37:​419-22. peptides with neuroexcitatory activity. ImmunoChip study implicates antigen
4. Roth T, Dauvilliers Y, Mignot E, et al. Proc Natl Acad Sci U S A 1998;​95:​322-7. presentation to T cells in narcolepsy. PLoS
Disrupted nighttime sleep in narcolepsy. 16. Lin L, Faraco J, Li R, et al. The sleep Genet 2013;​9(2):​e1003270.
J Clin Sleep Med 2013;​9:​955-65. disorder canine narcolepsy is caused by 28. Han F, Lin L, Warby SC, et al. Narco-
5. Pizza F, Tartarotti S, Poryazova R, a mutation in the hypocretin (orexin) re- lepsy onset is seasonal and increased fol-
Baumann CR, Bassetti CL. Sleep-disor- ceptor 2 gene. Cell 1999;​98:​365-76. lowing the 2009 H1N1 pandemic in China.
dered breathing and periodic limb move- 17. Chemelli RM, Willie JT, Sinton CM, et Ann Neurol 2011;​70:​410-7.
ments in narcolepsy with cataplexy: a sys- al. Narcolepsy in orexin knockout mice: 29. Aran A, Lin L, Nevsimalova S, et al.
tematic analysis of 35 consecutive patients. molecular genetics of sleep regulation. Elevated anti-streptococcal antibodies in
Eur Neurol 2013;​70:​22-6. Cell 1999;​98:​437-51. patients with recent narcolepsy onset.
6. Ohayon MM. Narcolepsy is compli- 18. Peyron C, Faraco J, Rogers W, et al. Sleep 2009;​32:​979-83.
cated by high medical and psychiatric co- A mutation in a case of early onset narco- 30. Nohynek H, Jokinen J, Partinen M, et al.
morbidities: a comparison with the gener- lepsy and a generalized absence of hypo- AS03 adjuvanted AH1N1 vaccine associ-
al population. Sleep Med 2013;​14:​488-92. cretin peptides in human narcoleptic ated with an abrupt increase in the inci-
7. Andlauer O, Moore H, Jouhier L, et al. brains. Nat Med 2000;​6:​991-7. dence of childhood narcolepsy in Finland.
Nocturnal rapid eye movement sleep la- 19. Thannickal TC, Moore RY, Nienhuis R, PLoS One 2012;​7(3):​e33536.
tency for identifying patients with narco- et al. Reduced number of hypocretin neu- 31. Partinen M, Saarenpää-Heikkilä O,
lepsy/hypocretin deficiency. JAMA Neurol rons in human narcolepsy. Neuron 2000;​ Ilveskoski I, et al. Increased incidence and
2013;​70:​891-902. 27:​469-74. clinical picture of childhood narcolepsy
8. Reiter J, Katz E, Scammell TE, Maski 20. Crocker A, España RA, Papadopoulou following the 2009 H1N1 pandemic vac-
K. Usefulness of a nocturnal SOREMP for M, et al. Concomitant loss of dynorphin, cination campaign in Finland. PLoS One
diagnosing narcolepsy with cataplexy in a NARP, and orexin in narcolepsy. Neurol- 2012;​7(3):​e33723.
pediatric population. Sleep 2015;​38:​859- ogy 2005;​65:​1184-8. 32. Mahlios J, De la Herrán-Arita AK,
65. 21. Thannickal TC, Nienhuis R, Siegel JM. Mignot E. The autoimmune basis of nar-
9. Littner MR, Kushida C, Wise M, et al. Localized loss of hypocretin (orexin) cells colepsy. Curr Opin Neurobiol 2013;​ 23:​
Practice parameters for clinical use of the in narcolepsy without cataplexy. Sleep 767-73.
multiple sleep latency test and the main- 2009;​32:​993-8. 33. Nishino S, Kanbayashi T. Symptom-
tenance of wakefulness test. Sleep 2005;​ 22. Rogers AE, Meehan J, Guilleminault C, atic narcolepsy, cataplexy and hypersom-
28:​113-21. Grumet FC, Mignot E. HLA DR15 (DR2) nia, and their implications in the hypo-
10. Trotti LM, Staab BA, Rye DB. Test- and DQB1*0602 typing studies in 188 nar- thalamic hypocretin/orexin system. Sleep
retest reliability of the multiple sleep la- coleptic patients with cataplexy. Neurology Med Rev 2005;​9:​269-310.
tency test in narcolepsy without cataplexy 1997;​48:​1550-6. 34. Estabrooke IV, McCarthy MT, Ko E,
and idiopathic hypersomnia. J Clin Sleep 23. Tafti M, Hor H, Dauvilliers Y, et al. et al. Fos expression in orexin neurons
Med 2013;​9:​789-95. DQB1 locus alone explains most of the varies with behavioral state. J Neurosci
11. Goldbart A, Peppard P, Finn L, et al. risk and protection in narcolepsy with 2001;​21:​1656-62.
Narcolepsy and predictors of positive cataplexy in Europe. Sleep 2014;​ 37:​
19- 35. Lee MG, Hassani OK, Jones BE. Dis-
MSLTs in the Wisconsin Sleep Cohort. 25. charge of identified orexin/hypocretin
Sleep 2014;​37:​1043-51. 24. Bourgin P, Zeitzer JM, Mignot E. CSF neurons across the sleep-waking cycle.
12. Baumann CR, Mignot E, Lammers GJ, hypocretin-1 assessment in sleep and J Neurosci 2005;​25:​6716-20.
et al. Challenges in diagnosing narcolepsy neurological disorders. Lancet Neurol 36. Blouin AM, Fried I, Wilson CL, et al.
without cataplexy: a consensus statement. 2008;​7:​649-62. Human hypocretin and melanin-concen-
Sleep 2014;​37:​1035-42. 25. Ollila HM, Ravel JM, Han F, et al. trating hormone levels are linked to emo-

n engl j med 373;27 nejm.org  December 31, 2015 2661

The New England Journal of Medicine
Downloaded from nejm.org at UNIVERSITY OF WOLLONGONG on December 30, 2015. For personal use only. No other uses without permission.
Copyright © 2015 Massachusetts Medical Society. All rights reserved.
 Narcolepsy

tion and social interaction. Nat Commun 42. Donadio V, Liguori R, Vandi S, et al. A closer look at amphetamine-induced
2013;​4:​1547. Lower wake resting sympathetic and car- reverse transport and trafficking of the
37. Diniz Behn CG, Klerman EB, Mochi­ diovascular activities in narcolepsy with dopamine and norepinephrine transport-
zuki T, Lin SC, Scammell TE. Abnormal cataplexy. Neurology 2014;​83:​1080-6. ers. Mol Neurobiol 2009;​39:​73-80.
sleep/wake dynamics in orexin knockout 43. Mignot EJ. A practical guide to the 48. Johns MW. A new method for measur-
mice. Sleep 2010;​33:​297-306. therapy of narcolepsy and hypersomnia ing daytime sleepiness: the Epworth Sleep-
38. Burgess CR, Scammell TE. Narcolepsy: syndromes. Neurotherapeutics 2012;​ 9:​ iness Scale. Sleep 1991;​14:​540-5.
neural mechanisms of sleepiness and 739-52. 49. Vienne J, Bettler B, Franken P, Tafti
cataplexy. J Neurosci 2012;​32:​12305-11. 44. Lecendreux M. Pharmacological man- M. Differential effects of GABAB recep-
39. Oishi Y, Williams RH, Agostinelli L, agement of narcolepsy and cataplexy in tor subtypes, gamma-hydroxybutyric acid,
et al. Role of the medial prefrontal cor- pediatric patients. Paediatr Drugs 2014;​ and baclofen on EEG activity and sleep
tex in cataplexy. J Neurosci 2013;​33:​9743- 16:​363-72. regulation. J Neurosci 2010;​ 30:​
14194-
51. 45. Scammell TE. Treatment of narcolepsy. 204.
40. Burgess CR, Oishi Y, Mochizuki T, UpToDate, 2015 (http://www​.uptodate​.com/​ 50. Boscolo-Berto R, Viel G, Montagnese
Peever JH, Scammell TE. Amygdala lesions contents/​t reatment-of-narcolepsy). S, Raduazzo DI, Ferrara SD, Dauvilliers Y.
reduce cataplexy in orexin knock-out mice. 46. US Modafinil in Narcolepsy Multi- Narcolepsy and effectiveness of gamma-
J Neurosci 2013;​33:​9734-42. center Study Group. Randomized trial of hydroxybutyrate (GHB): a systematic re-
41. Meletti S, Vaudano AE, Pizza F, et al. modafinil for the treatment of pathologi- view and meta-analysis of randomized
The brain correlates of laugh and cata- cal somnolence in narcolepsy. Ann Neu- controlled trials. Sleep Med Rev 2012;​16:​
plexy in childhood narcolepsy. J Neurosci rol 1998;​43:​88-97. 431-43.
2015;​35:​11583-94. 47. Robertson SD, Matthies HJ, Galli A. Copyright © 2015 Massachusetts Medical Society.

We are seeking a Deputy Editor to join our editorial team at the New England Journal
of Medicine. The Deputy Editor will review, select, and edit manuscripts for the
Journal as well as participate in planning the Journal’s content. He or she will also
ensure that the best available research is submitted to the Journal, identify potential
Journal contributors, and invite submissions as appropriate.
Applicants must have an M.D. degree and at least 5 years of research experience.
The successful candidate will be actively involved in patient care and will continue
that involvement (approximately 20% of time) while serving in this role.
To apply, visit https://mmscareers.silkroad.com.

2662 n engl j med 373;27 nejm.org  December 31, 2015

The New England Journal of Medicine

Downloaded from nejm.org at UNIVERSITY OF WOLLONGONG on December 30, 2015. For personal use only. No other uses without permission.
Copyright © 2015 Massachusetts Medical Society. All rights reserved.