Clinical manifestations

Presentation and course

Narcolepsy is characterized by excessive daytime sleepiness, which is the primordial symptom of this disorder. All patients with narcolepsy have excessive daytime sleepiness. In the classic form, now called narcolepsy type 1 (alternate names: hypocretin deficiency syndrome, narcolepsy-cataplexy, narcolepsy with cataplexy), cataplexy is added, often in association with sleep paralysis and hypnagogic hallucinations. Men and women are equally affected. Onset is usually gradual, typically in the second or third decade of life, although childhood onset may occur, and some patients are not diagnosed until middle age (119). Change in sleep schedule, psychological stress, head trauma, or infection sometimes appears to precipitate symptoms, but these associations may be coincidental.

Excessive sleepiness in narcolepsy is associated with repeated naps or lapses into sleep. Sleep tends to occur in boring or monotonous situations and may be temporarily forestalled by physical activity and mental stimulation. Following brief naps, the patient awakens refreshed. Sleep attacks occur on a background of essentially continuous drowsiness, and although the severity of sleepiness may vary considerably, remissions are exceedingly rare. Patients often report episodes of amnesia with "automatic behavior" during which they may carry out complex nonsensical activities, such as mixing inappropriate foods, writing nonsense words, or driving to the wrong destination or missing exits. Accidents due to sleepiness and automatic behavior may occur while driving or operating dangerous equipment. Sleepiness decreases with physical activity.

Cataplexy, a unique feature of narcolepsy, is characterized by sudden loss of muscle tone, usually provoked by strong emotion, particularly laughter. Consciousness is preserved and memory is intact. Duration of cataplexy ranges from a few seconds to several minutes, and recovery is complete. The loss of tone varies in severity from complete postural collapse to mild weakness with head droop, facial sagging, dropping of the jaw, slurred speech, or buckling of the knees. During the cataplectic attack, tendon reflexes disappear, and the F-wave of the nerve conduction exam also disappears (121). Respiratory and oculomotor muscles are not affected. The frequency of cataplexy varies from one to two events per year to many episodes per day. In some patients, episodes of cataplexy may occur almost continuously, a condition referred to as "status cataplecticus." Some patients learn to avoid conditions or situations that may lead to cataplexy. Cataplexy rarely precedes the onset of excessive sleepiness but may develop simultaneously with sleepiness or with a delay of 1 to 30 years. Isolated cataplexy has been described in the context of nondiagnostic multiple sleep latency tests and normal CSF-hypocretin-1 levels (greater than 217 pg/mL). One patient gradually developed excessive daytime sleepiness and low CSF-hypocretin-1 (less than 110 pg/ mL) (44).

Hypnagogic hallucinations are vivid perceptual experiences that occur at sleep onset, often associated with fear or dread. Hallucinations occurred more frequently and with more motor and multimodal manifestations in narcolepsy with cataplexy (59%) than in narcolepsy without cataplexy (28%) (74). Sleep paralysis is a transient, generalized inability to move or to speak during the transition between sleep and wakefulness. The experiences are often frightening. In addition, patients with narcolepsy often report memory lapses, diplopia, blurred vision, and ptosis, all of which are probably consequences of chronic sleepiness. In a controlled study, memory processes were preserved in patients with narcolepsy type 1, and the authors concluded that memory complaints may not be associated with an objective memory deficit (91). Depression and attention deficits could explain the subjective memory complaints. The predominant theta electroencephalography rhythm during sleep paralysis suggests that the brain during sleep paralysis is in a dreaming state (81).

Disrupted nocturnal sleep with frequent awakenings is common, a condition that worsens as age advances and may contribute to daytime sleepiness. Many patients report that narcolepsy seriously affects interpersonal, marital, work, and social relationships.

Narcolepsy type 1 may be due to a medical condition. Central nervous system disorders, such as autoimmune or paraneoplastic disorders associated with anti-Ma2 or antiaquaporin4 antibodies, head trauma, and tumors or other lesions of the hypothalamus may be associated with narcolepsy. If undetectable hypocretin-1 levels are reported and the condition fulfills diagnostic criteria for narcolepsy type 1, a diagnosis is made of narcolepsy type 1 due to medical condition (ICSD-3).

A group of patients with otherwise typical features of narcolepsy do not have cataplexy (narcolepsy type 2), a condition also referred to as "monosymptomatic narcolepsy," "narcolepsy without cataplexy," "atypical narcolepsy," or "hypersomnia with REM sleep abnormalities." Patients with sleepiness and low (< 110 pg/ml) or absent CSF hypocretin-1 levels are classified as having narcolepsy type 1, even if they do not manifest cataplexy. Sleep paralysis, hypnagogic hallucinations, or automatic behavior may be present. Such patients are a heterogeneous group, some of whom are in the evolving phase of classic narcolepsy and go on to develop cataplexy. Reports of bizarre dreams may lead to a diagnosis of psychosis, particularly in children. The coincidence of narcolepsy and schizophrenia is possible although rare (0.5 to 9 cases in a population of 1 million) (69).

Patients with narcolepsy may have a distinct odor. In one experiment, sweat samples from narcoleptic and healthy controls were tested independently by two trained dogs and their positive or negative detection compared to the gold standard diagnosis for narcolepsy (42). Eleven out of 12 patients with narcolepsy were detected by trained dogs, whereas only 3 out of 22 control subjects proved positive. The authors concluded that narcoleptic patients have a distinct typical odor that trained dogs can detect. The development of an olfactory test could be a useful adjunct in the screening of narcolepsy while opening a new research area.

Approximately 15% of narcoleptic patients develop REM sleep behavior disorder (RBD) (21). The age of onset is younger than in the other forms of chronic RBD. Clinical features are the same as in the other forms of chronic RBD, but the frequency of the episodes is less marked. Clinical variants include RBD induced or worsened by pharmacological agents, most of them being used to treat cataplexy, RBD in narcoleptic children, and RBD in the context of symptomatic narcolepsy (126).

Children with narcolepsy and cataplexy showed an abrupt increase of sleep time during the 24 hours with generalized hypotonia and motor overactivity in one study (106). At follow-up, sleep time and nocturnal sleep latency shortened in the absence of other polysomnographic or clinical changes. Over time, the picture of cataplexy evolved into a classic presentation including episodes of brief muscle weakness triggered by emotions, whereas total sleep time across the 24 hours decreased, returning to more age-appropriate levels. Children with narcolepsy type 1 close to disease onset showed persistent hypotonia with prominent facial involvement (cataplectic facies) and hyperkinetic movement abnormalities that increased during emotional stimulation. This clinical presentation progressively vanished leading to the typical picture of cataplexy. Childhood narcolepsy type 1 also showed behavioral abnormalities and psychiatric disorders with depressive feelings and hyperactive/aggressive behavior, even psychotic features. The association with obesity and precocious puberty in some children may reflect a wide hypothalamic derangement secondary to hypocretin neuronal loss (113). Caregivers may be the only source of information in younger children (13). Teachers and caregivers may report bad behavior, tiredness, laziness, hyperactivity, and poor concentration rather than sleepiness.

Prognosis and complications

Narcoleptic patients have an increased risk of accidents while operating automobiles and motor equipment. Other psychosocial complications include interpersonal and marital difficulties, loss of employment and academic opportunities, and depression. Personality changes and memory problems are common. Some patients appear to adapt well to narcolepsy; however, for most patients, the disease has a pervasive effect on social and occupational functioning.

The prognosis in narcolepsy is variable. Excessive sleepiness is almost always a lifelong problem. In some patients, cataplexy, hypnagogic hallucinations, and sleep paralysis become less problematic over time, but the REM sleep behavior disorder, which occurs in some patients with narcolepsy, may become more challenging as time evolves. Some patients with prominent hallucinations may develop delusions associated with the hallucinations and may be misdiagnosed with a schizophreniform illness (43). Many patients appear to cope better with sleepiness and cataplexy after several years of symptoms, and avoidance of difficult situations may lessen the stress associated with these problems. In some patients, periodic limb movement disorder and obstructive sleep apnea may appear years after diagnosis. Insomnia may progress as age advances, and occasional patients have been diagnosed with insomnia before the performance of appropriate diagnostic tests.

Psychosocial problems are frequent in narcoleptics. These include increased incidence of accidents, poor school performance, occupational problems, and interpersonal problems. Anxiety, depression, social withdrawal, and reduced self-esteem are present in at least 30% of patients and arise from a complex interaction between basic mechanisms of the disorder and psychoreactive mechanisms. Another study has shown that narcolepsy causes elevated direct and indirect costs (eg, unemployment) (41). Narcolepsy is associated with significant comorbid psychiatric illness burden and higher psychiatric medication usage compared with the non-narcolepsy population (118).

One study quantifying the problem of diagnostic delay for patients with narcolepsy in the United States highlights that symptoms are more likely to be missed if they develop before 18 years of age (87). Disease burden is high because of problems with fatigue, cognition, and persistence of residual symptoms despite treatment.

Clinical vignette

Mr. X remembered being a sleepyhead in high school. He would sleep through study hall and had major difficulty staying awake in morning classes. College had been a struggle as well. Now, 25 years old and recently married, Mr. X decided to visit a sleep specialist because of insomnia. He had no difficulty falling asleep but lately there were many awakenings through the night that were disturbing his newlywed wife. On several occasions he had awakened in the middle of the night unable to move or shout despite a sense of growing fright. The episodes had lasted less than a minute and had caused anxiety. On specific questioning, Mr. X acknowledged falling asleep driving only to be awakened by the noise caused when going over the rumble-strips on the side of the road. He would take naps almost daily that were refreshing and always associated with vivid dreams. He denied episodes of sudden falls but remembered that during his Catholic wedding ceremony, he suddenly became unable to stand from a kneeling position, and his head keeled over as if in deep prayer, an action that witnesses attributed to the intensity of the moment.

The neurologist-diplomate in sleep disorders evaluating Mr. X elicited a normal neurologic examination. He considered a working diagnosis of narcolepsy and ordered nocturnal polysomnography followed by a multiple sleep latency test (MSLT). He advised Mr. X not to drive unless accompanied and always during daytime hours as well as to avoid driving for more than 30 minutes or on long stretches of road. He also recommended 20-minute naps after lunch and when returning home from work. Medication would not be prescribed until the results of the sleep study became available. The study was done two weeks later and showed a nocturnal short-onset REM sleep latency of 7.5 minutes. The proportion of nocturnal REM sleep was 27%, which was high for the standards of the laboratory. Sleep apnea and periodic limb movements were not found. The MSLT showed presence of REM sleep in all four naps with a latency of five minutes. Overall sleep latency was two minutes. Given these results, the specialist made a diagnosis of narcolepsy with possible cataplexy and decided not to pursue a CSF hypocretin analysis in light of the diagnostic certainty of the polysomnographic results. During the follow-up visit the specialist gave Mr. X and his wife an overview of the test results, the prognosis, and management. He prescribed 200 mg of modafinil in the morning, a dose that could be increased to 400 mg if the previous dose became insufficient. Once again, he counseled the patient not to drive under certain conditions and gave him brochures with additional information on narcolepsy as well as the addresses of various narcolepsy organizations and support groups. A second follow-up visit was scheduled for three months later to review the efficacy of the medication and consider sodium oxybate.

Biological basis

The biological process responsible for the deletion of hypocretin neurons in the hypothalamus in narcolepsy type-I is suspected to be autoimmune. Narcolepsy type 2 is less well-defined disorder, with a variable phenotype and evolution, and few reliable biomarkers (16). There is a dearth of knowledge about this disorder, and its etiology remains unclear and needs to be further explored.

A significant aspect of the burden of narcolepsy relates to abundant comorbid conditions, including emotional, metabolic, sleep, and immune health conditions (54).

Etiology and pathogenesis

The combination of narcolepsy with cataplexy is closely associated with the human leukocyte antigen (HLA) subtypes DR2/DRB1*1501 and DQB1*0602. These two subtypes are always found together in Caucasians and Asians, but in blacks, DQB1*0602 is more specifically associated with narcolepsy. Almost all patients with cataplexy are positive for DQB1*0602, compared with 12% to 38% of the general population who have this HLA subtype (95). Persons homozygous for DQB1-0602 are at increased risk for narcolepsy but do not have more severe symptoms (104). The presence of typical cataplexy is associated with a 90% DQB1*0602 positivity, a percentage that is much lower (30% to 45%) in narcoleptics with atypical or no cataplexy (76). Rare cases of sporadic and familial narcolepsy with typical cataplexy, negative for both DR2 and DQB1*0602, have been reported (96).

However, narcolepsy can occur in HLA-DR1501-negative individuals, and HLA-DR1501 is present in 15% to 35% of the normal population, indicating that the HLA-DR1501 antigen is neither sufficient nor necessary for the development of narcolepsy (95). The existence of twins discordant for narcolepsy suggests that nongenetic, environmental factors have a critical role in the development of narcolepsy.

Although the HLA-associated susceptibility gene is present in the vast majority of narcoleptic patients, it is not sufficient to induce the disorder. The existence of families in which HLA markers do not cosegregate with narcolepsy suggests that a second gene plays a role.

Some family members of persons with narcolepsy-cataplexy may have all of the features of narcolepsy except cataplexy, whereas others may have symptoms and laboratory findings consistent with idiopathic hypersomnia. These family studies suggest that the narcolepsy genotype may be expressed as more than one phenotype (60).

In a study by Chen and colleagues, the frequency of narcolepsy in first-degree relatives was 85.3 times greater than that of the general population (31). Nearly 30% of relatives fulfilled the criteria of narcolepsy spectrum, defined as presence of abnormal multiple sleep latency test results, with shortened mean sleep latency (≤ 8 minutes) and/or presence of sleep onset rapid eye movement sleep periods (SOREMPs) unexplained by medical or psychiatric disorders or other sleep disorders. The findings further suggest that there is a spectrum of narcolepsy phenotypes and manifestations ranging from the most severe clinical symptoms of narcolepsy-cataplexy to narcolepsy without cataplexy, idiopathic hypersomnia, and the asymptomatic presence of biomarkers such as shortened mean sleep latencies in the MSLT and SOREMPs. These findings suggest that the population at risk for narcolepsy and narcolepsy-like symptoms is much larger than previously thought. The characterization of the narcolepsy spectrum as a diagnostic entity would enable the precise diagnosis of many obscure forms of hypersomnia and facilitate management of those at risk. In a subsequent study to determine the familial aggregation of narcolepsy, Wing and colleagues studied narcolepsy probands and their first-degree relatives and compared them to age- and sex-matched unrelated healthy controls (139). Among the relatives of 33 probands with narcolepsy and 81 first-degree relatives, 12.3% were diagnosed with narcolepsy and 39.5% had narcolepsy spectrum, as defined by unexplained abnormal MSLT (shortened MSL and SOREMP) results. The authors concluded that the familial risk of narcolepsy among first-degree relatives is high because of a spectrum of narcolepsy features among relatives, ranging from full clinical tetrads to asymptomatic abnormal MSLT findings.

The broad symptom presentation of narcolepsy, which is much more encompassing than the classical "tetrad" of sleepiness, cataplexy, hallucinations, and sleep paralysis, may delay the diagnosis. Furthermore, the presentation of symptoms can be markedly different between children and adults. One study provides a detailed description of the broad and dynamic symptom spectrum of narcolepsy, with specific attention to the different manifestations in both adults and children, which may aid in the prompt diagnosis of narcolepsy (110).

Immunology, cell biology, and pathology. Concern has been raised over reports of narcolepsy in northern Europe following H1N1 vaccination (34). In a retrospective analysis of narcolepsy onset in subjects diagnosed in Beijing, China, between 1998 and 2010, Han and colleagues found that the occurrence of narcolepsy onset was seasonal and significantly influenced by month and calendar year (55). Onset was least frequent in November and most frequent in April, with a 6.7-fold increase from trough to peak. A 3-fold increase in narcolepsy onset was observed following the 2009 H1N1 winter influenza pandemic, and the increase was unlikely to be explained by increased vaccination as only 5.6% of patients recalled receiving an H1N1 vaccination. The authors concluded that in China, narcolepsy onset is highly correlated with seasonal and annual patterns of upper airway infections, including H1N1 influenza. In a systematic review and metaanalysis to analyze the magnitude of H1N1 vaccination-related risk, the authors showed that the risk appears to be limited to only one vaccine (Pandemrix®) (122). The relative risk of narcolepsy during the first year after vaccination was increased 5- to 14-fold in children and adolescents and 2- to 7-fold in adults, with an attributable risk in children and adolescents of 1 per 18,400 vaccine doses. The authors concluded that benefits of immunization outweigh the risk of vaccination-associated narcolepsy because narcolepsy remains a rare disorder. In a study in Slovakia of 61 patients with narcolepsy diagnosed from 2000 to 2017, patients showed double the excess prevalence in mental disorders (OR 2.15, p < 0.05) and dyslipidemia (OR 2.22, p < 0.05) (47). There was a mild increase in autoimmune disorders and allergy in the narcolepsy group, and the thyroiditis of Hashimoto was the most frequent autoimmune disorder.

Discoveries in dogs and humans suggest that two peptides found in the hypothalamus and named hypocretin I and II, also called orexin A and B (120), are important in the pathophysiology of narcolepsy (127). Hypocretin I and II are synthesized from a precursor by a small number of cells in the posterior and lateral hypothalamus (39). These neurons project to monoaminergic and cholinergic centers of the ascending reticular activating system. Narcolepsy in dogs is caused by a deletion in the hypocretin 2-receptor gene (75), whereas REM sleep episodes while awake and cataplexy are observed in hypocretin knockout mice (30). These findings suggest that loss of the physiologic hypocretin excitatory influence on histaminergic, dopaminergic, and cholinergic systems may reduce thalamocortical arousal and underlie manifestations typically associated with narcolepsy. In a study of the histopathology of the hypothalamus in patients with narcolepsy, hypocretin-producing cells were reduced by 85% to 95% (129). Most cases of narcolepsy with cataplexy are associated with the loss of approximately 50,000 to 100,000 hypothalamic neurons containing hypocretin. Studies of hypocretin content in CSF in humans correlate with the evolving hypothesis implicating hypocretin in narcolepsy (96). Hypocretin was undetectable in CSF in seven of nine patients with narcolepsy (100). In another study of patients with narcolepsy, content of hypocretin I in CSF was either absent or reduced, whereas serum levels were normal, pointing to the brain locality of the disorder (33). In one case study, a patient with a hypothalamic infarct after resection of a craniopharyngioma developed sleepiness, and the subsequent study of hypocretin in CSF revealed a low concentration (124). Sleepiness following hypothalamic injury in the course of resection of an astrocytoma was also associated with a low concentration of hypocretin in CSF (12). Narcolepsy-cataplexy developed in an acromegalic man two weeks following irradiation of the pituitary gland. Interestingly, the patient had normal CSF concentrations of hypocretin, suggesting that the damage was inflicted to hypocretin receptors rather than secretors of the neurotransmitter (40). There is a case of a patient with multiple sclerosis with reversible ophthalmoplegia and hemianopsia who developed hypersomnia and had undetectable levels of Hcrt in CSF. MRI of the brain showed demyelinating lesions involving hypothalamic nuclei (61). There is a report of a 53-year-old man presenting with depressed alertness and severe excessive sleepiness in the setting of neurosarcoidosis. Neuroimaging demonstrated hypothalamic destruction due to sarcoidosis with a CSF hypocretin level of 0 pg/mL. The patient also experienced respiratory depression, which was likely the result of extensive diencephalic injury (90). These observations suggest that narcolepsy may be caused by damage to the hypocretin system.

Discovery of the involvement of the hypocretin-releasing system in narcolepsy may reveal the ultimate etiology of the disorder. So far, convincing mutations in the hypocretin or hypocretin-receptor genes have not been found in humans with narcolepsy (105). Nonetheless, the field is open to envision various injuries to the hypothalamic hypocretin-secreting cell group, from surgical injury to head trauma to immunologic aggression, which would explain acquired narcolepsy and its gradations of severity. Reports indicate that transplantation of Hcrt neurons into the lateral hypothalamus of rats lesioned with a neurotoxin named hypocretin-2-saporin, diminishes narcoleptic-like sleep behavior (11). Such experiments open the way to cell transplantation as an effective method to increase Hcrt production within the narcoleptic brain.

The immunopathologic hypothesis to explain narcolepsy has acquired some notoriety in view of recent discoveries. As noted before, narcolepsy is an HLA-associated disorder at risk of immunopathologic injury. Based on the very strong association with the HLA subtype DQB1*0602, it has been hypothesized that narcolepsy is caused by an autoimmune-mediated process directed at the hypocretin neurons. However, autoantibodies against the hypocretin neurotransmission system have not been found, and confirmation of the hypothesis that autoantibody-mediated dysfunction in the hypocretin system underlies the pathophysiology of narcolepsy remains to be confirmed.

The notion of an immunopathologic mechanism has received some support from the observation that other disorders of autoimmune cause may be associated with narcolepsy. Narcoleptic symptoms have been described in paraneoplastic encephalitis with anti-Ma2 antibodies. In one study, CSF hypocretin-1 levels were measured in six patients with anti-Ma2 encephalitis (102). Four patients with anti-Ma2 encephalitis had excessive daytime sleepiness and hypocretin-1 was not detectable in their cerebrospinal fluid, suggesting an immune-mediated hypocretin dysfunction.

Brain lesions in the upper brainstem and in the region of the third ventricle may precipitate narcolepsy (03). Kanbayashi and colleagues studied patients with neuromyelitis optica, multiple sclerosis, and excessive daytime sleepiness (66). They found bilateral and symmetrical hypothalamic lesions associated with marked or moderate hypocretin deficiency in seven cases of demyelinating disorder and sleepiness. Four patients met the ICSD-2 criteria for diagnosis of narcolepsy. Three patients, including two patients with narcolepsy, were seropositive for anti-AQP4 antibody and had been diagnosed as having neuromyelitis optica-related disorder. Because AQP4 is highly expressed in the hypothalamic periventricular regions, an immune attack on AQP4 might have been partially responsible for bilateral hypothalamic lesions and hypocretin deficiency in narcolepsy/excessive daytime sleepiness with autoimmune demyelinating disease.

A study in monozygotic twins concordant for narcolepsy-cataplexy showed absence of abnormality of the hypocretin system, suggesting that there are genetic forms of narcolepsy with cataplexy that do not depend on the hypocretin pathway (68).

Environmental factors may play a role in the causation of narcolepsy. In a study, 200 patients with narcolepsy/hypocretin deficiency were tested for markers of immune response to beta-hemolytic streptococcus (antistreptolysin O [ASO], anti-DNAse B [ADB]) and Helicobacter pylori [Anti Hp IgG], two bacterial infections known to trigger autoimmunity, and compared to age-matched healthy controls (10). All patients were DQB1*0602 positive, with low CSF hypocretin-1, or had clear-cut cataplexy. When compared to controls, ASO and ADB titers were highest close to narcolepsy onset and decreased with disease duration. The authors concluded that streptococcal infections are probably a significant environmental trigger for narcolepsy.

In a PET study, lower cortical amyloid burden was observed in narcolepsy type 1 than in Alzheimer and other control groups (p <  0.0001 and p = 0.0005, respectively) (50). Similar results were obtained with all subcortical reference regions and for all cortical regions of interest, except cingulum. In patients with narcolepsy type 1, the authors of the study showed that lower brain amyloid burden assessed by 18 F-florbetapir PET suggests delayed appearance of amyloid plaques. The study is interesting because experimental research in rats using dynamic contrast MRI has found that the clearance of neurotoxic proteins from the brain by glymphatic transport is most active during sleep (111). Increased sleep in narcolepsy type 1 could facilitate the clearance of β-amyloid, tau, alpha-synuclein, and other neurotoxins from the brain.

Epidemiology

A study conducted in the United Kingdom, Germany, Spain, and Italy uncovered a prevalence of 0.047% (101). Men and women are affected about equally. Cases of narcolepsy without cataplexy may represent 10% to 50% of the narcolepsy population. Narcolepsy may begin before 10 years or after 50 years of age, but gradual onset in the second decade of life is typical. Narcolepsy with cataplexy in early childhood is rare and difficult to diagnose (57). Narcolepsy in African Americans is characterized by earlier symptom onset, higher Epworth Sleepiness Scale score, higher HLA-DQB1*06:02 positivity, and low CSF hypocretin-1 level in the absence of cataplexy. In African Americans, more subjects without cataplexy have narcolepsy type 1 (67). The prevalence of narcolepsy in 2017 in Slovakia was 10.47 (CI 95% 8.26-14) cases per million inhabitants, with a mean incidence rate between 2000 and 2017 of 0.57 cases per million inhabitants (47).

In Norway, the data collected during three years following vaccination with Pandemrix® (58) showed a significantly increased risk for narcolepsy with cataplexy (P< .0001) and reduced CSF hypocretin levels in vaccinated children ages 4 to 19 years the first year after vaccination, with a minimum incidence of 10 of 100,000 individuals per year. The second year after vaccination, the incidence was 1.1 of 100,000 individuals per year, which was not significantly different from the incidence of 0.5 to 1 of 100,000 per year in unvaccinated children during the same period.

The Truven Health MarketScan Commercial Dissertation Database (THMCDD) was used to estimate prevalence and incidence of narcolepsy, with and without cataplexy, by age groups, gender, and region among patients under 66 years of age with continuous enrollment for years 2008 to 2010. THMCDD contains health claims information for more than 18 million people. Prevalence was expressed as cases per 100,000 persons. Average annual incidence (using varying criteria for latency between the diagnostic tests, polysomnograph coupled with multiple sleep latency test [MSLT], and the diagnosis) was expressed as new cases per 100,000 persons per year.

In a large study of 8,444,517 continuously enrolled patients in the United States of America, 6703 were diagnosed with narcolepsy (125). The overall prevalence was 79.4 per 100,000, without cataplexy was 65.4 per 100,000 and with cataplexy 14.0 per 100,000. The overall average annual incidence was 7.67, 7.13, and 4.87 per 100,000 persons per year, respectively. Incidence for narcolepsy type 2 was several times higher than narcolepsy type 1. Prevalence and incidence were approximately 50% greater for females compared to males across most age groups and highest among the 21 to 30 years age group. The incidence was highest among enrollees in their early 20s and late teens. The North Central United States had the highest prevalence and incidence, whereas the West was the lowest. The authors concluded that the prevalence and incidence of narcolepsy was higher than in most previous studies.

Prevention

No methods of prevention are known. Narcolepsy in a first-degree relative increases the risk of developing narcolepsy by 20- to 40-fold. Lesions of the diencephalon and upper brainstem also may increase the risk of developing narcolepsy (03). Vaccination with Pandemrix® (58) should be avoided in children ages 4 to 19 years.

Differential diagnosis

Confusing conditions

There are many causes of excessive sleepiness in addition to narcolepsy. Obstructive sleep apnea syndrome is common and produces prominent daytime sleepiness that improves following effective intervention. Chronically insufficient sleep is another common cause, particularly in adolescents. Disrupted sleep due to poor sleep hygiene, shift work or night work, medical or psychiatric illness, and the effects of medications and alcohol can result in daytime sleepiness. The periodic limb movement disorder can cause frequent nocturnal awakenings and excessive sleepiness. Less frequently, sleepiness may occur in association with hypothalamic lesions, Prader-Willi syndrome, and myotonic dystrophy. The most useful differentiating feature is cataplexy, which does not occur in any of these other disorders.

Some patients with complaints of excessive sleepiness and sleep-onset REM periods during a multiple sleep latency test do not have cataplexy. Diagnostic considerations include circadian rhythm disorders, chronic REM sleep disruption or deprivation associated with sleep apnea, periodic leg movements, affective disorders, or medications. Idiopathic hypersomnia is associated with excessive sleepiness but without sleep-onset REM periods. Clinically, patients with idiopathic hypersomnia and those with narcolepsy may have substantial overlap of symptomatology (19). A subgroup of narcoleptics (18% of narcoleptics) with a long sleep time was reported (131). Their symptoms combine the disabilities of both narcolepsy (severe sleepiness) and idiopathic hypersomnia (long sleep time and unrefreshing naps). These patients may constitute a group with multiple arousal system dysfunctions.

Posttraumatic hypersomnia occurs rarely as a sequela of CNS trauma and is usually seen in conjunction with other symptoms such as headache, fatigue, and memory impairment. Systemic infection, especially infectious mononucleosis, may be accompanied by sleepiness and fatigue. Recurrent hypersomnia or Kleine-Levin syndrome consists of periodic episodes of hypersomnia, hyperphagia, and abnormal behavior that recur on average twice a year.

The presence of sleep apnea or periodic limb movement disorder does not preclude a diagnosis if cataplexy is present. Cataplexy must be differentiated from transient ischemic attacks, hypotension, akinetic seizures, periodic paralysis, vestibular dysfunction, and psychiatric disorders. The association with laughter and other emotions and the response to tricyclic antidepressant medications may help in the diagnosis of cataplexy. Although narcolepsy is the only sleep disorder associated with cataplexy, the symptom may occur with type C Niemann Pick disease. Sleep paralysis and hypnagogic hallucinations are common in the general population and can occur with other sleep disorders; they are not specific for narcolepsy (02).

Narcolepsy with or without cataplexy may exhibit a pathogenetic association with a variety of medical conditions. If hypocretin is very low or absent in CSF, the condition is diagnosed as narcolepsy type 1 due to medical condition (ICDS-3). Patients with symptomatic narcolepsy report excessive daytime somnolence, cataplexy in many cases, and MSLT alterations that fit the criteria of narcolepsy. Symptomatic narcolepsy has been diagnosed in conjunction with multiple sclerosis (142), myotonic dystrophy (70), primary and secondary hypothalamic tumors (86), inflammation of the hypothalamus such as viral encephalitis, tuberculosis, Whipple disease, histiocytosis X, lymphocytic hypophysitis, radiation therapy, and paraneoplastic limbic encephalitis (83), strokes involving the diencephalons (124), acute disseminated encephalomyelitis (53), Prader-Willi syndrome, head trauma, progressive supranuclear palsy, Niemann-Pick disease type C (65), Norrie disease (133), and autosomal dominant deafness with ataxia and narcolepsy (92). Two cases of severe narcolepsy-cataplexy have been described in childhood in close temporal association with obesity and precocious puberty (108).

In children, narcolepsy with or without cataplexy may masquerade as behavioral alterations or attention-deficit/hyperactivity disorder (80). In one study, children with rapid weight gain were younger and sleepier, and the prevalence of obesity was higher at diagnosis (143). These patients had a higher risk of developing long-term obesity. The authors concluded that rapid weight gain could represent a marker of a more severe phenotype of childhood narcolepsy.

Associated or underlying disorders

In a study conducted in Slovakia, the comorbidities found were arterial hypertension (17%), ischemic heart disease (8%), dyslipidemia (18%), diabetes mellitus type 2 (10%), cardiac arrhythmia/atrial fibrillation (5%), autoimmune disorders (20%), allergy (11%), malignancy (3%), headache (15%), and mental disorders (20%) (47).

Diagnostic workup

Excessive daytime sleepiness may be identified using screening tools such as the Epworth Sleepiness Scale (ESS) and the modified scale for children known as the Epworth Sleepiness Scale-CHAD (64; 62; 134). The Swiss Narcolepsy Scale is a screening tool used to identify a clinical profile suggestive of narcolepsy and cataplexy (18). In patients without cataplexy, polysomnography helps to document the absence of other disorders that could explain the patient's symptoms and provides objective confirmation of narcolepsy. Nighttime polysomnography often shows a short sleep latency of less than 10 minutes, sleep-onset REM (appearing within 15 minutes after sleep onset), and disrupted sleep with increased stage 1 sleep and frequent arousals. SOREMP during nocturnal sleep is a highly specific finding in the absence of other sleep disorders.

Presence of nocturnal REM sleep without atonia (nRWA) in the polysomnogram may aid the pediatric diagnosis of narcolepsy. In a retrospective study of children 6 to 18 years of age who completed a nocturnal polysomnogram followed by MSLT, 11 patients had narcolepsy type 1 (NT1), six had narcolepsy type 2 (NT2), 12 had idiopathic hypersomnia (IH), and 11 had subjective hypersomnia (sHS) (22). Group nRWA indices (epochs of RWA/total stage R sleep epochs) were compared and analyzed from the nocturnal PSGs. The authors of the study found that the median nRWA index of patients with narcolepsy type 1 was 15 to 30 times higher compared to subjective hypersomnia and idiopathic hypersomnia (Ps < .005) but similar to that of the narcolepsy type 2 group (P = .46). The authors concluded that the nRWA index is a very good diagnostic biomarker of pediatric narcolepsy.

The multiple sleep latency test (MSLT) provides an objective measure of excessive sleepiness and demonstrates the presence of sleep-onset REM periods (SOREMP). For correct interpretation of the multiple sleep latency test, nocturnal polysomnography should be performed on the night immediately preceding the multiple sleep latency test, and the patient should be free of any medication effects that may influence sleep. Medications affecting sleep should be discontinued 15 days before testing. Sleep latencies of less than eight minutes and two or more sleep-onset REM periods are found in many narcoleptic patients; however, these findings are not specific and can occur in patients with other sleep disorders. Mean sleep latencies of 3.1+2.9 minutes have been shown in a metaanalysis. The ICSD-3 criteria require a mean sleep latency of eight minutes or less and two SOREMPS or more in at least four nap studies of an MSLT performed after a night of sufficient sleep (at least 6 hours). A nocturnal SOREMP (less than 15 minutes REM sleep onset) may substitute for one episode of daytime SOREMP in the MSLT. Furthermore, because sleep-onset REM periods do not always occur in patients with narcolepsy, a repeat MSLT is often valuable if clinical suspicion is high and an initial multiple sleep latency test does not show frequent sleep-onset REM periods (02). Coelho and colleagues showed that a repeat MSLT confirmed the diagnosis of narcolepsy in 20% of patients whose results had been nonconfirmatory on a first MSLT (32). The study provides support for a repeat MSLT in cases where clinical suspicion for narcolepsy is high despite an ambiguous first test.

To evaluate repeatability of MSLT results in narcolepsy type 1 and narcolepsy type 2, the authors evaluated retrospectively narcolepsy type 1 (n = 60) and narcolepsy type 2 (n = 54) cases and controls (n = 15) (117). All subjects had undergone two MSLTs, meeting criteria for narcolepsy. The authors found that both MSLTs in unmedicated patients were positive for narcolepsy in 78%, 18%, and 7% of narcolepsy type 1, narcolepsy type 2, and controls, respectively. Narcolepsy type 2 cases were changed to idiopathic hypersomnia or to a negative MSLT 26% and 57% of the time, respectively. Medication use (P = .009) significantly decreased the likelihood of a repeat positive MSLT. Thus, a positive MSLT for narcolepsy is more reproducible in narcolepsy type 1 than narcolepsy type 2.

The following conditions are highly desirable when pursuing a laboratory diagnostic test: (1) the patient must be free of drugs that influence sleep for at least 14 days, confirmed by a urine drug screen; (2) the sleep-wake schedule must have been standardized and, if necessary, extended to a minimum of seven hours in bed each night (longer for children) for at least seven days before polysomnography, documented by sleep log or actigraphy; and (3) nocturnal polysomnography should be performed on the night immediately preceding the MSLT to rule out other sleep disorders.

The maintenance of wakefulness test (MWT) is more useful as a tool to assess the response to treatment and to assess the risk due to sleepiness after treatment (78). The patient is instructed to stay awake and, therefore, the test is a measure of the ability to stay awake. The methods are variable and there are few normative data and less clinical experience. The MWT 40-minute protocol has been recommended using the first epoch of sleep as the definition of sleep onset. The trial is finished after 40 minutes if no sleep occurs, or after unquestionable sleep, defined as three continuous epochs of stage 1 sleep or 1 epoch of any other stage. The test is unable to identify SOREMPS or confirm their absence. It is considered unambiguously abnormal if the patient falls asleep with an average latency of eight minutes or less. Its reliability for driving fitness evaluation is insufficient, and motivation plays a major role (20).

Routine EEG offers little diagnostic assistance but may show persistent drowsiness or REM sleep. Although cataplexy with daytime sleepiness is nearly pathognomonic of narcolepsy, patients with these symptoms often undergo sleep studies to exclude drug-seeking individuals and to provide objective confirmation of a lifelong disorder that usually requires treatment with schedule II drugs.

The decrease of Hcrt-1 levels in CSF in patients with narcolepsy provides a new test to diagnose the condition. Mignot has determined that 110 pg/mL is the cut-off value to diagnose narcolepsy (96). Patients with idiopathic hypersomnia, sleep apnea, restless legs syndrome, or insomnia have normal Hcrt levels. Patients with narcolepsy-cataplexy have predicted values of less than 110 pg/mL with a specificity of 99% and a sensitivity of 87%, which is higher than the MSLT. Up to 10% of patients with narcolepsy and cataplexy will have a normal hypocretin-1 level in CSF so that a negative finding cannot exclude the diagnosis. These cases are generally DQB1*0602 negative and have familial incidence. When cataplexy is absent or atypical the predictive power is limited with high specificity (99%) but low sensitivity (16%). Ten percent to 20% of patients without cataplexy will have decreased CSF levels of hypocretin (less than 110 pg/mL), almost all with positive HLA DQB1 0602. Most of these patients will have normal Hcrt values, creating a dilemma for the clinician because this is precisely the group of patients in need of objective diagnostic studies (94). In some cases of narcolepsy without cataplexy, the Hcrt values have been measured in CSF between 110 and 200 pg/mL, 200 pg/mL being the normal level, raising the suggestion of partial Hcrt deficiency. On the other hand, up to 15% of brain-damaged individuals have decreased Hcrt values in the intermediate range, suggesting some form of abnormality of the Hcrt system. Somnolence in these individuals could then be linked to a partial Hcrt deficiency, a notion that needs confirmation. The clinical usefulness of measurement of Hcrt in CSF is highest in those situations where the clinical features and MSLT are difficult to interpret, such as when a psychiatric condition or the administration of drugs confounds the picture. It is also useful for the early diagnosis in children who might not have developed cataplexy yet and in general when the diagnosis of narcolepsy is uncertain. HLA typing, MSLT, and sleep study results predicted low concentration of CSF hypocretin-1 more than subjective manifestations (sleepiness and sleep paralysis) in patients with narcolepsy without cataplexy (05).

Overall, measuring levels of hypocretin-1 in CSF is a highly specific and sensitive test for the diagnosis of narcolepsy type 1. Commercially available radioimmunoassays are available for measuring hypocretin-1 in CSF. The Stanford reference sample value indicates that less than 110 pg/mL is highly specific.

Assessing for primary hypersomnia, fragmented and poor-quality sleep must be ruled out. Using a sleep diary or actigraphy, clinicians may assess sleep/wake habits, but accuracy comes under question. In a review to evaluate the accuracy of a sleep diary against actigraphy, data from 35 patients with suspected primary hypersomnia were collected (73). Least accurate was “sleep time,” with 14.7%, 23.5%, and 58.8% of patients within 20, 30, and 60 minutes of the actigraphy values, respectively. “Time to fall asleep” was most accurate, with 76.5%, 82.4%, and 100% similarity, respectively. The authors of the study recommend use of a diary alongside actigraphy because the diary is low cost and adds subjective information that cannot be gathered from actigraphy.

Management

Lifestyle adjustments combined with symptomatic pharmacologic treatment allow most patients to live a relatively normal life (71). Treatment of narcolepsy type-1 and 2 remains symptomatic targeting sleepiness, cataplexy, and disrupted nocturnal sleep. New psychostimulants have been recently developed, such as hypocretin receptor-2 agonists that will revolutionize the management of these less than rare disorders.

The American Academy of Sleep Medicine has published a guideline with clinical practice recommendations for the treatment of central disorders of hypersomnolence in adults and children (88).

Excessive daytime sleepiness. Effective treatment of narcolepsy often includes pharmacologic, educational, and lifestyle intervention. The primary treatment for excessive sleepiness consists of CNS stimulants: modafinil, amphetamine, methamphetamine, dextroamphetamine, methylphenidate, and selegiline (97). Pemoline is no longer recommended and no longer available because of rare instances of liver toxicity. Solriamfetol and pitolisant are novel FDA-approved wake-promoting drugs. Sodium oxybate is not a wake-promoting drug but is effective in dispelling excessive daytime sleepiness (vide infra). Dosages and schedules must be individualized to provide acceptable relief from sleepiness while avoiding the side effects that may accompany high doses of stimulants. Many patients with narcolepsy take stimulants daily for decades without significant adverse effects, but all patients on stimulants should be monitored for cardiac arrhythmias, tremor, anxiety, or high blood pressure. Occasionally, "drug holidays" or changes to other stimulants or wake-promoting drugs may be required because of the development of tolerance to the drugs. Overuse and abuse of stimulants are rare in narcoleptics; usage below the amounts prescribed is more common. Tolerance has not been described with solriamfetol. In clinical trials, modafinil significantly improved wakefulness on objective and subjective measures of excessive sleepiness (28; 06; 07). There have been no withdrawal symptoms following abrupt modafinil discontinuation. The recommended dose is 200 mg once daily in the morning. Dose adjustments are required for patients with hepatic disease and should be considered for the elderly and those with renal disease. Some patients have improved further on 400 mg and even 600 mg daily. Split dosing (morning 400 mg and noon 200 mg) may be effective for late afternoon or evening sleepiness. Post-marketing surveillance has shown no generalized interest in modafinil as a drug of abuse. Peak plasma concentrations are attained in 2 to 4 hours and may be delayed by food. Elimination half-life is 15 hours. Modafinil is metabolized by the liver and less than 10% of unchanged drug appears in urinary excretion. Modafinil levels may be affected by potent inducers (eg, phenytoin) or inhibitors of cytochrome P450 (CYP) enzymes such as steroidal contraceptives or cyclosporine. The levels of the following agents may increase: warfarin, phenytoin, diazepam, and propranolol.

Armodafinil (the longer half-life enantiomer of modafinil) has been shown to be effective for treatment of excessive sleepiness in patients with narcolepsy. Subjects receiving armodafinil experienced significant improvement in sleepiness as measured by the MWT mean sleep latency and in the Clinical Global Impression of Change (56). Once-daily armodafinil was effective in improving wakefulness in adult patients with excessive sleepiness associated with obstructive sleep apnea/hypopnea syndrome, narcolepsy, or shift work sleep disorder in four large (n > 195), double-blind, multinational trials of 12 weeks' duration (51). Compared with placebo, mean sleep latency was significantly improved with armodafinil 150 or 250 mg once daily in patients with narcolepsy, as assessed by the multiple sleep latency test (MSLT) or the maintenance of wakefulness test (MWT).

Solriamfetol is a selective dopamine and norepinephrine reuptake inhibitor that has been shown in clinical trials to be an effective wake-promoting drug in doses of up to 150 mg daily (14; 82). The Maintenance of Wakefulness test has shown efficacy through nine hours post-dose. Treatment-emergent adverse events with solriamfetol were headache, nausea, nasopharyngitis, insomnia, dry mouth, anxiety, decreased appetite, and upper respiratory tract infection (82). Tolerance has not been described. Solriamfetol can be used concomitantly with sodium oxybate.

Pitolisant is a histamine-3 receptor antagonist/inverse agonist that increases histamine levels in the brain and has wake-promoting efficacy (128). In addition, pitolisant has anticataplectic action (37; 93). The recommended dosage range is 17.8 to 35.6 mg taken in the morning. It is contraindicated in patients with hepatic impairment, whereas centrally acting antihistaminics should be avoided. Onset of clinical response for excessive daytime sleepiness or cataplexy was generally observed within the first 2 to 3 weeks of pitolisant treatment (135). Most common adverse reactions are headache, insomnia, and nausea (114). Rare patients have developed hallucinations in clinical trials. Overall, the drop-out rate in clinical trials has been low, below 5%. The post-marketing experience is still being evaluated. Pitolisant can be used concomitantly with sodium oxybate.

Sodium oxybate is beneficial for the treatment of excessive daytime sleepiness and cataplexy (140; 23) and also for control of disrupted sleep due to narcolepsy. The new preparation LXB (Xywav) has a low salt content (92% less sodium) and is the preferred choice for patients with hypertension, heart disease, or who complain of leg swelling. LXB is the calcium, magnesium, potassium, and sodium oxybate oral solution that is indicated for the treatment of cataplexy or excessive daytime sleepiness in patients seven years of age and older with narcolepsy (26). Xywav, like Xyrem, is available only through a restricted program under a Risk Evaluation and Mitigation Strategy (REMS). In a long-term analysis the safety and tolerability profile of Xywav was consistent with the safety profile of Xyrem (59; 25). The most common adverse events reported were headache, dizziness, and nasopharyngitis.

Cataplexy. Drugs useful for treatment of cataplexy are imipramine, clomipramine, or protriptyline. Fluoxetine, other specific serotonin reuptake inhibitors, venlafaxine (an antidepressant that increases serotonin and norepinephrine uptake), and reboxetine (a selective norepinephrine reuptake inhibitor not available for use in the United States) are also effective for cataplexy (97). Abrupt withdrawal of these agents may result in a rebound of cataplexy. In 2002, the FDA approved sodium oxybate for treating patients with narcolepsy who experience episodes of cataplexy. Safety concerns about the use of the “date rape drug” have imposed restrictions on the distribution of the drug. Sodium oxybate has been designated as a schedule III controlled substance for medical use. Sodium oxybate has fewer and less-severe side effects compared to tricyclic antidepressants and is presumed to work through a different mechanism of action. GHB (gamma hydroxybutyrate), the endogenous form of sodium oxybate, is a metabolite of GABA. It acts as a neuromodulator with complex effects on GABA, dopamine, serotonin, and endogenous opioids. Also, evidence supports a neurotransmitter role for GHB. Like other neurotransmitters, GHB is synthesized in neurons, stored in vesicles, and released via depolarization into the synaptic cleft, where it undergoes reuptake. In addition, specific receptors have been shown to exist for GHB (98). Sodium oxybate has been evaluated for safety and efficacy in a total of 755 individuals. Three placebo-controlled, randomized clinical trials evaluated the efficacy of sodium oxybate in the treatment of various symptoms of narcolepsy, whereas four open-label studies also assessed the effect of sodium oxybate on various symptoms of narcolepsy (08). Continued use of sodium oxybate, titrated to optimal individual dose, produces substantial, additional reductions in the frequency of cataplexy beyond those initially seen over four weeks (09). In yet another study, the results demonstrated a clear dose-related reduction in Epworth Sleepiness Scale scores across the three doses. In all sodium oxybate-treated groups, some patients improved beyond the defined narcolepsy range (08). The nocturnal administration of sodium oxybate in patients with narcolepsy was associated with statistically significant and clinically relevant improvements in functional status, an important component of quality of life (136). In a metaanalysis of nine randomized controlled trials and 771 patients in GHB-treated groups reviewing the effectiveness of GHB on clinical features of narcolepsy, the authors confirmed the clinical efficacy of GHB in various study outcomes (27). The metaanalysis showed that GHB reduced cataplexy attacks both on a daily and a weekly basis, subjective nocturnal awakenings, daytime sleep attacks on a weekly basis, subjective daytime sleepiness, and sleep stage shifts. GHB increased sleep stages 3 and 4 and improved the clinical global impression score. No significant changes were observed in night sleep latency, total sleep time, REM sleep, and sleep stages 1 and 2. Consequently, the authors concluded that the meta-analysis demonstrated the effectiveness of GHB in treating major, clinically relevant narcolepsy symptoms and sleep architecture abnormalities. Overall, sodium oxybate has shown improvements in the sleep continuity and nocturnal sleep quality that are characteristic of disrupted nighttime sleep (115).

In a study aimed to evaluate primarily safety, 202 sodium oxybate-naive patients were initiated on open-label sodium oxybate at 4.5 g per night and titrated in 1.5 g increments up to 9 g per night or down to 3 g per night, based on individual clinical response. Treatment was 12 weeks. Response was defined as "much improved" or "somewhat improved." At weeks 6 and 12, 171 (85%) completed treatment. Adverse events were reported in 114 patients (56%) and serious adverse events in five (2%). The most common adverse events were nausea (10%), headache (7%), and dizziness (5%). Overall, 60% of patients rated their symptoms at 12 weeks as "much improved," and this improvement was dose dependent. The authors concluded that the safety profile of sodium oxybate was consistent with other trials (84). In another study, sodium oxybate alone and in combination with modafinil improved subjective ratings of excessive sleepiness and an objective measure of the ability to stay awake to similar extents in patients with narcolepsy type 1 and narcolepsy type 2 (24).

In 2018, the FDA approved the use of sodium oxybate for children and adolescents aged seven years and older (109). The product demonstrated efficacy for both cataplexy and excessive daytime sleepiness in children. It is administered in two nocturnal doses based on patient weight (63). Sodium oxybate formulations require two nightly doses, one at bedtime, and another 2.5 to 4 hours later. Once-nightly administration to treat adults with narcolepsy has been approved by the FDA in 2023. In a survey, the one-time dose administration was significantly preferred over the twice nightly (45). In a clinical trial, a single nighttime dose of sodium oxybate significantly improved disrupted nighttime sleep in patients with narcolepsy (116).

Other methods for management of narcoleptic symptoms. Pilot studies have tested the clinical efficacy of hypocretin administered intranasally to patients with narcolepsy (15). The authors found beneficial changes sufficient to warrant future studies.

Planned daytime naps may help control sleepiness and may reduce the total required daily dose of stimulants. Education of the patient, family, and coworkers plays an important role in adaptation to narcolepsy. Families should understand that narcolepsy is a chronic neurologic disease, and that pathologic sleepiness should not be confused with poor motivation or low intelligence. Patients with narcolepsy should be cautioned about the hazards of driving an automobile or operating dangerous equipment. Recommendations regarding driving should be individualized based on pertinent laws, the patient's awareness of sleepiness, the severity of sleepiness, response to medication and medication compliance, and the patient's driving habits and judgment concerning driving.

Tricyclic antidepressants, selective serotonin reuptake inhibitors (SSRIs), and venlafaxine may be effective treatment of sleep paralysis and hypnagogic hallucinations (97). Extended-release venlafaxine generally is well tolerated, and it is effective for most of the day (123).

Combinations of long- and short-acting forms of stimulants may be indicated and effective for some patients.

Some stimulants like methylphenidate have a short (3 to 4 hours) effective period, whereas others like armodafinil, sustained-release amphetamine, and sustained-release methylphenidate have longer duration of activity and longer onset of action. The combination may achieve alertness quickly and for longer periods of time without causing insomnia.

Future treatments in narcolepsy may aim toward preventing the autoimmune response that results in the loss of hypocretin cells (72). Immunotherapy, based on current knowledge of the pathophysiology of narcolepsy with cataplexy is a potential treatment option.

Dauvilliers and colleagues treated four hypocretin-deficient narcolepsy patients with intravenous immunoglobulin and assessed the efficacy with repeated polysomnography and questionnaires (36). Three patients received the treatment within a few months after acute onset of narcolepsy. Improvement in the frequency and severity of cataplectic attacks was observed, with the beneficial effect lasting up to seven months without anticataplectic drugs. The authors concluded that the early diagnosis and treatment of narcolepsy may modify its long-term outlook. Dauvilliers and a different team of researchers reported improvement of hypersomnolence and cataplexy with administration of IV immunoglobulin (IVIG) (1 g/kg/day over 2 days, repeated three times at 4-week intervals) 15 days after onset of symptoms in a 28-year-old woman (35). A placebo effect was ruled out by confirming normalization of CSF hypocretin level at three months post-treatment. The authors hypothesized that an autoimmune-mediated inflammatory process affecting hypocretin-releasing neurons in the hypothalamus was reversed by the administration of IVIG.

New drugs are currently being tested in animal models and in humans (38). These include a wide variety of hypocretin agonists, melanin-concentrating hormone receptor antagonists, and antigen-specific immunopharmacology agents. Even though current treatment is strictly symptomatic, it is expected that more pathophysiology-based treatments will be available in the near future.

The effects of the orexin 2 receptor (OX2R)-selective agonist danavorexton were evaluated initially in mice and thereafter in single- and multiple-rising-dose studies in healthy adults, and in individuals with narcolepsy 1 and 2 (46). In humans with narcolepsy 1 and 2, danavorexton administered intravenously was well tolerated and was associated with markedly improved sleep latency. In subjects with narcolepsy 1, sleep latency increased in the Maintenance of Wakefulness Test up to 40 minutes. Consequently, OX2R-selective agonists are a promising treatment for both narcolepsy 1 and 2.

Special considerations

Pregnancy

Excessive sleepiness in narcolepsy may worsen or improve during pregnancy. Risk to the narcoleptic mother and her fetus is related to (1) risk of trauma secondary to excessive sleepiness or cataplexy and (2) possible adverse effects related to use of medication during pregnancy. Because no adequate human trials of the CNS stimulants during pregnancy have been performed, decisions regarding initiation or continuation of drug therapy must be individualized, with particular emphasis on the potential risks associated with the untreated state versus potential harm to the fetus or mother caused by medication. Monoamine oxidase inhibitors should be avoided in pregnancy (77).

One study reported more obstetric complications in patients with narcolepsy-cataplexy during pregnancy (89), although these were not severe. The group also had a higher body mass index, had a higher incidence of impaired glucose metabolism during pregnancy, and caesarean section was conducted more frequently, despite cataplexy being a rare event during delivery. Symptoms of narcolepsy may render care of the infant more difficult. The results of a survey and literature suggested that the perceived risks of narcolepsy medication during pregnancy to the mother and the fetus are overestimated, as the risk for teratogenic effects from narcolepsy medications in therapeutic doses is essentially nonexistent. In rare cases patients had cataplexy that interfered with delivery, but if caesarian is required there appears to be no increased anesthetic or surgical risks.

Methylphenidate is excreted in breast milk only in small amounts. There have been no reports of breastfed infants demonstrating any adverse effects. Methylphenidate appears to be compatible with breastfeeding; however, the long-term neurodevelopmental effects have not been adequately studied (85).

To determine GHB levels in breast milk, two women with narcolepsy and cataplexy collected breast milk for analysis of GHB concentration after resuming sodium oxybate postpartum (17). Levels were two to four times higher 4 hours after the first sodium oxybate dose and three to five times higher 4 hours after the second dose. GHB levels returned to endogenous levels 6 to 10 hours following the second dose. Sodium oxybate is transmitted to breast milk, and the authors recommend that to avoid excess GHB exposure, breastfeeding mothers who take sodium oxybate should consider expressing and discarding their morning milk.

Anesthesia

No specific data are available regarding the use of general anesthesia in patients with narcolepsy. To minimize potential drug-drug interactions, the clinician may elect to taper or discontinue medications in the perioperative period.