Elsevier

Sleep Medicine

Volume 4, Issue 1, January 2003, Pages 7-12
Sleep Medicine

Original article
The narcoleptic borderland: a multimodal diagnostic approach including cerebrospinal fluid levels of hypocretin-1 (orexin A)

https://doi.org/10.1016/s1389-9457(02)00191-0Get rights and content

Abstract

Objectives: Biological markers of narcolepsy with cataplexy (classical narcolepsy) include sleep-onset REM periods (SOREM) on multiple sleep latency tests (MSLT), HLA-DQB1*0602 positivity, low levels of cerebrospinal fluid (CSF) hypocretin-1 (orexin A), increased body mass index (BMI), and high levels of CSF leptin. The clinical borderland of narcolepsy and the diagnostic value of different markers of narcolepsy remain controversial and were assessed in a consecutive series of 27 patients with hypersomnia of (mainly) neurological origin.

Methods: Diagnoses included classical narcolepsy (n=3), symptomatic narcolepsy (n=1), narcolepsy without cataplexy (n=4), idiopathic hypersomnia (n=5), hypersomnia associated with psychiatric disorders (n=5), and hypersomnia secondary to neurological disorders or of undetermined origin (n=9). Clinical assessment included BMI, Epworth Sleepiness Scale (ESS), Ullanlinna Narcolepsy Scale (UNS), and history of REM-symptoms (sleep paralysis, hallucinations). HLA-typing, electrophysiological studies (conventional polysomnography, MSLT, 1-week actigraphy), and measurements of CSF levels of hypocretin and leptin were also performed.

Results: Hypocretin-1 was undetectable in three patients with classic narcolepsy and detectable in the remaining 24 patients. Other narcoleptic markers also frequently found in patients without narcolepsy included ESS>14 (78% of 27 patients), UNS>14 (75%), REM symptoms (30%), sleep latencies on MSLT<5 min (41%), ≥2 SOREM (30%), DQB1*0602 positivity (52%), BMI>25 (52%), and increased CSF leptin (48%). Hypersomnia was documented by an increased time ‘asleep’ in 41% of patients. Overlapping clinical and electrophysiological findings were seen mostly in patients with narcolepsy without cataplexy, idiopathic hypersomnia, and psychiatric hypersomnia.

Conclusions: (1) Hypocretin dysfunction is not the ‘final common pathway’ in the pathophysiology of most hypersomnolent syndromes that fall on the borderline for a diagnosis of narcolepsy. (2) The observed overlap among these hypersomnolent syndromes implies that current diagnostic categories are not entirely unambiguous. (3) A common hypothalamic, hypocretin-independent dysfunction may be present in some of these syndromes.

Introduction

Human narcolepsy is a neurological disorder characterized by hypersomnia, cataplexy, sleep paralysis and hypnagogic hallucinations [1]. This full tetrad of symptoms is present in only a minority of narcoleptic patients, and cataplexy is the only pathognomonic symptom of narcolepsy. Biological markers of narcolepsy include sleep-onset REM periods (SOREM) on multiple sleep latency tests (MSLT) and DR2/DQB1*0602 positivity on HLA-typing, which may however be absent in some patients with the disorder [1].

In the absence of a gold standard for the diagnosis of narcolepsy, the definition of its clinical borderland – and in particular the differentiation between such entities as narcolepsy without cataplexy, idiopathic hypersomnia and hypersomnia associated with psychiatric disturbances – remains a matter of controversy [2], [3].

Hypocretins are peptides, synthesized by neurons in the posterior and lateral hypothalamus, which have widespread projections within the brain and have been implicated in sleep-wake functions, feeding and metabolic control [4]. Recent studies suggest a dysfunction of the newly described hypocretin (orexin) neurotransmitter system in narcolepsy. First, positional cloning has identified hypocretin-receptor-2 gene mutations as the cause of narcolepsy in dogs and orexin knockout mice have a narcolepsy-like phenotype [5], [6]. Second, a mutation was found in one of 74 tested human patients and a polymorphism in the prehypocretin gene in six of 178 patients [7], [8]. Third, hypocretin-1 was found to be low or absent in the CSF of >90% of patients with classical narcolepsy (narcolepsy with cataplexy) [9]. In a recent study Ripley et al. confirmed these results, reporting levels of CSF hypocretin-1 <100 pg/ml in 37 of 42 patients with narcolepsy [11]. Fourth, an 85–95% reduction of hypocretin neuronal expression was demonstrated post mortem in the brains of patients with narcolepsy with cataplexy [7], [12]. Finally, an increased body mass index (BMI) and CSF leptin levels were described in narcoleptics, suggesting an altered regulation of food intake and/or metabolism [10], [13].

The aim of the study was twofold. First, we wanted to test the hypothesis that CSF hypocretin-1 is reduced also in narcoleptic patients without cataplexy (so-called monosymptomatic narcolepsy) and in patients with other forms of (mainly) neurological hypersomnia, which belong to the borderland of narcolepsy. Second, we wanted to assess the diagnostic value of a multimodal approach including clinical, electrophysiological, genetic, BMI and CSF leptin data in this same patient group.

Preliminary results of this work have been presented before [14]. A few patients reported here were included in a multicenter study [15].

Section snippets

Subjects and methods

We studied 27 consecutive Caucasian patients (12 women and 15 men; mean age 38 years, range 16–53) with hypersomnia of variable etiology. The main diagnoses (see also Table 1) included narcolepsy with cataplexy (classical narcolepsy, n=3), symptomatic narcolepsy following Bickerstaff's encephalitis (n=1), narcolepsy without cataplexy (n=4), idiopathic hypersomnia (n=5), hypersomnia associated with psychiatric disorders (n=6), HIV-encephalopathy (n=1), brainstem stroke (n=2), periodic

Results

The main results of the study are summarized in Table 1.

Hypocretin-1 (orexin A) levels in the cerebrospinal fluid

Hypocretin-1 levels in the cerebrospinal fluid, and therefore hypocretin transmission, were not deficient in patients with hypersomnia without cataplexy, including monosymptomatic narcolepsy and idiopathic hypersomnia. In other words, the loss of hypocretin neurons in the lateral hypothalamus may be relevant only for narcoleptics with cataplexy, but not necessarily for narcoleptics without cataplexy and patients with other hypersomnolent syndromes of (presumed) neurological origin. Recent

Note

After the submission of this work a paper by Kanbayashi et al. was published reporting normal (n=14) or intermediate (152–198 pg/ml, n=3) levels of CSF hypocretin-1 (orexin A) in a series of Japanese patients with narcolepsy without cataplexy (n=5) and idiopathic hypersomnia (n=12) (J Sleep Res 2002;11:91–3).

Acknowledgements

Dr. M. Okun helped in coordinating the collection and shipping of the CSF samples. Dr. J. Gottselig reviewed the English text of the paper.

References (27)

  • M Gencik et al.

    A prepro-orexin gene polymorphism is associated with narcolepsy

    Neurology

    (2001)
  • N Nishino et al.

    Hypocretin (orexin) deficiency in human narcolepsy

    Lancet

    (2000)
  • S Nishino et al.

    Low cerebrospinal fluid hypocretin (orexin) and altered energy homeostasis in human narcolepsy

    Ann Neurol

    (2001)
  • Cited by (0)

    View full text