CSF hypocretin-1 levels in narcolepsy, Kleine-Levin syndrome, and other hypersomnias and neurological conditions.

OBJECTIVE: To determine the role of CSF hypocretin-1 in narcolepsy with and without cataplexy, Kleine-Levin syndrome (KLS), idiopathic and other hypersomnias, and several neurological conditions. PATIENTS: 26 narcoleptic patients with cataplexy, 9 narcoleptic patients without cataplexy, 2 patients with abnormal REM-sleep-associated hypersomnia, 7 patients with idiopathic hypersomnia, 2 patients with post-traumatic hypersomnia, 4 patients with KLS, and 88 patients with other neurological disorders. RESULTS: 23 patients with narcolepsy-cataplexy had low CSF hypocretin-1 levels, while one patient had a normal hypocretin level (HLA-DQB1*0602 negative) and the other two had intermediate levels (familial forms). One narcoleptic patient without cataplexy had a low hypocretin level. One patient affected with post-traumatic hypersomnia had intermediate hypocretin levels. The KLS patients had normal hypocretin levels while asymptomatic, but one KLS patient (also affected with Prader-Willi syndrome) showed a twofold decrease in hypocretin levels during a symptomatic episode. Among the patients without hypersomnia, two patients with normal pressure hydrocephalus and one with unclear central vertigo had intermediate levels. CONCLUSION: Low CSF hypocretin-1 is highly specific (99.1%) and sensitive (88.5%) for narcolepsy with cataplexy. Hypocretin ligand deficiency appears not to be the major cause for other hypersomnias, with a possible continuum in the pathophysiology of narcolepsy without cataplexy and idiopathic hypersomnia. However, partial hypocretin lesions without low CSF hypocretin-1 consequences cannot be definitely excluded in those disorders. The existence of normal hypocretin levels in narcoleptic patients and intermediate levels in other rare aetiologies needs further investigation, especially for KLS, to establish the functional significance of hypocretin neurotransmission alterations.

Sleep architecture, slow wave activity and sleep spindles in mild sleep disordered breathing.

OBJECTIVE: A high degree of sleep fragmentation by arousals related to respiratory events would result in an abnormal distribution of slow wave activity (SWA) and a decrease in sleep spindle density in sleep disordered breathing (SDB) patients when compared to controls. METHODS: Eighteen mild SDB subjects (6 females and 12 males), aged 18-56 years with (5

Age at onset of narcolepsy in two large populations of patients in France and Quebec.

BACKGROUND: Narcolepsy usually starts around adolescence; however, there is great variability in the clinical presentation of narcolepsy. OBJECTIVE: To determine the age at onset in conjunction with severity of narcoleptic symptoms in two large populations of narcoleptic patients with a similar genetic background. METHODS: Information on age at onset and severity of the condition was obtained in 317 patients with well-defined narcolepsy-cataplexy from Montpellier (France) and in 202 from Montreal (Canada). RESULTS: The mean age at onset was 23.4 years in Montpellier and 24.4 in Montreal. The age at onset was bimodal in two independent patient populations: a first peak occurring at 14.7 years, and a second peak occurring at 35. Age at onset clearly differentiates patients with a positive family history of narcolepsy (early onset) from those without a family history. Other clinical and polygraphic findings may indicate that young age at onset is associated with increased severity of the condition (higher frequency of cataplexy and decreased mean sleep latency on the Multiple Sleep Latency Test). CONCLUSION: Bimodal distribution of age at onset of narcolepsy was found in two independent patient populations. Our data suggest that age at onset is genetically determined.

[Physiopathology of idiopathic hypersomnia. Current studies and new orientations]. / Physiopathologie de l'hypersomnie idiopathique. Données actuelles et nouvelles orientations.

In 1976 Bedrich Roth coined the term "idiopathic hypersomnia" and described two forms of the disease, one monosymptomatic, manifested only by excessive daytime sleepiness, and one polysymptomatic, characterized by excessive daytime sleepiness, nocturnal sleep of abnormally long duration and signs of "sleep drunkenness" on awakening. In comparison with that of narcolepsy, the pathophysiology of idiopathic hypersomnia remains poorly known. There are two main reasons for that: the absence of clinical and polysomnographic criteria pathognomonic or at least characteristic of the condition, as the cataplexies and the sleep onset REM periods of narcolepsy, and also the absence of a natural animal model comparable with the canine model of narcolepsy. The first investigations have stressed the frequent familial pattern of idiopathic hypersomnia. Later on biochemical assays have been performed in the CSF with results in favour of a dysfunction of noradrenergic systems. In the light of the two process model of sleep regulation in which sleep propensity is determined by a homeostatic process S and a circadian process C and of the later three-process model of regulation in which sleepiness/alertness are simulated by the combined action of a homeostatic process, a circadian process and sleep inertia, we suggest that idiopathic hypersomnia is not a pathological entity in itself, but rather the consequence of chronic sleep deprivation in very long sleepers.

[Disorders of arousal]. / Les troubles de l'éveil.

There are three slow wave sleep (SWS) parasomnias: confusional arousal, sleepwalking and night terrors grouped together in arousal disorders because impaired arousal from sleep has been postulated as a cause of these disorders. The onset of these disorders at the beginning of the night in slow wave sleep (SWS) is a typical feature. Night sleep is characterised by a great sleep fragmentation specially in SWS, and by a very strong SWS intensity reflected by both an increase of slow wave activity (SWA) before the parasomnia and a slower decay of wave activity (SWA) during the night. An abnormal deep sleep associated with a high SWS fragmentation is responsible for the occurrence of arousal disorders parasomnias.

[Advanced sleep phase syndrome]. / Le syndrome d'avance de phase du sommeil.

The Advanced Sleep Phase Syndrome (ASPS) is a sleep disorder characterized by an early sleep onset and early awakening without any disturbance of the sleep structure. The management of this disease requires clinical and laboratory investigations in an attempt to confirm the phase advance of body core temperature and melatonin rhythm. The use of light therapy, possibly associated with chronotherapy or melatonin intake has been proposed. The evolution is variable. Seven subjects, aged 15 to 72 were diagnosed in our sleep disorders unit by mean of sleep log, actigraphy, sleep and temperature recording. The sleep onset and sleep offset times were approximately the same according to sleep log, actigraphy and night polysomnography. The nadir of body core temperature was at 01:38 +/- 01:03. Two familial cases were identified of which 1 was investigated in constant routine condition with hourly blood sampling. An advanced phase of melatonin and cortisol was evidenced. The disease temporarily improved in 3 cases with light therapy and in one case with the association of light therapy and chronotherapy. These data show the difficulties of the management and the treatment of this rarely diagnosed disease.

Nocturnal sleep features in narcolepsy: a model-based approach.

The internal structure of night sleep in subjects with narcolepsy significantly differs as compared to the controls. The differences consist in the presence of sleep onset REM periods (SOREMPs) and in a longer duration (120 minutes versus 90 minutes) of the NREM-REM cycles. Another difference consists in the lack of increase in REM sleep duration during the night, cycle after cycle. These characteristics, taken as a whole, suggest that the sleep structure of narcoleptic patients is due to an imbalance between the homeostatic process and the enhanced pressure of REM inducing mechanisms. With regard to the intrasleep dynamics the current extension of the Two-process model cannot give a satisfactory explanation because REM sleep is included as an external trigger with no intrinsic rhythmic property. Neurobiologic studies lead to consider the REM cyclic occurrence as the result of reciprocal interaction between two populations of REM-ON and REM-OFF cells. The introduction in the Two-process-model extended to the intra-night dynamics of an ultradian oscillator, based on a REM-ON--REM-OFF reciprocal interaction, allows the theoretical possibility to simulate a SOREMP. This revised model accounts for the progressive extension of REM sleep-periods duration in the course of the night. The sleep structure of narcoleptic patients can be simulated leaving unaltered the pressure of the homeostatic process and enhancing the REM inducing pressure. Such an enhanced REM pressure can be achieved by increasing the value of the numerical coefficient, which represents the strength of connection between the two types of REM-ON and REM-OFF cells. This modification allows not only to obtain a stronger ultradian oscillator but also a longer periodicity of REM sleep occurrence. By coupling the homeostatic process, considered as normal, to such a modified ultradian oscillator, our model can explain the intra-night sleep dynamics of narcoleptic subjects. A REM-ON REM-OFF dysregulation can be hypothesized to explain the pathophysiological basis of nocturnal sleep features in narcolepsy.

[Circadian and ultradian cycles in narcolepsy]. / Processus ultradien et circadien dans la narcolepsie.

In narcolepsy, homeostatic process is preserved while sleep/wake circadian process is impaired. Other circadian components (body temperature, endocrine secretions, subjective sleepiness) are preserved. This circadian system weakness permits the occurrence of a very strong ultradian component, modulating sleep/wake rhythm. So, a 4 hour ultradian rhythmicity of Slow Wave Activity, and a 2 hour NREM/REM cycle longer than in normal subjects, has been evidenced in narcoleptic patients. These circadian and ultradian alterations can explain a major part of the narcoleptic symptoms.

EEG arousals and awakenings in relation with periodic leg movements during sleep.

It is known that periodic leg movements are frequently accompanied by full awakenings or by signs of EEG arousals. The time relationship of these EEG arousals with leg movements varies from patient to patient. They may precede or follow leg movements or occur simultaneously. It is not clear whether these arousals trigger leg movements or, alternatively, whether both EEG arousals and leg movements are separate expressions of a common pathophysiological mechanism. We investigated the temporal relationship of five EEG arousals, such as alpha activity, K-complexes, spindles, K-alpha, K-spindle activities and awakenings, with leg movements in 10 periodic leg movement patients. These EEG arousals were considered to be associated with leg movements if they occurred 10 s before/after or simultaneously with the onset of right or left tibialis muscle EMG potentials. It was found that 49.19% of EEG arousals occurred before leg movements, 30.61% occurred simultaneously and 23.18% occurred just after leg movements. The number of EEG arousals was significantly higher in the 10 s preceding leg movement than simultaneously or in the 10 s following. Alpha activity was the phenomenon associated most frequently with leg movements, irrespective of its temporal organization and was significantly higher during the 10 s preceding movement. Spindle and K-spindle activities were significantly higher before leg movement, whereas K-complex activity was significantly more frequent during leg movements. The number of awakenings was significantly higher after leg movements than simultaneously. These results indicated that leg movements are not primary, but rather are a phenomenon associated with an underlying arousal disorder.

Are periodic leg movements during sleep (PLMS) responsible for sleep disruption in insomnia patients?

On the basis of polygraphic findings, it has been suggested that periodic leg movements during sleep are not responsible for sleep impairment (Lugaresi et al., 1972). However, for some authors it is an important cause of insomnia (Guilleminault et al., 1975; Coleman, 1982). Thus, the relationship between periodic leg movements during sleep, sleep disruption and the complaint of patients is particularly complex. We investigated the macro- and micro-structure of sleep with and without leg movements in 10 PLMS patients complaining of insomnia to clarify whether periodic leg movements are responsible for sleep disruption. The total sleep time without periodic leg movements was significantly longer than sleep time with leg movements. Sleep time without leg movements was longer than sleep time with leg movements in stage 2, slow wave sleep (SWS) and rapid eye movement (REM) sleep. Short lasting awakenings were significantly more frequent during periodic leg movements associated sleep whilst long lasting awakenings were equally frequent during sleep with and without periodic leg movements. The percentage of the four electroencephalogram (EEG) activities (delta, theta, alpha and spindles) did not show any significant difference between periodic leg movements associated and not associated with sleep stages and total sleep time. The lack of significant differences in both the macro- and micro-structure of sleep and EEG activity content regarding the association with movements confirm the hypothesis that periodic leg movements did not primarily cause sleep disturbance.

Sleep architecture, slow wave activity, and sleep spindles in adult patients with sleepwalking and sleep terrors.

OBJECTIVES: A very strong SWS intensity reflected by both an increased level of SWA and an abnormal sleep spindles distribution would be responsible for the major difficulty of parasomniac subjects in waking up from SWS, leading to episodes of parasomnia. METHODS: Eleven adult parasomniac subjects, 6 females and 5 males, with sleepwalking (SW) and/or sleep terrors (ST) and 11 age- and sex-matched control subjects underwent polysomnography (PSG) during 2 consecutive nights. After an habituation and selection night followed by a 16 h period of controlled wakefulness, the sleep EEGs of the parasomniac and control subjects were analyzed on the second night by computer-aided visual scoring (integrated digital filtering analysis, IDFA) and spectral analysis (fast Fourier transform, FFT). Throughout the night subject behaviour was controlled and recorded by means of a video infra-red camera and videotape recorder. RESULTS: Fifteen episodes of parasomnia were recorded during the second night in the 11 subjects. Sleep analysis showed significantly (P<0.05) decreased sleep efficiency and stage 2 sleep (absolute values and percentage of total sleep time) and increased (P<0.05) slow wave sleep (absolute values and percentage of total sleep time). Arousal index and wake-time after sleep onset were significantly higher in parasomniac subjects. Sleep fragmentation was mainly concentrated in stages 3 and 4. The slow wave activity (SWA) absolute values averaged during the 2 min immediately preceding an episode of parasomnia were significantly higher than the SWA averaged during 2 min in the same stage 10 min before an episode of parasomnia. Moreover, SWA was higher in the slow wave sleep (SWS) episodes preceding the episode of parasomnia than in the episodes preceding an awakening without an episode of parasomnia. The temporal course of SWA showed a slower exponential decay in both groups, but the time constant of the curve was larger in parasomniacs than in controls. Finally, in control subjects the sleep spindle index increased from the beginning to the end of the night while it was equally distributed in parasomniacs. CONCLUSIONS: An abnormal deep sleep associated with a high SWS fragmentation might be responsible for the occurrence of SW or ST episodes.

The sleep tutorial.

A multimedia Sleep tutorial for General Practitioners was implemented from scratch. The implementation had into account the following features: 1) Specific needs of GPs evaluated in 3 different countries, related with Tutorial contents and technical features; 2) Multinational authorship from European experts; 3) Multilingual possibilities; 4) User friendliness and easy navigation. The tutorial was implemented and tested and its gama version is now available for commercialization.

Homeostatic process and sleep spindles in patients with sleep-maintenance insomnia: effect of partial (21 h) sleep deprivation.

OBJECTIVES: A low level of process 5 at bed time would be responsible for a reduced amount of slow-wave activity (SWA) leading to increased alpha activity and awakenings at the end of the night. METHODS: Following a base-line night (BLN) recording, 7 sleep-maintenance insomnia (SMI) subjects and 7 sex- and age-matched controls were maintained on 21 h of sleep deprivation. Thereafter, a recovery night (RN) was performed from 2300 h until spontaneous awakening. SWA (power density of the EEG delta band between 0.75 and 4.5 Hz) was monitored by means of spectral analysis (FFT). Sleep spindles and the occupation ratio of Rechtschaffen and Kales EEG bands were observed by integrated digital filtering analysis. RESULTS: SWA was lower in SMI subjects than in controls during RN but was higher than in BLN indicating that the homeostatic process was operating, but weaker in SMI subjects. On the other hand in SMI subjects the sleep spindle index (SSI) did not decrease during slow-wave sleep and was significantly lower than in controls. Moreover during RN the SSI decreased significantly during the first sleep cycle in controls and not in SMI subjects. The existence of an inverse relationship between SWA and SSI was therefore not observed in insomniacs. Finally the mean duration of alpha frequency significantly increased in SMI subjects. CONCLUSIONS: It is hypothesised that in SMI subjects, an alteration of the homeostatic process is responsible for insufficient sleep pressure leading to an inability to maintain sleep for an extended period.

Idiopathic hypersomnia.

Identification of idiopathic hypersomnia dates back 20 years only. It typically consists of prolonged nocturnal sleep, great difficulty waking up in the morning or at the end of a nap, and constant or recurrent excessive daytime sleepiness. Complete and incomplete forms are encountered. Twenty-three subjects fulfilling ICSD criteria are reported with clinical, polysomnographic and immunogenetic data. Considering differential diagnosis is an important step in the diagnosis of idiopathic hypersomnia. Idiopathic hypersomnia is much less frequent than narcolepsy. A strong genetic component is suggested by the high proportion of familial cases. No association with HLA has been evidenced to date.

[Regulation of sleep]. / Régulation du sommeil.

Sleep regulation calls for 3 processes: the first one is the homeostatic process increasing during wakefulness and decreasing during sleep, the second one is the circadian process depending on the circadian oscillator which controls temperature and alertness rhythms, and the third one is the ultradian process: it determines the NREM/REM periodicity. In the 2 process model of sleep regulation, the power density of the delta band (0.75-4.5 Hz) called slow wave activity (SWA) and obtained by spectral analysis, is supposed to reflect the variations of a homeostatic recovery process (process S) that increases in a saturating exponential way during wakefulness. Its decrease is expressed by the exponential decline of SWA during sleep. Process S interacts with the circadian process (process S) that determines the sleep timing. The 2 process model has been further modified to account for the semicircadian sleep propensity, although no satisfactory fit has been obtained with laboratory data. No impairment of NREM sleep homeostatic sleep regulation can be evidenced in narcoleptic patients who seem more sensitive to homeostatic regulation of sleep than normal subjects. On the other hand the circadian process appears to be weaker in narcoleptic patients than in normal subjects; this permits the occurrence of a strong ultradian component explaining diurnal sleep episodes.

[Circadian rhythm in normal and pathological sleep]. / Rythmes circadiens dans le sommeil normal et pathologique.

Different challenges, constraints for quality of sleep and wakefulness, sleep in extreme conditions, search for a better identification of disorders of sleep and wakefulness, spur researches on the circadian rhythm of sleep and on its disorders. Any advance in that field rests on the understanding of different physiologic phenomena: the intrinsic period of our internal clock is not 24-hours but a little less than 25-hours and it must be reset daily by the Zeitgebers. A sudden phase shift provokes an external desynchronisation. In addition not all rhythms resynchronize together resulting in an internal desyncronization. There is a close relationship between quantities of total sleep and REM sleep and the time of sleep onset within the 24-hours. The propensity to sleep follows a circasemidian rhythm with a major peak during the night and a secondary peak in the midafternoon. Disorders of the circadian rhythm of sleep are of two types: related to the misalignment of the sleep-wake schedule and the synchronizer-controlled rest-activity rhythm (shift work sleep disorder and time zone change syndrome) or related to an abnormal escape of the individual rest-activity rhythm from synchronizer control (delayed sleep phase syndrome, advanced sleep phase syndrome and non 24-hour sleep-wake syndrome.

Hypersomnia in association with dysthymia in comparison with idiopathic hypersomnia and normal controls.

Polysomnographic studies in hypersomniac patients with mood disorders are rare. Previous studies investigated patients with a severe mood disorder, but our study was done in patients with dysthymia, who complained of sleepiness. Mean sleep latency test (MSLT) and continuous polysomnographic recording (CPR) were done in 12 dysthymic patients, in comparison with 12 idiopathic hypersomnia patients, and 12 normal controls. In dysthymic patients mean sleep latency on the MSLT (13 +/- 1) was normal, and when CPR was done during 24 hours, no hypersomnia was found (553 +/- 24). Dysthymic patients showed an abnormal macrostructure of sleep (characterised by an excess of sleep stage I and a decrease of stages 3 and 4), which could be related to their complaint of hypersomnia.

Ultradian aspects of sleep in narcolepsy.

Following a baseline night recording, 9 narcoleptic subjects and 9 sex and age-matched control subjects were maintained on 16 hours of diurnal sleep deprivation. Thereafter subjects were submitted to a 32 hour bed rest protocol in a sound-light attenuated room. The EEG was recorded and processed using a Fast Fourier Transform. Narcoleptic patients did not differ from control subjects in total sleep time over the whole time-span. An ultradian tendency to sleep seems to be predominant in narcoleptic patients and evidence of a strong basic rest activity cycle is shown. The coupling between the homeostatic process of sleep regulation and an ultradian drive to sleep would explain the peculiar 4 hour distribution pattern of SWA in narcoleptic patients.

Use of modafinil in the treatment of narcolepsy: a long term follow-up study.

One hundred and forty patients (104 male and 36 female) aged 42.26 +/- 19.19 (range = 8 to 79.5 years) with narcolepsy-cataplexy were given modafinil (200 to 400 mg) at the Montpellier sleep disorders center from 1984 onwards. The follow-up focused on the reduction of excessive daytime somnolence (EDS), side effects and duration of treatment. In order to determine if any clinical aspect of narcolepsy could be involved in modafinil discontinuation, patients were divided into two groups according to continued or interrupted treatment. When modafinil effect on EDS was evaluated according to a scale varying from 0 (no effect) to 3 (excellent effect), 64.1% of the subjects, scored good or excellent. The mean duration of treatment was 22.05 months +/- 24.9, ranging from 1 to 114 months. Dependency signs were never observed.