Sleep Dynamics Beyond Traditional Sleep Macrostructure

In this issue of SLEEP, Sorensen and colleagues¹ report a study in which they evaluated the number of nocturnal transitions per hour between wakefulness and sleep, and between NREM and REM sleep, on the hypnograms of 63 narcolepsy patients with different representation of cataplexy and hypocretin-1 (hcrt-1) deficiency. They found that hcrt-1 deficiency was associated with more transitions between REM and NREM sleep, whereas cataplexy was associated with more transitions between sleep and wakefulness. Thus, Sorensen et al.¹ demonstrated that information extracted from conventional sleep scoring may be used to bridge the clinical evidence of nocturnal sleep disruption of narcoleptic patients to the sleep regulation model proposed after the discovery of the hypocretins system.² Accordingly, the hypocretinergic neurons sustain wakefulness (and suppress REM sleep) and stabilize wakefulness or sleep once one of these states has been reached through a flip-flop switch model.² Therefore, the loss of hypocretinergic signaling in human narcolepsy explains the emergence of full-blown REM sleep during daytime (i.e., sleep attacks) and a dissociated manifestation (e.g., cataplexy viewed as REM sleep atonia intruding into a wake state). Moreover, the incomplete sleep-wake switch could lead to a boundary dyscontrol with emergence of dissociated states during sleep (sleep paralyses, hallucinations, and REM sleep behavior disorder), to the inability to maintain sleep across the night (low sleep efficiency), and also to high transitional rates between different states (wakefulness, REM, and NREM sleep).³

Furthermore, the classification adopted by Sorensen et al.¹ to cluster their patients offers another interesting point: out of 63 patients, 43 (68%) presented with cataplexy, while 20 (32%) presented without cataplexy. However, hcrt-1 was in the normal range in 8 (19%) out of the 43 cataplectic patients, whereas it was deficient in 7 (35%) out of 20 patients without cataplexy. These data raise issues relative to the definition of cataplexy, which is currently based on anamnesis only.⁴ Indeed, hypocretin neurons proved to be absent in brains of narcolepsy with cataplexy patients,⁵ while hcrt-1 deficiency best identified cataplexy patients.⁶ Conversely, hypothalami of narcoleptic patients without cataplexy showed a partial hypocretinergic cell loss,⁷ and the threshold of cerebrospinal hcrt-1 able to identify these patients is as yet unknown: either normal values frequently occur despite high recurrence of sleep onset REM periods (SOREMPs) during the MSLT, or cataplexy appears several years after the lumbar puncture.⁸

While the prevalence of narcolepsy with cataplexy appears to be between 20 and 50 per 100,000 people,^9,10 the prevalence of narcolepsy without cataplexy is still unknown,⁴ but it is believed to be an even rarer disease and/or misdiagnosed with other conditions such as behaviorally induced insufficient sleep syndrome (BIISS) or circadian rhythm sleep disorders. The diagnosis of narcolepsy without cataplexy is even more challenging given the relatively frequent occurrence of SOREMPs in the general population.¹¹ At the same time, it should be emphasized that the current definition of cataplexy (i.e., a sudden loss of muscle tone triggered by strong emotions) seems elusive for several reasons. First, when cataplexy is assessed exclusively by history, it may be confused for other psychiatric symptoms including malingering,^12,13 thus suggesting the need to document it whenever possible, and to recognize that our knowledge on cataplexy and its mimics has to be expanded. Second, although cataplexy can rarely occur also in the absence of any apparent emotional stimulation in adult patients,¹⁴ this phenomenon is extremely frequent in children with narcolepsy with cataplexy, who display a persistent hypotonic condition involving the facial district or the whole body.^15,16 The latter point supports the need for further studies phenotyping cataplexy features across the lifespan.

Recent studies approach the complexity of narcolepsy using different polysomnographic techniques: the presence of SOREMP at nocturnal polysomnography showed good specificity but low sensitivity,¹⁷ whereas daytime SOREMP occurrence in spontaneous naps performed as well as the MSLT to identify narcolepsy patients.¹⁸ Moreover, the sleep onset period profile distinguished both idiopathic hypersomnia (less abrupt transition into a sustained sleep state)¹⁹ and BIISS (preferential entry into SOREMP after NREM sleep stage 2)^20,21 from narcolepsy. Also the analysis of the cyclic alternating pattern (CAP) during sleep showed peculiarities of narcolepsy with cataplexy patients (reduced CAP rate particularly affecting A1 components) versus both controls²² and other hypersomniacs.²³ Traditional quantitative EEG studies of the sleep onset period²⁴ or of sleep homeostasis²⁵ have thus far not identified any discrete markers for narcolepsy. Conversely, the integration of power spectra results in terms of state space analysis seems the most promising quantitative tool to characterize the different wake, NREM and REM sleep states, if it is shown to provide reliable results in humans.^26,27 Other quantitative polysomnographic features have shown consistent peculiarities in narcolepsy, such as the atonia index quantifying the degree of chin tone as a correlate of REM sleep behavior disorder^28,29 and the periodicity index of periodic leg movements during sleep.³⁰

Overall, the analysis of sleep stage transitions, such as conducted by Sorensen et al.¹ has the key advantage of utilizing already available data derived from sleep macrostructure to link the dynamics of sleep macrostructure to anatomical, physiological, and pathophysiological aspects of sleep and wakefulness.^31,32 Given the evidences of altered sleep transitions in other disturbances such as obstructive sleep apnea^33,34 or fibromyalgia,^35,36 further studies are warranted to verify whether the increased state transitions now reported in narcolepsy may identify narcolepsy patients,¹ either alone or in combination with one or a set of the other above mentioned parameters, and render information on disease severity, or ways to track the time course of narcolepsy.

CITATION

Plazzi G; Pizza F. Sleep dynamics beyond traditional sleep macrostructure. SLEEP 2013;36(8):1123–1124.

DISCLOSURE STATEMENT

The authors have indicated no financial conflicts of interest.