Review of Rapid Eye Movement Behavior Sleep Disorders Vivien C. Abad, MD, MBA, and Christian Guilleminault, MD
Address Stanford Sleep Disorders Center, 401 Quarry Road, Suite 3301, Stanford, CA 94305, USA. E-mail:
[email protected] Current Neurology and Neuroscience Reports 2004, 4:157–163 Current Science Inc. ISSN 1528–4042 Copyright © 2004 by Current Science Inc.
The spectrum of rapid eye movement behavior disorders (RBD) spans various age groups, with the greatest prevalence in elderly men. Major diagnostic features include harmful or potentially harmful sleep behaviors that disrupt sleep continuity and dream enactment during rapid eye movement sleep. In RBD patients, the polysomnogram during rapid eye movement sleep demonstrates excessive augmentation of chin electromyogram or excessive chin or limb phasic electromyogram twitching. RBD may be associated with various neurodegenerative disorders, such as multiple system atrophy, Parkinson’s disease, and dementia with Lewy bodies. Other co-morbid conditions may include narcolepsy, agrypnia excitata, sleepwalking, and sleep terrors. RBD is hypothesized to be caused by primary dysfunction of the pedunculo-pontine nucleus or other key brainstem structures associated with basal ganglia pathology or, alternatively, from abnormal afferent signals in the basal ganglia leading to dysfunction in the midbrain extrapyramidal area/ pedunculo-pontine nucleus regions.
Introduction In 1986, Quera-Salva and Guilleminault [1] reported two patients with olivopontocerebellar degeneration who acted out their dreams during rapid eye movement (REM) sleep and whose polysomnograms (PSG) demonstrated REM without atonia. They related these findings to Jouvet and Delorme’s [2] lesioned cats that exhibited dream enactment after para locus coeruleus lesions. Soon after, Schenck et al. [3] described four men and one woman who exhibited aggressive or disruptive behaviors during attempted dream enactment while asleep. These behaviors were accompanied by variable loss of chin atonia, increased limb-twitch activity, and increased REM ocular activity and density. These individuals had no associated neurologic syndromes, and Schenk and Mahowald [3,4••]
called this new category of parasomnia in humans “REM behavior disorder” (RBD). Rapid eye movement behavior disorder has a prevalence of 0.5% in the general population of United States, and approximately 87% of those afflicted are men [5]. Racial differences in incidence and prevalence have not been reported [5]. There is a strong association between RBD and certain neurodegenerative disorders, including Parkinson’s disease, multiple system atrophy, and dementia with Lewy bodies. Although animal models have been proposed, the spectrum of human RBD appears to be more complex. We review the pathophysiology of RBD, including current hypotheses on human pathophysiology, describe the typical clinical features associated with RBD patients and its various co-morbid conditions, and discuss the diagnostic and treatment modalities for RBD patients.
Pathophysiology
Animal model The seminal work by Jouvet and Delorme [2] in 1965 provides us with the animal model for RBD. Bilateral, symmetrical, mediodorsal pontine tegmental electrolytic lesions in cats resulted in loss of REM atonia and the enactment of phasic electrophysiologic and motor activity in non-rapid eye movement (NREM) sleep. Cats that had sustained lesions of either the caudal and ventral pons or bilateral lesions of the descending pathways responsible for REM atonia demonstrated oneiristic behavior (dream enactment) [2,4••,6]. Observed behaviors included attack behavior, generalized limb or truncal twitching/jerking, orienting and exploratory behaviors, stalking of imaginary prey, and locomotion. Attack behavior correlated with lesions extending rostro-ventrally into the midbrain [6]. In addition, Sanford et al. [7] reported that small unilateral lesions in the reticularis pontis oralis and reticularis pontis caudalis in rats resulted in abolition of REM atonia, whereas large bilateral lesions were necessary to release overt elaborate behaviors and abolish REM atonia. Based on such experiments, various researchers have suggested that RBD is produced by loss of REM atonia and disinhibition of mesencephalic motor pattern generators, thereby resulting in phasic motor overactivation with behavioral release during REM sleep [2,4••,6,7].
158
Sleep
Human model In humans, the pathophysiology of RBD has not been as clearly demonstrated. Rye [8] hypothesized that RBD may arise from injury to the pedunculopontine nucleus (PPN) or from abnormal afferent signals in the basal ganglia, transmitted via the pallidotegmental tracts, altering neural responsiveness in the midbrain extrapyramidal area (MEA)/PPN regions. Magnetic resonance imaging (MRI) studies in RBD patients have shown lesions consistent with lacunar infarcts in the dorsal pontomesencephalic regions and basis pontis, suggesting injury to the midrostral tegmental nuclei or the tegmento-reticular tracts or both [9,10]. Single photon emission computed tomography (SPECT) imaging has demonstrated significantly reduced striatal dopamine transporters in RBD patients, and positron emission tomography testing has shown reduced striatal dihydrotetrabenazine binding, indicating loss of dopaminergic midbrain neurons [11,12]. Albin et al. [12] proposed that RBD could result either from basal ganglia impairment leading to secondary dysfunction of the PPN or from primary dysfunction of the PPN or other key brainstem structures associated with basal ganglia pathology. Using MRI and SPECT imaging, Shirakawa et al. [13] demonstrated decreased blood flow in the upper part of both frontal lobes and in the pons and postulated that decreased blood flow to these areas might be associated with the pathogenesis of RBD. Proton magnetic resonance spectroscopy ( 1H MRS) studies have been inconsistent in demonstrating functional impairment in brainstem neurons [14,15]. Approximately one half of patients in Olson et al.’s [16•] and Schenck and Mahowald’s [17] series did not have identifiable neuropathologic processes involving the brainstem, whereas other patients demonstrated a marked reduction in noradrenergic locus coeruleus (LC) neurons. Despite the reduction in LC neurons, Schenck et al. [18] did not detect anti-LC antibodies to account for this. Although the causes are unclear, loss of LC neurons results in disinhibition of the PPN and the lateral dorsal tegmentum, thereby allowing motor activity during REM sleep [19]. Mahowald and Schenck [20] postulated that disinhibition of selective brainstem motor generators may produce both the behavior and dream disorder of RBD. The generator for violent behaviors may become disinhibited and coactivate an ascending output to forebrain dream-synthesizing centers and a descending output to the spinal motorneurons, thereby producing dreaming and simultaneous movement [20].
Idiopathic and Secondary RBD Between 25% and 79% of RBD in humans is idiopathic, but 36% to 67% of idiopathic RBD patients’ condition evolves into a “symptomatic” form (Parkinson’s/dementia) within a 3- to 29-year period, with a mean latency of 20.3 years [4••,21].
Rapid eye movement behavior disorder can also occur secondarily on an acute or chronic basis [4••,20–22]. Acute RBD can result from drug intoxication (biperiden, caffeine, tricyclic antidepressants, monoamine oxidase inhibitors) or from drug withdrawal (butalbital, meprobamate, pentazocine, nitrazepam) [20,23]. Chronic RBD can be produced by vascular causes (vasculitis, subarachnoid hemorrhage); tumors (pontine neoplasms, acoustic tumors); infectious/postinfectious diseases (GuillainBarre); degenerative or demyelinating conditions (amyotrophic lateral sclerosis, fatal familial insomnia, Parkinson’s disease, dementia, multiple sclerosis, Shy-Drager syndrome, olivopontocerebellar degeneration, multiple system atrophy), drugs (tricyclic antidepressants, fluoxetine, venlafaxine, mirtazapine, selegeline, anticholinergic medications); or from developmental, congenital, or familial diseases (Tourette’s syndrome, mitochondrial encephalomyopathy, narcolepsy, Group A xeroderma pigmentosum) [16•,20,24,25].
Presentation
Adult RBD Rapid eye movement behavior disorder is characterized by excessive motor activity during dreaming, loss of REM sleep atonia, and disruption of sleep, often accompanied by violent or injurious behavior during sleep [4••,16•,26]. An example of such behavior was described by a patient known as HW, a 75-year-old man who suffers from RBD. As he stood on the riverbank, HW was terrified because the assailants pursuing him were getting closer. He frantically looked around and perceived the river as an escape route. He jumped into the water, but ended up screaming in pain. For, alas, the riverbank was his bed, and the river was the hard floor, which fractured his toe. The major clinical features in RBD patients include male predominance, older age, prodrome, dream-enactment, altered dream process and content, sleep-related injury, and polysomnographic abnormalities of excessive loss of REM sleep atonia and/or increased phasic electromyogram (EMG) activity [4••,16•,21,26]. Sex hormones may play a role in mediating the aggressive and violent behaviors seen during REM sleep and could explain the male preponderance (87%) of RBD patients [4••]. Although adult RBD usually presents between the ages of 36 and 84 years, the mean age of onset for RBD patients ranges from 52 to 62 years [4••,16•,26]. RBD increases in complexity and intensity with age. In approximately 25% of patients, a prodrome of sleep talking, yelling, limb twitching, and gross limb and body jerking without complex behaviors can occur between 2 and 48 years prior to the frank onset of clinical RBD [2]. Dreamenacting behaviors occurred in 93% of patients in the Olson et al. [16•] (Mayo Clinic) series, in 87% of the Schenck and Mahowald [4••] series, and in 64% of the Sforza et al. [26] (Strasbourg) series. Behaviors included flailing and punch-
Review of Rapid Eye Movement Behavior Sleep Disorders • Abad and Guilleminault ing the arms, kicking, vocalization (swearing, yelling, talking, laughing), gesturing, reaching, grabbing, sitting, jumping out of bed, crawling, and running. Complex behaviors observed included attempting to leap through a window, attempting to set fire to a bed, firing an unloaded gun, and sexual assault [4••,16•,28]. Patients acted out distinctly altered dreams, but did not act out their customary dreams. Dream content included defense against attack (by people [57%] or animals [30%]), adventure dreams (9%), sport dreams (2%), and aggression by the dreamer (2%) [16•]. Eighty-seven percent of patients described feelings of fear and anger during dreams of being chased or attacked by unfamiliar people, animals, or insects [4••]. A subgroup of patients experienced recurrent dreams with presumed vestibular activation (spinning objects or angular motion with acceleration). Another subgroup recounted dreams of suddenly getting stuck in the mud, trapped in deep snow, or falling to the ground unable to get up, suggesting recurrent atonia intruding into their dream-enacting episodes. In the Strasbourg series, 64% of patients described vivid dreaming during dream enactment [26]. The frequency of expressed behaviors ranges from once every 2 weeks to a maximum of four times per night on 10 consecutive days [20]. Injuries to the patient and the bed partner are common (51% to 79%) [4••,21]. Sixty-four percent of spouses reported being assaulted (punched, slapped, kicked, hair pulled, and choked), whereas 32% of patients injured themselves by striking or bumping into furniture or walls or falling out of bed [16•].
Atypical sexual behavior and RBD Guilleminault and Silvestri [27] and Guilleminault et al. [28] reported that injuries to bed partners can also occur as part of atypical sexual behavior during sleep. A 29-year-old patient uttered profanities and forcibly held down and sexually assaulted his partner, who sustained cuts and bruises. A 38-year-old patient tore off his wife’s clothes, fondled her, tried sexual intercourse, and on one occasion attempted to choke her. A 33-year-old man grabbed his wife and forced intercourse [28]. Juvenile and adolescent RBD Secondary and idiopathic RBD have been described in children and adolescents as early as 11 months of age [4••]. RBD has been reported in children and adolescents with Tourette’s syndrome, Group A xeroderma pigmentosum, narcolepsy, juvenile Parkinson’s disease, and childhood autism [4••,29]. Like their adult counterparts, children with RBD can exhibit a wide range of behaviors, ranging from agitated crawling and violent jerking to complex behaviors associated with nightmares and vivid dreams. Status dissociatus Human existence encompasses three states: wakefulness, NREM sleep, and REM sleep. Mahowald and Schenck [30] described six patients with complete breakdown of
159
state boundaries, which they termed “status dissociatus.” Simultaneously occurring elements of wakefulness, NREM, and REM states were seen during both behavioral sleep and self-perceived sleep. Self-perceived sleep was behaviorally different from wakefulness, with dream-like mentation and complex motor activity and vocalization; the electroencephalogram (EEG) demonstrated a wakeful-like pattern with superimposed slowing and poorly developed sleep spindles and vertex-like activity. The EEG slowing during sleep disappeared during wakefulness, suggesting a NREM phenomenon. The only fully declared clinical or polygraphic state was wakefulness. Kushida et al. [31] also reported the occurrence of sleep walking during slow wave sleep and REM behavior disorder in a case of Machado-Joseph syndrome.
Diagnosis The International Classification for Sleep Disorders lists the following minimum criteria for diagnosis of RBD: limb or body movement associated with dream mentation together with at least one of the following: 1) harmful or potentially harmful sleep behaviors, 2) dreams that appear to be “acted out,” and 3) sleep behaviors that disrupt sleep continuity [32]. The overall sleep architecture is usually normal, with cycling of NREM and REM sleep. During NREM sleep, prominent aperiodic movements of all extremities may be seen. Using quantitative analysis of NREM sleep EEG in RBD patients, Massicotte-Marquez et al. [33] showed a higher percentage of NREM sleep compared with control subjects, and spectral analysis demonstrated higher spectral power in the slow oscillation frequency (0.25 to 1Hz) band. Polysomnography during REM sleep demonstrates excessive augmentation of chin EMG tone or excessive chin or limb phasic EMG twitching, as illustrated in Figure 1 [4••,32]. There is variability of involvement and distribution of the extremity movements and the chin EMG. The chin EMG may be atonic during extremity movements, or it may be augmented without body movements. EMG activity during REM in RBD patients consists of “long-lasting” muscle activity (increased muscle activity lasting ≥0.5 seconds) and “short-acting” muscle activity (increased muscle activity lasting
