Clinical manifestations

Presentation and course

Sleep paralysis is characterized by a brief period of inability to perform voluntary movements occurring at sleep onset (hypnagogic or predormital form) or when awakening, either during the night or in the morning (hypnopompic or postdormital form). Isolated sleep paralysis frequently occurs when awakening (in 64% of patients), whereas in narcolepsy and familial sleep paralysis, the episodes are more frequent at sleep onset. Spontaneous sleep paralysis episodes showed a skewed distribution over the night, with one quarter of episodes occurring within one hour of bedtime, although episodes were reported throughout the night with a minor mode around the time of normal waking.

Typically during sleep paralysis, patients are powerless to move the head, trunk, and limbs as well as to speak and even open the eyes, although ocular and respiratory movements are intact. Sometimes, patients feel unable to breathe deeply and have a feeling of suffocation frequently related to the sensation of a being that exerts pressure on the thorax (28). The episodes can be preceded by an “aura” with auditory, tactile, or visual perceptions (16). During sleep paralysis, the patient is fully awake or half awake, the sensorium is clear, awareness of surroundings and self is fairly well-preserved, and patients are able to recall the sleep paralysis afterward. During sleep paralysis, patients usually engage in a mental struggle to cry out and to move, yet they appear to be in deep sleep or dreaming. Many episodes of sleep paralysis, especially the first to take place in a patient's life, cause extreme anxiety associated with fear of dying during the episode. This anxiety is greatly intensified by terrifying dreamlike images or hallucinations (hypnagogic or hypnopompic hallucinations). Pleasant sensations during the episodes are less frequent (21) and usually associated with out of body experiences (16). Women report experiencing more phenomena and emotions, and more fearful emotions than men (especially the fear of going crazy), and are more likely to perceive concrete forms such as human figures or people they know (25).

In a study on 185 patients with isolated sleep paralysis who underwent a clinical interview, 57% reported to sense a presence in the room, usually a nonhuman presence (39). In addition, 21% of the sample experienced visual hallucinations of others, with the majority perceiving strangers as opposed to known individuals.

Sleep paralysis is transient, lasting only one to several minutes (average duration of episodes: 4 minutes) and disappearing either spontaneously or on external stimulation, especially by simple touch or movement induced by another person. Some patients note that repeated efforts to move the body or make vigorous eye movements may help to abort a paralytic episode. Patients may sometimes relapse into sleep paralysis if they remain supine; indeed, sleep paralysis has been shown to occur predominantly with the patient sleeping in the supine position. Both body position and timing of the sleep paralysis episodes were found to affect incidence and, to a lesser extent, quality of the sleep paralysis experience. Patients may also fail to establish the boundaries between real experience and dream mentation during the paralysis, suggesting a dissociated state of mind between wakefulness and REM sleep (42).

Once ended, sleep paralysis leaves no sequelae. Neurologic examination is normal. On rare occasions, some weakness or numbness in the extremities remains for a few minutes.

The current International Classification of Sleep Disorders (02) cites the following:

Prognosis and complications

No studies have demonstrated any long-term consequences on the health of patients who experience sleep paralysis. Although the underlying risk factors may contribute to health-related issues later in life (for example, anxiety poses a risk for hypertension), there is no reported independent association (13).

Isolated sleep paralysis is associated with shorter sleep duration, longer sleep onset latency, and greater insomnia symptoms (34).

Even if episodes are frequent, no sequelae or serious effects on the life of patients, are presented, with a normal state returning between episodes (10).

Sleep paralysis peak onset is between 17 and 19 years of age, with no sex difference (44). Episodes become less frequent with age, often with spontaneous remissions and relapses.

Clinical vignette

A 31-year-old male without relevant family history reported onset at 13 years of age of frightening episodes during which he was unable to move, speak, and even open his eyes. Episodes occurred first during naps and then during nocturnal sleep. During such episodes, he remained conscious and well in contact with the environment. He reported no dreaming or hallucinatory activity. The episodes lasted a few minutes and had a frequency of two to three times per night every night for the initial two years. There followed a spontaneous improvement until 22 years of age, when the frequency of the attacks returned to previous levels. The patient also reported recurrent nocturnal obsessive thoughts concerning his work and daily activities and increased perspiration. Frequency again decreased to one episode every 2 to 3 months until 28 years of age; the patient was now able to ask for help during the episodes. Immediately before admission, however, episodes recurred in clusters with free periods of up to 7 to 10 days. The patient felt tired and had tension-type headache but had never noticed sleep or drop attacks during the daytime. Neurologic examination and brain MRI were normal. Human leukocyte antigen typing was positive for the DRB1 01 and 0701, and DQB1 05 and 03032 antigens, not typically associated with narcolepsy. A night polysomnographic recording showed normal hypnic patterns and normal REM sleep duration and latency. The first REM episode was, however, associated with a lack of physiological atonia in chin muscles, and stages 2 of NREM sleep were characterized by several periods lasting 10 to 20 seconds, with diffuse alpha EEG activity and antigravitational muscle atonia. Mild snoring in the supine position was not associated with changes in arterial oxygen saturation. The physiological recordings ran for 48 hours; meanwhile, the patient suffered four nocturnal episodes of his typical "paralysis," all arising during REM sleep and with diffuse alpha mixed to theta EEG activity and atonia of chin muscles.

A multiple sleep latency test showed a normal sleep latency of 20' and no diurnal sleepiness. A diagnosis of isolated sleep paralysis was made.

Biological basis

Etiology and pathogenesis

The etiology of isolated recurrent sleep paralysis is unknown. Sleep paralysis may be found in families and occurs frequently as one of the classic symptoms of narcolepsy. The involvement of the hypocretin system has been proved in canine and murine narcolepsy and decreased or absent hypocretin levels have been found in the CSF of humans with narcolepsy. Hypocretins (also called orexins) are neuropeptides localized in synaptic vesicles in the hypothalamus that exert wake-promoting and REM sleep suppressant effects interacting with the aminergic and cholinergic systems. The hypocretin system projects densely from the lateral hypothalamus to the locus coeruleus, and hypocretin administration increases neuronal discharges in the locus coeruleus. Activity in the locus coeruleus ceases before and during cataplectic attacks, and locus coeruleus discharges are scant in narcoleptic dogs. Thus, deficient hypocretin pathways may remove excitation from the locus coeruleus, increasing the propensity for cataplexy and possibly sleep paralysis. However, no studies of hypocretins in isolated sleep paralysis have been executed; therefore, the relationship between isolated sleep paralysis and hypocretins remains unknown, and no association with human leukocyte antigens has been found differently from narcolepsy.

In a study on 16 normal subjects, polysomnogram recordings showed a REM/wake stage dissociated state during sleep paralysis (abundant alpha electroencephalographs and persistence of muscle atonia shown by the tonic electromyogram) (41). For this reason, sleep paralysis has been considered a condition characterized by a pathological dissociation between the level of alertness/self-awareness typical of wakefulness and the generalized muscle atonia typical of REM sleep. These abnormalities are probably attended by hyperactivity of cholinergic mechanisms in pontine or suprapontine neuronal structures as well as deficits in noradrenergic or serotonergic mechanisms in the brainstem that exert a gating influence on the cholinergic neurons responsible for muscle atonia during REM sleep. EEG studies on people with narcolepsy showed, with spectral analysis, that sleep paralysis is associated with EEG features typical of REM sleep (not of wakefulness), including mainly theta rhythms and occasional bursts of alpha rhythms (concomitant with rapid eye movements) as well as muscle atonia. Consequently, the concept of the sleep paralysis as an awake brain in a paralyzed body has been reviewed in favor of a rather partially aware but dreaming state (42; 24).

From a neuropsychological perspective, sleep paralysis hallucinations have been shown to revolve around a core experience of “sensed or felt presence,” consistent with the hypothesis of REM sleep initiation of a threat-activated vigilance system. Among the varied hallucinations associated with sleep paralysis, out-of-body experiences and vestibular-motor sensations represent a distinct factor. The results of self-report data from two online surveys of sleep paralysis were interpreted as consistent with the hypothesis that out-of-body experiences represent a breakdown in the normal binding of bodily-self sensations, suggesting that out-of-body feelings are consequences of anomalous vestibular motor experiences and precursors to a particular form of autoscopic experience, out-of-body autoscopy (06). The same “disintegration model” phenomenology (illusory movement experience, out-of-body feelings, out-of-body autoscopy) has been documented in recreational ketamine users, supporting a potential temporal-parietal dysfunctional interaction induced either by REM sleep or by substance abuse. The occurrence of dissociative experiences during wakefulness may link sleep paralysis and lucid dreaming occurrence in otherwise healthy subjects (11).

Epidemiology

Surveys of normal subjects indicate that isolated sleep paralysis occurs at least once in a lifetime in a substantial proportion of otherwise normal subjects.

The prevalence of sleep paralysis varies from countries and ethnic groups, and this disparity has been linked to different methodologies in determining the prevalence, including the different definitions of sleep paralysis (32).

Prevalence and incidence of the disorder are predicted by age, race, major depressive disorder, pain, hypersomnolence, cataplexy, hypnagogic and hypnopompic hallucinations, posttraumatic stress disorder, a reduction in sleep duration of ≥60 min, and the use of analgesic or antipyretic medication (30).

In a systematic literature review aggregating 36,533 subjects across different studies, 7.6% of the general population, 28.3% of students, and 31.9% of psychiatric patients reported at least one sleep paralysis event (37).

In a sleep survey of 8162 citizens, the cumulative experience of sleep paralysis was 39.6%, with the onset peaking at 16 years of age (23). However, Ohayon and colleagues found strikingly lower rates: 6.2% of 494 otherwise normal subjects reported at least one episode of isolated sleep paralysis in their lifetime, with 4% reporting less than one episode per month and 0.8% more than one per week (31). Using International Classification of Sleep Disorders criteria, 2.7% had sleep paralysis at least once per week, 18.2% once per month, and 75.5% less than once per month.

Fourteen percent of Nigerian young adults reported sleep paralysis.

Prevalence rates among students from various universities ranges from 30% to 40% (03; 12), with a peak onset between 17 and 19 years of age and higher prevalence in women (29).

In many patients with narcolepsy, sleep paralysis occurs every night in some periods of life but becomes much less frequent in other times of life.

Approximately 41% of patients with narcolepsy frequently suffer from sleep paralysis, which is sometimes associated with rhythmic movement disorder as motor manifestation of the attempt to "shake out" of the paralysis (33). The prevalence of sleep paralysis ranges from 25% to 74% in patients with narcolepsy type 1, and its progressive appearance along the narcolepsy course suggests it may be an age-dependent phenomenon (35).

Prevention

Sleep paralysis is frequently associated with narcolepsy and may occur in other conditions characterized by disturbed nocturnal sleep and wake-sleep rhythm. Age, BMI, smoking, alcohol consumption, sleep duration, and perceived stress are all factors modulating the incidence and severity of the disorder (45). In general, irregular sleep habits, sleep deprivation, disturbed sleep-wake rhythm, psychological stress, overtiredness, and sleeping in the supine position have been considered predisposing factors and should be avoided. Isolated sleep paralysis is frequent among night-shift workers (ie, nurses, associated with insomnia) (23) and in jet lag syndrome. Sleep paralysis may even be purposefully elicited by sleep interruption in normal subjects, especially in subjects with a low tolerance for disrupted sleep-wake rhythms in whom it appears to occur as a phenotype of REM dissociation during sleep onset REM periods.

Differential diagnosis

Confusing conditions

In general, the features of sleep paralysis are sufficiently clear, and its diagnosis is not difficult. The inter-observer reliability of the former International Classification of Sleep Disorders minimal diagnostic criteria for sleep paralysis was also found substantial, with a kappa value of 0.69 (range 0 to 1, ie, maximum agreement).

Isolated and familial sleep paralysis can be readily distinguished from narcolepsy, in which excessive daytime sleepiness, cataplexy, and vivid hypnagogic hallucinations frequently occur along with sleep paralysis. Cataplexy is differentiated by its occurrence in the waking state, rather than in sleep-wake transitions in the lying position, as well as by its precipitation by strong, mainly positive emotions.

Nightmares are REM-based parasomnias that can be confused for sleep paralysis for the associated hallucinations and dream imagery. The conscious awareness of one’s surroundings and atonia is missing in nightmares.

Atonic generalized epileptic seizures can be differentiated by their usual occurrence in the daytime waking state and by their epileptic EEG abnormalities.

Atonic drop attacks in patients with vertebrobasilar vascular insufficiency or in those with moyamoya disease usually occur in the waking state, are often accompanied by transient clouding of consciousness or unconsciousness, and are not related to sleep-wake transitions in bed.

Hypokalemic periodic paralysis often occurs on awakening but lasts hours or even days; patients cannot be so easily and promptly relieved of the motor paralysis by touching or moving their bodies. Hypokalemic periodic paralysis, which shows low serum potassium levels during attacks, may be provoked by high carbohydrate meals or alcohol and is readily reversed by correcting the hypokalemia.

Hysterical paralysis and catatonia must be distinguished but are usually evident by their associated clinical features. Occasionally, the dramatic symptoms associated with sleep paralysis may be mistaken for a psychotic state.

Nocturnal panic attacks such as sleep paralysis are both characterized by fear and acute distress. Nocturnal panic attacks lack paralysis and dream imagery and are unexpected, acute, and scary, whereas fear in sleep paralysis is often secondary to the paralysis/hallucinations (36).

The sensory features of sleep paralysis are mimicked by those of peduncular hallucinosis, a condition characterized by formed and colored visual images, which the patient knows are unreal, associated with lesions of the pons, midbrain, and diencephalon. Peduncular hallucinosis, however, lacks the motor features typical of sleep paralysis (43).

Associated or underlying disorders

Sleep paralysis is frequently associated with narcolepsy (41% of patients with narcolepsy suffer from this disorder). For this reason, sleep palsy is included in the "narcoleptic tetrad" together with sleep attacks, cataplexy, and hypnagogic hallucinations, but rarely they are the first symptom of narcolepsy (05). Further sleep disorders associated with sleep palsies are insomnia, obstructive sleep apnea, idiopathic hypersomnia, hypnopompic hallucinations, nocturnal leg cramps, and nightmares (09).

Psychiatric comorbidity has been associated with sleep paralysis, in particular posttraumatic stress disorder. Furthermore, panic disorder, generalized anxiety disorder, death anxiety, and social anxiety are frequent symptomatic accompaniments to sleep palsies (36).

Other associated disorders are generalized epilepsy, multiple sclerosis, automatic behaviors, and physical disease (31; 07).

Isolated sleep paralysis is often associated with substance use, stress, and trauma, in particular trauma related to bereavement (04), as well as genetic influences, physical illness, personality, intelligence, anomalous beliefs, symptoms of psychiatric illness in nonclinical samples (particularly anxiety symptoms), and psychiatric disorders (10).

Sleep paralysis and cataplexy occur more often in patients affected with Wilson disease. Patients with myotonic dystrophy type 1 with subjective sleepiness also reported more cataplexy-like and sleep paralysis symptoms.

Diagnostic workup

Sleep palsy diagnosis is mainly clinical. An accurate clinical history is essential to exclude other disorders associated with sleep paralysis such as narcolepsy. The Unusual Sleep Experiences Questionnaire was developed to assess the subjective phenomenology of during the episodes.

Polysomnographic recordings under audiovisual control may be useful in the diagnosis of sleep paralysis and should be entertained whenever attacks are frequent enough to permit recording. Polysomnography usually documents REM and wake-dissociated states or REM and theta and occasional alpha rhythms (concomitant with rapid eye movements) associated with muscle atonia.

Management

In general, patients who frequently experience sleep paralysis learn that the episodes are brief and benign, with no significant aftermath. When episodes of sleep paralysis are infrequent, medical treatment is often unnecessary. An increased frequency of sleep paralysis also appears to lead to habituation and de-dramatization in some individuals (25).

Behavioral strategies in these cases are important and are used to prevent or disrupt isolated sleep paralysis (36). Techniques used to prevent them include sleep position or sleep patterns changes (staying up late, eliminating naps, and attaining a more regulated sleep routine), environment changes (increasing light/illumination, not sleeping in an enclosed space, decrease room’s temperature), resisting falling asleep/wake up, relaxation techniques, diet modification, caffeine use cessation, and stress reduction. Disrupting strategies include attempting to move extremities or other body parts, calling out, and calming down strategies (use of reassuring self-talk, trying to let go, regulating breathing, and the use of prayer during the episodes) (34). Cognitive behavioral therapy helps patients to learn some of these techniques or to cope with frightening hallucinations, discuss catastrophic thoughts, and resolve sleep paralysis episodes with imaginary rehearsal (36). A similar approach has been used with meditation relaxation (MR therapy), which includes four steps: reappraisal of the meaning of the attack, psychological and emotional distancing, inward focused-attention meditation, and muscle relaxation (20).

To reduce the negative emotions usually associated with sleep palsy, Herrero and colleagues proposed a new clinical approach for the treatment of recurrent sleep palsy, based on the recognition of auditory, tactile, and visual sensations preceding the episodes (16). When the patient recognizes these symptoms, he or she could try to induce out of body experiences that are usually associated with more positive emotions (happiness, love, peace, tranquility, hope, euphoria, and curiosity).

When these methods are not successful and when sleep paralysis occurs frequently enough to cause fear or anxiety about sleep, pharmacologic therapies can be considered. These therapies include drugs that modify the REM stage of sleep where sleep paralysis appears. These drugs are tricyclic antidepressants such as clomipramine (25 to 50 mg qd), imipramine (25 to 150 mg qd), protriptyline (10 to 40 mg), and desmethylimipramine (25 to 150 mg qd); selective serotonin reuptake inhibitors should also be taken into account such as fluoxetine (40 to 80 mg qd), paroxetine (25 mg qd), and escitalopram (10 mg qd) (36; 17). In narcolepsy there is an ongoing debate on the use of sodium oxybate, a GABAergic compound used to treat sleepiness and cataplexy in narcolepsy with cataplexy (10).

Outcomes

Usually, behavioral strategies are effective to prevent sleep paralysis. In a study on 156 undergraduates changes to sleeping patterns were advocated by subjects as being the most successful prevention technique (36). Meditation-relaxation therapy (MR therapy) showed promising results in subjects with isolated and recurrent sleep paralysis or in subjects with narcolepsy and sleep paralysis (19).

When pharmacological therapy is necessary for sleep paralysis, the most commonly used drug are tricyclic antidepressants and selective serotonin reuptake inhibitors. Anticholinergic side effects are the most common sides effects of tricyclic antidepressants, whereas sexual disfunction, sleepiness, and weight gain are the most common of selective serotonin reuptake inhibitors. These drugs have been mainly utilized in the context of narcolepsy. Unfortunately, the assessment of sleep paralysis and the particular nature of the outcome measures are either inconsistent across studies or not well defined (36).

Special considerations

Pregnancy

The prevalence of sleep paralysis decreases during the first trimester of pregnancy, but increases during later pregnancy (from 5.7% to 13.3% in the second trimester), which is contrary to all of the other parasomnias (15). Sleep paralysis does not seem to affect the pregnancy.

Anesthesia

Anesthesia does not seem to increase or reduce the frequency of sleep paralysis and is not contraindicated for patients with sleep paralysis. However, increased frequency of sleep paralysis after general anesthesia has been reported in narcolepsy, possibly related to drug withdrawal around surgery (18).