Adult-Onset NREM Parasomnia with Hypnopompic Hallucinatory Pain: A Case Report

Abstract

We report the case of a 43-year-old woman presenting with nocturnal episodes of pain and screaming during sleep starting at age 30. There was no childhood or family history of parasomnia. The events had gradually become more frequent over the years, occurring in the first half of the night within 2 h of sleep onset. There were no triggers, and she had partial amnesia for the events. A diagnosis of adult-onset sleep terrors was made on clinical grounds and supported polysomnographically. Seizures and periodic limb movements were excluded as triggering factors. There was some mild sleep disordered breathing (predominantly non-desaturating hypopnea with a propensity for REM sleep of debatable significance). Imaging of the brain and spine and neurophysiological investigations ruled out lesions, entrapments, or neuropathies as possible causes of pain. Treatment (clonazepam, paroxetine, or gabapentin) was poorly tolerated and made no difference to the nocturnal episodes, while trazodone worsened them. This is the first report of hypnopompic psychic pain in association with a NREM parasomnia. We hypothesize that the pain may represent a sensory hallucination analogous to the more commonly recognized visual NREM parasomnia-associated hypnopompic visual hallucinations and that, as such, it may arise during arousal of the sensory neocortex as confabulatory response.

Citation:

Mantoan L; Eriksson SH; Nisbet AP; Walker MC. Adult-onset nrem parasomnia with hypnopompic hallucinatory pain: a case report. SLEEP 2013;36(2):287–290.

INTRODUCTION

NREM parasomnias are arousal disorders occurring in deep slow wave sleep (stage 3/4), which commonly manifest in childhood (with a peak incidence at 11–12 years) as sleep terrors and sleepwalking, but may be prevalent in 1% to 2% of adults.¹ They usually occur in the first half of the night, when slow wave sleep predominates. Frequently there is a family history, and factors increasing or fragmenting deep sleep may trigger episodes. The three most common NREM parasomnias are sleepwalking,² night terrors, and confusional arousal. Night terrors can be associated with a variety of symptoms and manifestations, the most common being screaming and prominent sympathetic activity such as tachycardia and sweating. The patient often recalls a deep sense of fear, often described as if fighting to live. In adults, NREM parasomnias can manifest as complex behavior that can sometimes be violent; most patients have at least partial amnesia for the event.¹ Here we report a case of NREM parasomnia in association with a previously undescribed semiology of hypnopompic psychic pain.

CASE REPORT

A 43-year-old semi-retired deputy head teacher was referred by her general practitioner to a sleep clinic** with a history of nocturnal screaming episodes within 1–2 h of sleep onset from the age of 30 years. Her husband was habitually awoken by his wife screaming loudly, usually flapping either her right or left hand against the bed in a semi-purposeful fashion. Her husband reported that the events were sometimes heralded by an inspiratory sigh, she looked terrified and would not respond to him. The screaming would usually last 5–10 sec, and she would then complain to her husband of intense pain affecting the fingers of either hand or arm and occasionally her legs, with no associated numbness or paraesthesia. She would become fully orientated within 30 sec and would be partially amnesic for the event, but would recall an accompanying sense of “fighting to stay alive” associated with intense panic and often accompanied by fast regular palpitations. Otherwise no dream mentation or visualizations were reported in association with the episodes. During an episode, she would usually remain lying down, but in more severe episodes she would sit up in bed. Occasionally, in a severe episode, she would leave the bed screaming. The frequency of the episodes varied. Initially they occurred monthly but then increased in frequency; at the time of initial assessment they were occurring 2–5 nights per week with 1–2 episodes per night. Since onset, she had had at most 3–4 weeks without symptoms. She was unable to identify any triggers for the episodes, and neither she nor her husband considered her to be stressed, anxious, or depressed. There was no history of sleep violence, sleep sex, sleep eating, or any other NREM parasomniac automatisms. There was no history of childhood parasomnia. Her self-reported sleep latency was about 15–20 min, her husband described her as a restless sleeper but without a history of jerks or kicking. There was no history of sleep paralysis or dream enactment and no history of daytime restless legs. Her husband said that she only occasionally snored and she was easily woken in the morning. She reported some mild neck stiffness and occasional pain in her shoulder blades following a very minor whiplash injury several years previously, but she did not complain of pain in arms or legs during the day, nor of any dysesthesias or motor symptoms. There was no previous history of radicular pain. She reported often feeling tired during the day and that her sleep was frequently unrefreshing. Her sleep hygiene was good. She avoided caffeine and was not on any medication other than hormone replacement therapy. She was a nonsmoker and consumed alcohol only on weekends (12 units over 3 days). Most of her episodes occurred during the week, but alcohol or sleep deprivation did not seem to affect the nocturnal behavior. She was the product of a normal birth with no history of febrile seizures, severe head injuries, meningitis, or encephalitis. Neurological examination was entirely normal. Her Epworth Sleepiness Scale score was 7/24.

Full polysomnography (PSG) showed 5 paroxysmal arousals from stage 3 and 4 sleep, 2 of which were clearly triggered by auditory stimuli (in one case a lorry passing by the window and in another a loud footstep outside her door). One of the other paroxysmal arousals was preceded by a debatable hypopnea (in this case an approximately 40% decrease in flow for 25 sec without any desaturation). The other 2 arousals, however, were spontaneous, and one of them was accompanied by an auditory gasp identified by her husband as characteristic of her episodes prior to the screaming. There was no auditory gasp after the debatable hypopnea described above. Total sleep time was 5 h 47 min, with 69% sleep efficiency. Sleep latency was normal. Sleep patterns showed a low percentage of slow wave sleep, with a manually scored arousal index of 28/h of sleep (American Academy of Sleep Medicine [AASM] criteria), mainly occurring during slow wave and light sleep. REM latency was normal and there was no evidence of movement during REM and she had normal REM atonia. There was some debatable non-desaturating sleep disordered breathing (all hypopneas). The apnea-hypopnea index (AHI) as measured by AASM recommended criteria was 2.1 and 3.8, indicating that there was no significant sleep apnea. There were no significant periodic leg movements or other abnormal movements during sleep.

Additional investigations were performed to rule out a thoracic outlet syndrome, cervical radiculopathy, or focal nerve entrapment. Magnetic resonance imaging (MRI) of the cervical spine revealed minor degenerative changes with slight foraminal narrowing at C4 on the right. Nerve conduction studies ruled out a median neuropathy at the wrist as a common cause for nocturnal upper limb discomfort, and showed very minor chronic denervation changes attributable to an old C6-7 radiculopathy. These results were not thought to be relevant to the patient's nocturnal pain which could occur in either arm or legs.

Initial treatment with a small dose (0.5–1 mg) of clonazepam was not tolerated, as it made her feel very depressed. The low dose tried did not have any effect on her nocturnal symptoms. Trazodone, a sedating antidepressant, was subsequently tried but made the symptoms worse.

In view of the unusual presentation with pain, the frequency of the episodes and the late-onset of the NREM parasomnia, the patient was reassessed with polysomnography at a tertiary referral neurology centre* to exclude seizures. Repeat PSG showed frequent sudden arousals from deep sleep with looks of pain and/or fear on one occasion associated with her typical hand shaking (Figure 1A). The episodes were accompanied by arousal rhythms on the EEG (Figure 1B), consistent with a NREM parasomnia. On the second PSG, the background EEG was normal and there was no evidence of sleep apnea, periodic limb movements, or seizures. No thalamic or hypothalamic lesions were identified on MRI brain, which revealed a Chiari type I malformation which was not thought to be relevant. The patient was unable to tolerate paroxetine due to its sedative effect and was subsequently started on gabapentin, which made no difference and was stopped due to weight gain.

Figure 1.

(A) Hypnogram showing frequent arousals from stage 3 sleep on the second night (D2) of PSG recording; Time in hours: minutes on the x-axis. Sleep stages on y-axis: stage W (wake), stages 1–3 (NREM), and stage R (REM). (B) Electroencephalogram (EEG) showing delta waves of stage 3 sleep before an episode of arousal with shaking of one arm and looks of fear. Key for PSG channels: 24 channels of scalp EEG were recorded, including superficial sphenoidal and mastoid electrodes. Channels 1–12 are EEG, labelled in accordance with the 10–20 international electrode placement system, displayed with parasagittal bipolar channels [1–8] and transverse bipolar channels [9–12]; channel 13 – Right electro-oculogram (EOG); channel 14 - Left EOG (EOG channels referenced to linked mastoid electrodes); channel 15 - submental electromyography (sub emg; bipolar); channel 16 - electrocardiogram (ECG); channel 17 - oxygen saturation by pulse oximetry (SpO₂). Calibration bars for this epoch are of 70 μV/cm for amplitude and 30 sec per page for timeline.

DISCUSSION

This report illustrates a case of NREM parasomnia with the unusual features of late onset (without childhood parasomnia), lack of family history, and a previously undescribed semiology of terror and screaming accompanied by intense limb pain disappearing within seconds of full awakening. Night terrors need to be differentiated from nocturnal panic attacks, which usually occur in late stage 2 or early stage 3 sleep, are associated with a panic disorder and engender no amnesia (none of which occurred in our case). To our knowledge, the association of a NREM parasomnia and hypnopompic limb pain has not been previously reported. Patients with these disorders frequently report single-scene, often frightening visual hallucinations without the narrative contents seen during REM sleep. A sense of having difficulty breathing is also commonly described in NREM parasomnias.³

Escape behavior is often reported in association with these illusions.⁴ The source of the phenomena is not well understood, but because the episodes frequently arise out of slow wave sleep when the cerebral cortex is unlikely to be capable of the complex processing necessary to generate these cognitions and visualizations, the episodes may well be confabulated during arousal in response to a more primitive fight or fright stimulus arising from subcortical brain regions.^5,6 In this case, we postulate that instead of visual cortex activation, the sensory cortex was being activated.

Similar to other descriptions, we found that the NREM parasomnias could be triggered by internal (hypopnea) or external stimuli (noises). It is possible that physical discomfort in the limbs triggered the arousals and acted as a cue for the sensory hallucination, but despite an intensive search we could not find convincing evidence of any ongoing radiculopathy or peripheral nerve entrapment.

Other etiologies of isolated nocturnal limb pain that may warrant exclusion by imaging or neurophysiology in such cases are detailed in Table 1 and include musculoskeletal causes such as the nocturnal bone pain of osteoid osteomas⁷ and joint pain in patients with arthritis⁸; nocturnal positional lumbo ischialgia can be a presenting symptom of spinal tumors.^9,10 Sleep-related leg cramps are a further diagnosis to be considered.

Table 1.

Differential diagnosis of paroxysmal nocturnal limb pain and features supporting or refuting each diagnosis in this case

In conclusion, to our knowledge this is the first report of a NREM parasomnia associated with painful paroxysms, for which we postulate the following underlying pathophysiological mechanism: an internal or external stimulus triggers arousal, facilitating the activation of innate motor pattern generators in the brainstem⁵ and activating somatosensory cortical areas to produce hypnopompic hallucinatory pain.

DISCLOSURE STATEMENT

This was not an industry supported study. Dr. Walker has received remuneration for speaking engagements and educational events sponsored by UCB Pharma, GSK, Viropharma, and Eisai. Dr. Eriksson has received remuneration for speaking engagements and educational events sponsored by UCB Pharma. The other authors have indicated no financial conflicts of interest.

ABBREVIATIONS

AASM

American Academy of Sleep Medicine

AHI

apnea-hypopnea index

EEG

electroencephalogram

MRI

magnetic resonance imaging

PSG

polysomnography