Clinical and polysomnographic profile of adult patients with NREM parasomnia – experience from a neurological sleep center

Summary

Objective: To map the anamnestic characteristics and polysomnographic profile of patients with NREM (non rapid eye movement) parasomnia compared to a control group and to determine whether sleep parameters are related to clinical qualitative parameters and comorbidities. Methodology: We included 53 patients (31 men), average age 31.5 ± 8 years, in the retrospective study. We evaluated anamnestic data, including data focused on sleep and sleep disorders, and the results of polysomnographic examination. Polysomnographic findings were compared with a control group of 42 healthy people (23 men), average age 34.1 ± 8 years. Results: The most frequent type of parasomnia evaluated according to the anamnesis was somnambulism (88.6%). The majority of patients reported onset in childhood (86.7%). The most important trigger of episodes was stress (26.4%) and less often alcohol (3.7%), and a foreign environment (9.4%). Among comorbid diseases, allergic diseases were the most frequent (45.2%), followed by psychiatric diseases (13%). During polysomnography, an episode of NREM parasomnia was noted in 50.9% of patients. Compared to the control group, patients with NREM parasomnia had shorter sleep latency, better sleep efficiency, lower proportion of wakefulness, and higher proportion of NREM 2 sleep. A lower proportion of NREM 3 sleep was found in the group of patients with a familial occurrence of parasomnia. Conclusion: The most common comorbid diseases in patients with NREM parasomnia were allergic diseases. Stress was the dominant trigger of parasomnia episodes in NREM parasomnias, so it is important to focus on its management in clinical practice. The association found between NREM 3 sleep loss and familial occurrence supports the current view that NREM 3 dysregulation has a heritable predisposition.

Keywords:

polysomnography – NREM parasomnia – NREM 3 sleep – familial occurrence – allergic disease

This is an unauthorised machine translation into English made using the DeepL Translate Pro translator. The editors do not guarantee that the content of the article corresponds fully to the original language version.

 

Introduction

Parasomnias with impaired awakening from NREM (non rapid eye movement) sleep (NREM parasomnias) are characterized by incomplete awakening from NREM 3 sleep, most commonly in the first third of the night. NREM parasomnias include several clinical manifestations -⁠ waking disorder with confusion, somnambulism, pavor nocturnus, sleep-related eating disorder [1].

Parasomnias disrupt sleep and are associated with a risk of injury to the patient or others. Understanding of the pathophysiological mechanisms of NREM parasomnias has been provided by studies that have captured parasomnias during video-EEG and polysomnographic (PSG) monitoring of patients and during single-photon emission tomography examinations. These studies have shown that patients experience functional dissociation between different brain regions -⁠ activation of motor centers, amygdala and cingulate, while a sleep state persists in the fronto-parietal region and hippocampus [2].

NREM parasomnias often have a familial occurrence. In somnambulism, genetic factors apply in 65% of cases [3]. A number of extrinsic and intrinsic triggers are known to induce parasomnias in dispositional individuals. Other sleep disorders, in particular obstructive sleep apnoea (OSA) or periodic limb movements in sleep (periodic limb movements in sleep; PLMS), may also act as triggers [4]. The findings of abnormalities in sleep macrostructure are heterogeneous across studies. At the level of sleep microstructure, the instability of NREM 3 sleep is evident [5].

The aim of our retrospective study was:

1)
To map the anamnestic characteristics of NREM parasomnias;

2)
To map the PSG profile of NREM parasomnias when compared with a control set;

3)
explore whether sleep parameters have some association with clinical parameters or comorbidities.

 

Methodology

Patients

We included adult patients who were examined at the Centre for Sleep Disorders of the Department of Neurology, 1st Faculty of Medicine, Charles University in Prague with NREM parasomnias between 2010 and 2020. The cohort consisted of 53 patients (31 males) in the age range 18-60 years, mean age was 31.5 ± 8 years. In all patients, anamnestic data were collected, including data targeting sleep and its disturbances, and neurological and video-PSG examinations were performed. PSG examination was indicated because of frequent occurrence of parasomnias, dangerous behavior during the episode, or de novo occurrence in adulthood.

Polysomnographic findings were compared with a control group of 42 healthy subjects (23 males), age range 20-47 years, mean age v was 34.1 ± 8 years.

 

Anamnestic data

Anamnestic data were obtained retrospectively from medical records, which included a detailed structured interview devoted to individual sleep disorders. From the records, we explored the following anamnestic data: Onset in childhood or adulthood, family history of parasomnias, frequency of episodes, nature of triggering factors, presence of dreams after a previous parasomnistic episode, presence of parasomnistic episodes with potential for injury (assault by another person, trauma), presence of restless legs syndrome/Willis-Ekbom disease; RLS/WED), nightmares, hypnagogic hallucinations, insomnia, excessive daytime sleepiness, bruxism, psychiatric and other comorbid conditions.

The Epworth Sleepiness Scale (ESS) was used to quantify the degree of excessive daytime sleepiness [6].

 

Polysomnographic examination

Patients underwent a nocturnal video-PSG examination using the RemLogic digital system, version 3.4.1 (Embla Systems, Broomfield, CO, USA) in standard montage, consisting of electrooculography, EEG (F3-M2, C3-M2, O1-M2, F4-M1, C4-M1, O2-M1), bilateral surface EMG of musculi mentales, musculi tibiales anteriores, electrocardiography, nasal and oral current pressure transducer, thoracic and abdominal respiratory effort transducer, blood oxygen saturation, microphone, and digitally synchronized video recording from 22 : 00 to 6 : 00 pm as recommended by the American Academy of Sleep Medicine (AASM) (Manual for the Scoring of Sleep and Associated Events, version 2.2 2015) [7].

The polysomnographic recording was assessed visually according to the AASM Manual [7].

Apnea-hypopnea index (AHI) ≥ 5 and number of periodic limb movements in sleep (periodic limb movement index; PLMI) > 15 were considered abnormal values [1].

 

Statistical methods

Normality of demographic, clinical, and PSG data was verified by the Shapiro-Wilks test. Two-sided Fisher's exact test was used to compare the difference in qualitative data between groups. Differences in quantitative data were evaluated using the Mann-Whitney test. Spearman's correlation coefficient (ρ -⁠ rho) was used to calculate correlations. A value of ρ > 0.4 and/or p < 0.001 was taken as the significance threshold. To limit the risk of type 1 error when multiple hypotheses were tested in parallel, Bonferroni correction was used with the significance threshold set at p < 0.05.

 

Results

Patient characteristics

The anamnestic and clinical characteristics of the cohort are shown in Table 1.

Somnambulism was the most common type of parasomnia assessed by history (88.6%). Some patients reported a combination of different clinical manifestations. In 6 patients, symptoms of waking with confusion and somnambulism overlapped, in 4 patients, symptoms of somnambulism and pavor nocturnus, and in 3 patients, symptoms of somnambulism and sleep-related eating disorder overlapped. The majority of patients reported an onset in childhood (86.7%). The prevalence was with weekly frequency of episodes (67.9%). Stress was the most frequent trigger of episodes, followed by alcohol and a foreign environment.

The mean ESS score was 6.7 points in patients and 6.4 points in controls. In 10 patients we recorded a score of more than 10 points, corresponding to moderate daytime sleepiness (6); in no patient did we record a score higher than 15 points.

The most common comorbid condition was allergic disease in 24 cases (45.2%), followed by psychiatric illness in 7 patients (13.0%), and 5 patients were diagnosed with cephalea (9.4%).

Three patients were taking long-term antidepressants, two patients clonazepam.

 

Polysomnographic findings

Polysomnographic findings are presented in Table 2.

Comparative analysis of patients with controls showed that patients with NREM parasomnias had shorter sleep latency and better sleep efficiency. There was a statistical trend in the proportion of sleep stages. NREM parasomnias had a lower proportion of wakefulness and a higher proportion of NREM stage 2 sleep. The groups did not differ in the other PSG parameters studied.

An episode of parasomnia was captured (most commonly in the form of waking with confusion) in 27 patients (50.9%), 6 of whom had a positive family history.

Sleep-disordered breathing assessed as AHI ≥ 5 was found in 14 patients (26.4%), only 3 of whom had AHI ≥ 15. In 3 patients, there was an abnormal finding of periodic limb movements in sleep (PLMI > 15).

 

Associations with anamnestic parameters and correlations

When comparing PSG data in relation to qualitative data, there was a statistical trend for a lower proportion of NREM phase 3 sleep in patients with a positive family history of NREM parasomnias in the exploratory phase, compared with patients without a positive family history (p = 0.024). However, none of the results of parameter comparisons as a function of qualitative anamnestic data were confirmed when corrected for multiple comparisons.

In correlation analysis, no associations were found between the clinical and anamnestic parameters we determined.

 

Discussion

Our retrospective study maps clinical and sleep characteristics in adults who were examined by video-PSG for NREM parasomnias because of frequent occurrence of the conditions, their atypical course, dangerous behavior during the course of the conditions, or de novo occurrence in adulthood.

The composition of the cohort is also likely to be influenced by the predominant type of manifestation, which was somnambulimus (88.6%) according to the history. The majority of patients had a persistent childhood presentation (86.7%), which is consistent with literature data [3]. A positive family history was found in 18.8%. Some studies reported familial occurrence in up to 48% of patients. However, a higher familial incidence has been documented in patients in whom symptoms began in childhood or adolescence than in patients in whom symptoms began in adulthood [5,8-11].

From the anamnestic data, we observed a high proportion of subsequent dream recall (56.6%) after the episode. Recently, research has suggested that up to 90% of adult patients with NREM parasomnias can recall dream experiences, which they mostly describe as bad dreams [12]. Compared to REM parasomnias, these dreams are shorter, less bizarre, complex and immersive [12,13] and often correspond to behaviors during the episode [14-16].

In addition to dream sensations tied to the parasomnistic episode, 45.2% of patients described vivid dreams (without explicit frightening content). This figure may indicate differences in dream activity not only in NREM but also in REM sleep. Literature data, regarding the occurrence of dreams other than those tied to parasomnistic experience, are still scarce [17].

From the anamnestic data, we also noted a high proportion of violent behaviour (35.8%) in line with the reason for detailed examination in some patients. Studies show that up to 50% of patients have violent behaviour during episodes, either towards themselves and episodes ending in injury or towards the other person [17,18]. Thus, our results correlate with recent research.

Of those patients who reported a trigger (21 patients), stress was the most common trigger for episodes, followed by alcohol and a foreign environment, which is consistent with the literature [17].

Sleep-disordered breathing and PLMS were considered as triggers of NREM parasomnias [4]. Most studies have not confirmed this finding [1,3,18]. Also in our study, PLMS was observed in only 3 patients (5.7%), OSA in 14 patients (26.4%), but only 3 of them had AHI of 15 or more, so they were more likely to be mild sleep-disordered breathing disorders. We did not observe an episode of NREM parasomnia related to the manifestation of OSA or PLMS in any of the patients.

Allergic diseases were the most frequent comorbid disease (45.2%). An increased incidence of nocturnal scares compared to healthy children has been described in children with bronchial asthma, depending on the lack of compensation of the disease [19]. The co-occurrence of allergic diseases and NREM parasomnias has also been reported in adult patients with NREM parasomnias: the prevalence of asthma was twice that of the general population, and atopy was significantly more prevalent in patients with onset of NREM parasomnias in childhood [8]. Our study also documents a high percentage of allergic diseases in adult patients with NREM parasomnias. We observed that 7 patients with a positive family history of NREM parasomnias also had a positive family history of allergies and atopy, and all patients had childhood onset. This could be suggestive for the same genetic basis of both diseases. So far, no genes have been conclusively identified to be associated with NREM parasomnias [20]. However, some authors have identified the presence of the HLA DQB1*05 : 01 allele in patients of Caucasian ethnicity with parasomnias. The DQB1*05 : 01 allele is also associated with asthma and pollen allergy [8,21,22] and it has been suggested that it could be a genetic marker for the co-occurrence of these diseases [8].

Sleep disruption accompanying allergic disease may be another factor that predisposes to the occurrence of NREM parasomnias [19,23]. Sleep-disordered breathing may be its cause [24,25], but sleep fragmentation has also been described as an independent manifestation of allergic disease [19,23].

Some studies have found an association with psychiatric comorbidity at the onset of NREM parasomnias in adulthood [3,26]. In our study, we observed the presence of a psychiatric comorbidity in 7 patients, which accounted for 13% (depressive disorder in 4, unspecified anxiety spectrum disorder in 2, and obsessive-compulsive disorder in 1). All cases were associated with somnambulism, but only 2 (1 with depressive disorder and 1 with anxiety disorder) had symptom onset in adulthood. Data were obtained by history taking; patients did not undergo a psychiatric examination or complete self-rating scales for depression or anxiety. The overall low number of psychiatric illnesses in our cohort may be related to the lack of focus on psychiatric illness at history taking. However, there are also studies showing that although patients with NREM parasomnias are more likely to suffer from depressive [27] and anxiety disorders [13], there is a large difference between results from clinical and epidemiological studies, with the association appearing very weak in clinical studies [27].

When comparing PSG data between patients and controls, we observed shorter sleep latency and higher sleep efficiency in patients compared to controls, which correlates with the results of some studies [13]; other studies report no significant difference in these parameters [5,28]. There is no clear explanation for the finding of higher sleep efficiency in NREM parasomnias. Carrilo-Solano et al. documented in a cohort of 158 patients a subphenotype of NREM parasomnia characterized by an increased need for sleep (sleep propensity) [13]. It is possible that the presence of this subphenotype is involved in the maintenance of sleep efficiency compared to the healthy population. We can also speculate that patients with NREM parasomnias are adapted to frequent arousals as a result of parasomnistic episodes, so that the phenomenon of maladaptation in the laboratory setting affects them less than it does healthy individuals. Future studies using adaptation nights or home PSG could resolve this question.

NREM parasomnias had a lower proportion of wakefulness and a higher proportion of NREM sleep phases 2. Other research in PSG examinations does not report a significant difference in the percentage of NREM phases between patients and controls [3,4]. One study reported a lower proportion of NREM 1 and NREM 2 sleep phases in patients [5], which was not confirmed by our research. However, the lower proportion of wakefulness and higher proportion of NREM 2 in NREM parasomnias compared to controls could be interpreted as a dependent phenomenon in the finding of short sleep latency and higher sleep efficiency.

When comparing PSG and anamnestic data in the parasomnia group, we found a lower proportion of NREM 3 sleep phase in patients with a positive family history, which could suggest a genetic basis for the dysregulation of NREM 3 sleep phase. The results of NREM 3 quantification at PSG in NREM parasomnias are not consistent in the literature. Most PSG studies have found no changes [4,8,17], others have reported lower NREM 3 or higher percentages of NREM 3 compared to healthy controls [8].

However, the conclusions of these studies agree on changes at the level of sleep microstructure i.e. demonstrate instability of NREM 3 sleep with its fragmentation [5].

 

Limitations of the study

A limitation of our study is that we did not assess the occurrence of arousals and cyclic alternating patterns and thus cannot comment on the microstructure of sleep. Several limitations to the work must be mentioned. We did not assess the microstructure of sleep (see above). Both patients and controls only completed one night of PSG without previous adaptation nights, which may have affected the results compared with the subjects' usual night in the home environment, but we assume that patients and controls are similarly situated in the degree of influence of the new environment. Retrospective data tracking is also a limitation, as is the lack of verification of data on psychiatric comorbidities. The fact that patients' anamnestic data are subject to some subjective bias could also be considered a limitation. However, the nature of the information usually does not allow objectification by measurement and we must therefore rely on the veracity of the statements.

 

Conclusion

Somnambulism was the most common type of parasomnia assessed by history. Most patients gave an onset in childhood, with a predominance of weekly frequency of episodes. The most common comorbid conditions in patients with NREM parasomnias were allergic diseases followed by psychiatric diseases. Stress was the dominant trigger of parasomnias in NREM parasomnias compared to other factors, so it makes sense to focus on its management in clinical practice. A comparative analysis of patients with controls showed that patients with NREM parasomnias had shorter sleep latency and better sleep efficiency. The association found between NREM 3 decline and familial occurrence supports the current view that NREM 3 dysregulation has a heritable predisposition.

 

Financial support

Supported by the National Institute for Neurological Research (EXCELES Program, ID: LX22NPO5107) -⁠ Funded by the European Union -⁠ Next Generation EU; by the Charles University Research Program: Cooperatio Neuroscience; by the General University Hospital in Prague MZ Czech Republic-RVO-VFN64165 project.

 

Conflict of interest

The authors declare that they have no conflict of interest in relation to the subject of the study.

 

Table 1. Qualitative parameters of the cohort and their abundance.

	Number	Prevalence (%)
Phenotype
pavor nocturnus	8	15
somnambulism	47	88,6
awakening with confusion	7	13,2
SRED	3	5,6
Characteristics of NREM parasomnias
onset in childhood	46	86,7
onset in adulthood	7	13,2
positive family history	10	18,8
frequency multiple times per night	5	9,4
weekly frequency (> 1/week)	36	67,9
Monthly frequency (> 1/month, < 1/week)	11	20,7
sporadic frequency (> 1/year, < 1/month)	6	11,3
trigger stress	14	26,4
Alcohol trigger	2	3,7
trigger foreign environment	5	9,4
dream equipment in relation to the state	30	56,6
violent behaviour during the state (assault of another person, accident)	19	35,8
Sleep comorbidities
RLS/WED	3	5,6
vivid dreams	24	45,2
hypnagogic hallucinations	6	11,3
insomnia	7	13,2
hypersomnia	2	3,7
bruxism	5	9,4
Clinical comorbidities
Depression	4	7,5
Anxiety	2	3,7
OCD	1	1,8
Allergic disease	24	45,2
cephalea	5	9,4
head injury	6	11,3
epilepsy	2	3,7
thyroid disorders	5	9,4
Hypertension	5	9,4

OCD -⁠ obsessive-compulsive disorder; OSA -⁠ obstructive sleep apnoea; NREM -⁠ non-rapid eye movement; PLMS -⁠ periodic leg movements in sleep; RLS/WED -⁠ restless legs syndrome; SRED -⁠ sleep related eating disorder

 

 

 

Table 2. Polysomnographic parameters.

	Patients (n = 53)		Controls (n = 42)
	Mean	SD	Mean	SD	p
Age	31,5	8,2	34,1	8,0	0,112
Sex (F/M)	41,5 %/58,5 %		45,2 %/54,8 %
SL (min)	15,0	15,7	23,7	25,8	0,002*
REM latency (min)	118,2	61,0	124,4	75,5	0,994
SEF (%)	85,5	10,3	79,6	11,2	0,003*
TST (min)	402,8	52,0	385,8	66,8	0,284
vigilance (%)	11,3	9,4	16,7	11,1	0,004
NREM 1 (%)	5,8	2,5	6,0	3,0	0,825
NREM 2 (%)	43,3	8,4	38,8	8,3	0,014
NREM 3 (%)	21,8	7,9	21,4	7,4	0,884
REM (%)	17,6	5,7	17,1	6,0	0,539
AHI	3,9	5,6	3,1	4,8	0,940
ODI	3,3	5,4	1,6	2,7	0,101
PLMI	4,2	14,6	3,7	12,1	0,586

 

Bold p-values indicate statistical significance. *Statistical significance supported by Bonferroni correction for 15 independent variables.

AHI, apnea-hypopnea index; M, male; NREM 1, NREM 2, NREM 3, non-REM 1, 2, 3 sleep stages; ODI, oxygen desaturation index; PLMI, periodic leg movements in sleep index; REM -⁠ rapid-eye-movement sleep; SD -⁠ standard deviation; SEF -⁠ sleep efficiency; SL -⁠ sleep latency; TST -⁠ total sleep time; F -⁠ female