Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2022 Nov 15.
Published in final edited form as: J Neurol Sci. 2021 Sep 20;430:119997. doi: 10.1016/j.jns.2021.119997

Genetic and Demographic Predisposing Factors Associated with Pediatric Sleepwalking in the Philadelphia Neurodevelopmental Cohort

Yuhei Chiba a,d,e,h, Owen R Phillips b,f, Shinichi Takenoshita a,g, Hanna M Ollila c, Joachim F Hallmayer b, Seiji Nishino a, Manpreet K Singh b
PMCID: PMC8595666  NIHMSID: NIHMS1741251  PMID: 34563919

Abstract

Objectives:

Sleepwalking is a parasomnia associated with non-rapid eye movement (NREM) sleep and is formally diagnosed using polysomnography (PSG). However, PSG are difficult to perform on children or adolescents due to needed compliance. To understand this condition in youth, few studies have been conducted on a large cohort of youths with a diverse distribution of ages and races to characterize it better in the absence of PSG. The present study aimed to evaluate the prevalence of sleepwalking in youth, as well as associated demographic and genetic characteristics, using questionnaires in a large pediatric cohort.

Methods:

Data from the Philadelphia Neurodevelopmental Cohort (PNC) of 7515 youths aged between 8 and 22 years were used in analyses. Demographic and clinical data, including age, sex, and race, and genetic data from 2753 African American (AA) and 4762 European American (EA) subjects were investigated. The age-wise prevalence of sleepwalking in AA and EA subjects was evaluated. Finally, race-specific genomewide association (GWAS) analyses of sleepwalking were also performed (N=155 AA cases and 2598 AA controls; N=512 EA cases and 4250 EA controls).

Results:

Lifetime history of sleepwalking correlated with male sex and EA race. A genetic risk locus that reached genome-wide significance was detected at rs73450744 on chromosome 18 in AA, but not EA youth.

Conclusion:

The present results suggest that male sex, EA race, and genetic factors may be associated with higher rates of sleepwalking among youth. Future studies should consider these variables to advance understanding of the complex pathogenesis of sleepwalking.

Keywords: disorders of arousal, Sleepwalking, pediatric parasomnia, race, GWAS

1. INTRODUCTION

Sleepwalking is a type of parasomnia observed in children,[1,2] and is characterized by complex behaviors that are generally initiated during partial arousals from slow wave sleep. The representative episode is comprised of a range of behaviors, such a toddler sitting up and crawling around the bed, walking quietly, or running around the house. The sleepwalking individual is disoriented in time and space, with slow speech, severely diminished mentation, and blunted responses to questions or requests.[3] Sleepwalking, sleep terrors, and confusional arousal are classified as disorders of arousal (DOA) in the International Classification of Sleep Disorders third edition (ICDS-3), and are considered to be provoked by the dissociation of non-rapid eye movement (NREM) sleep into wakefulness. Sleepwalking is diagnosed based on clinical criteria, such as recurrent episodes of incomplete awakening from sleep, inappropriate or absent responsiveness to the efforts of others to intervene or redirect the person during the episode, limited or no associated cognition or dream imagery, and partial or complete amnesia for the episode, and arousals are associated with ambulation and other complex behaviors out of bed.[3] Importantly, sleepwalking may be associated with a risk of injury and may be life threatening due to the unconscious performance of dangerous behaviors.[4] However, most children with sleepwalking do not require treatment because it naturally resolves with age, which either suggest that the mechanisms underlying sleepwalking likely correspond to brain plasticity or represent transient deviations in sleep regulation in otherwise typically developing youth.[1,3–5]

A pathophysiological model of DOA has broadly been described using the 3-P model, in which DOAs are attributed to the co-occurrence of various predisposing, priming, and precipitating factors.[6,7] Although priming and precipitating factors can be identified by sleep deprivation, various substances including Z-drugs, comorbid conditions, or arousing external stimuli, an assessment of predisposing factors requires evaluation in large cohorts. Moreover, predisposing factors precede the emergence of sleepwalking and may be related to genetic factors that have yet to be confirmed. Indeed, sleepwalking runs in families and manifests in the children of parents who sleepwalk.[8–10] A twin study reported a significantly higher concordance of sleepwalking in monozygotic pairs than in dizygotic pairs.[11,12] In addition to essential and detailed interviewing of children and their parents, polysomnography (PSG) is performed for comprehensive examination of sleep comorbidities (e.g. sleep apneas) and to explore various differential diagnoses (e.g. sleep epilepsy). However, PSG is not practical to conduct in large pediatric cohorts due to requisite hospitalization. Further, since sleepwalking does not always occur during a PSG, even an overnight PSG may not be sufficient to capture sleepwalking events, complicated further by poor compliance in youth. Consequently, few studies have been conducted to characterize factors predisposing youth to sleepwalk in a large enough cohort with diverse distributions of ages and races to be able to generalize toward a population based understanding of those predispositions.[13]

Data from the Philadelphia Neurodevelopmental Cohort (PNC) is publicly accessible from the National Institutes of Health database of Genotypes and Phenotypes (dbGap).[14,15] It includes a large sample of nearly 10,000 African (AA) and European American (EA) youths evaluated in the Children’s Hospital of Philadelphia (CHOP) network. This cohort provides multi-level information on youth ranging from demographic, to behavioral, and genetic. Although the PNC was not designed to specifically study sleep, the database includes a questionnaire that reports on the presence or absence of sleepwalking among youth to permit an exploratory investigation of demographics and genetic factors associated with the presence or absence of sleepwalking in a large cohort of youth. Thus, this study aimed to examine predisposing factors, namely, demographic and genetic factors, for sleepwalking based on data derived from the PNC database using statistical and genome-wide association (GWAS) analyses. This characterization may provide initial insights into the pathogenesis and treatment of sleepwalking in youths.

2. METHODS

2.1. Sample

Data in the PNC database were collected from the CHOP healthcare network. Subjects were not specifically recruited from psychiatric or sleep specialty clinics. Inclusion criteria were youths that were ambulatory, in stable health, proficient in English, and able to complete the study procedure, including neurocognitive, genetic, and clinical assessments. Youths with severe developmental delays, significant hearing loss, or limited mobility were excluded. Our requests for the research projects were approved and listed in the requester name of Yuhei Chiba and Manpreet Sigh in https://www.ncbi.nlm.nih.gov/projects/gap/cgi-bin/study.cgi?study_id=phs000607.v3.p2&phv.

2.2. Data assessment

Data analyses of 8719 subjects (ages: 8-22 years) from the PNC phs000607.v1.p1 were conducted. Demographic factors and medical histories were assessed using a structured computerized instrument, namely the GOASSESS.[16,17] We investigated the age, sex, and race as self-reported by AA and EA subjects, excluding 1061 subjects of mixed/other races to simplify the GWAS analysis. GOASSESS is designed to collect information on the past and present history of medical condition including sleepwalking, and was administered to subjects over 11 years of age, as well as their parents if subjects were between the ages of 8 and 17 years. PNC youth or their parents were asked “Have you (your child) ever had sleepwalking?”. More detailed questions were not asked on sleepwalking. Responses were coded as yes, no, or unsure. If the answer was yes, the follow-up question “Is it a current condition?” was added. Further information was not obtained on the “current condition”. After the exclusion of 143 subjects who responded “unsure” to the first question, the remaining 7515 subjects who were AA (2753 cases) and EA (4762 cases) were examined. Subjects who responded “yes” to the first question by themselves or via their parents were defined as those with a history of sleepwalking. Subjects who responded “no” by themselves or via their parents were defined as subjects without a history of sleepwalking.

2.3. Analysis of demographic factors associated with sleepwalking

In the analysis of all subjects, we compared the prevalence of sex, age, and EA using the Student’s t-test and chi-squared test, and then calculated the odds ratio using a univariate regression analysis between subjects with and without a history of sleepwalking.

The prevalence of sex and age in AA and EA youths were compared using the Student’s t-test and chi-squared test, and then the odds ratio between subjects with and without a history of sleepwalking was calculated using a univariate regression analysis. The lifetime and current prevalence of sleepwalking were compared between AA and EA youths.

The current age-wise prevalence of sleepwalking was plotted against race. The age-wise prevalence of current sleepwalking was compared between AA and EA subjects using Fisher’s exact test and False Discovery Rate correction of Benjamini & Hochberg method. Statistical analyses were performed using SPSS v 20.0 (IBM Corp., Armonk, NY, USA). The level of significance was set to p < 0.05.

2.4. Genetic factors of sleepwalking: A GWAS analysis

All blood samples were genotyped on one out of four Illumina arrays: HumanHap550v3, HumanHap550v1, HumanHap610, and Omni Array. Genotype data were imputed using the IMPUTE2 package[18] and reference haplotypes in Phase III of the 1000 genomes data (October 2015 release)[19] included approximately 88 million variants from 2,504 individuals from Africa, Asia, Europe, and the Americas. Excessive relatives with a pairwise proportion of identity by the descent estimate of PI_HAT > 0.25 were excluded. We conducted a principal component analysis (PCA) in PLINK1.9,[20] and calculated eigenvectors. A quality control process was performed to eliminate individuals with >2% missing data and SNPs with >1% missing data, minor allele frequencies (MAF) <0.01, and Hardy-Weinberg Equilibrium p values <10−6.

In the SNP-based association analysis, a logistic regression with the covariations of sex, age, and the first four eigenvectors calculated by the PCA was applied using an additive effect model between AA and EA subjects with and without a history of sleepwalking. A meta-analysis to combine p values from the GWAS analysis of AA and EA subjects was performed using a fixed-effect model with the --meta command of PLINK1.9. A QQ plot and lambda were calculated for GWAS and the meta-analysis using the R qqman[21] and GenABEL[22] packages. Genome-wide significance was set to P < 5×108. Regional association plots were generated using LocusZoom.[23] (http://locuszoom.org/) by summary statistics from GWAS and the meta-analysis. We selected the “1000 Genomes Nov 2014 AMR” panel for the background LD structure.

3. RESULTS

3.1. Investigation of demographic factors

The results obtained on demographic factors are summarized in Table 1.

Table 1.

Demographic factors of all, African, and European subjects with and without a history of sleepwalking.

All subjects (N=7515) African subjects (N=2753) European subjects (N=4762)
SW(+) SW(−) odds SW(+) SW(−) odds SW(+) SW(−) odds
Sleepwalking history (Number) 667 6848 155 2598 512 4250
Male 55.2% 48.1% 1.30 56.1% 45.3% 1.56 54.9% 49.8% 1.23
Age (mean ± standard deviation) 13.8±3.7 13.8±3.7 13.3±3.7 13.9±3.7 14.0±3.7 13.8±3.6
Race (European American) 76.8% 62.1% 1.76
Lifetime frequency 8.88% 5.63%* 10.75%
Current number 199 44 155
Current prevalence 2.64% 1.60%* 3.25%
Type of clinical interview
Adult proband (18y.o.~) 21.7% 20.4% 19.4% 22.3% 22.5% 19.3%
Middle proband (11 ~ 17 y.o.) 54.4% 55.2% 49.7% 54.0% 55.9% 55.9%
Young proband parents (8~10 y.o.) 23.8% 24.4% 31.0% 23.7% 21.7% 24.9%
Open in a new tab

The odds ratios of the items are shown when they correlated (p<0.05) with sleepwalking in a multivariate logistic regression analysis.

SW; sleepwalking, y.o.; years old

*

; Significantly different prevalence between African and European American subjects.

3.1.1. Prevalence of sleepwalking by race

The numbers of AA and EA subjects with sleepwalking during their lifetime were 155 (lifetime prevalence: 5.6%) and 512 (lifetime prevalence: 10.8%), respectively; therefore, 667 subjects (lifetime prevalence: 8.9%) had a history of sleepwalking. The numbers of AA and EA subjects currently sleepwalking were 44 (current prevalence: 1.6%) and 155 (current prevalence: 3.3%), respectively; therefore, 199 subjects were currently sleepwalking (current prevalence: 2.6%). The numbers of AA and EA subjects without a history of sleepwalking during their lifetime were 2598 and 4250, respectively, with a total of 6848.

3.1.2. Odds ratio of sleepwalking with demographic factors.

A history of sleepwalking correlated with male sex (odds ratio, 1.30) and EA race (odds ratio, 1.76). In AA and EA subjects, a history of sleepwalking correlated with male sex (odds ratios of 1.56 in AA and 1.23 in EA, respectively). The lifetime prevalence of sleepwalking was significantly higher in EA subjects than in AA subjects (10.8 and 5.6%, respectively, P<0.001 by the chi-squared test). The current prevalence of sleepwalking was significantly higher in EA subjects than in AA subjects (3.3 and 1.6%, respectively, P<0.001 by the chi-squared test).

3.1.3. The current prevalence of sleepwalking by age.

The current prevalence of sleepwalking was plotted against age in Figure 1 and detailed in Table 2. The current prevalence of sleepwalking in all subjects included in the analysis was the highest at approximately 4.5% in eight-year-old subjects, and gradually decreased with age; however, a transient increase was observed at approximately 18 years old. No significant differences at any ages by Fisher exact test were observed in the age-wise current prevalence of sleepwalking between AA and EA subjects, except at the age of 18 years old, with the current prevalence of sleepwalking being significantly higher in EA subjects than in AA subjects. But this significance resulted in lost by FDR correction using the Benjamini & Hochberg method. The characteristics of a bimodal distribution were apparent in EA subjects

Fig. 1.

Fig. 1.

Open in a new tab

An age-specific plot of the current prevalence of sleepwalking.

*; Significantly different prevalence by Fisher’s exact test between African and European American subjects.

Table 2.

Summary of the age-related current prevalence of sleepwalking.

Age All subjects African European P-value
8 4.52% 3.49% 5.06% 0.434
9 3.35% 1.90% 4.20% 0.226
10 3.33% 1.84% 4.27% 0.157
11 3.06% 1.50% 4.20% 0.062
12 3.62% 2.39% 4.02% 0.644
13 3.72% 2.42% 4.13% 0.644
14 2.28% 2.11% 2.39% 1.000
15 1.27% 0.00% 1.89% 0.058
16 1.23% 1.29% 1.19% 1.000
17 1.88% 0.48% 2.65% 0.108
18 2.56% 0.72% 3.77% 0.013 *
19 1.45% 1.06% 1.71% 0.710
20 2.02% 0.99% 2.74% 0.651
21-22 0.89% 0.00% 1.39% 1.000
Open in a new tab
*

; p value is < 0.05 by Fisher exact test, but is not significant when adjusted for multiple comparisons.

3.2. GWAS analysis

Due to the small sample size for analysis, we performed a GWAS analysis of AA and EA subjects combined with and without a history of sleepwalking. Figure 2a and b show a Manhattan plot and QQ plot of the GWAS analysis of AA and EA subjects with the covariates of age, sex, and the top 4 eigenvectors of PCA. The lambda values of AA and EA analyses were 0.971 and 1.013, respectively. Figure 2c shows a Manhattan plot and QQ plot of the meta-analysis of GWAS of AA and EA subjects. The lambda value of the meta-analysis was 1.009. Table 3 summarizes representative SNPs related to sleepwalking based on the results of the GWAS analysis and meta-analysis. A genetic locus that reached Genome-wide significance at P< 5×10−8 was detected at rs73450744 on chromosome 18 in the AA analysis. A locus zoom plot showed that rs73450744 was in a non-coding region and the nearest gene was Keratoconus Gene 6 (KC6). (Figure 3) Since EA subjects did not have this SNP, it showed no significance in the meta-analysis. Although rs1201967 and rs12197808 exhibited the lowest p values in the meta-analysis, these values did not reach Genome-wide significance.

Fig. 2.

Fig. 2.

Open in a new tab

Manhattan plots of sleepwalking in African and European Americans. The SNP, rs73450744, on chromosome 18 in African American subjects reached genome-wide significance at p < 5 × 10 8.

Table 3.

Summary of GWAS results

Results of GWAS
CHR SNP BP A1/A2 MAF in EA MAF in AA Case Control OR P Fq_case Fq_control Nearest Gene
African American
3 rs113999746 21220365 A/G 0.0001853 0.02285 109 1710 4.458 7.50E-07 0.07339 0.01849 VENTXP7
6 rs183412314 20663830 T/C 0.0002806 0.01127 109 1709 6.495 5.76E-07 0.05093 0.008269 CDKAL1
7 rs493724 29007667 G/C 0.00009259 0.02974 109 1710 3.641 9.18E-07 0.08716 0.02392 TRIL
9 rs113394219 115184393 A/G 0.0008329 0.07278 109 1710 2.822 3.63E-07 0.1667 0.0684 HSDL2
18 rs73450744 39150106 T/C 0 0.01529 109 1710 5.765 3.67E-08 0.06881 0.01267 KC6
European American
6 rs387722 92340664 C/T 0.3708 0.4144 472 4001 0.6795 3.88E-07 0.2966 0.3778 CASC6
7 rs157658 121798998 T/T 0.4534 0.3422 472 4001 0.679 6.21E-08 0.3715 0.4641 AASS
Meta-analysis
CHR SNP BP A1/A2 MAF in EA MAF in AA Case Control OR P Fq_case Fq_control Q I Nearest Gene
6 rs1201867 92340802 T/G 0.3689 0.224 581 5710 0.7222 3.03E-06 0.2814 0.3306 0.0734 68.79 CASC6
6 rs12197808 125201579 A/G 0.05128 0.1774 581 5710 1.6472 2.27E-06 0.1071 0.08376 0.1285 56.72 STL
Open in a new tab

CHR; Chromosome, SNP; single nucleotide polymorphism, BP; base pair, A1; A1 allele, A2; A2 allele, MAF; minor allele frequency, OR; odds ratio, P; P-value, Fq; frequency, AA; African American, EA; European American, Q; p-value for Cochrane’s Q statistic, I; I2 heterogeneity index (0-100)

Fig. 3.

Fig. 3.

Open in a new tab

Locus zoom plot of GWAS of rs73450744 in African American subjects. The nearest gene of rs73450744 is KC6.

4. DISCUSSION

We investigated the prevalence of sleepwalking and predisposing demographic and genetic factors in a cohort of 8-22 year old youth. Our results showed that the prevalence of sleepwalking is higher in EA subjects than in AA subjects in the PNC database, and that key SNPs were associated with sleepwalking in AA but not EA youth. The PNC study very broadly queried whether sleepwalking was present or absent in a subject based on their response to a structured computerized instrument, and subjects and their parents received no additional information on the definition of sleepwalking as a sleep disorder, its duration, severity, or frequency, and which informant (parent/child/both) provided information about sleepwalking. Subjects are usually not aware of their episodes unless episodes are witnessed and reported by others. Stallman et al., [24] performed a self-report investigation on sleepwalking, which showed that the 17.5% of 532 Australian adolescents, mean age = 17.0 years old, were unsure if they had sleepwalking in the previous month. The existence of current sleepwalking might be biased by the fact that some participants might experience recall bias or live alone, limiting the ability of others to witness or observe sleepwalking episodes. Reports on the detailed episodes of sleepwalking, sleep habits, socioeconomic status, sleep parameters, drug intake, and family history of sleepwalking are important to contextualize and confirm our findings. Our study provides a preliminary report on the key demographic and genetic predispositions that might merit consideration.

4.1. The prevalence of sleepwalking by race.

The lifetime prevalence of sleepwalking was approximately 9% in the PNC, with a current prevalence of sleepwalking of 2.64% across all youths examined in this cohort. Stallman and Kohler conducted a meta-analysis of the prevalence of sleepwalking,[13] and the findings obtained showed that the estimated lifetime prevalence of sleepwalking was 6.9%, while the current prevalence of sleepwalking (within 12 months) was 5.0% in children and adolescents younger than 18 years old. They noted that the prevalence of sleepwalking was affected by the definition of each sleepwalking-focused study and the methods used to assess a history of sleepwalking, such as self- or parental reports, the frequency of episodes, recall periods, types of behavior, and the presence of impairment or distress. Our results also showed that lifetime and current prevalence of sleepwalking were significantly higher in EA subjects (10.75 and 3.25%, respectively) than in AA subjects (5.63 and 1.60%, respectively), suggesting that there may be race differences in the prevelance rates of sleepwalking across race. To our knowledge, our analyses are the first to characterize sleepwalking in AA children and adolescents. Indeed, our findings among youth mimic those previously reported in adults [25] in which AA adults had a lower odds ratio of sleepwalking than EA adults. Others [26] have also shown that the lifetime prevalence of sleepwalking is roughtly 4.3% in young Nigerian adults aged 19-35 years old, which was similar to the rates observed in AA youths in the PNC. In our univariate logistic regressions, EA race determined higher odds of sleepwalking compared to AA race. Guglielmo et al.[27] elaborates on sleep characteristics in US school-aged children and adolescents, reporting more frequent sleep/wake issues and earlier despite longer sleep duration in EA compared to AA youths. These sleep characteristics were independent of the socioeconomic status, such as parent education, income, or poverty. Finally, Rao et al.[28] has reported differences in the electroencephalographic sleep patterns of adolescents by race. Specifically, AA youth showed lower sleep efficiency, spent more time in stage 2 sleep, and had less slow wave sleep compared to EA youths. Although the socioeconomic status of the PNC was not reported nor was a PSG performed, American youth of EA descent may have sleep habits that are related to deeper sleep and sleep fragmentation, resulting in an increased risk of sleepwalking compared to AA youth.

4.2.1. Sex differences

The present study showed a history of sleepwalking significantly correlated with male sex. There are some mixed reports about sex differences in sleepwalking. Archbold et al.[29] collected data from a pediatric sleep questionnaire from the parents of 1038 children (58.4% European American, 19.4% African American) aged 2.0 to 13.9. Boys showed significantly higher rate of lifetime sleepwalking (boys of about 17% and girls of about 12%), which is consistent with our results. In contrast, Petit et al., [10] performed questionnaires in mothers of 1940 children aged under 13 in Canada, and sex was not associated with the occurrence of sleepwalking during childhood. In addition, Laberge et al.,[30] provided data from mothers of 1333 children aged under 13 in French, and sex was not associated with the occurrence of sleepwalking during childhood. Considering brain maturation by sex differences, prior studies have reported that girls display a higher overall proportion of slow wave sleep during the night than boys; however, the amount of stage 4 sleep decreases during development with a steeper slope in boys than in girls.[31,32] These differences in the sleep architecture between boys and girls and over the course of development may explain sex differences in the prevalence of sleepwalking.

The PNC cohort contains information from many clinical evaluations, including the dimensionally granular phenotyping of cognitive and neurodevelopmental characteristics, which may be the focus of future studies to elucidate sex differences between sleepwalking, neurodevelopment, and sleep maturation.

4.3. Age-wise prevalence of sleepwalking

As expected, the prevalence of sleepwalking in EA and AA subjects generally decreased with age, likely reflecting a mechanism by which youth outgrow this problem over the course of development. In our study, current prevalence of sleepwalking in EA subjects showed a bimodal distribution, with a continuous decrease from 8 years old followed by a transient increase at approximately 18 years old. However, the age-wise differences of current prevalence of sleepwalking between EA and AA were not significant. There are few previous reports on age wise prevalence of sleepwalking. Klackenberg and colleagues,[33] investigated the age-wise current prevalence of sleepwalking in subjects aged between 6 and 16 years of age in Sweden. The prevalence of “sometimes or often sleepwalking” was the highest at eight years old and then decreased with age, but slightly increased at 15 and 16 years old, which looked consistent with our result of EA youth in the PNC. Fisher and Wilson[34] also reported the age wise prevalence of current sleepwalking of youth under 20 years old in Canada, which demonstrated that the current prevalence decreased by age and transient slight increase at 17 and 18 years old, consistent with our result. Longitudinal relations between brain maturation and sleep architecture change may impact the pathophysiology of sleepwalking[31,32].

4.4. Genetic predisposition

Our GWAS analysis identified one locus on chromosome 18 that predicted sleepwalking. These results need to be replicated in a larger sample with controls because the number of cases with lifetime sleepwalking was small for a GWAS analysis (581 EA and 109 AA youths with sleepwalking). However, to our knowledge, our pilot analysis is the first to use a GWAS analysis of sleepwalking, and rs73450744 on chromosome18 reached genome-wide significance in AA youths with versus without a history of sleepwalking. Since this SNP was not present in EA youths, it was not possible to analyze the data using a meta-analytic approach.

The SNP rs73450744 is in a non-coding region and the nearest gene is KC6. Rabinobitz et al.[35] reported that the mRNA of KC6 increased in corneas with keratoconus, which is a non-inflammatory corneal thinning disorder affecting younger individuals. Eye rubbing is considered to be an etiology of keratoconus. However, the pathophysiology of keratoconus and the function or contribution of KC6 to keratoconus remain unknown. Waisberg et al.[36] reported a case of keratoconus that presented with sleepwalking. In this case, keratoconus was associated with eye rubbing and eye pressing during sleepwalking. Therefore, keratoconus appears to be a consequence of sleepwalking; however, the causal relation is unclear. Patients with keratoconus are also reported to be at an increased risk of developing obstructive sleep apnea,[37,38] suggesting another potential association between keratoconus and sleepwalking. keratoconus may represent a sleep fragmentation factor, which is a priming factor, according the 3-P model. Further investigations on the relations among keratoconus, sleepwalking, and obstructive sleep apnea are needed.

No SNPs reached genome-wide significance in our overall meta-analysis. Human Leukocyte Antigen (HLA) DQB1 allele was reported the risk for sleepwalking.[39,40] Unfortunately our typical GWAS method cannot determine the HLA type. Licis et al.,[41] investigated the inheritance of sleepwalking in 4 generations of a single pedigree, and identified the genetic locus for sleepwalking on chromosome 20q12-13.12. The locus of KC6 and other candidate loci explored in our study are not located on 20q12-13.12. These variations in findings may be attributed to population differences or study inclusion/exclusion criteria, such as the exclusion of excessive relatedness in the PNC compared to pedigree studies. Nevertheless, the presence of sleepwalking in family pedigrees underscore the need to consider genetic contributions of this comlex phenotype, which like other neurodevelopmental conditions, may be more polygenic and multifactorial rather than monogenic or follow standard Mendelian patterns.

Although the results of the present GWAS analysis showed that KC6 reached genome-wide significance and had a relatively high odds ratio in AA youth, this variant is a low frequency variant that needs to be interpreted with caution. Low frequency variants will have a small effect at a population level. However, individuals carrying this risk variant may exhibit the sleepwalking phenotype. Unfortunately, no family histories are included in the PNC. Future studies are needed to investigate whether the expression of this gene increases or decreases in individuals currently sleepwalking combined with more information on family medical histories.

4.5. Limitations

The PNC is a large and diverse database of youth. We attempted to address challenges related to variance in race by only including the two largest race subgroups represented in the database (namely, AA and EA) in our analysis, while maintaining the value of diverse representation among participants. Although the overall cohort was large, the relatively smaller sample size for the GWAS analysis within race subgroups was a limiting factor, and may contribute to null findings (type I error) in EA youth and in the meta-analysis. The AA sample was even smaller, which raises a concern that the significant SNP may be the result of a type II error, and significant SNPs in small GWAS are very hard to replicate. Future studies including more subjects or combinations of cohort studies will be beneficial to confirm our preliminary observations of the effects of the genome on sleepwalking. Ultimately, this research merits replication in an even larger dataset aimed to evaluate multi-level sleep characteristics among a diverse group of children and adolescents.

5. CONCLUSION

Sleepwalking is prevalent among youth but tends to decrease in frequency with age. Further, sleepwalking may be associated with key predisposing demographic factors, including male sex and EA race. A significant genetic risk locus was identified at rs73450744 on chromosome 18 in exploratory analyses of AA youths. The regulation of sleep or sleepwalking behaviors themselves in youths may be altered by multiple factors. Although the pathogenesis of sleepwalking may be complex, our preliminary analysis points to potentially important demographic variables that may need to be considered in sleepwalking children. Additional studies examining the neurobiological etiology of sleepwalking will aid in translating toward novel and personalized treatments.

Highlight.

We explored the demographic and genetic characteristics of sleepwalking in 7515 youth.

Sleepwalking is prevalent among youth but tends to decrease in frequency with age.

Male sex was associated with higher rates of sleepwalking among youth.

European American race was associated with higher rates of sleepwalking among youth.

A risk locus was identified in exploratory analyses of African American youths.

6. FUNDING AND DISCLOSURES

The present study was supported by National Institute of Mental Health and the Office of Research in Woment’s Health grant R56 MH107243 to Dr. Singh. Dr. Chiba receives the support of a Postdoctoral fellowship from the Uehara Memorial Foundation. Dr. Phillips is a founder of Brain Key AI and a consultant for Ono Pharmaceuticals. Hanna M. Ollila receives research support from the Academy of Finland (grant #309643). Dr. Singh has received research support from Stanford’s Maternal Child Health Research Institute and Department of Psychiatry, the National Institute of Mental Health, the National Institute of Aging, Johnson and Johnson, Allergan, Patient Centered Outcomes Research Institute, and the Brain and Behavior Research Foundation. She is on the advisory board for Sunovion and Skyland Trail, has been a consultant for X, moonshot factory, Alphabet, Inc., and Limbix, and receives royalties from American Psychiatric Association Press and Thrive Global. No other authors report any other conflicts of interest.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  • [1].Howell MJ, Parasomnias: An Updated Review, Neurotherapeutics. 9 (2012) 753–775. doi: 10.1007/s13311-012-0143-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [2].Petit D, Touchette E, Tremblay RE, Boivin M, Montplaisir J, Dyssomnias and parasomnias in early childhood, Pediatrics. 119 (2007) e1016–1025. doi: 10.1542/peds.2006-2132. [DOI] [PubMed] [Google Scholar]
  • [3].International Classification of Sleep Disorders, American Academy of Sleep Medicine, 2014. [Google Scholar]
  • [4].Sauter TC, Veerakatty S, Haider DG, Geiser T, Ricklin ME, Exadaktylos AK, Somnambulism: Emergency Department Admissions Due to Sleepwalking-Related Trauma, West J Emerg Med. 17 (2016) 709–712. doi: 10.5811/westjem.2016.8.31123. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [5].Ekambaram V, Maski K, Non-Rapid Eye Movement Arousal Parasomnias in Children, Pediatr Ann. 46 (2017) e327–e331. doi: 10.3928/19382359-20170814-01. [DOI] [PubMed] [Google Scholar]
  • [6].Castelnovo A, Lopez R, Proserpio P, Nobili L, Dauvilliers Y, NREM sleep parasomnias as disorders of sleep-state dissociation, Nature Reviews Neurology. 14 (2018) 470–481. doi: 10.1038/s41582-018-0030-y. [DOI] [PubMed] [Google Scholar]
  • [7].Pressman MR, Factors that predispose, prime and precipitate NREM parasomnias in adults: Clinical and forensic implications, Sleep Medicine Reviews. 11 (2007) 5–30. doi: 10.1016/j.smrv.2006.06.003. [DOI] [PubMed] [Google Scholar]
  • [8].Kales A, Soldatos CR, Bixler EO, Ladda RL, Charney DS, Weber G, Schweitzer PK, Hereditary factors in sleepwalking and night terrors, Br J Psychiatry. 137 (1980) 111–118. [DOI] [PubMed] [Google Scholar]
  • [9].Lopez R, Shen Y, Chenini S, Rassu AL, Evangelista E, Barateau L, Jaussent I, Dauvilliers Y, Diagnostic criteria for disorders of arousal: A video-polysomnographic assessment, Annals of Neurology. 83 (2018) 341–351. doi: 10.1002/ana.25153. [DOI] [PubMed] [Google Scholar]
  • [10].Petit D, Pennestri M-H, Paquet J, Desautels A, Zadra A, Vitaro F, Tremblay RE, Boivin M, Montplaisir J, Childhood Sleepwalking and Sleep Terrors: A Longitudinal Study of Prevalence and Familial Aggregation, JAMA Pediatr. 169 (2015) 653–658. doi: 10.1001/jamapediatrics.2015.127. [DOI] [PubMed] [Google Scholar]
  • [11].Malkoff DonaldB., Mick BrooksA., SLEEP-WALKING IN TWINS, The Lancet. 296 (1970) 664. doi: 10.1016/S0140-6736(70)91434-0. [DOI] [PubMed] [Google Scholar]
  • [12].Hublin C, Kaprio J, Partinen M, Heikkila K, Koskenvuo M, Prevalence and Genetics of Sleepwalking: A Population-based Twin Study, Neurology. 48 (1997) 177–181. doi: 10.1212/WNL.48.1.177. [DOI] [PubMed] [Google Scholar]
  • [13].Stallman HM, Kohler M, Prevalence of Sleepwalking: A Systematic Review and Meta-Analysis, PLoS ONE. 11 (2016) e0164769. doi: 10.1371/journal.pone.0164769. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [14].Robinson EB, Kirby A, Ruparel K, Yang J, McGrath L, Anttila V, Neale BM, Merikangas K, Lehner T, Sleiman PMA, Daly MJ, Gur R, Gur R, Hakonarson H, The genetic architecture of pediatric cognitive abilities in the Philadelphia Neurodevelopmental Cohort, Mol Psychiatry. 20 (2015) 454–458. doi: 10.1038/mp.2014.65. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [15].Satterthwaite TD, Connolly JJ, Ruparel K, Calkins ME, Jackson C, Elliott MA, Roalf DR, Ryan Hopsona KP, Behr M, Qiu H, Mentch FD, Chiavacci R, Sleiman PMA, Gur RC, Hakonarson H, Gur RE, The Philadelphia Neurodevelopmental Cohort: A publicly available resource for the study of normal and abnormal brain development in youth, Neuroimage. 124 (2016) 1115–1119. doi: 10.1016/j.neuroimage.2015.03.056. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [16].Calkins ME, Merikangas KR, Moore TM, Burstein M, Behr MA, Satterthwaite TD, Ruparel K, Wolf DH, Roalf DR, Mentch FD, Qiu H, Chiavacci R, Connolly JJ, Sleiman PMA, Gur RC, Hakonarson H, Gur RE, The Philadelphia Neurodevelopmental Cohort: constructing a deep phenotyping collaborative, J Child Psychol Psychiatry. 56 (2015) 1356–1369. doi: 10.1111/jcpp.12416. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [17].Calkins ME, Moore TM, Merikangas KR, Burstein M, Satterthwaite TD, Bilker WB, Ruparel K, Chiavacci R, Wolf DH, Mentch F, Qiu H, Connolly JJ, Sleiman PA, Hakonarson H, Gur RC, Gur RE, The psychosis spectrum in a young U.S. community sample: findings from the Philadelphia Neurodevelopmental Cohort, World Psychiatry. 13 (2014) 296–305. doi: 10.1002/wps.20152. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [18].Howie BN, Donnelly P, Marchini J, A Flexible and Accurate Genotype Imputation Method for the Next Generation of Genome-Wide Association Studies, PLOS Genetics. 5 (2009) e1000529. doi: 10.1371/journal.pgen.1000529. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [19].1000 Genomes Project Consortium, Auton A, Brooks LD, Durbin RM, Garrison EP, Kang HM, Korbel JO, Marchini JL, McCarthy S, McVean GA, Abecasis GR, A global reference for human genetic variation, Nature. 526 (2015) 68–74. doi: 10.1038/nature15393. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [20].Purcell S, Neale B, Todd-Brown K, Thomas L, Ferreira MAR, Bender D, Maller J, Sklar P, de Bakker PIW, Daly MJ, Sham PC, PLINK: a tool set for whole-genome association and population-based linkage analyses, Am. J. Hum. Genet 81 (2007) 559–575. doi: 10.1086/519795. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [21].Turner SD, qqman: an R package for visualizing GWAS results using Q-Q and manhattan plots, BioRxiv. (2014) 005165. doi: 10.1101/005165. [DOI] [Google Scholar]
  • [22].Aulchenko YS, Ripke S, Isaacs A, van Duijn CM, GenABEL: an R library for genome-wide association analysis, Bioinformatics. 23 (2007) 1294–1296. doi: 10.1093/bioinformatics/btm108. [DOI] [PubMed] [Google Scholar]
  • [23].Pruim RJ, Welch RP, Sanna S, Teslovich TM, Chines PS, Gliedt TP, Boehnke M, Abecasis GR, Willer CJ, LocusZoom: regional visualization of genome-wide association scan results, Bioinformatics. 26 (2010) 2336–2337. doi: 10.1093/bioinformatics/btq419. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [24].Stallman HM, Kohler M, Wilson A, Biggs S, Dollman J, Martin J, Kennedy D, Lushington K, Self-reported sleepwalking in Australian senior secondary school students, Sleep Medicine. 25 (2016) 1–3. doi: 10.1016/j.sleep.2016.06.024. [DOI] [PubMed] [Google Scholar]
  • [25].Ohayon MM, Mahowald MW, Dauvilliers Y, Krystal AD, Léger D, Prevalence and comorbidity of nocturnal wandering in the U.S. adult general population, Neurology. 78 (2012) 1583–1589. doi: 10.1212/WNL.0b013e3182563be5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [26].Oluwole OSA, Lifetime prevalence and incidence of parasomnias in a population of young adult Nigerians, J Neurol. 257 (2010) 1141–1147. doi: 10.1007/s00415-010-5479-6. [DOI] [PubMed] [Google Scholar]
  • [27].Guglielmo D, Gazmararian JA, Chung J, Rogers AE, Hale L, Racial/ethnic sleep disparities in US school-aged children and adolescents: a review of the literature, Sleep Health. 4 (2018) 68–80. doi: 10.1016/j.sleh.2017.09.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [28].Rao U, Hammen CL, Poland RE, Ethnic differences in electroencephalographic sleep patterns in adolescents, Asian J Psychiatr. 2 (2009) 17–24. doi: 10.1016/j.ajp.2008.12.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [29].Archbold KH, Pituch KJ, Panahi P, Chervin RD, Symptoms of sleep disturbances among children at two general pediatric clinics, The Journal of Pediatrics. 140 (2002) 97–102. doi: 10.1067/mpd.2002.119990. [DOI] [PubMed] [Google Scholar]
  • [30].Laberge L, Tremblay RE, Vitaro F, Montplaisir J, Development of Parasomnias From Childhood to Early Adolescence, (n.d.) 10. [DOI] [PubMed] [Google Scholar]
  • [31].Ohayon MM, Carskadon MA, Guilleminault C, Vitiello MV, Meta-analysis of quantitative sleep parameters from childhood to old age in healthy individuals: developing normative sleep values across the human lifespan, Sleep. 27 (2004) 1255–1273. [DOI] [PubMed] [Google Scholar]
  • [32].Acebo C, Millman RP, Rosenberg C, Cavallo A, Carskadon MA, Sleep, breathing, and cephalometrics in older children and young adults. Part I -- Normative values, Chest. 109 (1996) 664–672. [DOI] [PubMed] [Google Scholar]
  • [33].Klackenberg G, Somnambulism in Childhood-Prevalence, Course and Behavioral Correlations, Acta Paediatrica. 71 (1982) 495–499. doi: 10.1111/j.1651-2227.1982.tb09458.x. [DOI] [PubMed] [Google Scholar]
  • [34].Fisher BE, Wilson AE, Selected Sleep Disturbances in School Children Reported by Parents: Prevalence, Interrelationships, Behavioral Correlates and Parental Attributions, Percept Mot Skills. 64 (1987) 1147–1157. doi: 10.2466/pms.1987.64.3c.1147. [DOI] [PubMed] [Google Scholar]
  • [35].Rabinowitz YS, Dong L, Wistow G, Gene expression profile studies of human keratoconus cornea for NEIBank: a novel cornea-expressed gene and the absence of transcripts for aquaporin 5, Invest. Ophthalmol. Vis. Sci 46 (2005) 1239–1246. doi: 10.1167/iovs.04-1148. [DOI] [PubMed] [Google Scholar]
  • [36].Waisberg Y, KERATOCONUS IN A PATIENT WITH SOMNAMBULISM AND EYE-PRESSING. [Letter], Journal. 17 (1991). [PubMed] [Google Scholar]
  • [37].Naderan M, Rezagholizadeh F, Zolfaghari M, Naderan M, Rajabi MT, Kamaleddin MA, Association between the prevalence of obstructive sleep apnoea and the severity of keratoconus, British Journal of Ophthalmology. 99 (2015) 1675–1679. doi: 10.1136/bjophthalmol-2015-306665. [DOI] [PubMed] [Google Scholar]
  • [38].Woodward MA, Blachley TS, Stein JD, The Association Between Sociodemographic Factors, Common Systemic Diseases, and Keratoconus: An Analysis of a Nationwide Healthcare Claims Database, Ophthalmology. 123 (2016) 457–465.e2. doi: 10.1016/j.ophtha.2015.10.035. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [39].Heidbreder A, Frauscher B, Mitterling T, Boentert M, Schirmacher A, Hörtnagl P, Schennach H, Massoth C, Happe S, Mayer G, Young P, Högl B, Not Only Sleepwalking But NREM Parasomnia Irrespective of the Type Is Associated with HLA DQB1*05:01, J Clin Sleep Med. 12 (2016) 565–570. doi: 10.5664/jcsm.5692. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [40].Lecendreux M, Bassetti C, Dauvilliers Y, Mayer G, Neidhart E, Tafti M, HLA and genetic susceptibility to sleepwalking, Molecular Psychiatry. 8 (2003) 114. doi: 10.1038/sj.mp.4001203. [DOI] [PubMed] [Google Scholar]
  • [41].Licis AK, Desruisseau DM, Yamada KA, Duntley SP, Gurnett CA, Novel genetic findings in an extended family pedigree with sleepwalking, Neurology. 76 (2011) 49–52. doi: 10.1212/WNL.0b013e318203e964. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES