Arousal Dysregulation in Children with Autism Spectrum Disorder

Background: Substantial heterogeneity exists in the degree of impairment, the collection of behavioral symptoms and the medical conditions associated with Autism Spectrum Disorder (ASD). Identifying explicit ASD phenotypes based on a constellation of symptoms may lead to greater understanding of the causal mechanisms of ASD and precise treatments. Arousal dysregulation in children with ASD may produce symptoms including anxiety and insomnia.

Objectives: The purpose of this study was to investigate the relationship between anxiety, biomarkers of arousal dysregulation and sleep parameters in ASD versus typically-developing controls (TDC). We hypothesize that individuals with ASD and anxiety as compared to individuals with ASD without anxiety and TDC will have: (1) greater rate of insomnia, defined as difficulty falling and maintaining sleep with daytime impairment, and a sleep latency of greater than 30 minutes as measured by actigraphy; (2) greater sleep latency, decreased sleep efficiency; (3) higher levels of urine catecholamines; (4) lower Respiratory Sinus Arrhythmia score.

Methods: Setting: Home and center visit at the Center for Autism Research (CAR) at the CHOP. Anxiety diagnosis was made with Anxiety Disorders Interview Schedule. Sleep parameters included CSHQ and 7 nights of actigraphy. Biomarkers of arousal dysregulation included urine catecholamine diurnal samples. Respiratory Sinus Arrhythmia (RSA) was obtained using a BioHarness ECG system taking a clean 5-minute interval from a 30-minute session at the CAR lab.

Results: Sample = 57 ASD, 15 TDC, ages 6-18.

47% of ASD subjects had anxiety. 53% of ASD subjects had insomnia. Our sample had a significant relationship between anxiety and insomnia (Χ²<.0001): 88.9% of children with ASD and anxiety also had insomnia, while insomnia was present in only 20% children with ASD without anxiety. In order to identify the relationship between a phenotype of arousal dysregulation and biomarkers, we separated the cohort into two groups: Group A including children with ASD without anxiety and no insomnia (n=24), and Group B including children with ASD, anxiety and insomnia (n=24).

ASD cohort had longer sleep latency than TDCs (30.5 min vs. 19.8 min, p=.025). Group A sleep latency was 19.7 min, and Group B sleep latency was 40.1 min (p=.001).

Group B had a higher CSHQ score than Group A (p=.082), specifically sleep anxiety (p=.072) and daytime sleepiness (p=.039). Group B sleep onset delay and sleep anxiety scores were greater than TDC (p=.065, p=.026).

Total sample mean evening urine epinephrine level was 9.39 μg/g creatinine, with Group A 7.57 μg/g creatinine, Group B 12.43 μg/g creatinine, and TDC 7.62 μg/g creatinine. Reference epinephrine level for children aged 10-15 years is 8 μg/g creatinine (Pussard). Urine epinephrine levels were greater in Group B than in TDCs and Group A (p=.1628).

Group B had lower RSA than Group A (6.1 vs. 6.9, p=.035).

Conclusions:  A potential link between insomnia and anxiety in children with ASD was found. Sleep latency differed significantly between TDC and ASD, with Group B having the longest sleep latency. Increased epinephrine in evening urine and lower RSA support a hypothesis of an arousal dysregulation phenotype in children with ASD.