Heart Rate Variability and Anxiety in Autism Spectrum Disorders

Background: Anxiety is a frequently occurring co-morbid condition of Autism Spectrum Disorders (ASD), with prevalence rates estimated to be approximately 40% or higher in some studies. As most anxiety measures are heavily loaded on verbal abilities, there is a substantial need for non-verbal indices of anxiety in ASD. Peripheral nervous system function may ultimately provide a language-free index of arousal dysregulation in ASD—one that has been linked to anxiety disorders in typically developing populations. In anxiety disorders, dysregulation of the parasympathetic nervous system (i.e., abnormal vagal tone) has been associated with an inability to adjust appropriately and quickly to environmental changes and stressful challenges (Beauchaine et al., 2007). One way to measure physiological dysregulation is with Heart Rate Variability (HRV), defined as the rhythmic beat-to-beat change in heart rate. Conversely, greater HRV is associated with superior adaptability and autonomic regulation to stressors. The present study examines the relationship of HRV and anxiety in ASD.  

Objectives: The aim of this study was to evaluate differences in HRV among children with ASD and comorbid anxiety, ASD alone, and TDC, when viewing 8 minutes of static and 8 minutes of dynamic sets of images of neutral, happy, and angry facial expressions. Angry faces have been shown to elicit an anxiety response in both neurotypical and ASD populations.

Methods: Participants were assessed for ASD and anxiety using gold-standard instruments: ADOS, ADI-R, and ADIS. Cognitive ability was evaluated with the DAS-II. HRV and respiration were recorded using the Biopac MP150 Respiration & ECG Modules, sampled at 1000 Hz and analyzed using power spectral analysis to determine power in the high frequency range of 0.12-1.00Hz. Participants were seated in a comfortable chair and static and dynamic facial stimuli were presented on a computer screen for 8-minute sessions. Response data was also recorded after each trial where participants pressed a button to discriminate neutral, happy, and angry facial expressions.   

Results: 46 participants (33 ASD, 13 TDC) completed both the face tasks and the anxiety disorder diagnostic battery. Of the 33 individuals with ASD, 20 met criteria for at least one anxiety disorder, yielding three groups: ASD with anxiety, ASD only, and TDC (who were preselected to be free of Axis I psychopathology). An ANOVA showed no group differences within the three groups on age; F(2,45)=0.125, p=.0.883 (mean age in years=11.51, 11.05, 11.46, respectively) and IQ; F(2,45)=2.184, p=0.125 (DAS-II GCA; mean IQ=99.05, 110.92, 113.85). Data analysis is ongoing.  

Conclusions: Although data analysis for this study is in preliminary stages, this study contains one of the first ASD samples that have been well-characterized in terms of both anxiety disorder status and parasympathetic nervous system function. Data from this study will fill a gap in our understanding of the psychophysiology of anxiety in ASD. Ultimately, measures of autonomic flexibility such as HRV may provide insight into how individuals with ASD and co-occurring anxiety manage the affective elements of interpersonal situations.