Cortisol, Sleep and Heart Rate Variability in Resting Neurotypical and Autistic Adults

Background: The autonomic nervous system modulates heart rate via two antagonistic systems: sympathetic and parasympathetic. Individuals with an autistic spectrum disorder (ASD) are reported to display high sympathetic activity and atypical diurnal cortisol secretion patterns.

Objectives: The objectives of the present study were two-fold: a) to compare evening and morning cortisol levels in ASD and typically developing (TD) adults; b) to explore the association between cortisol, self-reported sleep habits and heart rate variability (HRV).

Methods: Sixteen unmedicated ASD adults (22.0±3.7 years, 15M, 1F) without sleep complaints, intellectual disability and psychiatric nor neurological comorbidity were compared to 17 TD healthy participants (21.7±4.0 years, 16M, 1F). Five salivary cortisol samples were collected in the evening and two in the morning, 20 minutes apart. Sleep latency, nocturnal awakenings, total sleep time and sleep efficiency were assessed by self-reports. The electrocardiogram was recorded in the evening and in the morning to compute low (LF) and high (HF) spectral frequencies. Cortisol and HRV parameters were compared using two-way ANOVAs (ASD vs TD X evening vs morning). Sleep parameters were compared between the 2 groups using independent Mann-Whitney U tests. Pearson correlations between cortisol levels, sleep measures and HRV parameters were calculated.

Results: Groups did not differ on cortisol levels. Compared to the TD group, ASD self-reports showed longer sleep latencies (38.6±12.6 vs 13.3±2.3 minutes; p<0.01), longer nocturnal awakenings (11.6±3.3 vs 2.5±0.8 minutes; p<0.01), lower sleep efficiencies (90.2±3.3 vs 95.3±1.6 %; p=0.01) but same total sleep time (7.9±0.3 vs 8.0±0.2 hours). Morning HF values were higher in ASD than TD (p=0.05). In the TD group, higher evening cortisol levels correlated with longer nocturnal awakenings (r=0.7; p=0.02); higher morning cortisol levels correlated with longer total sleep time (r=0.7; p=0.02), lower morning LF (r=-0.9; p=0.002) and higher morning HF (r=0.9; p=0.002). No significant correlations were found in the ASD group.

Conclusions: Only the TD group showed significant correlations between cortisol levels, sleep and HRV. The lack of significant correlations in the ASD group could reflect the presence of poor self-reported sleep despite the lack of complaints or an alternative coupling between neuronal and endocrine mechanisms of sleep control in ASD.