EEG Markers of Alterated Sleep In Adults with Autism: Stage 2 K-Complexes

Background: Objective measures of sleep in adults with Autism Spectrum Disorders (ASD) show signs of alterated sleep, including a delayed sleep onset and increased time after sleep onset. K-complexes are sleep EEG phasic events mainly prevalent during stage 2 and generated by cortico-cortical loops. K-complexes are thought to reflect brain mechanisms involved in sleep protection by inhibiting access of incoming messages to the cortex. K-complexes can be used as EEG markers to quantify sleep maintenance capacity.

Objectives: To evaluate the sleep maintenance capacity of autistic adults through the quantification and scalp distribution of K-complexes, using a full EEG montage. 

Methods: Sixteen autistic adults (ASD: 14 M, 2 F; 22.1 ± 1.3 years) and normal IQ and a comparison group of 18 typically developed participants (TD: 17 M, 1 F; 21.1 ± 1.0 years) were recorded for two consecutive nights in a sleep laboratory. Sleep stage 2 K-complexes were visually identified and quantified for 14 recording electrodes (Fp1, Fp2, F3, F4, F7, F8, C3, C4, P3, P4, P7, P8, O1, O2) according to the following criteria: a negative-going biphasic wave with sharp onset and smoother offset, lasting 0.5 to 1.5 seconds, with an amplitude of at least 75 µV. Spectral analysis of the EEG 2 seconds before and 2 seconds after K-complexes was performed and spectral power was calculated for delta (0.5-3.5 Hz), theta (4.0-7.5 Hz), alpha (8.0-12.5 Hz), sigma (11.5-14.5 Hz) and beta (13.0-30.0 Hz) frequency bands. 

Results: The ASD group generated less K-complexes per hour of stage 2 over all recording sites compared to the TD group but a significant difference was reached only over parietal cortex bilaterally (P3 electrode: ASD = 49.6±6.5 vs. TD = 75.9±10.2, p<.05; P4 electrode: ASD = 44.8±6.0 vs. TD = 78.8±9.8, p<.01). The analysis of ASD sub-groups suggested that high functioning autism (HFA: n = 9) contributed more than Asperger Syndrome (n = 7) to this effect. Preliminary analysis of EEG spectral power showed a significant increase in relative delta activity after vs. before K-complexes in TD participants. 

Conclusions: : 1) The decrease density of K-complexes in ASD suggest that neural substrates of sleep stage 2 K-complexes, particularly cortico-cortical loops, are atypical in adults with ASD. 2) Impairment of this cortical sleep protective mechanism may be responsible for fragmented sleep in ASD. 3) The analysis of EEG activity in the TD group shows that K-complexes are indeed followed by cortical hyperpolarization, a sleep protective mechanism that will be assessed in further analyses in the ASD group.