SLEEP EEG ANALYSIS IN CHILDHOOD AUTISM

Objectives: The aim of the study was to investigate the role of sleep EEG analysis in autistic children. The main characteristics of sleep activated epileptiform abnormalities and results of nonlinear EEG analysis were evaluated.

Methods: We examined a group of 77 autistic children (61 boys, 16 girls) with an average age of 9.1±5.3 years. A twenty-one channel overnight sleep video-EEG monitoring was performed using a Schwarzer EPAS 32 system. The EEG records were divided into three groups: normal, abnormal non-epileptiform and abnormal with epileptiform discharges. Non-linear sleep EEG analysis (coarse-grained entropy rate) in 27 autistic children (7.1±3.6 years) were compared with control group of 20 mentally non-retarded deaf children (8.4±2.3 years) with cochlear implantation.

Results: Evaluable EEG records were obtained in 63 autistic children. Sleep EEG were significantly more likely to detect epileptiform abnormalities than the wake EEGs (39.5% versus 23.3%) The most frequent localisation of discharges were in frontocentral regions. CER analysis of sleep EEG in autistic children with epileptiform EEG differed from normal EEG in higher values of variation coefficient in frontal and central areas in NREM 3 and 4.

Conclusions: We found significant decrease in EEG coupling and information drive in autistic children group in comparison to age and IQ matched control group. Results of nonlinear analysis also support the hypothesis of underconnectivity in autism spectrum disorders. Supported by VZ MZO 00064203