Abstracts

Rationale: Children with autism spectrum disorders (ASD) have a higher frequency of epilepsy and interictal epileptiform discharges (IEDs) than typically developing children (TD); however, these findings have been noted predominantly in children with ASD, mental retardation( MR), and epilepsy. The goal of this study is to determine the presence of IEDs in children with ASD who are medication free, do not have a history of MR or epilepsy, compared to an age-comparative group of TD controls. To better asses IED prevalence, overnight sleep EEGs from non-REM sleep were reviewed. To determine whether the first hour of sleep will provide a valid means of detecting EEG abnormalities, the results of the first hour were compared to those of the entire study. We also identified paroxysmal, high amplitude rhythmic slowing (HRSA), which has been described in children with ASD, and may represent either a variant of epileptiform activity or an arousal response. This study was designed to provide pilot data for a larger analysis on the relationship of sleep IEDs to cognition and behavior in children with ASD. Methods: Overnight 21-channel EEGs using the International 10-20 system were completed on children participating in a study characterizing sleep in children with ASD at our Clinical Research Center. EEGs were examined on 20 children, ages 4-10yrs, 13(11 boys, 2girls) who met DSM-IV diagnostic criteria for ASD confirmed by the Autism Diagnostic Observation Schedule and 7(5boys, 2girls) who were TD controls. All children were medication-free, and without epilepsy or MR. EEGs were read independently by two readers who were blinded to the diagnosis. Results: In all 20 children, the first hour of sleep contained all NREM sleep stages. HRSA was seen predominantly at sleep onset in 45% of cases. It was present as paroxysmal, high amplitude, generalized theta range discharges, lasting for 2-4 seconds, containing low amplitude sharply contoured discharges. IEDs were present as intermittent multi-focal(57%)or focal epileptiform discharges, predominantly located in the right hemisphere(71%). In the 13 children with ASD, 7 had abnormal overnight EEGs, consisting of IEDs in 5(4 of whom also had HRSA)and 2 with only HRSA. In the children with IEDs, the first hour of sleep captured these IEDs in all cases. In the children with HRSA, the first hour of sleep captured HRSA in two cases and missed it in two cases. In the 7 TD children, 4 had abnormal overnight EEGs, consisting of IEDs in 2 (1 also with HRSA) and 2 with only HRSA. In all of the TD children, the first hour of sleep captured both IEDs and HRSA. Conclusions: The frequency of EEG abnormalities in children with ASD who are medication-free, seizure-free, and without mental retardation did not differ from age comparable TD children. Furthermore, EEG abnormalities detected in the first hour of sleep were comparable to the entire overnight study; of note, all stages of NREM sleep were recorded in the first hour of sleep. The significance of sleep IEDS and paroxysmal HRSA to measures of cognition and behavior in children with ASD is currently under investigation.