Abstracts

Rationale: There is abundant data from population series on the association of abnormal electroencephalograms (EEG) in children with autism spectrum disorders (ASD). However, there is very little retrospective or prospective data regarding the evolution of EEG findings in children with ASD. Further characterization of these changes will help evaluate the potential predictive value of EEG abnormalities in children with ASD.Methods: The study population was identified by querying the current neurophysiology database at Childrens Medical Center Dallas for patients with the clinical diagnosis of autism who had more than one electroencephalogram between January 2002 and May 2015. Medical records were reviewed and data on patient demographics, medical history, neurological examinations, EEG findings, neuroimaging, and anti-seizure medication management were extracted systematically. All EEGs were reviewed by board certified physicians in clinical neurophysiology and all patients met clinical criteria for the diagnosis of ASD as per the Diagnostic and Statistical Manual of Mental Disorders (DSM IV-TR).Results: 108 patients were identified with a total of 331 EEGs performed. 77 patients were male. 44 patients did not have seizures or had staring spells which were not clearly epileptic. The average age of seizure onset was 6 years (range: 1 week - 16 years). The average age of diagnosis of autism was 2.9 years. See demographic table for details. 39% (42/108) of patients had an EEG which was initially normal. For patients without clinical seizures 45% (20/44) of initial EEGs were normal whereas patients with clinical seizures 33% (21/63) of initial EEGs were normal (see chart 1). Of the 20 patients with normal EEG and no clinical seizures, 2 patients had subsequent abnormal EEGs whereas the 21 patients with normal EEG and clinical seizures, 7 patients had subsequent abnormal EEGs (p = 0.13; see chart 2). 50% (9/18) of patients with history of regression presented with an initially normal EEG whereas 34% (27/79) of patients without history of regression presented with a normal EEG (p = 0.28).Conclusions: EEG abnormalities were common in our population. Epileptiform and nonepileptiform EEG abnormalities were more frequent in patients with clinical seizures. 10% of patients without clinical seizures and an initially normal EEG had a subsequent abnormal study whereas 33% of patients with clinical seizures and an initially normal EEG had a subsequent abnormal study. Patients with a history of regression did not have higher rates of epileptiform or nonepileptiform abnormalities. Some patients appear to have intermittent abnormalities which may not be present on initial EEGs, thus, recurrent EEGs may help characterize these abnormalities. EEG is a useful clinical tool, but given the higher rates of abnormalities in patients with autism clinical judgement regarding treatment is vital.