Abstracts

Rationale: Understanding trends in the characteristics of autistic children with epilepsy may improve the anticipatory management of this vulnerable group. Yasuhara (2010) showed that in serial routine sleep EEGs of autistic children with epilepsy, spike discharges most frequently (65.6% of cases) originated from the frontal lobe. Further, Kanemura (2013) demonstrated that frontal interictal epileptiform discharges (IEDs) in autistic children are significantly associated with later development of epilepsy. In the present study, we sought to confirm whether or not frontal IEDs are associated with comorbid epilepsy in a sample of autistic children followed by our pediatric neurology department. Methods: This was an IRB-approved retrospective chart review of 154 autistic children (mean age 9 years; 124 males; 30 females) who were identified to have received at least one EEG from Stony Brook Child Neurology in the year 2017. Chart review also revealed the results of past EEGs, if present in the electronic medical record, dating back to 2012. Data were collected for age, gender, location of IEDs on EEGs, and presence or absence of comorbid epilepsy. Results: There were 46 epileptic autistics, and 108 non-epileptic autistics. There were 54 autistics with IEDs, and 100 autistics with no IEDs. There were 38 epileptic autistics with IEDs, and 8 epileptic autistics with no IEDs. There were 16 non-epileptic autistics with IEDs, and zero non-epileptic autistics with no IEDs. In the 46 epileptic autistics, the frontal lobe was the most common location of IEDs (27%) followed by generalized (17%), temporal (17%), central (13%), parietal (7%) and occipital (5%). In the 16 non-epileptic autistics, central (23%), temporal (23%), and occipital (23%) were equally common, followed by frontal (16%), parietal (15%), and generalized (0%).Formal data analysis was performed on the subdivision of autistics who satisfied any of three criteria: 1. had IEDs, 2. carried a comorbid diagnosis of epilepsy, 3. both (1) and (2). There were 62 autistics in this subdivision. Binary logistic regression was used to calculate increased or decreased odds of epilepsy by location of IEDs. Frontal IEDs did not predict comorbid epilepsy (OR=1.2, p=0.85). Generalized IEDs (OR=5.0, p=0.72) and temporal IEDs (OR=4.7, p=0.09) seemed to confer increased odds of epilepsy, although the p-values were not statistically significant. Occipital IEDs conferred decreased odds of epilepsy to a significant degree (OR=0.04, p=0.005). Conclusions: In contrast with previous studies that have shown a significant association between frontal IEDs and comorbid epilepsy in autistic children, we could not demonstrate this association to statistical significance. We did observe that in epileptic autistics, the frontal lobe was the most common location of IEDs (27%). Our statistical analysis also showed that occipital IEDs are significantly associated with decreased odds of epilepsy (OR=0.04, p=0.005).In some regards, the quality of our evidence approached that of a case-control study because some of our subjects had multiple historical EEGs. However, most of our subjects only had one EEG, limiting the quality to that of a cross-sectional design. Our next step is to identify a larger number of non-epileptic autistic children that have IEDs on EEG and follow their cases for development of comorbid epilepsy. In the meantime, the trends suggested by the present study may still be of value in the anticipatory management of autistic children. Funding: No funding