Autistic Children with EEG Abnormalities and/or Epilepsy: Clinical Characterization in Two Independent Samples

Background:  Several studies have reported a high frequency of epilepsy and/or co-occurrence of EEG abnormalities in children with autism spectrum disorder, although the incidence rates vary between 5% to 46%. The data heterogeneity reported in literature is probably due to the different characteristics of the samples examined in each study which may include different clinical features and ages. Many factors such as mental retardation, brain impairments and rare genetic diseases associated with ASD, increase the risk of epilepsy in symptomatic or syndromic autism. However, seizures or EEG abnormalities detected in “idiopathic” ASD remains higher than in general population, suggesting that autism is associated with a higher risk of epilepsy and EEG abnormalities. To this date it is not clear if this comorbidity represents an epiphenomenon of the neural dysfunction in autism, or if there may be a causal relationship between the two.

Objectives:  To reconsider the mentioned association between epilepsy and/or EEG abnormalities with clinical, biological, developmental and familial medical history. We examined data extracted from two different large and selected samples of ASD patients.

Methods:  We analyzed two different samples of non syndromic ASD patients: an original sample encompassing 432 Italian patients and a replica sample including 714 Caucasian American patients recruited by the Autism Genetic Resource Exchange Consortium (AGRE). Comparable clinical, biological and developmental variables were correlated in both samples with “presence/absence of epilepsy” or “presence/absence of EEG abnormalities”, and tested for association between each variable and biological endophenotypes (serotonin blood levels, head circumference and global peptiduria), by non parametric Kendall τ, Kruskal-Wallis ANOVA and Mann-Whitney U test.

Results:  In the experimental sample and in the replica sample, ASD patients positive for EEG abnormalities display a significant association with a lower risk of familial history for autoimmune/allergic diseases (Italian and AGRE sample: τ= -0.115 and -0.252, respectively; both P<0.05), and those affected by epilepsy are characterized specifically by verbal language delay in the absence of a generalized neurodevelopmental delay (Italian and AGRE sample: τ= 0.100 and 0.073; both p<0.05). Interestingly, Italian patients with EEG abnormalities show significantly lower serotonin blood levels (τ=-0.183; P<0.01).

Conclusions:  

Overall, epilepsy and EEG abnormalities remain a relatively isolated and independent features, not strongly associated with specific symptom patterns. Nonetheless, these results extend previous findings, by supporting a deleterious effect of EEG abnormalities on verbal language development, and by pointing toward a protective role against the development of epilepsy for a family history of autoimmune/allergic diseases and elevated serotonin blood levels. These results spur interest into the potential positive effects of antiepileptic drug treatment in facilitating language development among small non-verbal ASD children with EEG abnormalities, even in the absence of overt seizures.