Search

The occurrence of epilepsy in autism is variable; nevertheless, EEG paroxysmal abnormalities (PA) are frequently recorded in patients with autism, although the influence of epilepsy and/or EEG PA on the autistic regression has not been clarified yet. We examine a large sample of 345 inpatients with autism, divided into three groups: (1) patients without epilepsy and EEG PA; (2) patients with EEG PA but no seizures; (3) patients with epilepsy including febrile convulsions. The prevalence of epilepsy (24.9%) and EEG PA (45.5%) was higher than that reported in the general population. The significant differences among the three groups concerned autistic regression (comparison between groups 1 and 2, p0.05; comparison between groups 1 and 3, p0.01), cerebral lesions (comparison between groups 1 and 2, p0.05; between groups 1 and 3, p0.001), and symptomatic autism (comparison between groups 1 and 2 as much as comparison between groups 1 and 3, p0.001), which were prevalent in groups 2 and 3; while severe/profound mental retardation was more frequent in group 3 compared to group 1 (p0.01). Focal epilepsy (43.0%) and febrile convulsions (33.7%) were frequent in the third group with epilepsy. EEG PA were mainly localized in temporal and central areas (31.4%). Only 2.6% of patients had subcontinuous/continuous EEG PA during sleep. Seizures and EEG PA were not related to autistic regression. EEG PA occurred mainly in childhood, while epilepsy tended to occur (p0.001) as age increased. The age at onset of seizures had two peaks: between 0 and 5 and between 10 and 15 years with no difference between idiopathic and symptomatic cases. In 58.5% of subjects agedor = 20 years, epilepsy including febrile seizures occurred at some point of their lives, while cases with only EEG PA were less frequent (9.7%). The relationship among autism, EEG PA and epilepsy should be clarified and investigated. In autism, seizures and EEG PA could represent an epiphenomenon of a cerebral dysfunction independent of apparent lesions.