Effect of epilepsy on autism symptoms in Angelman syndrome

Kristin A. Bakke, Patricia Howlin, Lars Retterstøl, Øivind J. Kanavin, Arvid Heiberg, Terje Nærland

Autism spectrum disorder and epilepsy often co-occur; however, the extent to which the association between autism symptoms and epilepsy is due to shared aetiology or to the direct effects of seizures is a topic of ongoing debate. Angelman syndrome (AS) is presented as a suitable disease model to explore this association. Data from medical records and questionnaires were used to examine the association between age of epilepsy onset, autism symptoms, genetic aberration and communication level. Forty-eight participants had genetically verified AS (median age 14.5 years; range 1-57 years). A measure of autism symptoms (the Social Communication Questionnaire; SCQ) was completed for 38 individuals aged ≥ 4 years. Genetic cause was subgrouped into deletion and other genetic aberrations of the 15q11-q13 area. The number of signs used to communicate (< 20 sign and ≥ 20 signs) was used as a measure of nonverbal communication. Mean age of epilepsy onset was 3.0 years (range 3 months-7.8 years). Mean SCQ score for individuals without epilepsy was 13.6 (SD = 6.7) and with epilepsy 17.0 (SD = 5.6; p = 0.17); 58% used fewer than 20 signs to communicate. There were no age differences between groups according to presence of epilepsy, level of nonverbal communication or type of genetic aberration. SCQ scores were higher in individuals with the deletion than in those with other genetic aberrations (18.7 vs 10.8 p = 0.008) and higher in the group who used < 20 signs to communicate (19.4 vs 14.1 p = 0.007). Age of epilepsy onset was correlated with SCQ (r = - 0.61, p < 0.001). Multiple regression showed that age of seizure onset was significantly related to SCQ score (β = - 0.90; p = 0.006), even when the type of genetic abnormality was controlled (R2 = 0.53; F = 10.7; p = 0.001). The study provides support for the notion that seizures themselves contribute more to autism symptoms than expected from the underlying genetic pathology alone. The study demonstrates how a rare genetic syndrome such as Angelman syndrome may be used to study the relation between epilepsy and autism symptomatology.