A Family History Study of Autism and Epilepsy

Background: Previous studies have found similar rates of epilepsy among family members of individuals with and without autism. However, when epilepsy status of the person with autism is considered, higher rates of epilepsy are found in family members of individuals with co-occuring autism and epilepsy vs. autism only. There is increasing evidence that both autism and epilepsy are associated with a host of other developmental, psychiatric, and medical comorbidities. We propose to study the prevalence of these comorbidities in family members of individuals with autism and epilepsy.

Objectives:  This objective of this study was to examine family history data in an autism dataset as a function of the epilepsy status of the proband. We hypothesized that there would be a greater rate of developmental, psychiatric, and medical comorbidities in first-degree family members of individuals with autism and epilepsy.

Methods:  We collected standard family history data for 506 autism families (98 multipex/408 simplex); 11% of our probands had epilepsy based on parent report and medical record review. We examined 9 family history traits: seizures/epilepsy, migraines, gastrointestinal problems, OCD, schizophrenia, depression, bipolar disorder, dyslexia, and learning disabilities. Groups were defined using epilepsy status of affecteds (epilepsy/non-epilepsy) based on informant report.  We examined group differences in the presence/absence of each trait in first-degree family members as a function of epilepsy status of the proband using logistic regression. Next, we investigated the frequency of first-degree family members positive, again by epilepsy status of the proband, for each variable adjusting for intra-familial correlations using Poisson regression. All analyses were adjusted for family type (multiplex vs. simplex), age at examination, and sex.  

Results:  Adjusting for family type, age at examination, and sex, we found significantly higher rates of dyslexia (p=0.03; OR=2.8) and learning disability (p=0.04; OR=2.3) in family members of the autism-epilepsy group. In our second analysis, using a GEE approach to correct for intra-familial correlation, we found a significantly greater frequency of first-degree family members positive for dyslexia in the autism-epilpsy group (p=0.02) as well as a borderline significantly greater frequency of first-degree family members  positive for learning disability (p=0.06) in this same autism-epilepsy group.  

Conclusions:  Our results suggest an increased rate of learning problems in first-degree family members of individuals with co-occurring autism and epilepsy. Previous reports have cited an increased risk of various cognitive and learning problems in relatives of individuals with autism; our findings suggest that this risk in autism may be heightened in the presence of epilepsy. To our knowledge, this is the first study to test this hypothesis. Our results are intriguing in that increasing evidence suggests that genetic variants which confer risk to autism may also yield other neurobehavioral phenotypes in both individuals affected with autism as well as family members  (Miller 2009). Since our results are based solely on informant report they must be interpreted cautiously. However, this exploratory study provides new clues regarding the autism-epilepsy relationship and extends the range of neurobehavioral phenotypes that may be tied to potential underlying genes when autism and epilepsy co-occur.