Abstracts

Rationale: In clinical practice, there seems to be a correlation between language outcome and Rolandic epilepsy, language outcome seems to be impaired in these children. In this study we measured language outcome of children with Rolandic epilepsy and correlated these results on the working memory-model of Baddeley. Methods: Retrospective patient file study of 48 children (mean age 115 months, SD 19.7) with Rolandic epilepsy. These children were thoroughly evaluated on the epilepsy centre Kempenhaeghe, the Netherlands. All children had a 24 hour electroencephalogram and a neuropsychological assessment. Furthermore, questionnaires were given to the parents and the teachers of these children to measure the behaviour and the subjective expected school performance by them. Results: In children with Rolandic epilepsy, parents report their children to have more language problems, like reading, writing and perception and expression of language. There was a significant delay in months on the established reading skills at school (words mean delay in months 6.0 SD 11.9, p<0.002, sentences mean delay in months 8.6 SD 12.7, p<0.000), but not on the reached mathematics skills at school (mean delay in months 4.1 SD 12.5). Semantic language performance is more impaired in children with Rolandic epilepsy (mean delay of 8.6 months SD 12.7). The phonological loop was not impaired (mean score digit span WISC 9.2 SD 3.4). Conclusions: Language is impaired in children with Rolandic epilepsy. Our results suggest that this is not depending on the phonological loop or on the central executive component of the widely accepted working memory-model of Baddeley. Semantic language skills are more impaired than phonological language skills. Reading disorders are frequently noticed by the parents of children with language impairment. Children with Rolandic epilepsy should be screened for being at risk for reading problems.