Language in Childhood Epilepsy: A Systematic Review and Meta-Analysis

• In Event: Communication Outcomes for Differently-abled Individuals: Factors associated with Language Development and Accessibility

Abstract

Epilepsy, the most prevalent neurological disorder in children, involves one or more occurrences of a surge of abnormal brain activity (i.e., seizure; Plioplys et al., 2007). It is a heterogeneous condition with considerable variation in its origin (e.g., localization) and age of onset (Berg et al., 2010). The abnormal brain activity associated with childhood epilepsy can have an impact on the developmental trajectory of cognitive processes, like language, in this population. However, there is variation in how researchers study language ability in children with epilepsy and the results that are reported (Selassie et al., 2008; Vanasse et al., 2005). Some researchers find that children with epilepsy have deficits in their language ability (Selassie et al., 2008; Verotti et al., 2011) whereas others report no differences (Lopes et al., 2014. These contradictory results may be due, in part, to differences in seizure classification of the children being studied (e.g., generalized seizures vs. focal seizures) and what components of language are measured (e.g., syntax vs. semantic). The current systematic review and meta-analysis uses data from 13 available studies to consider the magnitude of language differences in children with epilepsy compared to their typically developing peers. Seizure classification (generalized seizures, focal seizures, no specification), age of onset, component of language measured, and instrument used to measure language were all considered as potential moderators of differences in language skill.
Consistent with the heterogenous nature of childhood epilepsy we found considerable variability with respect to studies reporting language deficits (or lack thereof). The results indicate a significant large effect size for language deficits in children with epilepsy compared to their peers. Seizure classification partially, but not fully, accounts for the variability in effect size. Children with epilepsy are at increased risk for language problems, especially children with generalized seizures. However, there is a paucity of research regarding the language profile of children with generalized seizures aside from those noting difficulty with lexical retrieval. In addition, effect sizes differ relative to component of language measured; effect sizes were greatest in magnitude for semantic language and verbal fluency, and minimal for syntax, but only when including all studies of children with epilepsy, regardless of seizure classification. These findings differ when considering language component in children with generalized or focal seizures only. Children with focal seizures are at increased risk for impaired language that can be largely attributed to poor semantic language. Given the research that is available, it appears earlier age of onset has a greater impact than later age of onset. This negates the idea that the plasticity of the brain will serve to overcome language difficulties.
Our results have important research and clinical applications. Based on this review, it is clear that researchers and clinicians should be cautious in the selection of measures used to assess language in children with epilepsy. Assessments should include subtests that discriminate ability for specific components of language. Moreover, knowledge of etiological factors (e.g., seizure classification) is fundamental for research and clinical applications.

Authors

• Katharine M Bailey, University of Alberta
  Presenting Author

• Nancie Im-Bolter, Trent University
  Non-Presenting Author