Abstracts

Rationale: Children with epilepsy have a higher incidence of atypical language dominance. Factors that have been associated with atypical language dominance in children include left handedness, focal seizures arising from the left hemisphere and lesions of the left hemisphere. The objective of this study was to identify the effect of acquired and developmental brain lesions on language dominance in children. Methods: This a retrospective cohort study of 64 children with MRI documented brain lesions who underwent presurgical fMRI for language lateralization from January 2005 - December 2008 at the Hospital for Sick Children in Toronto, Canada. Routine presurgical 1.5 tesla MRI scanning was performed to investigate for structural abnormalities, and reported by a pediatric neuroradiologist. Lesion type was classified as developmental (eg. brain malformation, cortical dysplasia) or acquired (eg. stroke, tumor). Blood oxygen level dependent fMRI for language lateralization was performed at 1.5 tesla, utilizing a minimum of 2 standardized language paradigms. Studies were analyzed using AFNI and visually inspected and rated for lateralization and localization of frontal and temporal lobe language areas. Atypical language dominance was defined as right or bilateral hemisphere dominance for frontal and/or temporal language areas. Results: Children with acquired lesions were more likely to demonstrate atypical language. Sixteen of 32 (50%) children with acquired lesions demonstrated atypical language in contrast to 9 of 32 (28%) of children with developmental lesions. Lesion location also had an effect on language dominance. Nine of 22 (41%) of children with left temporal lesions demonstrated atypical language in contrast to 3 of 16 (19%) of children with right temporal lesions (p<0.05). Furthermore, for lesions of the left temporal lobe, 8 of 14 (57%) children with acquired lesions demonstrated atypical language in contrast to 1 of 8 (13%) children with developmental lesions (p<0.05). This is also true for left frontal lesions, with 3 of 4 (75%) children with acquired lesions demonstrating atypical language in contrast to 3 of 8 (38%)with developmental lesions. Conclusions: Children with acquired brain lesions are more likely than children with developmental lesions to demonstrate right hemisphere or bilateral language dominance on fMRI. In this cohort of 64 children, only 28% of those with developmental lesions demonstrated atypical language dominance. This is true even of children with left hemisphere developmental lesions, with only 26% demonstrating atypical language. This is of critical importance in the planning of neurosurgical and epilepsy surgery procedures in children. It cannot be presumed that the critical language centres in children with left developmental lesions will undergo inter-hemispheric reorganization.