Abstracts

Rationale: Comprehensive language testing during intracarotid amobarbital procedure (IAP) has revealed right hemisphere language representation, particularly for individuals with early onset of seizures as well as left hemisphere seizure focus. However, individual language abilities (i.e. reading, naming, comprehension, repetition) appear to be differentially affected during anesthetization implying dissociated lateralization of these language functions. In addition, earlier case studies have described other language functions independently represented in opposite hemispheres (i.e. speech production in right hemisphere vs. comprehension in left hemisphere). This study was undertaken to identify possible dissociated language representation patterns. Methods: A retrospective database analysis was conducted of 608 patients who underwent presurgical evaluation including the IAP. The IAP protocol described by Loring et al. was modeled to assess hemispheric representation of language and memory function. In this protocol, the language functions of comprehension, naming, repetition, and reading were rated on a 4-point scale (0-3). Language scores were not included if there was evidence of early recovery of function based on EEG monitoring and/or motor assessment. The number of correct items was divided by the total number of items administered to generate a percentage correct for each hemisphere (range 0 to 100). The right percentage score was then subtracted from the left percentage score to produce a lateralization index of (range +100 to -100), where positive scores indicate left hemisphere language dominance and negative scores indicate right hemisphere language dominance. Modeling Loring et al., right hemisphere language dominance was then identified for those patients demonstrating a lateralization index of ≤ -15. For this analysis, only right hemisphere dominant subjects (n=44) were included. Scores were analyzed by nonparametric statistics as the language ratings were not normally distributed. Spearman's rho was utilized to determine if each of the language scores correlated with one another. Friedman's test was used to determine if there were significant differences between the four language function scores across all patients. Results: Spearman's rho was significant, indicating positive correlations between all language areas assessed (see table 1). Consistent with this finding, mean language scores were not significantly different from one another based upon Friedman's test (Chi Square 4.688, p > .05; see table 2). Conclusions: The strong positive correlations between the language scores indicate that these scores vary in concert with one another. Furthermore, group comparisons did not reveal any language function that was more strongly represented in the right hemisphere as compared to the others. Although previous case reports have described hemispheric dissociation of language functions, the above findings indicate that language functions generally tend to be represented in the same hemisphere together and there was no dissociated hemispheric language representation pattern.