Abstracts

Deficits in naming and memory are among the most common post-operative cognitive deficits reported among patients undergoing dominant temporal lobectomy for epilepsy. While some authors have suggested that naming and memory are related but independent abilities mediated by lateral temporal cortex and medial temporal lobe structures, respectively (Sass, Westerveld et al.), others have argued that naming is also a hippocampally mediated function (Davies et al.). The present study provides data supporting the dissociation of naming from hippocampal function and provides evidence in favor of naming as a primarily neocortically mediated ability.
A total of 59 patients undergoing dominant left temporal lobectomy with hippocampal resection were studied. All patients had complex partial seizures emanating from the dominant temporal lobe as defined by ictal EEG recording, and were without structural lesions. Neuropsychological testing, including memory, language, and intellectual function was performed before and after dominant left temporal lobectomy that included hippocampus. Neuropsychological test results (Boston Naming Test, WAIS-R IQ scores, verbal Selective Reminding Test) were analyzed in relation to the extent of hippocampal pathology (as measured by neuronal cell density) and the magnitude of neocortical resection in the superior, middle, and inferior temporal gyrus as determined from intraoperative measurement.
Correlational analysis and ANOVA were performed to analyze the relationship between naming outcome, temporal neocortical resection, hippocampal pathology, and other neurocognitive measures. Pearson correlation coefficient reveals significant relationships between baseline naming and IQ (r=0.61; p[lt].001), and postoperative naming and IQ (r=.38; p[lt].01). However, there were no significant relationships between baseline naming, postoperative naming or change in naming and hippocampal pathology (all correlations NS). In contrast, ANOVA reveals that, although they did not differ before surgery, patients whose surgery included partial resection of the superior temporal gyrus (range 2-3.5 cm) showed significantly greater decline in naming (average delta= -8.0) compared with those who had surgery sparing the STG (average delta = -2.3; p[lt].03).
These results indicate that naming is not a function primarily mediated by the hippocampus; the degree of hippocampal pathology has no effect on baseline naming ability or change following dominant left temporal lobectomy. In contrast, anterior portions of the superior temporal gyrus appear to have a significant role in naming. The implications for surgical treatment are that sparing superior temporal gyrus may result in less dysnomia following dominant left temporal lobectomy.