Abstracts

RATIONALE: Electrical stimulation studies show large individual patient variability in the number and location of language nodes in the temporal lobe. This study sought to identify determinants of the variability in number of language nodes, and determine if an ipsilateral temporal seizure focus contributes to this variability.
METHODS: We retrospectively reviewed language maps obtained over the past two years at our center by direct electrical stimulation of subdural grid electrodes. We identified 14 language maps from patients with left temporal lobe epilepsy, left hemispheric language dominance on the IAP, and adequate sampling of the temporal neocortex during mapping. There were 7 patients whose seizures began before the age of 13 years, and 7 whose seizures began later. Temporal language nodes were classified as those associated with receptive errors, dysnomia, or mixed dysfunction. The number of nodes of each kind, and their combinations, were the primary dependent variables examined. Variables examined for predictive value included age, gender, education, age of onset of seizures, duration of epilepsy, type of temporal lobe epilepsy (mesial versus neocortical), and number of temporal contacts tested.
RESULTS: For patients whose seizures began before 13 years (mean age 28 years, SD 18, range 12 to 47 : mean duration of epilepsy 22 years, SD 17, range 2 to 45), the number of mixed and receptive language nodes combined was strongly correlated to the age of onset of epilepsy (correlation coefficient r = 0.78). By contrast to this group, for patients whose seizures began after the age of 13 years (mean age 42 years, SD 11, range 25 to 55 : mean duration of epilepsy 18 years, SD 10, range 4 to 34), the number of language nodes was negatively correlated to age of seizure onset (r = -0.51), but positively correlated to duration of epilepsy, especially for dysnomia and mixed language deficits combined (r = 0.67). The mean number of language nodes for this group was 7.3 (SD 3.0), significantly higher (P[lt]0.01) than that for the previous group (mean 3.0, SD 1.73). None of the other variables were significantly related to the number of language nodes for either group.
CONCLUSIONS: Our data suggest that prior to maturation of language cortex, the earlier the age of seizure onset, the more likely it is for receptive and mixed language representations to be located outside their usual temporal locations. In patients whose seizures began after the age of 13 years, the number of language nodes correlates negatively to age of seizure onset, but positively with duration of epilepsy. This finding, together with the significantly greater abundance language nodes in this group, suggests that after maturation of the language cortex, the presence of a chronic ipsilateral epileptic focus may lead to an expansion of language representation in the temporal lobe. It is unclear whether this represents compensatory cortical plasticity, or an artifact of the electrical stimulation near an epileptic focus.