Abstracts

Rationale: Bilingual patients pose a particular challenge for presurgical language mapping, as previous studies have shown that results may be different for primary (L1) and secondary (L2) languages. Electrocorticographic (ECoG) spectral mapping is an emerging functional mapping technique that can complement routine electrocortical stimulation mapping (ESM). Event-related power augmentation in high gamma frequencies (>60 Hz) is a particularly promising index of functional activation. We hypothesized that this graded measure of cortical activation might help discriminate the cortical representations of L1 and L2 in bilingual patients. Methods: We studied bilingual patients for whom English was a second language (L2) and in whom subdural electrodes were implanted for epilepsy surgery over dominant perisylvian cortex. Patients underwent functional mapping with ESM and if possible, with ECoG, during a visual object naming task using pictures from the Boston Naming Test. In each case, mapping was repeated with responses in either L1 or L2. ESM was also performed during spontaneous speech, verbal comprehension (modified Token Test), and reading aloud in English. Time-frequency analysis of ECoG signals (1-300 Hz bandpass, 1000 Hz sampling) was performed using a matching pursuit algorithm, and event-related power augmentation in high gamma frequencies during the object naming task were estimated with respect to a 1-second pre-stimulus baseline. Results: Patient #1 (L1 = Igbo) underwent ESM at 34 electrode pairs, and patient #2 (L1 = Italian) underwent ESM at 70 pairs of electrodes. In both patients, there was no difference between L1 and L2 in the distribution of sites with ESM-induced impairments of visual object naming. ECoG was recorded in patient #2 during visual object naming with both English and Italian responses. High gamma power augmentation was seen in the majority of sites where ESM interfered with L1 and L2 naming (60% and 80%, respectively) and was absent in the majority of sites where ESM did not interfere with naming (80% and 82%, respectively). Although the sites where ESM interfered with naming were identical for L1 and L2 responses, the distribution of ECoG high gamma responses were distinct for L1 and L2. Conclusions: Electrocortical stimulation mapping revealed few differences in the cortical representations for primary versus secondary languages. However, the spatial distribution of ECoG high gamma activity was distinct during performance of visual object naming in primary and secondary languages. These results suggest that comprehensive language mapping prior to surgical resection should include maps for both primary and secondary languages. In addition, the results of ECoG spectral mapping may complement those of ESM when mapping language cortex.