Abstracts

Rationale: Cortical mapping of language-related naming function is an important component of the presurgical evaluation of patients with medically intractable focal epilepsy. Previous studies reported probability maps generated using either electrical brain stimulation (EBS) or functional magnetic resonance imaging (fMRI). However, there are reports that find EBS inadequate for language mapping in children younger than 10 years of age (Schevon et al., Epilepsia, 2007) and fMRI an unsuitable replacement at present (Giussani et al., Neurosurgery, 2010). A more recently developed approach of mapping language cortex involves detection of gamma-augmentation that represents cortical activation driven by a task. In the present study, we have generated naming probability maps using event-related gamma-augmentation on electrocorticography (ECoG) and determined the inter-individual variability in localization of cortices involved in auditory language functions. Methods: We studied 10 patients with focal epilepsy who underwent extraoperative ECoG and subsequent resective epilepsy surgery. All patients underwent preoperative MRI and an auditory naming task for cortical mapping using EBS as well as ECoG. They completed up to 100 question-and-answer trials recorded and integrated with ECoG. Questions were designed to elicit 1 or 2 word answers; e.g. Q: What flies in the sky? ECoG traces were temporally locked to response-onset and transformed into time-frequency matrices. To create an average probability map of naming function, all ECoG electrode positions in native space were transferred into MNI152 template space using landmark-constrained conformal cortical mapping (Muzik et al., Int J Biomed Imaging, 2007). Results: The question-and-answer task required up to 20 minutes of each patient s time. Our findings indicate gamma-augmentations sequentially involving bilateral posterior superior temporal gyri, posterior left middle and inferior frontal gyri, and bilateral inferior pre- and post-central gyri. Gamma-augmentations in the superior temporal gyri driven by auditory stimuli as well as those in the pre- and post-central gyri driven by vocal responses were highly consistent across individuals in localization. However, gamma-augmentation during a delay period between auditory stimulus and vocal response had highly variable inter-individual localization and involved variable portions of the left frontal and temporal areas. Conclusions: Naming-related gamma-augmentations may identify cortex participating in naming activity in both hemispheres. This time-efficient technique may reveal how cerebral cortex participates in auditory-naming processing in patients with focal epilepsy. We have found that gamma-augmentations during the delay period are highly variable in cortical localization among patients. These preliminary data are consistent with our working hypothesis that ECoG data can be used to generate a probability map of naming in humans.