How canonical is canonical? Examining the BOLD response to interictal discharges at the epileptogenic site using intracranial EEG-fMRI

• Published in: Canadian journal of neurological sciences 2014-05, Vol.41 (3), p.S25-S25
• Main Authors: Beers, C, Gaxiola-Valdez, I, Pittman, D, Kang, A, Agha-Khani, Y, Federico, P
• Format: Article
• Language: English
• ISSN: 0317-1671

Background: Simultaneous EEG-functional MRI (EEG-fMRI) is a well-established technique for imaging the Blood-Oxygen-Level Dependent (BOLD) response associated with interictal epileptiform discharges (IEDs). Essential to this analysis is application of the hemodynamic response function (HRF), which quantifies BOLD signal changes in response to neural events. The standard HRF was originally derived from an auditory response task performed on healthy subjects. However, it is not know if the HRF is truly generalizable to all tasks. Therefore, we quantified the BOLD signal changes associated with intracranial EEG-fMRI (iEEG-fMRI) IEDs recorded directly from the epileptogenic source. Methods: Nine subjects undergoing intracranial video-EEG monitoring underwent simultaneous iEEG-fMRI at 3T. Epileptiform discharges that were isolated 3s prior, and 10s post-spike were identified. Regions of interest encompassing 50 voxels nearest to the active electrode were created. BOLD signal changes for isolated spikes were averaged to generate HRFs characteristic of each subject. Results: Seven to 23 IEDs were identified for each subject. BOLD responses recorded from the epileptogenic tissue showed a later time-to-peak, ranging from 7.5-10s, compared to 4-6s for the standard HRF. Conclusions: The HRF associated with IEDs recorded intracranially has a longer time-to-peak than the canonical HRF. These results imply that epileptic discharges may generate BOLD signal changes unique from other neural events.