Abstracts

Rationale: Simultaneous EEG-(BOLD)fMRI (Electroencephalography-Blood Oxygen-Level Dependent functional Magnetic Resonance Imaging) has been widely used for measuring BOLD changes associated with IED’s (interictal epileptic discharges). However, BOLD change is not a direct measure of neuronal activity. Instead, it is influenced by multiple factors such as oxygen consumption, cerebral blood flow, and CBV (cerebral blood volume). Therefore, characterization of CBV during the IED is important for the understanding of the neurophysiology of epilepsy. In this research, we conducted simultaneous EEG-BOLD measurements and simultaneous EEG-CBV measurements in different runs from the same group of epilepsy patients, utilizing a non-invasive, MRI based CBV measurement method called VASO (Vascular Space Occupancy) imaging. Methods: Three localization related epilepsy patients (17 to 50 years old) with frequent IED’s have participated in the study after written consent. Each patient underwent four alternate runs (eight in total) of EEG-BOLD and EEG-CBV measurements in the resting state. EEG was recorded using a 32-channel carbon wire electrode system. The IEDs were manually identified from the EEG after scanner noise and ballistocardiogram removal. The BOLD image was obtained using an EPI (echo-planar imaging) sequence (TR/TE=1550/50 ms, 25 6mm axial slices). The CBV image was obtained using a Global Inversion Cycling VASO sequence (TR/TE=3000/8.8 ms, 21 5mm axial slices). BOLD and CBV images were then subject to the generalized linear model analysis to obtain IED related BOLD and CBV activation maps for each patient. Since the VASO signal decreases as the CBV increases, the sign of the VASO activation map was inverted. To compare BOLD and CBV timecourses, a ROI (region of interest) was defined using the BOLD activation map, and average BOLD and CBV signal timecourses in the ROI around the IED events was computed for each patient. Results: In all subjects, significant BOLD activation was found at regions concordant with clinical EEG reading. Although local maxima of CBV were found near these loci in all subjects, BOLD and CBV peaks were centered at different locations. However, voxelwise correlation of the BOLD map and CBV map within the cortex showed positive correlations in all subjects (p < 0.001, Bonferroni corrected). In all subjects, average EPI (BOLD) and VASO (negative CBV) timecourses showed positive and negative responses to the IED’s, respectively (See Figure). One second before the IED, both EPI and VASO signals were decreasing, suggesting that the BOLD was decreasing and the CBV was increasing at this time in all subjects. Conclusions: VASO (CBV) signals decreased in the suspected epileptogenic regions in response to IED events. VASO can be used to investigate neurovascular activity in response to IED and explore the complex relationships among BOLD, CBV and electrophysiology in epilepsy. This research was supported by NIH R01-NS47605. .