Abstracts

Rationale: Simultaneous EEG/functional MRI (EEG-fMRI) measures non-invasively hemodynamic responses to interictal epileptic discharges (IED), which have been related to the epileptogenic region. On the other hand, high frequency oscillations (HFOs) are an excellent indicator of the epileptogenic zone (EZ), but are primarily recorded from intracerebral EEG (iEEG). We studied the spatial overlap of these two important markers in patients with drug-resistant epilepsy. Methods: We included patients with drug-resistant epilepsy who underwent EEG-fMRI and later iEEG. We defined two groups based on iEEG findings: patients with focal seizures (FS) and patients with non-focal or unknown onset seizures (NF/US). We co-registered hemodynamic t-maps with anatomical imaging to determine the localization of the intracerebral electrodes. Each channel was classified as pertaining to one of the following EEG-fMRI categories: primary cluster (maximum absolute t-value), secondary cluster (absolute t-value > 90% of the primary cluster) or outside the primary and secondary clusters otherwise. We marked HFOs (ripples: 80-250 Hz; fast ripples: 250-500 Hz) during one hour of slow wave sleep. Rates of HFOs in each EEG-fMRI category were calculated. Channels were then classified as high- or low-HFO-rate. The proportion of observed high-HFO-rate channels that belonged to the primary or primary plus secondary clusters was compared to the proportion expected by chance. Results: We included 25 patients (16 female; mean age 29.4 ± 8.2 years), 11 with FS and 14 with NF/US. We found a significantly higher median HFO rate in the primary cluster compared to secondary cluster and outside these two clusters for the FS group (p < 0.0001), but not for the NF/US group. For the FS group, the observed number of high-HFO-rate channels (ripples and fast ripples) within the primary or primary plus secondary clusters was significantly higher than expected by chance. This remained true only for the ripples in the NF/US group. At the patient level, most patients (17/25; 68%) had at least one high-HFO-rate channel within a primary cluster.  Conclusions: In focal epilepsy patients the maximum IED-related hemodynamic response overlaps with the tissue generating HFOs at high rates. This EEG-fMRI response should be explored if iEEG is considered before surgery. If it shows the highest rates of HFO, it is highly possible that a focal EZ is demarcated by this hemodynamic response. Otherwise, the patient may have a widespread EZ or a focal one not covered by the implantation.  Secondary hemodynamic responses, close to the maximum response, may be of additional help in identifying the epileptogenic tissue. Funding: This study was supported by the Canadian Institutes of Health Research grant FDN 143208. K.G.O was also supported by the Frederick Andermann Clinical/Research Fellowship in Epileptology and EEG.