Abstracts

Rationale: Advances in human electrophysiology allow recording across wide spatial and temporal scales. This approach has identified multiple potential biomarkers of epileptogenic brain. Pathological high frequency oscillations (HFOs) show promise for clinical translation. In this study we assessed whether there was a difference between HFOs in temporal and neocortical structures in seizure onset channels (SOZ) and non-seizure onset channels (non-SOZs) in non-lesional class I patients following resection. Methods: Data was collected under a Mayo Clinic IRB investigation of wide-bandwidth electrophysiology recorded from hybrid electrodes in patients undergoing evaluation for epilepsy surgery. The decision to undergo intracranial EEG monitoring was a clinical one. The protocol involves custom hybrid electrodes, which contain standard macroelectrodes and additional microwire arrays (40 micrometer Platinum/Iridium wires spaced 0.5-1 mm; manufactured by Adtech Medical Instrument). Criteria for HFO automated detection are varied; we used an automated detector that uses a cascade of line-length and event to background ratio of the filtered signal. Non lesional patients with Engel Class 1 outcomes at last follow up, following resective surgery were selected. Results: Thirteen patients were included, 3 male and 10 females, 8 temporal and 5 neocortical resections. Combining all patients there was a suggestion of increased HFO counts in SOZ compared to non-SOZ, although this failed to reach significance (Figure 1). When assessing only temporal resections there was a significantly increased number of HFOs in SOZ (Figure 1). There was a non-significant suggestion of increased HFOs in SOZ compared to non-SOZ in neocortical resections (Figure 1). Comparing neocortical versus temporal HFOs in SOZ there appeared to be an increased number of HFOs in temporal structures, although the counts did not quite reach significance (Figure 2). Conclusions: This study helps to further clarify the significance of HFOs; suggesting that HFO counts in temporal structures are different in the SOZ compared to non-SOZ in nonlesional human patients. It further suggests that there is a difference in the frequency of HFOs in SOZ in temporal versus neocortical structures. The fact that all the patients have class 1 outcomes suggests that the resected areas were responsible for seizure generation This small study provides evidence of proof of concept, of the role of HFOs in temporal and neocortical structures in SOZ, suggesting that HFO counts in temporal structures may be a biomarker of seizure onset and provides support for ongoing investigations into the role of HFOs in nonlesional epilepsy.