Abstracts

Rationale: High frequency oscillations (HFOs) have recently been postulated as being associated with the seizure onset zone. We hypothesize that early ictal increases in the power of high frequency activity (70-175 Hz) from multiple seizure events aligned to seizure onset, and statistically compared to the preictal interval, can provide an accurate tool for localizing the seizure onset zone for epilepsy surgery planning. We also investigate the efficiency of different referencing strategies in detecting localized increases in high frequency power. Finally, we show that early ictal increases in power of low frequency activity (0-40 Hz) are not limited to the seizure onset zone.Methods: Intracranial EEG recordings obtained during multiple seizures (3-86) from six patients with grid, depth, or combination of grid and depth electrodes, were aligned to seizure onset identified by clinicians, and analyzed with a matching pursuit (MP) method to identify power increases in high (70-175 Hz) and low (0-40 Hz) frequency components of ictal ECoG signals, as compared to 5 minute preictal intervals. For patients who had both grid and depth electrodes implanted, different references were used to identify montages detecting high frequency power changes with the highest sensitivity: common average, distant 'silent' electrode, or 'silent' electrode in close proximity to the seizure foci ('silent' grid electrode or 'silent' depth contact). Seizure foci were identified independently by visual inspection of the ECoG recordings by experienced epileptologists. Results of MP analyses were compared to the clinical evaluation.Results: Early ictal increases in the power of high frequency activity (70-175 Hz) were observed only for signals recorded from electrodes localized within the seizure onset zone identified by clinicians, but not for signals recorded outside of the onset zone. Both common average and 'silent' distant referencing enabled detection of the early ictal increases in the power of high frequency activity. Referencing to 'silent' electrode (grid electrode or depth contact) in close proximity to the seizure focus was often not sensitive enough to detect these changes. However, more patients are required to investigate these referencing strategies further. Early ictal increases in power of low frequency activity (0-40 Hz) were not limited to the seizure onset zone, and in some patients were observed in all electrodes.Conclusions: Detecting early ictal increases in the power of high frequency activity in multiple seizures aligned to seizure onset, and compared to preictal intervals, may be a useful measure for localizing the epileptogenic zone for epilepsy surgery planning. This measure can be used with both common average referencing and 'silent' distant referencing. This project was funded by NINDS R01 NS40596.