Abstracts

Rationale:
Interictal Epileptiform Discharges (IEDs) are regarded as one of the most well-established biomarkers of epilepsy. Yet, they occur in large areas of the brain whose resection may be unnecessary for seizure freedom. It is critical to distinguish between more and less pathological IEDs to delineate the epileptogenic zone (EZ) from interictal recordings without waiting for a seizure to occur. High-frequency oscillations (HFOs) occur either in isolation or superimpose on IEDs. It is currently unknown whether the IEDs overlap with HFOs would facilitate the distinction between more and less pathological IEDs. Our goal is to assess whether IEDs overlapping on ripples and/or fast ripples are more specific biomarkers of the EZ compared to IEDs with no HFOs in children with medically refractory epilepsy (MRE).
Method:
We examined 20 children with MRE who underwent surgery. We dichotomized surgical outcome into seizure-free (Engel 1; 11 patients) and non-seizure-free (Engel≥2; 9 patients). On intracranial EEG (icEEG) recordings, we identified interictal spikes (1-70 Hz), ripples (80-250 Hz), and fast-ripples (250-500 Hz) using an automated algorithm based on RippleLab. All ripples and fast ripples were manually inspected by two independent readers to distinguish real HFOs from filtering artifacts. We classified the identified events into seven categories: spikes only, ripples only, fast ripples only, spikes with ripples, spikes with fast ripples, ripples with fast ripples, spikes with ripples + fast ripples (Fig. 1). For each patient, we estimated the number of these events per category in the entire recording. Using Wilcoxon signed-rank test, we tested differences between electrodes inside and outside the SOZ defined by the epileptologists. We also compared the number of events in resected and non-resected areas for seizure-free (good outcome) and non-seizure-free (poor outcome) patients.
Results:
In total, we identified 661,712 IEDs only, 15,283 ripples only, 10 fast ripples only, 55,022 IEDs with ripples, 4,720 IEDs with fast ripples, and 10,060 IEDs with ripples + fast ripples. We observed a higher number of IEDs with ripples in electrodes located inside the SOZ (mean: 48.51±61.82) compared to outside (mean: 26.74±43.59) for all patients (p=0.0079; Fig. 2a). We also observed a higher number of IEDs with ripples + fast ripples inside the SOZ (mean: 11.74±37.17) compared to outside (mean: 4.87±14.73) for all patients (p=0.0078). No differences were observed for the other categories. Finally, we show a higher number of IEDs with ripples in resected (54.58±69.86) versus non-resected areas (25.97±20.49) for patients with good outcome (p=0.0371; Fig. 2b); this differences was not observed for patients with poor outcome (p=0.1484; Fig. 2c).
Conclusion:
Our study shows that IEDs overlapping on HFOs (ripples and fast ripples) are specific biomarkers of the EZ compared to IEDs or ripples/fast ripples which occur in isolation. Such interictal biomarkers can help discriminate the EZ from areas of spread without having to wait for a seizure to occur and guide the EZ resection improving the surgical outcome of patients.
Funding:
:RO1NS104116-01A1 by NINDS R21NS101373-01A1 by NINDS