Abstracts

Rationale: High frequency oscillations (HFO) called ripples(80-250Hz) and fast ripples (250-500Hz) can be recorded from intracranial EEG(IC-EEG) macroelectrodes in patients with intractable epilepsy. We analyze various correlations of interictal HFO and spike in order to understand their relationship to each other and to the epileptogenic zone. Methods: We studied 10 subjects with refractory localization-related epilepsy without clear lesion on neuroimaging studies who underwent 2-stage surgery with subdural grid/depth electrode placement(80 to 124 electrodes per patient). Three 10-minute interictal EEG samples (one sample per 24-hours, roughly at the same time of the day and during awake, resting phase) at least six hours from the last ictus were analyzed. Each file was reviewed and manually corrected by a single EEGer. Each electrode was labeled by its ictal activity as seizure onset (SO) (electrographic involvement at the earliest point in seizure), seizure spread (SS) (EEG involvement within 10 seconds of seizure onset), or neither (N). Seizure onset and spread are defined as epileptogenic zone. EEG was sampled at 200 or 1000Hz and HFO were recorded by using high pass filter at 50Hz. Automatic computer program analysis was performed in Matlab (MathWorks) to calculate the occurence frequency of HFOs and spikes per electrode. Two hundred milliseconds of data were searched on either side of a marked HFO to look for co-occurrence of spike. Analysis included comparing the frequency of HFO occurrence, spike frequency, and frequency of HFOs with concurrent spikes, to each electrode based on designation as ictal onset, spread, or neither. Statistical analyses included ranked data correlations, ANOVAs with subject as a blocking factor to reduce the error term, followed by post-hoc Tukey tests as appropriate. Alpha was set at p < .01. Results: Total average spike frequency per patient was calculated as 9,390 and total average frequency of HFO per patient was 13,073. Rank score correlation of frequency of spikes and HFO is 0.60. Correlation of frequency of HFO with concomitant spikes within /-200ms of a marked HFO was .69. The ANOVA was significant for spikes, with post-hoc tests showing onset zone was significantly different (higher) than spread or neither. HFO was significant with onset being different (higher) than neither. HFO with a spike was significant, with onset being higher than spread or neither. Conclusions: We hypothesize that electrode with high spiking frequency also had high HFO frequency at the seizure onset and spread zone. Correlation of HFO and concomitant spike was statistically significant therefore we assume that there is a good probability for some of the HFO to be associated with spikes. These preliminary results suggest that interictal spike frequency and frequency of HFO occurrence both can discriminate between the epileptogenic zone from the surrounding brain area.