Abstracts

Rationale: Refractory epilepsy is frequently seen in adults with primary brain tumors. Lesionectomy alone may be inadequate in achieving seizure freedom. In addition to ictal EEG, quantitative interictal EEG analysis may help to further define the epileptogenic zone in patients with tumor-related epilepsy. Methods: Nine consecutive patients who underwent 2-stage surgery for resection of tumor and epileptogenic zone were studied. Chronic invasive EEG was recorded over 4-6 days using hippocampal depth electrodes and/or subdural grids (40 to 96 electrodes per patient). Two to four 10-minute interictal awake EEG samples taken ≥ 6 hours from the last clinical seizure were analyzed using automatic spike detection software (Stellate Systems) and reviewed by a single EEGer. Artifacts identified as spikes were discarded whereas spikes missed by the software were included. These corrected files were used to calculate average spike frequency for each electrode (n = 618). Each electrode was labeled by its ictal activity as “seizure onset” (electrographic involvement at the earliest point in seizure, n = 56); “seizure spread” (involvement of the electrode within 10 seconds of seizure onset, n = 85); or “neither” (n = 477). Similarly, electrode location was blindly identified as “tumoral” (within or overlying tumor, n = 154), “peritumoral” (adjacent to tumor, n = 104), or “non-tumoral” (n = 360) based on preoperative and post-implantation MRI and/or CT scans. For each patient, the mean (µ) and standard deviation (σ) of spike frequency at all electrode sites was calculated, and then used to transform the raw spike frequency to a z-score (Xi-µ/σ) at each electrode. This standardization process yielded 618 electrodes with a z-transformed spike frequency that can be treated independently. Data were analyzed with independent groups ANOVA, with spike frequency forming the dependent measure, and electrode location and ictal activity forming the independent measures. Significant results were further analyzed using Tukey’s pairwise comparison procedure. Results: There was an effect of electrode location (F(2,615) = 9.06, p < 0.0001) with “tumoral” electrodes showing greater interictal spiking than electrodes labeled as “neither”. Spike frequency at “peritumoral” electrodes fell midway between the two, but not significantly different from either. There was also an effect of ictal activity (F(2,615) = 18.78, p < 0.0001), with “seizure onset” and “seizure spread” electrodes showing greater interictal spike frequency than “neither” electrodes. Spiking at “seizure onset” electrodes was higher than at “seizure spread” without reaching statistical significance. Conclusions: Interictal spike frequency appears to be spatially related to tumor location. In addition, the interictal spiking correlates positively to the ictal onset and spread regions. Thus, interictal electrophysiology may assist in better defining the epileptogenic zone in patients with tumor-related refractory epilepsy.