Abstracts

Rationale: It has been shown that intra-operative electrocorticography (ECoG) can be used to track post-resection outcomes in patients with Temporal Lobe Epilepsy (TLE). However, prognostic value of intra-operative ECoG has been a controversial topic. Importantly, ECoG offers the benefit of delimiting the epileptogenic zone in cases with where MRI, PET, and SPECT scans are non-lesional. Thus, intra-operative ECoG carries the potential for more accurate targeting of epileptic brain tissue, which in turn could improve seizure outcomes following resection. Here we investigate the prognostic value of post-resection intra-operative ECoG on outcomes on a larger scale than any articles published to date on the same topic. Methods: A comprehensive epilepsy database compiled at our institution over the past 15 years was used to isolate those cases that underwent electrocorticography with resection, and in whom follow-up data were available for at least 6 months. Of the total of 444 patients operated on in this interval, 430 met these criteria. In all 444 cases, after cortical exposure but prior to resection, subdural strip electrodes were used in multiple locations around the proposed resection site, to obtain baseline intra-operative ECoG activity that was then compared with post-resection ECoG. Post resection ECoG was recorded for between 4 and 10 minutes in each case. The ECoG results were classified on a scale of 0 (no spikes), 1 (sparse spikes), and 2 (frequent spikes). Post-surgical outcomes were measured using the Engel and ILAE classifications at 6 months and at most recent follow-up. Results: The surgical techniques included temporal resections (n = 193) and extra-temporal resections (n = 203), and a combination of temporal and extra-temporal resections (n = 48). Results from ECoG data and Engel Classifications were designated into four groups: ECoG value of 0 with an Engel classification of 1 or 2, ECoG of 0 with an Engel of 3 or 4, ECoG of 1 or 2 with an Engel 1 or 2, and ECoG of 1 or 2 with an Engel 3 or 4. A Chi-squared analysis of the 4 groups demonstrated a relationship of statistical significance (p = 0.0076). Accordingly, similar group structures were created for the ILAE Classifications, and Chi-squared analysis again demonstrated a relationship of statistical significance (p = 0.0342). We found having a post-resection ECoG classification of 0 had a predictive value of 0.75 for the patient to have a follow-up Engel score of 1-2. Conversely, an ECoG value of 1-2 had a low predictive value (0.35) on the patient having a follow-up Engel score of 3-4. Also, further analysis illustrated a direct correlation between the recorded ECoG value (0,1, or 2) and the number of failed medications before the resection (R2 = 0.979). Ongoing analysis is currently being conducted to determine the relationship between the ECoG and etiology of the epilepsy (lesional or not) and temporal vs extra temporal localization, relative to the surgical outcome. Conclusions: This is the largest study ever that uses intra-operative ECoG to determine completeness of resection. We demonstrate the ability of ECoG measurements to be used to guide the reduction. Further, the presence of spikes post resection relate to the likelihood of seizure recurrence and this could be used to reduce the number of anticonvulsants use in these patients if they show favorable outcomes after epilepsy surgery. Overall, ECoG can potentially be a vital tool in the optimization of surgical management of intractable epilepsy. Funding: No funding