Predictive Value of Electrocorticography in Epilepsy Patients with Unilateral Mesial Temporal Sclerosis Undergoing Selective Amygdalohippocampectomy

Objective: We have attempted to evaluate the predictive significance of intraoperative electro-corticography (ECoG) in patients with unilateral mesial temporal sclerosis undergoing transsylvian selective amygdalohippocampectomy (sAHE). We presumed that ECoG-spiking restricted to the mesial structures supports the concept of a mesial temporal epileptogenic zone (EZ) whereas spikes recorded from the lateral cortex suggest a more extensive EZ which may not be appropriately treated by an sAHE. Methods: ECoG was recorded before and after resection in 22 patients with medication-resistant epilepsy and mesial temporal sclerosis without other structural lesions diagnosed by MRI and video-EEG monitoring. A standardized sAHE was performed, regardless of the ECoG findings. The follow-up was 12 months (n=20) or 6 months (n=2). The ECoG findings recorded from the mesiobasal temporal lobe (MTL) and the lateral temporal lobe (LTL) before and after sAHE were correlated with seizure outcome [Group 1: completely seizure free (Engel class Ia), group 2: not completely seizure free]. The main hypothesis which was tested using Fisher's exact test was: „Patients with pre-resection spikes restricted to the MTL have a higher seizure-free rate than others (with LTL, MTL+LTL or no spikes)“. Results: Fifteen of 22 patients (68%) remained seizure free (Group 1). Of the remaining seven patients (Group 2) four were Engel class Ib or Ic, one was IIa and two were IIIa. Patients with pre-resection spikes restricted to the MTL (n=11) more frequently remained seizure free (82%) as compared to the rest of the patients (55%). This difference was not statistically significant (P=0.361). Two of 4 patients with LTL-spikes and 4 of 7 patients without pre-resection spikes were in group 1. The frequency of pre-resection spikes was not related to seizure outcomes. No correlation between the seizure outcome and the presence of persistent spikes recorded on postoperative ECoG was found. Conclusion: Pre-resection ECoG may be helpful in the prediction of seizure outcome in patients undergoing sAHE for mesial temporal lobe epilepsy. Based on the data of this pilot study, a minimum n of 102 is needed to prove a 27% difference in outcome with a power of 80% using a two-sided test with a significance level of 0.05. In order to prove a 20% difference in outcome and assuming a less equal distribution into the ECoG-result groups, 220 patients would be necessary. Therefore, a consecutive prospective study would demand a multi-center design.