Abstracts

Rationale: Up to 35% of epilepsy patients become medically refractory. In these patients resective surgery or neurostimulation devices are considered as potential treatment options. The aims of this study are to 1) Assess clinical characteristics and surgical outcomes in medically refractory epilepsy patients. 2) To correlate between pre-surgical imaging (MRI) and electrodiagnostic studies (scalp EEG, intracranial EEG, MEG) with surgical outcomes. In addition, we will further explore non-traditional imaging platforms to evaluate for structural abnormalities that may not be readily identifiable by current imaging techniques. Clinical and imaging markers of poor surgical outcomes will aid in clinical decision making and better counseling of patients. Methods: This is a retrospective single center study of patients who have undergone epilepsy surgery at a level IV epilepsy center. Patients >19 years of age who had a pre-surgical 3T MRI scan and underwent epilepsy surgery (resection or resection followed by neurostimulation device placement) between 2012 and 2016 were included. Patients were excluded if 2 year post-surgical follow up data was unavailable. Of the 96 patients who had epilepsy surgery, 49 met our inclusion and exclusion criteria. Demographics, clinical history, surgical history, electrophysiological and imaging data was recorded. A good outcome was defined as an Engel score I. A poor outcome was defined as an Engel score >/= II or if the patient required further surgery or neurostimulation device within 2 years of the initial surgery. Results: Overall outcome in our study cohort showed a good outcome in 51% (n=25) and a poor outcome in 49% (n= 24) of the patients. In patients with risk factors for epilepsy 54.8% had a poor outcome compared to those with no risk factors where 38.9% had a poor outcome (p= 0.28). Patients with greater seizure frequency prior to surgery tended to have higher Engel scores (p = 0.003). Of the patients who had a unilateral temporal lobe focus based on electrophysiology, 31.5% had a poor outcome, compared to those who had data suggestive of extra-temporal or bilateral foci where 60% had a poor outcome (p= 0.05). 58.8% of patients with discordance of localization of the suspected epileptogenic foci in electrodiagnostic and imaging studies with the surgical resection site had a poor outcome, compared to only 26.7% when data was concordant (p = 0.04). In patients with MRI showing unilateral mesial temporal sclerosis (MTS), 27.3% had a poor outcome, compared to non-lesional and other lesional (non MTS) patients where 42.1% and 68.4% had a poor outcome, respectively (p = 0.07). We will further analyze the imaging characteristics of non-lesional MRI patients using Freesurfer imaging software to evaluate for abnormalities in cortical thickness, cortical surface area, gray matter volume and mean curvature. Limitations of this study include that pre vs post-surgical neuropsychological assessments and quality of life indices were not considered as an outcome measure. Conclusions: In patients with medical refractory epilepsy undergoing epilepsy surgery, higher proportion of patients had a poor surgical outcome if they had the following factors: known risk factors, extra-temporal or bilateral seizure foci, non-MTS lesions on MRI, discordance of presurgical work up with resection site. We will further explore whether epilepsy patients with non-lesional MRIs have subtle cortical abnormalities which might be uncovered when analyzed using non-traditional imaging tools. Funding: No funding