Abstracts

Rationale: Precise identification of the epileptogenic zone (EZ) is a challenge in many patients with medically refractory focal epilepsy undergoing presurgical evaluation. Intracranial stereo-encephalography (SEEG) can be a useful investigation when non-invasive tools such as scalp video-EEG and MRI alone have not been definitive. The presence of a lesion on magnetic resonance imaging (MRI) guides planning intracranial EEG. However, a proportion of patients have non-lesional MRI and in these cases other imaging modalities such as 18F-fluorodeoxyglucose Positron Emission Tomography (FDG-PET) and Ictal Single Photon Emission Computed Tomography (Ictal SPECT) may be undertaken to guide planning and further decisions about surgery.Methods: In this study we reviewed all patients with non-lesional MRI who underwent SEEG at the National Hospital for Neurology and Neurosurgery, London, UK between November 2010 and January 2015. We analysed whether FDG-PET and ictal SPECT data correlated with SEEG results, as well as how this related to recommendation for surgery, postoperative outcome and pathology.Results: In total 45 patients underwent SEEG implantation, 17 of which had non-lesional MRI (12 male; mean age 38 years; range 19 to 52); average duration of epilepsy was 21 years. All non-lesional patients underwent FDG-PET scans; FDG-PET scans were normal in 6/17 (35%) and showed unilateral hypometabolism in 11/17(65%). Ictal SPECT was carried out in 4 patients, 3 of whom were FDG-PET negative. See Figure 1. In the 11 FDG-PET positive cases, hypometabolism was seen in frontal (n=4), frontotemporal (n=3), temporal (n=2) regions and posterior quadrant (n=2). A focal or regional EZ was identified with SEEG in 8 cases (73%) and correlated with FDG-PET findings in all these cases. Surgery was proposed in all cases and has been carried out in 6 cases. Pathology showed focal cortical dysplasia (n=2), hippocampal sclerosis (n=1) and non specific changes (n=3) and all had favourable seizure outcomes (ILAE Class 1 to 2) (mean 15 months; range 6 to 27). Of the 3 cases (27%) in whom a focal or regional EZ was not identified, a diffuse epileptogenic zone was found in 2, which overlapped with the area of hypometabolism and in the third, a bitemporal ictal onset was seen with SEEG compared with unilateral temporal hypometabolism on FDG-PET, therefore, surgery was not recommended. In the 6 FDG-PET negative cases, a focal or regional EZ was identified in 3 cases (50%) and ictal SPECT was concordant with EZ in two of these cases. In one of these cases, surgery was not recommended due to EZ being in an eloquent region. In the other 2 cases, surgery has been carried out with pathology showing hippocampal sclerosis and gliosis, one case with favourable seizure outcome (ILAE Class 1-2) at 6 months follow up.Conclusions: In patients with non-lesional MRI, undergoing surgical investigation with SEEG, hypometabolism FDG-PET can guide implantation and correlated well with the epileptogenic zone. Ictal SPECT may represent an additional tool.