Abstracts

Rationale: Functional neuroimaging, in the form of Positron Emission Tomography (PET), has proven to be a useful technique in providing supportive evidence for the localization of seizure onset zone, as part of presurgical workup. However, not all PET studies may be precise in determining the seizure onset zone. In order to increase the sensitivity of PET, we have studied several novel techniques to isolate the cortical elements of PET, to analyze the data quantitatively. This new computer algorithm lets us determine potential seizure onset zone with better precision. Methods: We have used a new software algorithm that uses a gray matter segmentation to extract the cortical portions of the co-registered PET. We also use reproducible cerebral parcels from a modified brain atlas to perform symmetry analysis of the cortical PET, that has enabled us to concentrate solely on the abnormally asymmetric cortical metabolism. One caveat of successful analysis is the determination of the electrophysiological state of the cortex within the first 30 minutes (uptake phase) of radiotracer injection. Determining whether the radiotracer uptake took place during a true inter-ictal, ictal or inter-ictal/ictal continuum has been an important factor for the accuracy of this method. Results: We have retrospectively analyzed the utility of this technique in 6 patients with cerebral cortical dysplasia, who underwent presurgical phase I assessment with PET studies that were performed in ictal or ictal/interictal continuum, or frankly ictal states.The patients who had ictal PET proved to be precisely localizing to the seizure onset zone with intense hypermetabolism in the epicenter, whereas the patients whose PET studies were performed during inter-ictal/ictal continuum had precise localization with subtle hypermetabolism. Conclusions: This case series clearly demonstrates the utility of our computer algorithm as a supportive technique for the identification of the seizure onset zone. An important point to take into account is the electrophysiological state of the patients (ictal versus interictal versus ictal/interictal continuum). As opposed to the common notion that PET has to be performed in interictal state, our data renders support to the potential that ictal PET could be an equally invaluable tool supporting the localization of the seizure onset zone. Additionally, subtle hypermetabolism in case with frequent “interictal” spikes carefully delineated using our algorithm can also be correctly localizing the seizure onset zone. Funding: No funding