Abstracts

Rationale: Epilepsy is the most common manifestation of tuberous sclerosis complex (TSC), and is often refractory to medical treatment. Epilepsy surgery is often denied when epileptogenic tubers are located in the vicinity of primary cortices. Recent radiological studies suggest that widespread white matter abnormalities extend to tuber-free regions. It is unknown if tuber localization and white matter abnormalities affect primary cortex localization. Methods: Seven TSC patients with cortical tubers seen on MRI and 7 control subjects were included in the study. Auditory, motor, somatosensory, and visual evoked fields were acquired using a 306-channel whole-head MEG (VectorView, Elekta Neuromag), and were co-registered with isotropic MRIs (1.5 Tesla, 1 mm voxel) using 3D digitizer (Polhemus). A single equivalent current dipole (ECD) in a spherical head model was calculated and mapped on the co-registered MRI. Expected localization of primary cortices (ELPC) according to the known relationship to anatomical landmarks was mapped on subject’s MRIs. Localization of evoked fields (EF) was considered normal if it fell within a 2 cm radius surrounding ELPC. Tuber localization was defined on FLAIR sequence MRIs of TSC patients. Results: Normal subjects: Five of the normal subjects had the anticipated localization for all EFs. Two normal subjects had magnetic artifacts that obscured activity in the temporal regions, probably due to metallic dental fillings. TSC patients: 2/6 TSC patients had abnormal localization of at least one EF. The remaining 4 patients had EFs with a normal localization. While being within the 2 cm radius, the mean distance from EFs to the ELPC was longer in patients compared with the normal controls. None of the patients had a tuber localized in ELPCs. EPs with abnormal localization were shifted towards the closest tuber. One TSC patient could not be analyzed due to a prominent cardiac artifact (metallic clip from previous surgery). Conclusions: Our findings suggest that primary cortices could be shifted in TSC patients in the absence of obvious disruption of the anatomic area were the primary cortex is expected to be located. The fact that EFs appear to be shifted towards the closest tuber can be related to co-activation of tubers during activation of primary cortices. In this case, a single dipole will be fitted to an area in between the 2 sources. Another possible explanation is that brain abnormalities in TSC patients extend beyond tubers that can be visualized on MRIs and that brain organization is abnormal with occasional shifting of primary cortices. Special attention to pre-surgical localization of eloquent cortices in TSC patients is advised.