Abstracts

Rationale: Tuberous sclerosis complex (TSC) is an autosomal dominant multisystem disorder that causes intractable epilepsy in children. Multi-staged MRI-guided laser ablation in children with TSC has been an affective alternative treatment modality. However, determining the most active tuber conglomerates based on scalp EEG alone may be difficult. However, rapidly developing source localization techniques may improve the accuracy of target selection. Methods: We present a 5-year-old girl with generalized and complex partial epilepsy with spasms, hemispasms, head drops and staring episodes. Scalp EEG revealed hemispasms or spasms associated with generalized electrodecrements with right or left hemispheric predominance with a broad field involving frontal, temporal leads, staring episodes associated with left frontal polar, frontal, central, parietal, anterior and anterior inferior temporal or right frontal polar, frontal, central involvement. Despite corresponding tuber conglomerates in each of the above regions, determining the correct target tuber groups turned out to be more difficult than anticipated. Hence decision was made to analyze EEG data using source localization techniques. Results: Interictal spikes, spike waves, electrographic and clinical seizures were seen to arise, primarily from F7 and adjacent electrodes. An average of 188 such discharges was formed. The second most active site, both interictally and ictally was F4. An average was formed from 40 F4 spikes. Additional active regions included P3 and P7. The F7 spike average source analysis showed the primary source to be maximal in the anterior medial end of the left superior/anterior insula. A secondary source was observed in the inferior left posterior cingulate gyrus. The F4 analysis pointed to the primary source, maximal in the anterior right medial frontal cortex just posterior to an obvious right frontal tuber. The secondary source was located in the anterior‐inferior basal right temporal lobe. Both primary sources were seen to be in a network involving a portion of the limbic system; a well‐recognized pattern for epileptic networks. In the light of this new data, with corroborative PET-CT results, the patient underwent MRI-guided laser ablation of right frontal, left anterior medial and anterior superior insula, and left inferior posterior cingulate tuber conglomerates with subsequent control of spasms, hemispasms, head drops and staring episodes which preoperatively presented with more than 5 daily clusters for each semiology. Currently, 2 months post procedure the patient only has one or two brief myoclonic episodes in addition to improving behavioral and neurocognitive difficulties. Conclusions: Applying source localization techniques to scalp EEG may improve accuracy of target tuber conglomerate determination for minimally invasive MRI-guided laser ablation, and improve surgical outcomes.