Abstracts

Rationale: HMS in adults frequently arise from the frontal lobe (FL), although the insular region, temporal (TL) and parietal lobes (PL) have been reported. HMS data in children is limited and it is unclear whether the HMS mainly arise from FL. Methods: We retrospectively reviewed medical records of 317 patients with medically intractable epilepsy (MIE) who underwent RES between January 2009 and December 2015 at the Children's Hospital Colorado. All patients were younger than 20 years old at the time of RES. Patients with HMS were selected from chart reviews and then confirmed by video EEG (vEEG). Presurgical workups including vEEG, MRI, FDG-PET, ictal and interictal SPECT, MEG, intraoperative electrocorticography (ECoG), invasive vEEG and surgical outcome using the Engel's classification (EC) were reviewed. Results: We identified 18(5.6%) children with HMS. Eight were females. The mean age at seizure onset and at RES was 5.3 years and 12.8 years respectively. The epileptogenic zone (EZ) was located in TL in 10 (55.5%), FL in 5 (27.7%), occipital lobe (OL) in 2 (11%), and left hemisphere (LH) in 1 patients (5.5 %). The scalp EEG showed TL slowing with ipsilateral epileptiform discharges (ipEDs) in 10 cases, FL slowing with ipEDs in 3, bi-frontal EDs in 2, LH slowing with ipEDs in 1 and left OL slowing with ipEDs in 2 patients. Subdural EEG showed EZ in the left T-O in 2, right FL in 2 and 1 without seizures. Stereo-EEG showed EZ in the contralateral TL in 1. MRIs showed lesions in 16 and were negative in 2. FDG-PET was congruent in localization with MRI in 13, in lateralization in 1 and non-congruent in 2 of 16. Ictal SPECT was performed in 4 and was congruent with EZ in 2. MEG was performed in 2 and was congruent with EZ in 1. Pathology was available in 17 patients. Normal pathology was found in 1 and abnormal in 16 of 17. Focal cortical dysplasia (FCD) was found in 15 patients (1a 7, 1b 1, 2b in 2, 3b in 4 and mMCD in 1). Gliosis was found in 1 patient. Anterior temporal lobectomy was performed in 8, frontal lobectomy in 5, temporo-occipital resection in 2, occipital lobectomy in 1, hemispherectomy in 1 and laser ablation in 1 patient. At 1-year postsurgical follow-up, EC 1 outcome was found in 15, EC 2 in 2 and EC 4 in 1. At 2-year follow-up, EC 1 was found in 10, EC 2 in 2, and EC 4 in 1. Conclusions: HMS were seen in 5.6% of patients who underwent RES with 55.5% arising from TL, 27.7% from FL, 11% from OL and 5.5% from LH. No definite scalp or invasive EEG markers for HMS were found. In children, HMS arising from extra-FL, especially in TL is significantly more common than in adults and has been under-recognized. Since seizure semiology is essential in localizing/lateralizing the EZ, planning of intracranial EEG electrode placement and surgery, we suggest that TL should be considered as a potential EZ in all patients with HMS undergoing RES. This is also a first report of HSM arising from OL, therefore, OL should not be eliminated from intracranial electrode coverage in patients with HMS. Funding: NONE