Abstracts

Rationale: Hemispherectomy is an effective treatment for intractable epilepsy arising from one hemisphere. However, not all patients become seizure free post operatively. A number of factors including incomplete hemispheric disconnection or diffuse bilateral epileptogenesis may contribute to failed surgery. Our objective was to analyze the significance of scalp VEEG findings prior to anatomic hemispherectomy (AH) in children who had failed functional hemispherectomy (FH) for intractable epilepsyMethods: We identified 18 patients who underwent anatomic hemispherectomy (AH) from a subset of 145 patients (12.4%) who had functional hemispherectomy (FH) during 1997 - 2006. We reviewed their scalp video-EEG (VEEG) data, analyzing clinical seizure semiology, interictal and ictal EEG as well as clinical data, MRI and PET. In 10 patients we had both pre and post FH data sets. In 8 patients only post FH data was available as FH was performed elsewhereResults: Out of 18 children 9 were male and 9/18 underwent a right hemispherectomy. Age at initial presentation ranged from birth to 72 months (median 2.5 months). The age at FH ranged from 3 months to 17 years (median 18 months) and from 5 months to 20 years (median 63 months) for AH. The etiologies were divided into two groups: 13/18 had developmental lesions (malformations of cortical development, hemimegalencephaly) and 5/18 had acquired lesions (Rasmussen’s encephalitis and stroke). Post AH follow up ranged from 1 month to 24 months (median 6 months). VEEG of 10 patients with malformations of cortical development (5 hemimegalencephaly) showed unilateral interictal and ictal findings confirming that the residual seizures were still arising from the operated hemisphere and 7/10 (70%) became seizure free post AH. The remaining three had a worthwhile seizure reduction. Two cases with Rasmussen Syndrome had contralateral seizure onset, yet became seizure free after AH. A third patient with Rasmussen Syndrome had no interictal or ictal changes on EEG but semiology was lateralizing. After AH, seizure control improved considerably but did not become seizure free. 5 patients showed varying degrees of bihemispheric interictal and ictal findings, predominantly on the lesion side which guided the decision for AH. Of 3 children with malformations, 2 became seizure free and one improved. However, 2 patients with stroke failed AH. Conclusions: Detailed VEEG analysis after failing FH can help to confirm residual epileptogenicity in the operated hemisphere and support patient selection for reoperation. Repeat AH offered a good prognosis for seizure control in patients with persistent ipsilateral seizure onset and in some patients with bilateral EEG findings. Contralateral ictal EEG onset in patients with Rasmussen Syndrome may not be a contraindication for AH. Stroke patients with bilateral interictal and ictal EEG findings appeared to do less well after AH