Abstracts

Rationale: Hemispherotomy (HP) has been performed successfully in treatment of epilepsy, with similar prognosis but less morbidity than hemispherectomy. The optimal timing for surgery tends to be retarded since it is still considered the most radical surgical procedure, resulting in definitive deficits. This study reports pre-, intra- and post-operative (PO) data to better evaluate morbidity and functional consequences of HP. Methods: A retrospective study included 12 children (8 boys, age 1.3-13.5 years) with unilateral hemispheric diseases and pharmacoresistant seizures, who underwent HP at Hospital S o Paulo between 2003-2010. All had been assessed using a standard protocol involving clinical, neurophysiological and neuroradiological evaluation. Clinical information was selected to survey presurgical and operative variables, including PO seizure control and pre- and post-surgery language and motor functions. Functional level of each child was evaluated through Gross Motor Function Classification System (GMFCS) and Manual Ability Classification System (MACS), which classify the child s movement and manual abilities, respectively. Data were analyzed using a statistical program (SPSS for Windows ); level of significance was p < 0.05. Results: The underlying pathology was Rasmussen encephalitis in 6 patients (50%), ischemic insult in 5 (42%) and hemimegalencephaly in 1 (8%). Left hemisphere was involved in 9 (75%). Seizure onset ranged from 0.2 to 7.4 years (3.4), lasting from 0.1 to 11 years (mean 2.4). The mean duration of surgical procedure was 9.8 hours (8-15); of orotracheal intubation, 26.4 hours (10-48); of ICU and hospital stay, 2.5 (1-5) and 10.8 (5-18), respectively. One child presented intraoperative and all patients had PO mild or moderate reversible complications, such as hypertension, bleeding, ventriculitis, insipidus diabetes, atelectasis, trigeminal neuralgia, seizures and fever. The PO follow-up period ranged from 0.2 to 7 years (mean 2.4). Five/12 were classified in Engel s classes I or II at 2 years and 8 (67%) in Engel I, 2 Engel II and 2 Engel III at the last follow-up. In the latter two patients epilepsy duration was longer than 10 years (p=0.02). Prior to surgery, all children exhibited hemiparesis (from I to IV); 9 were functionally independent being 4 (33%) GMFCS I and 5 (42%), GMFCS II. Nine of these children were GMFCS II at PO period. Three were III and V prior to surgery and remained so afterwards; these patients presented the earliest seizure onsets (0.2 and 0.5 years). Before surgery, fine finger movement was preserved in 6 children; 5 who were MACS I turned to II at PO evaluation, with maintenance of functional manual ability. Of the 9 children with language skills, they remained unchanged or even improved in all, including 6 (67%) who had left-side resection. Four (44%) of these children were older than 6 years at surgery. Conclusions: HP was a well tolerated procedure, with mild to moderate and reversible complications, effective seizure control and no motor or language strong functional impact. These consequences might help in the process of decision making.