Abstracts

Rationale: Only 2 studies document the efficacy and complication of hemispherectomies in infants. Methods: This is a chart review of 11 infants with hemispherectomies from 2003 to 2010 (7 boys, 4 girls). All had presurgical evaluation including brain MRI and EEG monitoring. Preoperative data included development, age at surgery and seizure onset, seizure type, treatments tried prior to surgery, EEG, MRI and functional neuroimaging results. Perioperative data included blood loss during surgery and hospital length of stay. Postoperative data included surgical complications, pathology, and seizure frequency after 6 months, 1 year, 2 years and 4 years. Results: Seizures started on the first day of life for 2, in the first 2 weeks of life for 6, and within the first 6 months for 3. Five children had complex partial seizures (CPS), 2 had infantile spasms (IS) and 4 had both; all had EEG with onset from one hemisphere. Ten infants tried at least 2 medications prior to surgery except one with hemimegalencephaly and status epilepticus severe enough to warrant surgery at 2 weeks old. One patient had a trial of the ketogenic diet. On MRI, there were 4 with cortical dysplasia (CD), 1 with Sturge Weber, 2 with hemimegalencephaly (1 also with linear nevus syndrome), 1 with encephalomalacia from hemorrhage and 1 with a normal MRI. In this patient, a follow-up MRI showed hyperintensity in the basal ganglia and cerebellum consistent with vigabatrin toxicity and an electrocorticography demonstrated spikes over the entire hemisphere. Four had interictal PET scans with 3 showing hypometabolism on the affected side; one had an ictal PET with hypermetabolism over the affected side and in the peri-insular cortex on the opposite hemisphere. All patients had a functional hemispherectomy (5=right, 6=left). One patient had a 2 stage hemispherectomy performed first at 1 week of age then completed at 2 weeks of age due to autonomic instability. Age at surgery ranged from 14 days to 12 months of age. Hospital length of stay, estimated blood loss and fluid resuscitation were similar compared to older patients. Pathology revealed 9 with CD, one with gliosis and one with findings consistent with Sturge Weber. Follow-up ranged from 2 months to 5 years (mean 21.9 months). Six patients were seizure free, 3 with isolated seizures within the first 2 months of surgery and 1 with a 90% improvement in seizures. One had 50% improvement in seizures. Four had normal language development. No mortality occurred. All patients had hemiparesis and visual field cut. Four patients required a ventriculoperitoneal shunt (VPS) due to increased intracranial pressure. One had growth failure later. Conclusions: 1)No significant morbidity or mortality was noted. 36% required VPS. 2)Seizure freedom occurred in 55% with the rest benefiting significantly from surgery. 3)81% had CD. This may represent the importance of malformations as a primary etiology in this young population. 4)55% had IS demonstrating a predisposition for this catastrophic seizure type in this population. Further studies are needed to confirm these findings.