Abstracts

The surgical techniques for the treatment of catastrophic hemispheric epilepsy have evolved since first introduced in the 1950[rsquo]s. Techniques for hemispherectomy include anatomical, functional, hemispherical deafferentation, peri-insular hemispherotomy, and hemidecortication. Despite a multiplicity of techniques, surgical outcome remains overall excellent across the variety of pathologies treated. Little has been described regarding the applicability and efficacy of individual surgical techniques applied to specific pathologies. We examined 87 patients undergoing functional and anatomical hemispherectomy to determine whether one technique was more successful than another in relation to the underlying pathology. A retrospective review was performed of 87 patients who underwent hemispherectomy for treatment of intractable epilepsy. All of the operations were performed by a single neurosurgeon during 1996-2003. Surgical techniques included functional, anatomical, and modified anatomical hemispherectomy. Surgical outcome and complications were described. The mean follow-up was 26.3 months (6-79 months). The mean age of the patients was 7.9 years (4months-55yrs). The mean body weight at the time of surgery was 27.9 kgs (6.9-94kgs). The mean operative blood loss was 330 ml. The mean duration of admission was 10 days. Morbidity included coagulopathy (34%), aseptic meningitis (25%), hydrocephalus (5%), and infection (3%). There was no mortality. Pathology included hemispheric dysplasia (n =30, 34.5%), hemimegalencephaly (n=14, 16.1%), perinatal stroke (n= 29, 33.3%), Rasmussen[rsquo]s encephalitis (n= 10, 11.5%), and Sturge Weber disease (n=3, 3.4%). Forty eight percent of the hemispheric dysplasia group (including hemimegalencephaly) underwent modified anatomical or anatomical hemispherectomy. Seventy-one percent of perinatal stroke group underwent functional hemispherectomy. The chance of freedom from seizures (Engel I) after optimal resection was 100% in Sturge-Weber disease, 80% in hemispheric dysplasia, 71% in perinatal stroke, 60% in Rasmussen[rsquo]s encephalitis, and 53% in hemimegalencephaly. Re-operation for persistent seizures after surgery was more common in the hemimegalencephaly (21%) and hemispheric dysplasia groups (13%). The basal frontal region was the most common area of incomplete disconnection. Hydrocephalus was also more common in the hemimegalencephaly group. Hemispherectomy is an effective and safe operation for the treatment of catastrophic hemispheric epilepsy. Functional hemispherectomy was an excellent technique for perinatal stroke and Sturge Weber disease. Our experience in Rasmussen[rsquo]s encephalitis has led to additional removal of the insular cortex. Our experience with malformative lesions (especially hemimegalencephaly) has led to abandonment of the functional operation in favor of anatomical removal of the abnormal hemisphere.