Abstracts

RATIONALE: Resective surgery is accepted as standard care for some children with intractable localization-related epilepsy. When the epilepsy is non-lesional, identification of the ictal onset zone depends on functional imaging for directing phase II evaluations.This presentation reviews our experience with 19 subjects in whom we used PET for defining surgical candidacy.
METHODS: The surgical database of The Children[scquote]s Hospital Epilepsy Program, Denver, provided information regarding patients who had PET. Retrospective chart review was performed to define presentation and outcome of all patients evaluated with PET (performed at Children[scquote]s Hospital of Michigan) since June 1996.
RESULTS: Nineteen patients (mean age 13 years) evaluated at The Children[scquote]s Hospital, Denver, with phase I monitoring and routine imaging had PET scans because of a lack of clearly identified focal CNS lesion and/or localizing EEG to direct intracranial grid placement. Eight of 19 patients withdrew from further surgical evaluation, 5 because PET suggested diffuse or multifocal abnormalities or because of divergent data; 3 refused further evaluation. Eleven patients underwent Phase II monitoring. Resection was not performed in 5(1 had no seizures during prolonged monitoring at another institution; 3 because intracranial recordings failed to identify the zone of ictal-onset; 1 because of potential visual deficits if resection were performed). Seven had resective surgery, 1 without phase II (a child with tuberous sclerosis who underwent multilobar resection). Two patients had extra-temporal resections; 1 had multilobar resection; 4 had temporal lobectomies. Of the 7 who had resective surgery, 2 (29%) were not improved after surgery (both temporal lobectomies) while 5 (71%) are considered surgical successes: 1 (14%) is improved though has persistent seizures; 1 has had only very rare seizures; 3(42%) are seizure free.
CONCLUSIONS: PET provided important information for decision-making regarding surgical treatment of medically intractable epilepsy in our patients. 8/19 (42%) patients were excluded from further evaluation based on PET information. Of 11 patients who had Phase II monitoring after PET, 5 (45%) were not candidates for surgical resection. Seven had resections of whom 5 (71%) are considered surgical successes. These patients would not have undergone surgical resection without PET. For patients excluded from surgery and for patients improved by surgery, PET was useful and appropriate. However, for 5 patients who had phase II monitoring but no resection and for the 2 who were not improved after resective surgery, PET was not useful. PET aids in surgical planning in some but not all children with intractable partial epilepsy. Further refinement of non-invasive presurgical assessments is needed to optimize outcomes.