We sought to analyze our experience with pediatric epilepsy surgery patients to determine the place of [18F]fluorodeoxyglucose (FDG) positron emission tomography (FDG-PET) in the preoperative evaluation of such children relative to chronic invasive intracranial monitoring. Fifty-six children who received an interictal FDG-PET as part of a phase 1 epilepsy surgery evaluation were compared with 44 children who did not have this study in a retrospective analysis of 100 patients accrued over a 4-year period. There was no significant difference between the two groups of children in terms of age or follow-up or was there a significant difference between the FDG-PET group and the no-FDG-PET group in regard to the numbers of children who had surgery, the type of procedure done, whether chronic invasive intracranial monitoring was performed, or outcome. The hypometabolic area demonstrated on interictal FDG-PET was concordant with that of the epileptogenic zone as mapped out with ictal recordings from subdural electrodes in 2 of 13 patients in whom a complete data set was available for comparison. In the other 11 children there was either poor agreement between interictal FDG-PET and ictal electrocorticographic data or the interictal FDG-PET was normal in the face of an epileptogenic focus which was successfully mapped by invasive electrophysiologic techniques and excised. We conclude that one cannot exclude a child with intractable partial seizures from surgical consideration because the interictal FDG-PET is normal; nor is there sufficient correlation between the interictal hypometabolic area on FDG-PET and the epileptogenic zone in terms of anatomic location and size to justify forgoing chronic invasive intracranial monitoring in children with intractable partial seizures being evaluated for epilepsy surgery unless there is absolute concordance between all neuroimaging, clinical, and video-electroencephalographic data.