PURPOSE: To investigate the effects of topiramate on cerebral glucose metabolism, we performed 18F-fluorodeoxy glucose positron emission tomography (FDG-PET) in patients with new-onset epilepsy. METHODS: Thirteen patients with new-onset epilepsy or without medication after epilepsy diagnosis were included. Pre- and post-drug FDG-PET were performed (before and after topiramate administration) in all subjects (M/F=9/4, 28.2+/-11.4 years). For SPM analysis, paired pre- and linearly transformed post-drug FDG-PETs were spatially normalized into a standard PET template, provided in SPM-99, using a 12-parameter affine and a non-linear transformation. Spatially normalized images were then smoothed by convolution using an isotopic Gaussian kernel with a 14 mm full width at half maximum. The paired t-test was used to compare pre- and post-drug PET images. RESULTS: Mean dose of topiramate at the time of post-drug FDG-PET scanning was 163+/-71 mg. Mean duration of topiramate administration was 169+/-54 days. Responses to topiramate medication were seizure free in 7, reduced seizures in 3, and no changes in 3 patients. Reported adverse events were headache in 2 patients. SPM analysis between post-drug and pre-drug FDG-PET images showed post-drug hypometabolism in the white matters of both parietal and right temporal lobes, and corpus callosum, both thalami, right cingulate gyrus, left midbrain, both superior frontal gyri, left middle frontal gyrus, right inferior- and left superior parietal lobules, and left inferior temporal gyrus (corrected p<0.05). No brain region showed post-drug hypermetabolism. CONCLUSION: Topiramate reduced glucose metabolism more in the corpus callosum, thalamus and white matters, and less in the cerebral cortex.