11 C]Flumazenil binding in the medial temporal lobe in patients with temporal lobe epilepsy: correlation with hippocampal MR volumetry, T2 relaxometry, and neuropathology

OBJECTIVE: To detect reduced [11C]flumazenil in patients with temporal lobe epilepsy (TLE) and to relate binding to histopathology. METHODS: The authors studied 16 patients who underwent epilepsy surgery because of drug-resistant TLE using [11C]flumazenil PET and quantitative MRI. In 12 patients, resected hippocampus was available for histologic analysis. [11C]Flumazenil binding potential (fitted BP) was assessed with the simplified reference tissue model. RESULTS: [11C]Flumazenil fitted BP in the medial temporal lobe was reduced in all patients with abnormal hippocampal volumetry or T2 relaxometry on MRI. Fitted BP was also reduced in 46% of the patients with hippocampal volume within the normal range and in 38% of patients with less than 2 SD T2 prolongation. In all MRI-negative/PET-positive patients, the histologic analysis verified hippocampal damage. Also, [11C]flumazenil fitted BP correlated with the severity of reduced hippocampal volume, T2 prolongation, and histologically assessed neuronal loss and astrogliosis. CONCLUSION: [11C]Flumazenil PET provides a useful tool for investigating the hippocampal damage in vivo even in patients with no remarkable hippocampal abnormalities on quantitative MRI.