HIPPOCAMPAL ATROPHY IS NOT A MAJOR DETERMINANT OF REGIONAL HYPOMETABOLISM IN TEMPORAL-LOBE EPILEPSY

Purpose: The pathophysiologic basis for the [F-18]fluorodeoxyglucose p ositron-emission tomography (FDG-PET) temporal lobe hypometabolism in patients with hippocampal sclerosis (HS) is uncertain. We tested the h ypothesis that hippocampal atrophy, which is strongly correlated with hippocampal cell loss, is largely responsible for the regional hypomet abolism in HS. Methods: Regions of interest (ROIs) on FDG-PET scanning were determined in the medial, lateral, and posterior temporal lobe, thalamus, and basal ganglia. A right/left asymmetry index for each ROI was calculated. These results were correlated with hippocampal magnet ic resonance imaging (MRI) volume ratios. Results: There was no correl ation between the magnitudes of the FDG-PET asymmetry index and the MR I volume ratio for the mesial or lateral temporal regions (r = - 0.09, r = -0.04). When the right/left asymmetry index was compared with the right/left hippocampal volume ratio, correlations for the mesial temp oral ROI (r = 0.79, p < 0.0001) and lateral temporal ROI (r = 0.57, p < 0.0005) were found. These, however, simply indicated that both tests accurately reflect the side of the epileptogenic region. The concorda nce of the side of relative hypometabolism of the FDG-PET with the sid e of the hippocampal atrophy was higher for the mesial temporal region (100%) than for the lateral (77.5%). Conclusions: The lack of correla tion between the magnitudes of the ratios argues against hippocampal a trophy and cell loss having a central role in the FDG-PET temporal hyp ometabolism.