Quantitative evaluation of hippocampal gray-white matter boundary blurring in medial temporal lobe epilepsy with hippocampal sclerosis

Abstract

IntroductionThe magnetic resonance imaging (MRI) findings of hippocampal sclerosis (HS) include decreased volume, increased signal intensity, and hippocampal gray-white matter boundary blurring (HGWBB). Given that the layered structure is obscure in HS, there have been no reports on the quantitative evaluation of HGWBB and its relationship with the clinical outcome. Thus, this study aims to correlate the extent of HGWBB to its clinical manifestation of HS. MethodsFifty-four patients with temporal lobe epilepsy who underwent hippocampal resection were enrolled. To evaluate HGWBB quantitatively, we defined an index by calculating the standard deviation of the intrahippocampal signal on short tau inversion recovery. In addition, we created a prognostic scoring system using four criteria, including hippocampal signal intensity, size of hippocampal cross-sectional area, presence of temporal lobe lesions, and the HGWBB index. ResultsThe HGWBB index was significantly lower on the affected side than on the unaffected side (p < 0.001). This trend was more prominent in the poor prognosis group than that in the good prognosis group. The prognostic scoring system revealed that when three or more criteria were positive, the prognostic accuracy reached 87.5% sensitivity and 71.7% specificity. ConclusionThe HGWBB index is useful for the diagnosis of temporal lobe epilepsy with HS and for predicting seizure outcomes when used with another index of hippocampal volume loss and increased signal intensity.