Combined measurements of hippocampal N-acetyl-aspartate and T-2 relaxationtime in the evaluation of mesial temporal lobe epilepsy: Correlation with clinical severity and memory performances

Purpose: In this study we tried to find a correlation between the clinical severity and memory performances, by comparing proton magnetic resonance (M R) spectroscopy and T-2 relaxation time measurements in the hippocampi, in a homogeneous group of 27 patients with unilateral mesial temporal robe epi lepsy with ipsilateral hippocampal sclerosis on MR imaging, with a view to answer the following questions: (a) how sensitive is this approach for the assessment of the apparently normal contralateral hippocampus, (b) do the r esults relate to the clinical severity, and (c) does it allow evaluation of the degree of hippocampal dysfunction. Methods: Volume-selective proton MR spectroscopy of the head of both hippoc ampi was performed at 3 T-2 by using the PRESS sequence, with an echo time of 135 ms, to estimate NAA/(Cho + Cr) ratios. The relaxation times were mea sured at 0.28 T, by using a conventional Carr-Purcell-Meiboom-Gill sequence , with a repetition time of 2,000 ms, an echo time of 15 ms, and 48 echoes. Results: The combination of NAA/(Cho + Cr) ratio and T-2 relaxation time va lues was allowed to classify contralateral hippocampus abnormalities in two groups: first, decreased NAA/(Cho + Cr) ratio with strongly increased T-2 relaxation time values corresponding to abnormalities observed in sclerotic ipsilateral hippocampi; and second, decreased NAA/(Cho + Cr) ratio with no rmal or slightly increased T-2 relaxation time values. Whereas the NAA/(Cho + Cr) ratio or T-2 relaxation time value alone was not correlated with mem ory performances, their association shows that left hippocampal injury eval uated both by NAA and T-2 relaxation time measurements was clearly correlat ed with verbal memory scores, and right hippocampal injury, with visual mem ory scores. On the other hand, the maximal seizure frequency reported by th e patients was correlated with ipsilateral NAA/(Cho + Cr) ratio and T-2 rel axation time values but not with contralateral results. Conclusions: We showed that the combination of NAA and T-2 relaxation time measurements can be used to examine the degree of ipsi- and contralateral h ippocampal dysfunction or injuries and their relations with memory performa nces in the presurgical evaluation of patients.