Abstracts

More precise techniques, such as HDM-LD mapping of the hippocampus, may assist in detecting subtle abnormalities in MTS. We compare groups of patients with well-defined MTS with controls, to determine influences of normal hippocampal right/left asymmetries on results, and document the ability of HDM-LD-defined hippocampal changes to predict MTS. In this study, we objectively quantitate shape variations in patients with right and left mesial temporal epilepsy (MTLE) as compared to controls, using large deformation high dimensional mapping (HDM-LD) vector analysis. Subjects were identified retrospectively from consecutive cases from the epilepsy surgery series at Saint Louis University. All epilepsy subjects had post-surgical confirmation of MTS. Using a previously described technique, the right and left MTS groups were compared independently with the control group, resulting in eigenvector fields describing differences within the populations. This technique accounts for normal asymmetries of the right and left hippocampus, and results in a subset of eignevectors which maximally discriminate MTS groups from controls. A leave-on-out (jackknife) procedure comparing eigenvector-defined shape differences between MTS groups and controls was used to predict the side of MTS. The mean coefficient associated with the first nine shape eigenvectors for each group showed that the first three eigenvectors were large, and accounted for most of the differences between groups. The largest difference among the two MTS groups was found in the second eigenvector, while the MTS groups were rather similar in the first and third eigenvectors. This suggested that the laterality of the MTS was largely symmetric in the diseased side hippocampus and was characterized by eigenvector 2. When comparing the left MTS group with the controls, eigenvectors 1,2,3 were selected by a logistic regression procedure (Likelihood Ratio: 2=32.0, df=3, p[lt].0001). A leave-on-out procedure correctly predicted group classification in 14 out of 15 (93.3%) MTS subjects and in 14 out of 15 (93.3%) control subjects. When comparing the right MTS group with the controls, eigenvectors 1,2,3 were selected by a logistic regression procedure (Likelihood Ratio: 2=41.3, df=3, p[lt].0001). The leave-on-out procedure correctly predicted group classification in all 15 MTS and 15 control subjects. HDM-LD eigenvector analysis shows MTS affects the right and left hippocampi in a nearly identical pattern, after accounting for normal right/left hippocampal shape differences. Shape analysis also predicts the hippocampus which is affected by MTS.