Teasing out the anatomy of mesial temporal lobe epilepsy.

Commentary While most patients receiving neurosurgical treatment for medically intractable temporal lobe epilepsy have hippocam-pal atrophy (HA) on MRI, there is another subgroup with mesial temporal epilepsy with a normal appearing hippocampus, with different clinical features and surgical outcome (1). Both of these subtypes of mesial temporal epilepsy are associated with widespread atrophy outside the hippocampus, particularly in bilateral neocortices (2–6). What causes this extrahip-pocampal abnormality? What is its distribution? How does this distribution vary among individuals, and does it differ between the subgroups of mesial temporal epilepsy patients with and without HA? Does the degree of extrahippocampal abnormality predict surgical outcome? This work by Bernhardt et al. joins a growing body of investigation that is starting to address some of these questions. Extrahippocampal abnormalities in mesial temporal epilepsy have been documented by MR spectroscopy and morphometry. MR spectroscopy measures N-acetyl-aspartate, which is typically reduced in the epilepsy focus and reflects neuronal loss. Mueller et al. reported that N-acetyl-aspartate was reduced in a bilateral frontotemporal distribution in HA patients but was more diffusely decreased in those without HA (6), with considerable variation among individual patients. MRI morphometric analysis demonstrated bilateral atrophy in the amygdala, and in the entorhinal, cingulate, temporo-polar, lateral temporal and frontocentral cortices, as well as in the thalamus, in some patients with HA (2–7). Although past reports of MRI analysis in patients without HA have shown some inconsistency as to the distribution of extrahippocampal atrophy (4, 6), the present study by Bernhardt et al. shows a pattern of cortical thinning very similar to the HA group. Key findings that shed light on the significance of cortical thinning in mesial temporal epilepsy are that it is progressive and that it has a robust association with seizure control (8, 9). In mesial temporal epilepsy, with or without HA, serial MRIs over a mean interval of 2.5 years demonstrated progressive atrophy in the mesial and superior lateral frontal and parietal cortices, with more rapid worsening in patients with more than a 14-year history of epilepsy, and greater thinning in those with a higher seizure frequency (8). Serial MRIs in a control group documented that these changes could not be attributed to normal aging (8). Other investigators independently showed that MRIs with a median interval of 39 months showed progressive white and gray matter atrophy, especially in patients with a longer history of epilepsy, with more intense progression with a higher …