Abstracts

Rationale: Hippocampal sclerosis (HS) is traditionally considered to be the cause of seizures in patients with medial temporal lobe epilepsy (MTLE). HS can be observed in routine brain MRI as hippocampal atrophy. However, some patients with unilateral MTLE exhibit mild contralateral hippocampal atrophy. Atrophy in the hippocampus with HS (the generator of seizures) is considered to be a consequence of cell death with formation of aberrant epileptogenic neuronal circuitry, whilst the contralateral hippocampus is atrophied possibly as a consequence of the excitotoxic effects of seizures without epileptogenic reorganization. The distinction between the epileptogenic hippocampal atrophy of HS (the generator ), versus the non-epileptogenic atrophy (the receiver) may enable the correct identification of the seizure onset site and lead to better surgical results. This study aimed to compare the tridimensional distribution of hippocampal atrophy in the ipsilateral and contralateral hippocampi of patients with MTLE. Methods: We studied 14 patients with MTLE (6 with right and 8 with left MTLE) with unilateral seizure onset documented by video-EEG corresponding to the side of hippocampal atrophy. All patients were submitted to unilateral hippocampal resection with hippocampal sclerosis confirmed by histology. Importantly, all patients were seizure free at least 12 months after surgery. We also studied a control group of 34 healthy individuals. T1-weighted images were submitted to voxel-based morphometry preprocessing and analysis of the hippocampus through cytoarchitectonic regions of interest (ROIs). The hippocampi of patients and controls were compared employing a non-parametric voxel-wise Wilcoxon test. Results were corrected for multiple comparisons with the application of an FDR corrected threshold for q<0.05. Results: Patients with left MTLE showed atrophy within both hippocampi, but the atrophy within the ipsilateral hippocampus (both for left and right MTLE) was more intense and more widely distributed (Figure 1). The anterior-inferior aspect of the head of the hippocampus was the site of more significant damaged in the ipsilateral hippocampus (Figures 1 and 2) (site of maximal atrophy: Left MTLE, left hippocampus x=-28, y=-16, z=-24, Z=3.6; Right MTLE, right hippocampus x=22, y=-11, z=-27, Z=2.9). On the contralateral hippocampus, the atrophy was more noticeable in the posterior head and body areas (Left MTLE, right hippocampus x=25, y=-17, z=-21, Z=2.1; Right MTLE, left hippocampus x=-27, y=-19, z=-17, Z=2.1). There were no areas of increased gray matter volume in patients compared with controls. Conclusions: In this study, we demonstrated that the hippocampal atrophy observed in the hippocampus that generates seizures (the generator ) has an anatomically distinct pattern compared to the atrophy of the contralateral hippocampus (the receiver ). This was demonstrated on patients who achieved seizure freedom with surgery, thereby eliminating the possibility that the contralateral hippocampus may also be a site of epileptogenesis.