1HMRSI SHOWS DIFFERENT PATTERNS OF METABOLIC HIPPOCAMPAL DAMAGE IN TEMPORAL LOBE AND NEOCORTICAL EPILEPSY
Abstract number :
2.300
Submission category :
Year :
2004
Submission ID :
789
Source :
www.aesnet.org
Presentation date :
12/2/2004 12:00:00 AM
Published date :
Dec 1, 2004, 06:00 AM
Authors :
1Susanne G. Mueller, 2Kenneth D. Laxer, 1Nathan Cashdollar, 1Ria C. Lopez, and 1Michael W. Weiner
Hippocampal metabolic abnormalities (HMA) as detected by MR spectroscopic imaging (MRSI) not only can be found in different forms of temporal lobe epilepsy (TLE), i.e., with and without hippocampal sclerosis, but also in neocortical epilepsies (NE). The aim of this study was to learn more about the specific characteristics of HMA in TLE and NE by determining differences regarding severity, frequency of bilateral damage, concordance with EEG lateralization, and extent of HMA between these 3 patient groups. 41 patients with partial epilepsy (16 with NE, 15 with TLE with medial temporal sclerosis (TLE-MTS), and 10 with TLE and normal MRI (TLE-No)) and 16 controls were studied with hippocampal 2D MRSI (TR/TE: 1800/135 ms). In the control and patient groups 12 voxels were uniformly chosen in each hippocampus. Voxels with NAA/(Cr+Cho)[le](Meancontrols[ndash]2SDcontrols) were defined as pathological voxels (PV). HMA were classified according to severity (number of PV in both hippocampi), bilaterality (presence of PV in both hippocampi), concordance with EEG (higher number of PV in ipsilateral hippocampus) and extent (restricted: PV adjacent; diffuse: PV non adjacent). Statistical analysis was done by U-tests. HMA were found in 73% of TLE-MTS, 50% of TLE-no and 63% of NE. In TLE-MTS, HMA was significantly more restricted (p=0.007) and concordant with EEG lateralization (p=0.016) than in TLE-no in whom it was often bilateral and non-concordant. HMA in NE differed from TLE-MTS by being more often diffuse (p=0.036) and from TLE-no by being more often unilateral (p=0.038) and concordant with EEG lateralization (p=0.046). HMA had distinct characteristics in each of the 3 groups which very likely reflect differences of hippocampal neuronal loss/dysfunction in these groups. In patients with TLE-MTS the pattern is consistent with the hippocampus being the seizure origin. The fact that HMA in TLE-no and NE in comparison to TLE-MTS were more often diffuse, i.e., not restricted, bilateral or not concordant with EEG lateralization, might either indicate a secondary involvement of the hippocampal formation (NE) or a seizure onset zone not restricted to the hippocampus (TLE-no). (Supported by NIH grant ROI-NS31966 to KDL.)