Longitudinal MRI of the Limbic System in Medically Intractable Temporal Lobe Epilepsy
Abstract number :
2.141
Submission category :
5. Neuro Imaging
Year :
2010
Submission ID :
12735
Source :
www.aesnet.org
Presentation date :
12/3/2010 12:00:00 AM
Published date :
Dec 2, 2010, 06:00 AM
Authors :
Min Liu, L. Concha, C. Beaulieu and D. Gross
Rationale: There have been a small number of MRI studies demonstrating atrophy of mesial grey matter (GM) such as hippocampus and amygdala in patients with chronic pharmacoresistant temporal lobe epilepsy (TLE). However, it is unknown if limbic white matter (WM) including fimbria-fornix and cingulum are affected under the influence of long-term recurrent seizures in this case. Our objective was to assess (1) white matter tracts of limbic system using diffusion tensor imaging (DTI), (2) total GM and WM volumes, and (3) change of T2 relaxometry of hippocampi in medically intractable TLE patients scanned 5.5 years apart. Methods: Four patients with TLE and without mesial temporal sclerosis (3 males, 1 female, age at first scan: 33, 33, 45 and 46 yrs) and two healthy volunteers (both male, 55 and 48 yrs) were scanned twice on the same 1.5T scanner using both CSF-suppressed FLAIR DTI and standard DTI, a multi-echo T2 sequence and MPRAGE with a scan interval of 71, 62, 68 and 65 months for patients and 42 months for both controls. Three patients had independent bitemporal ictal abnormalities while one had unilateral right temporal ictal and interictal EEG abnormalities. Left and right fimbria-fornix (Fx) and parahippocampal cingulum (pCg) were identified separately with DTI tractography in DTIStudio to yield fractional anisotropy (FA) and mean diffusivity (MD). Hippocampal T2 was measured by manual region-of-interest. T1-weighted MPRAGE images were segmented into GM and WM to yield volumes using SPM5. The difference (?) of the 2nd scan relative to the 1st and the percentage of change were calculated for each parameter. Results: DTI demonstrated elevated MD bilaterally in the Fx of all patients over 5.5 years (mean ?MD=0.081 10-3 mm2s-1, 8.8%) with no difference in controls (mean ?MD=0.014 10-3 mm2s-1, 1.4%) whereas Fx FA reduced similarly for both groups (patients: mean ?FA=-0.026, -5.5%; controls: mean ?FA =-0.037, -7.3%). Notably, the pCg showed elevated FA bilaterally in all patients (mean ?FA=0.043, 10.7%), but not in controls (mean ?FA=-0.0025, -0.5%). MD was decreased in 5/8 but elevated in 3/8 of pCg in patients while MD was decreased in all controls. Brain segmentation demonstrated greater GM volume loss of 29 cm3 in patients compared to smaller reduction of 4 cm3 in controls. WM volume reduced similarly in both groups (patients: -8 cm3; controls: -6 cm3). T2 analysis revealed one patient with dramatic increased T2 on bilateral hippocampi and the one patient manifesting unilateral EEG abnormalities with increased T2 on the abnormal side (all ?T2>8 ms). T2 changes in controls were small (mean ?T2=3 ms). Conclusions: Five year longitudinal imaging of medically intractable TLE patients demonstrates significant grey and white matter changes. In general the observed changes (increased T2 and MD and decreased volume) suggest progressive grey and white matter degeneration secondary to seizures. In contrast, increased FA of the pCg is consistent with enhanced integrity as opposed to degeneration of the structure which suggests plasticity of some white matter tracts in response to seizures.
Neuroimaging