Abstracts

Rationale: Asymmetries in pathological presentation and the organization of brain functions, such as mesial temporal lobe (MTL) functions, are valuable indicators of seizure onset location, and helpful in guiding clinical-decision making prior to surgical interventions in temporal lobe epilepsy (TLE). In this project, we investigated the laterality patterns of hippocampal gyrus (HG) and parahippocampal gyrus (PHG) associated with a visual-spatial semantic memory (VSSM) task in TLE patients, with an emphasis on the degree to which the presence of mesial temporal sclerosis (MTS) modulates these patterns. Lastly, we tested the impact of aging on hemispheric asymmetry, potentially implicating mediators of cognitive reorganization. Methods: Fifty-six unilateral TLE patients (LTLE: 31; RTLE: 25) and 34 matched healthy controls (HC) were recruited. All participants underwent an fMRI scan during performing a VSSM task adopted from Binder et al. (2005) which includes 8 blocks of scene recognition (indoor vs outdoor) and 8 blocks of control condition (scrambled images comparisons). After standard postprocessing, the laterality index (LI) of HG and PHG were calculated based on the contrast of scene recognition minus control condition respectively. We assessed the LI pattern of these two MTL structures between patients and HC. In addition, we broke down patient groups into subgroups (with, without MTS) and tested for difference in LI patterns. The associations of aging and LI in all groups were explored at last. Results: TLE patients and HC did not differ in age, gender, or handedness. All the experimental groups did not differ in the performance accuracy of scene recognition (HC: 99.0%; LTLE: 98.3%; RTLE: 99.4%). In HC the task yielded a slight right hemisphere activation bias in both the HG (LI=.11) and the PHG (LI=.09). A significant main effect for Group was shown (F=4.8, p=.01), with the LTLE displaying a higher LI (greater rightward bias) than the NC or RTLE. A significant interaction (F=4.1, p=.02) was also found such that the LI difference between HG and PHG was higher in the RTLE (Figure 1). In MTS patients, an interaction was found between Group (LTLE, RTLE) and MTL Structure (HG, PHG) (F=6.09, p=.021, Figure 2A), with RTLE showing a greater HG-PHG LI difference. The same analysis conducted with non-MTS subjects revealed no such differences (Figure 2B). Lastly, in HCs, significant positive correlations were found between age and both the HG (r=.387, p=.024) and PHG LIs (r=.412, p=.016). This relationship was not observed in the patient groups. Conclusions: In our sample of matched TLE and HCs, these findings suggest the lateralization of HG activity implementing VSSM reorganized as a result of TLE pathology, with this pattern stronger in the setting of MTS. The PHG does not show the same dramatic shifts in laterality. A similar shift in response to pathology is observed in the RTLE patients, who display a shift to the left-hemisphere, with this pattern also stronger in patients with MTS. These shifts in TLE showed no association with age. In NC, however, both the HG and PHG LI's showed a shift to the right hemisphere with aging. These findings demonstrate adaptive memory reorganization to the healthy hemisphere in TLE. They also suggest that with normal aging important shifts in lateralization may occur for tasks that demand bi-hemispheric activity, implying that the epileptic brain may age differently. Future research will determine if the presence or absence of the cognitive reorganization demonstrated here can serve as a biomarker for either seizure or cognitive outcomes following surgical interventions for TLE. Reference: Binder, et al. A comparison of two FMRI protocols for eliciting hippocampal activation. Epilepsia 2005;46:1061-1070. Funding: N.A.