Abstracts

Rationale: Resections of the temporal lobe for epilepsy or neoplasms may result in a significant decline in confrontation naming. The resulting deficit can be socio-economically disabling and prevents the return of these patients to independent functioning. The exact substrates responsible for this are not known, and it has previously been assumed that injury to the superior temporal gyrus or resection size predicts these declines. A clear understanding of the most critical constituents might influence the design of surgical strategies to minimize language declines. We used multivariate voxel-based lesion symptom mapping (VLSM) to relate changes in cognitive function, measured as decline in performance on the Boston Naming Test (BNT), to lesions in the dominant temporal lobe.

Methods: Data were obtained from 95 patients who underwent surgical resection in the dominant left temporal lobe for medication-resistant epilepsy. This included a range of surgical approaches including anteromesial temporal lobe resections (n=57), mesial temporal laser ablations (n=26), and resections in other parts of the temporal lobe (n=12). Patients underwent neuropsychological testing and MRI prior to and 6 months following surgery. Lesion masks were traced on the postoperative MRI and aligned to a normative imaging space using nonlinear deformation. The effects of preoperative scores and seizure outcomes (ILAE scores) were regressed out of postoperative scores via a general linear model. VLSM using a multivariate support vector regression model was used to assess postoperative neuropsychological test scores. Resulting Beta maps were converted to p values and corrected using a permutation-based cluster level correction. Additionally, a surface-based mixed-effects multilevel analysis was used to estimate broadband gamma activity during a picture naming task across a subset of the subjects with ECoG recordings, and the resulting maps were integrated with VLSM results. Lastly, language-based task fMRI (picture naming) was used to assess SNR in regions found to be significant in VLSM analysis.

Results: Beta maps from VLSM analysis revealed that the loss of basal temporal regions (mid and anterior fusiform gyrus) was associated with a prominent decline in BNT scores (p< 0.005). Similarly, ECoG analysis showed a significant increase in activity in the mid fusiform gyrus (mFus) during picture naming immediately preceding articulation. It is notable that many of these voxels were spatially correlated with the susceptibility artifact on echo-planar imaging (tSNR=51.21+/-20.16), perhaps explaining why this region has been under-appreciated as the locus responsible for postoperative deficits.