Abstracts

Rationale: Patients with intractable epilepsy may undergo surgical intervention to remove temporal lobe seizure foci. Functional MRI is an informative presurgical tool for identifying laterality of language regions and avoiding postoperative deficits. Moreover, it can provide data on the effects of persistent seizures on the functional organization of language. In the current study, we aimed to measure post-operative changes in lateralization of language related regions after temporal lobectomy in patients with focal epilepsy. Methods: 15 normal controls (mean age=22.6 years, SD=3.1) and 19 patients (mean age= 26.2, SD=12.7) with temporal lobe epilepsy (13 left TLE, 6 right TLE) performed an auditory description decision task as part of presurgical evaluation for a temporal lobectomy and then again at least six months post-surgery (average 5.4 years). Controls performed the task with at least a six month time difference between scans (average 1.7 years). Images were processed in SPM8. A laterality index (LI) was calculated using a bootstrap method for anatomical regions of interest (ROI): inferior frontal gyrus (IFG), Wernicke's area (WA), and middle frontal gyrus (MFG). Lateralization was left if LI>0.20, bilateral if LI<0.20, right if LI<0.20. We conducted a repeated measures (2 x 2 x 4 way) ANOVA to evaluate effects of group (patient or control), time (as repeated scans), and ROI. Exploratory analyses (ANOVA) directed toward the patient group examined the effect of location of resection on LI. Results: Results showed no main effect of time but main effects of group [patients had lower LI in all ROIs (p<0.001)] and ROI [WA was the most lateralized and MFG was the least (p<0.001)]. There were no changes in LI from pre- to post-operative conditions, and this did not differ for group or ROI. Patient-directed analyses indicated that there was no effect of TLE on LI, nor was there an effect whether the surgery was in the dominant hemisphere (n=10). Clinically, there were no cases of aphasia or significant naming deficits. Conclusions: Postsurgical assessment of LI did not differ following resection—remaining lower than controls—and no significant language deficits were observed. This provides evidence that following resection, there are no substantial changes in lateralization which is likely due to the information provided by the presurgical fMRI results, and also indicates that despite the cessation of seizures and reducing of medications, LI does not normalize toward control levels. This may be because networks are no longer plastic, changes occur over a longer timeframe than were measured in this study, or greater bilateral activation is compensatory.