Abstracts

Rationale: Executive functions in temporal lobe epilepsy (TLE) remain much less characterized than memory and language function, despite the ample evidence to support the likelihood of executive dysfunction in both adult and pediatric TLE patients (Hermann et al., 2007; Stretton and Thompson, 2012; Longo et al., 2013). In this study, we built on past research that revealed associations between frontal lobe white matter compromise and executive dysfunction in TLE (Kucukboyaci et al., 2012). We studied how diffusion tensor imaging (DTI) and functional magnetic resonance imaging (fMRI) metrics interact to explain executive dysfunction in patients with left TLE. Based on extant literature, we hypothesized that patients with increased frontoparietal functional connectivity (FC) would demonstrate better executive function performance and that frontoparietal FC strengths and executive function performances will be related to integrity of the underlying white matter. Methods: We derived FC metrics for the frontoparietal networks recently implicated in executive function (Zhang et al., 2012, Janssen et al., 2013) and combined these metrics with average diffusivity and fractional anisotropy (FA) estimates obtained from DTI tractography of the frontal corticostriatal tracts (FCST). The fMRI processing used to tabulate task-regressed FC data was adopted from Kucukboyaci et al. (2013). We operationalized executive function as performance on tasks of response inhibition (Color Word Interference - Inhibition, CWI) and visuomotor set-shifting (Trailmaking Test - B). We analyzed how variations in structural white-matter integrity and frontoparietal FC interact with executive function performance using Spearman's rho correlations. Results: We found that higher performance on the CWI task was associated with higher FC between right superior frontal gyrus (SFG) and left inferior parietal lobule (IPL; rho= .816, p =.004), and between right SFG and right IPL (rho=.571, p = .085). There were no significant correlations between CWI task performance and FC between the left SFG and left or right IPL. We also found that, for a given level of FA in the FCST bilaterally, higher FC between these two regions still explained a significant portion (i.e., 60-80%) of the variation in CWI task performance. In contrast, FC between the left SFG and bilateral IPL did not. Conclusions: We report the FC-executive function correlations to be stronger between the left IPL and the contralateral frontal lobe of left TLE patients after controlling for FA of the underlying frontal white matter. Given the early age of onset and long disease duration in our patient sample, our results are suggestive of possible functional reorganization that involves increased contralateral frontoparietal FC that may occur as a result of increased diseased burden on the affected hemisphere.