Abstracts

Rationale: Anterior temporal lobectomy (ATL) is an effective treatment for medically refractory temporal lobe epilepsy, carries the risk of additional postoperative cognitive morbidity. Recent advances in resting state fMRI (rs-fMRI) have demonstrated that a novel resting state intrinsic connectivity distribution (rs-ICD) analysis of fMRI BOLD signal fluctuations has also shown to be a powerful tool in identifying neural underpinnings for various cognitive functions in normal subjects. Preliminary results in our lab have been the first to demonstrate correlations with performance on neuropsychological memory tests and resting state ICD in patients with temporal lobe epilepsy. However, the ability of these correlations to predict post-surgical outcome is less understood. The aim of this study is to determine if these ICD correlations can predict postoperative naming outcome in ATL patients. Methods: We recruited 20 TLE patients (11 left; 9 right) who underwent resting state fMRI, preoperative neuropsychological tests, standard temporal lobectomy, and one year of postoperative follow-up. A gradient-echo fMRI sequence was obtained from each patient, and connectivity was measured by a voxel based calculation of the degree of correlation coefficient for each voxel within the brain hemisphere and within medial or medial/lateral temporal lobes to ipsilateral and contralateral to the seizure focus. Analysis of rs-fMRI connectivity was correlated with preoperative and pre-postoperative change scores obtained on a battery of neuropsychological tests, including the Boston Naming Test (BNT).Results: Intrinsic connectivity distribution in the left temporal lobe was correlated with pre to postoperative change scores on the BNT. ICD predicted postsurgical naming outcome with the weaker the connectivity, the more likely a person was to decline post surgically. ICD values predicted postoperative naming decline above preoperative neuropsychological testing. Conclusions: This is the first study to show a voxel based measure of resting state functional connectivity without predetermined ROIs or arbitrary thresholds predicts postoperative naming decline in L-ATL patients. Our results suggest that L-ATL patients who have greater connectivity networks in the left anterior temporal lobe are at less risk for postoperative naming decline following left anterior temporal lobe resection. Conversely, those L-ATL patients with fewer connections extending into these areas are at higher risk for significant postoperative naming decline. These findings furhter our understanding of neural correlates of postsurgical naming decline in patients with epilepsy and provides a new, non-invasive technique that identifies patients who are at risk for such naming decline.