Abstracts

Rationale:
Previous studies have observed decreased connectivity between the default mode network (DMN) and the temporal lobes in temporal lobe epilepsy (TLE). However, in patients undergoing surgery for intractable TLE, the relationship between the pre-operative and post-operative DMN and the patients’ functional recovery remains unclear. Here, we used resting state functional MRI (rsfMRI) network mapping to more broadly explore DMN connectivity and its correlation with neuropsychological measurements both pre- and post-operatively in TLE patients. The objective of this study was to employ pre-operative, non-invasive network mapping to identify correlates between DMN and neuropsychological outcome.
Method:
We prospectively followed 27 medically refractory TLE patients who all consented to participating in this study. Each patient received a Phase 1 workup for epilepsy surgery including (but not limited to): long-term video-EEG monitoring, high resolution MRI with rsfMRI, and quantitative neuropsychological evaluation. We conducted modeling of DMN and epilepsy networks, as well as neuropsychological evaluation, both pre- and post-operatively. Connectivity of the DMN was defined as the average Pearson correlation coefficient of the adjacency matrix generated from all voxels within the DMN hub (either ventral or dorsal hubs).
Results:
Pre-operative data indicated that patients with higher ventral DMN connectivity had significantly worse visuoconstructional and right frontal executive function. Post-operatively, change in the DMN was determined by comparing the pre-operative connectivity to the post-operative level. A larger decrease in both ventral and dorsal DMN connectivity was associated with deterioration of verbal memory and logical memory after surgery. Additionally, lower pre-operative ventral DMN connectivity was associated with less disruption of the ventral and dorsal DMN after surgery. When comparing DMN and the epilepsy network, we observed that post-operative ventral and dorsal DMN connectivity was lower in patients with more widespread epilepsy networks – networks with greater involvement of the contralateral temporal lobe. However, there was not a significant correlation found between epilepsy network spread and the pre-operative connectivity of ventral and dorsal DMN.
Conclusion:
The relationship of subdivisions of the DMN, the epilepsy network, and neurocognitive function before and after epilepsy surgery was explored. Before surgery, higher ventral connectivity was associated with worse executive and visuoconstructional function. After surgery, disconnection of the ventral and dorsal DMN correlated with greater decrease in logical and verbal memory when compared with the pre-operation baseline. Pre-operative epilepsy networks that were more widespread were found in patients that were more likely to have a decrease in connectivity within the ventral and dorsal DMN after surgery. These results together suggest the possibility that the spread of the epilepsy network may be predictive of less “stable” DMN connectivity, and, when the DMN is disrupted with surgery, patients are more likely to have a decline in neuropsychology function.
Funding:
:No funding to report.