Abstracts

Hubness of the Left Piriform Cortex May Be Associated with Response to Antiseizure Medication in Patients with Temporal Lobe Epilepsy

Abstract number : 3.508
Submission category : 5. Neuro Imaging / 5A. Structural Imaging
Year : 2023
Submission ID : 1495
Source : www.aesnet.org
Presentation date : 12/4/2023 12:00:00 AM
Published date :

Authors :
Presenting Author: Felix Zahnert, MD – Philipps University Marburg

Susanne Knake, MD – Epilepsy Center Hesse; Reichert Paul, MD – Philipps University Marburg

Rationale:
Recently, the piriform cortex has gained substantial attention in the field of epilepsy surgery as the extent of its resection may be associated with postoperative seizure freedom in patients with TLE. However, little is known about the role of its connectivity in relation to the patient's response to surgery or anti-seizure medication.

Methods: We identified 65 patients with temporal lobe epilepsy who had received a diffusion MRI scan at our centre retrospectively. IRB approval was granted. Piriform cortex (PC) was segmented manually on both hemispheres for each patient according to an established protocol (Galovic et al 2019). Probabilistic tractography (FSL probtrackX) was conducted to generate connectomes from two separate cortical parcellations (Desikan, 2006 and Schaefer 200 regions functional parcellation, 2018) including the segmented PC. Graph theoretical (GT) measures of centrality (node degree and betweenness centrality) were computed. GT-measures were compared for the conditions "reponders to ASM" versus "non-responders" and "Engel_1A" after epilepsy surgery versus "not Engel_1A" using a permutation t-test with 100 k permutations. Subsequently, in an exploratory analysis, GT measures of all regions were compared across conditions and effect sizes were examined.

Results:
The left piriform cortex showed a strong effect of greater degree centrality in subjects who were refractory to treatment with ASM (n=17) as compared to responders to ASM within the 'Desikan" connectomes (n=44) (p=0.0094, t=2.63, d=0.81). The same was the case for betweeness centrality of the left PC (p=0.00898, t=2.73, d=0.58). No significant effect was observed for the right hemisphere. Furthermore, no differences in hubness of either PC was detected in patients who became seizure free after epilepsy surgery compared to those who were not. Within the finer-grained Schaefer parcellation, no significant effects were observed within piriform cortex.

Conclusions:
We present the first data on the structural connectivity of the PC in patients with TLE. The hubness of the left PC was associated with resistance to treatment with ASM, underlining its role in focal epilepsy and leading to speculation whether this projection of the PC, which can be part of the epileptogenic network, to a larger network of cortical regions may be one mechanism leading to treatment resistance. However, this result could not be validated using a second, more  fine-grained parcellation, which may be due to the low angular resolution protocols (30 gradient directions) used in this study.

Finally, this study provides an important first insight into the connectivity of the PC in TLE, and further investigations may 1) test this approach on different, high quality datasets with different parcellations and 2) might lead to the development of a connectome biomarker of prospective seizure freedom on ASM.

Funding: None

Neuro Imaging