The specific profile of structural connectivity based on DTI-DWI and correlations with cognitive performance in patients with hippocampal sclerosis and temporal lobe epilepsy.

VALLEJO AZAR, MARIANA NAHIR; PRINCICH, JUAN PABLO; VILLANUEVA, MANUELA; NASIMBERA, ALEJANDRO; KOCHEN, SILVIA; GONZALEZ, PAULA NATALIA

Simposio; Simposio LATBrain 2022; 2022

Introduction: Current evidence supports that epilepsy is due to alterations of neural networks rather than area-specific abnormalities, and multiple publications suggest connectivity abnormalities in patients with temporal lobe epilepsy and hippocampal sclerosis (HS). This work aims to determine whether the abnormalities found in HS patients are common in patients with epilepsy. For this, we compared the global properties of brain structural connectivity using graph theory-based indicators between two different types of refractory epilepsy (HS and focal cortical dysplasia -FCD) and a healthy control group (HC). In turn, it is known that alterations in the organization of structural connectivity networks in these patients would play an important role in the cognitive deficit and comorbidities of the disease. The second aim of this study is to evaluate the correlation between global network parameters in patients with epilepsy and cognitive performance in verbal and non-verbal memory tests.Methods: From an MRI sample of epilepsy patients composed of 23 subjects with FCD (right FCD: n= 8), 41 with HS (left HS: n= 25) and 39 HC, matched for sex and age, brain connectivity between 80 anatomical regions (cortical and subcortical) was calculated. These ROIs were segmented in FreeSurfer v6.0, using 3T T1 images previously obtained. Using DWI-DTI, a deterministic tractography analysis was performed employing the DSI studio software. Connectivity matrices were generated between nodes, using average fractional anisotropy-weighted as the measure. In this framework, graph parameters were calculated: clustering coefficient, average connection length, global efficiency, small worldness and transitivity. These measures were compared between groups using ANOVA, controlling the covariates of age, gender and time of disease progression. Significance was adjusted to minimize the effect of multiple comparisons (Bonferroni, Alpha: 0.05). Finally, Pearson's correlation coefficient was calculated between the before mentioned global parameters and the Z-scores obtained from the non-verbal (visual) memory test in a subsample of HD patients (n= 33, right: n= 12).Results: We found an increase in clustering coefficient (p=.041) and transitivity (p=.037) exclusively in HS patients concerning FCD. No differences were found in the rest of the indicators or between patients and healthy controls. A significant negative correlation was found between the clustering coefficient and the Z-score in the non-verbal (visual) memory test (r= -.632, p= .027).Conclusions: The alterations found here could be specific markers of brain connectivity reorganization in patients with epilepsy and HS, and may represent a biomarker for diagnosis, therapeutic response predictions, cognitive deficit or evolutionary follow-up in future work.