Structural And Functional Brain Networks In Alertness Of Unilateral Temporal Lobe Epilepsy,a Graph Theory Research

Objective:Temporal lobe epilepsy（TLE）is the most common form of adult medically intractable epilepsy.Accumulating evidence has shown that TLE is a disorder of abnormal epileptogenic networks,rather than focal sources.Graph theory allows for a network-based representation of TLE brain networks,and has potential to illuminate characteristics of brain topology conducive to the associated declined alertness in TLE.We use Diffusion Tensor Imaging（DTI）tractography to construct white matter networks of cerebral cortex;and from the resting-state functional MRI to the brain functional networks.We aimed to study the differences of small-world topological properties related brain structural and functional networks between right TLE patients and controls,respectively.Moreover,we investigate the relationship between the network properties and the alertness of TLE.Materials and methods:20 right TLE patients and 19 healthy controls were recruited,matching the gender and age.（1）The DTI scan was performed in all subjects.The DTI data was computed by using PANDA software,and then constructed the white matter network of cerebral cortex,parallel comparison between the two groups from network parameters(local efficiency,the shortest path length,cluster coefficient E_loc and global efficiency E_glo,modularity);（2）All the subjects underwent resting-state BOLD-f MRI scan.The images were segmented by using AAL-90 templates.Then a 90*90 binary matrix were built based on a range of sparsity threshold.The network topological parameters(local efficiency,the shortest path length,E_loc and E_glo,module)were described on the basis of the binary matrix.Then calculated the parallel comparison between the two groups;（3）the attentional networks test（ANT）was performed in all the subjects,conducted the score of alertness.Further correlation analyses between network parameters in which white matter network and function network and ANT scores were performed in the entire right TLE patients,with or without correction for clinical course.Results:（1）The FA-weighted matrix（90×90）generated from PANDA was threshold in different level to create the adjacency matrix.For each subject,the FA-weighted matrix was used for further graph analyses.Each matrix represented the white matter network of cerebral cortex,in which each row or column represented a brain region of the AAL template.The graph measurements of the DTI network were computed over a series of thresholds on FA values（FA=0.2～0.42）with a step of 0.02.Compared with healthy controls,RTLE patients were of higher cluster coefficient（FA<=0.28）and same path length,which led to significantly elevated small-worldness（FA<=0.26,and FA=0.30）.RTLE showed unchanged mean local efficiency and global efficiency.（2）The correlation matrix was threshold in different sparsity to create the adjacency matrix.Since there is currently no definitive way to select a sole threshold,we calculated graph measures of the BOLD-f MRI network over a series of thresholds in a wide range of network sparsity（5%～40%）with a step of1%.Compared with healthy controls,RTLE patients were of lower cluster coefficient（sparsity<11%）and higher path length（5%<sparsity<8%）,which led to significantly reduced small-worldness（sparsity<11%）.After controlling the functional connectivity strength difference using the sparsity threshold,RTLE showed lower mean local efficiency（sparsity<9%）,and lower global communication efficiency（5%<sparsity<8%）.（3）There was a negative correlativity between ANT scores andσ,γin white matter structural networks of right TLE patients（0.34<FA<0.42）.This was also significant when corrected for clinical course.There was no correlation between ANT scores andλ,E_loc,E_glo,modularity in white matter structural networks of right TLE patients.（4）ANT scores was negatively associated withσ,γin functional networks of right TLE patients（0.29<sparsity<0.40）.The same correlation were found in ANT scores and modularity as well（0.16<sparsity<0.38）.This was also significant when corrected for clinical course.There was no correlation between ANT scores andλ,E_loc,E_glo in functional networks of right TLE patients.But negative correlation appeared between ANT scores and E_loc when corrected for clinical course.Conclusion:（1）All subjects exhibited altered small-worldness in both brain structural and functional networks.The brain white matter structural networks of right TLE patients shifted towards a more regular network topology.And their brain functional networks shifted towards a more random network topology.（2）The white matter structure of RTLE patients’reconstructed possibly,and maintained the efficiency of network structure.The functional network may be destructed,and brain functional separation and integration declined.Thus functional network efficiency decreased as well.（3）Dynamics of functional network might be changing earlier than structural network.It suggested that it might be more sensitive to detect the brain function of epileptic patients,which could serve as a potential imagining biomarker for clinical detection of TLE patients.（4）There might be an abnormal alertness function when damage occurred in both brain structural and functional network.And the small world properties of the functional network would have a greater impact on the alert function.