| Objective: To research the gray matter changes by using voxel-based morphometry(VBM)method,to study the changes of topological properties of brain structural network by graph-based analysis and to investigate the nonlinear characteristics of EEG background activity by adaptive fractal analysis(AFA)in patients with drug-refractory temporal lobe epilepsy and negative 3.0T MRI finding.Methods: Recruited 35 patients with3.0T MRI-Negative Temporal Lobe Epilepsy(TLE-N),including 20 patients in the drug-refractory group(r TLE-N)and 15 patients in the drug-controlled group(c TLE-N),and 20 healthy controls(HC)with age and gender matching were recruited at the same time.All the subjects completed 3DT1 and DTI imaging scan of the brain.The VBM method was used to process the 3DT1 data,and then to compare and analyze the changes in the gray matter volume of the three groups.Based on DTI data,the deterministic fiber tracking technology and graph theory were used to construct the brain structure network and calculate its topological parameters,and then compare and analyze the changes of the topological properties of brain structure networks of the three groups.6-second EEG background activity segments of r TLE-N group and c TLE-N group were selected,and AFA method was used to calculate the Hurst exponent of EEG background activity of each lead,and then the nonlinear characteristics of EEG background activity in r TLE-N group and c TLE-N group were compared.Results:(1)VBM results: Compared with the HC group,the r TLE-N group had decreased gray matter volume in bilateral hippocampus,bilateral thalamus,bilateral middle temporal gyrus,bilateral median cingulate and paracingulate gyri,left gyrus rectus,left medial orbital superior frontal gyrus and right parahippocampa gyrus(P<0.001).Compared with the HC group,the c TLE-N group had decreased gray matter volume in right middle temporal gyrus and right medial orbital superior frontal gyrus(P<0.001).Compared with the c TLE-N group,the r TLE-N group had decreased gray matter volume in the bilateral insular,bilateral rolandic operculum,left parahippocampa gyrus,left medial orbital superior frontal gyrus,left fusiform gyrus,left cerebelum＿6,right thalamus and right inferior occipital gyrus.(P<0.001).(2)Topological properties of white matter structure network: The white matter structure networks of the three groups all exhibited small world attributes.Compared with the HC group,the r TLE-N group showed increased the shortest path length and decreased local efficiency and global efficiency(P<0.05),and the c TLE-N group showed increased the shortest path length and decreased global efficiency(P<0.05).Compared with the c TLE-N group,the r TLE-N group showed increased the shortest path length and decreased local efficiency and global efficiency(P<0.05).(3)Nonlinear characteristics of EEG background activity: the EEG data with less interference and pseudoerror of 13 patients with r TLE-N and 11 patients with c TLE-N were included.The difference of Hurst exponent between the r TLE-N group and the c TLE-N group in F8-P4,T4-F8 and T6-O2 leads was statistically significant(P<0.05).The range of Hurst exponent in the r TLE-N group was all between 0.5 and 1,while the range of Hurst exponent in the c TLE-N group was all >1.Conclusions:(1)TLE-N is a brain network disease,but there are different epileptic networks between c TLE-N and r TLE-N.Although the grey matter atrophy of c TLE-N was limited,the topological properties of the brain structural network have been damaged,while r TLE-N involved in the atrophy of gray matter in multiple brain regions and the damage of the topological properties of the brain structural network is more serious.(2)At the electrophysiological level,r TLE-N is also different from c TLE-N.The EEG background activity of the leads in the right temporal region of r TLE-N group tended to be of long-term memory,while c TLE-N group tended to be of non-stationary process. |
PDF Full Text Request