Functional And Structural Brain Hemispheric Networks In Major Depressive Disorder

A massive amount of evidence from neuroimaging studies indicates that major depressive disorder (MDD) is related to disrupted hemispheric asymmetry of both structural and functional brain networks. Despite these advances, the studies mainly focus on specific regions. However, hemispheric network topological alterations in MDD are largely unknown to date. Here, we systematically investigated MDD-related alterations in both structural and functional hemispheric brain networks by combining diffusion tensor imaging (DTI) and resting-state fMRI (R-fMRI). We collected both DTI and R-fMRI data from 35 patients with MDD and 35 age-, gender-, and education-matched normal controls (NC). Following the network construction, graph-based network properties (local efficiency, Eloc, global efficiency, Eglob, normalized Eloc, Gamma, normalized Eglob, Lambda and nodal degree) were then calculated for each hemispheric network. Two-way mixed ANCOVA was performed to examine hemispheric, group and hemispherexgroup interaction effects on the network measures with age, gender, education as covariates. Lateralization index was performed to examine the asymmetry of connections in hemispheric level. Spearman rank partial correlation was performed to examine the relation between the functional-structural hemispheric asymmetry and Pearson partial correlation was finally preformed to examine the relation with the clinical features. We first found significantly decreased intra-and inter-connectivity strength in MDD when compared with NC. Then, we found decreased global properties and decreased intra-and inter-hemispheric degree centrality in hippocampus, anterior cingulate gyrus and superior frontal gyrus in the MDD in functional network, not in structural networks. Finally, a decreased coupling of functional and structural hemispheric lateralization was found in MDD. Our findings suggest that MDD is related to hemisphere- and modality-specific topological alterations of brain networks and thus provide novel insights into disconnection nature of the disease.