Abstract
BACKGROUND: Anxious depression (AD) is a common subtype of major depressive disorder (MDD). Neuroimaging studies of AD have revealed inconsistent and heterogeneous brain alterations with the use of single-model methods. Therefore, it is necessary to explore the pathogenesis of AD using multi-model imaging analyses to obtain more homogeneous and robust results. METHODS: One hundred and eighty-two patients with MDD and 64 matched healthy controls (HCs) were recruited. Voxel-based morphometry (VBM) was used to estimate the gray matter volume (GMV) of all subjects. The GMV differences between the AD and non-anxious depression (NAD) participants were used as regions of interest (ROIs) for subsequent resting state functional connectivity (rs-FC) analyses. Correlation analysis was used to evaluate the associations between clinical symptoms and abnormal function in specific brain areas. RESULTS: Decreased GMV in the medial frontal gyrus (MFG) and the superior frontal gyrus (SFG) was observed in the AD group compared to the NAD group. Taking the MFG and SFG as ROIs, the rs-FC analysis revealed decreased FC between the left SFG and left temporal pole and between the left SFG and right MFG in the AD group compared to the NAD group. Finally, the FC between the left SFG and left temporal pole was negatively correlated with HAMD-17 scores in the AD group. CONCLUSION: By combining the GMV and rs-FC models, this study revealed that structural and functional disruption of the affective network may be an important pathophysiology underlying AD. The structural impairment may serve as the foundation of the functional impairment.