There is a pressing need to further our understanding of the mechanisms underlying the depression symptoms in patients with post-stroke depression (PSD) in order to inform targeted therapeutic approaches. While previous research has demonstrated a reorganization in the functional brain network of PSD, it remains uncertain whether, or not it also occurs in the structural brain network. We therefore aim to investigate the structural brain network of patients with PSD as compared to post-stroke non-depression (PSND) patients. In addition, our research considers the relationship between network metrics and functional measurements. Thirty-one PSD patients and twenty-three PSND patients were recruited. All patients underwent MRI and functional assessments, including the Barthel index, mini-mental state examination (MMSE), and Hamilton depression rating scale (HAMD). Diffusion tensor imaging was used to construct the structural brain network and to conduct the subsequent graph theoretical analysis. Network measures were computed and compared between PSD and PSND patients. Associations between functional assessments and network measures were studied as well. We successfully detected increased global and local efficiency in patients with PSD. Regions with disrupted local connections were located primarily in the cognitive and limbic systems. More importantly, PSD patients' global and regional network measures were associated with depression severity, as measured by HAMD. These findings suggest that disrupted global and local network topologies might contribute to PSD patients' depression symptoms. Therefore, connectome-based network measures could be potential bio-markers for evaluating stroke patients' depression levels.
Keywords: brain network; diffusion tensor imaging; network analysis; post-stroke depression; small-world.