Altered global topological properties of brain gray matter and white matter functional networks in major depressive disorder and bipolar depression
10.3760/cma.j.cn113661-20250409-00170
- VernacularTitle:抑郁症与双相抑郁患者脑灰质和白质功能网络拓扑结构改变的比较研究
- Author:
Taipeng SUN
1
;
Yue ZHOU
;
Gang CHEN
;
Wei XU
;
Linlin YOU
;
Yingying YIN
;
Yonggui YUAN
Author Information
1. 东南大学附属中大医院心身医学科，南京210009
- Publication Type:Journal Article
- Keywords:
Depressive disorder;
Bipolar disorder;
Network topology characteristics;
Differential diagnosis
- From:
Chinese Journal of Psychiatry
2025;58(12):891-902
- CountryChina
- Language:Chinese
-
Abstract:
Objective:To investigate the alterations in the topological properties of gray matter and white matter dynamic and static functional brain networks in patients with major depressive disorder (MDD) and bipolar depression (BDD) using graph theory analysis, and to evaluate the potential of their combination as biomarkers for differential diagnosis between unipolar and bipolar depression.Methods:From March 2021 to April 2024, inpatients were recruited from the Department of Psychosomatic Medicine, Zhongda Hospital, Southeast University, including 132 patients with MDD, 84 patients with BDD, and 91 healthy controls (HCs). Resting-state structural and functional MRI data were collected, and dynamic and static functional brain networks of gray matter and white matter were constructed. Graph theory analysis was applied to calculate global and nodal network properties, differences in topological attributes among the three groups were compared by One-way analysis of covariance, and Turkey′s post hoc test was used for further pairwise comparison. The network topology attribute indicators with statistically significant inter-group differences were selected using the Least Absolute Shrinkage and Selection Operator regression (LASSO) for feature classification. The diagnostic performance of combined gray and white matter network features for distinguishing MDD from BDD was assessed using receiver operating characteristic (ROC) curves and a random forest model.Results:In the analysis of the static gray matter functional network, both MDD and BDD patients showed abnormal local topological properties. Compared with HCs, the MDD group exhibited abnormal betweenness centrality (BC) in the left inferior frontal gyrus, left precuneus, left ventromedial occipital cortex, right ventromedial occipital cortex, and right anterior thalamus ( t=-3.95-3.62, all P<0.05). The degree centrality (DC) of the left and right anterior thalamus was also abnormal in the MDD group ( t=3.78,4.14, both P<0.001), as was the nodal efficiency (Ne) of the left precuneus and bilateral anterior thalamus ( t=2.37, 3.61, 3.82, all P<0.05). Compared with HCs, the BDD group showed abnormalities in DC and Ne of the left precuneus ( t=-2.76, P=0.014; t=-3.01, P=0.007). In the analysis of the dynamic white matter functional network, both MDD and BDD patients demonstrated abnormal temporal variability of local topological properties. Compared with HCs, the MDD and BDD groups showed reduced BC temporal variability in the left superior corona radiata ( t=-2.39, P=0.047; t=-4.28, P<0.001), and there were significant differences in DC temporal variability in the right posterior limb of the internal capsule and lentiform nucleus ( t=2.65, P=0.021; t=3.49, P=0.001) in MDD group compared with HCs and BBD. The differential diagnosis model combining gray and white matter dynamic and static network topological features achieved an area under the ROC curve of 0.80. Conclusion:Both MDD and BDD exhibit altered topological properties in static gray matter functional networks and dynamic white matter functional networks. The combination of these features may aid in the differential diagnosis of MDD and BDD.
