A Study On Effective Connectivity Of Major Depressive Disorder Based On Spectral Dynamic Causal Modeling

Major depressive disorder(MDD)is a serious affective disorder.The abnormal functions and connections within many functional networks including the default mode network(DMN),salience network(SN)and central executive network(CEN)were discovered in many previous studies by the functional magnetic resonance imaging(f MRI)technology.Compared with traditional functional connectivity,effective connectivity was used to investigate the causal interactions between two brain regions.It can not only compute the strength of causal interactions among the different brain regions,but also provide the direction of communication among multiple brain regions.And this method,which can reveal the mechanism of causal interaction among human brain regions more accurately,is important to disclose the neural mechanism of MDD.Hence,the present study used the spectral dynamic causal modelling(sp DCM)to model and analyze the effective connectivity of patients with MDD.The main work includes the following three parts:1.Effective connectivity analysis of single network(DMN)DMN plays an important role in the pathogenesis of MDD,and a large number of previous researches have reported the aberrant functional connectivity within this region.However,the causal interaction of DMN remains unclear.In order to explore the underlying relationship between abnormal effective connections within DMN and MDD or the relapse of MDD,the first part of this paper investigated the effective connections within the DMN by sp DCM.The present study collected the resting-state f MRI data of healthy controls,first-episode MDD patients and recurrent MDD patients respectively.Then,we used the sp DCM to construct the pattern of effective connectivity for three groups.By comparative analysis,our study discovered that the abnormal connection between the left parietal cortex(LPC)and the medial prefrontal cortex(m PFC)of left DMN might be closely related to the episode of MDD.Meanwhile,our research also found that the aberrant connection from m PFC to posterior congulate cortex(PCC)in DMN might be associated with MDD recurrence.2.Effective connectivity analysis of two networks(DMN/CEN)The DMN is also referred to as the task-negative network(TNN),and corresponds to the task-positive networks(TPNs)whose representative network is CEN.Previous studies have shown that the interaction between the DMN and CEN was crucial to support the cognitive and emotional function.In order to investigate the switching state of the TPNs and TNN,the second part of this paper studied the effective connectivity between the DMN and CEN using the sp DCM.The result of sp DCM showed that the effective connections of healthy controls were mainly concentrated in the connection between two networks,while the connections of MDD patients were mainly concentrated in the connection within the networks.This result implied that there may be a lack of communication between the two networks in patients with MDD.Secondly,we also found the influence from DMN to CEN was enhanced in MDD patients,while the influence from CEN to DMN was attenuated.That means the switching function between TPNs and TNN was abnormal,and this dysfunction leads to abnormally affective control in the MDD patients.3.Large-scale effective connectivity for brain networksMDD patients usually have multiple dysfunctions such as the emotion,cognitive control and attention deficit.According to the single network(DMN)and two networks(The CEN and DMN)analysis,this part of work further investigated the effective connectivity among nine networks including visual network,sensorimotor network,auditory network,salience network,dorsal attention network,DMN,cerebellum and left and right CEN using the large-scale effective connectivity analysis and graph theory analysis.Our results demonstrated that the symmetric connection of large-scale effective connectivity was disrupted.Meanwhile,we also found the mean strength of three abnormal effective connections within patients with MDD that were negative correlated with the clinical score.These aberrant connections include that left precuneus to right precentral,the Frontal_Inf_Tri_L to right Angular,and the VN2 of visual network to Parietal_Inf_L.In addition,the results of graph theory analysis discovered the impaired topological structure of the brain,and the efficiency of information transmission in the whole brain was significantly reduced.Meanwhile,the reduced local efficiency led to the ability of resisting external risks in a few brain networks attenuated.From the perspective of brain network,this paper deeply explored the abnormal effective connection between intra-and inter-network brain networks in patients with MDD using the analysis of the effective connection from a single network to large-scale brain network,which provides a new interpretation for the pathophysiological mechanism of MDD and some inspirations to the prevention and treatment of the recurrence of this disease.