Study On Brain Network Connectivity Of Different Clinical Features Of First-episode,Drug-naive Depression

Objective(s):Major depressive disorder(MDD)is a highly heterogeneous disease.The severity of the disease,the age of onset,and different clinical features or symptoms may be related to the heterogeneity of MDD.Some studies have found that MDD with different clinical features has different brain function and structural damage,but these research results are inconsistent.With the progress of methodology in recent years,it is possible to study the heterogeneity of MDD through the brain connectivity network.Therefore,we studied the heterogeneity of different severity and age of onset and its correlation with different symptom features.At the same time,in order to further explore the structural network neurobiological mechanism of different clinical features,we focused on neural network research on MDD with melancholic features and suicidal ideation.Methods:181 adult MDD patients(18-60 years old)with the first onset and without medication were recruited to participate in the study.They were divided into Slight MDD(SD;N=23),Moderate MDD(MD;N=77)and Heavy MDD(HD;N=81)groups.And the healthy control group(HC;N=81)matched with the patient’s age and sex were collected,and the general clinical data and neuroimaging data were collected.The study was approved by the ethics committee of Kunming Medical University,and all participants signed written informed consent.FreeSurfer 7.0 was used to preprocess T1 images of grey matter volume(GMV)and structural covariance network(SCN)analysis for the default mode network(DMN)and the executive control network(ECN).Using graph theory to analyze the node characteristics of white matter(WM)structural network,different symptom related networks were determined in three MDD groups with different onset-age(OA).Patients were grouped according to different OA:early-onset MDD group(18-29 years old;EOD,n=32),mid-onset MDD group(30-44 years old;MOD,n=30)and late-onset depression(45-60 years old;LOD,n=23),And 32(EHC),29(MHC)and 23(LHC)in the corresponding control group.In order to further explore the structural network of different clinical features(melancholic features and suicidal ideation,SI)and reduce the influence of age on them,participants younger than 45 years old were selected for analysis.First,they were grouped according to melancholic features:melancholic MDD(M-MDD,n=23),non-melancholic MDD(NM-MDD,n=59)and 63 healthy controls(HC).The graph theory method is used to analyze the characteristics of WM structure network,and the support vector machine(SVM)method is also used to find out the brain regions that distinguish MDD without and with melancholic features.MDD was grouped according to MADRS item 10th score(suicidal ideation,SI):No SI(n=32),Mild SI(n=64)and Severe SI(n=33)and matched healthy control group(n=69).Finally,based on DPABIsurf,fMRI data were preprocessed in the MDD group(n=50,4 cases excluded with FD>2mm)and control group(n=50),and cortical based dynamic functional network analysis and correlation analysis with HDRS-17 factor score were performed.Results:In the three groups of depression with different severity(Slight,Moderate and Heavy MDD),the brain regions with reduced gray matter volume(GMV)increased with the severity of the disease.Compared with HC group,the GMV of Sligh group decreased by 4 brain regions:left anterior cingulate gyrus,left hippocampus,left corpus callosum and right middle temporal gyrus;The GMV of moderate group(14 regions)and severe group(22 regions)decreased and spread to the left and right frontal parietal lobe and occipitotemporal lobe.Among them,the GMV of left anterior cingulate gyrus was negatively correlated with severity,and had nothing to do with the course of disease.The whole brain network connectivity of MDD increases,and increases with the disease.In the default modal network(DMN)and executive control network(ECN),the connectivity between the left cingulate gyrus and the left and right orbitofrontal cortex was enhanced.The study on the heterogeneity of depressive symptoms and brain network of white matter structure in patients of different ages found that:For early-onset MDD(18-29 years;EOD),in which abnormal brain regions are concentrated in DMN and ECN,and higher guilt and loss of interest were correlated with the insula,and inferior parietal lobe which in default mode network and salience network.For mid-onset MDD(3044 years;MOD),in which abnormal brain regions are more concentrated in the ECN and higher somatic symptoms were correlated with thalamus which in the corticostriatal-thalamic-cortical circuit.For later-onset MDD(45-60 years;LOD),poor sleep symptoms were correlated with the caudate in the basal ganglia,which suggests the cingulate operculum network in the control of sleep.Following,based on the research results of different clinical features.The white matter network analysis of MDD with melancholic features(18-45 years old)found that compared with HC,small-worldness was reduced and node attributes are abnormal in the right orbital inferior frontal gyrus(ORBinf.R),left orbital superior frontal gyrus(ORBsup.L),right caudate nucleus(CAU.R),right orbital superior frontal gyrus(ORB sup.R),right orbital middle frontal gyrus(ORBmid.R),left rectus gyrus(REC.L),and left median cingulate and paracingulate gyrus(DCG.L)of MDD.Compared with non-melancholic MDD(NM-MDD),small-worldness was reduced and node attributes are abnormal in ORBinf.R,ORBsup.L,and CAU.R of Melancholic MDD(M-MDD).Correlated with the 7th item score of the HDRS-17(work and interest),increased aBC in ORBinf.R was positive and decreased aBC in ORBsup.L was negative.SVM analysis results showed that abnormal aBC in ORBinf.R and ORBsup.L showed the highest accuracy of 81.0%(69/83),the sensitivity of 66.3%,and specificity of(85.2%)for discriminating MDD patients with or without melancholic features.And in MDD with suicidal ideation(18-45 years old),ReHo,gray matter volume and structure covariance network analysis showed that,ReHo values and GMV of the right orbital inferior frontal gyrus(ORBinf.R)in the MDD group decreased with the increase of SI.And left hippocampus(HIP.L)in the MDD group decreased.Compared with HC group,Compared to the HC group,the MDD patients showed enhanced structural connectivity of three pairs of brain regions(ACC.L-left superior frontal gyrus(SFG.L),ACC.L-HIP.L and HIP.L-ORBinf.R).The structural connectivity of three pairs was gradually enhanced from NSI group to SSI group(HIP.L-ORBinf.R,ORBinf.R-right posterior cingulate cortex(PCC.R),PCC R-left postcentral gyrus(PoCG.L)).Further analysis of the hippocampal subregion showed that the left hippocampal CA1 body was atrophic and correlated with ORBinf.R in structural connectivity is enhanced.Finally,the complex brain network constructed based on cortical dynamic functional network once again proves that there is a small world attribute model in MDD,which is basically consistent with our previous research results.The global information efficiency and the small-world attribute are weakened,the orderly and efficient network changes to a random network,and the connectivity of brain regions is abnormal,such as prefrontal lobe and cingulate gyrus.The right orbital inferior frontal gyrus left hippocampus was correlated with the score of suicidal ideation.The right orbital inferior frontal gyrus and the right dorsolateral superior frontal gyrus were associated with decreased work and activity.Both the right orbitofrontal were related to the scores of HDRS-17 items 3(suicidal ideation,SI)and 7(decreased work and activity).Conclusion(s):Our study found that the brain functional and structural network of patients with MDD were damaged in the early stage of the disease,and the damage increased with the severity of the disease,while patients with small GMV of left anterior cingulate gyrus may be more likely to develop into severe depression,and found that their connectivity with left orbitofrontal in the default network and with right orbitofrontal in the executive network were enhanced.In the early-onset patients,the changes of the default network and executive network related brain regions were found.In the middle-onset patients,the changes of executive network related brain regions were more likely to be affected by internal and external environmental factors.In the late-onset patients,more brain network damage was concentrated in the basal ganglia.The study also found that the brain structure and functional network connectivity of patients with MDD(less than 45 years old)suggest that the right orbital forehead located in the executive network is closely related to the cognitive impairment of depression,and the mild suicide ideation has changed.In conclusion,the brain changes of MDD have appeared in the early stage of disease onset and mild clinical symptoms.This change increases with the severity of the disease.With the increase of age,this trend becomes more obvious,and the damaged brain regions are age-related;The changes of brain structural and functional connectivity in MDD patients with melancholic and suicidal tendency also suggest the biological phenotype of their symptoms.Therefore,patients with MDD should be prevented,diagnosed and treated early.The study on the relationship between different clinical features and neuroimaging of patients with depression may provide a new direction for early diagnosis and individualized treatment of patients with depression in the future.