| Objective:Depressive disorder is a very harmful mental disease,and its pathological mechanism is not clear.Studies have shown that 2-week improvement in antidepressant treatment can effectively predict long-term clinical cure,so it is of great significance to identify biomarkers that are more likely to respond to specific treatment in individuals with depressive disorder.Mri imaging of the brain in a non-invasive,non-invasive manner is used to study the structure and function of the brain,and has the potential to identify objective neurobiological markers that reflect the underlying pathophysiological processes of depressive disorder,and to promote the development of personalized treatment based on biomarkers.We used brain imaging to predict the long-term efficacy of ssris in patients with first-episode depression for 2 weeks,which can help clinical decision-making and individualized treatment.Methods:167 cases of patients with first-onset depressive disorder and 80 cases of normal control were collected.Gender,age,education and other general data were collected according to the self-made case report form(CRF)at the baseline of all subjects,and MRI scanning was performed.All patients with first-episode depression were treated with SSRI,and symptoms were assessed at baseline,2 weeks,4 weeks,and 8 weeks using HAMD-17.According to the reduction rate of HAMD-17≤20%and HAMD-17≥50%,the patients were divided into the ineffective group and the effective group.The steps of this study are as follows:1.Study on the incidence of depression disorder and the efficacy of SSRI drugs based on structural magnetic resonance imaging(1)Comparison of gray matter volume between depressed patients and normal controlsExcluding the unqualified brain structure image data,159 patients with first-episode depressive disorder and 53 normal controls were included.The gray matter volume was used as the index,and the difference between the gray matter volume of depressive disorder and normal controls was compared by independent sample t test,and the different brain areas were extracted.The diagnostic value of differential gray matter volume between depressive disorder and normal controls was analyzed by using subject operating characteristic curve(ROC).(2)Comparison of gray matter volume between the effective group and the ineffective groupThe gray matter volume difference between depressive disorder and normal control was used as the mask.Independent sample T test was used to compare the differences between the effective group(38 cases)and the ineffective group(41 cases)of SSRI drugs in depressive disorder,and time series of different brain areas were extracted.(3)Correlation between gray matter volume and clinical symptomsThe time series of gray matter volume differences between the effective group and the invalid group were extracted,and gender,age and education background were taken as covariables to make partial correlation analysis with HAMD-17 minus score.2.Study on the incidence of depression disorder and the efficacy of SSRI by diffusion tensor imaging(1)Comparison of white matter integrity between depressed patients and normal controlsExcluding unqualified brain image data,there were 105 patients with first-episode depressive disorder and 71 normal controls.Anisotropic score(FA)value was used as the index,and independent sample T test was used to compare the differences in white matter integrity between depressive disorder and normal controls,and different brain areas were extracted.The diagnostic value of DTI brain regions in depression disorders and normal controls was analyzed by using ROC.(2)Comparison of white matter integrity between the effective group and the ineffective groupDTI brain regions of depression disorder and normal control were used as the mask.Independent sample T test was used to compare the differences between the effective group(38cases)and the ineffective group(36 cases),and time series of different brain regions were extracted.(3)Correlation between white matter integrity and clinical symptomsTime series of DTI differences between the effective group and the ineffective group were extracted,and gender,age and education background were taken as covariables to conduct partial correlation analysis with HAMD-17 minus score rate.3.Study on the incidence of depression disorder and the efficacy of SSRI drugs based on resting state magnetic resonance imaging(1)Differences in resting brain function between depressed disorder and normal controlsExcluding unqualified brain image data,101 patients with first-episode depressive disorder and 53 patients with normal control were selected.Local consistency of resting state(ReHo)was used as the index.Independent sample t-test was used to compare the brain regions of ReHo between depressive disorder and normal control,and the brain regions were extracted.The diagnostic value of different brain regions of depression disorder and normal control ReHo in depression disorder was analyzed by using subject operating characteristic curve(ROC)Independent sample T test was used to compare the changes in brain connectivity and whole brain function between the depression disorder group and the normal control group,and to extract the different brain regions.The diagnostic value of different brain regions of depression disorder and normal control ReHo in depression disorder was analyzed by using subject operating characteristic curve(ROC).(2)Dfference in resting brain between the effective group and the ineffective groupDifferent brain regions of ReHo between the depression disorder group and the normal control group were taken as ROI.Independent sample T test was used to compare the different brain regions of ReHo between the effective group(32 cases)and the ineffective group(31 cases),and time series of different brain regions were extractedIndependent sample T test was used to compare the functional connectivity differences between the effective group(32 cases)and the ineffective group(31 cases),and time series of the differences were extracted.(3)Correlation between resting brain image and clinical symptomsTime series of ReHo differences between the effective group and the ineffective group were extracted,and gender,age and education background were taken as covariables to conduct partial correlation analysis with hamd-17 minus score rate.Time series of resting state functional connectivity differences between the effective group and the ineffective group were extracted,and gender,age and education background were taken as covariables to conduct partial correlation analysis with hamd-17 minus scoreResults:1.Study on the incidence of depression disorder and the efficacy of SSRI drugs based on structural magnetic resonance imaging(1)General demographic data and clinical characteristicsThere was no significant difference in age(t=-1.38,p=0.167)and gender(x~2=1.108,p=0.292)between the two groups(t=-1.206,p=0.229).There was no significant difference in age(t=-1.63,p=0.106),gender(x~2=3.65,p=0.056)and educational background(t=1.268,p=0.202)between the two groups.(2)Comparison of gray matter volume between depressed patients and normal controlsThe gray matter volume of the right superior temporal gyrus,right medial orbital gyrus and right inferior orbital gyrus decreased in patients with depression.Multiple comparisons were performed using the gaussian random fieldmethod(GRF),and the statistical level was P<0.05,which was statistically significant.According to ROC curve analysis,AUC of gray matter volume difference brain region was 0.316(95%CI=0.233-0.399).Poor ability to recognize depression.(3)Comparison of gray matter volume between the effective group and the ineffective groupAdjusted by AlphaSim(single voxel threshold of p<0.01 and cluster≥169.Compared with the effective group,the volume of right superior temporal gyrus in the ineffective group decreased significantly.(4)Correlation between gray matter volume and clinical symptomsWhen gender,age and education background were controlled,changes in gray matter volume were positively correlated with HAMD-17 score reduction rate(r=0.230,P=0.046).2.Study on the incidence of depression disorder and the efficacy of SSRI drugs based on diffusion tensor imaging(1)General demographic data and clinical characteristicsThere was no significant difference in age(t=0.315,p=0.753),gender(x~2=0.09,p=0.76)and educational background(t=0.188,p=0.096)between the two groups.There was no significant difference in age(t=-0.566,p=0.577),gender(x~2=1.724,p=0.189)and educational background(t=1.175,p=0.245)between the two groups.(2)Comparison of white matter integrity between depressed patients and normal controlsThe integrity of fibrous white matter in the left hippocampus and the parahippocampal gyrus,the left posterior cingulate gyrus and the right posterior cingulate gyrus was decreased in patients with depression.Multiple comparisons were performed using the gaussian random field method(GRF),and the statistical level was P<0.05,which was statistically significant.The AUC of gray matter volume difference brain region was 0.275(95%CI=0.194--0.356),which showed poor ability to recognize depression(3)Comparison of white matter integrity between the effective group and the ineffective groupAdjusted by AlphaSim(single voxel threshold of p<0.01 and cluster≥82).The white matter integrity damage was more serious in right posterior cingulate gyrus.(4)Correlation between gray matter volume and clinical symptomsWhen gender,age and education background were controlled,changes in white matter integrity were not correlated with HAMD-17 score reduction rate(r=0.205,P=0.076).3.Study on the incidence of depression disorder and the efficacy of SSRI drugs based on resting state magnetic resonance imaging(1)General demographic data and clinical characteristicsThere was no significant difference in age(t=-0.236,p=0.814),gender(x~2=0.104,p=0.747)and educational background(t=0.980,p=0.329)between the two groups.There was no significant difference in age(t=-0.566,p=0.577),gender(x~2=1.724,p=0.189)and educational background(t=1.175,p=0.245)between the two groups.(2)Comparison of resting brain function between depressed patients and normal controlsReHo increased in left and right parahippocampal gyrus in patients with depression.ReHo decreased in the right occipital gyrus,and the multiple comparisons were performed using the gaussian random field(GRF)method.The statistical level was P<0.05,which was statistically significant.In patients with depression disorder,functional connections between the right and left parahippocampal gyrus,the right middle occipital gyrus and the middle temporal gyrus were reduced.According to ROC curve analysis,AUC in the brain region with local consistency difference in resting state was 0.544(95%CI=0.445--0.644),and AUC in the brain region with local consistency difference in resting state was 0.822(95%CI=0.734--0.909).The mean time series of local consistency at rest state and the mean time series of functional connection were summarized by logistic regression and analyzed by ROC curve,and the AUC was 0.83.2(95%CI=0.752-0.911).(3)Comparison of resting brain function between the effective group and the ineffective groupAdjusted by AlphaSim(single voxel threshold of p<0.01 and cluster≥17).The local consistency of right parahippocampal gyrus decreased in the effective group.Adjusted by AlphaSim(single voxel threshold of p<0.01 and cluster≥10).The ineffective group had lower functional connectivity in the medial temporal gyrus.(4)Correlation between resting brain function and clinical symptomsWhen controlling for gender,age and education background,locally consistent changes in resting state were negatively correlated with hamd-17 minus score(r=-0.265,P=0.028).There was no correlation between the change of resting state functional connection and hamd-17 subtraction rate(r=0.116,P=0.386).Conclusion:1.There were abnormalities in gray matter,white matter and resting state function in patients with depression.2.Abnormal resting brain function may be a biological marker for the diagnosis of depression3.The degree of local consistency impairment in baseline resting state in patients with depression is expected to be a predictor of SSRI efficacy. |
PDF Full Text Request