| Objective:Major depression disorder(MDD)is a common,harmful disorder,and the pathological mechanism remains unclear.The primary pharmacotherapy regimen for MDD is selective serotonin reuptake inhibitors(SSRIs),but fewer than 40% of patients with MDD are in remission following initial treatment.In this study,we use magnetic resonance imaging based on the morphological structure of the cerebral cortex,including cortical thickness(CT),cortical surface area(SA),cortical curvature(CC)and cortical volume(CV),to explore the brain imaging pathogenesis of MDD.Further we analyze the morphological and structural differences of cerebral cortex with different treatment effects in patients with MDD treated with SSRI drugs for 2 weeks,and explore whether these morphological and structural characteristics of cerebral cortex can be used as predictors of efficacy SSRI drugs.Neuroimaging biomarkers of treatment efficacy can be used to guide personalized treatment in MDD.This study aims to provide help in the formulation of personalized drug treatment plans for patients with major depressive disorder.Methods:A total of 126 first-episode,drug-naive MDD patients(MDDs)and 71 healthy controls(HCs)were enrolled in our study.Demographic data were collected according to the self-made case report form(CRF)at the baseline of all subjects.Magnetic resonance imaging(MRI)scanning was performed for all the participants at baseline,and all imaging was processed using the DPABISurf software.All MDDs were treated with SSRIs,and symptoms were assessed at both the baseline and 2 weeks using the 17-item Hamilton Rating Scale(HAMD-17).According to HAMD-17 total score improvement from baseline to the end of 2 weeks,the MDDs were divided into the non-responder group(defined as ≤20% HAMD-17 reduction)and responder group(defined as ≥50%HAMD-17 reduction).The steps of this study are as follows:1.Comparison of the cortical morphological structure between the MDD group and the HC groupAfter excluding unqualified MRI data,there were 100 cases in MDD group and 52 cases in HC group.CT,SA,CC and CV were used as indicators,and surface-based morphometry(SBM)was used for analysis.The receiver operating characteristic(ROC)curve was used to analyze the diagnostic value of MDDs’ and HCs’ CT,SA,CC and CV for MDD.2.Comparison of the cortical morphological structure between the effective group and the ineffective groupThe different brain regions of CT,SA,CC and CV between the MDD group and the HC group were used as MASK,and the values of the different brain regions of the CT,SA,CC and CV were extracted from the effective group(29 cases)and the ineffective group(34 cases)respectively.Independent-sample t-test were used to compare the CT,SA,CC and CV values between the effective group and the ineffective group.ROC curves were used to analyze whether CT,SA,CC and CV were sufficient to distinguish SSRI effective from ineffective.3.Correlation between cortical morphological structure and reduction rate of clinical symptom scaleCorrelation analysis was performed for the effective group and the ineffective group separately to identify the relationship between CT,SA,CC,CV and SSRI treatment efficacy.The CT,SA,CC and CV differences between the effective group and the ineffective group were extracted,and the partial correlation analysis was performed with the 2-week reduction rate of HAMD-17(sex,age,and education were covariates).Results:1.General demographic data and clinical characteristicsThere were no significant differences in demographic data between the MDD group and the HC group.There were no significant differences in demographic data between the effective group and the ineffective group,and there were significant differences in the HAMD-17 score reduction rate and the HAMD-17 total score at the end of 2 weeks between the treatment effective group and the treatment ineffective group(pc0.001).2.Comparison of the cortical morphological structure between the MDD group and the HC group(1)Significant decreases were found in the CT of the right supplementary motor area(SMA)in MDDs(corrected by the Monte Carlo permutation correction,cluster-wise significant threshold at p<0.025 and vertex-wise threshold at p = 0.001),area under the curve(AUC)= 0.732 [95% confidence interval(CI)= 0.233–0.399],sensitivity = 76.9%,and specificity = 63%,and the cutoff value was 0.399.(2)Significant decreases were found in the CV of the right visual association cortex(VSA)in MDDs(corrected by the Monte Carlo permutation correction,cluster-wise significant threshold at p<0.025 and vertex-wise threshold at p = 0.001),area under the curve(AUC)= 0.676 [95% confidence interval(CI)= 0.602–0.750],sensitivity = 54.3%,and specificity = 76.5%,and the cutoff value was 0.308.(3)There was no significant difference in the SA and CC between the MDD group and the HC group.3.Comparison of the cortical thickness and cortical volume between the effective group and the ineffective group(1)In the effective group,the CT of the right SMA was significantly thinner than in the ineffective group.The results of ROC curve analyses of the effective group and ineffective were AUC = 0.885(95% CI = 0.803–0.968),sensitivity = 73.5%,and specificity = 96.6%,and the cutoff value was 0.701.(2)There was no difference in the CV of the right VSA between the effective group and the ineffective group.4.Correlation between cortical thickness and reduction rate of clinical symptom scaleThere was a negative correlation(r =-0.373,p = 0.044)between the CT of SMA(0weeks)and HAMD-17 reductive rate(2 weeks)in the effective group.There was no correlation(r =-0.252,p = 0.150)between the CT of SMA(0 weeks)and HAMD-17 reductive rate(2 weeks)in the ineffective group.Conclusion:1.Cortical thickness and cortical volume are abnormal in patients with major depressive disorder.Lower cortical thickness of the right supplementary motor area patients may be a neuroimaging biomarker for major depressive disorder diagnosis.2.A greater extent of thinner cortical thickness in the right supplementary motor area may predict good treatment outcome of SSRIs. |
PDF Full Text Request