| Objective:Major depressive disorder(MDD)is a highly recurrent psychiatric disorder characterized by abnormalities in negative emotional processing.It is widely recognized that bias in processing negative emotional stimuli plays a crucial role in the onset and development of MDD.However,the specific neuroimaging features associated with negative emotional processing during different stages of MDD remain unclear.Building upon the aforementioned context,this study aimed to investigate the neurocircuitry alterations related to negative emotional processing,independent of clinical state(trait-related)and changes accompanying depressive symptom remission(state-related),using an emotional face matching paradigm and magnetic resonance imaging(MRI).Furthermore,we aimed to examine which neuroimaging features could predict symptom remission and treatment response in MDD patients,providing empirical evidence for elucidating the psychopathological mechanisms underlying MDD.Method:(1)Study 1 is a neuroimaging study of negative emotional processing in which a total of 49 first-episode drug-na(?)ve major depressive disorder(c MDD)patients,43 remitted depressed(RD)patients who were untreated,and 99 healthy controls(HC)completed an emotional facial matching functional magnetic resonance imaging(f MRI)task.The thalamus,amygdala,insula,ventrolateral prefrontal cortex(VLPFC),and dorsolateral prefrontal cortex were selected as regions of interest(ROIs).Brain activation analysis and parameter-based empirical Bayesian dynamic causal modeling were employed to evaluate the trait-related and state-related characteristics of brain functional(activation and effective connectivity)abnormalities during the negative emotional processing in patients with MDD.(2)Study 2 is a neuroanatomical study related to the negative emotional processing,in which a total of 126 c MDD patients,66RD patients,and 230 HC completed magnetic resonance imaging(MRI)scans.Based on the results of Study 1,ROIs were defined for group differences analysis of morphometric measures(cortical thickness,cortical surface area,and gray matter volume)of brain structures.(3)Study 3 is based on the results from Studies 1 and 2 to construct a predictive model for MDD.In this section,a total of 54 drug-na(?)ve first-episode MDD patients who completed an emotional facial matching f MRI paradigm before treatment and then received selective serotonin reuptake inhibitor(SSRI)antidepressant treatment.The severity of depression was assessed using the 17-item Hamilton Depression Rating Scale(HAMD-17)at baseline and after 8 weeks of treatment(with treatment response defined as a reduction of≥50%in HAMD-17 scores and remission defined as HAMD-17≤7).A predictive regression models were constructed to identify neuroimaging features related to negative emotional processing that can predict the degree of improvement in depressive symptoms before and after an 8-week treatment period.Furthermore,develop prognostic regression models to examine which features effectively differentiate treatment response(remission vs.non-remission)and determine the neuroimaging features that can accurately predict depressive symptom remission and treatment efficacy.Results:(1)Brain activation analysis revealed that c MDD and RD groups exhibited significantly higher activation in bilateral thalamus,insula,and VLPFC than HC group(p<0.05,small volume correction)when matching negatively emotional faces vs.neutral stimuli.In addition,c MDD group showed significantly higher activation in bilateral amygdala compared to HC group(p<0.05,small volume correction).Further effective connectivity analysis showed that compared to HC group,c MDD and RD groups exhibited significantly weakened effective connectivity from VLPFC to insula under negative emotional modulation(posterior probability>95%,all validated by cross-validation).Moreover,c MDD group showed significantly enhanced effective connectivity from amygdala to insula and increased self-inhibition of VLPFC(posterior probability>95%,validated by cross-validation).(2)Both c MDD and RD groups had significantly thinner cortical thickness in bilateral VLPFC and insula compared to HC group(p<0.05,permutation correction),while the corresponding cortical surface area was significantly larger than HC group(p<0.05,permutation correction).Additionally,c MDD and RD groups exhibited significantly smaller gray matter volume in several subregions of the thalamus than HC group(p<0.05,FDR correction).(3)Among the 54c MDD patients,25 achieved remission after an 8-week treatment with SSRIs.Multiple linear regression analysis demonstrated a significant model fit with R~2 of 0.26 and adjusted R~2 of 0.22(F=5.37,p=0.04).The baseline f MRI activation of left amygdala in the remitted c MDD group had a significant effect on the severity of depression score between baseline and 8 weeks(p=0.04).Furthermore,logistic regression analysis demonstrated that the prognostic model,which incorporated both f MRI and clinical data,had a better fit(AIC=24.86,BIC=26.85).The left amygdala f MRI activation values at baseline showed a marginal significance(OR=3.03,p=0.067).Conclusion:(1)Aberrant enhancement of bilateral thalamus,insula,and dorsolateral prefrontal cortex activation,as well as weakened effective connectivity from dorsolateral prefrontal cortex to insula,observed in patients with major depressive disorder(MDD)during negative emotional processing,indicates trait-related functional impairment in the brain.These regions also exhibit corresponding structural abnormalities.(2)Abnormal hyperactivation of bilateral amygdala,enhanced effective connectivity from amygdala to insula,and suppressed self-connectivity within the dorsolateral prefrontal cortex during negative emotional processing are indicative of state-related functional impairments in the brains of individuals with MDD.Particularly,the degree of left amygdala activation during negative emotional processing may serve as a neuroimaging marker for predicting symptom remission and treatment efficacy in MDD patients. |
PDF Full Text Request