BACKGROUND

Disturbance of the triple network model was recently proposed to be associated with the occurrence of posttraumatic stress disorder (PTSD) symptoms. Based on resting-state dynamic causal modeling (rs-DCM) analysis, we investigated the neurobiological model at a neuronal level along with potential neuroimaging biomarkers for identifying individuals with PTSD.

METHODS

We recruited survivors of a devastating typhoon including 27 PTSD patients, 33 trauma-exposed controls (TECs), and 30 healthy controls without trauma exposure. All subjects underwent resting-state functional magnetic resonance imaging. Independent components analysis was used to identify triple networks. Detailed effective connectivity patterns were estimated by rs-DCM analysis. Spearman correlation analysis was performed on aberrant DCM parameters with clinical assessment results relevant to PTSD diagnosis. We also carried out step-wise binary logistic regression and receiver operating characteristic curve (ROC) analysis to confirm the capacity of altered effective connectivity parameters to distinguish PTSD patients.

RESULTS

Within the executive control network, enhanced positive connectivity from the left posterior parietal cortex to the left dorsolateral prefrontal cortex was correlated with intrusion symptoms and showed good performance (area under the receiver operating characteristic curve = 0.879) in detecting PTSD patients. In the salience network, we observed a decreased causal flow from the right amygdala to the right insula and a lower transit value for the right amygdala in PTSD patients relative to TECs.

CONCLUSION

Altered effective connectivity patterns in the triple network may reflect the occurrence of PTSD symptoms, providing a potential biomarker for detecting patients. Our findings shed new insight into the neural pathophysiology of PTSD.