Abstract:BACKGROUND:Numerous studies have revealed the abnormal static functional connectivity (FC) among different brain regions in patients with generalized anxiety disorder (GAD). However, little is known about the dynamic changes of FC in patients with GAD.METHODS:This study investigated the whole-brain dynamic changes of FC in patients with GAD by combining global FC density (FCD) and sliding window correlation analyses. The standard deviation of dynamic FCD (dFCD) was calculated to evaluate its temporal variability along time. Support vector regression was then employed to predict the symptom severity of patients based on abnormal dynamic connectivity patterns.RESULTS:The abnormal dFCD variability between 81 GAD patients and 80 healthy controls showed that the patients had higher dFCD variability in the bilateral dorsomedial prefrontal cortex (dmPFC) and left hippocampus while lower dFCD variability in the right postcentral gyrus. The abnormal dFCD variability of the left dmPFC is an important feature for anxiety prediction.LIMITATIONS:The selection of sliding window length remains controversial, and most of our patients have been treated with medications. Future studies are expected to rule out the potential confounding effects from applying different parameters of the sliding window and recruiting large samples of medication-free patients.CONCLUSION:The altered patterns of time-varying brain connectivity in the frontolimbic and sensorimotor areas may reflect abnormal dynamic neural communication between these regions and other regions of the brain, which may deepen our understanding of the disease.

Abnormal Dynamic Functional Connectivity Density in Patients with Generalized Anxiety Disorder.