Alterations of static and dynamic functional connectivity of the nucleus accumbens subregions may be associated with chronic process of migraine: a resting-state fMRI study

The nucleus accumbens (NAc) plays an important role in the chronic pain process and consists of core and shell subregions with distinct connectivity patterns. However, their roles in migraine progression remain unclear. This study explored static and dynamic functional connectivity (FC) patterns of NAc subregions across healthy controls (HCs), patients with episodic migraine (EM) and patients with chronic migraine (CM). 97 participants were enrolled: 70 migraine patients (24 EM, 46 CM) and 27 HCs. All underwent functional magnetic resonance imaging on a GE 3.0T system. Demographic and clinical characteristics were collected. The NAc core and shell were defined as ROIs, and static and dynamic seed-based FC analyses were performed across groups. Partial correlations between FC values and clinical variables were conducted. Significant static FC and dynamic FC were selected through multivariable logistic regression. Receiver operating characteristic curves were used to assess their diagnostic performance for distinguishing CM patients. Compared with HCs, EM showed increased sFC between the right NAc core and right supplementary motor area (SMA), whereas CM exhibited decreased sFC, particularly between the left NAc shell and the left fusiform gyrus and between the right NAc core and the left cerebellum crus II. Relative to EM, CM showed reduced sFC mainly between the left NAc shell and the bilateral precentral gyrus and right superior frontal gyrus (SFG), and between the right NAc core and right SMA, right inferior frontal gyrus, and left inferior temporal gyrus (ITG). The dFC analysis revealed decreased connectivity between the left NAc shell and left precentral gyrus and right SFG, and between the right NAc core and left SMA. Reduced sFC between the right NAc core and left ITG and dFC between the left NAc shell and right SFG effectively distinguished CM from EM. The combined FC calculated from sFC and dFC further improved discrimination (AUC = 0.919). NAc subregions in CM showed reduced connectivity with multiple brain regions, mainly in sensorimotor, executive control, and default mode networks. Significant sFC and dFC of the NAc subregions may serve as potential imaging biomarkers for CM.