Mapping PTSD-Related Brain Dysregulation With Connectome Gradient Analysis.

Abstract

Background: Hierarchical architecture is a fundamental organizational principle of the human brain. Previous studies have suggested that posttraumatic stress disorder (PTSD) may be characterized as disorders in the cerebral hierarchical organization. However, the specific abnormalities and underlying mechanisms are unclear.

Purpose: To investigate whether there are disorders of cerebral hierarchical organization in patients with PTSD and their underlying mechanisms of alteration.

Study type: Prospective, case control.

Filed strength/sequence: 3.0T, gradient echo echo-planar imaging sequence.

Subjects: Forty-nine patients with PTSD (11 males and 38 females; Clinician-Administered PTSD Scale (CAPS) score > 40) and 38 trauma-exposed controls (TEC) (13 males and 25 females; CAPS score < 40).

Assessment: Connectome gradient analysis was used to systematically examine disorders of cerebral hierarchical organization. Gradient metrics included range and variance of gradient scores. Graph theory analysis was also employed to explore underlying mechanisms of gradient abnormalities, and system segregation (quantifying the degree of separation between functional networks) and participation coefficients (PC) (quantifying the degree of connectivity that a given node has to other networks) were calculated.

Statistical tests: Two-sample t-tests were used to compare differences in gradient and graph theory metrics between groups. The association between gradient scores and CAPS scores was assessed using partial correlation analysis. p < 0.05 was set as the statistical significance threshold, with false discovery rate (FDR) correction.

Results: Compared with TEC, patients with PTSD showed significantly increased global gradient variance and altered gradient indicators in networks. At global and network levels, patients with PTSD exhibited significantly increased system segregation and significantly reduced PC, which were significantly associated with gradient variance (global system segregation: r = 0.84, global PC: r = 0.93, system segregation in SMN: r = 0.59, PC in DAN: r = -0.62 and PC in FPN: r = -0.53). Moreover, gradient scores in DAN (r = 0.319) and some regions of DMN (ANG.L: r = 0.294), SMN (PreCG.L: r = 0.319), and LIM (HIP.R: r = 0.319) were significantly correlated with CAPS score.

Data conclusion: This study, integrating connectome gradient analysis with graph theory, showed hierarchical disruptions across multilevel brain networks in PTSD, potentially explaining clinical symptoms such as hypervigilance and dissociation.

Evidence level: 2.

Technical efficacy: Stage 1.