Abstract Functional MRI (fMRI) provides complementary insights into brain network organization during rest (rs-fMRI) and external stimulation (t-fMRI). While rs-fMRI reveals intrinsic connectivity patterns, t-fMRI reflects stimulus-dependent network responses. However, how these states relate to each other in animal disease models remains incompletely understood. We performed sequential acquisition of rs-fMRI and t-fMRI during low- and high-intensity mechanical paw stimulation in a rat model of post-surgical pain (PSP) and SHAM controls. Functional connectivity and network organization were analyzed using network-based statistics and graph-theoretical metrics. In addition, supervised classification of node-level network parameters was performed using linear discriminant analysis (LDA), and regional contributions to network separation were quantified using Mahalanobis distance metrics. Global connectivity strength and small-world organization were preserved across different imaging conditions in both groups, indicating stable network topology. However, multivariate classification revealed clear modality-dependent network signatures in SHAM animals that were markedly reduced in PSP. Region-wise Mahalanobis analyses showed that stimulus-related network shifts were more heterogeneous in SHAM animals, with higher dispersion and recurrent regional “hotspots”, whereas PSP animals exhibited more uniform and spatially diffuse responses. These findings indicate that post-surgical pain primarily affects the flexibility of regional network reconfiguration rather than global topology. Combining rs- and stimulus-based fMRI thus provides complementary insight into disease-related network alterations that are not detectable from resting-state data alone.
Pradier et al. (Tue,) studied this question.