Cortical functioning within the default mode network (DMN) and salience network (SN) is altered in chronic pain patients. The mechanisms underlying these alterations are unknown, but a novel unexamined source is cross-network communication. Aberrant functional connectivity (FC) between the DMN and SN, whose activity is normally anticorrelated, reflects disease severity in many brain disorders. Further, stronger FC between the posterior cingulate cortex (PCC) and anterior insula has been reported in chronic pain, pointing to abnormal DMN-SN interactions. Here, we tested the hypothesis that cross-network FC between the DMN and SN is abnormal in chronic pain, and is related to pain and associated symptoms. We used resting state fMRI to examine FC within and between the DMN and SN in 20 patients with chronic pain due to ankylosing spondylitis and 20 healthy controls. A whole-network analysis revealed that compared to healthy controls, patients exhibited less anticorrelated FC between the SN and DMN, and the degree of cross-network abnormality tracked pain and disease-related symptoms. This suggests that cross-network FC is a metric of functional brain abnormality in chronic pain. In a complementary seed-based analysis, the PCC was strongly connected with the SN and weakly connected with the DMN in chronic pain compared to healthy controls, suggesting that the PCC acts as a hub for altered network interaction. Sensorimotor cortex cross-network FC correlated with measures of physical function, suggesting that physical functioning also impacts brain network interaction in chronic pain. Our study implicates altered communication between brain networks as a key factor underlying chronic pain.

Chronic pain is a common occurrence in multiple sclerosis (MS) that severely affects quality of life, but the underlying brain mechanisms related to these symptoms are unknown. Previous electroencephalography studies have demonstrated a role of alpha-band and beta-band power in pain processing. However, how and where these brain signals change in MS-related chronic pain is unknown. Here, we used resting state magnetoencephalography to examine regional spectral power in the dynamic pain connectome-including areas of the ascending nociceptive pathway, default mode network (DMN), and the salience network (SN)-in patients with chronic MS pain and in healthy controls. Each patient was assessed for pain, neuropathic pain (NP), and pain interference with activities of daily living. We found that patients with MS exhibited an increase of alpha-band power and a decrease of beta-band power, most prominently in the thalamus and the posterior insula of the ascending nociceptive pathway and in the right temporoparietal junction of the SN. In addition, patients with mixed-NP exhibited slowing of alpha peak power within the thalamus and the posterior insula, and in the posterior cingulate cortex of the DMN. Finally, pain interference scores in patients with mixed-NP were strongly correlated with alpha and beta peak power in the thalamus and posterior insula. These novel findings reveal brain mechanisms of MS-related pain in the ascending nociceptive pathway, SN, and DMN, and that these spectral abnormalities reflect the impact of pain on quality of life measures.