Default mode network changes in fibromyalgia patients are largely dependent on current clinical pain

Abstract

Differences in fMRI resting-state connectivity of the default mode network (DMN) seen in chronic pain patients are often interpreted as brain reorganization due to the chronic pain condition. Nevertheless, patients' pain at the time of fMRI might influence the DMN because pain, like cognitive stimuli, engages attentional mechanisms and cognitive engagement is known to alter DMN activity. Here, we aimed to dissociate the influence of chronic pain condition (trait) from the influence of current pain experience (state) on DMN connectivity in patients with fibromyalgia (FM). We performed resting-state fMRI scans to test DMN connectivity in FM patients and matched healthy controls in two separate cohorts: (1) in a cohort not experiencing pain during scanning (27 FM patients and 27 controls), (2) in a cohort with current clinical pain during scanning (16 FM patients and 16 controls). In FM patients without pain during scanning, the connectivity of the DMN did not differ significantly from controls. By contrast, FM patients with current clinical pain during the scan had significantly increased DMN connectivity to bilateral anterior insula (INS) similar to previous studies. Regression analysis showed a positive relationship between DMN-midINS connectivity and current pain. We therefore suggest that transient DMN disruptions due to current clinical pain during scanning (current pain state) may be a substantial contributor to DMN connectivity disruptions observed in chronic pain patients.

Fig. 1.. DMN connectivity in FM patients with and without current clinical pain during the resting-state FMRI scan.

A) Patients with current pain have increased DMN-left aINS and DMN-right aINS connectivity compared to matched controls (z > 2.3, p > 0.05 cluster-corrected across INS ROI). Inset shows patients > controls (scanner pain-free cohort), uncorrected and does not survive cluster correction; B) Single subject DMN-INS connectivity values averaged across significant aINS clusters for each group of each cohort. P-values indicate post-hoc comparisons between groups of each cohort (black underline) and between cohorts (red underline; p < 0.001 for (patients > controls)_{scanner pain}-(patients > controls)_{scanner pain-free}, two-sample t-test on difference scores); C) In patients with current pain, DMN-left midINS connectivity is positively correlated with current pain score (z > 2.3, p > 0.05 cluster-corrected across INS ROI), scatter plot shows mean connectivity value across the significant cluster for each patient plotted against their current pain score (NRS, numerical rating scale); D) Cluster showing the relationship with current pain is adjacent to and overlapping with the cluster showing group differences in DMN connectivity. MNI standard brain, showing values above z = 2.3, but all results (except inset) are significant at z > 2.3, p < 0.05 cluster-corrected for spatial extent.

Fig. 2.. DMN-INS connectivity is related to chronic pain duration in scanner pain-free patients.

In patients who were pain-free during the scan, DMN-mid/pINS connectivity is positively correlated with pain duration (yrs, years) at z > 2.3, p < 0.05 cluster-corrected for spatial extent; scatter plot shows mean connectivity value across the significant cluster for each patient plotted against their pain duration in years. MNI standard brain, showing values above z = 2.3.