Behavioral Significance of Altered Resting State Network in Fibromyalgia

Abstract

Fibromyalgia (FM) is a chronic widespread pain syndrome characterized by increased sensitivity to painful stimuli (hyperalgesia) as well as nonpainful stimuli (allodynia). Neuroimaging studies have demonstated that fibromyalgia is characterized by augmented sensory processing mediated by abnormal central nervous system, although the underlying mechanism of pain is not fully understood. Recent studies have revealed intrinsic connectivity alteration during resting state associated with clinical symptoms in FM patients. However, the network construction is inherently affected by thresholding, which is operator-dependent. In this study, we employed persistent brain network homology which is a multiple scale network modeling framework building the brain network at every threshold, thus not requiring thresholding to understand the topological features of brain network in fibromyalgia (FM). Spontaneous MEG activity was analyzed for 200 seconds in 17 healthy controls (HCs) and 18 FM patients. All participants were instructed to close their eyes at rest. Resting MEG data were filtered into theta (4-7 Hz), alpha (8-12 Hz), beta (13-30 Hz), and gamma (31-48 Hz) bands, and we constructed source activities in 76 nodes based on Automated Anatomical Labeling (AAL) in each frequency band using a beamformer. To construct brain network using persistent brain network homology, the distance matrix (1 – absolute value of the correlation coefficient) was calculated. Then, barcode, single linkage dendrogram and single linkage matrix (SLM) were generated for each subject in two groups based on the proposed modeling framework. Barcode represents the global topological change at every threshold. Single linkage dendromgram represents the local network features that show what regions could be merged over different thresholds. Single linkage distance (SLD) is the component of single linkage dendrogram and used to estimate functional connectivity. In theta band, the slope of decrease in the number of connected components in barcodes showed steeper in HC, suggesting FM patients had decreased global connectivity than the healthy controls. Patients with FM had reduced connectivity within default mode network region (mPFC, ACC, PCC, and precuneus). They also disclosed longer SLDs between middle temporal lobule and visual area (calcarine fissure, cuneus, lingual gyrus, superior occipital cortex, and fusiform gyrus) in theta. The longer SLDs between MTG and visual cortex were correlated with pain duration. Our results provided disrupted intrinsic connectivity within resting state network of FM patients in theta band. Moreover, our findings suggest that the persistent pain in FM patients would be associated with the connectivity within the DMN and sensory network.