Global and local properties of brain resting state networks in chronic pain.

I. CIFRE; DANIEL FRAIMAN; PABLO BALENZUELA; V. APKARIAN; DANTE CHIALVO

Chicago, Illinois, EEUU

Conferencia; Meeting of Neuroscience 2009; 2009

Society for Neuroscience, USA.

           Previous findings of disrupted correlations in chronic back pain patients (Baliki et al, J.N. 2008) suggest that brain resting state networks can be altered in chronic pain patients. The aim of the present study is to examine the integrity of global and local correlation properties of brain resting-state networks in chronic pain patients. The emphasis of the study is the identification of robust markers able to differentiate health from disease. Data analyzed comprises fMRI scans from subjects resting quietly with closed eyes. First, probabilistic independent component analyses (PICA) were used to identify resting-state networks from low-frequency fluctuations of fMRI BOLD time-series in each subject. Then the spatial group maps for the default-mode network, the core network, and two lateralized parietal-frontal networks were extracted and compared with controls. Second, voxel-to-voxel linear BOLD correlations where computed in all patients and groups and its individual distributions calculated. We found severe alterations in the resting state networks spatial maps both in the chronic back pain group (n=18) as well as in the CRPS patients (n=22). Surprisingly, these alterations were found to correlate very well with changes in the skewness of the distribution of the voxel-to-voxel correlation, a simple but significant global estimator of functional connectivity in this case. These findings confirm that spontaneous pain in CBP and CRPS patients alters cortical network dynamic beyond the pain matrix itself, a change that can be detected at the level of the correlations of brain resting activity.