Reduced small-world but hyper connected in cocaine users: A resting state fMRI study

Background and Aim : The changes in functional connections between brain networks in cocaine users may be related to cognitive and behavioral disruptions. As potential circuit level biomarkers of cocaine dependence, the alterations in functional connectivity strength may be used in treatment development outcome. Overall, models of cocaine addiction emphasize the role of disrupted frontal circuitry supporting cognitive control processes. In cocaine addicts, prefrontal and cingulate cortices, inferior frontal regions and cerebellum have shown hyper activation during response inhibition. In the past decade, resting state functional MRI (rsfMRI) connectivity analysis using graph metrics has been widely used to assess functional interactions between large-scale brain networks. A central goal of this study is to identify possible disruptions in interactions between resting-state networks in cocaine-dependent using graph theory. We hypothesize that alterations in interactions between large-scale brain networks in cocaine-dependents, will be significantly different from those observed in healthy subjects.Methods : The rsfMRI data were collected from 18 cocaine-dependent participants (age: 34.8 ± 8.7) in the NYU institute for Pediatric Neuroscience and 18 age-matched controls (age: 34.4 ± 8.9) from NewYork_a dataset of 1000 functional connectome. MPRAGE structural scans were processed using automated algorithms FreeSurfer version 6 software package. Data were preprocessed using FS-FAST (http://surfer.nmr.mgh.harvard.edu), includes slice-timing correction, head-motion correction, spatial smoothing, intensity normalization, registration to annatomical, normalization to common space, removal four first time points, fit trend with polynomial of order 2, Temporal filtering (0.009 to 0.1 HZ), removal CSF/white matter signals and six head motion parameters. We used Desikan atlas with 68 parcels and Pearson correlation coefficients to calculate resting state Functional Connectivity matrix (rsFC) for each subject separately. Then the group average rsFC matrix were converted into binary one using a fix density threshold from 0% to 50% with step one. The resulting binary matrix was used to build an undirected graph model of the brain network. Global graph metrics including clustering, local efficiency and small-worldness were calculated to investigate functional integration. Degree was calculated as functional segregation. Graph parameters were obtained using braph software (http://braph.org). We compared two groups with statistical hypothesis test (ttest).Results : Among 68 brain subdivisions, after controlling the resultant network density, local efficiency, clustering and small-worldness decreased in cocaine uses(CU) compared to control(CTL). Degree, reduced in frontal regions of CU, however inferior parietal, posterior cingulate, anterior cingulate and insula regions, showed stronger degree among CU than CTL.Conclusion : The increased degree strength in cocaine users’ brain may suggests an elevated dynamic resting state in addicted brain. The reduced local efficiency and reduced small-worldness suggest a loss of normal inter-regional communications that may underlie the loss of cognitive control and inhibition in drug addiction.

Keywords:

Cocaine addiction , resting state fMRI , functional connectivity , graph theory