Structural brain connectivity network alterations following mild traumatic brain injury
Timo Roine1, Mehrbod Mohammadian2,3, Timo Kurki2,3, Jussi Hirvonen2,3,4,5, and Olli Tenovuo2,3

1Turku Brain and Mind Center, University of Turku, Turku, Finland, 2Department of Neurology, University of Turku, Turku, Finland, 3Division of Clinical Neurosciences, Traumatic Brain Injury Centre, Turku University Hospital, Turku, Finland, 4Turku PET Centre, Turku University Hospital, Turku, Finland, 5Department of Radiology, University of Turku, Turku, Finland


We used graph theoretical analysis to investigate structural brain connectivity networks in mild traumatic brain injury (mTBI). Global and local measures of structural connectivity were investigated in acute/sub-acute and chronic phases after TBI. There were no statistically significant differences in the global network measures between patients and controls at either of the stages after TBI. Node-level differences were found between patients and controls in local efficiency, strength, and betweenness centrality in several brain regions. However, only betweenness centrality in the right pars opercularis endured the Bonferroni correction for multiple comparisons.


Traumatic brain injury (TBI) is a global public health burden with more than 50 million new cases each year (1), more than 90% of which are considered to be mild (mTBI)(2). TBI often results in cognitive and behavioral dysfunction, hence disrupting the normal functioning of the patients (1). Conventional neuroimaging methods often fail to show the subtle changes in the brain due to lack of sensitivity (3). Diffusion-weighted (DW) MRI is capable of detecting the subtle microstructural abnormalities associated with mTBI (4). Most of the studies, which have investigated these changes are based on diffusion parameters extracted from DW-MR images. However, it is shown that due to the heterogeneity of mTBI, modeling the brain networks could better characterize brain’s complex topology and regional connectivity (5). In this study, we investigated structural brain network alterations in patients with mTBI.


We investigated 85 patients with mTBI (47±20 years) and 30 orthopedic trauma controls (50±20 years) in this study. All subjects underwent MRI with Siemens Magnetom Verio 3T (Siemens Healthcare, Erlangen, Germany). T1-weighted and DW-MRI data were acquired using MPRAGE and echo-planar imaging sequences, respectively. Sixty-four gradient directions with a b-value of 1000 were used for the acquisition of the DW images. In addition, one b=0 s/mm2 image was also acquired. The DW-MR images were preprocessed by correcting for bias field, subject motion, eddy current induced, and echo planar imaging distortions by using MRtrix3 (6) and FSL (7). T1-weighted images were then nonlinearly co-registered to the DW images. Fiber orientations were then estimated by using constrained spherical deconvolution (CSD) and probabilistic tractography was performed in MRtrix3. Ten million anatomically feasible streamlines were generated to obtain whole brain tractograms (8). T1-weighted images were parcellated with FreeSurfer by using the Desikan-Killiany atlas (9). As a result, 84 gray matter areas were extracted and used as nodes of the structural brain networks, and the number of streamlines was used as an edge weight (Figure 1). Finally, graph theoretical analysis was performed by using Brain Connectivity Toolbox (10) and in-house MATLAB scripts to investigate global and local network differences between the groups. Seven global network measures (betweenness centrality (BC), normalized clustering coefficient, normalized global efficiency, normalized characteristics path length, small-worldness, degree, and strength) and three local network measures (BC, local efficiency, and strength) were investigated. Statistical analysis was performed using IBM SPSS (version 23, SPSS IBM, New York, NY) and a confidence interval of 95% was used to assess the significance of the results. Age and gender were used as covariates in the statistical analyses. Results were then corrected for multiple comparisons by using Bonferroni correction.


No significant differences were found in the global network measures between patients and controls at either of the stages after the injury. Several areas were found to have significant differences between patients and controls at both acute/sub-acute and chronic stage in all of the local network properties (Tables 1 and 2). However, only the right pars opercularis (p<0.00059) remained significant after the Bonferroni correction for multiple comparisons, as shown in Figure 2, and only at the chronic stage of TBI.


Our results indicate that the organization of the structural brain connectivity network was altered only regionally but not at a global level. Local network differences were observed both at acute/sub-acute and chronic stages. Right pars opercularis, left caudal middle frontal, and right caudate were affected at both stages of TBI indicating that network changes persist months after injury.


Our results show alterations in brain connectivity after an mTBI, which may well be an explanation for the cognitive deficits frequently seen in these patients.


T.R. received funding from the Emil Aaltonen Foundation (Finland), the Finnish Cultural Foundation (Finland), and Maud Kuistila Memorial Foundation (Finland). M.M received funding from the University of Turku graduate school.


1. Maas AIR, Menon DK, Adelson PD, et al. Traumatic brain injury: integrated approaches to improve prevention, clinical care, and research. Lancet. Neurol. 2017;16:987–1048.

2. Feigin VL, Theadom A, Barker-Collo S, et al. Incidence of traumatic brain injury in New Zealand: a population-based study. Lancet Neurol. 2013;12:53–64.

3. Shenton ME, Hamoda HM, Schneiderman JS, et al. A review of magnetic resonance imaging and diffusion tensor imaging findings in mild traumatic brain injury. Brain Imaging Behav. 2012;6:137–192.

4. Eierud C, Craddock RC, Fletcher S, et al. Neuroimaging after mild traumatic brain injury: Review and meta-analysis. NeuroImage Clin. 2014;4:283–294.

5. van der Horn HJ, Kok JG, de Koning ME, et al. Altered Wiring of the Human Structural Connectome in Adults with Mild Traumatic Brain Injury. J. Neurotrauma 2017;34:1035–1044.

6. Tournier JD, Calamante F, Connelly A. MRtrix: Diffusion tractography in crossing fiber regions. Int. J. Imaging Syst. Technol. 2012;22:53–66.

7. Jenkinson M, Beckmann CF, Behrens TEJ, Woolrich MW, Smith SM. FSL. Neuroimage 2012;62:782–790.

8. Smith RE, Tournier J-D, Calamante F, Connelly A. Anatomically-constrained tractography: improved diffusion MRI streamlines tractography through effective use of anatomical information. Neuroimage 2012;62:1924–38.

9. Desikan RS, Ségonne F, Fischl B, et al. An automated labeling system for subdividing the human cerebral cortex on MRI scans into gyral based regions of interest. Neuroimage 2006;31:968–980.

10. Rubinov M, Sporns O. Complex network measures of brain connectivity: uses and interpretations. Neuroimage 2010;52:1059–69.


Figure 1. Visualization of the differences in betweenness centrality of the structural brain connectivity networks between patients with mild TBI in the chronic stage and control subjects. Nodes of the network correspond to the 84 gray matter regions, and the edges are weighted with the number of streamlines. Size of the nodes corresponds to the volume of the gray matter region and the colour describes the statistical significance (p-value) of the differences.

Figure 2. Box plot of the betweenness centrality values in the right pars opercularis in patients with traumatic brain injury at the chronic stage and control subjects.

Table 1. Local network properties with significant difference between patients with mild traumatic brain injury at the acute/sub-acute stage and controls.

Table 2. Local network properties with significant difference between patients with mild traumatic brain injury at the chronic stage and controls.

Proc. Intl. Soc. Mag. Reson. Med. 27 (2019)