Functional connectivity in mild traumatic brain injury

Abstract

Objectives: Research suggests that the majority of mild traumatic brain injury (mTBI) patients exhibit both cognitive and emotional dysfunction within the first weeks of injury, followed by symptom resolution 3-6 months postinjury. The neuronal correlates of said dysfunction are difficult to detect with standard clinical neuroimaging, complicating differential diagnosis and early identification of patients who may not recover. This study examined whether resting state functional magnetic resonance imaging (fMRI) provides objective markers of injury and predicts cognitive, emotional, and somatic complaints in mTBI patients semiacutely (<3 weeks postinjury) and in late recovery (3-5 month) phases.

Methods: Twenty-seven semiacute mTBI patients and 26 gender, age, and education-matched controls were studied. Fifteen of 27 patients returned for a follow-up visit 3-5 months postinjury. The main dependent variables were spontaneous fluctuations (temporal correlation) in the default-mode (DMN) and fronto-parietal task-related networks as measured by fMRI.

Results: Significant differences in self-reported cognitive, emotional, and somatic complaints were observed (all P < 0.05), despite normal clinical (T1 and T2) imaging and neuropsychological testing results. Mild TBI patients demonstrated decreased functional connectivity within the DMN and hyper-connectivity between the DMN and lateral prefrontal cortex. Measures of functional connectivity exhibited high levels of sensitivity and specificity for patient classification and predicted cognitive complaints in the semi-acute injury stage. However, no changes in functional connectivity were observed across a 4-month recovery period.

Conclusions: Abnormal connectivity between the DMN and frontal cortex may provide objective biomarkers of mTBI and underlie cognitive impairment.

Copyright © 2011 Wiley-Liss, Inc.

Figures

Figure 1

Regions demonstrating group differences in functional connectivity for the rostral anterior cingulate (rACC; Panel A, green coloring) and posterior cingulate (PCC; Panel B, green coloring) seeds during visit 1. Red coloring indicates regions where the absolute measure of functional connectivity (i.e., correlation or anticorrelation) was greater for mild traumatic brain injury patients (mTBI), whereas blue refers to regions where connectivity was greater for healthy controls (HC). The graphs depicting connectivity coefficients (error bars = one standard deviation) follow an identical color scheme and are grouped according to whether selected regions are from the default‐mode [DMN; PCC, right supramarginal gyrus (R SMG), and right superior frontal gyrus (R SFG)) or task‐related (TRN; ventral lateral prefrontal cortex and insula (VLPFC+), and bilateral inferior parietal lobule (IPL)] networks. Coordinates for slice locations are presented according to the Talairach atlas (R, right; L, left), and cluster volumes are presented in supplementary tables.

Figure 2

Regions demonstrating group differences in functional connectivity for the right lateral prefrontal cortex (R PFC; Panel A, green coloring) and right inferior parietal lobule (R IPL; Panel B, green coloring) seeds during the first visit. Red coloring indicates regions where the absolute measure of functional connectivity (i.e., correlation or anticorrelation) was greater for mild traumatic brain injury patients (mTBI), whereas blue refers to regions where connectivity was greater for healthy controls (HC). The graphs depicting connectivity coefficients (error bars = one standard deviation) follow an identical color scheme. Connectivity coefficients from the R PFC seed are depicted for the bilateral superior parietal lobule (SPL), left superior frontal gyrus (L SFG), and right inferior temporal gyrus (R ITG). The R IPL results include the right posterior parietal cortex (R PPC), bilateral medial frontal gyrus (MeFG), and right middle frontal gyrus (R MdFG). Coordinates for slice locations are presented according to the Talairach atlas (R, right; L, left; Z, axial view), and cluster volumes are presented in supplementary tables.

Figure 3

Selected functional connectivity data (error bars = one standard deviation) from the 15 mild traumatic brain injury patients (mTBI; red bars) and 15 matched healthy controls (HC; blue bars) who completed visits 1 (V1; dark‐colored bars) and 2 (V2; light‐colored bars). The different panels correspond to the different seeds used in the analyses, and abbreviations for regions are identical to Figures 1 and 2.

Figure 4

Panel A depicts five white matter tracts that have previously been implicated in connecting different nodes of the DMN and TRN, including the superior longitudinal fasciculus (SLF), cingulum bundle (CB), anterior corona radiata (ACR), external capsule (EC), and anterior limb of the internal capsule (ALIC). Coordinates for tract locations are presented according to the Talairach atlas. Panel B depicts fractional anisotropy (FA) and radial diffusivity (RD) values for both mild traumatic brain injury patients (mTBI; red bars) and healthy controls (HC; blue bars) at visit 1. Double asterisks denote significant group differences, whereas single asterisks denote statistical trends. Error bars equal one standard deviation.