Pratik Mukherjee

Bruce McCandliss

DTI characterizes the microstructural properties of biologic tissues through measurements of the orientation-dependent diffusivity of water protons. This renders DTI particularly sensitive to human brain white matter pathology at microscopic spatial scales, because the longitudinal arrangement of the axolemma, microtubules, neurofilaments, and myelin sheaths along the direction of axonal fibers creates “anisotropy” of water diffusion. Water diffuses more freely along the orientation of fiber bundles than it does across fibers. For this reason, we felt that DTI would be useful for the study of mild traumatic brain injury (MTBI), including concussions. The most devastating long-term effects of MTBI on cognition and behavior are thought to be due to rotational acceleration forces causing shearing of white matter tracts, also known as “diffuse axonal injury” (DAI) or “traumatic axonal injury” (TAI). This shearing injury is associated with axonal misalignment, swelling, and retraction balls under light microscopy of human brain specimens. We therefore hypothesized that DAI would cause reductions in fractional anisotropy (FA), a measure of the microstructural “integrity” of white matter. We further postulated that the overall load of white matter microstructural injury on DTI in a particular patient would correlate with the degree of persistent postconcussive cognitive impairment.

We compared patients with MTBI and persistent symptoms months to years after injury to normative control subjects matched for age, gender, and years of education. From DTI measurements of FA in many white matter tracts, we found that, consistent with our hypothesis, most of these patients with chronic MTBI showed regions of reduced microstructural

integrity in the white matter, defined as FA decreased by more than 2.5 standard deviations below the mean of the normative controls at that region of interest. The white matter tracts that most commonly showed microstructural injury on DTI were the anterior corona radiata (ACR), the uncinate fasciculus (UF), the genu of the corpus callosum, and the cingulum bundle. Frontal and temporal lobe tracts were most likely to demonstrate abnormally low FA. Importantly, 10 of 11 patients with MTBI with no abnormalities on conventional 3T MRI sequences had DTI evidence of microstructural white matter injury, indicating that DTI could be more sensitive than even high-field structural MRI sequences. Finally, the load of white matter injury, measured as the number of tracts in each patient with abnormally low FA, showed a significant correlation with slowed reaction times in a cognitive control task, supporting our hypothesis that DTI findings would be associated with persistent postconcussive cognitive impairment. Strikingly, we also found that the number of focal lesions such as microbleeds on T2*-weighted gradient-echo imaging did not correlate significantly with reaction time in the patients with MTBI, indicating that DTI is a better predictor of cognitive processing speed.

The predominantly frontal and temporal pattern of microstructural white matter injury we found in the study fit well with the types of cognitive deficits typically suffered after concussion, including impairments of attention, memory, and information processing speed. In a follow-up study that we published later in 2008, we found that FA of the two most commonly injured tracts, the ACR and the UF, correlated with performance on tests of cognitive control and verbal memory, respectively. These findings provide insight into the relationships between microstructure damage to particular white matter tracts and specific functions impacted in TBI. Our work on DTI of TBI has led to further grant support from the McDonnell Foundation, the Department of Defense, the National Institutes of Health, and the GE Healthcare–National Football League Head Health Initiative. With this generous funding, we are currently embarking on large-scale studies of acute MTBI and of sports concussions, involving thousands of enrolled patients and controls using advanced 3T MRI methods, including DTI and resting-state fMRI, to evaluate structural and functional connectivity in the injured brain. These new studies aim to establish these techniques as routine clinical imaging tools for diagnosing MTBI, for predicting the occurrence of persistent postconcussive syndrome, and for triaging patients to the most effective interventions, including cognitive training and rehabilitation.

Read this article at AJNR.org . . .