Brain-behavior relationships in young traumatic brain injury patients: Fractional anisotropy measures are highly correlated with dynamic visuomotor tracking performance

Abstract

Traumatic brain injury (TBI) patients have a high incidence of eye–hand coordination deficits. Diffuse axonal injury is common in TBI and is presumed to contribute to persistent motor problems. Using Diffusion Tensor Imaging (DTI), this study sought to identify changes in (sensori)motor white matter (WM) pathways/regions in a TBI group during the chronic recovery stage. A secondary objective was to examine the relationship between WM integrity and upper-limb visuomotor tracking performance. A young TBI (n = 17) and control (n = 14) group performed a dynamic tracking task, characterized by increasing information processing speed and predictive movement control. DTI scans were administered along with standard anatomical scans. The TBI group was found to perform inferior to the control group on the tracking task. Decreased fractional anisotropy was found in the TBI group in dedicated pathways involved in transmission of afferent and efferent information, i.e., corticospinal tract, posterior thalamic radiation, and optic radiation, due to increased diffusivity parallel and perpendicular to axonal fibre direction. This decrease in WM integrity was associated with inferior visuomotor tracking performance. Moreover, discriminant function analysis demonstrated that the model, based on the combined application of DTI and behavioral measures, was the most effective in distinguishing between TBI patients and controls. This study shows that specific eye–hand coordination deficits in a young TBI group are related to microstructural abnormalities in task-specific cerebral WM structures. Measures of white matter integrity are potentially important biomarkers for TBI that may support prognosis of motor deficits.

Introduction

Traumatic brain injury (TBI) is a major health problem in children and adolescents, demanding early and accurate diagnosis, assessment of severity, and prognostication of likely outcomes (Langlois, Rutland-Brown, & Thomas, 2005). Traditional imaging techniques, such as computerized tomography (CT) and conventional (T1, T2, and T2*-weighted) magnetic resonance imaging (MRI) have proven to be highly effective in identifying macroscopic lesions, which is a necessary component in managing acute trauma. However, they show marked limitations in assessing microscopic lesions, such as those associated with diffuse axonal injury (DAI). Diffusion tensor imaging (DTI) is more sensitive to changes in the microstructure of white matter, and shows considerable improvements in sensitivity, specificity, and predictive value.

DTI is a relatively new technique in structural neuroimaging. It allows the estimation of the orientation of fiber bundles in white matter (WM) on the basis of the diffusion characteristics of water. The directionality of diffusion can be quantified by the fractional anisotropy (FA) index and is determined by several factors, including the thickness of the myelin sheath and of the axons as well as the organization of the fibers and properties of the intracellular and extracellular space around the axon. FA values range from 0 to 1, where 0 represents maximal isotropic diffusion (i.e., free diffusion in all directions) and 1 represents maximal anisotropic diffusion (i.e., movement parallel to the major axis of a WM tract). Changes in tissue structure (in this case axonal injuries) can lead to a modification of the degree directionality, which can be detected by DTI. Therefore, in damaged WM tracts of patients with TBI one would expect to find changes in the anisotropy of diffusion in comparison with healthy subjects.

DTI has already produced promising results in assessing DAI in adults with TBI. DTI has been shown to provide evidence of axonal injury in the presence of normal standard MR imaging in mild TBI (Arfanakis et al., 2002, Inglese et al., 2005, Nakayama et al., 2006). DTI measures of anisotropy have been shown to correlate with initial Glasgow Coma Scale score and outcome scores in TBI adults (e.g., Benson et al., 2007, Kraus et al., 2007, Salmond et al., 2006). Consequently, FA analysis may be a valuable noninvasive biomarker of tissue injury for use in predicting outcomes (Huisman et al., 2004).

The literature involving the application of DTI in pediatric TBI is limited but shows promise. Multiple groups have demonstrated reductions in integrity in the chronic phase of mild, moderate, and severe pediatric TBI (Wilde et al., 2006, Wozniak et al., 2007, Yuan et al., 2007). Moreover, a few studies have focused on regional anisotropy changes and their relationship to neuropsychological/cognitive outcome in pediatric TBI. For example, Yuan et al. (2007) demonstrated in children at least 1 year after TBI that FA across a number of white matter tracts correlates with initial rating of severity of TBI. Levin et al. (2008) have also demonstrated a relationship between a composite FA score obtained 3 months after injury and global outcome, cognitive processing speed, and speed of resolving interference. Further support for the relationship between FA and specific chronic cognitive outcomes comes from studies of children demonstrating correlation between regional FA in frontal and ‘supracallosal’ areas and measures of executive functioning and motor speed (Wozniak et al., 2007) and correlation between corpus callosum FA and processing speed (Wilde et al., 2006).

To the best of our knowledge, the present study is the first to relate DTI measures with motor functioning in a young TBI group. Movement disorders are related to gross-motor (such as balance, running speed, strength, coordination) as well as fine-motor (upper-limb speed, dexterity, reaching) function, which may lead to long-term impairments of motor proficiency (Kuhtz-Buschbeck et al., 2003). Here, we used detailed kinematic measures to assess performance on a dynamic eye–hand coordination task, requiring gradually increasing perceptual information processing speed and predictive movement control. Eye–hand coordination deficits are of central concern in rehabilitation because they limit flexible behavioral adjustments to the ever-changing environment. Previous behavioral studies revealed that predictive visual and manual tracking of a target is impaired in children (Caeyenberghs et al., 2009) and adolescents with TBI (Heitger et al., 2006, Suh et al., 2006a, Suh et al., 2006b).

Here, we studied WM changes in dedicated anatomical structures associated with visuomotor performance. Moreover, we applied an ROI-based method instead of a voxel-wise analysis, which is not hypothesis-driven and sensitive to user-defined parameter settings (Jones, Symms, Cercignani, & Howard, 2005) and can therefore end up in scattered, underpowered, and difficult-to-interpret results. We hypothesized that the TBI as compared to control group would show deficient tracking performance, i.e., they would make less use of feedforward (predictive) control and rely more on (slower) feedback control instead. From a structural perspective, decreases in WM integrity of afferent (medial lemnicus, posterior thalamic/optic radiation) and efferent (corticospinal tract) pathways were predicted in the TBI group. Finally, significant correlations between tracking performance and structural brain deficits were expected in the TBI group.