25 Associations between Diffusion Kurtosis Imaging, Tau, and Cognitive Outcomes in TBI

Abstract

Objective:

Determine associations between cognitive outcomes in remote TBI (i.e., at least 6 months post injury), a blood marker of neural degeneration (i.e., Tau), and diffusion kurtosis imaging (DKI) measures (e.g., mean or radial kurtosis). Because DKI imaging is sensitive to the environmental complexity of the imaged area, we sought to investigate regions known to be associated with the cognitive and emotional sequalae of TBI, such as the anterior thalamic radiations, uncinate fasciculus, and the corpus callosum.

Participants and Methods:

41 individuals with mild-to-moderate TBI and a mean age(SD) of 36.1(10.4) years underwent DKI, a blood draw, and neuropsychological assessments. 23 healthy controls (HC) with a mean age(SD) of 35.2(15.2) years underwent the blood draw and assessments, but no imaging. Higher diffusion kurtosis indicates more restricted diffusion, possibly due to greater complexity within the imaged region. Thus, in the context of TBI, DKI can be used as a proxy measurement for biological processes that alter the complexity of imaged environments, such as reactive gliosis. Some people show cognitive deficits long after TBI and this could be associated with increased inflammation and membrane protein aggregates in damaged brain regions. We used bivariate correlations and general linear models to investigate the association of mean kurtosis (MK) in long white matter tracts and Tau (total or phosphorylated) to color-word Stroop scores; a measure of fronto-subcortical function.

Results:

In patients with TBI, MK was significantly associated with serum total Tau (TTau) in the right (r=-0.396) and left (r=-0.555) uncinate fasciculus (UF), right (r=-0.402) and left (r=-0.504) anterior thalamic radiations (ATR), and the genu (r=-0.526) and body (r=-0.404) of the corpus callosum (CC). TTau had a significant association with word Stroop scores, F(1,63)=-2.546, p=0.013. However, there was no significant effect of group (i.e., TBI or HC), F(2,63)=-0.426, p=0.672, on cognitive performance. When models were implemented that included both TTau and MK in either the UF or ATR as explanatory variables to predict word Stroop scores, TTau levels and MK in the right UF explained a significant amount of the variance in Stroop performance, F(1,29)=2.215, p=0.025. Further, there was also a significant association between radial kurtosis in the right UF and Stroop word scores (r= 0.366).

Conclusions:

Our results show that an indicator of biological complexity (DKI) in cognitively important brain regions is associated with cognitive performance and Tau in patients with remote mild-to-moderate TBI. The UF is a critical fronto-temporal/subcortical pathway that has previously been implicated in the manifestation of executive dysfunction and mood dysregulation in TBI. Tau is an important marker of neurodegeneration implicated in Alzheimer’s disease, Parkinson’s disease, and chronic traumatic encephalopathy (CTE), and DKI is potentially sensitive to markers of neurodegeneration. The association of Tau and DKI measures is novel and shows concordance between blood and brain imaging markers and cognitive performance in patients with mild to moderate TBI.