Abstract Details

To assess the relationship between white matter hyperintensities (WMH), diffusion tensor imaging (DTI), and blood-based biomarkers among patients with traumatic brain injury (TBI).

Patients with non-penetrating TBI were enrolled within 24 hours of admission. Participants underwent MRI at two weeks and blood was collected at day one, day three, two weeks, and six months post-injury. Following preprocessing of DTI data, mean fractional anisotropy (FA) and mean diffusivity (MD) were calculated. WMH were counted in juxtacortical, subcortical, and periventricular areas. Blood was processed using Simoa HD-X Neuro4plex assay to obtain concentrations of GFAP, NfL, and UCH-L1.

100 individuals with TBI were included (mean age 38.3 years, 76% male, 52% black; median Glasgow Coma Scale score 15 (IQR:14-15)). Of these individuals, 39% had WMH identified by a neuroradiologist. Those with WMH had significantly lower FA (p=0.007) and higher MD (p=0.04) compared to those without WMH. More specifically, Spearman’s correlations demonstrated significant relationships (p<0.05) between MD and number of WMH in subcortical areas (r=0.28) and between FA and number of WMH in juxtacortical (r=-0.34), subcortical (r=-0.42), and periventricular areas (r=-0.34).  Logistic regression, controlling for age, revealed no significant relationships between WMH presence and any blood-biomarkers of interest. Zero inflated negative binomial regressions, controlling for age, also revealed no significant relationships between number of WMH in juxtacortical regions and GFAP or NfL. There was a significant relationship for Day 1 UCH-L1 where, for every one unit increase, there is 1.36 times more WMH (p=0.023, 95%CI: 1.04, 1.77).

WMH are associated with more severe and widespread white matter damage as measured by DTI. Blood biomarkers may also provide both complementary and distinctive information regarding physiological processes following TBI.