Impact of mild head injury on diffusion MRI brain characteristics in midlife: Data from the PREVENT Dementia Study

Abstract

AbstractBackgroundTraumatic brain injury (TBI) has been associated with an increased risk of dementia, but the link between mild TBI across the life‐course remains unclear. The PREVENT Dementia study is recruiting healthy volunteers in middle age (aged 40 – 59) to identify factors that may increase their risk of dementia. This analysis assessed white matter integrity and its association with the incidence, frequency and severity of TBIs.Methods140 subjects (mean age=51.9 years) from the PREVENT Dementia study West London site underwent a 3T MRI scan, including diffusion‐weighted (DW) imaging, and completed the self‐report Brain Injury Screening Questionnaire (BISQ). DW images were processed using FSL tools (EDDY, DTIFIT, TBSS) and resulting fractional anisotropy (FA) and mean diffusivity (MD) spatial maps were fed into voxelwise cross‐subject analysis (FWE‐corrected at p<0.05). BISQ summary scores were used as regressors in cross‐subject analyses to test the effect on WM integrity of a) no TBI (N=29) vs. any TBI (N=107), b) number of TBIs, and c) severity of TBIs. Gender, years of education, NART score, presence of psychiatric diagnoses and substance abuse were included as covariates. One participant with severe TBI (BISQ severity=6) and three participants with moderate TBI (BISQ severity=4) were excluded from analyses.ResultsHigher FA was observed in bilateral body of the corpus callosum, superior corona radiata and forceps minor in those that reported at least one TBI compared to those that had none. There was no effect of this on MD. There was no significant association of the number of TBIs with either FA or MD. However, increased severity of TBI was significantly associated with higher FA in similar regions showing an effect of any vs. no TBI. Increased severity of TBI was also significantly associated with lower MD in the same regions.ConclusionsIt is unclear if higher FA reflects a mechanism of recovery post‐TBI or the injury itself. We plan to analyse a larger sample, as this sub‐group may not be representative, and include other possible confounders, such as cognition and age of and time since injury, to explore what these results may infer about mild TBI and dementia risk.