Regional volume differences in blast-related TBI with cognitive implications: A LIMBIC-CENC Study
Poster No:
2285
Submission Type:
Abstract Submission
Authors:
Emily Dennis1, Jared Rowland2, Mary Newsome3, Brian Avants4, Erin Bigler5, Nicholas Davenport6, Jessica Gill7, Sidney Hinds II8, Elizabeth Hovenden5, Kimbra Kenney8, Hannah Lindsey5, MJ Pugh3, Robert Shih8, Randall Schiebel9, James Stone4, Maya Troyanskaya9, Nicholas Tustison4, Samuel Walton10, J Kent Werner8, Gerald York11, William Walker10, David Cifu10, David Tate3, Elisabeth Wilde3
Institutions:
1University of Utah, Emerald Hills, CA, 2W.G. (Bill) Hefner VA Medical Center, Salisbury, NC, Salisbury, NC, 3George E. Wahlen Veterans Affairs Medical Center, Salt Lake City, UT, 4University of Virginia, Charlottesville, VA, 5University of Utah, Salt Lake City, UT, 6Minneapolis VA Health Care System, Minneapolis, MN, 7Johns Hopkins University, Baltimore, MD, 8Uniformed Services University, Bethesda, MD, 9Michael E. DeBakey Veterans Affairs Medical Center, Houston, TX, 10Virginia Commonwealth University, Richmond, VA, 11Alaska Radiology Associates, Anchorage, AK
First Author:
Co-Author(s):
Introduction:
Blast-related traumatic brain injuries (TBIs) have been called the "signature injury" of the recent conflicts in Iraq and Afghanistan, with more than 100,000 injuries sustained by service members between 2001 and 2018 and many more "subthreshold" exposures (1). Blast-related TBI is thought to arise through unique mechanisms resulting from the over- and underpressure of the blast; however, there is a significant challenge in studying blast-related injuries outside of the experimental setting as they nearly always also include blunt-force impacts. Leveraging the multi-site Long-term Impact of Military-relevant Brain Injury Consortium/Chronic Effects of Neurotrauma Consortium (LIMBIC-CENC) sample, we sought to identify differences in brain structure in individuals with blast-related mild TBI (b-mTBI), and to determine if these differences related to cognitive function.
Methods:
Using a structured interview, lifetime history of all possible concussive events was assessed (2). The total sample size was n=774, with n=598 having a history of any mTBI, n=391 deployment-related mTBI, and n=260 b-mTBI. Participants completed a cognitive battery, including the Trail Making Test (TMT) and WAIS-IV, along with the PCL-5 for PTSD symptoms, PHQ-9 for depression symptoms, and AUDIT-C for problematic alcohol use. T1-weighted MRI was collected with prospectively harmonized sequences, and a multi-site template was created. We used tensor-based morphometry (TBM) to create Jacobian determinant images using unbiased_pairwise_registration. Voxel-wise linear mixed effects models were implemented with site as a random effect and age and gender as covariates. Results were corrected for multiple comparisons using searchlight FDR (3).
Our primary analysis compared individuals with a history of b-mTBI to those without (which included TBI-negative controls and those with only non-blast mTBI). We conducted additional analyses examining multiple potentially confounding variables. We also examined voxel-wise associations with cognitive performance, and in the case of overlap between blast and cognitive clusters, we conducted causal mediation analyses with the R package, mediation.
Our primary analysis compared individuals with a history of b-mTBI to those without (which included TBI-negative controls and those with only non-blast mTBI). We conducted additional analyses examining multiple potentially confounding variables. We also examined voxel-wise associations with cognitive performance, and in the case of overlap between blast and cognitive clusters, we conducted causal mediation analyses with the R package, mediation.
Results:
We found several regions with smaller volumes in the b-mTBI group (Fig 1). These effects remained after controlling for current PTSD symptoms, depression symptoms, and problematic alcohol use within the past three months. They also remained significant when comparing the b-mTBI group to non-blast deployment mTBI. There was a significant positive association with the number of b-mTBIs.
Cognitive performance was associated with volume in multiple clusters overlapping with those associated with b-mTBI. Volumes of clusters extracted from the blast analysis were negatively associated with TMT part B completion time and positively associated with WAIS-IV Total Digit Span performance in the b-mTBI group (ps < .03). Further, across the whole sample, the volumes of these clusters significantly mediated the association between b-mTBI and cognitive performance (Fig 2).
Cognitive performance was associated with volume in multiple clusters overlapping with those associated with b-mTBI. Volumes of clusters extracted from the blast analysis were negatively associated with TMT part B completion time and positively associated with WAIS-IV Total Digit Span performance in the b-mTBI group (ps < .03). Further, across the whole sample, the volumes of these clusters significantly mediated the association between b-mTBI and cognitive performance (Fig 2).
Conclusions:
We report smaller white matter (WM) and subcortical gray matter (GM) cluster volumes in those with a history of b-mTBI, along with evidence for a dose effect for repeated b-mTBI. Tissue stiffness may be relevant in the setting of b-mTBI, as stiffer tissues may be more prone to disruption by a blast-related pressure wave. Magnetic resonance elastography (MRE) has shown that subcortical GM and WM projection tracts have greater shear stiffness than other brain regions, possibly underlying selective vulnerability of these areas (4). Causal mediation analysis linked alterations in volume to deficits in cognitive performance. Volumes in the brain regions associated with blast-related TBI mediated the association between b-mTBI history and cognitive performance. These results suggest that volumetric loss as a result of blast injury may be a mechanism by which clinically-measurable cognitive deficits occur in military veterans.
Neuroanatomy, Physiology, Metabolism and Neurotransmission:
Cortical Anatomy and Brain Mapping 2
Novel Imaging Acquisition Methods:
Anatomical MRI 1
Keywords:
MRI
Neurological
STRUCTURAL MRI
Trauma
Other - Traumatic brain injury
1|2Indicates the priority used for review
·Fig 1. Differences in regional volume between participants with and without a history of blast-related TBI. Color corresponds to Cohen’s d. Left in image is right in brain.
·Fig 2. Causal mediation analyses.
Provide references using author date format
2. Walker WC et al. Structured interview for mild traumatic brain injury after military blast: inter-rater agreement and development of diagnostic algorithm. J Neurotrauma 2015;32:464–73.
3. Langers DR et al. Enhanced signal detection in neuroimaging by means of regional control of the global false discovery rate. Neuroimage 2007;38:43–56.
4. Hiscox LV et al. Standard-space atlas of the viscoelastic properties of the human brain. Hum Brain Mapp 2020;41:5282–300.