Tract Profile Analysis Reveals Focal White Matter Changes in Individuals with HIV
Poster No:
2191
Submission Type:
Abstract Submission
Authors:
Alan Finkelstein1, Jianhui Zhong1, Giovanni Schifitto1
Institutions:
1University of Rochester, Rochester, NY
First Author:
Co-Author(s):
Introduction:
Individuals with chronic HIV suffer from HIV-associated neuroinflammation1. Chronic neuroinflammation compromises white matter (WM) integrity, which can be evaluated using diffusion-weighted imaging (DWI) within regions of interest (ROI)3. Yet, recent work has shown that tissue properties vary along individual tracts due to differentially vulnerable axonal populations8. Thus, tract profiles can reveal subtle differences between distinct locations along a tract. We performed tract-profile analysis using automated fiber quantification (AFQ) to investigate susceptible sites along WM bundles affected by HIV-associated neuroinflammation. DWI metrics exhibited differential variability and sensitivity to cognition along tracts and between groups, providing complementary information to understand disease processes better.
Methods:
80 individuals without HIV (mean age: 52.03 +/- 17.1 [SD], M/F: 51/29) and 85 individuals with HIV (mean age: 53.5 +/- 10.3 [SD]; M/F: 61/24) were evaluated using DWI. Images were acquired using a 3T Siemens MAGNETOM PrismaFit (Erlangen, Germany), using a 64-channel head coil. DWI was performed using a single shot SE-EPI sequence along 64 directions with two non-zero b-values (1,000 and 2,000 s/mm2; TR/TE = 4300/69 ms; FOV: 172x172, 1.5x1.5x1.5 mm3). Preprocessing of multi-shell DWI data is described previously4. Free-water, fractional anisotropy (FA), radial diffusivity (RD), axial diffusivity (AD), and mean diffusivity (MD) were estimated using diffusion tensor imaging. Axial kurtosis (AK), mean kurtosis (MK), and radial kurtosis (RK) were obtained using diffusion kurtosis imaging. The orientation dispersion index (ODI), isotropic volume fraction (fiso) and the intracellular volume fraction (icvf) using neurite orientation dispersion and density imaging10. Whole brain probabilistic tractography was performed in MRtrix36 following estimation of fiber orientation distribution functions (fODF)5. RecoBundles2 was used to extract 80 WM bundles from tractograms based on similarity with template streamlines using an atlas9. Tract-profile analysis was performed using AFQ, splitting each tract profile into 100 nodes and sampling ODI, FA, MD, and RK at each node. Neurocognitive evaluations were performed to evaluate six cognitive domains, and a composite total cognitive z-score was determined7. Independent t-tests were performed to compare tract profiles and mean tract values between groups. Pearson correlation test was used to test associations between two continuous variables.
Results:
Figure 1A shows mean tract profiles for the left and right superior longitudinal fasciculus (SLF), and the left and right middle longitudinal fasciculus (MdLF), selected for clarity. Similar changes were observed in other tracts. Differential sensitivity to neuroinflammation is observed along the tract, compared to mean tract values between groups (Figure 1B). Figure 2A shows the correlation coefficient between ODI, FA, MD, and AK at each node along a tract with the composite total cognitive z-score for individuals with and without HIV . Along each tract, we observed high variability of the correlation coefficient, illustrating which areas are associated with cognitive changes, not discernable with conventional analysis (Figure 2B).
Conclusions:
Tract profile analysis revealed WM bundles are differentially affected by HIV-associated neuroinflammation. We observed differential sensitivity to cognitive measures using tract profile analysis, suggesting more nuanced manifestations of HIV-associated neuroinflammation are appreciated at specific locations compared to conventional tract-averaged measurements. This work is consistent with prior results implicating the SLF in cognitive impairment in individuals infected with HIV3, and associates the MdLF as a distinct entity involved in HIV-associated cognitive decline.
Modeling and Analysis Methods:
Connectivity (eg. functional, effective, structural)
Diffusion MRI Modeling and Analysis 2
Neuroanatomy, Physiology, Metabolism and Neurotransmission:
White Matter Anatomy, Fiber Pathways and Connectivity 1
Keywords:
Tractography
White Matter
WHITE MATTER IMAGING - DTI, HARDI, DSI, ETC
Other - HIV-associated Neuroinflammation
1|2Indicates the priority used for review
Provide references using author date format
2. Garyfallidis E. (2018). ‘Recognition of white matter bundles using local and global streamline-based registration and clustering’, Neuroimage, vol. 170, pp. 283-295.
3. Ma J. (2022). ‘Application of Diffusion Tensor Imaging (DTI) in the Diagnosis of HIV-Associated Neurocognitive Disorder (HAND): A Meta-Analysis and a System Review’, Front Neurol, vol. 13, pp. 898191.
4. Murray KD. (2020). ‘Pathomechanisms of HIV-Associated Cerebral Small Vessel Disease: A Comprehensive Clinical and Neuroimaging Protocol and Analysis Pipeline’, Front Neurol, vol. 11, pp. 595463.
5. Tournier JD. (2007). ‘Robust determination of the fibre orientation distribution in diffusion MRI: non-negativity constrained super-resolved spherical deconvolution’, Neuroimage, vol. 35, no. 4, pp. 1459-1472.
6. Tournier JD. (2019). ‘MRtrix3: A fast, flexible and open software framework for medical image processing and visualisation’, Neuroimage, vol. 202, pp. 116137.
7. Weber MT. (2022). ‘Longitudinal Effects of Combination Antiretroviral Therapy on Cognition and Neuroimaging Biomarkers in Treatment-Naive People With HIV’, Neurology, vol. 99, no. 10, pp. e1045-e1055.
8. Yeatman JD. (2012). ‘Tract profiles of white matter properties: automating fiber-tract quantification’, PLoS One, vol. 7, no. 11, pp. e49790.
9. Yeh FC. (2018). ‘Population-averaged atlas of the macroscale human structural connectome and its network topology’, Neuroimage, vol. 178, pp. 57-68.
10. Zhang H. (2012). ‘NODDI: practical in vivo neurite orientation dispersion and density imaging of the human brain’, Neuroimage, vol. 61, no. 4, pp. 1000-1016.