Characterizing the Relationship between Mechanical and Diffusive Properties in the Aging Hippocampus

Introduction: : Neurodegenerative diseases are becoming increasingly prevalent with the global rise in life expectancy, as approximately 6 million Americans alone live with Alzheimer’s Disease (AD) (Committee, 2017). The hippocampus, a medial temporal lobe structure crucial for memory and cognition, is especially vulnerable to neurodegenerative pathologies, including cerebral amyloid angiopathy and tauopathies (Delgorio, 2023). While there are pharmaceutical methods to clear these neurotoxic proteins from the brain, no current treatments can restore lost cognitive function. Consequently, there is an urgent need to identify early age-related neurophysiological deterioration to establish effective interventive techniques.

Quantitative MRI techniques, including magnetic resonance elastography (MRE) and diffusion-weighted imaging (DWI), are sensitive to subtle changes in the brain’s cellular microstructure. MRE uses external shear waves to induce micron-scale displacements to brain tissue in vivo, which are used to calculate neuromechanical properties including stiffness (Hiscox, 2021). Neuromechanics are associated with cellular tissue structure and provide a unique window into the physiological responses to aging and disease onset, as whole-brain stiffness has been shown to decrease with age and cognitive impairment (Hall, 2021). DWI measures the Brownian motion of water, and is capable of revealing fiber track orientation, axonal deterioration, and demyelination (Melhem, 2002). DWI is commonly used to study brain health through microstructure, but little is known about commonalities or differences with MRE, and how they can be used together. Considering the intrinsic relationship between neuromechanical behavior and cellular organization, this study aims to investigate alterations to hippocampal stiffness and diffusivity across cognitively impaired and unimpaired adults.

Materials and

Methods: : This study involved 56 participants (male=17, female=39, aged 46-85) consisting of mild cognitively impaired (MCI=20) and cognitively unimpaired (CU=36) groups. Participants underwent a comprehensive MRI protocol using a Siemens 3T Prisma scanner, which included T1-weighted, MRE, and DWI scans. MRE involves a pneumatic pillow driver sending shear waves through the head at 50 Hz (Figure 1A). MRE sequence parameters include 8 interleaved spiral readouts (R=2), TR/TE=140/76 ms, and a 1.5 mm isotropic resolution. Oscillatory tissue displacement (Figure 1B) is captured with motion encoding gradients in three orthogonal directions and is nonlinearly inverted (Hiscox, 2016) (Figure 1C) into maps of brain stiffness (Figure 1D). Diffusion imaging includes 137 encoding directions with b-values of 0 (reference), 1500, and 3000 s/mm/mm acquired with 1.5 mm isotropic resolution. Forward- and reverse-polarity fieldmaps are taken for eddy current distortion and motion artifact corrections performed with the FMRIB Software Library (FSL) topup feature (Anderson, 2003). Maps of diffusion metrics were derived from the diffusion tensor, in particular fractional anisotropy (FA) and radial diffusivity (RD). The stiffness, FA, and RD maps of each subject were registered to their T1-weighted anatomical scans for hippocampal alignment (Figure 2) and segmentation using the FSL tool flirt (Jenkinson, 2001). Following segmentation, the means of hippocampal stiffness, FA, and RD for each subject were calculated as the average of all voxels in the hippocampus mask. Statistical analysis included two-way ANOVA to investigate main effects and interactions between FA/RD and cognitive status on stiffness, and linear regression to assess within group relationships.

Results, Conclusions, and Discussions:: We found that hippocampal stiffness is significantly correlated with diffusivity metrics when considering cognitive status. Using a 2-way ANOVA, FA was significantly correlated with stiffness (p=0.00232), whereas cognitive status alone was not. However, the interaction term between FA and cognitive status was also significant (p=0.00022), indicating that stiffness is modulated by cognitive status. RD showed similar results, being significantly correlated with stiffness (p=0.016) and having a significant interaction term (p=0.012), whereas cognitive status remained non-significant. These findings establish that the MCI group experiences atypical aging pathology in the hippocampus that is inseparable from stiffness and diffusivity metrics.

Separate linear regressions modeled stiffness with FA and RD in each group (Figure 3). In the CU group, no significance was found between FA (r=0.23, p=0.577) or RD (r=-0.015, p=0.619) and stiffness. In contrast, in the MCI group, FA and stiffness had a significant positive association (r=0.80, p=0.0008) where RD and stiffness were inversely correlated (r=-0.62, p=0.0495). All results were found controlling for age. This suggests that microstructural degradation of the hippocampus in diseased populations is observable with changes in diffusivity and tissue stiffness. Though, in a population without pathology, the two measures are unrelated, likely due to greater contributions to each from other components of tissue microstructure.

This differential relationship reveals a pathophysiological mechanism driving tissue decomposition in exclusively cognitively impaired populations. Decreased FA and increased RD are associated with demyelination, which inhibits neuronal communication (Melhem, 2002). Processes involving demyelination and extracellular remodeling precede symptoms of cognitive impairment and tangentially influence tissue biomechanics. MRE findings described here are consistent with previously seen diffusivity metrics specifying decreased cellular density and impaired tissue integrity (Alexander, 2008).

Our results highlight a significant interaction between DWI outcomes and cognitive status in predicting hippocampal stiffness, suggesting a pathologically driven mechanism influencing microstructural and biomechanical degeneration. These findings may reflect early Alzheimer’s-related amyloid and tau accumulation, which subtly decomposes tissue architecture that is detectable with MRE and DWI. This study supports the utility of combining elastography and diffusion in detecting neurophysiological changes, reinforcing the need for comprehensive multimodal imaging in identifying evidence of neurodegeneration preceding cognitive decline.

Acknowledgements and/or References (Optional):: Acknowledgements
This research was supported in part by NIH grants R01-AG058853 and R01-AG080052, as well as the Delaware Community Foundation Boerger Fund. Recruitment of research subjects was supported by the Delaware Center for Cognitive Aging Research.

References
1. L. V. Hiscox, H. Schwarb, M. D. J. McGarry, and C. L. Johnson, “Aging brain mechanics: Progress and promise of magnetic resonance elastography,” NeuroImage, vol. 232, p. 117889, May 2021, doi: 10.1016/j.neuroimage.2021.117889.
2. L. V. Hiscox et al., “Magnetic resonance elastography (MRE) of the human brain: technique, findings and clinical applications,” Phys. Med. Biol., vol. 61, no. 24, pp. R401–R437, Dec. 2016, doi: 10.1088/0031-9155/61/24/R401.
3. C. M. Hall, E. Moeendarbary, and G. K. Sheridan, “Mechanobiology of the brain in ageing and Alzheimer’s disease,” Eur. J. Neurosci., vol. 53, no. 12, pp. 3851–3878, Jun. 2021, doi: 10.1111/ejn.14766.
4. A. Manduca et al., “Magnetic resonance elastography: Non-invasive mapping of tissue elasticity,” Med. Image Anal., vol. 5, no. 4, pp. 237–254, Dec. 2001, doi: 10.1016/S1361-8415(00)00039-6.
5. Committee on Preventing Dementia and Cognitive Impairment, Board on Health Sciences Policy, Health and Medicine Division, and National Academies of Sciences, Engineering, and Medicine, Preventing Cognitive Decline and Dementia: A Way Forward. Washington, D.C.: National Academies Press, 2017, p. 24782. doi: 10.17226/24782.
6. P. L. Delgorio et al., “Hippocampal subfield viscoelasticity in amnestic mild cognitive impairment evaluated with MR elastography,” NeuroImage Clin., vol. 37, p. 103327, 2023, doi: 10.1016/j.nicl.2023.103327.
7. J. L. R. Andersson, S. Skare, and J. Ashburner, “How to correct susceptibility distortions in spin-echo echo-planar images: application to diffusion tensor imaging,” NeuroImage, vol. 20, no. 2, pp. 870–888, Oct. 2003, doi: 10.1016/S1053-8119(03)00336-7.
8. S. M. Smith et al., “Advances in Functional and Structural MR Image Analysis and Implementation as FSL Technical Report TR04SS2”.
9. Jenkinson M., Bannister, P., Brady, J. M. and Smith, S. M. Improved Optimisation for the Robust and Accurate Linear Registration and Motion Correction of Brain Images. NeuroImage, 17(2), 825-841, 2002.
10. Jenkinson, M. and Smith, S. M. A Global Optimisation Method for Robust Affine Registration of Brain Images. Medical Image Analysis, 5(2), 143-156, 2001.
11. Greve, D.N. and Fischl, B. Accurate and robust brain image alignment using boundary-based registration. NeuroImage, 48(1):63-72, 2009.
12. E. R. Melhem, S. Mori, G. Mukundan, M. A. Kraut, M. G. Pomper, and P. C. M. Van Zijl, “Diffusion Tensor MR Imaging of the Brain and White Matter Tractography,” Am. J. Roentgenol., vol. 178, no. 1, pp. 3–16, Jan. 2002, doi: 10.2214/ajr.178.1.1780003.
13. A. L. Alexander, J. E. Lee, M. Lazar, and A. S. Field, “Diffusion Tensor Imaging of the Brain,” 2008.