Back to 2017 Annual Meeting Program
UTE-MRI Strategies for Imaging Peripheral Nerve Injury
Justin M. Brown, MD; Jiang Du, PhD; Eric Y Chang, MD; Yajun Ma, PhD; Ross Mandeville, MD; Sameer B. Shah, PhD
University of California, San Diego, San Diego, CA
Introduction: Functional recovery following peripheral nerve damage is often poor. The structural environment of a regenerating nerve, including progressive changes to the extracellular matrix and basal lamina, can profoundly influence recovery. Current non-invasive imaging approaches are limited in their ability to characterize distal nerve stump quality.
Materials and Methods: We have developed novel 3D-ultrashort-time-to-echo with cones readout (3D-UTE-Cones) magnetic resonance imaging (MRI) strategies to provide rapid, noninvasive, and high-resolution quantitative assessment of nerve architecture. These strategies exploit the properties of nerve collagen and myelin, which have much shorter T2s than nerve fibers, and are not well characterized using conventional MRI. In particular, we applied 3D-UTE-Cones strategies to a Lewis rat sciatic nerve injury model. Using both 11.7T high-resolution and 3T clinical MRI scanners, 3D-UTE-Cones-DWI (diffusion weighted imaging), 3D-UTE-Cones bi-component analysis, and 3D-UTE-Cones-MT (magnetization-transfer) sequences were applied to the proximal and distal stumps of uninjured control nerves and nerves injured and capped without repair for 1, 3, and 12 weeks, corresponding to different phases of degeneration. Several MRI biomarkers, including diffusion parameters, bound/free water T2*s and fractions, water/macromolecule fractions and exchange rates were correlated with structural features of the stumps, including connective tissue area fractions, axon density, and myelination, which were assessed using immunohistochemistry (IHC; trichrome labeling, beta3-tubulin, laminin, and myelin basic protein antibodies). Imaging and analyses were also piloted on human cadaveric nerves and in vivo.
Results: Consistent with IHC outcomes, scanned control rat nerves showed well-defined differences between epineurial, perineurial, and endoneurial/nerve fiber compartments, and subtraction images selectively showed the external and internal epineurium. Diffusion imaging also showed well-defined differences between nerve compartments, providing a baseline for changes that may occur with neuropathy. After injury, IHC revealed that the near distal stump (close to the injury site) lost integrity at 3 weeks, and was unanalyzable at 2 months, while the far distal stump (away from injury site) maintained its integrity at 3 weeks, but lost its integrity by 2 months. 3D-UTE-Cones outcomes also showed differences in nerves injured for 2 months, compared to uninjured controls. Consistent with increased fibrotic infiltration into the nerve, bicomponent analysis revealed an increase in short component (collagen content) from 33% to 55% in the stump margin compared to control. MT modeling indicated fewer macromolecule-bound protons after injury, likely reflecting fewer neurofilaments and/or reduced myelin.
Conclusions: We successfully deployed novel methods to image nerves, which may have significant clinical impact on surgical decision making.
Back to 2017 Annual Meeting Program