Diffusion Tensor Peripheral Nerve Tractography ~ Histological Changes and Diffusion Anisotropy ~

INTRODUCTION: The standard repertoire in the diagnosis of peripheral nervous system disorders includes clinical and electrophysiological examinations, supplemented by more invasive procedures, because peripheral nerves are difficult to delineate on conventional MRI due to their poor contrast with the surrounding tissues. Recently, the visualization of peripheral nerves using MRI has been attempted using special techniques such as whole-body MR neurography. However, the interpretation of the images obtained by the neurography is based on visual inspection, and is therefore qualitative and subjective. Furthermore, since the neurography cannot image continuous nerve fibers over their entire length, it is not considered useful for examining the growth of regenerating peripheral nerves. With the progress in MRI technology and techniques, several researchers have reported on successful diffusion tensor tractography (DTT), which the orientation of nerve fibers can be followed to trace specific neural pathways using the movements of water molecules, such as that of the corticospinal tract in the brain or the spinal cord. When no proper tools are available yet for visualizing the peripheral nerves, it is important to evaluate the validity of assessment of peripheral nerve degeneration and regeneration using DTT. METHODS: Animals and surgical procedures. Adult female Sprague-Dawley rats were used. The sciatic nerves were exposed and subjected to a crush injury using a brain aneurysm clip under chloral hydrate anesthesia. Magnetic resonance imaging. MRI was performed using a 7.0-Tesla MRI, PharmaScan 70/16 (Bruker BioSpin, Ettlingen, Germany) with a coil dedicated for small animals. To verify the feasibility of peripheral nerve DTT, we first performed DTT of the sciatic nerves of an in vivo model. For analyzing the degeneration and regeneration of nerves in detail, diffusion tensor MRI was conducted with phantoms consisting of excised sciatic nerve specimens (1 day, and 1, 3, 6, and 12 weeks after crush injury and intact nerves, n=6 each) embedded in 2% agarose gel with 5mM copper sulfate, using a spin-echo sequence. Diffusion tensor analysis. DTT images were computed with the Volume One and dTV II SR software. We used one region of interest (ROI) technique, by choosing a seed ROI through which the fibers were tracked. To delineate the fibers, the ROI was placed 5mm proximal to the contusion. We measured the fractional anisotropy (FA) values in each sample at points 5 mm (proximal), 0 mm (epicenter) and -5 mm (distal) proximal to the contusion. The scanning parameters were as follows; TR = 4500 ms, TE = 40 ms, flip angle = 90 deg, FOV = 30 x 30 mm, slice thickness = 1.25 mm, reconstructed image resolution = 0.31 mm, matrix size = 96 x 96, b-value = 1000 sec/mm, MPG orientations = 12 axes. Histological analysis. We performed electron-microscopic analysis of 6 nerve samples at the same time points as the diffusion tensor analysis. At a point 5 mm distal to the contusion, 80-nm-thick ultrathin sections were cut cross-sectionally and stained with uranyl acetate. The following parameters were calculated for each nerve: axon density, axon diameter, myelin sheath density, and myelin sheath thickness. Functional analysis. The leg muscle contraction test, the Rota-rod test (to evaluate motor coordination), and the von Frey filament test (to evaluate mechanical sensitivity) were used to assess the recovery of function after the sciatic nerve injury. 6 rats were used for these functional evaluations. The tests were conducted at the same time points as the diffusion tensor analysis. Statistical analysis. Pearson’s correlation coefficients were calculated to determine correlations between these parameters and the FA. RESULTS SECTION: Peripheral nerves could be reliably distinguished from the surrounding tissues by DTT in the in vivo study (Fig. 1A-B). Using a threshold FA value of 0.6, the recovery process of the contused peripheral nerves could be clearly visualized (Fig. 2A-F). The fibers could be tracked distally at 3 weeks after the injury, with the number of fibers increasing thereafter with time. For analyzing the histology in detail, we performed electron-microscope examination. Ultrathin sections from a site distal to the site of injury indicated that the axonal structures remained relatively stable until 1 day after the operation (Fig. 3B), with the beginning of disintegration (Fig. 3C). Regenerating nerve fibers, with myelinated fibers of a smaller caliber and thinner myelin sheath as compared with that of normal nerve fibers, were clearly detectable from 3 weeks after the injury (Fig. 3D), almost reaching the pre-injury level by 12 weeks after the injury (Fig. 3F). The FA value, a parameter for constructing the DTT, was correlated with the axon density (r=0.8034, p<0.05) and axon diameter (r=0.9023, p<0.05). In addition, the FA value at the epicenter was strongly correlated with each functional parameter (p<0.01) . DISCUSSION: Advantage is taken of the higher anisotropy values of peripheral nerves than the surrounding tissues in the detection and evaluation of peripheral nerves using DTT. FA values of the peripheral nerves were more strongly correlated with axon-related parameters, namely, axon density and axon diameter, than with myelin-related parameters. These findings support the theories that axonal membranes play a major role in anisotropic water diffusion in neural fibers and that myelination can modulate the degree of anisotropy. DTT of the peripheral nerves may become a novel tool for the evaluation of peripheral nerves if it is applied correctly and its properties and limitations are clearly understood. ACKNOLEDGEMENT: This work was supported by grants from the Leading Project for Realization of Regenerative Medicine from the Ministry of Education, Culture, Sports, Science and Technology (MEXT), Japan, from the Japan Science and Technology Corporation (JST), and from the General Insurance Association of Japan. This work was also supported by a Keio University grant-in-aid for encouragement of young medical scientists, by a grants-in-aid for scientific research of MEXT, Japan, and by a grant-in-aid from the 21st COE Program of MEXT, Japan to Keio University. REFERENCES: 1. Yamashita T, et al. (2009) N Engl J Med 96:538-539. 2. Mori S, et al. (2006) Neuron 51, 527–539 3. Fujiyoshi K, et al. (2007) J Neurosci 27: 11991-11998. 4. Masutani Y, et al. (2003) Eur J Radiol 46: 53-66. 5. Takagi T, et al. (2009) Neuroimage 44: 884-892. Fig. 1. In vivo sciatic nerve DTT. A:FA>0.15.B:FA>0.35. Using a more stringent threshold, DTT can clearly delineate nerves and exclude the surrounding tissues. Fig. 2. Temporal analysis of injured sciatic nerves DTT. A:intact, B:1d, C:1w, D:3w, E:6w, F:12w after injury. The region of interest (ROI) was placed 5 mm proximal to the contusion (white line). It recovered gradually and reached the pre-injury level. Poster No. 1381 • 56th Annual Meeting of the Orthopaedic Research Society