Post-mortem diffusion MRI of cervical spine and nerves roots

Wieke Haakma^1,2,3, Lidy Kuster², Martijn Froeling¹, Lars Uhrenholt², Michael Pedersen^3,4, Jeroen Hendrikse¹, Alexander Leemans⁵, and Lene Warner Thorup Boel²

¹Radiology, University Medical Center Utrecht, Utrecht, Netherlands, ²Forensic Medicine, Aarhus University, Aarhus, Denmark, ³Comparative Medicine Lab, Department of Clinical Medicine, Aarhus University, Aarhus, Denmark, ⁴MR Research Center, Department of Clinical Medicine, Aarhus University, Aarhus, Denmark, ⁵Image Sciences Institute, University Medical Center Utrecht, Utrecht, Netherlands

Synopsis

In this work we examined the architecture and diffusion measures of the cervical spine and nerves in non-fixated post-mortem subjects. We were able to display the architectural configuration of the cervical nerves at the level of C4-C8 and we computed reference values for the diffusion measures in these nerves. We showed with great detail the ventral and dorsal nerve roots with fiber tractography. Microscopic examination revealed normal anatomy. We expect that post-mortem diffusion MRI will be valuable for understanding of pathological mechanisms underlying degenerative neurological diseases, as it is possible to compare any findings directly to histological examinations.

Purpose

The exact pathological mechanisms behind degenerative nerve diseases affecting spinal and peripheral nervous tissue are still largely unknown. The examination of spine and nervous tissue in the cervical area can be tedious. Therefore, a method is needed that is able to examine peripheral nervous tissue in an efficient way. Diffusion tensor imaging (DTI), allowing for the evaluation of microstructural properties of nervous structures, is an excellent candidate. In this context, DTI has already been used to investigate the lumbar nerve roots¹, and the cervical nerves in vivo^2,3. To what extent DTI is able to identify the cervical myelum and nerve roots in great detail remains to be shown. Validation with histology is needed to assess the added value of DTI for investigating peripheral nervous tissue. The aim of this work is to examine the architectural configuration and the microstructural substrate of the cervical spine and its nerve roots with post-mortem (PM) DTI and histological dissection in non-fixated subjects.

Methods

Five non-fixated PM subjects with normal anatomy of the cervical spine were included; 5 men (4-6 days after death) with a mean age of 51 years (range 25–90 years). Subjects were scanned at the level of the C3-C8 on a 1.5 Tesla MR system (Achieva; Philips Healthcare, Best, The Netherlands) using a head-neck 32-channel phased-array surface coil. Two DTI protocols were obtained with diffusion-weighted spin echo single-shot EPI sequence, i.e., (1) an “isotropic” protocol to investigate the nerves (C4-C8) and (2) a “high in-plane resolution” protocol to investigate the myelum and ventral and dorsal nerve roots (C5-C7). Both were repeated two times in the same scan session. As an anatomical reference, a multi-echo fast field echo (mFFE) sequence was acquired at the level of C5-C7. An overview of the parameters of the scan sequences is displayed in Table 1. Tissue samples of the myelum and peripheral nerve roots at the level of C5-C7 were obtained during the autopsy on the next day. Processing of the diffusion MRI data was performed with ExploreDTI⁴ and comprised the following steps. First, the diffusion MRI data were corrected for eddy current distortions, and EPI deformations^5,6. Second, diffusion tensors were fitted using the iteratively weighted linear regression procedure⁷. Finally, DTI based fiber tractography (FT) was performed with a fractional anisotropy (FA) threshold of 0.15-0.9, minimum fiber length of 10 mm, and an angle of 30°. Estimates for the FA, mean (MD), axial (AD), and radial (RD) diffusivity of the cervical nerves (C4-C8) were computed.

Results

With FT performed on DTI data obtained with the “isotropic” protocol it was possible to reconstruct the 3D architecture of the myelum, cervical nerves, and nerve roots in all 5 PM subjects (Fig. 1). We were able to show the dorsal and ventral nerve roots with great detail with the “high in-plane resolution” protocol (Fig. 2 and Fig. 3) and we identified a low FA in the grey matter and a high FA in the white matter. Histology revealed normal anatomy of the myelum and peripheral nerves in all 5 cases, which was in accordance with the architectural configuration found with FT. Reference values for the DTI measures are shown in Table 2. MD values were approximately 5 times lower than in vivo results, whereas FA values were comparable with previous in vivo studies of the cervical nerves^1,2.

Discussion

The observed differences in MD, AD, and RD between the PM and in vivo results are expected to be mainly caused by temperature differences of the tissue⁸. Diffusion values were comparable to those obtained in an earlier PM study investigating the lumbosacral nerves⁹. With the “high in-plane resolution protocol” we were able to show with great detail the dorsal and ventral nerve roots with FT. This information is relevant, since it provides insight to what extent these nerve bundles can be visualized with FT PM, and it can be used as an example for future in vivo studies. The proposed DTI method may contribute to better understanding of nervous and muscular diseases, where only the motor or the sensor neurons are affected, as these findings can directly be compared to histological examinations.

Conclusion

This PM DTI and FT study shows that the cervical spine and nerves can be identified with great detail. Histological examination identified normal anatomy of the myelum and peripheral nerves, which was in accordance with the architectural configuration found with FT. We were able to quantify the diffusion properties of these nerves PM and we showed the dorsal and ventral nerve roots with great detail.

Acknowledgements

This study was supported by the KNAW Van Leersum fund.

References

[1] Karampinos DC, Melkus G, Shepherd TM, et al. Diffusion tensor imaging and T2 relaxometry of bilateral lumbar nerve roots: feasibility of in-plane imaging NMR Biomed. 2013;26(6):630–7

[2] Vargas MI, Viallon M, Nguyen D, et al. Diffusion tensor imaging (DTI) and tractography of the brachial plexus: feasibility and initial experience in neoplastic conditions. Neuroradiology. 2010;52(3):237-45

[3] Tagliafico A, Calabrese M, Puntoni M, et al. Brachial plexus MR imaging: accuracy and reproducibility of DTI-derived measurements and fibre tractography at 3.0-T. Eur Radiol. 2011;21(8):1764-71

[4] Leemans A, Jeurissen B, Sijbers J, et al. ExploreDTI: a graphical toolbox for processing, analyzing, and visualizing diffusion MR data. Proc Intl Soc Mag Reson Med. 2009;17:3536.

[5] Leemans A, Jones DK. The B-matrix must be rotated when correcting for subject motion in DTI data. Magn Reson Med. 2009;61(6):1336–49

[6] Irfanoglu MO, Walker L, Sarlls J, et al. Effects of image distortions originating from susceptibility variations and concomitant fields on diffusion MRI tractography results. Neuroimage. 2012;61(1):275-88

[7] Veraart J, Sijbers J, Sunaert S, et al. Weighted linear least squares estimation of diffusion MRI parameters: Strengths, limitations, and pitfalls. NeuroImage. 2013;1(81):335-46

[8] D'Arceuil HE, Westmoreland S, de Crespigny AJ. An approach to high resolution diffusion tensor imaging in fixed primate brain. Neuroimage 2007;35(2):553–565.

[9] Haakma W, Pedersen M, Froeling M, et al. ISMRM 2015, 3015

Figures

Figure 1: Left) fiber tractography of the cervical spine and peripheral nerves in a post-mortem subject displaying the architecture of the nerves. Fractional anisotropy (FA) color map of the nerves C4-C8 and spine. Right) a tissue sample of the myelum and branching nerves at the level of C5-C7.

Figure 2: Cervical myelum and dorsal and ventral nerve roots, A) FA color-encoded map, B) MD color-encoded map, C) schematic overview based on [10], D) mFFE displaying the grey and white matter and branching nerve roots, E) microscopic image (Klüver Barrera staining) showing normal anatomy of grey and white matter.

Figure 3: A) directionally color-encoded map with fiber tractography of the myelum and (ventral and dorsal) nerve roots, B) FA directional color-encoded map displaying the grey matter (low FA, dark blue), and white matter (high FA, bright blue) in the myelum and dorsal and ventral nerve roots with low FA.

Table 1: Acquisition parameters of MRI protocols.

Table 2: Diffusion parameters (fractional anisotropy, FA, mean diffusivity, MD, axial diffusivity, AD, and radial diffusivity, RD) of post-mortem nerves C4-C8.

Proc. Intl. Soc. Mag. Reson. Med. 24 (2016)

0926