Ratthaporn Boonsuth1,2, Marco Battiston1, Rebecca S. Samson1, Alberto Calvi1,3, Claudia A. M. Gandini Wheeler-Kingshott1,4,5, and Marios C. Yiannakas1
1NMR Research Unit, Queen Square MS Centre, Department of Neuroinflammation, UCL Queen Square Institute of Neurology, University College London, London, United Kingdom, 2Department of Radiologic Technology, Faculty of Associated Medical Sciences, Chiang Mai University, Chiang Mai, Thailand, 3Center of Neuroimmunology, Laboratory of Advanced Imaging in Neuroimmunological Diseases; Institut d’Investigacions Biomèdiques August Pi i Sunyer (IDIBAPS), Barcelona, Spain, 4Brain Connectivity Research Centre, IRCCS Mondino Foundation, Pavia, Italy, 5Department of Brain and Behavioural Sciences, University of Pavia, Pavia, Italy
Synopsis
Keywords: Multiple Sclerosis, Quantitative Imaging, Nerves, Multimodal, Neurography
The peripheral nervous system is
not routinely examined objectively in multiple sclerosis (MS), despite evidence
from neuropathology that demonstrates its implication. In this pilot
in
vivo study, the sciatic nerve was examined using multi-shell
diffusion-weighted imaging, quantitative magnetisation transfer and T1
relaxometry to investigate whether pathological neural tissue damage could be
detected in people with relapsing-remitting MS as compared to healthy controls.
Introduction
Multiple sclerosis (MS) has long
been considered a central nervous system (CNS) disorder, characterised by diffuse
demyelination and axonal degeneration. However, neuropathological,
electrophysiological, and imaging investigations have demonstrated that the
peripheral nervous system (PNS) is also involved in MS, with demyelination and
axonal degeneration representing the main pathophysiological processes1-3.
Magnetic resonance imaging (MRI),
specifically magnetic resonance neurography (MRN), has previously been used to
explore the PNS in a wide spectrum of neurological conditions. Moreover, MRN
has been used to facilitate semi-quantitative MRI (qMRI) investigations, which
provide more specific information about neural tissue composition, thus
allowing a better understanding of the pathophysiological mechanisms
involved3-6. Given the neuropathological evidence that demyelination
and, to a lesser extent, axonal degeneration are the primary pathophysiological
mechanisms of damage in the PNS in MS, emerging qMRI methods, previously
validated in the CNS, such as multi-shell diffusion-weighted imaging (DWI),
quantitative magnetization transfer (qMT) and T1 relaxometry, would also be
relevant in the context of PNS investigations in MS.
The pilot study aimed to apply the
aforementioned qMRI methods to the PNS of people with relapsing-remitting MS
(RRMS) and to determine whether these methods are sensitive to neural tissue
damage in the sciatic nerve as compared to healthy controls (HCs).Methods
Participants:
Fifteen HCs
(mean age 33.8 years, 9 female, range 25-44) and 15 people with RRMS (mean age 39.7 years, 10 female, range 29-46)
were recruited. Written informed consent was obtained from all study
participants and the study was approved by local ethical committees.
MR
imaging: A 3T Philips Ingenia CX MRI system was used with the product SENSE
spine and torso coils. In all participants the right sciatic nerve (upper
thigh) was examined; both sciatic nerves were first located using the 3D
‘nerve-SHeath signal increased with INKed rest-tissue RARE Imaging’ (SHINKEI)
sequence acquired in the coronal plane with a large field-of-view
(FOV)
7,8 (Figure 1A). The parameters used were
as follows: repetition time (TR)=2200ms, echo time (TE)=180ms, FOV=300×420 mm
2,
voxel size=1.2×1.2×2 mm
3, number of excitations (NEX)=1, turbo
spin-echo (TSE) factor=56, improved motion-sensitized driven-equilibrium
(iMSDE) duration=50ms, 170 slices, scanning time 05:43 min. The 3D SHINKEI
images were then used to facilitate the planning of high-resolution
fat-suppressed T2-weighted and reduced FOV echo-planar imaging (ZOOM-EPI)
acquisitions in the axial plane, for image segmentation and estimation of qMRI
metrics, respectively. The following parameters were used for the
fat-suppressed T2-weighted acquisition: TR=5000 ms, TE=60 ms, FOV=180×180 mm
2,
voxel size=0.5×0.5×4 mm
3, NEX=1, TSE factor=11, 30 slices, scanning
time of 08:08 min. The multi-modal qMRI protocol was performed in separate
acquisitions using ZOOM-EPI in a unified readout fashion (i.e., same resolution
and FOV), and with identical scan geometry to the high-resolution
fat-suppressed T2-weighted acquisition, as follows:
- Multi-shell
DWI acquisition: TR=4000ms, TE=80ms, FOV=64×40 mm2; voxel size=1×1×10
mm3; 12 slices;
b-value=700s/mm2 (16 directions); b-value=1200s/mm2 (20
directions); b-value=2000s/mm2 (32 directions); 7
interleaved non-diffusion-weighted (b=0) images were also collected; scanning
time=16:36 minutes.
- qMT with TE=51 ms; TR=7377ms; FOV=64×40mm2; voxel size=1x1x10mm3; 12 slices; 10 MT-weighted images; scanning time=9:13 minutes.
- T1 relaxometry with inversion
recovery (IR) with TE=51ms; TR=8335ms; FOV=64×40mm2;
voxel size=1x1x10mm3;
12 slices; 8 inversion times,
scanning time=07:05 minutes.
Image analysis: The sciatic nerve was segmented manually
in FSLview (http://www.fmrib.ox.ac.uk/fsl/) on fat-suppressed T2-weighted images
(Figure1B & 1C), with separate binary masks created for each slice. DWI images were
additionally motion corrected using FLIRT
(FMRIB's Linear Image Registration Tool)
9. The T2-weighted images and binary masks were
subsequently resampled to the same resolution as the qMRI data and
co-registered to the qMRI data using NiftyReg
10. Diffusion
kurtosis (DKI) signals were fitted to the acquired multi-shell DWI data using
the DiPy dkifit
11,12 tool; the fitting provided both standard
diffusion tensor (DTI) and kurtosis metrics. qMT data
were analysed using a simplified two-pool model for the estimation of the bound
pool fraction (BPF). T1 maps were obtained from the IR data, by fitting a
mono-exponential recovery model.
Results
Table 1 shows the mean (SD) of all
DWI metrics, T1 and BPF calculated in all HCs (n=15) and people with RRMS (n=15).
Figure 2 shows example maps of standard DTI metrics i.e. fractional anisotropy
(FA) and axial/radial/mean diffusivity (AD/RD/MD), DKI axial/radial/mean
kurtosis (AK/RK/MK), the inversion recovery and qMT with the maps of T1 and BPF, respectively.
Figure 3 shows the distribution of all qMRI metrics in HCs and people with
RRMS. Discussion and Conclusion
In
this pilot study, we present preliminary data from a qMRI assessment of the
sciatic nerve in HCs and people with RRMS using multi-shell DWI, qMT and
relaxometry for T1 mapping acquired with ZOOM-EPI in a unified readout fashion.
Differences in the distribution of qMRI metrics between HCs and people with
RRMS were demonstrated, with trends that are in agreement with
histopathological studies in the literature. However, a larger sample
population will be required to clarify the findings of this pilot study. Acknowledgements
This study is funded by the National Brain Appeal (MCY). Horizon2020 (Human Brain Project SGA3, Specific Grant Agreement No.
945539 ), BRC (#BRC704/CAP/CGW), MRC (#MR/S026088/1), Ataxia UK, MS Society
(#77), Wings for Life (#169111). CGWK is a shareholder in Queen Square
Analytics Ltd. RSS has received research funding from
the UK MS Society (grant #77), CureDRPLA and Ataxia UK & we thank the
UCL-UCLH Biomedical Research Centre for ongoing support. References
- Pollock M, Calder C, Allpress S. Peripheral nerve
abnormality in multiple sclerosis. Ann Neurol. 1977;2:41-48
- Misawa S, Kuwabara S, Mori M, et al. Peripheral nerve
demyelination in multiple sclerosis. Clin Neurophysiol.
2008;119:1829-1833.
- Jende J, Hauck GH, Diem R, et al. Peripheral nerve
involvement in multiple sclerosis: Demonstration by magnetic resonance neurography. Ann
Neurol. 2017;82:676-685
- Boonsuth R, Samson RS, Tur C, et al. Assessing Lumbar Plexus
and Sciatic Nerve Damage in Relapsing-Remitting Multiple Sclerosis Using
Magnetisation Transfer Ratio. Front Neurol. 2021;12:763143.
- Yiannakas MC, Schneider T, Yoneyama M, et al. Magnetisation
transfer ratio combined with magnetic resonance neurography is feasible in the
proximal lumbar plexus using healthy volunteers at 3T. Sci Rep.
2020;10(1):14568.
- Granziera C, Wuerfel J, Barkhof F, et al. Quantitative magnetic
resonance imaging towards clinical application in multiple
sclerosis. Brain. 2021;144(5):1296-1311.
- Yoneyama M, Takahara T, Kwee TC, Nakamura M, Tabuchi T. Rapid
high resolution MR neurography with a diffusion-weighted pre-pulse. Magn
Reson Med Sci. 2013;12(2):111-119.
- Kasper JM, Wadhwa V, Scott KM, et al. SHINKEI-a novel
3D isotropic MR neurography technique: technical advantages over 3DIRTSE-based
imaging, Eur. Radiol. 2015;25(6):1672–1677.
- Jenkinson M, Smith S. A global optimisation method for robust
affine registration of brain images. Med Image Anal. 2001;5(2):143-156.
- Modat M, Mcclelland J, Ourselin S. Lung registration using the NiftyReg
package. Medical image analysis for the clinic-a grand Challenge. 2010;33-42.
- Hansen B, Jespersen SN. Data
for evaluation of fast kurtosis strategies, b-value optimization and
exploration of diffusion MRI contrast. Sci Data. 2016;3:160072.
doi:10.1038/sdata.2016.72.
- Garyfallidis E, Brett M, Amirbekian
B, et al. Dipy, a library for the analysis of diffusion MRI data. Front
Neuroinform. 2014;8:8.