Unravel the MRI-based microstructural signatures behind the primary phenotypes of progressive and relapsing-remitting multiple sclerosis
This article was originally published here
J Magn Reson Imaging. June 30, 2021. doi: 10.1002 / jmri.27806. Online ahead of print.
BACKGROUND: The mechanisms underlying the primary phenotypes of progressive relapsing-remitting multiple sclerosis (PPMS / RRMS) are unknown. Magnetic resonance imaging (MRI) studies support the involvement of gray matter (GM) in degeneration, highlighting its damage as an early feature of both phenotypes. However, the role of the GM microstructure is unclear, calling for new methods for its decryption.
OBJECTIVE: To study the GM morphometric and microstructural differences between PPMS and RRMS in order to characterize the degeneration of GM tissues using MRI.
TYPE OF STUDY: Prospective cross-sectional study.
SUBJECTS: Forty-five PPMS (26 women) and 45 RRMS (32 women) patients.
FIELD STRENGTH / SEQUENCE: 3T scanner. T1-weighted three-dimensional (3D) fast field echo (T1-w), 3D turbo spin echo (TSE) T2-w, TSE 3D fluid-attenuated inversion recovery and MRI of echo-echo planar imaging scattering spin (dMRI).
EVALUATION: T1-w and dMRI data were used to provide information on morphometric and microstructural characteristics, respectively. For dMRI, diffusion tensor imaging and 3D single harmonic oscillator-based reconstruction and estimation models were used for feature extraction from a predefined set of regions. A support vector machine (SVM) was used to perform patient classification based on all of these measurements.
STATISTICAL TESTS: Differences between MS phenotypes were investigated using analysis of covariance and statistical tests (P
RESULTS: All dMRI indices showed significant microstructural alterations between the MS phenotypes considered, for example, the mode and the median of the probability of return to plan in the hippocampus. Conversely, thalamic volume was the only morphometric characteristic significantly different between the two groups of MS. Ten of the 12 characteristics retained by the selection process as discriminating in the two groups of MS concerned the hippocampus. The SVM classifier using these selected features achieved 70% accuracy and 69% accuracy.
DATA CONCLUSION: We have provided evidence to support the ability of dMRI to discriminate between PPMS and RRMS, as well as highlight the central role of the hippocampus.
LEVEL OF EVIDENCE: 2 TECHNICAL EFFICIENCY STAGE: 3.