Can T₁w/T₂w ratio be used as a myelin‐specific measure in subcortical structures? Comparisons between FSE‐based T₁w/T₂w ratios, GRASE‐based T₁w/T₂w ratios and multi‐echo GRASE‐based myelin water fractions

Abstract

Given the growing popularity of T -weighted/T -weighted (T w/T w) ratio measurements, the objective of the current study was to evaluate the concordance between T w/T w ratios obtained using conventional fast spin echo (FSE) versus combined gradient and spin echo (GRASE) sequences for T w image acquisition, and to compare the resulting T w/T w ratios with histologically validated myelin water fraction (MWF) measurements in several subcortical brain structures. In order to compare these measurements across a relatively wide range of myelin concentrations, whole-brain T w magnetization prepared rapid acquisition gradient echo (MPRAGE), T w FSE and three-dimensional multi-echo GRASE data were acquired from 10 participants with multiple sclerosis at 3 T. Then, after high-dimensional, non-linear warping, region of interest (ROI) analyses were performed to compare T w/T w ratios and MWF estimates (across participants and brain regions) in 11 bilateral white matter (WM) and four bilateral subcortical grey matter (SGM) structures extracted from the JHU_MNI_SS 'Eve' atlas. Although the GRASE sequence systematically underestimated T w/T w values compared to the FSE sequence (revealed by Bland-Altman and mountain plots), linear regressions across participants and ROIs revealed consistently high correlations between the two methods (r = 0.62 for all ROIs, r = 0.62 for WM structures and r = 0.73 for SGM structures). However, correlations between either FSE-based or GRASE-based T w/T w ratios and MWFs were extremely low in WM structures (FSE-based, r = 0.000020; GRASE-based, r = 0.0014), low across all ROIs (FSE-based, r = 0.053; GRASE-based, r = 0.029) and moderate in SGM structures (FSE-based, r = 0.20; GRASE-based, r = 0.17). Overall, our findings indicated a high degree of correlation (but not equivalence) between FSE-based and GRASE-based T w/T w ratios, and low correlations between T w/T w ratios and MWFs. This suggests that the two T w/T w ratio approaches measure similar facets of subcortical tissue microstructure, whereas T w/T w ratios and MWFs appear to be sensitized to different microstructural properties. On this basis, we conclude that multi-echo GRASE sequences can be used in future studies to efficiently elucidate both general (T w/T w ratio) and myelin-specific (MWF) tissue characteristics.

Gray matter T1‐w/T2‐w ratios are higher in Alzheimer's disease

“…Previous studies have suggested that myelin is the predominant source of MR contrast in T1‐weighted and T2‐weighted images (Bock, Kocharyan, Liu, & Silva, ; Eickhoff et al, ; Laule et al, ; Wallace et al, ), and by using the ratio of these images the contrast sensitivity to detect myelin content is increased further due to attenuation of the shared intensity biases (Glasser & Van Essen, ). Although, other studies have questioned the sensitivity of the T1‐w/T2‐w ratio for myelin content, and argued that it is likely not the only factor contributing to the T1‐w/T2‐w image (Arshad et al, ; Righart et al, ; Uddin et al, ). Still, it has been shown that T1‐w/T2‐w ratio differentiates between high and low myelinated cortex as determined by myelin proteolipid protein staining (Nakamura et al, ).…”

“…In disorders with a strong demyelinating component such as multiple sclerosis (MS), the T1‐w/T2‐w ratio is lowered in pathologically vulnerable regions (Beer et al, ; Nakamura, Chen, Ontaneda, Fox, & Trapp, ). However, recent studies have raised controversy on the interpretation of the T1‐w/T2‐w ratio, showing that this measure may also reflect tissue microstructure other than myelin, such as axon and dendrite density or iron content (Arshad, Stanley, & Raz, ; Righart et al, ; Uddin, Figley, Marrie, & Figley, ; van Rooden et al, ). Because disrupted brain connectivity is an important feature of AD (Dicks et al, ; Fornito et al, ; Pievani et al, ), an intracortical reduction of myelin content in AD could be hypothesized.…”

MR‐based age‐related effects on the striatum, globus pallidus, and thalamus in healthy individuals across the adult lifespan

“…Resultantly, it has been suggested T1w/T2w is a less than optimal index of white matter myelin (Arshad et al, 2016;Uddin et al, 2018).…”

“…One prominent technique is myelin water imaging (MWI), which exploits the different T2-relaxation times of water trapped between myelin sheaths, yielding a myelin water fraction (MWF) parameter (MacKay et al, 1994;Uddin et al, 2018). The accuracy of MWI has been validated in studies correlating histopathological myelin samples and neuroimaging-derived MWF in animal (Beaulieu et al, 1998 (Laule et al, 2006(Laule et al, , 2008.…”

“…Quantitative MRI techniques to measure myelin have primarily consisted of T1 relaxation time and magnetization transfer ratio (MTR) maps, as well as T2 relaxation techniques that separate the MR signal to obtain an indirect measure of myelin.However, these quantitative techniques have also been shown to be influenced by other underlying biological processes (e.g., inflammation) that contribute to the myelin MRI signal obtained(Laule et al, 2007).One prominent technique is myelin water imaging (MWI), which exploits the different T2-relaxation times of water trapped between myelin sheaths, yielding a myelin water fraction (MWF) parameter(MacKay et al, 1994;Uddin et al, 2018). Specifically, with recent advances and increasing availability of quantitative MRI acquisition techniques, researchers are beginning to examine myelin using more direct measurements of microstructure, such as myelin.…”