Toward understanding transverse relaxation in human brain through its field dependence

Abstract

Apparent transverse relaxation rate constants (R2† = 1/T2†) were measured in various regions of the healthy human brain using a multi-echo adiabatic spin echo (MASE) sequence at five different magnetic fields, 1.5, 1.9, 3, 4.7, and 7T. The R2† values showed a clear dependence on magnetic field strength (B0). The regional distribution of the R2† was well explained by the sum of three components: 1) regional non-hemin iron concentration ([Fe]), 2) regional macromolecular mass fraction (fM), 3) a region independent factor. Accordingly, R2† = α[Fe] + βfM + γ, where coefficients α, β, and γ were experimentally determined at each magnetic field, by a least square fitting method using multiple regression analysis. Whereas the coefficient α linearly increased with B0, β showed a quadratic dependence on top of a field independent component. The coefficient γ also increased slightly with B0, on top of a field independent component. The linear dependence of α on B0 was consistent with that observed for the transverse relaxation rate of water protons in ferritin solutions, as found previously by others. The quadratic dependence of β on B0 was accounted for by iso- and anisochronous exchange mechanisms using intrinsic relaxation parameters obtained from the literature.

Cognitive Implications of Deep Gray Matter Iron in Multiple Sclerosis

“…While R2* and QSM are complementary measures, they have distinct differences. First, the iron sensitivity of R2* is highly field-dependent, with higher fields being advantageous, 29 while QSM is largely field-insensitive. 30 Second, R2* is more susceptible to water content with inflammation weakening the R2* signal but having little effect on QSM.…”

Value of transverse relaxometry difference methods for iron in human brain

“…In-vitro R 2 measurements in aqueous solution of ferritin showed very high correlation between R 2 and ferritin concentration (33) however, complex tissue structures, including variable water and macromolecular content, reduce in-vivo correlations to iron (14,31).…”

“…For example, tissue neurodegeneration may have contributions from both iron-independent and iron-related processes (25,26). A correlation between R 2 and post-mortem iron concentration in deep grey matter at 1.5 T was first reported by Drayer et al (10) and has since been investigated extensively by other groups (12,14,20,(27)(28)(29)(30)(31) at different field strengths ranging from 0.5 T to 7.0 T.…”

“…Note several previous studies (2,14,20,24,25,(28)(29)(30)(31)(39)(40)(41)(42)(43)(59)(60)(61) have used the age versus regional iron content best fits of post-mortem data reported by Hallgren and Sourander (5) to estimate iron content using relaxometry such as R 1 (14,60), R 2 (28,32), R 2 * (32,59) and R 2 ' (28,32) .…”

“…To overcome high field RF inhomogeneity, methods such as twice refocused adiabatic pulses (31) or stimulated echo compensation (38) are required.…”