Quantification of the neurochemical profile using simulated macromolecule resonances at 3 T

• Published in: NMR in biomedicine Vol. 26; no. 5; pp. 593 - 599
• Main Authors: Schaller, Benoît, Xin, Lijing, Cudalbu, Cristina, Gruetter, Rolf
• Format: Journal Article
• Language: English
• Published: England Blackwell Publishing Ltd 01.05.2013; Wiley Subscription Services, Inc
• Subjects: Brain; Brain - metabolism; gamma-Aminobutyric Acid - analysis; Glutamic Acid - analysis; Glutamine - analysis; Humans; in vivo 1H MRS; Macromolecular Substances - metabolism; macromolecule; Magnetic Resonance Spectroscopy - methods; Mathematics; quantification differences; spin echo full intensity acquired localized (SPECIAL); spline baseline
• ISSN: 0952-3480; 1099-1492; 1099-1492
• DOI: 10.1002/nbm.2896

The broad resonances underlying the entire 1H NMR spectrum of the brain, ascribed to macromolecules, can influence metabolite quantification. At the intermediate field strength of 3 T, distinct approaches for the determination of the macromolecule signal, previously used at either 1.5 or 7 T and higher, may become equivalent. The aim of this study was to evaluate, at 3 T for healthy subjects using LCModel, the impact on the metabolite quantification of two different macromolecule approaches: (i) experimentally measured macromolecules; and (ii) mathematically estimated macromolecules. Although small, but significant, differences in metabolite quantification (up to 23% for glutamate) were noted for some metabolites, 10 metabolites were quantified reproducibly with both approaches with a Cramer–Rao lower bound below 20%, and the neurochemical profiles were therefore similar. We conclude that the mathematical approximation can provide sufficiently accurate and reproducible estimation of the macromolecule contribution to the 1H spectrum at 3 T. Copyright © 2013 John Wiley & Sons, Ltd. The determination of the macromolecule contribution to the 1H spectrum varies with the field strength. The present study evaluates two different macromolecule estimations, experimentally measured macromolecules and mathematically estimated macromolecules, and studies their impact on the quantification. Although small, but significant, differences in the metabolite quantification (<23%) were noted, the neurochemical profile provided by the two approaches was similar. The LCModel built‐in simulated macromolecule is thus expected to yield a sufficient approximation of the macromolecule signal at 3 T.