Reproducibility of phase rotation STEAM at 3T: Focus on glutathione

Abstract

Purpose The purpose of this study was to determine the reproducibility of a very short echo time (TE) phase rotation stimulated echo acquisition mode (STEAM) sequence at 3T with a focus on the detection of glutathione. Methods Ten healthy subjects were scanned on two separate visits. Spectra were acquired from voxels placed in the anterior and posterior cingulates. Reproducibility was assessed using mean coefficients of variation (CVs) and mean absolute differences (ADs), and reliability was assessed using standard error of measurement (SEM) and intraclass correlations (ICCs). Phantoms containing glutathione and metabolites with overlapping resonances were scanned to test the validity of glutathione quantification. Results Excellent reproducibility as illustrated by CVs ≤8.3% and ADs ≤11.6% for both regions was obtained for glutathione and other commonly reported metabolites. Reproducibility measures for γ-aminobutyric acid and glutamine were good overall with CVs ranging from 6.4%-10.5% and ADs ranging from 8.6%-15.5% for both regions. Glutathione absolute and relative reliability were very good (SEMs ≤9.9%) and fair (ICCs = 0.42-0.51), respectively. Phantom studies demonstrated the ability to accurately detect glutathione from other metabolites with overlapping resonances with great precision (R2 = 0.99). Conclusion A very short TE phase rotation STEAM sequence proved reproducible for metabolites difficult to quantify but important for the study of psychiatric and neurological illness. Magn Reson Med 72:603-609, 2014. © 2013 Wiley Periodicals, Inc.