Optimized multi-voxel TE-averaged PRESS for glutamate detection in the human brain at 3T

Abstract

Purpose: To optimize possible combinations of echo times (TE) for multi-voxel TE-averaged Point RESolved Spectroscopy (PRESS) while reducing the total number of TEs required to separate glutamate (Glu) and glutamine (Gln) within a clinically feasible scan time.

Methods: General Approach to Magnetic resonance Mathematical Analysis (GAMMA) was used to implement 2D J-resolved PRESS technique, and the spectra of 14 individual brain metabolites were simulated at 64 different TEs. Monte Carlo simulations were used for selecting the best TE combinations to separate Glu and Gln using TE-averaged PRESS with a total number of two, three, four and five TEs. Single-voxel ¹H-MRS data were acquired using 64 different TEs from a healthy volunteer on a clinical 3T MR scanner to validate the echo time combinations selected with simulations. Additionally, 2D ¹H-MRSI data of eight healthy volunteers were acquired on a clinical 3T MR scanner using four different TEs that were determined by Monte Carlo simulations. Optimized TE-averaged PRESS spectra were created by averaging the spectra acquired at selected TEs. LCModel was used for spectral quantification. A Wilcoxon signed-rank test was used to detect statistically significant differences in Glu/Gln ratios between 35 ms PRESS and optimized TE-averaged PRESS data.

Results: Glu could be clearly separated from Gln at 2.35 ppm, using optimized TE-averaged PRESS with only four TEs (35, 37, 40, and 42 ms) that were selected through Monte Carlo simulations. Glu/Gln ratios were significantly higher in the optimized TE-averaged PRESS data of healthy volunteers than in the 35 ms PRESS data (P = 0.008).

Conclusion: Optimized multi-voxel TE-averaged PRESS enabled faster and unobstructed quantification of Glu at multiple voxels in the human brain in vivo at 3T.

Keywords: 3T; Human brain; In vivo; Multi-voxel; Proton MR spectroscopic imaging; Spectral editing; TE-Averaged PRESS.