Reliability of glutamate quantification in human nucleus accumbens using proton magnetic resonance spectroscopy at a 70-cm wide-bore clinical 3T MRI system

Abstract

The human nucleus accumbens is a challenging region to study using proton magnetic resonance spectroscopy (1H-MRS) on a 70-cm wide-bore clinical 3T MRI system. The aim of this study was to investigate the reliability for quantitative measurement of glutamate concentration in the nucleus accumbens using a 70-cm wide-bore clinical 3T MRI. 1H-MRS of the nucleus accumbens was acquired using the Point-Resolved Spectroscopic Sequence (PRESS) with echo time of 40 ms from 10 healthy volunteers (5 female; age range: 18-30 years) on two separate visits (a baseline, and 1-month time point). The Java-based Magnetic Resonance User Interface (jMRUI) software package was used to quantitatively measure the absolute metabolite concentrations. The test-retest reliability and reproducibility were assessed using intraclass correlations coefficients (ICC), and coefficients of variation (CV). Glutamate concentrations were similar across visits (P = 0.832). Reproducibility measures for all metabolites were good with CV ranging from 7.8 to 14.0%. The ICC values of all metabolites for the intra-class measures were excellent (ICC > 0.8), except that the reliability for Glx (glutamate + glutamine) was good (ICC = 0.768). Pearson correlations for all metabolites were all highly significant (r = 0.636-0.788, P < 0.05). In conclusion, the short-echo-time PRESS can reliably obtain high quality glutamate spectrum from a ~3.4 cm3 voxel of the nucleus accumbens using a 70-cm wide-bore clinical 3T MRI.