Investigating GABA concentrations measured with macromolecule suppressed and unsuppressed MEGA-PRESS MR spectroscopy and their relationship with BOLD responses in the occipital cortex

Niall-William Duncan, Jianfeng Zhang, Georg Northoff, Xuchu Weng

Research output: Contribution to journalArticle


Background: A combination of magnetic resonance spectroscopy (MRS) and functional (f)MRI is a promising method for studying brain activity. Negative results have, however, produced uncertainty as to the validity of the approach. Using a MEGA-PRESS sequence adapted to suppress the macromolecule signal (GABA–) has been suggested as a key methodological improvement, but there is some doubt as to the relationship between such estimates and those from the standard sequence (GABA+), making interpretation difficult. Purpose: To investigate the relationship between GABA+ and GABA– estimates from the posterior cingulate and occipital cortices. The second aim was to test for a correlation between occipital GABA and blood oxygenation level-dependent (BOLD) responses in the visual cortex to establish which of the two MEGA-PRESS sequences was more related to the functional responses. Study Type: Prospective. Subjects: Thirty-one healthy participants. Field Strength/Sequence: 3T/single-voxel 1 H-MRS and gradient-echo echo planar imaging (EPI). Assessment: GABA estimates were made using the Gannet toolbox. fMRI data were analyzed with FSL and Python scripts. Statistical Test: Relationships between different variables were tested with Pearson's correlation. Results: GABA+ and GABA– concentrations were found to be correlated in both regions (r = 0.52, 95% confidence interval [CI] = 0.35 0.66, p FDR = 0.002). No relationship was found between either the GABA+ or the GABA– concentrations and the amplitude of the BOLD response in the occipital cortex (GABA+, r = –0.14, p FDR > 0.1; GABA–, r = –0.29, p FDR >0.1). However, adding these results to those of prior studies in a meta-analysis of correlation coefficients did provide overall support for a negative correlation between GABA and BOLD response amplitudes (r = –0.39, 95% CI = –0.15–0.64). Data Conclusion: The current findings highlight potential methodological issues that continue to interfere with relating MRS GABA estimates with fMRI responses but, taken in sum, provide support for this general approach. Level of Evidence: 1. Technical Efficacy: Stage 1. J. Magn. Reson. Imaging 2019.

Original languageEnglish
Pages (from-to)1285-1294
Number of pages10
JournalJournal of Magnetic Resonance Imaging
Issue number4
Publication statusPublished - Oct 1 2019


ASJC Scopus subject areas

  • Radiology Nuclear Medicine and imaging

Cite this