In vivo quantification of intracerebral GABA by single-voxel 1H-MRS—How reproducible are the results?☆
Introduction
Gamma aminobutyric acid (GABA) is the chief inhibitory neurotransmitter in the central nervous system and plays a decisive role in a variety of nervous system disorders, such as anxiety disorders, epilepsy, schizophrenia, insomnia and many others [1].
To better understand the mechanisms of diseases like epilepsy, or to monitor medical treatment that affects GABA metabolism such as antiepileptic therapy, non-invasive techniques for detection of intracerebral GABA such as proton MR spectroscopy (1H-MRS) are of great importance. However, quantifying intracerebral GABA by 1H-MRS is not straightforward, as resonances of GABA overlap with signals of other metabolites such as glutamate (Glu), N-acetylaspartate (NAA), total creatine (tCr) and macromolecules (MM). In addition, the concentration and subsequently the signal-to-noise ratio (SNR) of GABA in cortical grey matter is low (∼1 mM in vivo) compared to commonly quantified metabolites [2], [3]. Both make reliable quantification of GABA difficult.
Therefore, quantification requires sophisticated editing sequences, which have been the topic of many studies [2], [4], [5], [6], [7], [8]. These sequences can be divided into the following groups: (1) J-difference editing methods that cancel the tCr signal at 3.03 ppm by taking advantage of the J-coupling between the GABA-3 and GABA-4 resonance [6]; (2) multiple quantum filtering methods that suppress overlapping signals of GABA-3/4 [7]; and (3) two-dimensional spectroscopy methods to separate GABA-3/4 from overlapping signals in the second spectral dimension [5]. Due to the conceptual simplicity and the ease of implementation, J-difference editing using Point Resolved Spectroscopy with MEGA suppression (MEGA-PRESS) [6] has evolved the most commonly used method.
Different ways of quantification have been employed in the past [2], [9]. Either internal referencing to tCr or external referencing to phantom solutions was applied. Advantages and disadvantages of different referencing methods have been discussed previously [10]. While external referencing may be prone to system instability, it is known that tCr that is frequently used for internal referencing may not be stable for comparisons between patients and healthy controls, as reported previously in many disorders [11], [12].
The aims of this study were fourfold: first, to determine the in vivo reproducibility of GABA quantification by a MEGA-PRESS editing sequence, comparable to the one used by Kaiser et al. [4]; second, to compare reproducibility for external referencing, and internal reference to water (H2O) and to tCr; third, to evaluate the inter- and intra-subject variability and scan–rescan reproducibility; and fourth, to compare two post-processing tools for quantification of signals. The first quantification method is based on integration using MestreC [13] spectroscopy processing software. The second method is time-domain fitting (AMARES) [14].
Section snippets
Subjects
Eleven healthy volunteers (age 30.0 ± 7.1 years, five male, six female) were included in this study. The study complied with the Declaration of Helsinki and was approved by our Institutional Review Board. Written, informed consent was obtained from all subjects. None of the healthy volunteers demonstrated pathological findings on MR imaging or had any known history of neurological or psychiatric disorders. All subjects were light drinkers (1–14.9 g/day ethanol), non-smokers or light smokers (<20
Results
The quantification of GABA was performed for 87 voxels in 11 subjects. Fig. 4 illustrates representative in vivo spectra in comparison to acquired phantom data (Fig. 3). The triplet structure of GABA is clearly visible in the unsuppressed phantom spectrum. In contrast a pseudo-doublet can be observed in the edited spectrum after subtraction. This doublet can be also resolved in vivo (Fig. 4). All results are shown in ratios to an internal reference and not in absolute values. However, this does
Discussion
In the present study, the intra- and inter-subject reproducibility of GABA quantification results employing a MEGA suppression method combined with different post-processing procedures and referencing methods was investigated. We compared internal reference to water and tCr, as well as quantification by time-domain fitting in contrast to signal integration. This knowledge is important for non-invasive determination of intracerebral GABA in many psychiatric and neurological diseases [1], [15],
Conclusion
The results of our study show that in vivo1H-MRS of intracerebral GABA can be a fast, and reproducible method on higher field (3.0 T) clinical routine scanners, if proper post-processing and internal referencing are employed. This method has the potential to be used in clinical and scientific settings to detect even small alterations in intracerebral GABA.
Conflict of interest statement
None declared.
Acknowledgments
This study was supported by ELAN (Fonds für Forschung und Lehre am Universitätsklinikum Erlangen), Wilhelm Sander Foundation and Österreichische Forschungsförderungsgesellschaft (FFG) mbH (Grant #812104).
References (19)
- et al.
In vivo detection of gray and white matter differences in GABA concentration in the human brain
Neuroimage
(2006) - et al.
Assessment of GABA concentration in human brain using two-dimensional proton magnetic resonance spectroscopy
Psychiatry Res-Neuroimag
(2000) - et al.
A practical guide to robust detection of GABA in human brain by J-difference spectroscopy at 3 T using a standard volume coil
Magn Reson Imag
(2007) - et al.
Absolute metabolite quantification by in vivo NMR spectroscopy. II. A multicentre trial of protocols for in vivo localised proton studies of human brain
Magn Reson Imag
(1998) - et al.
Increased thalamic N-acetylaspartate in male patients with familial bipolar I disorder
Psychiatry Res-Neuroimag
(2001) - et al.
Improved method for accurate and efficient quantification of MRS data with use of prior knowledge
J Magn Reson
(1997) - et al.
In vivo GABA detection with improved selectivity and sensitivity by localized double quantum filter technique at 4.1 T
Magn Reson Imag
(2004) - et al.
In vivo single-shot three-dimensionally localized multiple quantum spectroscopy of GABA in the human brain with improved spectral selectivity
J Magn Reson
(2005) - et al.
GABA, gamma-hydroxybutyric acid, and neurological disease
Ann Neurol
(2003)
Cited by (109)
The role of MRS-assessed GABA in human behavioral performance
2022, Progress in NeurobiologyCitation Excerpt :NAA is the largest concentration metabolite signal whereas unsuppressed water has the largest SNR but requires additional water-unsuppressed scans. In a small sample size, the measure of GABA/tCr was shown to have less inter- and intra-subject variability compared with the measure of GABA/water (Bogner et al., 2010). However, a recent multi-center study showed that the variation in the measure of GABA/water was similar to GABA/tCr measurement (Mikkelsen et al., 2019).
Alteration of gamma-aminobutyric acid in the left dorsolateral prefrontal cortex of individuals with chronic insomnia: a combined transcranial magnetic stimulation-magnetic resonance spectroscopy study
2022, Sleep MedicineCitation Excerpt :For GABA+, the ratio to creatine (Cr) was computed for analysis later on. This step intended to remove the influence of inter-scan signal intensity changes [40]. Additionally, the GABA level was the total amount of some contributions (GABA moiety of homocarnosine, free GABA and so on) and was therefore referred to as GABA+ [41].
Reduced GABA levels in the medial prefrontal cortex are associated with cognitive impairment in patients with NMOSD
2022, Multiple Sclerosis and Related DisordersCitation Excerpt :Gannet used a Gaussian model to fit GABA+ and Glx signals, and a Lorentzian model to fit the tCr signal. MRS findings were reported as the ratios of metabolites to tCr (GABA+/tCr, Glx/tCr) because (1) this quantification method exhibits a performance equal to or better than that of water reference (Bogner et al., 2010); (2) the tCr concentrations were shown to be stable and independent of NMOSD (Aboul-Enein et al., 2010; Duan et al., 2017); (3) normalization to tCr reduces inter-subject variance due to differences in global signal intensity and CSF fraction within voxels (Bathel et al., 2018; Ragland et al., 2020). For both VOIs, individual metabolic peak was excluded if the signal fitting error was above 15%.
Lower cortical gamma-aminobutyric acid level contributes to connectivity in sensory-motor inter-connected regions in progressive MS
2020, Multiple Sclerosis and Related DisordersReduced in vivo visual cortex GABA in schizophrenia, a replication in a recent onset sample
2020, Schizophrenia Research
- ☆
Parts of this study were presented at: Bogner, W., et al., Quantification of intercerebral GABA by 1H-MRS—How reproducible are the results? (scientific paper). ECR, 2008. Vienna, Control number: 5763 (B-429); Bogner, W., et al., Strategies for reliable quantification of intracerebral GABA by 1H-MRS (electronic poster). ISMRM, 2008. Toronto, ID# 4955.
- 1
Tel.: +43 1 40400 3772; fax: +43 1 40400 1773.
- 2
Tel.: +49 9131 85 39388; fax: +49 9131 85 36179.
- 3
Tel.: +49 9131 85 34259; fax: +49 9131 85 34271.
- 4
Tel.: +49 9131 85 34570; fax: +49 9131 85 34551.
- 5
Tel.: +43 1 40400 1773; fax: +43 1 40400 7631.
- 6
Tel.: +49 9131 85 34326; fax: +49 9131 85 36179.
- 7
Tel.: +49 9131 85 39116; fax: +49 9131 85 36469.
- 8
Tel.: +49 9131 85 34580; fax: +49 9131 85 36469.