Synopsis
In
one-dimensional (1D) MR Spectroscopy (MRS), it is difficult to resolve the
multitude of metabolite peaks that exist over a small spectral range.
Spectral-editing techniques target a particular J-coupled metabolite
selectively, such as lactate, GABA, glutamate, etc. with a drawback that only
one metabolite is selected for each recording. Due to the added 2nd dimension, two-dimensional (2D) MRS can unambiguously resolve
many overlapping peaks non-selectively. Instead of a standard 1D spectrum
plotting intensity versus a single-axis (i.e., chemical shift + J-coupling), 2D
MRS techniques produce a 2D spectrum plotting intensity versus two frequency
axes, the dimensions of which depend on the specific 2D MRS technique. A major
goal of this presentation is to give an overview of the basics of 2D MRS and
describe several localized 2D MRS sequences which have been implemented on the
whole body 1.5T, 3T, and 7T MRI scanners.
Target Audience
Basic Scientists and Medical Researchers interested in the basics of localized 2D MR Spectroscopy
Outcome/Objectives
To record localized 2D MR spectra with improved resolution that will facilitate better peak assignments and metabolite quantitation
Purpose
1) To discuss problems due to overlap and metabolite quantitation in water suppressed proton (1H) MR Spectroscopy;
2) To describe the basic theory of 2D MRS including localized correlated spectroscopy (L-COSY) and PRESS localized J-resolved Spectroscopy (JPRESS);
3) To discuss the necessary pulse sequences for 2D MRS: L-COSY, JPRESS, localized exchange spectroscopy (L-EXSY), constant time COSY (CT-COSY) and constant time PRESS (CT-PRESS)
Methods
Shown in Fig.1 and Fig.2 are 2D MRS sequences:
L-COSY (Fig.1A), JPRESS (Fig.1B), L-EXSY (Fig.1C), CT-COSY (Fig.2A) and
CT-PRESS (Fig.2B). As presented In
Fig.1A, the 2D L-COSY sequence uses a combination of three slice-selective RF
pulses (90
0-180
0-90
0) to localize a volume of
interest (VOI) or region of interest (ROI) in a single shot (1,2).
An incremental period for the second spectral dimension (t
1) was
inserted immediately after the Hahn spin-echo using the first 90
0
and 180
0 RF pulse pair. The last slice-selective 90
0 RF
pulse acted also as the coherence transfer pulse, critical for recording the 2D
spectrum (1-4). Refocusing B
0 gradient crusher pulses
around the slice-selective 180
0 RF pulse and also, before and after
the last 90
0 RF pulse enabled minimizing unwanted coherences. The
coherence transfer between J-coupled protons of different metabolites crucial
for the L-COSY sequence is presented in Fig.3. Fig. 4A shows a simulated 3T L-COSY
of more than 16 metabolites using the GAMMA library (5). 2D cross peaks due to NAA,
lactate (Lac). Glutamate (Glu) and myo-inositol (mI) are highlighted. Shown in
Fig.4B and 4C are the same 3T 2D L-COSY
spectrum recorded in a brain phantom containing all metabolites shown in Fig.4A
showing the J-connectivity of NAA and glutathione (GSH) protons.
The basic pulse sequence for JPRESS is shown in
Fig.1B where the VOI is localized using the conventional PRESS sequence (6) containing
three slice-selective RF pulses (90
0-180
0-180
0).
The incremental duration (t
1) was inserted immediately after the
first Hahn echo and the signal was read-out along the t
2 dimension (2,
7-8). The 2nd spectral bandwidth of JPRESS will be smaller than that
of L-COSY due to the refocusing of chemical shifts between different protons of
metabolites. The L-EXSY sequence is shown in Fig.1C, where the VOI is localized
using the conventional STEAM sequence (9) including three slice-selective 90
0
RF pulses. The 2nd spectral dimension encoding variable t
1
was inserted before the 2
nd 90
0 RF pulse (10). Multiple
averages could be used in combination with or without a multi-step RF
phase-cycling to improve the SNR from the localized volume and also, to
minimize any artifacts stemming from RF pulses with inaccurate flip-angles (90
0
and 180
0).
Two more variants of the localized 2D MRS, namely
CT-COSY and CT-PRESS are shown in Fig.2A and 2B, respectively (11-13). Compared to the basic localized COSY
spectrum, the CT-COSY spectrum can be easier to interpret with the reduced number
of cross peaks for each metabolite due to decoupling along F
1. CT-COSY spectra have more T2-weighting
due to longer constant time (T
c) values. In contrast to the basic L-COSY and CT-COSY
spectra, a further increase in SNR is possible by CT-PRESS for coupled
resonances since there is no coherence transfer of magnetization between the
J-coupled protons leading to the disappearance of cross-peaks. A second rf channel is not required to
achieve broadband decoupling. However,
two major drawbacks of the 2D CT-PRESS are: 1) The signal amplitude in
CT-PRESS depends on T
c. 2) The spectrum has to be acquired in the 2D
mode, which requires long acquisition time.
Since the raw data acquired using 2D MRS has two dimensions (t
1 and t
2), a double Fourier transformation will be necessary to obtain the final 2D MR spectrum. Hence, shown in Fig.5 are the raw 2D data, data after the 1
st FFT along the t
2 dimension, data after the 2
nd FFT along the t
1 dimension and 2D stack/contour plots after the double FFT.
Conclusion
Our preliminary results clearly indicate that 2D
L-COSY and JPRESS have better resolution than 1D MRS. Several metabolite
multiplets have been unambiguously resolved in contrast to 1D MRS. The 2D MRS
sequences have been implemented in both single and multi-voxel modes and
feasibility of recording 2D MR spectra in a clinical environment is established.
Acknowledgements
Grants support from 1) National Institute of Health (5R21NS080649-02,5R21NS090956-02, 5R21NS086449-02, P50CA092131) and Department of Defense CDMRP prostate cancer research program (PCRP #W81XWH-11-1-0248).References
1) Thomas MA, Yue K, Binesh N et al. Localized
Two-dimensional Shift Correlated MR Spectroscopy of Human Brain. Magn Reson in
Med 2001; 46: 58-67
2) Thomas MA,
Hattori N, Umeda M, Sawada T and Naruse S. Adding a new spectral dimension to
localized 3T 1H MR Spectroscopy- From Phantoms to Human Brain in vivo. NMR
Biomed 2003; 16:245-251.
3) Ernst RR,
Bodenhausen G and Wokaun A. Principles of NMR Spectroscopy in one and two
dimensions. Oxford Publications, Oxford, 1987.
4) Aue WP, Bartholdi E, and Ernst RR. Two-dimensional
spectroscopy - application to nuclear magnetic resonance. J Chem Phys. 1976;
64(5):2229-2246.
5) Smith S, Levante T, Meier BH, Ernst
RR. Computer simulations in magnetic resonance. An object-oriented programming
approach. J. Magn. Reson., Ser. A 1994; 106(1): 75–105.
6) Bottomley PA. Spatial localization in NMR spectroscopy in
vivo. Annals of the New York Academy of Sciences 1987;508:333-348.
7) Ryner LN,
Sorenson JA and Thomas MA. 3-D Localized 2D-NMR Spectroscopy on an MRI Scanner.
J Magn Reson B 1995; 107:
126-137.
8) Ryner LN, Sorenson JA and
Thomas MA. Localized 2D J-Resolved H-1 MR Spectroscopy: Strong Coupling Effects in vitro and in
vivo. Magn Reson in Imag
1995; 13: 853-869.
9) Bruhn H, Frahm J, Gyngell ML et al. Cerebral metabolism in man
after acute stroke: New observations using localized proton NMR spectroscopy.
Mag Reson Med 1989; 9: 126.
10) Thomas MA,
Chung HK, Middlekauff H. Localized two-dimensional 1H magnetic resonance exchange spectroscopy: a preliminary evaluation
in human muscle. Magn
Reson Med. 2005 Mar;53(3):495-502.
11) Chung HK,
Banakar S, Thomas MA.1143. Localized Constant-Time Correlated Spectroscopy
(CT-COSY) of Human Brain In Vivo. Proc ISMRM. 2003, 1143.
12) Watanabe H,
Takaya N, Mitsumori E. Absolute quantitation of glutamate, GABA and glutamine
using localized 2D constant-time COSY spectroscopy in vivo. Magn Reson Med Sci
2014;13:25-32
13) Chung HK,
Banakar S, Thomas MA. Broadband Decoupled and Single Voxel Localized
2D MR Spectroscopy. Proc ISMRM. 2004, 687.