Fast CT-PRESS-based spiral chemical shift imaging at 3 Tesla.

A new sequence is presented that combines constant-time point-resolved spectroscopy (CT-PRESS) with fast spiral chemical shift imaging. It allows the acquisition of multivoxel spectra without line splitting with a minimum total measurement time of less than 5 min for a field of view of 24 cm and a nominal 1.5x1.5-cm2 in-plane resolution. Measurements were performed with 17 CS encoding steps in t1 (Deltat1=12.8 ms) and an average echo time of 151 ms, which was determined by simulating the CT-PRESS experiment for the spin systems of glutamate (Glu) and myo-inositol (mI). Signals from N-acetyl-aspartate, total creatine, choline-containing compounds (Cho), Glu, and mI were detected in a healthy volunteer with no or only minor baseline distortions within 14 min on a 3 T MR scanner.     

Optimized glutamate detection at 3T

Purpose:

To identify the pulse sequence and acquisition parameters that result in the most accurate and repeatable measurements of glutamate (Glu) concentration in the brain at 3T.

Materials and Methods:

Simulations were performed to compare the accuracy and repeatability of 11 pulse sequences and acquisition parameters, within four general classes (PRESS, STEAM, Carr–Purcell PRESS [CPRESS] and TE averaged PRESS [JPRESS]), the majority of which were previously suggested as optimal for Glu detection. Three of the simulated acquisitions were implemented in a clinical scanner and measures of repeatability in vivo were compared to their simulated values.

Results:

Good agreement was demonstrated between simulated and experimentally determined measures of repeatability. Among the acquisitions considered, a CPRESS sequence with minimal echo time, together with, possibly, a short TE PRESS sequence, result in the most repeatable within session Glu measurements, while slightly overestimating the Glu concentration. Excellent accuracy is demonstrated by the simulations for a JPRESS sequence, at the expense of lower repeatability than optimal PRESS or CPRESS sequences.

Conclusion:

Further proof of concept is presented toward validation of a simulation approach to understand pulse sequence performance in measuring the concentration of a given metabolite. Improved within session Glu measurement repeatability is predicted for CPRESS and demonstrated in vivo. J. Magn. Reson. Imaging 2009;30:1155–1162. © 2009 Wiley‐Liss, Inc.     

Measurement of glycine in human prefrontal brain by point‐resolved spectroscopy at 7.0 tesla in vivo

Measurement of glycine in human frontal brain by an optimized point‐resolved spectroscopy sequence at 7 T is reported. Echo time dependencies of the overlapping coupled resonances of myo‐inositol, free choline, and threonine were investigated with density matrix simulations, incorporating the slice‐selective radiofrequency and gradient pulses. The numerical simulations indicated that the selectivity of the 3.55‐ppm glycine singlet is maximized at (TE₁, TE₂) = (101, 51) ms. Phantom experiments indicated that the myo‐inositol peak amplitude between 3.5 and 3.6 ppm is reduced by a factor of 30 following the optimized point‐resolved spectroscopy, as predicted by the simulation. From LCModel analyses, the glycine concentration in the medial prefrontal cortex in healthy adults was estimated, with a mean Cramér‐Rao lower bound of 7 ± 1% (mean ± standard deviation; n = 7), to be 0.8 ± 0.1 mM, with reference to total creatine at 8 mM. Magn Reson Med, 2009. © 2009 Wiley‐Liss, Inc.     

Interindividual reproducibility of glutamate quantification using 1.5-T proton magnetic resonance spectroscopy.

The goal of this study was to measure the interindividual reproducibility of glutamate quantification in 1.5-T (1)H MRS of human brains. To determine the effective echo time (TE) for glutamate quantification, spectra from a phantom and 12 participants were obtained with TE = 30, 35, 40, and 144 ms (repetition time (TR) = 2000 ms and volume of interest = 4 cm(3)). The average Cramer-Rao lower bounds for glutamate quantification using LCModel was lowest in two experiments when TE = 40 ms.Twenty-one subjects participated in experiments that measured interindividual reproducibility of glutamate quantification. Spectra were acquired with TR = 6000 ms and TE = 40 ms. Results showed that the coefficients of variance were 11.0 and 13.1% in the anterior cingulate cortex and insula, respectively. This suggests that glutamate can be reproducibly measured from 1.5-T (1)H MRS with long TR, effective TE, and the LCModel.     

13C editing of glutamate in human brain using J-refocused coherence transfer spectroscopy at 4.1 T

The method of single quantum ¹³C editing is analyzed and implemented with water suppressed J-refocused coherence transfer spectroscopy. Analysis of the ¹³C inversion pulse demonstrates that it is optimally placed into the second echo of the J-refocused sequence. We have used this method to acquire ¹³C-edited spectra of glutamate from phantoms and in vivo. The turnover of ¹³C4-labeled glutamate in human brain in vivo was observed in parasagittal gray matter using a volume head coil at 4.1 T with a time resolution of 5.3 min.     

Fast proton spectroscopic imaging with high signal-to-noise ratio: spectroscopic RARE.

A new fast spectroscopic imaging (SI) method is presented which is based on spatial localization by the fast MRI method of rapid acquisition with relaxation enhancement (RARE) and encoding of the chemical shift information by shifting the position of a refocusing 180 pulse in a series of measurements. This method is termed spectroscopic RARE. In contrast to spectroscopic ultrafast low-angle RARE (U-FLARE), the formation of two echo families (odd and even) is suppressed by using a train of 180 RF pulses with an internal four-step phase cycle. By this means a high signal-to-noise ratio (SNR) per unit measurement time is obtained, because the separation of odd and even echoes, as well as dummy echoes to stabilize the echo amplitudes, is not needed anymore. The method is of particular interest for detecting signals of coupled spins, as effective homonuclear decoupling can be achieved by use of constant evolution time chemical shift encoding. The pulse sequence was implemented on a 4.7 T imaging system, tested on phantoms, and applied to the healthy rat brain in vivo. Spectroscopic RARE is particularly useful if T2* double less-than sign T2, which is typically fulfilled for in vivo proton SI measurements at high magnetic field strength.     

Spectral simplification for resolved glutamate and glutamine measurement using a standard STEAM sequence with optimized timing parameters at 3, 4, 4.7, 7, and 9.4T.

The C4 multiplet proton resonances of glutamate (Glu) around 2.35 ppm and glutamine (Gln) around 2.45 ppm usually overlap in MR spectra, particularly at low- and mid-field strengths (1.5-4.7T). A spectral simplification approach is introduced that provides unobstructed Glu and Gln measurement using a standard STEAM localization sequence with optimized interpulse timings. The underlying idea is to exploit the dependence of response of a coupled spin system on the echo time (TE) and mixing time (TM) to find an optimum timing set (TE, TM), at which the outer-wings of C4 "pseudo-triplet" proton resonances of Glu and Gln are significantly suppressed while the central peaks are maintained. The spectral overlap is thus resolved as the overlap exists exclusively at the outer-wings and the central peaks are readily separated due to the approximate 0.1-ppm difference in chemical shift. Density matrix simulation for Glu, Gln, and other overlapping metabolites at 2.3-2.5 ppm was conducted to predict the optimum timing sets. The simulated, phantom, and in vivo results demonstrated that the C4 multiplet proton resonances of Glu and Gln can be resolved for unobstructed detection at 3T, 4T, and 4.7T. For simplicity, only simulated data are illustrated at 7T and 9.4T.     

Measurement of brain glutamate using TE-averaged PRESS at 3T.

A method is introduced that provides improved in vivo spectroscopic measurements of glutamate (Glu), glutamine (Gln), choline (Cho), creatine (Cre), N-acetyl compounds (NAtot, NAA + NAAG), and the inositols (mI and sI). It was found that at 3T, TE averaging, the f1 = 0 slice of a 2D J-resolved spectrum, yielded unobstructed signals for Glu, Glu + Gln (Glx), mI, NA(tot), Cre, and Cho. The C4 protons of Glu at 2.35 ppm, and the C2 protons of Glx at 3.75 ppm were well resolved and yielded reliable measures of Glu/Gln stasis. Apparent T1/T2 values were obtained from the raw data, and metabolite tissue levels were determined relative to a readily available standard. A repeatibility error of <5%, and a coefficient of variation (CV) of <10% were observed for brain Glu levels in a study of six normal volunteers.     

Detection of brain glutamate and glutamine in spectroscopic images at 4.1 T

Brain glutamate and glutamine were detected in healthy human volunteers in spectroscopic images with a nominal voxel size of 2.25 cm³ at an echo time of 15 ms. Due to the increased frequency separation and simplification of J-coupling patterns, the separate detection of brain glutamate and glutamine at short echo times was possible. Creatine, choline, and N-acetylaspartate with other N-acetylated compounds were also detected. The ratios of the metabolite resonance intensities were in agreement with previously published values.     

Comparative reliability of proton spectroscopy techniques designed to improve detection of J‐coupled metabolites

Improved detection of J‐coupled neurometabolites through the use of modified proton magnetic resonance spectroscopy (1H‐MRS) techniques has recently been reported. TE‐averaged point‐resolved spectroscopy (PRESS) uses the J modulation effects by averaging FIDs with differing echo times to improve detection of glutamate, while standard PRESS detection of glutamate can be improved by using an appropriate single echo determined from J‐modulation simulations. In the present study, the reliabilities of TE‐averaged PRESS, standard PRESS with TE = 40 ms, and standard PRESS with TE = 30 ms in detecting metabolite levels in the cingulate gyrus of the human brain at 3T were compared in six subjects. TE‐averaged PRESS measures showed a mean variability of 9% for N‐acetyl aspartate, choline, and creatine, compared with < 4% for the 30‐ and 40‐ms PRESS techniques. The coefficients of variation for glutamate were 10%, 7%, and 5% for TE‐averaged, 30‐ms, and 40‐ms PRESS, respectively. PRESS with a TE of 40 ms also demonstrated improved reliability for GABA and glutamine concentrations. These results show that with the appropriate selection of echo time standard PRESS can be a reliable ¹H‐MRS technique for the measurement of J‐coupled neurometabolites in the human brain and, moreover, compares favorably with at least one J‐edited technique. Magn Reson Med 60:964–969, 2008. © 2008 Wiley‐Liss, Inc.     

Spectroscopic imaging of glutamate C4 turnover in human brain.

One-dimensional spectroscopic imaging of (13)C-4-glutamate turnover is performed in the human brain with a 6 cc nominal voxel resolution at 4T. Data were acquired with an indirect detection approach using a short spin echo single quantum (1)H-(13)C heteronuclear editing method and a 7 cm surface coil with quadrature (13)C decoupling coils. To analyze the data as a function of tissue type, T(1)-based image segmentation through the surface coil was performed to determine the gray and white matter contributions to each voxel. The tricarboxylic acid (TCA) cycle rate in gray and white matter was then determined using a two-compartment model with the tissue fractionation derived from the image segmentation. The mean values for the TCA cycle rate for occipital gray and white matter from three volunteers was 0.88 +/- 0.12 and 0.28 +/- 0.13 respectively, in agreement with literature data.     

Measurement of glycine in the human brain in vivo by 1H‐MRS at 3 T: application in brain tumors

Glycine is a key metabolic intermediate required for the synthesis of proteins, nucleic acids, and other molecules, and its detection in cancer could, therefore, provide biologically relevant information about the growth of the tumor. Here, we report measurement of glycine in human brain and gliomas by an optimized point‐resolved spectroscopy sequence at 3 T. Echo time dependence of the major obstacle, myo‐inositol (mI) multiplet, was investigated with numerical simulations, incorporating the 3D volume localization. The simulations indicated that a subecho pair (TE₁, TE₂) = (60, 100) ms permits detection of both glycine and mI with optimum selectivity. In vivo validation of the optimized point‐resolved spectroscopy was conducted on the right parietal cortex of five healthy volunteers. Metabolite signals estimated from LCModel were normalized with respect to the brain water signal, and the concentrations were evaluated assuming the total creatine concentration at 8 mM. The glycine concentration was estimated as 0.6 ± 0.1 mM (mean ± SD, n = 5), with a mean Cramér‐Rao lower bound of 9 ± 1%. The point‐resolved spectroscopy sequence was applied to measure the glycine levels in patients with glioblastoma multiforme. Metabolite concentrations were obtained using the water signal from the tumor mass. The study revealed that a subset of human gliomas contains glycine levels elevated 1.5–8 fold relative to normal. Magn Reson Med, 2011. © 2011 Wiley‐Liss, Inc.     

T2 measurement and quantification of glutamate in human brain in vivo.

The proton NMR transverse relaxation time T(2) of glutamate (Glu) in human brain was measured by means of spectrally selective refocusing at 3.0 T in vivo. An 81.4-ms-long dual-band Gaussian 180 degrees RF pulse, designed for refocusing at 2.35 and 3.03 ppm, was employed within point-resolved spectroscopy (PRESS) to generate the Glu C4-proton target multiplet and the total creatine (tCr) singlet. Six optimal echo times (TEs) between 128 and 380 ms were selected from numerical analysis of the filtering performance for effective detection of the Glu signal with minimal contamination from glutamine (Gln), N-acetylaspartate (NAA), and glutathione (GSH). The magnetization of Glu and tCr was extracted from spectral fitting of experimental and calculated spectra. Apparent T(2) values of Glu and tCr were estimated as 201 +/- 18 and 164 +/- 12 ms for the medial prefrontal (PF) cortex, and 198 +/- 22 and 169 +/- 15 ms (mean +/- SD, N = 5) for the left frontal (LF) cortex, respectively. With water segmentation data, the magnetization values of Glu and tCr of the two adjacent voxels, calculated from the T(2) values and spectra following the thermal equilibrium magnetization, were combined to give the Glu and tCr concentrations as 10.37 +/- 1.06 and 8.87 +/- 0.56 mM for gray matter (GM), and 5.06 +/- 0.57 and 5.16 +/- 0.45 mM (mean +/- SD, N = 5) for white matter (WM), respectively.     

In vivo 1H NMR spectroscopy of the human brain at 7 T.

In vivo 1H NMR spectra from the human brain were measured at 7 T. Ultrashort echo-time STEAM was used to minimize J-modulation and signal attenuation caused by the shorter T2 of metabolites. Precise adjustment of higher-order shims, which was achieved with FASTMAP, was crucial to benefit from this high magnetic field. Sensitivity improvements were evident from single-shot spectra and from the direct detection of glucose at 5.23 ppm in 8-ml volumes. The linewidth of the creatine methyl resonance was at best 9 Hz. In spite of the increased linewidth of singlet resonances at 7 T, the ability to resolve overlapping multiplets of J-coupled spin systems, such as glutamine and glutamate, was substantially increased. Characteristic spectral patterns of metabolites, e.g., myo-inositol and taurine, were discernible in the in vivo spectra, which facilitated an unambiguous signal assignment.     

In vivo detection of serine in the human brain by proton magnetic resonance spectroscopy (1H‐MRS) at 7 Tesla

A single‐voxel proton magnetic resonance spectroscopy (¹H‐MRS) filtering strategy for in vivo detection of serine (Ser) in human brain at 7T is proposed. Spectral difference of coupled resonances arising from different subecho times of triple refocusing at a constant total echo time (TE) was utilized to detect the Ser multiplet and cancel the overlapping creatine (Cr) 3.92‐ppm singlet via difference editing. Dependence of the Ser signal on subecho times was investigated using density‐matrix simulation incorporating the slice‐selective radio frequency (RF) pulses. The simulation indicated that the difference‐edited Ser CH₂ multiplet at ～3.96 ppm is maximized with (TE₁, TE₂, TE₃) = (54, 78, 78) and (36, 152, 22) ms. The edited Ser peak amplitude was estimated, with both numerical and phantom analyses of the performance, as 83% with respect to 90° acquisition for a localized volume, ignoring relaxation effects. From the area ratio of the edited Ser and unedited Cr 3.03‐ppm peaks, assuming identical T₁ and T₂ between Ser and Cr, the Ser‐to‐Cr concentration ratio for the frontal cortex of healthy adults was estimated to be 0.8 ± 0.2 (mean ± SD; N = 6). Magn Reson Med, 2009. © 2009 Wiley‐Liss, Inc.     

Determination of regional brain temperature using proton magnetic resonance spectroscopy to assess brain-body temperature differences in healthy human subjects.

Proton magnetic resonance spectroscopy ((1)H MRS) was used to determine brain temperature in healthy volunteers. Partially water-suppressed (1)H MRS data sets were acquired at 3T from four different gray matter (GM)/white matter (WM) volumes. Brain temperatures were determined from the chemical-shift difference between the CH(3) of N-acetyl aspartate (NAA) at 2.01 ppm and water. Brain temperatures in (1)H MRS voxels of 2 x 2 x 2 cm(3) showed no substantial heterogeneity. The volume-averaged temperature from single-voxel spectroscopy was compared with body temperatures obtained from the oral cavity, tympanum, and temporal artery regions. The mean brain parenchyma temperature was 0.5 degrees C cooler than readings obtained from three extra-brain sites (P < 0.01). (1)H MRS imaging (MRSI) data were acquired from a slice encompassing the single-voxel volumes to assess the ability of spectroscopic imaging to determine regional brain temperature within the imaging slice. Brain temperature away from the center of the brain determined by MRSI differed from that obtained by single-voxel MRS in the same brain region, possibly due to a poor line width (LW) in MRSI. The data are discussed in the light of proposed brain-body temperature gradients and the use of (1)H MRSI to monitor brain temperature in pathologies, such as brain trauma.