Incorporating lactate/lipid discrimination into a spectroscopic imaging sequence

Lactate measurement by neurochemical profiling in the dorsolateral prefrontal cortex at 7T: accuracy, precision, and relaxation times

PURPOSE

This assesses the potential of measuring lactate in the human brain using three non-editing MRS methods at 7T and compares the accuracy and precision of the methods.

METHODS

¹ H MRS data were measured in the right dorsolateral prefrontal cortex using a semi-adiabatic spin-echo full-intensity acquired localized sequence with three different protocols: (I) TE = 16 ms, (II) TE = 110 ms, and (III) TE = 16 ms, TI = 300 ms. T₁ and T₂ relaxation times of lactate were also measured. Simulated spectra were generated for three protocols with known concentrations, using a range of spectral linewidths and SNRs to assess the effect of data quality on the measurement precision and accuracy.

RESULTS

Lactate was quantified in all three protocols with mean Cramér-Rao lower bound of 8% (I), 13% (II), and 7% (III). The T₁ and T₂ relaxation times of lactate were 1.9 ± 0.2 s and 94 ± 13 ms, respectively. Simulations predicted a spectral linewidth-associated underestimation of lactate measurement. Simulations, phantom and in vivo results showed that protocol II was most affected by this underestimation. In addition, the estimation error was insensitive to a broad range of spectral linewidth with protocol I. Within-session coefficient of variances of lactate were 6.1 ± 7.9% (I), 22.3 ± 12.3% (II), and 5.1 ± 5.4% (III), respectively.

CONCLUSION

We conclude that protocols I and III have the potential to measure lactate at 7T with good reproducibility, whereas the measurement accuracy and precision depend on spectral linewidth and SNR, respectively. Moreover, simulation is valuable for the optimization of measurement protocols in future study design and the correction for measurement bias.

Proton MRS and MRSI of the brain without water suppression

Water suppression (WS) techniques have played a vital role in the commencement and development of in vivo proton magnetic resonance spectroscopy (MRS, including spectroscopic imaging - MRSI). WS not only made in vivo proton MRS functionally available but also made its applications conveniently accessible, and it has become an indispensable tool in most of the routine applications of in vivo proton MR spectroscopy. On the other hand, WS brought forth some challenges. Therefore, various techniques of proton MRS without WS have been developed since the pioneering work in the late 1990s. After more than one and a half decades of advances in both hardware and software, non-water-suppressed proton MRS is coming to the stage of maturity and seeing increasing application in biomedical research and clinical diagnosis. In this article, we will review progress in the technical development and applications of proton MRS without WS.

Improving the spectral resolution and spectral fitting of (1) H MRSI data from human calf muscle by the SPREAD technique

Proton magnetic resonance spectroscopic imaging ((1) H MRSI) has been used for the in vivo measurement of intramyocellular lipids (IMCLs) in human calf muscle for almost two decades, but the low spectral resolution between extramyocellular lipids (EMCLs) and IMCLs, partially caused by the magnetic field inhomogeneity, has hindered the accuracy of spectral fitting. The purpose of this paper was to enhance the spectral resolution of (1) H MRSI data from human calf muscle using the SPREAD (spectral resolution amelioration by deconvolution) technique and to assess the influence of improved spectral resolution on the accuracy of spectral fitting and on in vivo measurement of IMCLs. We acquired MRI and (1) H MRSI data from calf muscles of three healthy volunteers. We reconstructed spectral lineshapes of the (1) H MRSI data based on field maps and used the lineshapes to deconvolve the measured MRS spectra, thereby eliminating the line broadening caused by field inhomogeneities and improving the spectral resolution of the (1) H MRSI data. We employed Monte Carlo (MC) simulations with 200 noise realizations to measure the variations of spectral fitting parameters and used an F-test to evaluate the significance of the differences of the variations between the spectra before SPREAD and after SPREAD. We also used Cramer-Rao lower bounds (CRLBs) to assess the improvements of spectral fitting after SPREAD. The use of SPREAD enhanced the separation between EMCL and IMCL peaks in (1) H MRSI spectra from human calf muscle. MC simulations and F-tests showed that the use of SPREAD significantly reduced the standard deviations of the estimated IMCL peak areas (p < 10(-8) ), and the CRLBs were strongly reduced (by ~37%).

J-difference lactate editing at 3.0 Tesla in the presence of strong lipids

PURPOSE

To implement in vivo detection of lactate in the presence of lipids by proton magnetic resonance spectroscopy at a 3 Tesla (T) field strength for potential applications in human tumors outside of the brain.

MATERIALS AND METHODS

The BASING J-difference sequence was implemented in the presence of high lipid concentrations in phantoms and in vivo at 3 Tesla.

RESULTS

The effectiveness of the lactate editing scheme is demonstrated in phantoms containing both lactate and lipids and in vivo in ischemic induced human muscle.

CONCLUSION

The ability of the BASING J-difference technique to detect lactate in the presence of strong lipid signals outside the brain at 3T is feasible. This robust technique should permit noninvasive lactate measurements in human tumors to investigate its potential as a prognostic indicator.

Lactate imaging with Hadamard-encoded slice-selective multiple quantum coherence chemical-shift imaging

The ability to generate in vivo maps of lactate may have significant diagnostic utility in staging and treatment planning of a wide variety of cancers. The double selective multiple quantum filter technique (SelMQC) has been shown to be effective for nonlocalized detection of lactate with little or no interference from other signals. Here the SelMQC technique has been combined with longitudinal Hadamard slice selection and chemical shift imaging (CSI) to yield slice-selective images of lactate. The technique is shown to be effective in phantoms and in WSU-DLCL2 xenografts implanted in flanks of SCID mice. Tumors exhibited an annulus of elevated lactate concentration surrounding a necrotic tumor core.

Water excitation as an alternative to fat saturation in MR imaging: preliminary results in musculoskeletal imaging

PURPOSE

To compare fat suppression methods by using spectrally selective fat saturation and section-selective water excitation in standard magnetic resonance (MR) imaging sequences used in day-to-day musculoskeletal practice.

MATERIALS AND METHODS

Eighty-three patients underwent MR examination with a 1.5-T system. The two methods were compared by using three common sequences: T1-weighted spin-echo (SE) imaging performed after contrast material injection (n = 24), intermediate-weighted fast SE (n = 36) imaging, and T2-weighted fast SE (n = 36) imaging. Acquisition times of the sequences and signal-to-noise and contrast-to-noise ratios of bone, muscle, fat, and water for the two methods were compared quantitatively. Images were then qualitatively reviewed by two radiologists who were blinded to the type of fat suppression used. Image quality was scored according to four criteria (homogeneity of fat suppression, susceptibility and foldover artifacts, conspicuousness of lesion, and overall image quality) by using a five-point scale (0, bad; 1, poor; 2, fair; 3, good; and 4, excellent). A paired Student t test was used to compare the quantitative data, and a nonparametric paired-data Wilcoxon signed rank test was used for qualitative analysis.

RESULTS

Water excitation allowed a substantial decrease in acquisition time (by up to 50%) for T1-weighted sequences. Quantitative measurements revealed a greater signal-to-noise ratio (P <.01) with water excitation for all three sequences, whereas the contrast-to-noise ratio was greater with water excitation only in intermediate-weighted sequences (P <.01). Qualitatively, water excitation proved statistically better than or equal to fat saturation for all criteria in all imaging sequences (P <.05). Mean scores of overall image quality ranged between 2.5 and 3.0 for fat saturation and 3.4 and 3.7 for water excitation, respectively (P <.05).

CONCLUSION

Section-selective water excitation is faster than conventional fat saturation and produces images of better quality.

Lactate discrimination incorporated into echo-planar spectroscopic imaging

A technique for discriminating a lactate signal from overlapping lipid signals in (1)H spectroscopic imaging is presented. It is based on J-coupling between lactate protons and on the broad spectral bandwidth of lipid signal. Measurement parameters used in the technique are determined so that TE is separated from n/J (n: a natural number, J: J-coupling constant) enough to suppress the lipid signal at the time when the lactate signal is strongest. Data processing is used to calculate the lactate signal intensity from the reconstructed spectra. This technique enables lactate to be discriminated in a single measurement and enables spectra of other metabolites to be acquired simultaneously. However, it necessitates a homogeneous magnetic field, long TE, and supplementary lipid suppression. Discrimination of the lactate signal is demonstrated by applying lactate-discriminating echo-planar spectroscopic imaging (EPSI), which combines this discrimination technique with the standard EPSI, to rat focal cerebral ischemia models. Magn Reson Med 45:568-574, 2001.

In vivo lactate editing in single voxel proton spectroscopy and proton spectroscopic imaging by homonuclear polarisation transfer

Volume-selective lactate editing has been performed successfully in vitro and in vivo in the brain on a clinical scanner using a PRESS-based single voxel 1H spectroscopy and a 1H spectroscopic imaging sequence. The PRESS sequence was made sensitive to homonuclear polarisation by replacing the standard 180 degree refocusing pulses with 90 degree pulses. Two acquisitions were made at a total echo time around 2/J (J is the coupling constant for CH and CH3 spins in lactate approximately 7 Hz) whose individual echo times differed by 5.5 ms. Subtraction of one signal from the other yielded the lactate resonance alone. The technique is an effective method of separating the overlapping signals of lactate and lipids. Furthermore this editing method can be performed without state of the art MRI scanner hardware.

Feasibility study of lactate imaging of head and neck tumors

A proton spectroscopic imaging sequence was used to investigate the feasibility of lactate imaging in head and neck tumors. The sequence employs a two-shot lactate editing method with inversion recovery for additional lipid suppression, and a restricted field of view to suppress motion artifacts. Variations in acquisition parameters and two different receive coils were investigated on twelve patients. Elevated lactate was detected in three patients, no lactate was observed in seven patients, and two studies were inconclusive because of severe motion or inhomogeneity artifacts. Best results were obtained with an anterior/posterior neck coil at a 288 ms echo time (TE).

Strategy for lipid suppression in lactate imaging using STIR-DQCT: a study of hypoxic-ischemic brain injury

In vivo lactate detection using gradient enhanced double quantum coherence transfer (DQCT) was significantly improved by addition of short-time-inversion-recovery (STIR). Phantom studies demonstrated lipid suppression down to the background noise level with 33% loss of lactate signal. In vivo studies using a rabbit model of hypoxic and unilateral-ischemic brain injury showed reduction down to 29 +/- 11% in lipids with inversion times between 140 and 170 ms. Lactate signals on the ischemic side were 51 +/- 53% higher than the nonischemic side at the peak of hypoxia. STIR-DQCT can be a useful robust method of obtaining metabolic maps of lactate in vivo.

In vivo lactate editing with simultaneous detection of choline, creatine, NAA, and lipid singlets at 1.5 T using PRESS excitation with applications to the study of brain and head and neck tumors

Two T2-independent J-difference lactate editing schemes for the PRESS magnetic resonance spectroscopy localization sequence are introduced. The techniques, which allow for simultaneous acquisition of the lactate doublet (1.3 ppm) and edited singlets upfield of and including choline (3.2 ppm), exploit the dependence of the in-phase intensity of the methyl doublet upon the time interval separating two inversion (BASING) pulses applied to its coupling partner after initial excitation. Editing method 1, which allows for echo times TE = n/J (n = 1, 2, 3, . . . . ), alters the BASING carrier frequency for each of two cycles so that, for one cycle, the quartet is inverted, whereas, for the other cycle, the quartet is unaffected. Method 2, which also provides water suppression, allows for editing for TE > 1/J by alternating, between cycles, the time interval separating the inversion pulses. Experimental results were obtained at 1.5 T using a Shinnar Le-Roux-designed maximum phase inversion pulse with a filter transition bandwidth of 55 Hz. Spectra were acquired from phantoms and in vivo from the human brain and neck. In a neck muscle study, the lipid suppression factor, achieved partly through the use of a novel phase regularization algorithm, was measured to be over 10(3). Spectra acquired from a primary brain and a metastatic neck tumor demonstrated the presence of lactate and choline signals consistent with abnormal spectral patterns. The advantages and limitations of the methods are analyzed theoretically and experimentally, and significance of the results is discussed.

Slice-selective fat saturation in MR angiography using spatial-spectral selective prepulses

Presaturation of fat signals by frequency-selective radiofrequency (RF) pulses is often applied in MR angiography to improve the visualization of the blood vessels. Unfortunately, standard fat saturation methods might cause a considerable reduction of the blood signal in the measured slices. This effect is caused by an attenuation of blood magnetization in remote tissue regions with water protons showing a similar Larmor frequency as the fat protons in the recorded slice. The affected blood water protons subsequently flow into the recorded slice and provide low signal intensity. Suitable spatial-spectral selective methods for slice-selective fat saturation were developed to avoid this unwanted effect. A spatial-spectral fat saturation technique was compared with a corresponding only spectrally selective approach. Both saturation techniques were included in a standard two-dimensional (2D) cine sequence and applied in angiographic examinations of the thighs. The results indicate that spatial-spectral saturation (acting slice selectively) leads to a clearly higher blood signal intensity in fat-suppressed MR angiography compared with standard techniques, especially in measurements performed during the systolic phase of the cardiac cycle.

[Proton magnetic resonance spectroscopy: a metabolic approach of cerebral tumors and their follow-up after external radiation therapy]

The same physical principles are the basis of magnetic resonance spectroscopy (MRS) and magnetic resonance imaging (MRI). Proton MRS is easily performed with clinical magnets (> or = 1.5 T) and may be added to routine MRI studies to provide metabolic information on pathological tissues. It represents an important tool to detect several metabolic compounds. The article will review the current status of proton MRS with a particular emphasis upon its clinical utility for the diagnosis of brain tumors and for the evaluation of the efficacy of radiotherapy.

Metabolite-specific NMR spectroscopy in vivo

An outline is presented of metabolite-specific in vivo NMR spectroscopy (particularly in brain). It reviews from a physical spectroscopist's perspective, the need for and the methods of observation of, individual metabolite resonances.

Consistent fat suppression with compensated spectral-spatial pulses

Reliable fat suppression is especially important with fast imaging techniques such as echo-planar (EPI), spiral, and fast spin-echo (FSE) T2-weighted imaging. Spectral-spatial excitation has a number of advantages over spectrally selective presaturation techniques, including better resilience to B0 and B1 inhomogeneity. In this paper, a FSE sequence using a spectral-spatial excitation pulse for superior fat suppression is presented. Previous problems maintaining the CPMG condition are solved using simple methods to accurately program radio-frequency (RF) phase. Next an analysis shows how B0 eddy currents can reduce fat suppression effectiveness for spectral-spatial pulses designed for conventional gradient systems. Three methods to compensate for the degradation are provided. Both the causes of the degradation and the compensation techniques apply equally to gradient-recalled applications using these pulses. These problems do not apply to pulses designed for high-speed gradient systems. The spectral-spatial FSE sequence delivers clinically lower fat signal with better uniformity than spectrally selective pre-saturation techniques.

MRS imaging using anatomically based k-space sampling and extrapolation

A comprehensive strategy for the acquisition, reconstruction, and postprocessing of MR spectroscopic images is presented. The reconstruction algorithm is the most critical component of this strategy. It is assumes that the desired image is spatially bounded, meaning that the desired image contains an object that is surrounded by a background of zeros. The reconstruction algorithm relies on prior knowledge of the background zeros for k-space extrapolation. This algorithm is a good candidate for proton MR spectroscopic image reconstruction because these images are often spatially bounded and prior knowledge of the zeros is easily obtained from a rapidly acquired high resolution conventional MRI. Although the reconstruction algorithm can be used with the standard 3DFT k-space distribution, a distribution that relies on anatomical features that are likely to occur in the spectroscopic image can produce better results. Prior knowledge of these anatomical features is also obtained from a conventional MRI. Finally, the postprocessing component of this strategy is valuable for reducing subcutaneous lipid contamination. Overall, the comprehensive approach presented here produces images that are better resolved than standard approaches without increasing acquisition time or reducing SNR. Examples using NAA data are provided.

A fast method for in vivo lactate imaging

A robust method for fast lactate imaging is presented using a combination of a lactate editing sequence and a one-shot imaging experiment. The lactate editing method is based on manipulation of the phase of the lactate CH3 signal via J-modulation. This is applied as a preparation experiment to the U-FLARE imaging sequence. Phantom experiments are presented in which the quality of water and lipid suppression is established. Edited lactate images with a spatial resolution of 3 microL and total measuring time of 15.8 min are shown. These were obtained from a hemispherical ischemia in the gerbil brain. The images are compared with diffusion-weighted water images.

Parametric multiecho proton spectroscopic imaging: application to the rat brain in vivo

A parametric multiecho variant of proton spectroscopic imaging (SI) is presented using a multiecho SI sequence with uniform phase-encoding of all echoes within each echo train. The acquisition of SI data sets at different echo times (TE) increases the amount of information obtained within the same total measuring time as in standard SI measurements. The gain in information can be used: (a) to choose the most appropriate TE for each metabolite signal with respect to T2, spin coupling, or problems caused by peak overlap; (b) to measure the relaxation time T2 of metabolite signals with high spatial resolution; or (c) to improve the signal-to-noise ratio for metabolite signals with long T2 values by adding spectra calculated from consecutive echoes. The method was tested in vivo on healthy rat brain and applied to study metabolic changes in rat brain lesions.

Incorporation of lactate measurement in multi-spin-echo proton spectroscopic imaging

An improved multi-slice, multi-spin-echo proton spectroscopic imaging method for human brain is presented. The technique allows the reconstruction of lactate images, along with choline plus creatine, N-acetylaspartate, and lipid images from one single data set processed in three separate ways. The discrimination between resonances of lipid protons and lactate methyl protons is based on homonuclear spin-spin coupling. The reliability of the separation of the lipid and lactate contribution depends on the T2 of the lipid resonances. Measurements were performed on a standard 1.5 Tesla clinical scanner on healthy volunteers and a patient with high grade CNS lymphoma, demonstrating the ability to obtain high quality metabolite maps within 11 min.

What might be the impact on neurology of the analysis of brain metabolism by in vivo magnetic resonance spectroscopy?

In vivo nuclear magnetic resonance spectroscopy (MRS) of the human brain is a recently developed technique which allows to assay noninvasively in vivo key molecules of brain metabolism. After a review of the origin of the signals detected by phosphorus and proton MRS of human brain, the impact of MRS on clinical neurology is examined. MRS of the brain does not purport to be a metabolic "biopsy", but unique applications for brain MRS are (1) quantitating the oxidative state of the brain and defining neuronal death, (2) assessing and mapping neuron damage, (3) evaluating membrane alterations, and (4) characterizing encephalopathies. In the near future brain MRS will be performed routinely after conventional MRI, as a valuable metabolic (and functional) complement to the anatomical evaluation of cerebral pathologies, particularly the toxic, metabolic and infectious encephalopathies.