Quantitative lactate‐specific MR imaging and ¹H spectroscopy of skeletal muscle at macroscopic and microscopic resolutions using a zero‐quantum/double‐quantum coherence filter and SLIM/GSLIM localization

Abstract

Quantitative lactate imaging and spectroscopy were performed on phantoms and on electrically stimulated, excised frog skeletal muscle at macroscopic and microscopic resolutions. Lactate selectivity was achieved by use of a zero-quantum/double-quantum coherence (ZQC/DQC) lactate filter, which suppressed all signals besides lactate, including water and lipid, to below noise level. Three-dimensional lactate data sets were acquired in 1-3 h; one of these spatial dimensions was frequency-encoded and the other two were phase-encoded. High-resolution images were reconstructed using the spectral localization by imaging (SLIM) and generalized SLIM (GSLIM) techniques. Lactate quantitation was achieved by employing an external lactate concentration standard and was verified by comparison to quantitative STEAM-localized and nonlocalized spectra that used total creatine as an internal concentration reference. Additionally, quantitatively accurate behavior of the SLIM and GSLIM techniques as applied to data sets of low signal-to-noise ratio and to macroscopically heterogeneous objects was verified using simulations and real muscle lactate data sets with known heterogeneity.

Lactate quantitation in a gerbil brain stroke model by GSLIM of multiple-quantum-filtered signals

“…Also, due to the absence of interhemispheric blood flow in the Mongolian gerbil, the demarcation of lactate signal between the ischemic and normal hemispheres is expected, and is not a GSLIM artifact. Prior computer simulations (13), as well as the results shown in Fig. 4, which depicts high lactate in the ischemic hemisphere despite defining the whole brain as a single compartment, verify that GSLIM can accurately disclose inhomogeneity within compartments.…”

“…The accuracy of these non-Fourier techniques has been verified previously (11)(12)(13). Although SLIM and GSLIM have been employed earlier in 31 P studies of excised human uterus (21), in 1 H studies of lactate in isolated frog skeletal muscle (13), and in studies of diffusion of water and metabolites (22), the work presented here is the first full description of an application of either SLIM or GSLIM in vivo. Preliminary investigations in vivo included 31 P studies of rat (23) and human (24) brain.…”

“…Both of these post-acquisition spectral localization methods employ a priori spatial information obtained from a high-resolution water image to constrain the spectral function from the low-resolution lactate data set to enhance resolution at compartment boundaries. Detailed discussions of the theory and practical issues pertaining to SLIM and GSLIM can be found in previous work (10)(11)(12)(13).…”

“…This protocol has previously been verified with simulations and experiments on phantoms and has been applied to studies of electrically stimulated isolated skeletal muscle (13). However, successful application of the method has not previously been shown in vivo.…”

“…An 8 ϫ 8 ϫ 8 DQC/ZQC-filtered lactate image was then acquired in 75 minutes beginning 100 minutes after ligation, with TR 2.2 seconds and identical FOV, using the pulse sequence presented in Fig. 1 and described in detail elsewhere (13,19). This sequence has previously been shown to suppress completely all signals besides lactate (13), and this behavior was again verified using a phantom of geometry similar to a gerbil head and containing a 10 mM lactate solution and vegetable oil.…”