| Sign In to gain access to subscriptions and/or personal tools. |
Imaging Biochemistry Noninvasively: Magnetic Resonance Spectroscopy in Liver DiseaseFrom the Departments of Anesthesiology and Radiology, University of Colorado at Denver and Health Sciences Center, Anschutz Medical Center, Aurora, CO. Address correspondence to: Natalie J. Serkova, PhD, University of Colorado Denver, Anschutz Medical Center, 12631 East 17th Pl, AO1, Suite L15-2011, Aurora, CO 80045; e-mail: Natalie.serkova{at}ucdenver.edu.
In the recent issue of the Journal of Parenteral and Enteral Nutrition, Dr Woodward and coauthors report on the quantitative changes in endogenous metabolites, for example, hepatic lipids and phospholipid precursors, in the liver of patients with intestinal failure–associated liver disease (IFALD) using a noninvasive magnetic resonance spectroscopy (MRS) approach.1 IFALD patients receiving parenteral nutrition for longer than 1 year, as well as their body mass index–matching healthy volunteers, underwent MRS scans (using a 1.5 T clinical MR scanner) for quantitative assessment of the biochemical changes in the liver due to disease severity. In vivo proton 1H-MRS was applied to integrate the MR peaks of total CH, CH2, and CH3 lipids, as well as choline and water; the ratios of water:lipid and choline:lipid were also calculated. Moreover, in vivo phosphorous 31P-MRS was used to assess the peak integrals of phosphomonoesters (PME, precursors for hepatic phospholipids) and phosphodiesters (PDE, catabolic products for hepatic phospholipids). This noninvasive approach demonstrates a significant clinical potential for MRS evaluations of hepatic metabolism in IFALD patients for early prediction of disease severity and for following up on treatment efficacy without requirement of invasive, biopsy-based protocols. In the past 2 decades, in vivo MR methods—which include both magnetic resonance imaging (MRI) and MRS—have been introduced into clinical protocols for noninvasive imaging of anatomical, physiologic (MRI), and metabolic (MRS) endpoints. Although MRI is routinely used in clinical practice, MRS is less frequently used in standard-of-care clinical radiologic protocols, mostly due to low sensitivity, large voxel size, prolonged scan times, and often special software and hardware requirements. Despite these experimental limitations, MRS remains an attractive clinical research tool because various metabolites—including neuronal markers, lipids, membrane constituents, osmolytes, and high-energy phosphates—can be measured for the diagnosis of various diseases and therapeutic monitoring in humans. Both techniques, MRI and MRS, are based on the nuclear magnetic resonance (NMR) phenomenon and require the very same clinical MR scanners. NMR was first demonstrated experimentally in 1946, and for its discovery, Felix Bloch (Stanford University, Stanford, CA) and Edward M. Purcell (Harvard University, Cambridge, MA) received the Nobel Prize in Physics in 1956 (http://nobelprize.org/nobel_prizes/physics/). This was followed by the Nobel Prize in Chemistry to Richard Ernst (Swiss Federal Institute of Technology, Switzerland) in 1991 for 2-dimensional Fourier transformation (allowing clinical application of NMR; http://nobelprize.org/nobel_prizes/chemistry/). The first NMR image (MRI) was obtained in 1972 by Paul C. Lauterbur (University of Illinois-Urbana), who shared the Nobel Prize in Physiology or Medicine (http://nobelprize.org/nobel_prizes/medicine/) with Sir Peter Mansfield (University of Nottingham, UK) in 2003 for their discoveries concerning MRI. MR generates images (MRI) or spectra (MRS) by applying an external varying magnetic field to the body. The magnetic field aligns hydrogen atoms parallel and antiparallel to the magnetic field. When a signal in the form of a radio wave pulse is applied to the body using surface coils, atomic distribution between parallel and antiparallel alignment is changed, and after the pulse is gone, the system relaxes to its original status. Hydrogen atoms in different tissue water (MRI) or from different chemicals (MRS) have different relaxation properties that can be detected by radio frequency MR receivers. Characteristics of tissue relaxation after a radio frequency pulse of energy can be then translated into information about the spatial distribution, mobility, and concentration of hydrogen (1H-MRI or 1H-MRS) and, less frequently, other tissue elements (31P-MRS).2 Whereas MRI is focused on hydrogen atoms of tissue water to produce anatomical images, MRS is based on observation of hydrogens in other than water metabolites (such as choline, triglycerides, total lipids in various tissues by 1H-MRS) or phosphorous metabolites by 31P-MRS (membrane phospholipid intermediates in various organs, adenosine triphosphate [ATP], and phosphocreatine in the muscle). Because the same clinical MR scanners are used for MRI and MRS, anatomical imaging is usually carried out before spectral analysis to determine the exact position of the voxel. In 1981, Ross et al3 applied 31P-MRS to a patient with McArdle syndrome, an inborn error of metabolism caused by a lack of glycogen phosphorylase activity in skeletal muscle, and noninvasively showed excessive reduction in muscle phosphocreatine in response to exercise. Since then, 31P-MRS has been extensively used to characterize metabolic state of skeletal muscle, followed by clinical application of 1H- and 31P-MRS in the brain.
More recently, liver biochemistry has been a focus of interest for
noninvasive 1H- and 31P-MRS
studies.4,5
First, the MRS experience, including determination and characterization of
liver fat content, liver tumors, and focal lesions, was rather successful,
which is not surprising taking into account that the liver is one of the most
metabolically active organs, has a large organ size, and is readily
accessible. However, one of the main concerns with abdominal MRS is the
ability to deal with motion artifacts (such as from respiratory movement,
vessel pulsation). This can be reduced by applying respiratory gating, breath
holding, or signal averaging. The diagnostic value of abdominal MRS also
relies on effective water suppression because lipid and choline concentrations
in the liver are various orders lower than that of water. On the other hand,
for quantitative analysis, the full water signal should be available as a
concentration reference. As such, a quantitative MRS study (as in the present
work of Woodward et
al1) will require
two 1H-MRS scans—one with and one without water suppression. In a liver
proton spectrum, 4 major resonances are usually detected, including the
dominant methylene proton signals [C-CH2-)n and (-CH2-CH=CH-)], the methyl
proton signal (CH3-), and the trimethylammonium resonances (made up of betaine
and total choline). Unresolved resonances originate from hepatic myo-inositol,
glycogen, and glucose moieties. Increased lipid resonances (including those of
intrahepatocellular and intramyocellular lipids) play an important role in
noninvasive detection of steatotic liver
disease,6-8
whereas decreases in the lipid signal were observed in chronic hepatitis
patients.9 In vivo
31P-MRS allows for detection of 6 major resonances of PME, PDE, inorganic
phosphate, and Noninvasiveness is a major factor when studying chronic diseases (which will become even more important with the increasing average age of the population). As such, the goal of the present study by Dr Woodward et al1 was to evaluate the potential benefit of noninvasive MRS in the assessment of adults at risk of IFALD of long-term parenteral nutrition. The authors showed that parenteral feeding is associated with a highly increased lipid:water ratio in the liver, which was independent of body mass index of the patient. In addition, the ratio choline:lipid was significantly decreased in patients with parenteral feeding. Although no significant differences were found between patients and the control subjects in 31P-MRS, the PME:PDE ratio of >0.3 was predictive for 2 patients at most risk of advanced liver disease. The increased PME:PDE ratios have been previously reported as a marker for advanced liver disease and cirrhosis.2,5,10 In summary, in vivo MRS provides a noninvasive means of studying body biochemistry, both for early disease diagnosis and for monitoring the biochemical response to treatment. MRS can be performed as an "add-on" sequence at the end of most clinical MRI examinations in the liver. As with all other existing modalities, MRS has its advantages and disadvantages. Compared to MRI, the sensitivity and spatial resolution of MRS are limiting factors. Compared to other metabolic imaging modalities—positron emission tomography and single-photon emission computed tomography, both of which detect metabolic tracers in micromolar concentrations—MRS is less sensitive and requires tissue concentrations in the millimolar range. However, one of the major advantages of MRS is its ability to simultaneously detect various endogenous tissue metabolites without requirement of injection of radioactive tracers. Some of the MRS limitations can be partly overcome by future clinical use of stronger magnetic fields (3 T scanners) because a concomitant increase in signal intensity and spatial and spectral resolution of metabolites can be achieved. Further work is require to provide reliable clinical MRS protocols to ascertain the "baseline" level of endogenous metabolites in "normal" liver and to metabolically characterize various liver abnormalities. After that, abdominal MRS may become a valuable clinical tool that can answer questions beyond the anatomical diagnosis of liver lesions.
Financial disclosure: none declared.
Journal of Parenteral and Enteral Nutrition, Vol. 33, No. 6,
726-728 (2009)
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
|
 |
|
Sign In to gain access to subscriptions and/or personal tools.
-, 
-, and β-ATP. Various clinical studies on
patients with cirrhosis and hepatic malignancies consistently demonstrate
elevated PME levels, which can be explained by increased membrane synthesis
and
turnover.
Top