(Circulation. 1996;93:1464-1470.)
© 1996 American Heart Association, Inc.
Articles |
1H NMR Spectroscopic Imaging of Myocardial Triglycerides in Excised Dog Hearts Subjected to 24 Hours of Coronary Occlusion
From the University of Alabama at Birmingham, Department of Medicine, Division of Cardiovascular Disease; the Department of Pharmacology (D.D.K.); and the Department of Pathology (J.B.C.).
Correspondence to Gerald M. Pohost, MD, Department of Medicine, Division of Cardiovascular Disease, 1900 University Blvd, THT 311, Birmingham, AL 35294-0006.
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
Background Myocardial ischemic insult causes depression of fatty-acid ß-oxidation and increased fatty-acid esterification with triglyceride (TG) accumulation. This accumulation has been demonstrated to occur in the territory with diminished blood flow surrounding an infarct, ie, the region at risk. To evaluate whether the extent of TG accumulation in the canine heart after 24 hours of ischemia could be detected, we applied myocardial 1H nuclear magnetic resonance (NMR) spectroscopic imaging (SI).
Methods and Results Seven adult mongrel dogs underwent 24 hours of left anterior descending coronary artery occlusion. Postmortem, the hearts were excised and the size and location of the infarct were determined. With a Philips 1.5-T clinical NMR imaging/spectroscopic system, two-dimensional (2D) 1H NMR SI was performed. TG 1H NMR chemical shift images were reconstructed from the frequency domain spectra by numerical integration. A statistically significant (P<.05) increase in TG signal intensity was demonstrated in the region at risk compared with the nonischemic control region. There was an intermediate quantity of TG in the infarct region. Biochemical determination of tissue TG content (milligrams per gram wet weight) in the control, at-risk, and infarct regions confirmed the 1H NMR measurements. Histological evaluation with oil red O staining also demonstrated graded TG accumulation in myocytes. The highest TG levels were found in the at-risk region and the lowest levels in the control region.
Conclusions By use of 2D 1H NMR SI, the present study confirms and extends previous work that demonstrates preferential accumulation of TG in the reversibly injured myocardium after 24 hours of coronary occlusion. This study provides an important step toward the clinical application of TG imaging. When TG imaging is ultimately possible, resultant data would have diagnostic, prognostic, and therapeutic implications.
Key Words: lipids • metabolism • myocardium • ischemia • magnetic resonance imaging
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Under aerobic conditions, myocardial contractile function is supported mainly by ß-oxidation of fatty acids.1 Reduction in regional myocardial blood flow results in well-documented changes in the enzymatic activities2 3 and flux characteristics4 5 6 of the myocardial ß-oxidation pathway. These ischemia-induced changes in myocardial fatty-acid metabolism can be divided into two groups. First, there are short-term (ie, <60 minutes) changes occurring in the central zone of infarction.3 7 Drastic reduction of blood flow in the central infarct region results in myofibrillar and mitochondrial disruption with short periods of ischemia.3 8 Other potentially harmful effects include acylcarnitine and lysophospholipid accumulation9 10 that can lead to alterations in ion transport,11 permeability,12 and/or membrane-associated intracellular signaling pathways.13 Such changes have been hypothesized to precipitate or exacerbate the course of ischemia-induced myocyte death.9 14
Surrounding this central infarct zone is a second type of ischemic myocardium considered to be region at risk. This reversibly injured myocardium experiences less severely reduced levels of blood flow than the central infarct zone, and although flow is reduced, morphological integrity is maintained. However, morphological analysis of the region at risk demonstrates the occurrence of a second type of alteration in fatty-acid metabolism. This alteration is characterized by accumulation of significant quantities of oil red O–staining lipid droplets in intact but at-risk myocytes, which does not occur in infarcted or in nonischemic (control) myocardium.15 16 These droplets contain essentially pure TGs17 and result from increased esterification due to reduced ß-oxidation. Prolonged myocardial TG accumulation requires a continuous supply of both exogenous fatty acids and ATP for the generation of acyl-CoA precursors of TGs18 and a continued supply of glycolysis-derived glycerol-3-phosphate for the glycerolipid backbone.19 20 Thus, it would appear that active but altered fatty-acid metabolism is occurring in the region at risk compared with the infarcted or the control myocardium. Identification of this region at risk in patients with myocardial ischemic insult would be of potential clinical importance because this region will most likely have reduced wall motion and slower recovery after the insult. Armed with such knowledge, we could differentiate the true infarct size from ischemic but viable myocardium. A clinically applicable method to detect the extent of this hypofunctional region at risk would have great diagnostic and prognostic value and could provide a means to track the effects of therapy.
We have been interested in the development of an NMR-based method for characterization of this region of insulted but viable myocardium. A substantial increase in 1H NMR–visible myocardial lipid content relative to control induced by short-term21 or long-term22 coronary occlusion has been observed in myocardial samples analyzed by ex vivo high-resolution NMR spectroscopy. However, delineation of intramyocardial lipid is technically challenging because of heart and respiratory motion and adjacent epicardial fat. Although 1H NMR SI methods like the Dixon method23 or single-volume spectroscopy24 can detect alterations in myocardial lipid metabolism, these approaches are not optimal for clinical applications. Therefore, we have extended previous studies by applying, for the first time, an NMR SI method25 to noninvasively evaluate myocardial TG accumulation in the intact heart. The aim of the present study was to apply this method to delineate the extent of TG accumulation in a canine ischemic insult model known to be associated with TG accumulation, ie, after 24 hours of coronary occlusion. By allowing direct visualization of myocardial TG, this new NMR-based imaging application appears to provide one means for differentiating viable region at risk, nonviable infarct, and control myocardium.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Canine Model of Infarction Induced by LAD Occlusion
All of the experimental protocols described were conducted in accordance with the guidelines of the American Physiological Society and the University of Alabama at Birmingham Institutional Animal Care and Use Committee.
Seven mongrel dogs (15 to 25 kg) were fasted overnight before the study and premedicated with acepromazine (3 mg IM) and atropine (0.04 mg/kg IM). Anesthesia was induced with sodium pentothal (300 to 500 mg IV to effect) and, after endotracheal intubation, was maintained by mechanical ventilation with isoflurane (0.5% to 1.5%) and oxygen (2 L/min). A lead II ECG was monitored, and an infusion of 0.9% saline was administered throughout the procedure. A left lateral thoracotomy was performed in the fifth intercostal space under aseptic conditions. The pericardium was opened, the LAD was isolated below its first diagonal branch, and the blood flow to the region of the heart perfused by the LAD was interrupted by use of a silk ligature. The thoracotomy was then closed, and the dog was allowed to recover under observation. Analgesics were given as needed for the next 24 hours (5 mL bupivacaine [0.5%] through a chest tube; morphine 0.25 to 1.0 mg/kg IV). Twenty-four hours after coronary ligation, blood was collected in four of the seven dogs for determination of serum TG levels, and the dogs were killed by KCl infusion. Hearts were removed and evaluated as follows.
Dual-Perfusion Technique Procedure
Region at risk and extent of irreversible damage were determined
by a previously described dual-perfusion technique.26
Briefly, a cannula was inserted into the aorta and secured with a
suture. An 18-gauge catheter was inserted immediately distal to the
point of LAD ligation and fixed in place. Then, the aorta was
retrograde-perfused with a solution of 0.25% Evans blue dye, and
the LAD was perfused with 1.5% TTC in 20 mmol/L potassium phosphate
buffer (pH 7.4). The solutions were maintained at a temperature of
37°C and infused simultaneously for 5 minutes at a
constant pressure of 100 mm Hg. After the staining process, fatty
tissue was carefully removed from the left ventricular
epicardium, and the heart was mounted in a glass beaker for SI in a
1.5-T whole-body Philips Gyroscan.
2D 1H NMR SI of the Ex Vivo Canine Heart
The beaker containing the heart was placed in a home-built
11-cm-diameter two-turn vertical Helmholtz coil tuned to 64
MHz. The heart was positioned in the 1.5-T MR system, and multislice 2D
Fourier-transformed scout images were obtained in horizontal and
vertical planes with a TE of 30 ms and a TR of 400 ms to verify the
position of the heart in the center of the radiofrequency coil and the
isocenter of the magnet. Volume-selective shimming was applied to
encompass a volume of interest (60x240x240 mm) through the heart. The
water resonance was suppressed by application of two chemically
selective inversion pulses at the water frequency. The first inversion
pulse was applied at 1100 ms before the start of the NMR SI sequence,
the second at 190 ms. The inversion times were adjusted to achieve
optimal water suppression. The 2D 1H NMR SI was performed
over a 10-mm-thick short-axis slice through the left ventricle.
The long-axis 2D Fourier-transformed scout images were used to
ensure reasonable alignment of acquired NMR images with the subsequent
myocardial slices.
For those acquisitions, a TR of 2000 ms and a TE of 46 ms were used. The field of view for the SI examination was 100 mm, with 32x32 gradient-phase encoding steps and one measurement per profile. Typically, data were collected from three or four slices interrogating the LAD perfusion bed, ie, myocardium subjected to 24 hours of occlusion. The SI data sets were windowed with a sine-bell function and Fourier transformed in k-space. The resulting time-domain signals were numerically fitted to a sum of exponentially decaying sinusoids by use of a noniterative time-domain fitting technique.27 Components within a frequency window of 60 Hz around the water signal were numerically removed to reduce the residual water signal. Residual magnetic field homogeneity was corrected in the time domain with a reference data set obtained without water suppression.28 29 A gaussian window was applied in the time domain and Fourier transformed to obtain frequency-domain spectra. Lipid (TG) images were reconstructed from the frequency-domain spectra by numerical integration.
This 1H NMR SI approach provides a means to depict the
level and the distribution of myocardial TGs in normal,
ischemic but reversibly insulted, and infarcted
myocardium. To establish the relation between NMR
SI–derived TG images and histological and biochemical
assessment of myocardium subjected to a 24-hour
ischemic insult, lipid and water 2D image files were
transferred to a Macintosh IIci computer for image processing and
analysis using Image version 1.49 (Wayne Rasband, NIMH). The 2D
image files were organized as a three-dimensional array called a
stack, with the water image being the first slice of the stack, the
lipid image being the second slice, and the x,
y coordinates being the same between the two slices.
Regional mean intensity analysis was performed by placing the
transparent acetate sheets (described below, and Fig 1d
) on the
computer screen near the water images and redrawing the irregular three
regions of interest that relate to the control, infarct, and
at-risk regions on the transparent acetate sheets, thus being
blinded to the lipid distribution as defined by the lipid image. The
stack was then switched to the second (lipid-image) slice, with the
predrawn regions remaining on the screen overlying the lipid image.
Measurements were taken from these images as mean intensity values of
the pixels within the control, infarct, and at-risk regions.
Background regions were also drawn and measurements taken for
background subtraction of any outside noise or interference.
|
Determination of Infarct Size
At the conclusion of 2D 1H NMR SI data acquisition,
all hearts were cut into four to six 1-cm-thick slices in a
bread-loaf fashion from apex to base. Both the basal and the apical
surfaces of these transverse sections were traced onto transparent
acetate sheets for determination of infarct size. The control region
was stained with Evans blue dye; the region at risk was identified by
the absence of Evans blue dye but presence of TTC stain (red formazan
precipitate, TTC positive), and the irreversibly injured
myocardium was indicated by the absence of TTC stain (TTC
negative).30 31 The respective regions corresponding to
the entire left ventricle, the region at risk, and infarcted
myocardium were computed with an Olympus Cue-2 image
analysis system (C. Squared Corp) followed by measurement of
the weight of the respective regions. Previous studies demonstrated
that infarct size derived by the planimetric method correlates well
with the direct gravimetric measurement of infarct size.32
The infarct region was expressed as a percentage of the area at risk
and as a percentage of the total left ventricular
weight.
Tissue TG Content Analysis
Frozen sections of control myocardium (posterior),
infarct area, and region at risk were thawed, patted dry, and weighed.
Each tissue sample was then finely minced and suspended in 2 mL
ice-cold 0.1 mol/L KCl buffered with 0.02 mol/L MOPS (pH 7.5).
Total lipid extraction was effected by adding 3 mL
chloroform:methanol (2:1)33 to each sample, stirring
for 4 to 6 hours at 4°C, and then centrifuging the extracts at
2000g for 30 minutes at 4°C. The organic phase was saved,
the samples were reextracted, and the two organic phases were combined.
The combined organic phases were then dried under nitrogen, and the
oily residue was suspended in 1 mL 0.1N tetraethyl ammonium hydroxide
in ethanol. TGs were selectively hydrolyzed by incubation at 60°C for
2 hours.34 At the end of the hydrolysis, 1.2 mL 0.1N HCl
was added, and the nonsaponified lipids and TTC were separated from
free glycerol by extraction with 2 mL chloroform:methanol. After
separation, aliquots of the aqueous phase were mixed with 0.2 mol/L
K+ HEPES (pH 7.2), and the samples were placed under vacuum
for 6 to 8 hours at room temperature to remove residual organic
solvent.
Tissue glycerol content derived from TG saponification was measured by
a glycerokinase-based assay with p-iodonitrotetrazolium
violet as indicator (Sigma Chemical Co). Values tabulated as milligrams
of TGs were derived from a standard curve after saponification and
extraction of a known quantity of triolein. All data reported
represent the average of duplicate measurements of myocardial
samples of the control, at-risk, and infarct regions of six dogs.
Transmural distribution of TTC/Evans blue was used to define the
viability state of the myocardium (these data are plotted
in Fig 3). If TTC staining was predominant, the segment was considered
TTC positive (region at risk). If TTC was predominantly absent, the
segment was considered nonviable (infarct). Control samples were
defined by location (posterior wall) and by Evans blue staining.
|
Histological Evaluation
Myocardial sections from the three regions of interest were
sampled and fixed in 2% phosphate-buffered (pH
7.4)
glutaraldehyde. Each myocardial sample was then washed
with 0.1 mol/L phosphate buffer and further fixed in
phosphate-buffered formalin overnight. The samples were then
incubated, stirred, and mixed with sucrose for 6 hours before freezing
with OCT embedding medium for frozen tissue specimens. Frozen sections
5 to 7 µm thick were then prepared and stained with oil red O to
visualize the oil droplet deposits.35
Statistical Analysis
All data are expressed as mean±SEM. Differences were considered
significant if P<.05. Changes in tissue TG levels and
spectroscopic methylene resonance intensities among the control,
at-risk, and infarct regions within each dog were first
analyzed by ANOVA, followed by Student's unpaired t
test.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
2D 1H NMR SI
By use of TTC and Evans blue staining, the control, at-risk (TTC-positive, dotted area), and infarct (TTC-negative, striated area) regions of the myocardium (Fig 1d
) were
identified. The infarct size in percentage of the left ventricle was
22.3±0.08%, the infarct size in percentage of the region at risk was
48±0.18%, and the size of the region at risk in percentage of the
left ventricle was 47±0.18%. These findings are
consistent with our published reports with comparable
hemodynamic changes.
To correlate the development of myocardial ischemic injury and
lipid accumulation, myocardial 1H NMR spectra from these
dogs subjected to 24 hours of coronary occlusion were obtained
(Fig 1). Myocardial water "scout" images (Fig 1b) were shown to
correlate with "lipid" images (Fig 1c) derived from all regions
of the myocardium.
Myocardial 1H NMR spectra were selected (Fig 1a) to
demonstrate tissue TG levels for control, infarct, and at-risk
regions (Fig 1d). The numbers of the spectra correspond to the numbers
shown in the panels (Fig 1a through 1c). Spectra 1 and 2 (Fig 1a) were
obtained from the ventricular cavity (Fig 1b and 1c) and
demonstrate substantially reduced methylene resonances. Spectra 3 and 4
(Fig 1a) were obtained from the control region (Fig 1b and 1c) and
contain methylene and creatine peaks of equivalent sizes, whereas
spectra 5 and 6 (Fig 1a) from the infarcted region (Fig 1b and 1c) show
methylene peaks that are slightly higher (Fig 1a) than those from the
control region. Finally, spectra 7 and 8 (Fig 1a) were obtained from
the ischemically insulted but not infarcted region at risk (Fig 1b and 1c) and contain markedly elevated methylene peaks.
The region at risk (dotted area, Fig 1d) also demonstrates increased
intensity in the lipid images (Fig 1c). These
ischemia-related myocardial lipids are substantially less
bright than those of epicardial fat, as demonstrated by a bright signal
from epicardial fat around the region of the right ventricle (9 to 1
o'clock positions, Fig 1c). Epicardial lipid is also observed at the 4
o'clock position (Fig 1c). The control myocardium exhibits
low image intensity (1 to 3 o'clock positions; Fig 1c).
Fig 2 summarizes the mean density of the images of the
methylene resonances in the infarct, risk, and control
myocardium in the seven hearts studied. This
analysis shows that a statistically significant increase in the
1H NMR lipid resonance intensity was measured between the
at-risk (TTC-positive) and control myocardium
(P<.01) (Fig 2). There were no such statistically
significant differences between the at-risk and the infarct regions
(TTC negative) (P<.11) or between the infarct and control
regions (P<.35) (Fig 2).
|
Biochemical Analysis of TG Content Correlates With
1H NMR SI TG Image
Analytical quantification of the TG content of the combined
endocardial and epicardial sections from control, infarcted
(TTC-negative), and at-risk (TTC-positive) myocardium
gave the following results. Samples from control myocardium
showed low levels of glyceride-containing lipids (Fig 3). Myocardium obtained from the central
regions of the infarct showed higher levels of TG compared with the
control myocardium (P<.016) (Fig 3). Samples
obtained from the region at risk (TTC positive) showed marked increases
in their TG content (P<.019) (Fig 3). This pattern is
similar to that observed with NMR-based approaches, with a
significantly higher level of TGs in the region at risk compared with
the control region.
Histological Evaluation Correlates With
1H NMR SI and Biochemical Analysis
Control myocardial tissue showed normal morphology as evaluated by
light microscopy with intact, nonswollen myocytes and a low level of
oil red O–staining, TG-containing droplets (Fig 4A). In
contrast, myocardial samples from the central zone of infarction showed
numerous swollen, ruptured myocytes (Fig 4B). The myofibrillar arrays
in these cells were in the process of dissolution. In addition, this
region contained significant numbers of neutrophils, as anticipated in
response to prolonged myocardial ischemia.36
Within this region, it was possible to observe groups of myocytes
retaining some of their structural integrity and containing a
significant number of oil red O–staining droplets (Fig 4B). Finally,
the region at risk showed a unique morphological pattern (Fig 4C). As
expected, myocyte morphology by light microscopy was normal. These
cells were not swollen, and the sarcolemmal and mitochondrial membranes
were intact. Little nuclear clearing was observed. However, oil red O
staining showed that these cells contained large numbers of TG
droplets, substantially greater than the infarcted and control regions
(Fig 4C).
|
Results of Serum TG Analyses
The serum TG level was evaluated in four of the seven dogs
immediately before they were killed. The mean level was 28±8 mg/dL,
which is consistent with the results found in normal dogs but
substantially lower than levels found in normal humans.
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Various methods have been used to demonstrate abnormal fatty-acid metabolism in ischemic myocardium. For example, positron emission tomography has demonstrated abnormal myocardial extraction and utilization of radiolabeled long-chain fatty acids in at-risk myocardium.6 Specifically, such studies have shown that 11C-palmitate is retained in ischemic but viable myocardium.4 37 Similarly, single-photon radionuclide methods have shown abnormalities of fatty-acid metabolism in the region at risk using 125I-labeled long-chain fatty acids.38 These studies and the analytic techniques of others5 7 indicate that abnormalities of fatty-acid retention (ie, enhanced esterification and/or slowed oxidation) characterize ischemically insulted but viable myocardium. Prior studies from our laboratory and by others indicate the potential utility of 1H-NMR spectroscopic methods for the study of changes in fatty-acid metabolism associated with ischemic myocardial insults, particularly changes in fatty-acid esterification.21 22 39 40 Both high-resolution21 and the Dixon23 1H NMR methods have demonstrated increased levels of lipids in region-at-risk myocardium. Based on chemical measurements and in vitro results, these increases in NMR signal intensity can be attributed to intracellular TGs.22 However, neither of the above-mentioned approaches allows accurate, quantitative measurements of changes in the lipid content in affected myocardium.
Usefulness of the NMR Approach
The present study applied a new 1H NMR SI approach
that offers several advantages over previous approaches. This approach
provides the intermediate step between previous results and in vivo
application. It offers high spatial resolution, spectroscopic
specificity, the ability to quantify, and the capacity to image the
distribution of myocardial lipids. In combination with standard NMR SI
techniques, it allows evaluation of the relation between cardiac
structure and changing lipid content. Previous approaches were limited
in the number of 1H NMR spectra obtained per heart due to
time constraints. The present approach allows collection and
analysis of myocardial slices, each containing 1024 spectra, in
a comparable total measurement time. This high number of acquired
spectra allows simultaneous creation of both a lipid image
and a water image of an entire myocardial slice. The resulting NMR SI
examination is more comprehensive and allows depiction of myocardial TG
distribution. In addition, our approach demonstrated both creatine and
choline signals, allowing the exact assignment of chemical shifts to
resonances, thereby verifying and identifying the signal generating the
spectroscopic image. Differences in lipid resonances within the same
slice could be quantified because of this high degree of
specificity.
This new approach was applied to evaluate TG distribution in control,
at-risk, and infarct zones. The lipid images revealed high
background signal intensities from the right ventricle and adipose
tissue on the epicardial surface of the heart (Fig 1c). This high lipid
signal from the right ventricle has been observed in previous
studies.41 Despite these high signal intensities, adequate
spectroscopic data from the left ventricular
myocardium could be obtained.
Validity of the NMR Approach
It was found that after a 24-hour ischemic insult,
1H NMR SI–detected myocardial TGs are highest in the area
at risk, intermediate in the infarct, and lowest in the control
myocardium. These results are readily apparent on the lipid
images (Fig 1c) and quantification (Fig 2). TGs are significantly
higher in the TTC-positive region at risk than in the control region
(Fig 2). The 1H NMR SI–determined TG content is higher in
the TTC-negative (infarct) region than in the control region, but this
difference does not quite achieve statistical significance by NMR
criteria (Fig 2). As anticipated, variability in NMR SI data is greater
than that of analytically determined data (Fig 3), accounting in large
part for the differences in statistical significance between
NMR-derived and biochemically derived data.
The biochemical and histological analyses (Figs 3 and 4) clearly demonstrate a pattern of response similar to that
observed with 1H NMR SI approaches. The region at risk
contained substantial increases in both the number of oil red
O–staining droplets and TG content (Fig 4), consistent with
previous reports.15 16 However, the infarct zone contained
a level of TG that was between the low level in the control region and
the high level in the region at risk, a finding that is not
consistent with the results of other
investigators.15 16 These observations can be explained as
follows. When flow is reduced and a myocardial region becomes
ischemic, ß-oxidation and fatty-acid transport into
mitochondria becomes impaired. With collateral-derived residual
flow, glycolysis continues to generate ATP, allowing the
energy-dependent process of fatty-acid esterification and
resultant TG generation to occur. As the transition from viable to
nonviable myocardium occurs, esterification slows and then
ceases, and TGs are broken down as the infarct region becomes necrotic.
Thus, increased TG is detectable even in the nonviable infarct region
but to a level substantially lower than that observed in the viable
region at risk.
Previous work from this laboratory has demonstrated that myocardial TGs are visible by 1H NMR.22 40 42 43 The results reported in this article extend this work by indicating that abnormal fatty-acid metabolism in ischemic myocardium may be amenable to investigation by the current and more sophisticated 1H NMR SI approaches. In combination with 31P NMR SI, it may be possible to acquire significant new information to characterize the state of the insulted myocardium. Since active lipolysis and oxidation40 42 of these risk-zone TGs should occur after reestablishment of normal flow, the combination of both 31P and 1H NMR SI could be useful in a clinical approach for evaluating the reversibility of underperfused myocardium associated with a myocardial infarction. This assumes that it will ultimately be possible to clinically image myocardial TGs by use of 1H NMR SI.
TG Accumulation and Viability
The important pathophysiological findings
in our study are suggested by the fact that active accumulation of TG
has been demonstrated in intact region-at-risk
myocardium. These findings suggest that this territory is
reversibly injured and potentially could recover function after
reperfusion. Conversion of myocardial intracellular fatty acids to
fatty acyl-CoA, the substrate for esterification, requires a continuous
supply of ATP.18 Because fatty-acid activation
involves ATP hydrolysis,18 it is clear that
region-at-risk myocardium must maintain both high
rates of metabolic energy production and membrane
integrity. Also, the glycerol-3-phosphate required for continued
myocardial synthesis of TGs arises solely from
glyceraldehyde-3-phosphate derived from ongoing glucose
metabolism.19 20 44 These conclusions support
previous studies demonstrating that active but altered
metabolism occurs in the region at risk consistent
with viable myocardium. If such altered
metabolism could be clinically detected, it would provide a
basis for the delineation of viable myocardium in patients
who have sustained a myocardial ischemic insult.
Clinical Implications
The ability to observe the accumulation of TGs in
myocardium in patients with ischemic heart disease
could have several clinical implications for diagnosis,
prognostication, and evaluation of the impact of therapy. For
diagnosis, TG accumulation in a potentially ischemic territory
suggests moderate to severe ischemic insult. In the present
study, TG accumulation was minimal in the nonischemic area,
mild in the infarcted territories, but substantial in the region at
risk surrounding the infarct. This region at risk was most likely
moderately to severely ischemic during the course of the
myocardial injury. For prognostication, myocardial territory with
substantial TG accumulation suggests viability,14 and the
more extensive this territory, the more extensive the injured but
viable myocardium. Thus, even if wall motion abnormality is
extensive, the presence of substantial TG should be an indication of
reversible injury. Accordingly, we would anticipate that in patients
with reduced ejection fraction after myocardial infarction, those with
substantial TG accumulation in the infarct territory would have a
better prognosis than those with little TG accumulation. Finally, TG
accumulation also should be useful for the assessment of therapy in
dysfunctional myocardium. Those zones in which TGs
accumulate substantially should have a reversible functional deficit.
It could also be anticipated that TGs decrease in parallel with
improvement in function. Accordingly, the time course with which TG
content decreases in serial spectroscopic images would provide an
indication of the beneficial effects of therapy and predict the
potential for functional improvement.
Approaches to Distinguish Between Epicardial and Myocardial
TGs
The ability to apply TGs clinically depends on overcoming a number
of technical impediments. As anticipated, epicardial fat can be readily
observed by clinical 1H NMR measurements.45
Also, it has been possible to visualize lipids within the
interventricular septum.45 However, it has
not yet been possible to distinguish between epicardial and
intramyocardial lipids in the left ventricular free wall.
Nevertheless, myocardial creatine and trimethylamines have been
observed in both left ventricular free wall and
interventricular septum by use of 1H NMR SI
methods. The locations of the peaks representing creatine
and trimethylamine molecules are quite distinct in their NMR spectral
position, and therefore, there is no contamination from epicardial fat.
It is anticipated that with further technical and methodological
improvements, it will be possible to distinguish between free-wall
myocardial TG and epicardial fat. In particular, advances in turbo
spin-echo imaging methods, fast spin-echoes for SI, and if
needed, higher magnetic fields should lead to the improvements
necessary to visualize intramyocardial TGs clinically. Accordingly,
although the present study suggests the potential for the clinical
use of 1H NMR spectroscopic methods to image myocardial TGs
associated with an ischemic insult, several problems, including
the contamination of free-wall myocardial lipids with epicardial
fat in 1H NMR SI, await resolution. Also, the study does
not evaluate the time-dependent changes in TGs associated with and
those that occur after the myocardial ischemic insult. These
issues will be addressed in future work.
|
Selected Abbreviations and Acronyms |
|---|
|
|
Acknowledgments |
|---|
This work was supported in part by NIH HD SCOR grant 2P50-HL-17667-19. The authors gratefully acknowledge the invaluable contributions of Roger A. Orr, CNMT, BS, in the areas of computation and data and image analysis. We are saddened by the recent untimely death of Roger A. Orr. In addition, we acknowledge the expert surgical assistance of Charlye A. Brocks.
Received June 1, 1995; revision received October 26, 1995; accepted October 29, 1995.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
1. Neely JR, Rovetto MJ, Oram JF. Myocardial utilization of carbohydrate and lipids. Prog Cardiovasc Dis. 1972;15:289-329. [Medline] [Order article via Infotrieve]
2.
Pauly DF, Yoon SB, McMillin JB.
Carnitine-acylcarnitine translocase in ischemia: evidence
for sulfhydryl modification. Am J Physiol. 1987;253:H1557-H1565.
3. Schwartz A, Wood JM, Allen JC, Bornet EP, Entman ML, Goldstein MA, Sordahl LA, Suzuki M. Biochemical and morphological correlates of cardiac ischemia membrane systems. Am J Cardiol. 1973;32:46-61. [Medline] [Order article via Infotrieve]
4. Schwaiger M, Schelbert HR, Ellison D, Hansen H, Yeatman L, Vinten-Johansen J, Selin C, Bario J, Phelps ME. Retention and clearance of 11C-palmitate in ischemic and reperfused canine myocardium. J Am Coll Cardiol. 1985;6:336-347. [Abstract]
5. Opie LH, Owen P, Riemersma RA. Relative rates of oxidation of glucose and free fatty acids by ischemic and non-ischemic myocardium after coronary artery ligation in the dog. Eur J Clin Invest. 1973;3:419-435. [Medline] [Order article via Infotrieve]
6.
Rosamund TL, Abendschein DR, Sobel BE, Bergmann SR,
Fox KA. Metabolic fate of radiolabeled palmitate in
ischemic canine myocardium: implications for
positron emission tomography. J Nucl Med. 1987;28:1322-1329.
7.
Liedtke AJ, DeMaison L, Eggleston AM, Cohen LM, Nellis
SH. Changes in substrate metabolism and effects of
excess free fatty acids in reperfused myocardium.
Circ Res. 1988;62:535-542.
8. Jennings RB, Raum JH, Herdson PB. Fine structural changes in myocardial ischemic injury. Arch Pathol. 1965;79:135-143. [Medline] [Order article via Infotrieve]
9.
Corr PB, Gross RW, Sobel BE. Amphipathic
metabolites and membrane dysfunction in ischemic
myocardium. Circ Res. 1984;55:135-154.
10. Corr PB, Snyder DW, Lee BI, Gross RW, Keim CR, Sobel BE. Pathophysiological concentration of lysophosphatides and the slow response. Am J Physiol. 1982;243:H187-H195.
11. Wood JM, Bush B, Pitts BJR, Schwartz A. Inhibition of bovine heart Na+,K+-ATPase by palmitoylcarnitine and palmitoyl-CoA. Biochem Biophys Res Commun. 1977;74:677-684. [Medline] [Order article via Infotrieve]
12. Spedding M, Mir AK. Direct activation of Ca2+ channels by palmitoylcarnitine, a putative endogenous ligand. Br J Pharmacol. 1987;92:457-468. [Medline] [Order article via Infotrieve]
13.
Brockenbrough JS, Korc M. Inhibition of
epidermal growth factor binding in rat pancreatic acini by
palmitoylcarnitine: evidence for Ca2+ and protein kinase C
independent regulation. Cancer Res. 1987;47:1805-1810.
14.
Katz AM, Messineo FC. Lipid-membrane
interactions and the pathogenesis of ischemic damage in the
myocardium. Circ Res. 1981;48:1-16.
15.
Bilheimer DW, Buja LM, Parkey RW, Bonte FJ, Willerson
JT. Fatty acid accumulation and abnormal lipid deposition in
peripheral and border zones of experimental myocardial
infarcts. J Nucl Med. 1978;19:276-283.
16. Wartman WB, Jennings RB, Yokoyama HO. Fatty change of the myocardium in early experimental myocardial infarction. Arch Pathol. 1956;62:318-323.
17. Christiansen K, Jensen PK. Membrane-bound lipid particles from beef heart. Biochim Biophys Acta. 1972;260:449-459. [Medline] [Order article via Infotrieve]
18.
Oram JF, Wenger JI, Neely JR. Regulation of long
chain fatty acid activation in heart muscle. J
Biol Chem. 1975;250:73-78.
19. Robinson J, Newsholme EA. Glycerol kinase activity in rat heart and adipose tissue. Biochem J. 1967;104:2C-4C. [Medline] [Order article via Infotrieve]
20. DeGroot MJM, Willemsen PHM, Coumans WA, van Bilsen M, van der Vusse GJ. Lactate-induced stimulation of myocardial triacylglycerol turnover. Biochim Biophys Acta. 1989;1006:111-115. [Medline] [Order article via Infotrieve]
21. Reeves RC, Evanochko WT, Canby RC, McMillin JB, Pohost GM. Demonstration of increased myocardial lipids with post-ischemic dysfunction (`myocardial stunning') by proton nuclear magnetic resonance spectroscopy. J Am Coll Cardiol. 1989;13:739-744. [Abstract]
22. Bouchard A, Doyle M, Wolkowicz PE, Wilson R, Evanochko WT, Pohost GM. Visualization of altered myocardial lipids by 1H NMR chemical-shift imaging following ischemic insult. Magn Reson Med. 1991;17:379-389. [Medline] [Order article via Infotrieve]
23.
Dixon WT. Simple proton spectroscopic
imaging. Radiology. 1984;153:189-194.
24. Ordidge RJ, Connelly A, Lohman JAB. Image-selected in-vivo spectroscopy (ISIS): a new technique for spatially selective spectroscopy. J Magn Reson. 1986;66:283-294.
25. Luyten PR, Marien AJH, den Hollander JA. Acquisition and quantitation in proton spectroscopy. NMR Biomed. 1991;4:64-69. [Medline] [Order article via Infotrieve]
26.
Romson JL, Hook BG, Rigot VH, Schork MA, Swanson DP,
Lucchesi BR. The effect of ibuprofen on accumulation of
indium-111 labeled platelets and leukocytes in experimental
myocardial infarction. Circulation. 1982;66:1002-1022.
27. DeBeer R, Michels F, van Ormondt D, van Tongeren BPO, Luyten PR, van Vroonhoven H. Reduced lipid contamination in in vivo 1H MRSI using time-domain fitting and neural network classification. Magn Reson Imaging. 1993;11:1019-1026. [Medline] [Order article via Infotrieve]
28. Ordidge RJ, Cresshull ID. The correction of transient B0 field shifts following the application of pulsed gradients by phase correction in the time domain. J Magn Reson. 1986;69:151-155.
29. Den Hollander JA, Oosterwaal LJMJ, van Vroonhoven H, Luyten PR. Elimination of magnetic field distortions in 1H NMR spectroscopic imaging. Proc. 10th Annual Meeting, Society of Magnetic Resonance in Medicine; 1991:472.
30. Vivaldi MT, Kloner RA, Schoen FJ. Triphenyltetrazolium staining of irreversible ischemic injury following coronary artery occlusion in rats. Am J Pathol. 1985;121:522-530. [Abstract]
31. Romaschin AD, Rebeyka I, Wilson GJ, Mickle DAG. Conjugated dienes in ischemic and reperfused myocardium: an in vivo chemical significance of oxygen free radical mediated injury. J Mol Cell Cardiol. 1987;19:289-302. [Medline] [Order article via Infotrieve]
32.
Ku DD, Lucchesi BR. Effects of dimethyl
propranolol (UM-272; SC-27761) on myocardial
ischemic injury in canine heart after temporary
coronary artery occlusion.
Circulation. 1978;57:541-548.
33. Bligh EG, Dyer WJ. A rapid method of total lipid extraction and purification. Can J Biochem Physiol. 1959;37:911-917.
34. Brockerhoff H. Breakdown of phospholipids in mild alkaline hydrolysis. J Lipid Res. 1963;4:96-99. [Abstract]
35. Shmitka TK, Nachlas MM. Histochemical alterations in ischemic heart muscle and early myocardial infarction. Am J Pathol. 1963;42:507-519.
36. Caulfield JB. The cardiovascular system in histochemistry in pathologic diagnoses. In: Spicer SS, ed. Histochemistry in Pathologic Diagnoses. New York, NY: Marcel Dekker; 1986:115-126.
37.
Buxton DB, Mody FV, Krivokapich J, Phelps ME, Schelbert
HR. Quantitative assessment of prolonged metabolic
abnormalities in reperfused canine myocardium.
Circulation. 1992;85:1842-1856.
38.
Miller DD, Gill JB, Livni E, Elmaleh DR, Aretz T,
Boucher CA, Strauss HW. Fatty acid analogue accumulation: a
marker of myocyte viability in ischemic-reperfused
myocardium. Circ Res. 1988;63:681-692.
39. Ugurbil K, Petein M, Maidan R, Michurski S, Cohn JN, From AH. High resolution proton NMR studies of perfused rat hearts. FEBS Lett. 1984;167:73-78. [Medline] [Order article via Infotrieve]
40. Madden MC, van Winkle WB, Kirk K, Pike MM, Pohost GM, Wolkowicz PE. 1H NMR spectroscopy can accurately quantitate the lipolysis and oxidation of cardiac triacylglycerols. Biochim Biophys Acta. 1993;1169:176-192. [Medline] [Order article via Infotrieve]
41. Robbins SL, Cotran RS, Kumar V. Cellular injury and adaptation. In: Robbins SL, Cotran RS, Kumar V, eds. Pathologic Basis of Disease. Philadelphia, Pa: WB Saunders; 1984:1-39.
42. Madden MC, van Winkle WB, Vaughn JM, Pohost GM, Wolkowicz PE. Morphometric analysis demonstrates that metabolically active cardiac triglycerides are 1H NMR visible. J Mol Cell Cardiol. 1993;25:587-597. [Medline] [Order article via Infotrieve]
43. Evanochko WT, Reeves RC, Canby RC, Pohost GM. Proton NMR spectroscopy in myocardial ischemic insult. Magn Reson Med. 1987;5:23-31. [Medline] [Order article via Infotrieve]
44. Schwaiger M, Neese RA, Araujo L, Wyns W, Wisneski JA, Sochor H, Swank S, Kulber D, Selin C, Phelps M, Schelbert H, Fishbein M, Gertz EW. Sustained nonoxidative glucose utilization and depletion of glycogen in reperfused canine myocardium. J Am Coll Cardiol. 1989;13:745-754. [Abstract]
45. Den Hollander JA, Evanochko WT, Pohost GM. Observation of cardiac lipids in humans by localized 1H magnetic resonance spectroscopic imaging. Magn Reson Med. 1994;32:175-180.[Medline] [Order article via Infotrieve]
This article has been cited by other articles:
 |
|
 |
|
  I. Barba, L. Chavarria, M. Ruiz-Meana, M. Mirabet, E. Agullo, and D. Garcia-Dorado Effect of intracellular lipid droplets on cytosolic Ca2+ and cell death during ischaemia-reperfusion injury in cardiomyocytes J. Physiol., March 15, 2009; 587(6): 1331 - 1341. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. K Mcguire, R. See, S. M Abdullah, P. G Snell, J. M Mcgavock, C. R Ayers, and L. S Szczepaniak The effect of rosiglitazone on integrated cardiovascular performance, cardiac structure, function and myocardial triglyceride: trial design and rationale Diabetes and Vascular Disease Research, January 1, 2009; 6(1): 43 - 50. [Abstract] [PDF]
|
|
|
|
 |
|
 L. J. Rijzewijk, R. W. van der Meer, J. W.A. Smit, M. Diamant, J. J. Bax, S. Hammer, J. A. Romijn, A. de Roos, and H. J. Lamb Myocardial Steatosis Is an Independent Predictor of Diastolic Dysfunction in Type 2 Diabetes Mellitus. J. Am. Coll. Cardiol., November 25, 2008; 52(22): 1793 - 1799. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. W. van der Meer, J. Doornbos, S. Kozerke, M. Schar, J. J. Bax, S. Hammer, MSc, J. W. A. Smit, J. A. Romijn, M. Diamant, L. J. Rijzewijk, et al. Metabolic Imaging of Myocardial Triglyceride Content: Reproducibility of 1H MR Spectroscopy with Respiratory Navigator Gating in Volunteers Radiology, October 1, 2007; 245(1): 251 - 257. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. Li, K. Tanaka, A. Ihaya, Y. Fujibayashi, S. Takamatsu, K. Morioka, M. Sasaki, T. Uesaka, T. Kimura, N. Yamada, et al. Gene therapy for chronic myocardial ischemia using platelet-derived endothelial cell growth factor in dogs Am J Physiol Heart Circ Physiol, January 1, 2005; 288(1): H408 - H415. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||