From the Division of Cardiology (R.W.J., L.C.B.) and the Division of
Nuclear Medicine (K.M.), Departments of Medicine and Radiology, The Johns
Hopkins Medical Institutions, Baltimore, Md, and the Department of
Experimental Cardiology, Max Planck Institut, Bad Nauheim, Germany (J.S.).
Correspondence to Dr Lewis C. Becker, Halsted 500, Johns Hopkins Medical Institutions, 600 N Wolfe St, Baltimore, MD 21205.
Methods and Results—In 12 anesthetized dogs, the left
anterior descending coronary artery was occluded for 90 minutes
before 4 hours reperfusion. Myocardial blood flow was measured by
microspheres and the tracers 14C-2-deoxyglucose and
18F-2-deoxyglucose were injected intravenously
after 5 and 180 minutes of reperfusion, respectively. After 240
minutes, the heart was stained with thioflavin-S (size of no-reflow
zone) and triphenyl-tetrazolium chloride (TTC, extent of necrosis).
Samples from normal, salvaged, and necrotic myocardium were
counted for 14C- and 18F-deoxyglucose and
microspheres. With the use of a three-compartment model of
2-deoxyglucose uptake, the rate constant k3
for phosphorylation of 14C- and
18F-2-deoxyglucose was calculated for each sample.
Viability was defined as k3
Conclusions—A large proportion of samples from infarcted
myocardium are viable at the end of the ischemic
period but lose viability during the first hours of reperfusion.
Myocardial Distribution and Retention of 2-Deoxyglucose
Retention of 14C-2-deoxyglucose in reperfused
myocardium was studied in three other dogs. After 90
minutes of LAD occlusion and 5 minutes of reperfusion,
14C-2-deoxyglucose was given
intravenously. Transmural biopsies (15 to 30 mg wet wt)
were taken in triplicate from LAD and circumflex artery territories by
biopsy drill at 1, 2, and 3 hours after tracer injection. The heart was
arrested after 4 hours and further biopsies taken. The biopsies were
weighed and counted for 14C activity. In two
other dogs, myocardial retention of
18F-2-deoxyglucose was compared with myocardial
blood flow during ischemia and reperfusion. The LAD was
occluded for 90 minutes and blood flow measured by radionuclide-labeled
15–µm microspheres (141Ce,
113Sn, 46Sc; Du Pont Co)
injected into the left atrium, with reference sampling from the right
femoral artery, at 10 minutes before reperfusion. After 10 minutes
reperfusion, 18F-2-deoxyglucose was injected
intravenously. A biopsy was excised by scalpel from
reperfused myocardium 35 minutes later and divided into
endocardial and epicardial halves. After 3 hours of reperfusion, blood
flow was measured with microspheres before repeat biopsy of
reperfused myocardium and the normal circumflex artery
territory. Samples were weighed and counted for
18F activity and microspheres.
Comparison of 18F-2-Deoxyglucose Uptake With
Histology
Operators studied all samples by light and electron microscopy, without
knowledge of sample location or regional blood flow. Samples were
embedded in epon with the use of a Wakura automatic tissue processor,
after fixation in 2% osmic acid anhydride, dehydration in an ethanol
series, and substitution by propylenoxide. Semithin (1 to 2 µm)
sections were stained with toluidine blue. Artifact-free areas were
selected for preparation of thin sections (50 to 60 nm), which were
attached to uncoated copper grids, stained with uranyl acetate and lead
citrate, and viewed in a Phillips EM 300-electron microscope. For each
myocardial sample, 30 to 40 micrographs were examined according to
previously established criteria.15 16 Reversible
injury was identified by absence of contraction bands, an intact
sarcolemma, absence of mitochondrial amorphous densities, and absence
of nuclear clearing and shrinkage. Irreversible injury was identified
by the presence of amorphous densities, matrix clearing and/or cristae
breakage in the mitochondria, clearing and shrinkage of nuclei, and/or
disruption of the sarcolemma. A sample was deemed to have suffered
irreversible injury if >50% of the micrographs from that sample
showed evidence of irreversible injury.
Sequential Measurements of 2-Deoxyglucose Uptake in Reperfused
Myocardium
The left ventricle was isolated and sectioned into five transverse
slices (8 to 10 mm thick), which were weighed and examined under
ultraviolet light to define no-reflow zones (absent thioflavin-S
fluorescence). Endocardial and epicardial surfaces of each
myocardial slice and borders of the ischemic region and the
no-reflow zones were traced on acetate sheets. Transmural sections were
excised at multiple sites in the LAD and circumflex territories of each
heart and divided into fifths (70 to 100 mg wet wt per sample). A total
of 60 to 70 samples were obtained from each heart, and sample sites
were recorded on the acetate sheets. The myocardial slices were
incubated in 2,3,5-TTC solution at 37°C for 30 minutes to
differentiate infarcted myocardium (absent or negative TTC
staining) from salvaged myocardium (brick-red or positive
TTC staining).17 18 19 The borders of infarct and
noninfarct regions were traced on the corresponding acetate sheets and
planimetered to measure the ischemic risk region, infarct
region, and no-reflow region.
Radionuclide Measurements
Because reperfusion of lethally injured myocardium is
associated with tissue edema and an increase in tissue wet weight of
The measure of viability of a myocardial sample is the
phosphorylation rate of 2-deoxyglucose. When the time
from tracer injection to sampling exceeds 30 minutes, the amount of
nonmetabolized tracer in the sample approaches
zero.22 With the use of a three-compartment model
of myocardial uptake and phosphorylation of
2-deoxyglucose,14 the overall reaction rate in
the sample of interest Ri can be described
as:
The rate constants of 2-deoxyglucose uptake and
phosphorylation in reperfused canine
myocardium have been previously
described,14 allowing calculation of
k3i for the infarct
region. The previous data show that k3
increases in normal myocardium after an infarct, but
k3 in salvaged postischemic
myocardium remains comparable to that of the control state.
Therefore, the salvaged (TTC-positive) myocardium was used
as the reference region. For all samples from reperfused
myocardium the rate constants used were
K1r=0.61 mL ·
min-1 · g-1,
k2r=0.87
min-1. For samples from normal
myocardium, the rate constants used were
K1i=0.83 mL ·
min-1 · g-1 and
k2i=1.44
min-1.14 The same rate
constants were used for calculation of k3
for 14C- and
18F-2-deoxyglucose.
Statistics
Comparison of 2-Deoxyglucose Uptake With Histopathology
Serial Studies of 2-Deoxyglucose Uptake in Reperfused
Myocardium
A total of 850 myocardial samples were examined (340 from the control
circumflex territory and 510 from the reperfused LAD territory). Among
samples from reperfused myocardium, 237 were from
TTC-negative regions, including 58 from the no-reflow zone, and 164
were from TTC-positive regions. There were 109 samples from borders of
TTC-negative and TTC-positive myocardium, which were not
included in the data analysis. Among the 401 samples from
reperfused myocardium, the
18F-2-deoxyglucose k3
threshold of 0.125 min-1 identified 235 of the
237 TTC-negative samples as nonviable and 155 of the 164 TTC-positive
samples as viable (sensitivity for identifying viable
myocardium, 93%; specificity, 99%; predictive accuracy,
97%).
Contrasting examples of 2-deoxyglucose uptake during early and late
reperfusion are shown in Figs 4
Data for a different heart are shown in Fig 5
The 2-deoxyglucose k3 values in
TTC-positive and TTC-negative myocardium are summarized for
the group in Table 2
Ischemic Necrosis, Reperfusion Necrosis, and Infarct
Size
The proportion of anatomic infarct size due to necrosis during
ischemia or reperfusion in each dog was calculated from the
number of infarct samples that were nonviable by both
14C-2-deoxyglucose and
18F-2-deoxyglucose (ischemic necrosis) or
viable by 14C-2-deoxyglucose but nonviable by
18F-2-deoxyglucose (reperfusion necrosis) for
that particular dog. The proportions of infarct size due to
ischemic or reperfusion necrosis varied according to the level
of collateral blood flow (Fig 7
Myocardial Viability and 2-Deoxyglucose
The k3 in the reference region and the
value of the viability threshold might change during reperfusion, but
our pathology comparisons indicate the same viability threshold at 35
minutes and at 4 hours after reperfusion. There was no systematic
change in k3 in the normal
myocardium in this study and the previous PET
study14 found no change in
k3 in salvaged myocardium. The
proportion of samples deemed viable by
14C-2-deoxyglucose and
18F-2-deoxyglucose in TTC-positive
myocardium were the same across a wide range of
k3 values. To avoid any bias related to
selection of the threshold value of k3, we
determined the number of samples that were viable in the infarct region
according to a wide range of threshold values of
k3. Irrespective of the
k3 threshold used, many samples from the
TTC-negative infarct region, which were viable at the time of
14C-2-deoxyglucose injection after 5 minutes of
reperfusion, were nonviable by the time of
18F-2-deoxyglucose injection 3 hours later.
Interpretation of Findings
Impaired delivery of 18F-2-deoxyglucose to the
infarct region after 3 hours of reperfusion is also unlikely to account
for our findings. Blood flow to the infarct zone was mildly reduced
after 3 hours, but mean flows in the TTC-negative and TTC-positive
regions were similar, and myocyte uptake of
18F-2-deoxyglucose at steady state is independent
of blood flow. Reperfusion of infarcted myocardium is
associated with myocyte swelling and interstitial
edema.21 The true uptake of
14C-2-deoxyglucose during early reperfusion might
be underestimated when the sample is weighed after 4 hours. This error
would underestimate the calculated k3 for
early reperfusion but could not explain the differences in
k3 found in this study. Correction for
tissue edema in the infarcted myocardium was used in this
study, but even in the absence of any such correction, 60% of samples
from TTC-negative myocardium were viable after 5 minutes of
reperfusion, with k3>0.125
min-1 for
14C-2-deoxyglucose.
Reperfusion of irreversibly injured myocytes is associated with
contraction bands, cell swelling, and sarcolemmal
disruption.27 28 Such cells would be unlikely to
accumulate 14C-2-deoxyglucose, consistent
with our finding in samples from the no-reflow zones. An increase in
sarcolemmal permeability29 leads to loss of
enzymes such as creatine kinase. The reduced
18F-2-deoxyglucose content seen during later
reperfusion might reflect washout of hexokinase from necrotic myocytes,
but 14C-2-deoxyglucose would also be lost from
the same myocytes. Our observations cannot be explained as an artifact
of tissue edema or tracer washout. The most likely explanation is that
myocytes, which were viable during early reperfusion, subsequently lost
viability during the next 3 hours of reperfusion.
Irreversible Myocardial Injury During Reperfusion
The small residual uptake of 2-deoxyglucose observed in infarcted
myocardium may represent a few surviving myocytes,
or uptake in endothelial cells or fibroblasts, but the
volume of these elements is small compared with myocyte volume.
Neutrophil leukocytes accumulating during
reperfusion36 37 are a potential site of
2-deoxyglucose uptake, but leukocyte uptake of
18F-2-deoxyglucose in reperfused
myocardium is small compared with overall myocyte
uptake.38 Furthermore, any such error would
result in an increase in tissue
18F-2-deoxyglucose uptake, which is opposite to
our observations.
Two other studies have reported data consistent with the
occurrence of irreversible myocardial injury during reperfusion. One
study in rabbits, using sequential tissue staining with horseradish
peroxidase and TTC, found an apparent increase in infarct size during 3
hours of reperfusion.39 A canine study, using
radionuclide-labeled antimyosin antibodies, found a progressive
increase in antibody binding in reperfused
myocardium,40 which suggests lethal
injury, although these observations could also be explained by
increasing sarcolemmal permeability in infarcted myocytes.
This study does not define the mechanism of lethal myocardial injury
occurring during reperfusion, although the concordance between the time
course of reperfusion injury and neutrophil
infiltration36 37 is compelling. Intervention
studies have implicated neutrophil leukocytes in the pathogenesis of
irreversible injury during reperfusion.7 8
Although it is possible that myocytes are "programmed" for
inevitable necrosis during reperfusion, as a result of an irreversible
ischemic insult, our data show that a significant proportion of
samples from the infarct region are viable at the time of reperfusion
and many interventional studies argue that reperfusion necrosis is not
inevitable.
Received June 5, 1997;
revision received September 29, 1997;
accepted October 7, 1997.
© 1998 American Heart Association, Inc.
Basic Science Reports
Progression of Myocardial Necrosis During Reperfusion of Ischemic Myocardium
Abstract
Top
Abstract
Introduction
Methods
Results
Discussion
References
Background—The occurrence of myocyte
necrosis during reperfusion of ischemic myocardium
is controversial. This study measured myocardial 2-deoxyglucose uptake,
correlated with histology, to determine whether loss of viability
occurred during reperfusion of ischemic myocardium. 
0.125
min-1 (predictive accuracy 88% versus electron microscopy
and 97% versus TTC). Among 58 samples from no-reflow regions, 97%
were nonviable after 5 minutes of reperfusion
(k3=0.096±0.027 min-1). Among
164 samples from salvaged myocardium, 95% were viable
after both 5 and 180 minutes of reperfusion
(k3=0.170±0.056 min-1
P<.01 versus no-reflow). Among 179 samples from
infarcted myocardium, mean k3
after 5 minutes of reperfusion was 0.184±0.070 min-1 and
65% of samples were viable, but after 180 minutes of reperfusion mean
k3 had decreased to 0.077±0.032
min-1 (P<.0001) and 98% of samples were
nonviable.
Key Words: metabolism • myocardial infarction • ischemia • reperfusion • radioisotopes
Introduction
Top
Abstract
Introduction
Methods
Results
Discussion
References
Coronary reperfusion
improves ventricular function and survival after
infarction,1 2 but concern persists that damaged
but otherwise viable myocytes may undergo necrosis during
reperfusion.3 4 Although interventional studies
with scavengers of oxygen radicals,5 6 inhibition
or removal of neutrophil leukocytes,7 8 and
administration of adenosine9 suggest that
myocardial necrosis does occur during reperfusion, direct evidence has
been lacking. Sequential measurements of viability, using a marker of
basic cellular metabolism, are required to address this
question. Radionuclide-labeled 2-deoxyglucose is used as a tracer of
glucose uptake and phosphorylation in
brain10 and heart.11
Although rapidly phosphorylated by
hexokinase,12 2-deoxyglucose is not a substrate
for further glycolytic metabolism and is trapped in the
cell.13 The rate constant for
phosphorylation of
18F-2-deoxyglucose by hexokinase
(k3) is correlated with glucose
metabolism in the reperfused
myocardium.14 We used 2-deoxyglucose
for sequential measurements of viability in reperfused
myocardium, with injection of
14C-2-deoxyglucose immediately after reperfusion
and 18F-2-deoxyglucose 3 hours later. Uptake of
2-deoxyglucose was compared with histochemical and ultrastructural
evidence of reversible and irreversible myocardial injury and
correlated with collateral blood flow during ischemia to
differentiate lethal injury occurring during ischemia from that
occurring during reperfusion.
Methods
Top
Abstract
Introduction
Methods
Results
Discussion
References
Thirty-three mongrel dogs of either sex (weight, 20 to 27 kg)
were studied after an overnight fast. The dogs were
anesthetized with sodium thiamylal (12.5 mg/kg IV) followed by
chloralose (14 mg/kg IM) in urethane (136 mg/kg). Polyvinyl catheters
were placed in the right femoral artery and vein for reference sampling
of microspheres and administration of intravenous
fluids, respectively. After left thoracotomy, a catheter was placed in
the left atrium for microsphere injections and a catheter-tip
pressure transducer (5F, Millar Instruments) was advanced to the left
ventricle from the left atrium. An inflatable occluder was placed
around the proximal LAD. Hemodynamics were continuously
recorded on chart paper (Gould Instruments).
Because 14C-2-deoxyglucose and
18F-2-deoxyglucose are not structurally
identical, the distribution of the two tracers in reperfused
myocardium was studied in three dogs. After 90 minutes of
LAD occlusion, the dogs received simultaneous
intravenous 14C-2-deoxyglucose (25
µCi; specific activity, 59 mCi/mmol; Sigma) and
18F-2-deoxyglucose (0.25 to 1.5 mCi; Division of
Nuclear Medicine, Johns Hopkins Medical Institutions). Tracers were
injected after 5 minutes of reperfusion in one dog, after 15 minutes in
the second, and after 3 hours in the third. At 60 minutes after tracer
injection, the LAD was reoccluded and monastral blue dye injected into
the left atrium to define the ischemic region. Multiple (n=30
to 40) biopsies (30 to 85 mg wet wt) were obtained from reperfused
myocardium for counting of 14C and
18F activities.
In seven dogs the LAD was occluded for 90 minutes and myocardial
blood flow measured by microspheres at 10 minutes before
reperfusion. After 5 minutes of reperfusion,
18F-2-deoxyglucose was injected
intravenously, followed by a second microsphere
injection. After 35 minutes of reperfusion, the heart was arrested by
intravenous potassium chloride, excised, and sectioned into
short-axis slices. Multiple transmural sections from normal and
reperfused regions were divided into fifths from endocardium to
epicardium (30 to 100 mg wet wt per sample). Samples were randomly
selected from the control and reperfused myocardium in each
dog for electron microscopy. A small section of each sample was
immersed in cold (4°C) 3% glutaraldehyde in 0.1
mol/L cacodylate buffer, pH 7.4, and kept in fixative at 4°C for 24
hours before rinsing in 0.1 mol/L cacodylate buffer (with saccharose
added, pH 7.4) and storage at 4°C before examination. The remainder
of the sample was counted for 18F activity and
microspheres.
Myocardial uptake of 2-deoxyglucose after 5 minutes and 3 hours
of reperfusion was studied in 18 dogs in which the LAD was occluded for
90 minutes before free reperfusion was allowed. Myocardial blood flow
was measured by microspheres 10 minutes before reperfusion and
10 minutes and 3 hours after reperfusion. After 5 minutes of
reperfusion, 14C-2-deoxyglucose (25 µCi) was
injected intravenously and after 3 hours of reperfusion,
18F-2-deoxyglucose (
1 mCi; range, 0.25 to 2
mCi) was injected intravenously. The
18F-2-deoxyglucose was used as the second tracer
because of the short half-life of 18F. One hour
after injection of 18F-2-deoxyglucose, the
fluorescent dye thioflavine-S (2% solution) was injected
into the left atrium to define no-reflow zones. Two minutes later the
LAD was reoccluded and monastral blue dye (20 mL) was injected into the
left atrium to define the ischemic risk region. The heart was
then arrested with potassium chloride and excised for tissue
sampling and measurement of infarct size.
Myocardial samples were weighed and counted with flow
reference samples and radionuclide standards in an NaTl crystal well
counter (Packard 5986) set for photopeaks of 18F,
141Ce, 113Sn, and
46Sc. Decay-corrected counts were corrected for
crossover between radionuclides and blood flow calculated according to
standard methods.20 Samples were then solubilized
(Protosol, Du Pont) and incubated at 50°C for 48 hours, before
addition of 10 mL scintillation cocktail (EconoFluor, Du Pont) and
liquid scintillation counting (Packard Tri-Carb 2660) for
14C activity. To eliminate any error in
14C measurement due to the presence of gamma
emitters in the samples, calibration curves were determined for beta
activity observed in the presence of 141Ce,
113Sn, and 46Sc. After
counting for 18F activity, tissue samples were
counted twice in both gamma and beta counters at an interval of 4
weeks, with decay correction for each nuclide. Beta count activity due
to the gamma emitters was then subtracted from total observed beta
counts to derive the true 14C count activity. 25% during the first few hours,21
radionuclide count data for microspheres and
14C-2-deoxyglucose (injected during
ischemia or early reperfusion) in samples from the infarct
region were corrected to allow for a 25% increase in tissue wet weight
during reperfusion.
where BG=blood glucose level;
K1i=rate of transport of
2-deoxyglucose into myocyte from plasma;
k2i=rate of reverse
transport of 2-deoxglucose from myocyte to plasma; and
k3i=rate of
phosphorylation of 2-deoxyglucose by hexokinase.
Because the lumped constant (LC) and arterial input
function are the same for all samples in each heart, the ratio of
reaction rates in two samples can be calculated as the ratio of tracer
activities:
where CiT=tracer content in sample of
interest; and CrT=tracer content in reference
sample.
The proportion of samples in reperfused myocardium
that were viable with 14C-2-deoxyglucose (early
reflow) was compared with the proportion that were viable with
18F-2-deoxyglucose (late reflow) by
2 analysis.
Hemodynamic and regional myocardial blood flow
measurements during ischemia and reperfusion were compared by
ANOVA.23 Myocardial 2-deoxyglucose contents were
compared between control and postischemic regions in each
dog by ANOVA. The proportions of samples that appeared to undergo
necrosis during the ischemic period, or necrosis during
reperfusion, or remained viable were compared with collateral blood
flow by regression analysis. Results are reported as mean±SD,
and a value of P<.05 is described as significant.
Results
Top
Abstract
Introduction
Methods
Results
Discussion
References
Uptake and Distribution of 2-Deoxyglucose
The distribution of
14C-2-deoxyglucose and
18F-2-deoxyglucose in reperfused
myocardium, after simultaneous injection of
both tracers, is shown in Fig 1
. Data are
shown for 108 myocardial samples from three dogs. The slope (0.93)
reflects slightly lower 18F-2-deoxyglucose uptake
after 5 minutes of reperfusion, but neither the group regression nor
individual regressions differed from the line of identity, indicating
that tissue distributions and retention of the tracers were equivalent.
Retention of 14C-2-deoxyglucose in normal and
reperfused myocardium is shown for another three dogs in
Fig 1. One hour after injection, mean
14C-2-deoxyglucose content in reperfused
myocardium was half of that in normal
myocardium (P<.01). Myocardial count activities
were similar to those in subsequent experiments. During 3 hours after
injection, there was no significant change in mean
14C activity in normal or reperfused
myocardium. Biopsies obtained 4 hours after reperfusion had
14C activity similar to that of biopsies taken
after 1 hour. In the other two dogs, which had both flow and
18F-2-deoxyglucose uptake measurements, samples
were grouped according to myocardial blood flow. Samples with severe
ischemia (collateral flow <10% of control) and impaired
reperfusion (flow <40% of control), which are likely to have the most
severe necrosis, had similar 18F-2-deoxyglucose
content after 35 minutes (40% of control) and after 3 hours (41% of
control) reperfusion. These data show that radiolabeled 2-deoxyglucose,
injected during early reperfusion, is retained over 4 hours in
reperfused myocardium.
View larger version (23K):
[in a new window]
Figure 1. Top, Uptake of 14C-2-deoxyglucose and
18F-2- deoxyglucose after simultaneous
injection of both tracers. Count activities in reperfused
myocardium were normalized to mean activity of normal
region for comparison between dogs. Lines of identity (dashed) and
group regression (solid) are shown. Bottom, Retention of
14C-2-deoxyglucose in normal (open symbols) and reperfused
(closed symbols) myocardium. Data are shown for each of
three dogs.
The uptake of 18F-2-deoxyglucose in normal
and reperfused myocardium was compared with electron
microscopy findings in seven dogs. Blood flow to control (1.45±0.54
mL/min per gram) and reperfused (1.54±0.34 mL/min per gram)
myocardium was similar, but
18F-2-deoxyglucose content was less in reperfused
myocardium (61 465±30 328 counts/min per gram) than in
the control region (106 760±71 013 counts/min per gram,
P<.05). Samples were randomly selected from control (n=6)
and reperfused (n=28) regions. All control samples had the
ultrastructural features of viable myocardium (Fig 2A). Among samples from reperfused
myocardium, 8 manifested reversible ischemic injury
(Fig 2B) and 20 had irreversible injury (Fig 2C). The mean
k3 for these samples were 0.320±0.152
min-1 for control, 0.194±0.070
min-1 for reversible injury, and 0.098±0.055
min-1 for irreversible injury (P<.01
versus reversible injury). The individual
k3 values were compared with the electron
microscopy findings to determine which value of
k3 was the best discriminator between
reversible and irreversible injury (Fig 3). A value of
k3=0.125 min-1
appeared to be the best indicator of viability (sensitivity=93%,
specificity=85%, predictive accuracy=88%). If a value of
k3=0.100 min-1 was
used, specificity decreased to 60%, and if a value of
k3=0.150 min-1 was
used, sensitivity decreased to 86%. For the subsequent serial studies
of 2-deoxyglucose uptake, samples with k3
<0.125 min-1 were considered nonviable and those
with k3 0.125 min-1 were
considered viable.
View larger version (84K):
[in a new window]
Figure 2. A, Ultrastructure of normal myocardium
with intact sarcolemma and normal nuclear chromatin. The
k3 for this sample was 0.296. B, Reversibly
injured myocardium with clumping of nuclear chromatin and
mitochondrial swelling but intact sarcolemma. The
k3 for this sample was 0.212. C,
Irreversibly injured myocardium with mitochondrial swelling
and loss of cristae with calcium precipitates and marked nuclear
changes. The k3 for this sample was 0.085. B
and C are from the same dog.
View larger version (18K):
[in a new window]
Figure 3. Sensitivity and specificity of different levels of
k3 for detection of viability in reperfused
myocardium, in comparison with electron microscopy
findings.
Among 18 dogs included in the group, two had
ventricular fibrillation shortly after reperfusion and were
not resuscitated. Four dogs with collateral blood flows >30% control
flows and no evidence of infarction on TTC staining were excluded from
analysis. Data are reported for 12 dogs that completed 90
minutes of ischemia and 4 hours of reperfusion with TTC
evidence of infarction. Heart rate did not change from before
ischemia (136±19 bpm) to 3 hours of reperfusion (135±28 bpm),
but mean arterial pressure was lower during
ischemia (92±22 mm Hg) than before ischemia
(102±24 mmHg) or after 3 hours of reperfusion (102±20
mm Hg, P<.05 versus ischemia). Blood flow in
the circumflex territory was 1.07±0.57 mL/min per gram after 5 minutes
of reperfusion and 1.03±0.55 mL/min per gram after 3 hours. Collateral
flow during LAD occlusion was 0.07±0.04 mL/min per gram, increasing to
1.19±0.34 mL/min per gram during early reperfusion. After 3 hours,
flow in the LAD myocardium was 0.68±0.23 mL/min per gram
(P<.05 versus early reperfusion). Collateral flow to the
TTC-positive region was greater than flow to the TTC-negative region
(P<.01) (Table 1). Blood flow
was reduced in the no-reflow zones, but there was no other difference
in flow between TTC-negative and TTC-positive regions during
reperfusion.
View this table:
[in a new window]
Table 1. Regional Blood Flow in Reperfused
Myocardium and 5. Data from one heart are shown in Fig 4. Samples are grouped by origin from TTC-positive, TTC-negative, or
no-reflow regions. This heart had reduced
14C-2-deoxyglucose uptake in TTC-negative
myocardium after 5 minutes of reperfusion (Fig 4, A and
B). In TTC-negative samples the k3
for 14C-2-deoxyglucose was 0.082±0.019
min-1, and all but one of these samples had
k3<0.125 min-1,
indicating necrosis by 5 minutes after reperfusion. The same
TTC-negative samples all had k3<0.125
min-1 for
18F-2-deoxyglucose (Fig 4, C and D), with
k3=0.067±0.015 min-1
(NS versus 14C-2-deoxyglucose). Eight samples
from no-reflow regions all had k3<0.125
min-1 at both early and late reperfusion. The
data in this heart are consistent with irreversible injury
occurring during ischemia.
View larger version (22K):
[in a new window]
Figure 4. Uptake of 14C-2-deoxyglucose and
18F-2-deoxyglucose in reperfused myocardium of
one heart. The TTC-negative region had low uptake of both
14C-2-deoxyglucose (A) and 18F-2-deoxyglucose
(C). The TTC-negative samples were below the viability threshold
(k3=0.125 minutes-1 shown as
dashed line) for both 14C-2-deoxyglucose (B) and for
18F-2-deoxyglucose (D), suggesting that necrosis was
complete by the end of the ischemic period.
View larger version (20K):
[in a new window]
Figure 5. Uptake of 14C-2-deoxyglucose and
18F-2-deoxyglucose in reperfused myocardium of
another heart. The TTC-negative samples had uptake of
14C-2-deoxyglucose similar to TTC-positive samples with
k3 values above the viability threshold
(k3=0.125 min-1 shown as dashed
line) (A and B). Later, TTC-negative samples had low uptake of
18F-deoxyglucose (C) with k3
values below the viability threshold (D), consistent with the
occurrence of irreversible injury during reperfusion. 
. There was avid uptake of
14C-2-deoxyglucose during early reperfusion in
both TTC-positive and TTC-negative regions (Fig 5A). The calculated
k3 for
14C-2-deoxyglucose in the TTC-negative region was
0.228±0.077 min-1. All samples in the
TTC-negative region had k3>0.125
min-1, indicating viability at 5 minutes of
reperfusion (Fig 5B). After 3 hours of reperfusion the
k3 for
18F-2-deoxyglucose in the same TTC-negative
samples was 0.057±0.016 min-1
(P<.001 versus 14C-2-deoxyglucose),
and all TTC-negative samples had k3<0.125
min-1, indicating irreversible injury by the time
of 18F-2-deoxyglucose injection (Fig 5, C and D).
The data in this heart are consistent with the occurrence of
necrosis during the reperfusion period. . In the no-reflow
region, k3 after 5 minutes of reperfusion
was 0.096±0.027 min-1 and after 3 hours
0.060±0.023 min-1 (NS). All but 2 of these
samples were classified as nonviable after 5 minutes and all were
nonviable after 3 hours of reperfusion. In the TTC-negative region,
k3 decreased from 0.184±0.070
min-1 after 5 minutes of reperfusion to
0.077±0.032 min-1 (P<.0001) after 3
hours. After 5 minutes, 117 of these 179 samples were viable according
to the k3 threshold, but after 3 hours only
4 were viable (P<.0001). In the TTC-positive region, mean
k3 after 5 minutes (0.170±0.087
min-1) and after 3 hours (0.170±0.056
min-1) were similar, and 155 of these 164 samples
were viable. Among 237 samples from no-reflow and TTC-negative
infarct regions, 119 (50.2%) were viable at the time of
14C-2-deoxyglucose injection, but only 4 (1.7%)
were viable at the time of 18F-2-deoxyglucose
injection, consistent with the development of irreversible
injury during reperfusion. The proportion of samples classified as
viable in the no-reflow, TTC-negative, and TTC-positive regions,
according to different values of k3 are
shown in Fig 6. For
k3 between 0.100 and 0.200
min-1, almost no samples in the no-reflow zones
were classified as viable by 14C-2-deoxyglucose
or 18F-2-deoxyglucose. In TTC-positive
myocardium, the proportions of samples classified as viable
by 14C-2-deoxyglucose and
18F-2-deoxyglucose were similar, whichever
value of k3 was used. In TTC-negative
myocardium, there was a marked difference between the
proportions of samples classified as viable by
14C-2-deoxyglucose and those classified as
viable by 18F-2-deoxyglucose.
View this table:
[in a new window]
Table 2. Regional 2-Deoxyglucose Uptake in Control and
Reperfused Myocardium
View larger version (17K):
[in a new window]
Figure 6. Comparisons of the number of viable samples in
infarcted and salvaged regions of the LAD territory at time of
injection of 14C-2-deoxyglucose and of
18F-2-deoxyglucose. Data are shown for the proportion of
viable samples in each region, according to different levels of
k3. A, Samples from within the no-reflow
zone. All but two samples appeared nonviable by 5 minutes of
reperfusion and all were nonviable after 3 hours of reperfusion. B,
Samples from the TTC-negative region. Almost all samples were necrotic
after 3 hours of reperfusion, but only 35% of these samples were
nonviable after 5 minutes of reperfusion. At each
k3 value, many more samples were viable at
the time of 14C-2-deoxyglucose injection than at the time
of 18F-2-deoxyglucose injection. *P<.05,
**P<.01 vs 18F-2- deoxyglucose at same
k3. C, Samples from within the salvaged
TTC-positive zone. There is no difference in viability determined by
14C-2-deoxyglucose and 18F-2-deoxyglucose. For
the threshold of k3=0.125 min-1,
nearly all samples are deemed viable at both 5 minutes and 3 hours
after reperfusion.
The ischemic risk region occupied 30.0±4.0% of the left
ventricle. The mean infarct size was 37.6±21.0% of the risk region
(range, 5.7% to 68.3%), and the mean size of the no-reflow region was
8.3±7.7% of the risk region. Total infarct size was inversely related
to collateral blood flow during ischemia (r= -.87),
and the extent of the no-reflow zone was also inversely related to
collateral blood flow (r= -.64). ). The
four dogs with the lowest collateral blood flow (2.8±1.1% of control
flow) had the largest infarcts (54±7% of risk region) and in these
dogs 86±12% of the infarct samples were irreversibly injured by 5
minutes of reperfusion. The four dogs with intermediate collateral
flows (4.8±0.9% of control) had an infarct size of 46±15% of risk
region, but only 30±23% of infarct samples were irreversibly injured
by 5 minutes of reperfusion. In contrast, the four dogs with the
highest collateral flows (12.7±3.1% of control) had small infarcts
(13±10% of risk region) and only 7±7% of infarct samples were
nonviable by 5 minutes of reperfusion. As collateral blood flow
increased, the proportion of infarction due to irreversible injury
during reperfusion increased.
View larger version (22K):
[in a new window]
Figure 7. A, Relationship between collateral flow (% of
normal resting flow) and infarct size due to ischemic necrosis
according to 2-deoxyglucose data. B, The proportion of total infarct
size due to ischemic necrosis was inversely related to
collateral flow. Low collateral flows were associated with large
infarcts, almost all due to ischemic necrosis. C, Relationship
between collateral flow and infarct size due to reperfusion necrosis,
according to 2-deoxyglucose data. Regions most likely to manifest
necrosis during reperfusion were those with only mild or moderate
ischemia during LAD occlusion. D, The proportion of total
infarct size due to reperfusion necrosis. As collateral flows
increased, a greater proportion of necrosis appeared to occur during
reperfusion.
Discussion
Top
Abstract
Introduction
Methods
Results
Discussion
References
This study examined changes in viability of reperfused
myocardium. In salvaged myocardium, uptakes of
14C-2-deoxyglucose during early
reperfusion and 18F-2-deoxyglucose 3 hours later
were similar and above a threshold of viability. In samples from
severely injured no-reflow regions, both
14C-2-deoxyglucose and
18F-2-deoxyglucose uptakes were below the
viability threshold. In most samples from the infarct region,
14C-2-deoxyglucose uptake during early
reperfusion was above the viability threshold, but 3 hours later uptake
of 18F-2-deoxyglucose in the same samples was
reduced to levels associated with irreversible injury,
consistent with the occurrence of myocyte necrosis during
reperfusion. Regions with low collateral flows had nearly complete loss
of viability by the end of ischemia, but regions with higher
collateral flows appeared to undergo necrosis during reperfusion.
The tracer 18F-2-deoxyglucose is widely used
as a viability marker in clinical studies with
PET.24 The phosphorylated tracer
accumulates within myocytes, reaching steady state by 60
minutes,25 because
dephosphorylation is slow and alternate
metabolic pathways are limited.26
Myocardial uptake of 2-deoxyglucose depends on tracer delivery and
kinetics of 2-deoxyglucose transport and
phosphorylation. We calculated
k3, the rate constant for 2-deoxyglucose
phosphorylation, in individual myocardial samples,
using parameters derived from PET studies of reperfused
infarcts in canine hearts.14 The values of
k3 calculated for normal, salvaged, and
infarcted myocardium in this study are similar to the
previous PET data. To define an appropriate value of
k3 as a marker of viability, we compared
tissue k3 with ultrastructural appearance
and histochemical staining of reperfused myocardium.
Myocardial samples with an ultrastructural pattern of irreversible
injury, after 1 hour of reperfusion, had
k3<0.125 min-1, but
samples showing reversible injury had k3 0.125
min-1. Similarly, samples that were necrotic by
TTC stain after 4 hours of reperfusion had
k3<0.125 min-1, but
samples that were viable by TTC stain had k3
0.125 min-1. The lowest
k3 values were found in the no-reflow
regions with the most severe ischemic injury. Small differences
in predictive accuracy of k3=0.125
min-1 for detection of viability compared with
electron microscopy or TTC stain reflect different numbers of samples
in each comparison. It should be noted that this threshold may not be
universally applicable, particularly for clinical PET studies when
patients are given a glucose load.
Approximately half of myocardial samples from the TTC-negative
infarct region were apparently viable during early reperfusion, but
during the next 3 hours exhibited a decrease in 2-deoxyglucose
phosphorylation to the levels found in necrotic
myocardium from the no-reflow zone. This observation is
consistent with the occurrence of myocardial necrosis during
reperfusion, but several other possible interpretations should be
examined. Loss of 18F-2-deoxyglucose during late
reperfusion is unlikely to account for the observed differences in
myocardial 2-deoxyglucose content between 5 minutes and 3 hours of
reperfusion. First, our initial experiments showed that myocardial
content of 2-deoxyglucose was largely unchanged during 4 hours of
reperfusion. Second, if 18F-2-deoxyglucose were
lost from necrotic myocytes, then
14C-2-deoxyglucose would also be lost. Third,
increased metabolism of
18F-2-deoxyglucose-phosphate is unlikely, as the
rate constant for dephosphorylation remains an order of
magnitude below k3 during
reperfusion.14
Many samples from infarcted myocardium appear to have
undergone necrosis after restoration of coronary blood flow.
Myocytes can undergo necrosis in the presence of apparently adequate
coronary perfusion, as with catecholamine
stress,30 31 loss of calcium
homeostasis,32 33 or
reoxygenation after anoxia.34 It
is also now known that myocytes may be programmed to die through the
process of apoptosis.35 Although normally
a mechanism for removal of senescent cells, it is possible that
apoptosis may be responsible for large scale cell loss under
pathological conditions.
Selected Abbreviations and Acronyms
LAD
=
left anterior descending coronary artery
TTC
=
triphenyl-tetrazolium chloride
PET
=
positron emission tomography
Acknowledgments
This study was supported by USPHS grant 17655 (Specialized
Center of Research in Ischemic Heart Disease) from the National
Heart, Lung, and Blood Institute, Bethesda, Md. Dr Jeremy was supported
by an Overseas Research Fellowship of the National Heart Foundation of
Australia and a Telectronics Research Fellowship of the Royal
Australasian College of Physicians. The authors wish to thank Dr Robert
Dannals of the Division of Nuclear Medicine for the generous supply of
18F-deoxyglucose, Anthony Di Paula for laboratory
assistance, Dr Hugh McCarron for statistical analyses, and
Christine Holzmueller for secretarial assistance.
References
Top
Abstract
Introduction
Methods
Results
Discussion
References
This article has been cited by other articles:
 |
|
 |
|
  M. T. Dirksen, G. J. Laarman, M. L. Simoons, and D. J.G.M. Duncker Reperfusion injury in humans: A review of clinical trials on reperfusion injury inhibitory strategies Cardiovasc Res, June 1, 2007; 74(3): 343 - 355. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. W. Bar, D. Tzivoni, M. T. Dirksen, A. Fernandez-Ortiz, G. R. Heyndrickx, J. Brachmann, J. H.C. Reiber, N. Avasthy, J. Tatsuno, M. Davies, et al. Results of the first clinical study of adjunctive CAldaret (MCC-135) in patients undergoing primary percutaneous coronary intervention for ST-Elevation Myocardial Infarction: the randomized multicentre CASTEMI study Eur. Heart J., November 1, 2006; 27(21): 2516 - 2523. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Wang, P. Crisostomo, G. M. Wairiuko, and D. R. Meldrum Estrogen receptor-{alpha} mediates acute myocardial protection in females Am J Physiol Heart Circ Physiol, June 1, 2006; 290(6): H2204 - H2209. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
O. Gateau-Roesch, L. Argaud, and M. Ovize Mitochondrial permeability transition pore and postconditioning Cardiovasc Res, May 1, 2006; 70(2): 264 - 273. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Philipp, X.-M. Yang, L. Cui, A. M. Davis, J. M. Downey, and M. V. Cohen Postconditioning protects rabbit hearts through a protein kinase C-adenosine A2b receptor cascade Cardiovasc Res, May 1, 2006; 70(2): 308 - 314. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. P. Loukogeorgakis, A. T. Panagiotidou, D. M. Yellon, J. E. Deanfield, and R. J. MacAllister Postconditioning Protects Against Endothelial Ischemia-Reperfusion Injury in the Human Forearm Circulation, February 21, 2006; 113(7): 1015 - 1019. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Banz, O. M. Hess, S. C. Robson, D. Mettler, P. Meier, A. Haeberli, E. Csizmadia, E. Y. Korchagina, N. V. Bovin, and R. Rieben Locally targeted cytoprotection with dextran sulfate attenuates experimental porcine myocardial ischaemia/reperfusion injury Eur. Heart J., November 1, 2005; 26(21): 2334 - 2343. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Kitagawa, T. Yamazaki, T. Akiyama, M. Sugimachi, K. Sunagawa, and H. Mori Microdialysis separately monitors myocardial interstitial myoglobin during ischemia and reperfusion Am J Physiol Heart Circ Physiol, August 1, 2005; 289(2): H924 - H930. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Tsang, D. J. Hausenloy, and D. M. Yellon Myocardial postconditioning: reperfusion injury revisited Am J Physiol Heart Circ Physiol, July 1, 2005; 289(1): H2 - H7. [Full Text] [PDF]
|
|
|
|
 |
|
 X.-M. Yang, J. B. Proctor, L. Cui, T. Krieg, J. M. Downey, and M. V. Cohen Multiple, brief coronary occlusions during early reperfusion protect rabbit hearts by targeting cell signaling pathways J. Am. Coll. Cardiol., September 1, 2004; 44(5): 1103 - 1110. [Abstract] [Full Text] [PDF]
|
|
|