From the Institut für Herz- und Kreislaufphysiologie (R.L.) and the
Institut für Klinische Anästhesiologie (C.W.F.),
Heinrich-Heine-Universität Düsseldorf, and the Institut für
Physiologie, Medizinische Fakultät Carl Gustav Carus der Technischen
Universität Dresden (A.D.), Germany.
Correspondence to Dr Andreas Deussen, Professor and Chairman, Institut für Physiologie, Medizinische Fakultät Carl Gustav Carus der Technischen Universität Dresden, Fetscherstr 74, D-01307 Dresden, Germany.
Methods and Results—In anesthetized beagle dogs (n=10),
the relationship between local blood flow versus
S-adenosylhomocysteine (SAH) concentration, a
measure of the free intracellular adenosine concentration, and
lactate, a measure of the myocardial NADH/NAD+ ratio, were
determined under control conditions and after coronary
constriction. Control local myocardial blood flow was 0.99±0.46
mL · min-1 · g-1, with a
coefficient of variation of 0.36±0.12 (n=256 per heart; sample wet
mass, 125±30 mg). Tissue concentrations of SAH (3.4±2.5 nmol/g) and
lactate (1.88±0.80 µmol/g) were not elevated in low-flow
samples. However, after coronary artery constriction,
poststenotic blood flow decreased from 1.00±0.27 to 0.49±0.22
mL · min-1 · g-1
(P<0.04), with significant correlation between local
SAH and flow (r=-0.59) and lactate and flow
(r=-0.50). Although nearly all samples from control
high-flow regions showed increased SAH concentrations if relative flow
after stenosis was <1.0, control low-flow samples frequently
displayed low SAH concentrations. The percent reduction in flow
determined the changes in the local SAH and lactate concentration,
independent of the local control blood flow.
Conclusions—When the coronary inflow is unrestricted, the
oxygen supply to control low-flow regions meets metabolic
demand. Flow to control high-flow regions reflects a higher local
demand rather than overperfusion. Thus, blood flow
heterogeneity most likely reflects differences in
aerobic metabolism.
Although control low-flow areas with absolute flow rates <0.4 mL
· min-1 · g-1 do
not show elevated adenosine
concentrations,5 local blood flows <0.6 mL
· min-1 · g-1
during epicardial coronary artery constriction are associated
with a reduction of myocardial contractile
function14 and a change of energy
metabolism.15 16 Thus, the question
arises whether for a given global cardiac work, an absolute local flow
threshold can be defined below which the myocardium must be
considered underperfused. Alternatively, according to the concept of
supply-demand balance, local flow is only one determinant of myocardial
oxygenation, which also depends on the local
metabolic demand for oxygen. This balance could possibly be
achieved over a wide range of local blood flow rates if the local
oxygen requirement is adjusted in proportion to blood flow.
Impairment of myocardial oxygenation is sensitively
reflected by an increased formation of degradation products of
adenine nucleotides, eg, adenosine, inosine, and
hypoxanthine,17 indicative of a fall in the
myocardial phosphorylation potential and an increase in
the free AMP concentration (for review, see Reference 1818 ). The
S-adenosylhomocysteine (SAH)
technique15 19 permits the local assessment of
the free cytosolic adenosine concentration in cardiac tissue.
Another index of the extent of ischemia, myocardial lactate
production, reflects changes of the cytosolic
(NADH+H+)/NAD+
ratio.20 Limitations in the availability of
oxygen decrease the mitochondrial utilization of
NADH+H+ in the electron transport chain,
resulting in an increased conversion of pyruvate to
lactate.16 21
Using measurements of SAH and lactate concentrations along with blood
flow determinations in myocardial samples of 125 mg wet mass, the
present study tested the following hypotheses: (1) There is an
absolute local flow threshold under resting control conditions below
which a sample must be considered underperfused. (2) High-flow samples
are in a condition of luxury perfusion, ie, they tolerate a greater
relative reduction of blood flow than low-flow samples.
Materials
In six experiments, the relationship between regional myocardial blood
flow and concentrations of SAH and lactate during epicardial
coronary artery stenosis was studied. A first tracer
microsphere batch was injected, and arterial
PO2 and oxygen content were
determined under resting control conditions. Some 10 minutes after the
first tracer microsphere injection, a small side branch
(<1 mm in diameter) of the left circumflex coronary
artery was cannulated with a thin vinyl catheter for measurement of
distal coronary artery pressure. A constrictor device and a
Doppler flow probe (T206, Transonic Systems Inc) were placed around
the left circumflex artery proximal to the cannulated side branch.
After a second regional myocardial blood flow measurement had been
obtained (some 20 minutes after the first measurement), the constrictor
was narrowed to lower distal coronary artery pressure to
Analytical Procedures
Statistics
Local SAH and Lactate Under Physiological Conditions
Figure 3
Local SAH and Lactate During Coronary Constriction
To address the question of whether control low- and high-flow areas
differ in their coronary reserve, samples were selected that
exhibited a normalized flow of <0.5 (low-flow samples) or >1.5
(high-flow samples) under resting physiological
conditions. Those criteria identified 73 low-flow and 67 high-flow
samples. Production of a stenosis reduced blood flow in
the low- and high-flow class to similar extents, namely, 62% in low-
and 61% in high-flow samples. Initially, we analyzed whether
the control high-flow samples (flow before stenosis >1.5 times
mean flow) had a different flow threshold than low-flow samples (flow
before stenosis <0.5 times mean flow) to increase the SAH
concentration (Figure 6
The observed coefficient of variation of local blood flow as determined
in the present study (CVobs=0.36) is similar
to that reported recently.5 Further
analysis of the observed coefficient of variation in this
previous study indicated that the contributions of the methodological
variability and the temporal variability to the observed coefficient of
variation were small, which was largely accounted for by the effects of
a heterogeneous spatial blood flow distribution. The sample
mass in the present study was 34% higher than that in the previous
study, whereas the number of microspheres injected and the mean
myocardial blood flow were comparable in both studies. Hence, these
considerations can also be applied to the present study. Taken
together, measurement of the local microsphere deposition in
the present study was most likely a reliable index for spatial
blood flow distribution.
Another point to be addressed concerns the question to what extent
local myocardial blood flow reflects the local supply of oxygen, which
is also dependent on local hematocrit. Analyzing the local volumes of
distribution for the capillary and the plasma region in relation to
local myocardial blood flow, Gonzalez and
Bassingthwaighte24 observed a 20% higher
hematocrit in low- than in high-flow regions. Thus, the lower oxygen
supply of low-flow areas as determined by the flow variable may be
partly compensated for by a higher local hematocrit. With data shown in
their Figure 7
To assess the presence of critically oxygenated
areas, the present study used two methods: measurement of the
accumulation of SAH in the presence of homocysteine infusion and
measurement of the local lactate concentration. As described in detail
previously,15 19 25 the SAH technique permits the
sensitive detection of local differences in the free cytosolic
adenosine concentration. Because other determinants of the SAH
accumulation, ie, homocysteine concentration and SAH hydrolase
activity, do not differ significantly between high- and low-flow
myocardial areas,23 differences in the local SAH
accumulation depend largely on differences in the cytosolic
adenosine concentration. Although the SAH concentration is not
augmented in control low-flow samples, it increases dramatically in
samples supplied with a low blood flow after constriction of an
epicardial coronary artery (Figure 4
In contrast to SAH, which integrates the free cytosolic
adenosine signal over time,19 measurement
of the local lactate and pyruvate concentrations reflects the local
(NADH+H+)/NAD+
ratio21 at the moment of tissue freezing. During
a lack of oxygen availability, the NADH produced is not effectively
utilized by the electron transport chain, and hence, it accumulates.
This is expected to drive the equilibrium reaction from pyruvate to
lactate until a new steady state is obtained.20
Thus, in theory, the local relationship between lactate and pyruvate
mirrors the (NADH+H+)/NAD+
ratio. Because the pyruvate concentration is normally much smaller than
the lactate concentration (by a factor of
The present study used normalized flow values for ease of
comparison. Because physiological control
conditions were compared with conditions of regional ischemia,
it is important to critically examine the flow ranges for both
experimental conditions. Under physiological
control conditions, mean blood flow averaged 1.08 mL ·
min-1 · g-1, with
a range from 0.10 to 3.51 mL · min-1
· g-1 (n=4). Under conditions of a
coronary stenosis, mean blood flow of the entire tissue
block was 0.65 mL · min-1 ·
g-1, with a range from 0.01 to 3.15 mL ·
min-1 · g-1 (n=6).
Although mean blood flow was lower during conditions of
coronary constriction, it is important to note that the
absolute flow ranges overlapped in the range from 0.10 to 3.15 mL
· min-1 · g-1.
Concerning a flow threshold below which local concentrations of
adenosine and lactate are changed, it can be said that such a
threshold was not found under control physiological
conditions, that is, above a local flow of 0.10 mL ·
min-1 · g-1, which
was the lowest control blood flow measured in samples of 125 mg wet
mass. Because of the fractal nature of spatial blood flow
distribution,7 such a flow threshold is probably
sample size–dependent. This means that a low absolute flow, if
present in a contiguous large myocardial volume (eg, distal to a
coronary stenosis), may indicate myocardial
underperfusion, whereas the same flow present in smaller myocardial
volumes may be an expression of physiological blood
flow heterogeneity and, as the present study
demonstrates, may not be associated with ischemia. Hence, to
use local flow measurements for pinpointing ischemia, the
low-flow myocardial volume needs to be taken into account.
Several studies have shown that local myocardial
oxygenation may be heterogeneous. Under
physiological conditions, the coronary
venous oxygen saturation exhibited a coefficient of variation of
On the basis of the local determination of SAH and lactate
concentrations along with blood flow measurements, the present
study sheds first light on the coronary reserve of control
high-flow regions. To characterize areas as low- or high-flow,
microsphere measurements were performed under resting control
conditions. For the metabolite analysis, samples were selected
that showed a more than threefold flow difference according to this
flow measurement (<0.5 times mean flow versus >1.5 times mean flow).
It was found that after coronary constriction, SAH and lactate
concentrations in both groups were significantly related to the degree
of flow reduction induced by coronary constriction (Table 2
Previous investigators attempting to study the effects of
myocardial ischemia have almost exclusively used experimental
models in which coronary blood flow was acutely lowered or
completely interrupted. This approach has greatly fostered the view
that myocardial blood flow is the prime factor controlling myocardial
ischemia. However, it has recently been noted that variations
in tissue oxygen consumption may be more effective in adjusting tissue
PO2 than changes of blood flow or
oxygen content.31 The concept that matching of
these variables determines myocardial energy balance has led to the
theory of supply-to-demand relationship for
oxygen.32 An important result of the present
study is that as long as the coronary vessels were left
"untouched" by the investigator, the local oxygen supply matched
the oxygen demand, as evidenced by the lack of significantly increased
concentrations of SAH or lactate although local flow may have been
surprisingly low (lowest local flow, 10% of mean). The important
question that remains unanswered at present is how this local
matching of flow and metabolism is achieved. There are
several modes by which flow and metabolism may interact on
the local level. The local supply of blood may in part determine the
steady-state oxygen consumption of a myocardial region. Conversely,
local oxygen demand may control myocardial blood flow. Other factors
may influence local flow and metabolism in parallel and
thereby help to maintain a proper balance. These interactions are
clearly not sufficiently understood to date.
Received September 12, 1997;
revision received January 27, 1998;
accepted February 4, 1998.
© 1998 American Heart Association, Inc.
Basic Science Reports
Coronary Reserve of High- and Low-Flow Regions in the Dog Heart Left Ventricle
Abstract
Top
Abstract
Introduction
Methods
Results
Discussion
References
Background—Left
ventricular myocardial blood flow is spatially
heterogeneous. The hypothesis we tested was whether
myocardial areas with a steady-state flow <0.5 times mean flow are
underperfused and areas with flow >1.5 times mean flow are
overperfused.
Key Words: blood flow • homocysteine • microspheres • perfusion
Introduction
Top
Abstract
Introduction
Methods
Results
Discussion
References
The mean resting
myocardial blood flow in larger mammals (dog, pig, goat, sheep, human)
is consistently 0.6 to 1.0 mL ·
min-1 · g-1 (for
review, see Reference 11 ). However, when analysis of the local
deposition density of microspheres2 3 4 5
and diffusible plasma flow markers6 7 8 in volume
elements of a few millimeters diameter showed that the local flow
distribution within individual hearts is markedly
heterogeneous, covering a range from 20% to 200% of the
mean value. This flow heterogeneity is remarkably
stable over periods of several hours,4 even if
the heart is intermittently stimulated.9 10 The
physiological significance of blood flow
heterogeneity in left ventricular
myocardium results from the high average myocardial oxygen
extraction, which is 
70% under resting
physiological conditions.11
Thus, areas with a local oxygen supply of <50% of the average cannot
have a normal oxidative metabolism. In fact, measurements
of the local uptake of deoxyglucose5 12 and fatty
acids13 indicate that local metabolic
rates and transport capacities in left ventricular
myocardium mirror those of local blood flow, ie, high-flow
areas have a higher metabolic rate of glucose and a higher
uptake of fatty acids than low-flow regions. This indicates that
control high-flow areas are regions of high metabolism and
consequently have a higher demand for oxygen. However, it is still
unclear whether the full myocardial blood flow of high-flow areas is
required to meet higher local oxidative needs or whether high flow in
such regions represents a state of luxury perfusion. It seems
important to differentiate between these two possibilities to
understand the effects of local blood flow reductions, as may occur
during coronary stenosis, on local tissue
metabolism. If high-flow areas were in a state of luxury
perfusion, their oxygen extraction might be small, and it might be
expected that a flow reduction could be compensated over a wider flow
range in contrast to the same situation in low-flow areas. Conversely,
if high-flow areas have a higher requirement for oxygen, their
coronary reserve may even be smaller than that of low-flow
areas.
Methods
Top
Abstract
Introduction
Methods
Results
Discussion
References
Animal Preparation and Blood Flow Measurements
Beagle dogs (n=10) weighing 12 to 16 kg were
anesthetized with piritramide (3 mg/kg IV) and midazolam (2
mg/kg IV), relaxed with vecuronium bromide (0.2 mg/kg IV), intubated,
and artificially ventilated. Anesthesia and analgesia were
maintained by continuous infusions (0.6 mg ·
kg-1 · h-1
piritramide, 0.6 mg · kg-1 ·
h-1 midazolam IV). Additional vecuronium bromide
was given at a dose of 0.12 mg/kg IV every 30 to 60 minutes. Blood gas
and acid-base parameters as well as body temperature were
maintained within physiological limits. The blood
pressure signal from the femoral artery and the surface standard ECG
were continuously recorded on a polygraph. After thoracotomy
through the fourth or fifth intercostal space and pericardiotomy, the
left atrium was cannulated via the auricle with a 15-cm-long, 1.5-mm-OD
flexible PVC tubing advanced 3 cm. Regional myocardial blood flow
was determined with the tracer microsphere technique using
microspheres (New England Nuclear/Du Pont de Nemours) with a
diameter of 15.2±0.1 µm labeled with
46Sc, 85Sr,
113Sn, or 153Gd suspended
in 10% dextran and supplemented with 0.01% Tween 80 as described
previously.5 10 The microsphere solutions
were dispersed by repetitive vortexing and continuous sonication for at
least 20 minutes. Approximately 6x106 to
9x106 microspheres per injection were
aspirated in a 5-mL syringe, mixed with 3 mL atrial blood, and
immediately infused over 30 seconds followed within 30 seconds by a
flush of 2 mL warm saline. "Reference organ" samples were withdrawn
from the aortic catheter at a rate of 10 mL/min beginning 30 seconds
before the start and ending 2 minutes after the end of the tracer
injection. The whole reference blood sample was distributed into test
tubes and prepared for counting. The individual tracer radioactivity in
each tissue sample, along with microsphere standards and
blanks, was determined in a gamma counter (WIZARD 1480, Wallac Oy) at
window settings of 84 to 104 keV (153Gd), 430 to
470 keV (113Sn), 490 to 530 keV
(85Sr), and 850 to 1000 keV
(46Sc). Deposition densities of the different
isotopes were corrected for spectral overlap with MultiCalc software
(version 1.84, Wallac).
In four experiments, the relationship between regional
myocardial blood flow and concentrations of SAH or lactate under
physiological control conditions was studied. After
a first batch of tracer microspheres had been injected and
arterial PO2, oxygen
content, and homocysteine concentration measured, homocysteine
(DL-homocysteine thiolactone, Sigma) was given at a total
dose of 48 mg/kg IV, of which two thirds was administered within 5
minutes and one third during the next 25 minutes. Tracer
microspheres with a different label were injected during
homocysteine infusion (15 to 20 minutes). The arterial
homocysteine plasma concentration was determined after 1, 3, 10, 15,
20, and 25 minutes; the arterial oxygen content and
PO2 were measured after 10 and 20
minutes of homocysteine infusion. After 30 minutes of homocysteine
infusion, the free left ventricular wall was rapidly
excised with sharp scissors and frozen between two copper ingots (5 and
7.5 kg) precooled to the temperature of liquid nitrogen. 40 mm Hg. After the distal coronary pressure had been
in a new steady state for 5 minutes, arterial
PO2, oxygen content, and plasma
homocysteine concentration were measured, and a third tracer
microsphere injection was performed, followed by a homocysteine
infusion protocol identical to that described above. A final tracer
microsphere injection was made after 20 minutes of homocysteine
infusion. The experiments were terminated after 30 minutes of
homocysteine infusion by excision of the left ventricular
free wall as described above.
According to previously published standardized methods, the
myocardial tissue was freeze-dried and dissected into 256 samples with
an average dry mass of 25 mg.22 For
measurement of metabolite concentrations (SAH, ATP, ADP, homocysteine,
lactate, pyruvate), acid extracts were
prepared.15 19 An aliquot of the acidic
supernatant was used for later measurement of the homocysteine
concentration, and the remaining aliquot was neutralized for the
analysis of the other metabolites. Extracts were stored at
-70°C until metabolite analysis was performed. For plasma
homocysteine measurement, blood samples were centrifuged at
800g (20 minutes, 4°C) within 15 minutes after withdrawal.
Plasma (2 mL) was deproteinized with 0.1 mL of perchloric acid (60%)
and stored at -70°C. Samples were neutralized with 1 mol/L
K3PO4 immediately before
analysis. Homocysteine concentrations were measured with a
combined enzymatic–high-performance liquid
chromatography (HPLC) method as described
before.23 This in vitro method converts
homocysteine in the presence of a saturating concentration of
adenosine to SAH after addition of SAH hydrolase. The SAH
formed is then quantified by HPLC. The test was optimized for sample
homocysteine concentrations between 1 and 1000 µmol/L. Sample
concentrations of SAH, ATP, and ADP were determined by sensitive HPLC
techniques.15 19 Tissue concentrations of lactate
and pyruvate were determined with standard enzymatic methods (lactate,
Boehringer Mannheim, and pyruvate, Sigma
Diagnostic).
Data in this study are given as mean±SD. To facilitate an
interindividual comparison of myocardial blood flow data, relative flow
was calculated for individual myocardial samples from each experiment
by dividing blood flow of each sample by the average flow of the
particular microsphere injection. SAH and lactate
concentrations were normalized to the mean value of the respective
heart. Differences between experimental groups were tested with
Mann-Whitney U test/Wilcoxon rank sum W
test. Effects of coronary constriction on mean
poststenotic myocardial blood flow and coefficient of variation
were assessed by Wilcoxon matched-pairs signed rank test
(two-sided). Regression analyses were based on individual
measurements using Spearman's rank correlation coefficient. In
addition, nonlinear regression analyses were compared with the
rank correlation analysis to obtain information about the model
that gives the optimal fit. Statistical analyses were performed
with SPSS for Windows, version 6.0.1. A value of P<0.05
(two-sided) was assumed to indicate a significant difference.
Results
Top
Abstract
Introduction
Methods
Results
Discussion
References
In the 10 dogs studied, control heart rate, mean aortic pressure,
and mean arterial PO2 and
PCO2 were 82±15 bpm, 95±16
mm Hg, 103±18 mm Hg, and 36±5 mm Hg, respectively.
Arterial oxygen content was 16.2±2.1 mL/100 mL blood, and
hematocrit was 0.38±0.04. The coronary venous
PO2 under
physiological control conditions was 29±5
mm Hg, equivalent to an average oxygen extraction of 0.62±0.04 (n=4).
Neither homocysteine infusion nor regional underperfusion had any
significant effect on global hemodynamic or blood gas
parameters. Mean myocardial blood flow was 0.99±0.46
mL · min-1 ·
g-1, and the mean coefficient of variation of
local blood flow was 0.36±0.12 under resting control conditions. Local
myocardial blood flow measured before and during homocysteine infusion
(time interval, 30 minutes) correlated linearly
(r=0.86±0.03), as shown in Figure 1
, top. Because this correlation
coefficient is indistinguishable from that obtained for
simultaneous injection of differently labeled tracer
microspheres under comparable experimental conditions
(0.87±0.03),23 this indicates that the local
myocardial blood flow was temporally stable under our experimental
conditions. The control plasma homocysteine concentration was
5.1±0.5 µmol/L. Infusion of homocysteine at a rate of 6.4
mg · kg-1 ·
min-1 for 5 minutes and 0.64 mg ·
kg-1 · min-1 after
5 minutes increased the plasma concentration to 526±70 and
490±71 µmol/L after 10 and 25 minutes, respectively
(P=NS, 25 versus 5 minutes).
View larger version (17K):
[in a new window]
Figure 1. Plot of repetitive local flow measurements using
tracer microsphere technique in samples with a mean wet mass of
125 mg. Top, Flow measurements in the absence of coronary
stenosis before and during homocysteine (HCY) infusion. Middle,
Flow under control conditions vs flow during constriction of left
circumflex coronary artery (stenosis). Bottom, Flow
after 5 minutes (no homocysteine) vs 20 minutes of coronary
stenosis (homocysteine). Data at top are from 4 and data at
middle and bottom from 6 experiments. Blood flows are normalized to
mean value of individual tracer microsphere injection.
The relationship between the normalized SAH concentration and
normalized local blood flow is shown in Figure 2. In these experiments (n=4), in which
homocysteine was infused under resting control conditions, the average
myocardial blood flow was 1.08±0.69 mL ·
min-1 · g-1 and
the average SAH concentration 3.4±2.5 nmol/g after 30 minutes of
homocysteine. Local myocardial blood flow ranged from 20% to 200% of
the mean in the different experiments, and local SAH exhibited a
similar degree of heterogeneity. A weak, albeit
significant positive relationship existed between normalized myocardial
blood flow and the SAH concentration (r=0.30, n=214,
P<0.001). Results from the individual experiments are
summarized in Table 1. Most notably, not
a single experiment showed higher than average SAH concentrations in
low-flow areas. Local ATP and ADP concentrations did not differ between
low- and high-flow samples. The average ATP concentration was 2.7±0.2
(SEM) µmol/g and the average ADP concentration was 0.80±0.15
(SEM) µmol/g in 35 samples from the four hearts selected evenly
over the entire flow range. The ATP/ADP ratio was 3.5±0.2 and not
significantly related to local myocardial blood flow
(r=0.04).
View larger version (19K):
[in a new window]
Figure 2. Relationship between local SAH and myocardial
blood flow after 30 minutes of homocysteine infusion under control
physiological conditions. Data are from 4
experiments, indicated by different symbols. Blood flow was normalized
to mean value of respective experiment.
View this table:
[in a new window]
Table 1. Relationship Between SAH and Lactate Versus Local
Blood Flow Under Physiological Conditions shows the relationship between
the normalized lactate concentration and normalized local blood flow.
The average lactate concentration was 1.88±0.80 µmol/g. Like
local SAH (Figure 2
), the lactate concentration exhibited a
considerable heterogeneity. There was no significant
relationship between normalized blood flow and lactate
(r=0.09, n=164, P>0.05). Table 1 shows the
results from the individual experiments. The average pyruvate
concentration was 0.015±0.008 µmol/g. There was no significant
relationship between local blood flow and pyruvate (r=0.08,
n=99, P>0.05). The relationship between local pyruvate and
lactate was weak and did not reach the level of significance
(r=0.22, n=99, P>0.05).
View larger version (18K):
[in a new window]
Figure 3. Relationship between local lactate and myocardial
blood flow under control physiological conditions.
Data are from 4 experiments, indicated by different symbols. Blood flow
was normalized to mean value of respective experiment.
Blood flow in the poststenotic myocardium was
1.00±0.27 mL · min-1 ·
g-1 under resting control conditions and fell to
0.49±0.22 mL · min-1 ·
g-1 (P<0.04) on reduction of distal
coronary artery pressure to 38±5 mm Hg. Mean myocardial
blood flow of the entire tissue block analyzed was 0.94±0.29
mL · min-1 ·
g-1 before and 0.65±0.23 mL ·
min-1 · g-1 after
constriction of the circumflex coronary artery
(P=NS). This relative insensitivity of average blood flow
was because the left ventricular myocardium
included in the analysis contained circumflex as well as left
anterior descending coronary artery perfused regions. Although
flow of the poststenotic (circumflex) region fell, that in
adjacent areas frequently increased. This is illustrated in the middle
panel of Figure 1. Local flow measurements before and after
coronary stenosis correlated weakly (r=0.19,
P<0.05). In contrast, blood flow measurements in the
presence of persisting stenosis (Figure 1, bottom) correlated
well (r=0.91, P<0.0001) before versus during
homocysteine infusion. As a result of regional myocardial
underperfusion, the coefficient of variation of local flow increased
from 0.41±0.11 to 0.77±0.37 (before versus after stenosis,
P<0.02). During coronary constriction, there was an
inverse relationship between flow and SAH (Figure 4) and between flow and lactate (Figure 5), that is, low-flow samples exhibited
steeply increased concentrations of SAH and lactate. The relationships
between the metabolite concentrations and flow could be approximated
with exponential functions. The overall correlation coefficients were
-0.59 for SAH versus flow and -0.50 for lactate versus flow (both
P<0.0001). The results obtained under conditions of
coronary constriction differ qualitatively as well as
quantitatively from those obtained during resting control conditions
(Figures 2 and 3). Replotting of the data obtained under control
conditions (right panels of Figures 4 and 5) with the same ordinate
scaling as used to plot results from underperfusion experiments shows
that local SAH and lactate concentrations were low in absolute terms
over the entire flow range during physiological
conditions.
View larger version (18K):
[in a new window]
Figure 4. Relationship between local SAH concentration and
myocardial blood flow during regional underperfusion (left) and during
control physiological conditions (right). Data for
myocardial underperfusion are from 6 experiments with a total of 379
SAH measurements. Blood flow is normalized to mean blood flow of
individual experiments during coronary artery constriction.
Data for control conditions are taken from Figure 2 . Different symbols
indicate different experiments.
View larger version (17K):
[in a new window]
Figure 5. Relationship between local lactate concentration
and myocardial blood flow during regional underperfusion (left) and
during control physiological conditions (right).
Data for myocardial underperfusion are from 5 experiments with a total
of 228 lactate measurements. Blood flow is normalized to mean blood
flow of individual experiments during coronary artery
constriction. Data for control conditions are taken from Figure 3 .
Different symbols indicate different experiments.). The cutoff for
an increase of the SAH concentration was set to the maximal SAH
concentration determined in the four control experiments (12.9 nmol/g).
In nearly all control high-flow samples with a relative flow >1.0 at
the time of coronary stenosis, the SAH concentration
was below this cutoff concentration of 12.9 nmol/g. However, at
normalized flows <1.0, 85% of all samples in the high-flow group
showed increased SAH concentrations. At a normalized flow <0.5, all
samples showed increased SAH concentrations. These results clearly
differ from those obtained for control low-flow samples: nearly all
low-flow samples with relative flow >0.5 during coronary
constriction had normal SAH concentrations. This indicates that control
high-flow regions have a higher flow threshold than control low-flow
samples to increase their SAH concentration. In a further step of the
analysis, the SAH and lactate concentrations determined were
analyzed with respect to the associated flow reduction. Data in
Table 2 show that an exponential model
gave optimal fits for these relationships. When plotted as a function
of the relative flow reduction, the distributions of SAH and lactate
concentrations were similar for the two sample groups (Figure 7). If measurements were grouped into
four ranges of flow change, no statistically significant differences
were observed within these groups between control low- and high-flow
samples (Table 3). This suggested that
local underperfusion occurs to similar degrees of relative flow
reduction, independent of the control blood flow.
View larger version (17K):
[in a new window]
Figure 6. Plot of local SAH concentration as a function of
relative flow during coronary constriction. For selection
criteria of low- and high-flow samples see text. Horizontal dashed line
indicates cutoff between control and increased SAH concentrations.
Cutoff was set to 12.9 nmol/g SAH, the highest SAH concentration
determined in control experiments (Figure 2 ). Vertical dashed lines
indicate approximate threshold for an increase of SAH concentration.
Note different abscissa scaling for low- and high-flow samples.
View this table:
[in a new window]
Table 2. Regression Analysis of SAH and Lactate
Concentrations Versus Local Blood Flow Reduction
View larger version (17K):
[in a new window]
Figure 7. Plot of SAH concentration vs percent flow change
(top) and lactate concentration vs percent flow change (bottom).
Percent flow change is calculated from absolute local flow obtained
after coronary constriction relative to absolute local flow
before coronary constriction. Low-flow ( 
) and high-flow
(
) samples were selected on the basis of flow determinations
before coronary constriction. Statistical results from these
analyses are summarized in Tables 2 and 3.
View this table:
[in a new window]
Table 3. Comparison of SAH and Lactate Concentrations of
Control Low- and High-Flow Regions With Respect to Different Ranges of
Flow Change After Coronary Artery Constriction
Discussion
Top
Abstract
Introduction
Methods
Results
Discussion
References
The present study addresses the importance of local blood flow
heterogeneity for the occurrence of ischemic or
hypoxic islands within the left ventricular free wall.
Before the implications of the new results are discussed, it seems
appropriate to address the underlying assumptions. Because
microspheres have different physical properties (size, specific
gravity, rigidity) compared with red blood cells, it was necessary to
show that microsphere trapping in samples of 125 mg mass can be
taken as a valid estimate of local myocardial blood flow. The
reproducibility of the local flow measurements is evidenced by the
results shown in Figure 1. If the hemodynamic
conditions remained unchanged, repetitive injections of differently
labeled microspheres gave local deposition densities that
correlated closely (r=0.80 to 0.94). Most notably, areas
with deposition densities of <0.5 times or >1.5 times the mean after
the first microsphere injection exhibited similarly low or high
deposition densities, respectively, after the second
microsphere injection. Using samples with a wet mass as low as
54 mg, Bassingthwaighte and colleagues6 7 8 showed
that the spatial blood flow distribution in myocardium as
determined with the tracer microsphere technique is
quantitatively similar to that obtained with the molecular plasma flow
marker iododesmethylimipramine over a flow range from 0.20 to 1.50
mL · min-1 ·
g-1. Because in the present study, low- and
high-flow samples were selected on the basis of two independent
microsphere measurements, it seems safe to conclude that blood
flow of low-flow areas was greatly different from that of high-flow
areas (low flow times 3 
high flow). , which gives hematocrit values of 0.44 and 0.35 for low-
and high-flow regions, respectively (relative flows, 0.5 and 1.5,
respectively), it is calculated that despite the difference in
hematocrit, the oxygen supply of high-flow areas exceeds that of
low-flow areas by 140%. The question addressed in the present
study was whether this higher local oxygen supply reflected a higher
local oxygen demand or whether these high-flow areas should be
considered luxury perfused. ). The latter result, which
is in good agreement with previously reported
data,15 26 reveals the sensitivity of the method
to detect ischemic myocardial areas under the present
experimental conditions. The lack of myocardial areas with SAH
concentrations >12.9 nmol/g under physiological
control conditions (Figure 2) then indicates that myocardial samples
that receive a low steady-state oxygen supply are not underperfused in
the sense of being ischemic or hypoxic. 10) and because the
relative differences of the local pyruvate concentration tend to be
moderate compared with those of lactate (see "Results"),
differences in the lactate concentration are the major factor
reflecting the local redox potential of
(NADH+H+)/NAD+. The lactate
concentration steeply increased in low-flow areas during experimental
coronary artery stenosis (Figure 5, left). However,
like the results obtained with SAH, such an inverse relationship
between lactate and flow was not obtained during resting control
conditions (Figure 5, right). This supports the view that a low local
oxygen supply is not a sufficient requirement to indicate myocardial
ischemia. 0.4.27 Because the coefficient of variation of
arterial PO2 values was
0.08, the large variability of venous
PO2 values in the same experiments
must have been induced by local differences in oxygen extraction.
Myocardial tissue PO2 measurements
have revealed broad PO2 distributions
with a considerable fraction in the range of potentially critical
PO2 values between 0 and 5
mm Hg.28 29 30 Values in this range accounted for
10% to 14%29 and 25%28
of all PO2 measurements. In these
studies, local myocardial blood flows were not determined. Thus, it is
unknown whether low PO2 values
corresponded preferentially to low- or high-flow areas. If one assumes
that locally high SAH or lactate concentrations indicate the presence
of local ischemia or hypoxia, the present study
indicates that low-flow samples are not more prone to be associated
with a critical PO2 than areas that
receive a high blood flow. If we take the highest concentrations of SAH
measured under control conditions to indicate ischemia (eg, SAH
concentrations >200% of average, equivalent to 6% of all
measurements; see Figure 3), then such samples may be found in any flow
range. The same conclusion would be derived if the 25% of the samples
(estimate from Lösse et al28 of probability
of samples with a PO2 of <5
mm Hg) with highest SAH or lactate concentrations were selected. ).
However, there was a clear quantitative difference in the
SAH-versus-flow relationship of control low- and high-flow samples,
respectively, after coronary constriction (Figure 6). During
coronary constriction, high-flow areas that had normalized
flows <1.0 had high SAH concentrations. By contrast, low-flow areas
had elevated SAH concentrations only if normalized flow was <0.5. This
indicates that control high-flow regions have a requirement for high
flow and, therefore, a high oxygen supply. Furthermore, the data
suggest that the oxygen extraction of control high-flow areas is close
to 60%, similar to those of average perfused myocardium,
because a fall of local relative flow of >30% is associated with an
increased SAH concentration (reduction of relative flow from 1.5 to
1.0). As a corollary, the lower flow requirement of control low-flow
samples is then indicated by the rare occurrence of enhanced SAH
concentrations in the relative flow range of 0.5 to 1.0 after
stenosis (Figure 6). These findings strengthen the view that
control high-flow myocardial samples are high-metabolism
samples, as indicated by the results of fatty
acid13 or deoxyglucose uptake
measurements.5 12 This interpretation is
supported by the findings that changes in the local SAH and lactate
concentration were similar for equivalent percent reductions of flow
(Table 3) independent of the control blood flow rate.
Acknowledgments
This work was supported by a grant from the Deutsche
Forschungsgemeinschaft (project C11) within the program project
SFB 242 "Koronare Herzkrankheit. Diagnose und Therapie akuter
Komplikationen" at the Heinrich-Heine-Universität
Düsseldorf. Furthermore, the financial support of the
Igler-Stiftung is gratefully acknowledged. Dr Deussen was a Heisenberg
fellow of the Deutsche Forschungsgemeinschaft. We wish to thank Barbara
Patzer for her expert technical assistance during the course of the
study and Dr John Williams for reading and correcting of our
manuscript.
References
Top
Abstract
Introduction
Methods
Results
Discussion
References
125 mg wet mass were
analyzed for local blood flow and local tissue concentrations
of S-adenosylhomocysteine (SAH), an index of free
cytosolic adenosine, and lactate. Under control conditions,
low-flow samples (normalized flow <0.5) exhibited no increase of SAH
or lactate. However, after coronary constriction, control
high-flow samples (normalized flow >1.5) showed increased SAH
concentrations in a flow range similar to that of control low-flow
samples before stenosis. After coronary constriction,
SAH and lactate increased in proportion to the relative reduction of
local flow, independent of control flow rate. Thus, local myocardial
blood flow mirrors a spatially heterogeneous aerobic
metabolism.
This article has been cited by other articles:
 |
|
 |
|
  M. Stoll, M. Quentin, A. Molojavyi, V. Thamer, and U. K.M. Decking Spatial heterogeneity of myocardial perfusion predicts local potassium channel expression and action potential duration Cardiovasc Res, February 1, 2008; 77(3): 489 - 496. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. P. Johnson and K. L. Gould Clinical Evaluation of a New Concept: Resting Myocardial Perfusion Heterogeneity Quantified by Markovian Analysis of PET Identifies Coronary Microvascular Dysfunction and Early Atherosclerosis in 1,034 Subjects J. Nucl. Med., September 1, 2005; 46(9): 1427 - 1437. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. S. Kirton, A. J. Taberner, P. M. F. Nielsen, A. A. Young, and D. S. Loiselle Strain softening behaviour in nonviable rat right-ventricular trabeculae, in the presence and the absence of butanedione monoxime Exp Physiol, September 1, 2004; 89(5): 593 - 604. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. C. Hughes, M. J. Post, M. Simons, and B. H. Annex Translational Physiology: Porcine models of human coronary artery disease: implications for preclinical trials of therapeutic angiogenesis J Appl Physiol, May 1, 2003; 94(5): 1689 - 1701. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Lauer, R. Loncar, and A. Deussen Tracer Adenosine: A Novel Myocardial Flow Marker J. Nucl. Med., April 1, 2003; 44(4): 641 - 648. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 U. K. M. Decking Spatial Heterogeneity in the Heart: Recent Insights and Open Questions Physiology, December 1, 2002; 17(6): 246 - 250. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Berlin, K. E. Challoner, and R. D. Woodson Low-O2 affinity erythrocytes improve performance of ischemic myocardium J Appl Physiol, March 1, 2002; 92(3): 1267 - 1276. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. F.M van Oosterhout, T. Arts, J. B Bassingthwaighte, R. S Reneman, and F. W Prinzen Relation between local myocardial growth and blood flow during chronic ventricular pacing Cardiovasc Res, March 1, 2002; 53(4): 831 - 840. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Matsumoto, H. Tachibana, Y. Ogasawara, and F. Kajiya New double-tracer digital radiography for analysis of spatial and temporal myocardial flow heterogeneity Am J Physiol Heart Circ Physiol, January 1, 2001; 280(1): H465 - H474. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Sambuceti, M. Marzilli, A. Mari, C. Marini, P. Marzullo, R. Testa, I. Raugei, M. Papini, M. Schluter, and A. L'Abbate Clinical evidence for myocardial derecruitment downstream from severe stenosis: pressure-flow control interaction Am J Physiol Heart Circ Physiol, December 1, 2000; 279(6): H2641 - H2648. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
U. Schwanke, A. Deussen, G. Heusch, and J. D. Schipke Heterogeneity of local myocardial flow and oxidative metabolism Am J Physiol Heart Circ Physiol, September 1, 2000; 279(3): H1029 - H1035. [Abstract] [Full Text] [PDF]
|
|
|
|