Download Print - Circulation

730
In Vivo Relation of Intramyocardial Blood
Volume to Myocardial Perfusion
Evidence Supporting Microvascular Site for Autoregulation
Xue-si Wu, MD; Daniel L. Ewert, PhD; Yun-He Liu, MD; and Erik L. Ritman, MD, PhD
Downloaded from http://circ.ahajournals.org/ by guest on June 11, 2017
Background. The goal of this study was to explore the role of several factors that affect
intramyocardial blood volume by using minimally invasive computed tomography. Anesthetized
dogs were scanned with the dynamic spatial reconstructor, a high-speed tomographic scanner,
during injection of a bolus of iohexol into the aortic root.
Methods and Results. In control dogs, it is indicated that the fraction of myocardium that is
blood (FMB, %) relates to myocardial perfusion (F, milliliters per gram per minute) in that
region as FMB--a *F.2, where a=9.5± 1.2% (milliliters times minute per gram) '2 (mean± SD)
in the subendocardium and a=9.6+1.1% in the subepicardium. In another group of dogs, for
the myocardium perfused by a stenosed epicardial artery, a increased to approximately 10 for
a 25-43% stenosis (or pressure gradient of 9 mm Hg across narrowing) and to greater than 11
for a 50-55% stenosis (or pressure gradient of 40 mm Hg across narrowing). In these dogs, flow
was not impaired under control hemodynamic conditions, but the usual increase of flow (i.e.,
flow reserve) observed under maximum vasodilation conditions was impaired. In another group
of dogs, progressive embolization (using 15-,um-diameter microspheres) of the left ventricular
myocardial microcirculation caused the value of a to remain at approximately 9.5 with
embolization up to 501% of the fatal dose of microspheres, but it then decreased progressively
with embolization to 4.6 at the fatal dose.
Conclusions. We conclude that the FMB/F relation reflects hemodynamic conductance at the
microvascular level. (Circulation 1992;85:730-737)
ethods for estimating intramyocardial blood
volume (expressed as fraction of myocardium that is blood, FMB) are either quite
invasive and/or indirect'; hence, it is questionable
whether those estimates are representative of the
FMB throughout the in vivo heart within a neveropened thorax. Whether FMB reflects microvascular
behavior within the intact subject could provide insight into pathophysiological mechanisms affecting the
microcirculation. As demonstrated by Kanatsuka et
al,2 autoregulation occurs primarily at the < 100-,umdiameter arteriolar level. Hence, arteriolar vasodilation, combined with capillary recruitment,3,4 should
manifest as increased intramyocardial blood volume.
In this study, we evaluate fast computed tomography
M
From Beijing Anzhen Hospital (X.-S.W.), Beijing, China; the
Department of Electrical and Electronics Engineering (D.L.E.),
North Dakota State University, Fargo, N.D.; and the Department
of Physiology and Biophysics (E.L.R., Y.-H.L.), Mayo Medical
School, Rochester, Minn.
Supported in part by research grants HL-04664 and HL-43025,
National Institutes of Health, Bethesda, Md.
Address for correspondence: Erik L. Ritman, MD, PhD, Mayo
Medical School, 200 First Street SW, Rochester, MN 55905.
Received March 1, 1991; revision accepted September 17, 1991.
(CT) as a relatively noninvasive method for estimating
FMB by comparing the magnitudes of estimated intramyocardial blood volume and their changes in
response to hemodynamic manipulations against the
literature values.
We performed this evaluation by changing coronary flow by progressively altering left ventricular
work load, by locally constricting the cross-sectional
area of a proximal epicardial coronary artery lumen,
or by controlled embolization of terminal arterioles
with 15 -,m- or 16.5 -am-diameter microspheres. The
hypothesis underlying these studies is that maintenance of local myocardial blood flow results from an
ischemia-driven hyperemia downstream to a partial
obstruction to blood flow (e.g., stenosis or microembolization). Consequently, epicardial coronary stenosis should result in increased FMB in its perfusion
territory, whereas microvascular embolization should
result in no change in FMB until more than the
number of microvessels needed to maintain the perfusion status quo are embolized.
Methods
Experimental Sequence
A total of 48 dogs in five experimental groups were
anesthetized with Innovar Vet (0.4 mg fentanyl and
Wu et al Intramyocardial Blood Volume by Fast CT
Downloaded from http://circ.ahajournals.org/ by guest on June 11, 2017
20 mg droperidol per milliliter) plus a 2:1 N20-02
gas mixture. The dogs were ventilated with a Harvard
pump. All had a No. 8 Rodriques catheter position
with its tip in the aortic root for injection of approximately 30 ml iohexol contrast agent. Of these five
groups of dogs, groups 2 and 3 were part of other
studies by Chung et a15 and Spyra et al,6 respectively.
A bipolar catheter was positioned with its tip in the
coronary sinus to permit atrial pacing (generally 112
or 120 beats per minute). An additional Rodriques
catheter was positioned with its tip in the right atrium
for injection of 30 ml iohexol contrast agent. A 6F
Millar catheter with a pressure transducer at its tip
was positioned in the left ventricular cavity for monitoring left ventricular (LV) pressure.
In group 1 (four dogs, weight, 21-26 kg), a balloontipped catheter was positioned in the proximal descending intrathoracic aorta to permit, by balloon
inflation, rapid increase of hemodynamic afterload.
Infusion of adenosine into the right atrium via the
pacing catheter lumen (1 mg/kg/min) was used to
cause coronary arterial bed vasodilation while the
inflated aortic balloon maintained aortic pressure at
the control level.
In group 2, 20 dogs (weight, 22-28 kg) were
prepared identically to those in group 1. An additional balloon-tipped catheter was positioned in the
inferior vena cava to permit, by balloon inflation,
rapid decrease of hemodynamic preload.
In group 3, 11 dogs (weight, 22-27 kg) had 2-3mm-diameter hollow plastic cylinders embolized, via
catheter and guide wire, into a selected proximal
epicardial coronary artery.7 The sizes of the cylinders' lumina resulted in stenoses ranging from 25%
to 55% of the vessel lumen diameter proximal to the
731
was administered (in seven of the dogs), angiotensin
II could be infused (0.01-0.1 ug/kg/min) to maintain
aortic pressure at control levels.
After catheterization of the animals was completed, each dog was placed supine on a molded
cast and positioned in the dynamic spatial reconstructor (DSR) scanner so that the heart was centered in the imaging field.
cylinder.
DSR Scan
Details of the DSR scanner have been described
elsewhere.9 Briefly, the DSR scanner consists of a
rotatable cylindrical gantry housing 14 x-ray sources
positioned in a semicircle. Fourteen video cameras are
arranged on the semicircle opposite the x-ray tubes.
These cameras record the x-ray projection images on a
hemicylindrical fluorescent screen. The projection images, each recorded as 120 horizontal video scan lines
on a multichannel video disc recording system, provide input for tomographic images of parallel transverse sections at 0.9-mm intervals over the full cephalocaudal extent of the heart. The analog video signal
is digitized and the digital information for the same
line of all 14 cameras is simultaneously submitted to a
reconstruction algorithm. Computation results in an
image of a transverse section at the level of the
digitized video lines covering the region of interest.
The entire LV is imaged in three dimensions by the
stack of parallel transverse sections reconstructed in
this manner. Each DSR scan involved sequential
pulsing of 14 x-ray sources within 11 msec, and this
sequence of exposures was repeated 60 times per
second over the 10- or 20-second duration of the scan
sequence. X-ray voltage was set at 110 kV, with x-ray
current ranging from 500 mA to 800 mA, depending
on the size of the dog.
In group 4, three dogs (weight, 21-23 kg) had a 4F
balloon-tipped catheter positioned in either the proximal left anterior descending coronary artery (LAD)
or the left circumflex coronary artery (LCx). The
balloon was partially inflated to produce a selected
pressure gradient between the aortic root and catheter tip distal to the intracoronary balloon.
In group 5, 10 dogs (weight, 21-32 kg) had a 4F
catheter tip positioned in the left main coronary
artery, proximal LAD, or LCx. Nonradioactive microspheres were injected via this catheter to progressively embolize the myocardial territory perfused by
the catheterized vessel. In these dogs, the atrioventricular conduction tissue was previously selectively
destroyed with a local injection of -1 ml 34%
Formalin under fluoroscopic monitoring by using the
needle assembly developed for the percutaneous
technique described by Williams et al.8 An additional
pacing catheter was positioned in the right ventricle
and paced independently from the right atrial pacing
catheter so that, if needed, atrial pacing could be
performed at twice the ventricular rate to maintain
atrial contractions in phase with that of the ventricles. In this group, an additional aortic catheter was
positioned so that, after hexamethonium (30 mg/kg)
Experimental Protocol
In group 1, all dogs were infused with adenosine (1
ml/kg/min in all four; two also had 0.5 ml/kg/min,
making a total of six evaluations) during injection of
a thorium oxide (one dog) or ethiodol (three dogs)
emulsion into the aortic root. The thorium oxide
(Electron Microscopy Sciences, Fort Washington,
Pa.) and ethiodol (Savage Laboratories, Melville,
N.Y.) contrast agents were emulsified in saline by use
of a sonication device (Dynatech Laboratories, Inc.,
Chantilly, Va.): Two milliliters per kilogram were
injected over a 3-second period.
In group 2, four study sets were performed. Each
set involved an aortic root injection of nonionic
contrast agent (iohexol, 1 ml/kg) for the estimation of
myocardial perfusion and intramyocardial blood volume. The levophase of a right atrial injection of
contrast agent was used to provide the LV chamber
volume throughout the cardiac cycle and myocardial
wall delineation at end diastole, the same timing used
for making the myocardial perfusion estimates.
These studies were performed under control, elevated aortic pressure, decreased inferior vena caval
inflow, and adenosine infusion conditions.
732
Circulation Vol 85, No 2 February 1992
Contrast~injection
e!
o-rtc
:
ROP 1
Se
Myoar&t RQ
4
tA
i.
FIGURE 1. Illustration of densitometnc method for generation of indicator
dilution curves from fast computed tomography (CT) images showing regions
of interest (ROI) over aortic root and
myocardium used to estimate the image
brightness values from which indicator
dilution curve is generated. (Reproduced with permission by E. L. Ritman,
Mayo Clin Proc 1990;65:1336-1349.)
...
Downloaded from http://circ.ahajournals.org/ by guest on June 11, 2017
In group 3, a study sequence much as indicated for
group 2 was performed with a hollow plastic cylinder
in situ under control and under adenosine infusion
conditions.
In group 4, the study sequences were performed
under control conditions and sequential conditions of
increasing intracoronary balloon inflation.
In group 5, the study sequences were performed
under control, adenosine infusion, repeat control,
and a sequence of bolus injections of microspheres
until ventricular fibrillation occurred. For seven of
the 10 dogs, 16.5 -,m-diameter microspheres (Du
Pont/New England) were used. In the other three
dogs, the first injection into the coronary artery was
with the 16.5-gm microspheres, and subsequent injections used 15 -gm latex microspheres (E-Z Trac,
Los Angeles, Calif.). The microspheres were prepared according to manufacturers' instructions and
were injected slowly into the coronary catheter. Typically, an average of 8.5 million microspheres were
injected at a time. Immediately after the injection,
the DSR scan was performed. New quantities of
microspheres were prepared and administered at
30-45 -minute intervals. This process was repeated
until the heart arrested.
DSR Data Analysis
Details of estimation of myocardial blood flow by
analyzing DSR images of the aortic root angiograms
were described by Wang et al.1"1 The images generated by each of the 14 television cameras on the DSR
scanner were reconstructed into the images of transverse sections (-5-6 mm thick) through the aortic
root and at the midventricular level of the LV. The
image reconstructions were performed at the enddiastolic phase of all heart cycles throughout the
20-second scanning period. Dye dilution curves of
the aortic root and of the myocardium were then
generated from the sequence of DSR images (Figure
1). These curves represent the appearance and departure of the contrast agent in the region of the
image being measured. Any region (global, transmural, and small pieces) within the image could be
selected for measurement. These curves were used to
calculate myocardial perfusion (F) in milliliters per
gram per minute and FMB expressed as percent of
myocardial volume.
Nonionic contrast agent has less impact on coronary flow primarily because it has lower osmolality
than ionic contrast agents.1' Nonetheless, like ionic
contrast agent, up to 15% of the contrast agent enters
the extravascular space on the first pass.12 Consequently, the opacification of the myocardium does
not return to baseline immediately after the bolus
passes through the myocardial circulation.'3 The area
under the curve, therefore, would be overestimated.
To overcome this problem, we used a mathematical
technique developed by Bentley et a114 to correct for
this phenomenon in the kidney. To evaluate the
validity of this mathematical correction in myocardial
perfusion, we studied four dogs (group 1) so that we
could compare this mathematically corrected analysis
of an iohexol curve with a curve generated from
images recorded during injection of a thorium oxide
suspension or ethiodol emulsion, which do not enter
the extravascular space. This involved two or three
injections of contrast agent into the aortic root at
30-minute intervals under very similar hemodynamic
conditions. The mathematical technique generates a
sigmoidal curve (dash and dot line in middle panel of
Figure 2) describing the progressive accumulation of
extravascular dye by integrating the arterial input
curve (not shown in Figure 2) and scaling the plateau
of the resulting sigmoidal curve to equal the value of
the original myocardial curve (dashed in middle
panel of Figure 2) when it settles at a stable, abovebaseline value. This sigmoidal curve is then subtracted from the original myocardial curve to gener-
Wu et al Intramyocardial Blood Volume by Fast CT
733
space is illustrated in Figure 2. When the original
(upper panel) was used to compute FMB, a
value of 26.8% was obtained. If the computed curve
(middle panel) is used, an FMB value of 20.2%
results. This latter result compares closely with the
value of 20.0% obtained from the thorium oxide
Nonionic Contrast Agent
curve
(0)
CJ)
suspension curve (lower panel).
The results from the four dogs (group 1) studied to
evaluate the reproducibility and accuracy of this
approach for estimating FMB are summarized by two
regression equations describing the linear leastsquares best fit to the totally intravascular (ethiodol
or thorium, Y%) and partially extravascular (iohexol,
X%) contrast agent-based estimates of FMB. When
the fact that some iohexol remains in the extravascular space is ignored, the regression equation is
Y=0.74X+5.26, r=0.951, whereas, when the Bentley
method is used to correct for that extravascular
component, Y=0.98X+3.70, r=0.946. The intercept
3.70 is not distinguishable from zero at the level of
p=0.05. Because of the near 1:1 slope of the Bentley
method, all our dilution curves were analyzed using
this mathematical correction for loss of contrast to
the extravascular space.
In group 2, the aortogram image data were used to
estimate the transmural intramyocardial blood volume and flow under different hemodynamic loading
situations without abnormal coronary circulation. As
shown in Figure 3, it was observed that the intramyocardial blood volume relates curvilinearly to myocardial perfusion in that region. The curve FMB=a . F`12
fits the data points fairly well at flows higher than 0.8
ml/g/min. The magnitudes of the average (+SD) a
were 9.5+1.2% and 9.6+1.1% (milliliters times minutes per gram)'/2 for the subendocardial third and the
subepicardial third of the wall, respectively. Although other analytic curves may have a better
statistical fit to all the data points, we use the square
root relation for reasons detailed in the "Discussion." That heart rate impacts on the transmural
Curve (O-C)
(k
:t
FMB 20.2%
Original
Curve
11
1
(0) ,5t
Cm
\\\X
l
Computed
Extravascular
N
\
" Curve (C)
Thorium Oxide Suspension
Downloaded from http://circ.ahajournals.org/ by guest on June 11, 2017
FMB 20.0 %
Original
Curve
TIME
FIGURE 2. Schemes of contrast dilution curves generated by
passage ofthorium oxide suspension and by iohexol injected to
aortic root for calculation of intramyocardial blood volume
(FMB) and flow. Upper panel derived from contrast of
iohexol, which partially enters the extravascular space; lower
panel fiom thorium oxide, which remains intravascular.
ate the intravascular curve (solid line in middle panel
of Figure 2), which is the curve of interest for our
FMB and F calculations.
Results
One example of the mathematical technique used
to correct for loss of iohexol to the extravascular
30
Subepicordium
30
Subendocardium
*
_
°20
*
20
*
4D
0
P- 9.5 x
112
C)
o
to
I
O
2
4
6
8
0
2
4
6
8
FLOW (mL/g/min)
FIGURE 3. Scatterplots show curvilinear relations between intramyocardial blood volume (FMB) and flow (F) in subendocardium
and subepicardium. Solid curves are the least-squares best fit of formula FMB =a *F"P2 to the experimental data points.
Circulation Vol 85, No 2 February 1992
734
12
P<.05
12
12
l*
#
12
*
.It
E
.E to
E
IN.
i
-4
1O
to
,c
10,1
8 IE
8
-4
11..
10
_
8_
N
3 Meon ± SD
_
C
NS
IL
6 E
k
6 _
4
4 _
IN
* =
** =
2_
r|
T
i-
I
2
9
1
*
a
0
0
2
r-- I
*- 4 e *P<0*0
Downloaded from http://circ.ahajournals.org/ by guest on June 11, 2017
0
25-43
50-55
CORONARY DIAMETER STENOSIS (%)
myocardial blood volume and perfusion distribution
was illustrated in one anesthetized dog. In the control (resting) state, the endocardial/epicardial ratios
of F and FMB at 72 beats per minute were 1.6 and
1.5, respectively, and fell to 0.94 and 0.92 at a heart
rate of 112 beats per minute. In maximum vasodilation with adenosine, the endocardial/epicardial ratios
of F and FMB at 72 beats per minute were 0.86 and
0.89, respectively, and fell to 0.81 and 0.85 when
heart rate increased to 112 beats per minute.
In group 3, each region of interest was identified as
one of three categories: a control region, a region
perfused by a 25-43%, or by a 50-55% stenosed
artery. The average myocardial F and FMB of each
group is shown in Figure 4. The flow was not decreased under control conditions; however, the
FMB/F relation increased significantly in territories
perfused by stenosed vessels with values of a=
9.1+0.7 for no stenoses, 10.1+0.9 for 25-43%
stenoses, and 11.1+0.9 for 50-55% stenoses, respectively. There is a significant difference (p<0.01)
between any two values.
In group 4, flow started to decrease (see Figure 5)
at a pressure gradient across the coronary narrowing
(due to partial inflation of intracoronary balloon)
greater than 40 mm Hg, but the value of a was
already significantly (p<0.01) elevated when the
pressure gradient was greater than 9 mm Hg.
0
PcO.OOt5
0
Ui
FIGURE 4. Graph shows myocardial perfusion characteristics from dogs with/without intracoronary hollow plug, which
show the role of coronaty stenosis. The value of a was
obtained from the least-squares best fit of formula FMB=
a *Fl2 (FMB, intramyocardial blood volume; F, flow) to the
experimental data points for all dogs' regions perfused by the
indicated percent stenosis. The value a (i.e., FMB at flow =1
mlIg/min) increased gradually but significantly in the temritory
perfused by a coronary artery with 24-43% or 50 -55%
stenosis.
.1
p<0.o01
NS
.- l-.I
P<O.05
P<o.of
0
-4
42..
0
0
6 -4IE
Compared toAP=0
1au
-L_
9
40
85
112
A P ACROSS CORONARY
NARROWING (mm Hg)
FIGURE 5. Graph shows myocardial perfusion characteristics in dogs with an intracoronary balloon. In resting state, the
value at a (intramyocardial blood volume at flow=] ml/g/
min) increased significantly at moderate narrowing (pressure
gradient, AP, across balloon -40 mm Hg), whereas flow was
impaired only at more severe narrowing (AP -85 mm Hg).
However, flow relative to myocardium perfused by a nonstenosed vessel decreased significantly at moderate narrowing
with maximum vasodilation (not shown). FMB, intramyocardial blood volume.
In group 5, progressive embolization of the myocardial microcirculation obtained by sequential coronary
injections of microspheres induced the modulation of
FMB and F as shown in Figure 6. Both FMB and F
12
NS
NS
12
Pc0.05 PcO.05 PcO.001
~~~~~~~~1
i Mwn±FSD
10
E
4-.c_
Ik.
,c
6
to
,
E
Lat.
4 °
ik
06
NS
IT
1UL
NS
P<aOOO
NS
P'O.OO1
2
L~~
a
0
1-25 26-50 51-75 76-9
100
PERCENT OF FATAL#&SPHERE DOSE
'
M%
FIGURE 6. Graph shows effect of microsphere embolization
on modulation of intramyocardial blood volume (FMB) and
perfusion. Both the value of a (FMB at F=] ml/g/min) and
flow (F) decreased gradually but significantly in the teritory
perfused by a vessel with more than 50% offatal microsphere
dose.
Wu et al Intramyocardial Blood Volume by Fast CT
20
1.4
0
0
t
15
0
0.0
*
0
S
(0 -50% of FatoPIDose)
,
JO _
*
>,<y=9.6=9.x"t1
0
1U
P<0a01
7fx "2 P'0O05
~~~y=
(75-99%)
1X
*
5
92
0
4.6x"
yz
P<.001
(100%)
U
I
0
05
.
I
1.0
*
.
2.0
1.5
FLOW (mL/g/min)
2.5
Downloaded from http://circ.ahajournals.org/ by guest on June 11, 2017
FIGURE 7. Plot shows that intramyocardial btlood volume
(FMB) to flow (F) relation remained unch tanged, with
FMB=9.6 FPI2, until more than 50% of a fi;atal dose of
microspheres was embolized into the coronary ci, rculation. As
the fatal dose was approached, the relation's ,value of a
decreased progressively.
decreased significantly in the range of 51-7 5% of fatal
dose of microspheres. However, as shown in Figure 7,
the FMB to F relation in the 0-50% of fatal dose group
remained indistinguishable from that in gr(Dup 2.
Discussion
Previous studies confirmed that the myocardial
perfusion estimated by CT relates linearily to perfusion estimated by the generally accepted radioactive
microsphere method.10"15,16 We are quite (encouraged
about our dye dilution-based estimates o Ifintramyocardial blood content because our data arev consistent
with values in the literature,1"17-19 which range from
5% to 27% with most values ranging fr( )m 10% to
20% (several authors used ml/100 g LV)). That our
fast CT estimates of FMB and F are comsistent with
those presented by investigators using other methods
is illustrated in Table 1.
Although we have no direct head-to-he;ad comparison of the CT-based method and other generally
more invasive methods, an additional cir(cumstantial
confirmation of the validity (or at least nneaning) of
the FMB is that it is consistent with Poisezuille's law.
This law predicts that vascular flow (F) is proportional to AP R4/L, where R is the radius of a vessel
735
lumen and AP is the pressure gradient along vessel of
length L. If we assume that the volume (V) of the
intramyocardial vessels is proportional to LR' and if
L remains essentially unchanged with vasomotion
and recruitment, it then follows that F oAP* V2.
Thus, if AP remains essentially constant (which we
attempted to do by maintaining aortic pressure at
control levels with the aortic balloon inflation), we
expect Vo F"', which, if we equate FMB with V, is
what we observe.
A decrease of vascular resistance in acutely ischemic myocardium is well documented. A possible
mechanism is described by the adenosine hypotheSiS,20 which states that ischemia results in accumulation of adenosine, which in turn leads to vasodilation,
which would in turn tend to reverse ischemia because
of the resulting increase in perfusion. This hypothesis
is based
in
part
sure
the assumption that normal blood flow is
maintained by the resulting increased
gradient
increased
by
downstream
across
the stenosis. This gradient is
dropping
to
the
intra-arterial
the stenosis by further opening of the
vascular beds. This opening of the vascular beds
should manifest as increase in intramyocardial blood
volume to a level higher than expected for the flow.
Our data, therefore, support the hypothesis that
distal to an epicardial coronary artery stenosis, there
should be an increase in blood volume reflecting the
vasodilation. With stenosis causing a pressure drop
across the stenosis greater than 40 mm Hg, we observed a slight progressive decrease in FMB an
observation that is consistent with the observations of
Kanatsuka et al,2 who found that despite dilation of
the <100-,um vessels during ischemia, the larger
vessels constricted.
The data from this present study are consistent
with the flow reserve phenomenon, but, as indicated
in Figures 4 and 5, also suggest that the FMB-to-flow
relation is a possibly more sensitive index of the
hemodynamic impact of a mild coronary artery stenosis than is the flow reserve index.21 Thus, for a
stenosis less than 55% or generating a pressure
gradient less than 40 mm Hg, there is no discernible
change in blood flow, but at a stenosis of 25% or
generating a pressure gradient of 9 mm Hg, there is a
discernible increase in FMB.
The effect of microsphere embolization indicates
that absolute value of FMB decreased during injection of a cumulative microsphere dose equal to and
TABLE 1. Intramyocardial Blood Volume and Percent of Myocardial Volume
Comment
Reference
Value (%)
Animal
Radioactive tagging of RBC and/or plasma
25-27
7-26
Rat
Radioactive tagging of RBC and/or plasma
8-10
Swine
28,29
Radioactive tagging of RBC and/or plasma
6-10
Dog
30,31
Mean indicator transit time
11-17.8
Dog
1,32
24
Ultrasonic measurement of wall thickness
10-20
Dog
Left ventricular hypertrophy
33
27
Dog
RBC, red blood cells.
-
736
Circulation Vol 85, No 2 February 1992
greater than 50% fatal dose, beyond which dose the
FMB (under control conditions) progressively diminishes. We observed that the value of a was 4.6 just
before cardiac arrest. This could mean that, under
control hemodynamic conditions, approximately half
(i.e., 4.6/9.5) of control FMB is available when the
heart ceases to function. The number of microspheres required to arrest the heart (2.5 x 105/g)
matches those needed (5 x 105/g) by Hori et al,22 who
injected 15-,um microspheres selectively into the
LAD of open-chest dogs that were treated with
prazosin, a postsynaptic a-blocker, but aortic pressure was not maintained at control levels.
Downloaded from http://circ.ahajournals.org/ by guest on June 11, 2017
Summary
The magnitude of the FMB, estimated by fast CT,
behaves quantitatively like the invasively measured
microvascular volume as reported in the literature.
Second, this volume has a predictable relation to
myocardial perfusion. Because this relation is independent of the location within the heart wall, the
observed spatial heterogeneity of blood flow is most
likely due to the heterogeneity of blood volume. Local
blood volume appears to be strongly driven by local
perfusion needs as illustrated by increase of blood
volume beyond a coronary stenosis and by the steady
blood volume up to a 50% fatal-dose embolization of
the myocardial microcirculation. This observation is
quite consistent with the recent direct observations3,23
of epicardial arterioles that suggest that the bulk of
autoregulation occurs at the microvascular level.
Acknowledgments
The authors wish to thank Don Erdman and his
colleagues for their help with the preparation of the
animals and DSR scans. Christine Welch made the
illustrations, and Marissa Miller typed the manuscript.
References
1. Morgenstern C, Holjes U, Arnold G, Lochner W: The influence of coronary pressure and coronary flow on intracoronary
blood volume and geometry of the left ventricle. Pflugers Arch
1973;340:101-111
2. Kanatsuka H, Lamping KG, Eastham CL, Marcus ML: Heterogeneous changes in epimyocardial microvascular size during graded coronary stenosis. Circ Res 1990;60:389-396
3. Rose CP, Goresky CA: Vasomotor control of capillary transit
time heterogeneity in the canine circulation. Circ Res 1976;39:
541-554
4. King RB, Bassingthwaighte JB, Hales JPS, Rowell LB: Stability of heterogeneity of myocardial blood flow in normal awake
baboons. Circ Res 1985;57:285-295
5. Chung N, Wu X, Bailey KR, Ritman EL: LV pressure-volume
area and oxygen consumption: Evaluation in intact dog by fast
CT. Am J Physiol 1990;258(Heart Circ Physiol 27):H1208H1215
6. Spyra WJT, Bell MR, Bahn RC, Zinsmeister AR, Ritman EL,
Bove AA: Detection of mild coronary stenoses using the
dynamic spatial reconstructor. Invest Radiol 1990;25:472-479
7. Gewirtz H, Most AS: Production of a critical coronary arterial
stenosis in closed chest laboratory animals. Am J Cardiol
1981;47:589-596
8. Williams JC, Lambert EH, Titus JL: Use of intracardial A-V
nodal potentials in producing complete heart block in dogs.
JAppi Physiol 1969;27:740-744
9. Ritman EL, Robb RA, Harris LD: Imaging Physiological
Functions: Experience With the Dynamic Spatial Reconstructor.
Philadelphia, Praeger, 1985, p 302
10. Wang T, Wu X-S, Chung N, Ritman EL: Myocardial blood
flow estimated by synchronous, multislice, high-speed computed tomography. IEEE Trans Med Imaging 1989;8:70-77
11. Higgins CB: Cardiotolerance of lohexol, survey of experimental evidence. Invest Radiol 1985;20(suppl):S65-S69
12. Whitaker JW, Tancredi RG, Yipintsoi T, Donald D: Myocardial transport of diatrizoate meglumine (abstract). Physiologist
1980;23:6
13. Canty JM Jr, Judd RM, Brody AS, Klocke FJ: Extravascular
contrast loss significantly overestimates fast CT measurements
of intramyocardial blood volume (abstract). Circulation 1990;
82(suppl III):III-125
14. Bentley MD, Lerman LO, Hoffman EA, Fiksen-Olsen JJ,
Ritman EL, Romero JC: In vivo measurement of renal blood
flow (RBF) with dynamic CT (abstract). Physiologist 1989;32:
208
15. Gould RG, Lipton MJ, McNamara MT, Sievers RE, Koshold
S, Higgins CB: Measurement of regional myocardial blood
flow in dogs by ultrafast CT. Invest Radiol 1988;23:348-353
16. Rumberger JA, Feiring AJ, Lipton MJ, Higgins CB, Ell SR,
Marcus ML: Use of ultrafast computed tomography to quantitate regional myocardial perfusion: A preliminary report. J
Am Coll Cardiol 1987;9:59-69
17. Feigl EO: Coronary physiology. Physiol Rev 1983;63:1-2053
18. Crystal GJ, Downey HF, Bashour FA: Small vessel and total
coronary blood volume during intracoronary adenosine. Am J
Physiol 1981;241(Heart Circ Physiol 10):H194-H201
19. Salisbury PF, Cross CE, Rieben PA: Physiological factors
influencing coronary blood volume in isolated dog hearts. Am
JPhysiol 1961;200:633-636
20. Berne RM: The role of adenosine in the regulation of coronary blood flow. Circ Res 1980;47:808-813
21. McGinn AL, White CW, Wilson FR: Interstudy variability of
coronary flow reserve: Influence of heart rate, arterial pressure, and ventricular preload. Circulation 1990;81:1319-1330
22. Hori M, Tamai J, Kitakaze M, Ivalwa K, Gotoh K, Iwai K,
Koretsune Y, Kagiya T, Kitababahe A, Kamada T: Adenosineinduced hyperemia attenuates myocardial ischemia in coronary microembolization in dogs. Am J Physiol 1989;257(Heart
Circ Physiol 16):H244-H251
23. Dellsperger KC, Jahzen JL, Eastham CL, Marcus ML: Effects
of acute coronary artery occlusion on the coronary microcirculation. Am J Physiol 1990;259(Heart Circ Physiol 28):
H909-H916
24. Gaasch WH, Bing OHL, Franklin A, Rhodes 0, Bernard SA,
Weintraub RM: The influence of acute alterations in coronary
blood flow on left ventricular diastolic compliance and wall
thickness. Eur J Cardiol 1981;7(suppl):147-161
25. Everett NB, Simmons B, Lasher EP: Distribution of blood
(Fe59) and plasma (1131) volumes of rats determined by liquid
nitrogen freezing. Circ Res 1956;4:419-424
26. Gibson JGH, Seligman AM, Peacock WC, Aub JC, Fine J,
Evans RD: The distribution of red cells and plasma in large
and minute vessels of the normal dog, determined by radioactive isotopes of iron and iodine. J Clin Invest 1946;25:
848-857
27. Rakusan K, Wachtlova M, Poupa 0: An attempt to determine
indirectly the vascularity of heart muscle by measuring the
tissue concentration of haemoglobin in normal and anaemic
rats. Physiol Bohemoslov 1969;18:1-5
28. Howe BB, Winbury MM: Effect of acute coronary occlusion
on myocardial nutritional circulation in swine. Am J Physiol
1969;216:1069-1074
29. Winbury MM, Losada M, Kessel D, Howe BB, Pensinger RR:
Pentaethyritoltetranitrate and dipyridamole on cardiac nutritive flow and blood content. Am J Physiol 1971;220:1558-1563
30. Howe BB, Windbury MM: Effect of pentrinitrol, nitroglycerin
and propanolol on small-vessel blood content of canine myocardium. J Phannacol Exp Ther 1973;187:465-474
Wu et al Intramyocardial Blood Volume by Fast CT
31. Weiss HR, Windbury MM: Nitroglycerin and chromonar on
small-vessel blood content of the ventricular walls. Am J
Physiol 1974;226:838-843
32. Hirsch EF, Lochner W: Messung der Durchblutung und der
Blutftillung des coronaren Gefassbettes mit der Testoffinjection Methode am narkotisierten Hund bei geschlossem Thorax. Pflugers Arch 1962;274:624-632
737
33. O'Keefe DO, Hoffman JIE, Cheitlin R, O'Neill MJ, Allard
JR, Shapkin E: Coronary blood flow in experimental canine
left ventricular hypertrophy. Circ Res 1978;49:342-353
KEY WORDS * coronary stenosis *
embolization * fast CT
microcirculation
.
Downloaded from http://circ.ahajournals.org/ by guest on June 11, 2017
In vivo relation of intramyocardial blood volume to myocardial perfusion. Evidence
supporting microvascular site for autoregulation.
X S Wu, D L Ewert, Y H Liu and E L Ritman
Downloaded from http://circ.ahajournals.org/ by guest on June 11, 2017
Circulation. 1992;85:730-737
doi: 10.1161/01.CIR.85.2.730
Circulation is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 1992 American Heart Association, Inc. All rights reserved.
Print ISSN: 0009-7322. Online ISSN: 1524-4539
The online version of this article, along with updated information and services, is located on
the World Wide Web at:
http://circ.ahajournals.org/content/85/2/730
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally
published in Circulation can be obtained via RightsLink, a service of the Copyright Clearance Center, not the
Editorial Office. Once the online version of the published article for which permission is being requested is
located, click Request Permissions in the middle column of the Web page under Services. Further
information about this process is available in the Permissions and Rights Question and Answer document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Circulation is online at:
http://circ.ahajournals.org//subscriptions/