Download Salvage of ischemic myocardium by prostacyclin during

lACC Vol 2. No 2
August 1983 279- 86
279
Salvage of Ischemic Myocardium by Prostacyclin During Experimental
Myocardial Infarction
JACQUES A. MELIN, MD,* LEWIS C. BECKER, MD, FACC
Baltimore, Maryland
The effects of prostacyclin (PGI z) on infarct size and
regional myocardial blood flow were studied in 28 anesthetized dogs subjected to 5 hours of coronary occlusion. A region of myocardial hypoperfusion was defined
by injection of dye into the left atrium just before sacrifice. Infarct size was determined by planimetry of left
ventricular slices after incubation in triphenyl tetrazoIium chloride. The animals received either PGI z in Tris
14) or Tris
buffer solution (20 to 40 ng/kg per min, n
buffer alone (control, n = 14) beginning 10minutes after
anterior descending coronary artery occlusion. During
=
Prostacyclin (PGh), the major active metabolite of arachidonic acid in vascular endothelium (1,2), has potent vasodilating properties in addition to other important biologic
effects. Infusion of exogenous PGh has been shown to
favorably modify cellular injury produced by a regional or
global reduction in myocardial blood flow . After coronary
artery ligation, a decrease in enzyme depletion and stabilization of myocardial Iysosomes in ischemic tissue have
been observed in the anesthetized cat (3), and a decrease in
the extent of myocardial necrosis has been shown to occur
in the conscious dog (4). In the latter study, PGI 2 infusion
was associated with an increase in collateral blood fl ow to
the ischemic myocardium , accounting for at least part of its
myocardial salvaging effect. However, some animals demonstrated myocardial salvage despite low levels of collateral
flow , suggesting a direct myocardial or nonfl ow-mediated
effect. In the study of Araki and Lefer (5), ischemic myocardial injury was diminished in the isolated cat heart perFrom the Department of Medicine. Division of Cardiology. The Johns
Hopkins University School of Medicine, Baltimore, Maryland This study
was supported by the Ischemic Heart Disease SCOR Grant 2 P 50 HL
17655 from the National Heart, Lung, and Blood Institute . National Institutes of Health, Bethesda, Maryland. Dr. Melin was a Fogarty International Fellow of the National Institutes of Health Manuscript received
November 2, 1982; revised manuscriptreceived February9. 1983, accepted
February 13, 1983.
*Present address: Cliniques Universitaires Saint-Luc, Avenue HlPpocrate 10, 1200 Brussels, Belgium
Address for reprints: Lewis C. Becker, MD, Johns Hopkins Hospital,
600 North Wolfe Street. Baltimore. Maryland 21205 .
© 1983 by the American College of Cardiology
PGIz infusion, mean arterial pressure decreased by 8% ,
but heart rate was unchanged. Infarct size was significantly less (p<O.OOS) in PGIz-treated dogs compared
with the control group , both as percent of left ventricle
(8.1 versus 17.7%) and as percent of the hypoperfused
zone (39.8 versus 77.3%). No significant changes in regional myocardial blood flow occurred over the 5 hour
infusion period in either group . Thus, under the conditions of this study, prostacyclin appeared to protect
ischemic myocardium by a direct flow-independent
mechanism.
fused at constant flow. In addition to its vasodilatory properties, PGh can prevent or reverse in vivo platelet aggregation
(6- 8) and stabilize membranes (3).
This study was designed to determine whether direct
(nonflow-mediated) effects of PGh are important in contributing to myocardial salvage in the intact dog heart after
coronary artery ligation. We utilized a model in which a
region of myocardial hypoperfusion was defined by injection
of dye into the left atrium just before sacrifice. A reduction
by PGIz in the extent of infarction within the nondyed , or
hypoperfused, myocardial zone was likely to reflect myocardial salvage mediated by direct mechanisms.
Methods
Surgical preparation. Thirty-two mongrel dogs weighing 18 to 30 kg were anesthetized with intravenous alpha
chloralose (1 00 rug/kg) and urethane (I g/kg). Respiration
with roomair was maintained by a Harvard pumpand cuffed
endotracheal tube. The heart was exposed by a left thoracotomy and suspended in a pericardial cradle. The left anterior descending coronary artery was isolated proximally
just beyondits first or second diagonal branch and a ligature
was passed underneath. Polyethylene catheters (8 French)
were placed in the left atrium for injections of microspheres
and dye, and in the common carotid artery for withdrawal
of reference blood samples and the recording of systemic
0735-1097/83/$3 ()()
280
lACC Vol. 2. No.2
August 1983:279-86
MELIN AND BECKER
PROSTACYCLIN IN EXPERIMENTAL INFARCTION
arterial pressure. An additional catheter was placed in the
left jugular vein for administration of drugs. Aortic and left
atrial pressures and heart rate were continuously monitored
on a Gould (Brush 200) recorder.
Experimental protocol. After control hemodynamic and
electrocardiographic measurements, the left anterior descending coronary artery was ligated. Five minutes after
occlusion, 2 X 106 microspheres, 9 ± I JL in diameter,
labeled with scandium-46 (3M Company) were injected into
the left atrium. A reference arterial sample was drawn at a
constant rate of 2.16 mlImin with a calibrated Harvard pump,
beginning 30 seconds before microsphere injection and continuing for 2 minutes after injection. After the microsphere
injection, a continuous infusion of prostacyclin (PGIz) or
drug vehicle (50 roM isotonic Tris hydrochloride buffer
adjusted to a pH of 9.4 and infused at room temperature)
was started into the left jugular vein. Before each experiment, 500 JLg of PGI z was mixed with the Tris buffer solution to a volume of 150 ml. For PGI z, the infusion rate
was adjusted to reduce the mean arterial pressure approximately 10% below the pre-Pfll, (control) level (20 to 40
ng/kg per min). Ifnecessary, the rate was readjusted 2 hours
later to return arterial pressure to the desired level. The
infusion rate for the drug vehicle was constant at 0.15 ml/
min.
Five hours after occlusion, a second injection of microspheres labeled with niobium-95 was made into the left
atrium. After the myocardial blood flow determination,
monastral blue dye, 1 cc/kg (DuPont Company), was infused into the left atrium over a 1 minute period. Immediately afterward, the heart was arrested by injection of
saturated potassium chloride solution.
Postmortem tissue preparation. The hearts were excised and the ventricles were sectioned parallel to the atrioventricular groove, forming five slices I to 1.5 em thick.
After removing the right ventricle, each slice was weighed
and the left ventricular outlines were traced on clear acetate
sheets. The outlines of the hypoperfused region (risk region), indicated by the absence of blue staining, were added
to the tracing. Each slice was then placed in a solution of
triphenyl tetrazolium chloride (ITC) at 37°C for 30 minutes.
The TIC solution consisted of 5 mg/ml TTC, 1 M Sorenson's phosphate buffer and water in a 1:1:8 ratio. TTC is
reduced by dehydrogenase contained in viable cells, staining
noninfarcted myocardium a dark red. Infarcted dehydrogenase-deficient myocardium is a pale grey color and is
clearly demarcated from noninfarcted myocardium (9,10).
The edges of the infarct were drawn on the same transparent
sheets used for the left ventricular and risk region outlines.
The tracings of each ring were electronically measured by
planimetry to determine the areas of left ventricle, infarct
(unstained by TTC) and risk region (not perfused by monastral blue dye). Areas on the top and bottom surfaces were
averaged. Masses of infarcts and risk regions for each ring
were computed by relating their relative areas to the weights
of the rings.
Myocardial blood flow calculation. Sampling for regional myocardial blood flow was made transmurally in the
center of the infarct (area unstained by TTC), in the lateral
risk region (area unstained by monastral blue but stained by
TTC), in the region just lateral to the risk region and in the
nonischemic posterior papillary muscle. The samples were
divided into equal epicardial and endocardial halves, which
were weighed (0.5 to I g), placed in vials and counted with
the reference blood samples for radioactivity in a well-type
gamma scintillation counter (Packard 5986).
Regional myocardial bloodfiow (mllmin per g) was calculated as: counts/g in myocardial samples times (reference
blood flow/counts in the reference blood sample). The mass
of myocardium at risk before treatment (5 minutes after
occlusion) was determined by summing the weights of the
samples having myocardial blood flow less than 50% of the
flow to the nonischemic posterior wall. To assess the accuracy of monastral blue staining as a marker of the hypoperfused or "risk" region, flows were compared in stained
and unstainedsamples taken from sites adjacent to the stainedunstained border. For blue stained samples, both endocardial and epicardial halves were used in the analysis. For
unstained samples, epicardial halves were excluded because
small areas of "contamination" with blue staining were
often visible in the lateral subepicardial portion. Flows were
expressed as a percent of the posterior wall nonischemic
flow.
Statistical analysis. Student's t tests for paired and unpaired data were utilized for statistical comparisons between
groups. Linear regression analysis was performed by the
least-squares fit method. The X axis intercepts and their
standard error were obtained by reentering the data with X
and Y axes reversed. Probability (p) values of 0.05 or less
were considered statistically significant. All results are presented as mean ± standard error of the mean.
Results
Of the 32 dogs that underwent coronary artery occlusion,
4 were excluded because of ventricular fibrillation during
the initial 10 minutes after occlusion. Data in the remaining
28 dogs form the basis for this report: 14 received the drug
vehicle and 14 received prostacyclin (PGI z).
Hemodynamic changes. Heart rate, mean arterial pressure and left atrial pressure were similar before coronary
occlusion in the two groups (Table I). After coronary occlusion, mean arterial pressure decreased and left atrial pressure increased to a similar extent in both groups. During
the remainder of the experiment, the control group showed
no further significant hemodynamic changes. In the group
treated with PGlz, PGlz produced an average 8% decrease
in mean arterial pressure at 10 minutes after treatment (from
l ACC Vol 2. No 2
August 1983 279-86
PROST ACYCLI N
VEHI CLE
25
28 1
MELIN AND BECKER
PROST ACYCLI N IN EXPERIM ENT AL INFARCTI ON
25
IIU JE
UE
1JI
W
oJ
a.
20
20
15
15
10
10
5
5
::I:
0(
1JI
LL
0
a:
w
CD
::I:
~
Z
0
o
25
10 20 30
IOf
0
40 50 60 70 80 90 100 1 10 120
25
a.UE
10 20 30
IOf
40 50 60
70 80 90 100 110 120
a.UE
1JI
w
20
20
1JI 15
15
oJ
a.
::I:
0(
LL
0
a:
10
10
5
5
w
CD
::I:
~
Z
o
0
o
10 20 30 40 50 80
70 80 90 100 110120
MYOCARDI AL BLOOD FLOW
(AS PERCENT OF NO ISCHEMI C FLOW)
Figure l. Histograms of the flow values in blue and nonblue
samples adjacent to the stain border in control and prostacyclin
(PGh)-trea ted groups. The blue stained samples (n == 88 in each
group) are both endocardial and epicardial and the unstained samples (n == 44 in each group) are endocardial only. Flow is expressed
as percent of nonischemic, and groupings are 0 to 10%, II to
20%, 21 to 30% and 3 1 to 40%, and so forth. The data mdicate
that staining technique separates myocardium with flow 50% or
less and greater than 50% of nonischemic with little overlap and
that this fl ow " separation" level is the same m control and PGI!treated groups.
99 to 9 1 mm Hg, p < 0.005). At I hour after treatment,
the mean arterial pressure was reduced by 5% (to 94 mm
Hg, p < 0.05) and at 5 hours by only 2% (to 97 mm Hg).
The mean arterial pressure at 5 hours in the POh group was
5% less than that in the control group (97 versus 102 mm
Hg; NS = not significant). After administration of POll ,
left atrial pressure decreased (p < 0.005) and this change
was maintained during the 5 hours of treatment. Heart rate
did not change significantly. Thus, hemodynamically, the
major difference between control and POh -treated dogs was
a decrease in mean arterial pressure and left atrial pressure
during POh treatment.
Risk region determination. Figure I shows the myocardial blood flow values in the blue and nonblue samples
taken adjacent to the stained border in control and POlztreated groups. A lack of staining identified myocardium
0
10 2 0 30 40 50 80 70 80 90 100 1
10 120
MYOCARD IAL BLOOD FLOW
(AS PERCENT OF ONISC EMIC FLOW)
Table I, Hemodynamic Changes in Open Chest Dogs Treated
With Vehicle (control group) and Prostacyclin (PGh)
Heart Rate
min)
Mean Artenal
Pressure (mm Hg)
Mean Left
Atnal Pressure
(mm IIg)
151 ± 8
105 ± 4
4 ± 0.3
162
162
161
163
97 ±
99 ±
100 ±
102 ±
47
4
5 ± 0.3*
5 ± 0.3
5
3
5 ± 04
6 ± 0.5
149 ± 6
108 ± 4
3 ± 03
153
152
148
154
99 ±
91 ±
94 ±
97 ±
5
4
4
4
(beats/
Vehicle (n= 14)
Preocclusion
Postocclusron
5 nu n
20 mm
Ih
5h
PGl 2 (n= 14)
Prcocc lusion
± 8*
± 8
±
8
± 7
Postocclusion
5 01 111
20 mm
Ih
5h
7
7
± 8
± 6
±
±
4t
4§
4+
5
O.4t
0.4§
± 0.4§
± 04§
±
±
Valuesareexpressedas mean ± standard error of the mean. Statistical
analysis was performed to compare preocclusion With 5 minute postocc1usion values (*p<O 05. t p<0.OO5l. or 5 nunute postocclusron with later
postocc lu sron values rj:p<o OS . §p<0.005) Treatment was begun 10 mi nutes after occlusion.
h = hourts), mi n = minutes, n = number.
282
lACC Vol. 2, No 2
August 1983:279-86
MELIN AND BECKER
PROSTACYCLIN IN EXPERIMENTAL INFARcnON
with flow less than 50 % of the nonischemic value in both
control and PGh groups, with relatively little overlap between stained and unstained sample s. Of 88 stained samples,
2% (2 of 88) in the control group and 6% (5 of 88) in the
group treated with PG h had flows between 41 and 50 % of
the nonischemic flow , and in none was flow 40% or less.
In all of the remaining samples, flow was greater than 50%
of the nonischemic value . The slight difference between
control and PGl z-treated groups was not significant. Of 44
nonblue endocardial samples, 7% (3 of 44 ) in both control
and PGh groups had flows 51 to 60% of nonischemic flow
and none had flows greater than 60% . In all of the remaining
samples flow was 50 % or less of the nonischemic value.
Effect of PGIz on infarct size. Prostacyclin (P0I 2 ) significantly reduced infarct size, both as a percent of the left
ventricle and as a percent of the hypoperfused region (" risk
region ") (Table 2) . A wide range of infarct and risk region
sizes was seen in both groups. There was no significant
difference in risk region mass betwe en control and PGh
groups , although risk region mas s was slightly smaller in
the PGh group (17 versus 2 1 g; t = I. 32; not significant
[NSJ) .
Expressed as a percent of the left vent ricle , the mass of
the infarct was linearly related to the mass of the risk region
in both groups (Fig. 2). The slopes of the linear regres sion
were not different for the two groups (control dogs = 0.85;
PGh = 0.79) but the horizontal intercept, repre senting the
minimal risk region required for infarction , was smaller for
the control group than for the PGh group (6 versus 10%;
p < 0 .05 ).
Regional myocardial blood flow. Regional myocardial
blood flow data were available in 22 of the 28 dogs ( l O in
the control group and 12 in the PGh group). In all dogs ,
there was a gradient of collateral flow from lateral to central
Figure 2. Infarct size was directly related to the size of hyperperfused region ("risk region" ). Data for the prostacyciin (PGIz)treated dogs were shifted downward, indicating a smaller infarct
for given size risk region. PGI2 displaced the horizontal intercept
to the right (p < 0.05), although the slope was unchanged.
"l
..J
50
Z
40
....U
....a:
"l
• PROSTACYCLIN
>
0 .111
0 .115
0 .9'
6 VEHICLE
....
r-
SLOPE
0 .19
c
n
••
••
e
DO'
< .00.
IL.
"l
...J
30
....
Z
UJ
u
a:
20
UJ
Q.
.....
A
til
.
U
....Z
• •
A
a:
lJ..
.
A
10
0
0
10
20
30
40
RISK REGION AS PERCENT LEFT VENTRICLE
50
Table 2. Effect of Prostacyclin on Infarct Size in Open
Chest Dogs
Control Group (vehicle)
(n = 14)
PGIr Treated Group
(n = 14)
Infarct mass (g)
16.6 :!: 2.3
(4.4 to 31.4)
7.6 :!: 1.4*
(0 to 19.3)
Risk region mass (g)
21.2 :t 2.4
(4 4 to 40.3)
17.2 :t 1.8
(7 to 29.9)
92.8 :t 5.1
(57.4 to 137.5)
96.9:!: 3.1
(76.9 to 115.6)
InfarctlLV (%)
17.7 :t 2.2
(5.9 to 33.4)
8.1 :t 1.7*
(0 to 23)
Risk region/LV (g)
22.9 :!: 2.4
(5.9 to 43)
18.3 :!: 2.0
(8 to 31)
77.3 :!: 4. 1
(56.2 to 100)
39.8 :t 4.9*
(0 to 76)
LV mass (g)
Infarctln sk region (%)
*p < 0.005, PGIz versus vehicle.
Values are expressed as mean :!: standard error of the mean; values
in parentheses represent range.
portions of the risk region, and epicardial flows exceeded
endocard ial flows throughout the risk region . Table 3 shows
the time related chan ges in blood flow after coronary occlusion in different zone s of the myocardium. In dogs from
the control group, myocardial blood flow did not change
significantly in any zone or layer between 5 minutes and 5
hours of coronary occlu sion (paired t test). In dog s from
the PGI 2 group, mean flow chan ges were also not statistically significant. Howe ver , in three dogs, there was an increase in collateral flow, which explains the slight increase
in mean collat eral flow seen in the risk region of the PGI 2 treated dogs . Figure 3 shows the changes in subepicardial
collateral flow after 5 hours of treatment in individual dog s.
Only three dogs in the PGI 2 group and one in the vehicle
group had an increase of more than 0.05 ml/min per g .
The lack of a significant mean effect of PGl 2 on collateral
flow was confirmed by the comparable sizes of the ischemic
region 5 minutes after occlusion (before treatment) and ju st
before sacrifice (after treatment ) (Fig. 4). The extent of
initial ischemia was determined by summing the wei ghts of
samples with flow less than 50% of nonischemic flow , while
after treatment the size of the nond yed area was determined
by planimetry. The slope of the linear regression was clo se
to the line of identity and similar for the two groups (control
= 1.07; PGI 2 = 0.96), suggesting that no significant collateral development occurred in the two groups between 5
minutes and 5 hours after occlu sion. If the PGIz-treated
group had shown significant collateral development after
treatment , the slope should have been shifted downward .
The relation between subepicardial collateralflow 5 minutes after occlusion (bef ore treatment) and infarct size is
shown in Figure 5 . Subepicardial collateral flow and infarct
283
MELIN AND BECKER
PROSTACYCLIN IN EXPERIM ENTAL INFARCTION
JACC Vol. 2. No 2
August 1983 279-86
Table 3. Regional Myocardial Blood Flows in Dogs Given Vehicle or PGh
Nonischemic Region
Vehicle (n = 10)
Before treatment
5 hours after occ lusion
PG h (n = 12)
Before treatment
5 hours after occlusion
Infarct Center
Noninfarcted Risk RegIOn
Inner
Outer
Inner
Outer
Inner
Outer
089 ::!: 0 .07
0 .93 ::!: 0 .08
0 .83 ::!: 0.07
0 .87 ::!: 007
03 1 ::!:OO4
0 .32 ::!: 0.03
0 .34 1: 00 5
03 4 ::!: 0 .04
0 .05::!:OO I
0 .04 =: 0 .0 1
0 .1 3 ::!: 0 .02
0 . 12 :!: 0 .02
0 .80 ::!: 0 .08
0 .92 + 0 .09
0 .82 1: 0 . 10
0 .93 ::!: 0. 10
0 .28 ::!: 0 .03
0 .36 ::!: 0 .08
0 .39 1: 0.07
0.4 6 ::!: 0 .09
0 .0 7 ::!: 001
0 .07 ::!: 0 .0 1
0. 13 ::!: 0 .02
0 . 15 ::!: 002
Values for flows are mean z standard error of the mean (mllmin per g)
size were inversely related in control and treated groups
(linear r = - 0.77, - 0.76, respectively) with larger infarcts for lower flows; at flows above 20% of nonischemic
flow, the relation plateaued with infarct size ranging between 5 and 10% of left ventricle . The amount of infarction
associated with a given level of initial collateral flow was
reduced in the PGl z group , as shown by the downward shift
of the curve. The relation between infarct size and posttreatment subepicardial collateral flow 5 hours after occlusion is shown for compari son in Figure 6. Two dogs treated
with PGI z had a relatively high post-treatment subepicardial
flow greater than 0.25 mllmin per g, and had an infarct
involving less than 50% of the nondyed (risk) region. However, the remain ing PGIz-treated and control dogs had comparable flows in the range of 0 .05 to 0. 20 mllmin per g;
despite this , the dogs treated with PGIz clearly tended to
have a smaller infarct, relative to the size of the risk region ,
than did dogs in the control group .
Figure 3. Relation between initial and 5 hour ~post-tr.eatme.nt)
subepicardial collateral flow in the center of the ischemic region
in individual dogs . Flow values are in mllmin per g . Only three
prostacyclin (PGh)-treated dogs and one control dog had increases
of more than 0 .05 mllmin per g.
~
0 .4
/
•
0
..J
•
U.
/
Z
UJ
lI::X
Ill ...
Hol
a:UJ
a:
/
/
/
~..J
/
Z..J
UJO
XU 0 .2
~
ol..J
UJol
a:H
•
/
.
~o
a:
IIlU 0 .1
OH
ll.ll.
UJ
~ol
/
/
.
./ •
/
/
/
a:~
UJ
... Ul
UJUl
0<
I X
UJ
~
96
1.07
n
o 95
o 97
12
10
p
< .001
V
< 001
30
•
•
20
10
>z
/
00
H
0.0
l!l
0 .0
o
UJ~
ZUl
HO
Xll.
.
. /
/
m
• PFlOSTACYCLIN
A VEHICLE
40
O~
...... ):. .
/
/
r
SLoPE
;::
/
•
Figure 4. Relation between dye-determined risk region mass (posttreatment) and flow-determined risk region mass (pretreatment) .
The mass of myocardium at risk before treatment (5 minutes after
occlusion) was determined by summing the weights of the samples
with myocardial blood flow less than 50% of the flow to the
nonischemic myocardium . The slopes of the linear regression were
close to the line of identity and were similar for the two groups
(P = prostacyclin; V = vehicle) . The agreement between pretreatment and post-treatment values for ischemic mass suggests
that no significant collateral development occurred.
/
/
<
Effect of prostacyclin on infarct size. Our results demonstrate that prostacyclin (PGIz) reduces infarct size by about
50% after coronary artery occlu sion in the anesthetized dog .
Infarct size was measured relative to the size of the ischem ic
region, identified by left atrial injection of blue dye after
treatment with PGI2 and ju st before sacrifice of the animal.
The fact that salvage occurred within this hypoperfused zone
strongly suggests that the beneficial effect of PGIz in this
study was mediated by direct cellular. or nonflow effect s
of the agent. Microsphere measurements indicated no significant mean change in the level of collateral flow in the
group treated with PGI z, and topographic measurements
showed no change in the size of the ischemic region as a
result of treatment. Analysis of the relat ion between collateral flow and infarct size indicated that dogs treated with
/
/
~~
III
V£HICL£
A
t:..J
.col 0.3
a:
IOUJ
::l
PAClSTACYCLIN /
Discussion
0.1
0.2
0 .3
0.4
PRETREATMENT
SUBEPICARDIAL COLLATERAL FLOW
UJ
a:
0
0
10
20
30
40
FLOW-DETERMINED RISK REGION MASS (PRETREATMENT)
284
lACC Vol 2. No 2
August 1983 279-86
MELIN AND BECKER
PROSTACYCLIN IN EXPERIMENTAL INFARCTI ON
uJ
.J
u
H
35
ct
~
• PROSTACYCL!N
VEHICLE
30
uJ
A
>
n
12
- 17
- 16
to
D
< 005
< Ot
... 25
II..
w
.J
20
~
15
...z
ct
uJ
Q.
.
...
.
10
Vl
u
ct
II..
Z
5
oL.---'---_L-_---'-_ _. L - _ - ' - -_
o
00
0 10
_
0 20
.L-_
--'
0 30
SUBEPICARDIAL COLLATERAL FLOW AS A RATIO
OF NONISCHEMIC FLOW (PRETREATMENT )
Figure5. Pretreatment subepicardial collateral flow was inversely
related to infarct size in both control and prostacyclin (PGI2 ) treated groups. The downward shift of the relation in the POI 2
group indicates a smaller infarct size for a given level of initial
collateral blood flow.
PGh had a smaller infarct for a given level of either initial
pretreatment or post-treatment flow. However, within the
PGh group, three dogs may have had an increase in fl ow
which contributed to myocardial protection; the two dogs
with the highest flows after treatment (0.27 to 0.35 ml/min
per g) were from the PGI group.
Identification of region at risk for infarction . The
method used in this study for delineation of the region " at
risk" was specifically chosen to allow clear interpretation
of the results. We used the technique described by Kloner's
laboratory (II) of in vivo left atrial injection of dye just
Figure 6. Post-treatmentsubepicardial collateral flow (ml/min per
g) is compared with mass of infarct normalized to the mass of risk
region. In the prostacyclin (PG I2)-t reated group, two dogs had
fl ows greater than 0.25 mllmin per g and a small infarct (less than
50% of the risk region) . The remaining PGlz-treated dogs had low
fl ow values similar to those of control dogs « 0. 2 mllmin per g)
and smaller infarct size than that of control dogs .
It>
100
4
A
II..
13
A
I-
•
A
Z
wz
uo
CIH
Wl!l
D.W
CI
Ul
1- ...
60
A
••
•
40
<I:
...z
A
A
UCI
CI
II..
A
•
Ul
<I:~
o PAOSUcYCLIN
AVEHICl.E
60
20
•
•
•
•
•
It is unclear why an increase in collatera l flo w occurred
after administration of PGl z in our earlier study (4) but not
in the present one . Other investigators found no change ( (5)
0
0.0
0 .1
0.2
before sacrifice of the animal. The dye distributes to the left
ventricle under prevailing perfusion pressure and stains
myocardium with flow of more than 50% of nonischemic
value. Importantly, this method delineates a zone of ischemic myocardium present at the end of the experiment that
may be different in size from the zone that existed early
after coronary occlusion and before treatment. Therefore,
the method differs from another recently described approach
of defin ing the hypoperfused region (12), in which radioactive microspheres are injected early after coronary occlusion and autoradiography is subsequently performed after
sacrifice.
For the purposes of our study, it was advantageous for
the hypoperfused zone to be identifi ed after treatment. We
were interested in whether PGlz could reduce ischemic damage within a defined area of low fl ow, indicating that direct
cellular effects, independent of fl ow, were important. However, this method may have serious drawbacks in screening
for potential myocardial protective agents. The beneficial
effects of a treatment that only increased flow but did not
possess direct cytoprotective actions might be missed because infarct size is measured relative to the low flow zone
existing at the end of the experiment.
It is also useful to point out that the in vivo dye method
delineates a zone of myocardium smaller than the anatomic
occluded bed size. determined by postmortem intracoronary
injection of dyes or barium gels ( 13 , 14). Anatomic bed size
includes a certain amount of myocardium that is relatively
well protected by collateral flow. Values for infarct size,
expressed relative to anatomic bed size, are smaller than
similar values normalized to ischemic zone size.
Comparison with previous studies . Our finding of a
direct cytoprotective, nonflow-mediated effect of PGlz is
supported by at least two other studies. Araki and Lefer (5)
found that PGh is beneficial during myocardial ischemia in
the isolated perfused cat heart. Significant improvement of
cardiac function and moderation of depletion of tissue creatine kinase activity by PGI2 were observed in this model,
independent of effects on platelet aggregation and coronary
vasodilation. Previous results from our laboratory (4) showed
that a 6 hour infusion of PGh , sufficie nt to reduce mean
arterial pressure by 5%, reduced infarct size by 67% and
increased mean collateral fl ow by about threefold compared
with that in control animals. Analysis of individual responses, however, suggested that PGh could produce considerable myocardial salvage despite low levels of collateral
flow after treatment, implying that at least a portion of the
effect was direct and not flow-mediated.
0 .3
POSTTREATMENT (5hr)
SUBEPICARDIAL COLLATERAL FLOW
0.4
or a decrease (1 6) in collateral fl ow in anesthetized dogs,
but in each of these studies, PGh produced a decrease in
mean arterial pressure of 15 to 45%, which decreased the
lACC Vol. 2. No 2
MELIN AND BECKER
PROSTACYCLIN IN EXPERIMENTAL INFARCTION
August 1983 279-86
driving pressure for collateral perfusion. In the present study,
mean arterial pressure decreased only 8% 10 minutes after
beginning the infusion and was decreased only 5% at I hour
and 2% at 5 hours, respectively. There were several differences between this study and our earlier one. Previously,
we used saline solution to mix PGlz , with frequent changes
of fresh solution to avoid loss of activity, but in this study
we used Tris buffer with a pH of 9.4. Although unlikely,
it is possible that the small amounts of alkaline solution
somehow inhibited collateral function in this study. A more
important difference may be that the earlier study used conscious dogs and this study used anesthetized animals. The
effects of anesthesia on collateral function have not been
well explored, but increased sympathetic tone after anesthesia increases heart rate (17) and results in less diastolic
time per minute and relatively more extravascular resistance
to collateral flow . In addition, anesthesia could cause an
increase in coronary vascular resistance, affecting nonocc1uded arteries that provide the source of collateral flow,
and interferring with the usual response to vasodilators.
Mechanism of direct myocardial salvage by prostacyelin. Our results point to a direct cytoprotective effect of
PGlzas the major mechanism for myocardial salvage in this
study. Collateral flow was unchanged for the group of animals receiving PGlz, although flow did increase in three
individual dogs. A reduction in myocardial oxygen demand
was probably not a major factor, because mean arterial
pressure decreased only 8% and heart rate did not change.
The stabilizing action of PGl z on lysosomal and other
membranes may represent an important aspect of the direct
protective effect of the drug. During ischemia, leakage of
lysosomal enzymes (hydrolases and proteases) is reported
to occur before the irreversible damage of myocardium(18) .
Release of these enzymes into the cytoplasm could contribute to the degradation of structural proteins and membrane
phospholipids. PGlz has been shown to stabilize lysosomes
in the isolated cat liver (19) and in ischemic myocardium
of intact animals (20).
Activation of platelets and fo rmation of aggregates in
obstructed coronary vessels (21,22) could decrease coronary flow on a microvascular level . Platelet aggregation
causes increases in platelet thromboxane levels and secretion
of granular constituents that include coagulation factors,
vasoconstrictors and lysosomal enzymes (23). The granular
products could lead to a vicious cycle of platelet plugging,
vasoconstriction and necrosis, especially in marginal zones
(24). PGh has been found to: I) prevent platelets from
clumping , 2) disaggregate preformed platelet aggregates
(25,26), and 3) inhibit the conversion by platelets of arachidonic acid to the potent coronary vasoconstrictor,
thromboxane A z (27). Although these effects on platelets
could serve as the basis for the protective action of PGIz ,
myocardial protection has been demonstrated in the isolated
cat heart perfused with platelet free solution (5). Recently,
285
a PGlz mimetic compound, a new carbocyclin derivative
(ZK 36 375), has been shown to have significant cardioprotective activity during acute myocardial ischemia in the
cat. This action was dissociated from antiplatelet effects
because the drug did not reverse ischemia-induced formation
of platelet aggregates in vivo (28).
A recent study (29) suggested that PG/ z pres erves intraneuronal catecholamines in adrenegic nerves in the ischemic myocardium, presumably by inhibition of catecholamine release . Overflow of catecholamines stimulates
myocardial oxygen demand and platelet aggregation and
also increases myocardial ischemia. PGlz may antagonize
these potentially deleterious effects by stabilizing catecholamine-containing membranes within the nerve terminals.
References
I. Dustmg GJ, Moncada S, Vane JR. Prostacyclin (POX) is the endogenous metabolite responsible for relaxation of coronary arteries induced by arachidonic acid. Prostaglandins 1977;13:3-15 .
2 Moncada S. Higgs EA, Vane JR. Human arterial and venous tissues
generate prostacyclin (prostaglandin X), a potent inhibitor of platelet
aggregation. Lancet 1977; I:18-21.
3. Ogletree ML. Lefer AM, Smith JB. Nicolaou KC. Studies on protective effect of prostacyclin In acute myocardial ischemia. Zur J
Pharmacol 1979;56.95-1 03.
4. Jugdutt BI, Hutchins GM, Bulkley BH. Becker LC. Dissimilar effects
of prostacyclin, prostaglandin E1, and prostaglandin E2 on myocardial
infarct size after coronary occlusion in conscious dogs. Circ Res
1981;49:685- 700.
5. Araki H. Lefer AM. Role of prostacyclin In the preservation of ischemic myocardial tissue In the perfused cat heart Cire Res 1980;47:75763.
6. Bayer B, Blass K. Forster W. Antiaggregatory effect of prostacyclin
(PGI2) in vivo Br J Pharmacol 1979;66.10-2.
7 Gryglewski R, Korbut R, Ocetkiewicz A. De-aggregatory action of
prostacyclin in VIVO and its enhancement by theophylline. Prostaglandins 1978;15:637- 44.
8. Romson JL, Haack OW. Abrams GO. Lucchesi BR. Prevention of
occlusive coronary artery thrombosis by prostacyclin infusion in the
dog. Circulation 1981;64:906- 14.
9. Lie JT. Pairolero PC, Holley KE, Titus JI.. Macroscopic enzymemapping verification of large, homogeneous experimental myocardial
Infarcts of predictable size and location in dogs. J Thorac Cardiovasc
Surg 1976;69:599-605
10. Fishbcm MC. Meerbaum S, Rit J, et al. Early phase acute myocardial
Infarct size quantification: validation of the tnphenyl tetrazolium chloride tissue enzymestaining technique. Am Heart J 1981;101:593-600 .
II . Darsee JR. Kloner RA, Braunwald E. Demonstration of lateral and
epicardial border zone salvageby flurbiprofen using an in vivo method
for assessing myocardium at risk. Circulation 1981;63:29-35.
12. DeBoer LWV, Strauss HW, Kloner RA, et al. Autoradiographic method
for measuring the ischemic myocardium at risk: effects of verapamil
on infarct size following experimental coronary artery occlusion. Proc
Natl Acad SCI USA 1980;77:6119-23.
13. Jugdutt BI, Hutchins GM, Bulkley BH, Becker LC. Myocardial infarction in the conscious dog: three-dimensional mapping of infarct,
collateral flow and region at risk. Circulation 1979;60:1141-50 .
14. Reimer KA. Jennings RB The " wavefront phenomenon" of myocardial Ischemic cell death. Lab Invest 1979:40 633-44 .
15. Ribeiro LGT, Brandon TA. Hopkins TA. Reduto LA, Taylor AA,
Miller RR. Prostacyclin In experimental myocardial Ischemia: effects
286
MELIN AND BECKER
PROSTACYCLIN IN EXPERIMENTAL INFARCTION
on hemodynamics, regional myocardial blood flow, infarct size and
mortality. Am 1 Cardiol 1981;47:835-40.
16. Jentzer IH, Sonnenblick EH, KIrk ES. Specificity of prostacyclin as
a coronary artery vasodilator (abstr). Clin Res 1979;27: 177A.
17. Manders WT, Vatner SF. Effects of sodium pentobarbital anesthesia
on left ventricular function and distribution of cardiac output in dogs
with particular reference to the mechanism for tachycardia. Circ Res
1976;39:512-7.
18. Wildenthal K, Decker RS, Poole AR, Griffin EE, Dingle IT. Sequentiallysosomal alterations dunng cardiac ischemia. I. Biochemical
and immunohistochemical changes. Lab Invest 1978;38:656-61.
19. Araki
f1,
Lefer AM. Cytoprotective effect of prostacyclin during hy-
poxia in the isolated perfused cat liver. Am 1 PhysIOI1980;238:HI7681
20. Ogletree ML, Lefer AM, Smith 18, Nicolaou KC. Studies on the
protective effect of prostacyclin in acute myocardial Ischemia. Eur 1
Phannacol 1979;56:95-103.
21. Vik-Mo H. Platelet accumulation In the myocardium during acute
nonthrombohtic coronary artery occlusion in dogs. Scand 1 Haematol
1978;21:225-32.
22. Ruf W, McNamara 11, Suehiro A, Suehiro G, Wickline SA. Platelet
trapping in myocardial infarct in baboons: therapeutic effect of aspirin.
Am 1 Cardiol 1980;46:405-12.
23. Packham MA, Mustard IF. Clinical pharmacology of platelets. Blood
1977;50:555-73.
JACC Vol. 2, No.2
August
1983:279~86
24. Kraikitpanitch S, Haygood CC, Baxter 01, Yunice AA, LIndeman
RD. Effects of acetylsalicylic acid, dipyridamole, and hydrocortisone
on epinephrine-induced myocardial injury in dogs. Am Heart 1
1976;92:615-22.
25 Aiken lW, Gorman RR, Shebushi Rl. Prevention of blockage of
partially obstructed coronary arteries with prostacyclin correlates with
inhibition of platelet aggregation. Prostaglandins 1979; 17:483-94.
26. Ohlendorf R, Petzborn E, Schror K. Prevention of infarction induced
decrease in circulating platelet count by prostacyclm. Thromb Res
1980;19:447-53.
27. Lefer AM, Ogletree ML, Smith 18, Silver Ml, Nicolaou KC, Barnette
WE, Garic GP. Prostacyclin: a potentially valuable agent for preserving myocardial tissue in acute myocardial ischemia. Science
1978;200.52-4.
28. Schror K, Darius H, Ohlendorf R, Matzky R, Klaus W. Dissociation
of antiplatelet effects from myocardial cytoprotective activity during
acute myocardial ischemia In cats by a new carbacyclin derivative (Z
K 36 375) 1 Cardiovasc Phannacol 1982;4:554-61.
29. Schror K, Addicks K, Darius H, Ohlendorf R, Rosen P. PGh inhibits
ischemia induced platelet activation and prevents myocardial damage
by inhibition of catecholamine release from adrenergic nerve tenninals. Evidence for cAMP as common denominator. Thromb Res
1981;21: 175-81.