Download Systole Has Little Effect on Diastolic Coronary Artery Blood Flow

443
Systole Has Little Effect on Diastolic Coronary
Artery Blood Flow
Stephen A. Katz and Eric O. Feigl
Downloaded from http://circres.ahajournals.org/ by guest on May 7, 2017
The hypothesis that myocardial systolic contraction attenuates diastolic coronaryflowwas tested by
comparing flow during diastole to flow during a prolonged asystole. The circumflex coronary artery
was cannulated and perfused at constant pressure in closed-chest, morphine- and a-chloraloseanesthetized dogs. The heart was paced at 80,120,160, or 200 beats/min after atrioventricular heart
block under control, intracoronary adenosine, and intravenous norepinephrine treatment conditions.
Cessation of pacing while holding coronary pressure constant at the previous diastolic pressure
resulted in asystolic circumflexflowthat initially equaled the previous diastolicflowduring heart rates
of 80, 120, and 160 in all treatment groups. Initial asystolic circumflex flow was approximately 5%
higher than the previous diastolic flow at a heart rate of 200 beats/min, but this was probably due
to an artifact. It is concluded that systolic contraction does not limit diastolic coronary flow at heart
rates less than 160 beats/min and probably does not at higher heart rates. (Circulation Research 1988;
62:443-451)
I
n 1957, Sabiston and Gregg1 reported that producing asystole with vagal stimulation resulted in a
rapid increase in coronary flow above the previous
diastolic level and a sustained elevated plateau in
coronary flow during asystole. This observation has
been interpreted to indicate that repeated systoles limit
diastolic coronary flow. That is, with a regular heart
beat, diastole is too brief for coronary flow to reach a
steady value, even if coronary artery pressure is held
constant.
The purpose of the present investigation was to
reexamine the effect of systole on diastolic flow by
using the prolonged asystole that follows the cessation
of ventricular pacing in a heart-blocked preparation.
With the circumflex coronary artery perfused at constant pressure, flow during a prolonged asystole did not
rise above the previous diastolic flow level when the
heart rate was below 200 beats/min. At a heart rate of
200 beats/min, asystolic flow rose slightly above the
previous diastolic flow, but this may have been due to
an artifact of the experimental preparation. It is
concluded that diastolic coronary flow is not limited by
the regular systoles of the beating heart.
Materials and Methods
General Preparation
Male dogs weighing 17-31 kg were sedated with
morphine sulfate (2.0 mg/kg s.c.) and then anesthetized with a-chloralose (100 mg/kg i.v.). The anesthesia was maintained by a continuous infusion of
a-chloralose (5 mg/kg/hr i.v.), supplemental injections
From the Department of Physiology and Biophysics, University
of Washington, Seattle, Washington.
Supported by National Institutes of Health grant HL 16910.
Address for correspondence: Eric O. Feigl, MD, Department of
Physiology and Biophysics SJ-40, University of Washington,
Seattle, WA 98195.
Dr. Katz's present address: Hennepin County Medical Center,
M.M.C., Regional Kidney Disease Program, D-Building, 5th
Floor, 701 Park Avenue, Minneapolis, MN 55415.
Received September 24, 1986; accepted September 24, 1987.
of 0.5 g a-chloralose as needed, and an additional
injection of morphine sulfate (0.5 mg/kg s.c.) approximately 3 hours after the initial sedation. The metabolic
acidosis associated with chloralose anesthesia was
counteracted by a continuous infusion of 150 mM
NaHCO3 (5 ml/kg/hr) and occasional intravenous bolus
injections of 8.4% NaHCO3.2 Positive-pressure ventilation (model 607, Harvard, South Natick, Massachusetts) against 0-1 cm H2O end-expiratory back pressure
was adjusted to keep end-expiratory CO2 between 4.5%
and 5%, as monitored by infrared absorption (model
LB-2, Beckman, Fullerton, California). Inspired O2
concentration was supplemented to maintain an arterial
O2 tension of 120-150 mm Hg. Body temperature was
held constant at 37° C with a heating pad controlled by
an esophageal thermostat (model 73ATA, Yellow
Springs, Yellow Springs, Ohio). Blood coagulation
was prevented by sodium heparin (bolus of 750 U/kg
plus 250 U/kg/hr i.v.). Occasionally, lidocaine (1
mg/kg) was given when asystolic time periods after the
cessation of pacing were observed to be less than the
sum of two cardiac cycles.
A diagram of the experimental preparation is shown
in Figure 1. A double pressure transducer catheter
(Millar Instruments, Houston, Texas) was inserted in
the left femoral artery and advanced until the distal
transducer was in the left ventricle and the proximal
transducer was in the ascending aorta. The maximum
of the first derivative of left ventricular pressure with
respect to time, dP/dt, was obtained by differentiating
the intraventricular pressure signal using a linear
electronic differentiator. The right and left femoral
veins were cannulated for continuous NaHCO 3 administration and norepinephrine infusion, respectively.
Heart Block, Pacing, and Atrial Fibrillation
Closed-chest atrioventricular heart block was produced by injecting formaldehyde into the atrioventricular nodal region during fluoroscopic examination. 3 A
pacing catheter (model 5651, USCI, Billerica, Massachusetts) was placed in the right ventricle via the right
444
Circulation Research
Vol 62, No 3, March 1988
EMF
flowmeter
Adenosine
Windkessel pressure
Downloaded from http://circres.ahajournals.org/ by guest on May 7, 2017
Norepinephrine
FIGURE 1. Diagram of closed-chest experimental preparation.
Circumflex branch of left coronary artery was pump perfused
through a stainless steel cannula. Roller pump in coronary
perfusion line maintained coronary pressure during asystole at
previous diastolic level via a servo-control system operating off
of the Windkessel pressure. Circumflex flow was measured with
an electromagnetic flowmeter. Catheter electrodes in right
atrium and right ventricle were used for production of atrial
fibrillation and ventricular pacing, respectively Atrioventricular heart block not shown.
external jugular vein for ventricular pacing. An electrode placed against the wall of the right atrium via the
left external jugular vein was used to produce temporary atrial fibrillation (4-8 V, 1-msec duration, 50-100
Hz) during experimental runs.
Circumflex Artery Perfusion
The circumflex coronary artery was cannulated with
a wedge-tip stainless steel cannula (a modification of
the coronary flowmeter described by Smith et al4) and
perfused with blood from the right femoral artery
without opening the chest. Circumflex artery pressure
was measured at the tip of the cannula via an inner steel
tube and strain gauge manometer (model P23 ID,
Statham, Cleveland, Ohio).
Circumflex blood flow was measured in the perfusion line with an extracorporeal electromagnetic flow
transducer (model SWF-4RD, Zepeda, Seattle, Washington) with a 70-Hz cutoff filter. Mean coronary blood
flow was obtained with an averaging circuit with a
2.0-second time constant. The extracorporeal coronary
perfusion capacitance between the flowmeter and the
circumflex artery was minimized by placing the flowmeter directly proximal to the stainless steel cannula.
With the heart paced, mean circumflex coronary
artery pressure was set to equal mean arterial pressure
with a servo-controlled roller pump.5 During prolonged
asystoles, coronary artery pressure was held constant
and equaled the previous diastolic pressure within a
range of ± 3 mm Hg. It was more effective to servo
control the Windkessel pressure with high loop gain
than to servo control coronary tip pressure.
The seal between the cannula tip and the circumflex
artery was verified with four tests: 1) 10 \i% nitrogrycerin dissolved in 0.5 ml saline was given down a
side tube that opened to the outside of the cannula just
proximal to the cannula tip. With a satisfactory seal,
no increase in circumflex flow was observed, indicating that the nitroglycerin was unable to reach the
circumflex bed across the wedged cannula tip. 2)
Circumflex cannula tip perfusion pressure was briefly
increased to 50 mm Hg above aortic pressure. With
a satisfactory seal, circumflex flow increased only
slightly. 3) The inflow to the cannula was occluded for
5 seconds. Cannula tip pressure fell rapidly to
approximately 25 mm Hg or less with a satisfactory
seal. 4) At the termination of each experiment, a
saturated solution of crystal violet in ammonium
hydroxide was injected into the cannula while circumflex perfusion pressure was held at 50 mm Hg
above aortic pressure. A poor seal was indicated by
staining of the initial circumflex endothelium proximal to the cannula tip. Data were accepted only if all
four criteria were met. The weight of the stained
myocardium was used to calculate circumflex blood
flow per gram of myocardium.
Experimental Design
In six dogs with atrioventricular heart block and
temporary atrial fibrillation, asystoles were produced
by cessation of ventricular pacing. Circumflex perfusion pressure during the asystole was held at the
previous diastolic level.
Experimental runs were performed at four separate
paced heart rates of 80, 120, 160, and 200 beats/min
during each of three separate treatments for each of the
six dogs. The three treatments consisted of control
conditions, intracoronary adenosine infusion (1 raM
adenosine in saline) adjusted to maximize circumflex
coronary flow, and intravenous norepinephrine (0.04
mg/ml saline) adjusted to approximately double left
ventricular dP/dt. The order of heart rates was randomized within treatment groups. The order of treatment groups was control, intracoronary adenosine, and
intravenous norepinephrine. For some experimental
runs, it was necessary to briefly stop the ventilation
pump to prevent aortic pressure oscillations induced by
ventilation.
In three additional dogs, heart block was not
performed, and cardiac arrest was produced by stimulating the distal cut ends of both cervical vagi with
platinum electrodes using pulses of 0.4-msec duration,
Katz and Feigl
ASYSTOLE
Asystolic Coronary Flow
445
HR = 80
CORONARY FLOW |irt/mm • t~')
1.0-i
0.5-
CORONARY PRESSURE (mm K||
2<Xh
FIGURE 2. Control circumflex coronary
blood flow response to cessation of ventricular pacing at 80 beats/min. Asystolic flow is not greater than previous
diastolic flow, although mean blood flow
increased during asystole. Coronary
pressure during asystole was held constant at previous diastolic level
MEAN CORONARY FLOW
1.0-i
0.5-
AORDC PRESSURE (mm Hi)
20<h
Downloaded from http://circres.ahajournals.org/ by guest on May 7, 2017
100-
T1ME (Mtondt)
7-12 V, and 30 Hz. Circumflex perfusion pressure
during vagal arrest was held at the previous diastolic
level. Experimental runs were performed under all
treatment conditions but only at the prevailing intrinsic
sinus rate after vagotomy. Vagal stimulation was also
performed with the right ventricle paced at the intrinsic
rate. This was done to find the time from the onset of
vagal stimulation to the beginning of coronary parasympathetic vasodilation.
Data Analysis
Circumflex diastolic coronary blood flow was digitized for five cardiac cycles immediately previous to
a prolonged asystole. The moment of maximum left
ventricular dP/dt was used for timing purposes in the
cardiac cycle, and diastolic flow points were digitized
at 100, 50, 40, or 20 msec prior to the moment of
maximum dP/dt for the respective heart rates of 80,
120,160, and 200 beats/min. Additional diastolic flow
points were taken prior to this time point every 100
msec as the duration of diastole permitted (for example, two additional points at a heart rate of 80
beats/min, one additional point at a heart rate of 120
beats/min).
Asystolic flow points were digitized as above at 100,
50,40, or 20 msec prior to when a projected maximum
left ventricular dP/dt would have occurred had there
been a systole with respective heart rates of 80, 120,
160, or 200 beats/min. Additional asystolic flow points
were taken every 100 msec before and after this point
extending for at least two projected cardiac cycles
during the asystole. The asystolic flow test point used
for comparison was the determination made 100 msec
after the time-adjusted projected maximum left ventricular dP/dt point described above.
Average diastolic flow for a given experimental run
was calculated as the average of all diastolic flow point
determinations for five cardiac cycles immediately
previous to an asystole. Individual diastolic and
asystolic points were expressed as a percent of the
average diastolic value for that experimental run and
then averaged over the six animals on a time-pointby-time-point basis. This normalization was used to
remove the variability in absolute blood flow among
dogs so that changes between diastole and asystole
would be apparent.
The Wilcoxon signed rank test was used to test for
differences between absolute diastolic and asystolic
flows. Ap value of ^0.05 was considered significant
(two-tailed).
Results
A representative record of an asystole after cessation
of a ventricular paced rate of 80 beats/min under control
conditions is shown in Figure 2. Asystolic flow was not
greater than diastolic flow, although mean flow increased during asystole. For comparison, a small increase in asystolic flow above the previous diastolic level
for a heart rate of 200 beats/min is shown in Figure 3.
Asystoles after cessation of ventricular pacing were
performed during atrial fibrillation to prevent the
interference of atrial contraction upon circumflex
coronary flow (atrial coving)6"8 and perfusion pressure.
An example of two asystoles, before and after atrial
fibrillation, is shown in Figure 4. Although atrial coves
complicated the flow records, they did not appear to
alter the basic relation between diastolic and asystolic
flows, as can be seen from Figure 4.
Table 1 shows the diastolic flow averaged from five
cardiac cycles immediately prior to the cessation of
pacing, the corresponding asystolic test point flow, and
other hemodynamic data for control, intracoronary
446
Circulation Research
Vol 62, No 3, March 1988
ASYSTOLE
HR = 200
CORONARY FLOW (nH/min. f"*)
l.M-i
0.750-
\J
CORONARY PRESSURE (mm Hg)
FIGURE 3.
200
onary blood flow response to cessation of ventricular pacing at 200
beats/min. A small increase in asystolic flow compared with previous
diastolic flow occurs during initial
part of asystole. Mean coronary
blood flow also increased during
asystole. Coronary pressure during
asystole was held constant at previous diastolic level.
100
AAAAAAAA/VVVV
MEAN CORONARY FLOW (irt/mjn.g"1)
0.750
Control circumflex cor-
Downloaded from http://circres.ahajournals.org/ by guest on May 7, 2017
AORTIC PRESSURE (mm Hg)
200-,
100-
TIME (ucomb)
adenosine, and intravenous norepinephrine treatments
at heart rates of 80, 120, 160, and 200 beats/min.
Figures 5, 6, and 7 show the average normalized
diastolic and asystolic flows as a function of time. The
asystolic test points are marked with arrows. At heart
rates of 80, 120, and 160 beats/min, the asystolic flow
was not significantly greater than the average diastolic
flow from five previous corresponding cardiac cycles for
all three treatment groups (Table 1 and Figures 5 , 6 , and
7). At a heart rate of 200 beats/min, six out of six dogs
in the control and intracoronary adenosine treatment
groups exhibited significantly greater {p = 0.026, Wilcoxon signed rank test) asystolic test point flows than the
corresponding average diastolic flows found from the
five previous cardiac cycles (Table 1 and Figures 5 and
6). At a heart rate of 200 beats/min, four out of six dogs
EFFECT OF ATRIAL FIBRILLATION
BEFORE
AFTER
CORONARY FLOW (ml/min-g"')
1.50-1
1.50-1
CORONARY PRESSURE (mm Hg)
2O0-I
200-1
FIGURE 4. Two asystoles performed before
and after initiation of atrial fibrillation.
Electrocardiogram P-waves are coincident
with both coronary flow and pressure oscillations before atrial fibrillation. After onset
of atrial fibrillation, the electrocardiogram
channel records artifact from atrial stimulation at 50 Hz.
Katz and Feigl
Asystolic Coronary Flow
447
TABLE 1. Hemodynainlcs During Asystole
Diastolic
conductance
Circumflex
diastolic
flow
(ml/min/g)
Circumflex
asystolic
flow
(ml/min/g)
Mean
circumflex
flow
(ml/min/g)
Left
ventricular
dP/dt
(m Hg/sec)
Mean
arterial
pressure
(mm Hg)
Mean
conductance
I mm Hg /
I mm Hg /
Circumflex
diastolic
pressure
(mm Hg)
0.72±0.12
0.94±0.12
1.19±0.16
1.46 + 0.18*
0.70±0.12
0.92±0.12
1.23±0.17
1.55±0.16*
0.54±0.09
0.65 ±0.09
0.77±0.12
0.93±0.13
1.65 ±0.27
1.83±0.31
2.03 ±0.22
2.63±0.33
106±5
109±7
104±7
107 ±10
5.0±0.6
5.9±0.5
7.4±1.1
8.7±0.9
7.7±0.9
9.7±1.1
13.8±2.0
17.5±2.1
95 ±5
96±8
87 ±6
84±8
4.59±0.51
4.39 ±0.39
4.48±0.69
3.86±0.5O*
4.55 ±0.49
4.35 + 0.37
4.60 + 0.70
4.06±0.48*
4.24 ±0.49
4.27±0.53
3.89±0.58
3.45 ±0.47
1.90±0.17
2.03 ±0.20
2.31 ±0.21
2.69±0.23
94±7
101+7
99±8
94±7
44.8±3.6
41.4±3.3
38.4±3.7
35.7±2.4
63.6±3.4
62.7±4.4
62.5 ±6.3
52.7±4.0
71±5
74 + 5
70±5
72 + 5
0.95 ±0.19
1.51 ±0.49
1.43 ±0.27
1.13 + 0.11
3.00 + 0.39
3.70±0.26
4.26 ±0.39
4.66±0.48
110±9
132±17
135 ±14
122±14
8.6±1.3
11.0±2.9
10.7 + 1.7
9.6+1.0
13.1 =t 1.7
16.7±2.9
17.5±1.8
19.2±2.3
95 ±9
113±15
112±12
93 ±13
1.14 + 0.21
1.09±0.25
1.90 ±0.47
1.88 ±0.49
1.92 ±0.22
1.88±0.25
1.65 ±0.08
1.70 + 0.08
Values are mean±SEM; n = 6.
*p = 0.026, Wilcoxon signed rank test.
CONTROL
DIASTOLIC
1
1
tolic flow decreased despite constant coronary asystolic pressure (Figures 5,6, and 7). Mean coronary flow
was always increased during asystole compared to the
mean flow found for the five previous corresponding
cardiac cycles.
in the norepinephrine treatment group exhibited greater
coronary flows at the asystolic test point than the average
diastolic flows found from the five previous corresponding cardiac cycles (Table 1 and Figure 7).
After the first asystolic test point, succeeding asysAND
ASYSTOLIC
CORONARY
FLOW
100%-
o
—;
50%-
H R ' 80 b/min
HR • 120 b/min
100% =0.72 mi/mki.g"'
1 0 0 % - 0 . 9 4 mj/mtn - g
n.6
-I
n.6
-PACE
—
ASYSTOLE •
• PACE
- ASYSTOLE -
<
0%
IRCU
Downloaded from http://circres.ahajournals.org/ by guest on May 7, 2017
Heart
rate
(beats/min)
Control
80
120
160
200
Adenosine
80
120
160
200
Norepinephrine
80
120
160
200
100%-
I'""
o
Q
H R . 160 b/min
100%.
50% -
1. 19 m)/min-g"'
n. 6
—
o
-«
0
a
PACE -
1
2
—ASYSTOLE —
ASYSTOLE
3
4
5 0
1
2
3
4
TIME (seconds)
TIME (seconds)
FIGURE5. Normalized diastolic flow averages for control treatment group at heart rates of 80,120, 160, and 200 beats/min. Asystolic
test points are indicated by arrows. Asterisk denotes that at rate of 200 beats/min, asystolic flow was significantly greater than average
diastolic flow. Standard error bars for normalized averages are not shown because they would lie within flow points.
448
Circulation Research
Vol 62, No 3, March 1988
ADENOSINE
DIASTOLIC
AND ASYSTOLIC
CORONARY
FLOW
100% -
o
50%-
HR • 80 b/min
HR =120 b/min
100% '4.59 mi/min • g"'
100%. 4.39 ml/min -g
n .6
-l
n .6
• PACE
- I — ASYSTOLE -
• PACE
-ASYSTOtE-
0%
100% -
*
•
•
o
Downloaded from http://circres.ahajournals.org/ by guest on May 7, 2017
o
Q
50% -
HR .160 b/min
HR -200 b/min
100% = 448 ml/min-g" 1
100% "3.86 ml/min • g"'
n.6
n .6
— PACE
cc
o
—
-ASYSTOLE-
PACE
— I ASYSTOtE
0%
1
2
1
T I M E (seconds)
3
4
T I M E (seconds)
FIGURE 6. Normalized diastolic flow averages for adenosirie treatment group at heart rates of 80, 120, 160, and 200 beatslmin.
Asystolic test points are indicated by arrows. Asterisk denotes that at rate of 200 beatslmin, asystolic flow was significantly greater
than average diastolic flow. Standard error bars for normalized averages are not shown because they would lie within flow points.
NOREPINEPHRINE
DIASTOLIC
AND ASYSTOLIC
100% -
o
Q
HR i 80 b/min
HR .- 120 b/min
100* = 1.14 ml/min-g"'
100X = 1.90 ml/min-g''
n .6
^
co
FLOW
"r
50% I—
CORONARY
n .6
PATF
•
—-ASYSTOLE — •
- PACE
- I -^ASYSTOLE - ~
n% -
X
100% -
I"'
o
cr
o
HR = 160 b/min
HR >200 b/min
Q
UJ
100)5 • 1.92 ml/mm-g' 1
100% = 1.65 ml/min-|
50% -
n :6
i
en
O
P^CE
n.6
» -ASYSTOLE-
-.
• I ASYSTOLE
PACE
nx 23
T I M E (seconds)
1
2
T I M E (seconds)
FIGURE 7. Normalized diastolic flow averages for norepinephrine treatment group at heart rates of80, 120, 160, and 200 beatslmin.
Asystolic test points are indicated by arrows. Standard error bars for normalized averages are not shown because they would lie
within flow points.
Katz and Feigl
VAGAL ARREST
Asystolic Coronary Flow
449
HR = 1 7 4
CORONARY FLOW ( m i / m i n - g )
2.251.500.75CORONARY PRESSURE (mm Hg)
200-
A
100-
~i
"v
<\
r
*
A
•'•
v-*-
•^~-
r
•"• r- r\ C\ r\
^- v ^ v/^' \^- v^- v*-'
n
MEAN CORONARY FLOW ( m i / m i n . g"1)
Z251.500.75AORTIC PRESSURE (mm Hg)
200
n
Downloaded from http://circres.ahajournals.org/ by guest on May 7, 2017
100-
TIME
(seconds)
FIGURE 8. Circumflex coronary blood flow response to vagal arrest from intrinsic heart rate of 174 beats/min. A small increase in
asystolic flow compared with previous diastolic flow can be seen. Mean coronary blood flow also increased during vagal arrest.
Coronary pressure during vagal arrest was held constant at previous diastolic level.
In three additional dogs without heart block, vagal
arrest was produced by vagal stimulation at intrinsic
vagotomized heart rates between 104 and 185 beats/
min. The protocol was identical to the pace cessation
asystole protocol except that atrial fibrillation was not
used and cardiac arrest was produced by vagal stimulation. For all three treatment groups, initial circumflex flow during vagal arrest was not greater than the
corresponding average diastolic flow unless the heart
rate was greater than approximately 160 beats/min. A
representative record of a vagal arrest is shown in
Figure 8. In each of the three dogs, at least one
experimental run was performed where right ventricular pacing at the previous intrinsic vagotomized rate
was initiated immediately after the onset of vagal
stimulation. The latency between the onset of vagal
stimulation and the beginning of coronary vasodilation
was 2 or more seconds in every case.
Discussion
The purpose of the present study was to determine
the effect of myocardial systolic contraction on diastolic coronary flow. This was accomplished by recording coronary flow during prolonged asystoles
while maintaining diastolic coronary pressure constant. Adenosine and norepinephrine treatments were
used to examine vasodilation and increased contractile
states, respectively. The results indicate that systole
does not impede diastolic flow at heart rates of 80,120,
and 160 beats/min. At heart rates of 200 beats/min,
asystole resulted in a small (5%) but significantly
increased flow in six out of six dogs in both control and
adenosine treatment groups and in four of six dogs in
the norepinephrine treatment group. However, at heart
rates of 200 beats/min, the experimental preparation
may have limited diastolic flow.
There are two possible ways the experimental
preparation may have limited diastolic flow at 200
beats/min. First, with a very brief diastolic period,
there may not have been sufficient time to accelerate
blood through the tubing and cannula from the Windkessel. In effect, the cannula system lengthens the
"coronary artery" supplying the circumflex artery from
the pressure source in the Windkessel.
Second, contraction and relaxation of the left ventricle is less synchronous than normal in the presence
of heart block and right ventricular pacing because the
Purkinje system is not active in the initial phase of
depolarization.9"13 An asynchronous contraction and
relaxation may tend to abbreviate diastole; thus, the
diastolic period at 200 beats/min may have been shorter
than normal.
Considering the two potential artifacts discussed
above and that the increase in flow during asystole was
only about 5% at 200 beats/min, it seems unlikely that
systole limits diastolic flow in the usual range of heart
rates. However, the present results do not rule out the
possibility that systole limits diastolic flow at heart
rates above 200 beats/min. It was not possible to obtain
steady coronary pressures during diastole at ventricular rates higher than 200 beats/min with the present
experimental preparation, probably because of the
hydraulic impedance of the cannula system. Thus,
measurements at heart rates above 200 beats/min
were not attempted. Sabiston and Gregg1 did not
report heart rates, but heart rates of 180, 150, 214,
450
Circulation Research
Vol 62, No 3, March 1988
Downloaded from http://circres.ahajournals.org/ by guest on May 7, 2017
and 240 beats/min can be estimated from their figures.
Sabiston and Gregg1 reported an increase in mean
coronary flow in the transition from regular heart beats
to sustained asystole, and the present study confirms
their observation of mean coronary flow (Figures 2 and
3). Although asystolicflowdid not exceed the previous
diastolic flow at heart rates of 160 beats/min or less,
mean coronary flow invariably increased during asystole. This is not evidence that systole limits diastolic
flow; it occurs because diastolic flow is no longer
phasicalry interrupted by systole, and meanflowduring
asystole therefore converges toward the previous diastolic flow.
Sabiston and Gregg also observed that phasic coronaryflowincreased rapidly at the onset of asystole and
then remained at an elevated plateau well above the
previous peak diastolic flow level. The present results
fail to confirm the phasic flow results of Sabiston and
Gregg. In the present study, asystolic flow did not
exceed the previous diastolic flow at heart rates of 160
beats/min or less. Furthermore, a sustained elevated
plateau of phasic coronary flow was never observed at
any heart rate. In fact, asystolic flow tended to decline
with time. Asystolic flow was observed over a time
period equivalent to two cardiac cycles. In general, at
the end of that period, flow was less than average
diastolic flow despite constant coronary perfusion
pressure (Figures 5, 6, and 7). For heart rates of 80,
120, and 160 beats/min in all treatment groups,
including the adenosine group, when local metabolic
factors would be expected to be blunted, only three of
54 individual asystoles did not exhibit declining
asystolicflowsduring constant coronary perfusion. For
heart rates of 200 beats/min in all treatment groups, six
of 18 individual asystoles did not exhibit decreasing
flow rates with time, although the experimental preparation may have been inadequate at this heart rate. The
usual decline in flow was presumably caused by a
combination of a decreased rate of filling of intramyocardial capacitance (previously emptied by
systole)14 and local metabolic vasoconstriction secondary to the cessation of cardiac contractions.15
The primary reason for the difference in phasic flow
results between the present study and Sabiston and
Gregg's report is probably that the rotameter used by
Sabiston and Gregg had a limited frequency response.
Mean coronary flow increases during asystole (Figures
2 and 3), and a damped phasic recording will tend to
show this effect. However, the difference in phasic flow
results between the two studies probably involved more
than just a simple frequency limitation. The phasic
records published by Sabiston and Gregg show rapid
transitions between systolic and diastolic coronary
flow and do not appear sinusoidal, as would be expected
from a simple frequency limitation. However, it is
likely that the rotameter and perfusion system used by
Sabiston and Gregg had a markedly nonlinear hydraulic impedance with the impedance high at cardiac
frequencies (3-4 Hz) but low at steady (DC) perfusion.
Such a nonlinear impedance would result in phasic flow
being limited and steady flow being less restricted for
a given perfusion pressure. This nonlinear impedance
effect, plus a simple limited frequency response, would
explain the prompt increase to a plateau above the
previous diastolic level that was observed by Sabiston
and Gregg but was not observed in the present study.
Coronary artery pressure during asystole was servocontrolled to match the previous diastolic coronary
pressure in the present experiments. The reason for this
is because the critical experimental comparison was
between diastolic flow and the flow at the onset of
asystole. If pressure had been allowed to change from
the preceding diastolic value during the transition to
asystole, there might have been a corresponding
change in coronaryflow,especially during vasodilation
with adenosine. Such a pressure-dependent flow
change would have confused interpretation of the
results. Pressure in a cannulated coronary artery results
from the perfusion pump and from cardiac contraction
during systole (Figures 2 and 3). Cardiac contraction
compresses intramyocardial vessels and results in a
retrograde pressure wave14 so that mean coronary
pressure is greater than diastolic coronary pressure. If
mean coronary pressure had been kept constant during
the transition from diastole to asystole, then it would
have been necessary to increase pump pressure to
compensate for the absent systolic compressions. Such
an upward ramp or step of pressure at the onset of
asystole would tend to increase asystolic flow by a
pump hydraulic mechanism rather than a physiologic
mechanism. With the present pump and cannula
system, coronary pressure did not reach a steady value
during diastole at heart rates of 200 beats/min or more
(Figure 3). Thus, the hydraulic irnpedance of the
present perfusion apparatus became a limiting factor at
heart rates of 200 beats/min and above. This impedance
limitation is somewhat analogous to the limitation in
the Sabiston and Gregg experiments discussed above.
In the present experiments, asystolic perfusion
pressure was kept within 3 mm Hg of the previous
diastolic perfusion pressure, and asystole was produced by cessation of pacing after heart block, or by
vagal arrest. Other investigators have previously interrupted systolic contractions with similar techniques
and observed coronary flow in the course of investigations concerning coronary pressure-flow relations.1416"20 However, the prior studies were not designed to test the hypothesis considered in the present
study, and a systematic examination of the problem
over a range of heart rates and treatments has not been
done previously. Furthermore, asystolic perfusion
pressures were often somewhat greater than diastolic
pressures, and the complicating factor of atrial coves
was often present. Nevertheless, asystolic flows were
not usually greater than the preceding diastolic flow
level in the studies mentioned above.
In the control treatment group, both mean and
diastolic circumflex conductance increased with increasing heart rate (Table 1), presumably as a result of
functional hyperemia. In the adenosine treatment
group, when functional hyperemia was impaired with
adenosine vasodilation, mean circumflex conductance
Katz and Feigl
Downloaded from http://circres.ahajournals.org/ by guest on May 7, 2017
decreased with heart rate because less time was spent
in diastole at progressively increasing heart rates. The
corresponding diastolic circumflex conductance did
not decrease until a heart rate of 200 beats/min was
achieved.
Asystoles were also produced via vagal arrest since
this method was used in the original Sabiston and Gregg
report. After vagal arrest, the resulting asystolic flow
was not greater than the preceding average diastolic
flow unless heart rates were greater than approximately
160 beats/min (Figure 8). When the ventricles were
paced at a constant rate, parasympathetic coronary
vasodilation began approximately 2 seconds after the
start of vagal stimulation.21-22 Thus, the onset of
parasympathetic vasodilation was too slow to account
for the prompt increase in flow observed by Sabiston
and Gregg.
In summary, systolic contraction did not significantly impede diastolic coronary flow at heart rates up
to 160 beats/min. At a heart rate of 200 beats/min,
asystolic flow was initially 5% greater than diastolic
flow, but this was probably due to an artifact. It is
concluded that systolic contraction does not limit
diastolic coronary flow at heart rates below 160
beats/min and probably not at higher rates within the
physiological range.
Acknowledgments
We thank Stephanie Belanger and Fellner Smith for
expert technical assistance in these studies.
References
1. Sabiston DC Jr, Gregg DE: Effect of cardiac contraction on
coronary blood flow. Circulation 1957;15:14-20
2. Arfors KE, Arturson G, Malmberg P: Effect of prolonged
chloralose anesthesia on acid-base balance and cardiovascular
functions in dogs. Ada Physiol Scand 1971;81:47—53
3. Ito BR, Feigl EO: Technique for producing heart block in
closed-chest dogs without electrical recording. Pflugers Arch
1983;397:160-163
4. Smith FD, D'Alecy LG, Feigl EO: Cannula-tip coronary blood
flow transducer for use in closed-chest animals. JAppl Physiol
1974;37:592-595
5. Mohrman DE: A servo-controlled roller pump for cardiovascular research. Am J Physiol 1980;238:H269-H274
Asystolic Coronary Flow
451
6. Gregg DE, Khouri EM, Rayford CR: Systemic and coronary
energetics in the resting unanesthetized dog. Circ Res
1965;16:102-113
7. Reid JVO, Bhagat Cl: Effect on coronary flow of atrial
contraction. S Afr J Med Sci 1975;40:l 17-131
8. McHale PA, Greenfield JC Jr: Origin of atrial coving in canine
phasic coronary artery blood flow. Am J Physiol 1986;
251:H774-778
9. Wiggers CJ: The muscular reactions of the mammalian
ventricle to artificial surface stimuli. Am J Physiol
1925;73:346-378
10. Lister JW, Klotz DH, Jomain SL, Stuckey JH, Hoffman BF:
Effect of pacemaker site on cardiac output and ventricular
activation in dogs with complete heart block. Am J Cardiol
1964; 14:494-503
11. Kralios AC, Tsagaris TS: Effect of activation sequence on
MV02 before and after coronary ligation. Am J Physiol
1978;234:H260-H265
12. Heyndrickx GR, Vilaine JP, Knight DR, Vatner SF: Effects of
altered site of electrical activation on myocardial performance
during inotropic stimulation. Circulation 1985;71:1010—1016
13. Burkhoff D, Oikawa RY, Sagawa K: Influence of pacing site
on canine left ventricular contraction. Am J Physiol
1986;251:H428-H435
14. Spaan JAE: Coronary diastolic pressure-flow relation and zero
flow pressure explained on the basis of intramyocardial
compliance. Circ Res 1985;56:293-3O9
15. Feigl EO: Coronary physiology. Physiol Rev 1983;63:1-205
16. Dole WP, Bishop VS: Regulation of coronary bloodflowduring
individual diastoles in the dog. Circ Res 1982;50:377-385
17. Eng C, Jentzer JH, Kirk ES: The effects of the coronary
capacitance on the interpretation of diastolic pressure-flow
relationships. Circ Res 1982;50:334-341
18. Lee J, Chambers DE, Akizuki S, Downey JM: The role of
vascular capacitance in the coronary arteries. Circ Res
1984^5:751-762
19. Uhlig PN, Baer RW, Vlahakes GJ, Hanley FL, Messina LM,
Hoffman JIE: Arterial and venous coronary pressure-flow
relations in anesthetized dogs. Circ Res 1984^5:238-248
20. Klocke FJ, Mates RE, Canty JM Jr, Ellis AK: Coronary
pressure-flow relationships. Controversial issues and probable
implications. Circ Res 1985;56:310-323
21. Feigl EO: Parasympathetic control of coronary blood flow in
dogs. Circ Res 1969;25:509-519
22. Reid JVO, Ho BR, Huang AH, Buffington CW, Feigl EO:
Parasympathetic control of transmural coronary blood flow in
dogs. Am J Physiol 1985;249:H337-H343
KEY WORDS
relations •
• coronary conductance
coronary vascular resistance
•
•
pressure-flow
capacitance
Systole has little effect on diastolic coronary artery blood flow.
S A Katz and E O Feigl
Downloaded from http://circres.ahajournals.org/ by guest on May 7, 2017
Circ Res. 1988;62:443-451
doi: 10.1161/01.RES.62.3.443
Circulation Research is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 1988 American Heart Association, Inc. All rights reserved.
Print ISSN: 0009-7330. Online ISSN: 1524-4571
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://circres.ahajournals.org/content/62/3/443
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published in
Circulation Research 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 Research is online at:
http://circres.ahajournals.org//subscriptions/