The Effect of Increasing Preload
Output and Ejection in
on
Ventricular
Man
Limitations of the Frank-Starling Mechanism
DENNIS T. MANGANO, PH.D., M.D., DONALD C. VAN DYKE, M.D., AND ROBERT J. ELLIS, M.D.
Downloaded from http://circ.ahajournals.org/ by guest on July 31, 2017
SUMMARY The effects of preload augmentation on left ventricular output (Frank-Starling mechanism),
ejection fraction, and the relationship between pulmonary wedge pressure and end-diastolic volume were
studied in 15 patients immediately after myocardial revascularization. The patients were studied under
morphine sulfate anesthesia with the chest temporarily closed. Sequential measurements of radionuclide ejection fraction, thermodilution cardiac output, and systemic and pulmonary arterial pressures were made before
and after infusion of 500, 1000, 1250 and 1500 ml of whole blood. Cardiac performance was assessed by means
of (1) ventricular function curves (stroke volume and stroke work vs end-diastolic volume); (2) sequential
changes in ejection fraction and end-systolic volume; (3) the relationship between pulmonary capillary wedge
pressure (PCW) and end-diastolic volume; and (4) changes in estimated wall tension.
Volume transfusion increased the left ventricular end-diastolic volume index (EDVI) from an average control value of 41 ml/m2 to 88 ml/m2 (after infusion of 1500 ml of whole blood), and increased PCW from 2 mm
Hg to 8 mm Hg. Although cardiac index increased (2.1 to 3.6 1/min/m2) and left ventricular stroke work index
increased (20 to 38 g-m/m2), the largest increases occurred when EDVI was less than 70 ml/m2 and PCW less
than 7 mm Hg. Above these values the ventricular function curves were relatively flat. In contrast, ventricular
ejection progressively decreased with increasing preload. Ejection fraction decreased from 0.70 to 0.49 and
end-systolic volume index increased from 13 mi/m2 to 45 m]/m2, despite a decreasing systemic vascular
resistance.
Thus, in these patients: (1) the Frank-Starling mechanism plays little role in augmentation of stroke volume
above a PCW of 7 mm Hg or an EDVI of 70 ml/m2; (2) increases in ventricular output achieved by increasing
preload were accompanied by decreases in ventricular ejection fraction; and (3) the filling pressure (volume)
associated with maximal ejection fraction was significantly lower than that associated with maximal ventricular output.
STARLING'S LAW of the heart and the associated
Frank-Starling mechanism characterize cardiac performance in terms of ventricular function curves.`3
These curves express the functional consequences of
alterations in preload, and serve as the rationale for
improvement of cardiac output by volume administration. Studies in animals and man have attempted to
define the relationship between ventricular output
(stroke work, stroke volume) and filling pressure (enddiastolic, pulmonary wedge, central venous) in normal
and disease states.3'-" In patients with significant ventricular dysfunction, the maximal ventricular output is
lower than in normal patients and is achieved at higher
filling pressures, in some cases 18 mm Hg or more.4 "b'"
These results have served as guidelines for volume
transfusion therapy in patients with depressed ventricular function.4' 1 Maximal ventricular work
or volume output could be achieved by progressive
volume infusion until the "optimal" filling pressure is
obtained.
Although progressive increases in filling pressure
due to volume loading may increase ventricular output, it is not certain whether other indexes of ventricular function, such as ejection fraction and endsystolic volume, improve as well. If they do, the
benefits of volume loading should include improved
output and ejection; on the other hand, volume loading may result in diminished ejection fraction and may
alter the choice of an optimal filling pressure. These
considerations are especially important after cardiopulmonary bypass, when the heart must satisfy the
perfusion requirements of the rest of the body as well
as maintain adequate ejection to overcome the stresses
encountered during cardiac surgery. Accordingly, we
determined ventricular function in the immediate
postbypass period, and evaluated the effects of increasing filling pressure by volume loading on ejection
fraction and end-systolic volume, as well as on ventricular stroke work and stroke volume.
Methods
The protocol for this study was approved by the
Committee on Human Research of the University of
California, San Francisco, and informed consent was
obtained from all patients.
Fifteen men, ages 39-64 years, admitted for coronary artery surgery because of unstable angina (seven
patients) and stable progressive angina (eight patients) were studied. Thirteen had one or more myocardial infarctions before admission. None had valvular disease, a history of right- or left-heart failure, or
From the Departments of Anesthesia, Nuclear Medicine and
Surgery, University of California, and the Veteran's Administration
Medical Center, 4150 Clement Street, San Francisco, California.
Supported in part by the Medical Research Service of the
Veteran's Administration.
Address for correspondence: Dennis T. Mangano, M.D., Department of Anesthesia, Veteran's Administration Hospital, 4150
Clement Street, San Francisco, California 94121.
Received September 5, 1979; revision accepted February 5, 1980.
Circulation 62, No. 3, 1980.
535
CIRCULATION
536
Downloaded from http://circ.ahajournals.org/ by guest on July 31, 2017
evidence of right or left ventricular hypertrophy or
dilation. None had evidence of mitral regurgitation
using left ventriculography. Cardiac catheterization
revealed 90% stenosis of two or more coronary
arteries, ejection fractions of 0.39-0.82 (normal 0.66
± 0.06), left ventricular end-diastolic volume indexes
of 44-107 ml/m2 (normal 70 ± 20 ml/m2), pulmonary capillary wedge pressures of 1-14 mm Hg,
and central venous pressures of 0-9 mm Hg.
Medications included isosorbide and nitroglycerin in
all patients, and oral propranolol (40-160 mg four
times per day) in 13 patients. Medications were continued until 8 hours before surgery.
All patients were premedicated with morphine sulfate (10 mg i.m.) and oral diazepam (10 mg).
Anesthesia consisted of morphine sulfate (1.5-3
mg/kg i.v.) and diazepam (0.25-0.50 mg/kg i.v.).
Pancuronium (0.1 mg/kg i.v.) provided muscle relaxation, and ventilation (with 100% oxygen) was controlled. Hemodynamics were monitored by means of
radial artery and triple-lumfen thermodilution pulmonary artery catheters. All pressure measurements
were recorded on a Gilson polygraph from equisensitive Bell and Howell transducers calibrated with a
mercury manometer. Before each set of measurements, the zero reference point was located 5 cm
posterior to the sternal angle in a direction perpendicular to the frontal plane of the chest.
Ejection fraction was measured from radiocardiograms obtained with a coaxial cardiac scintillation
probe.12 The radiocardiogram is a recording of the
amount of a blood-borne bolus of radionuclide in the
heart chambers as a function of time, and is generated
from the first pass of the bolus through the central circulation.'3' 14 Figure 1 shows a typical radiocardiogram from a normal subject and illustrates the method
of measuring ejection fraction directly from the dual
time-activity (strip chart) recording as the difference
between diastolic and systolic counts divided by the
DIASTOLE
z
X y
EF
y
J
I-
SYSTOLE
o
BACKGROUND
AI AAAAhIA
Z
TIME (seconds)
FIGURE 1. A typical radiocardiogram obtained from a
coaxial probe system with time-activity curves from the left
ventricle (central port) and background (annular detector).
EF ejection fraction.
=
VOL 62, No 3, SEPTEMBER 1980
difference between diastolic and background
(baseline) counts. 12-17
The probe was positioned over the left ventricle to
record the passage of 1.5 mCi of 99mTc-sodium
pertechnetate injected into the right atrial orifice of
the thermodilution catheter. The probe incorporates a
central, collimated detector that accepts gamma
photons primarily from the left ventricle, surrounded
by an annular detector collimated for simultaneous
monitoring of background activity around the left ventricle. 12-15 High-frequency recording of count rate
from within the left ventricle, when corrected for
counts from surrounding structures, provides an accurate measure of left ventricular ejection fraction. 14-16
The radionuclide was flushed through the catheter
with the cold solution so that thermodilution cardiac
outputs and radiocardiograms were recorded simultaneously for all measurements. The position of
the radionuclide probe was not altered throughout the
measurement periods. All measurements were made
at end-expiration. Given stroke volume from thermodilution (cardiac output divided by heart rate) and
left ventricular ejection fraction from radiocardiography, left ventricular end-diastolic volume can be
calculated as stroke volume divided by ejection fraction. Because regurgitant flow would invalidate this
formula, patients with valvular incompetence were excluded from the study. In patients with coronary
artery disease without valvular regurgitation, the correlation of end-diastolic volume determined from
cineangiographic and radioisotope measures is 0.88.16
During cardiopulmonary bypass, cold cardioplegia
was used for myocardial protection.19 Ischemic arrest
time averaged 51 ± 4 minutes (range 35-74 minutes),
and 3.4 vessels per patient (average) were bypassed.
Cardiopulmonary bypass was discontinued at the
lowest pulmonary capillary wedge pressure to maintain a systolic blood pressure greater than 75 mm Hg.
After the administration of protamine, the aortic and
vena cava cannulas were removed, and the sternum
was temporarily reapproximated. Surgery was
stopped, the probe was placed over the left ventricle
and a 5-minute period of hemodynamic stability was
established (systolic blood pressure varying by less
than 1 0 mm Hg and heart rate varying by less than 5
beats/min). The following values were recorded at
end-expiration: radial artery systolic and diastolic
pressures, heart rate, pulmonary artery systolic and
diastolic, central venous and pulmonary capillary
wedge pressures, radionuclide ejection fraction, and
two thermodilution cardiac outputs (within 10%).
After these control measurements, progressive volume
loading was begun. A total of 1500 ml of whole blood
was transfused into a peripheral vein over 30 minutes.
The hemodynamic measurements were repeated after
infusion of 500, 1000, 1250 and 1500 ml.
No pharmacologic agent was given and no
respiratory adjustment was made for 15 minutes
before or during the measurement period.
Data were analyzed using two-way analysis of
variance comparing hemodynamics at each volume
EFFECT OF INCREASING PRELOAD/Mangano
et
537
al.
0
_
cm
I100-
Cl
75
6
cs
E
50
E
>
_i
23
25
FIGURE 2. Effect of 1500-ml volume
transfusion in a typical patient. L VEDVI
left ventricular end-diastolic volume index;
BP = blood pressure; HR = heart rate;
SVR = systemic vascular resistance; EF
ejection fraction; CI = cardiac index;
L VSWI = left ventricular stroke work index; T = estimate of left ventricular wall
tension at start of systole.
E0
=
=
5S
58
62
84
82
83
81
82
941
945
825
848
900
(mnHg) n
b/mn }
SVR
{dyne. sec)
\
cm&Oo
rhz
&oO
-10
AMOUNT TRANSFUSED
Downloaded from http://circ.ahajournals.org/ by guest on July 31, 2017
transfusion state with the one immediately preceding.
(The control hemodynamics were compared with the
500-ml values, the 500-ml values with the 1000-ml
values, and so forth.) In addition, the control values
were compared with the 1 500-ml values using a paired
t test.
Results
Figure 2 demonstrates the effects of acute volume
loading in one patient. Infusion of 1500 ml of whole
blood over 30 minutes significantly (p < 0.05, paired t
test) increased end-diastolic volume, and wedge
pressure increased from 4 mm Hg to 8 mm Hg. Ventricular output and work both increased. However,
ejection fraction progressively decreased, and endsystolic volume rose from 13 to 59 mI/M2. No significant change occurred in either systemic vascular
resistance or heart rate.
Table 1 summarizes the changes in hemodynamics
for all 15 patients. Significant (p < 0.05, paired t test)
increases in indexes of ventricular output (cardiac
index, stroke work index and blood pressure) were
accompanied by decreases in ejection fraction and
associated increases in the end-systolic volume index,
despite a decreased systemic vascular resistance. Pul-
capillary wedge pressure and end-diastolic
volume increased; PCW increased by approximately 1
mm Hg for each 16-ml increase in end-diastolic
volume. For purposes of discusion, an estimate of the
left ventricular wall tension at the start of systole (T)
was computed from the Laplacerelationship.20 Here,
wall thickness is unknown, and T = P X R, where P
represents mean systolic blood pressure and the radius
(R) is derived from EDV (EDV = 4/3 irR3). Significant increases in T were found in all patients.
In table 2, a comparison of the preoperative
catheterization findings with the intraoperative
volume loading results is shown for each patient. In
most cases the preoperative ejection fraction,
pulmonary wedge pressure and left ventricular enddiastolic volume were near the intraoperative ranges.
In patient 9, the values for ejection fraction differed
significantly. This patient suffered an acute myocardial infarction while on cardiopulmonary bypass.
Ventricular function curves for the 15 patients are
shown in figures 3A and 3B. For most patients, the
ascending limbs of the stroke volume and enddiastolic volume index curves were not particularly
steep (slopes generally less than 0.6), even over the
lower range of end-diastolic volumes. These curves
flattened at end-diastolic volumes of 70 mI/M2 or
monary
TABLE 1. Hemodynamic Changes After Volume T-ansfusion
cl
SVR
SVI
HR
BP
(dyn-secA\( 1
PCW
EDVI
cm5
min/m2/ (mi/M2)
(ml/m2) (mm Hg) (mm Hg) (beats/min)
SWI
(g-m/m2)
EF
ESVI
(ml/m2)
T
( dyn\
\cm/
Control
2.3
-0.6
41
-1-3
88
52
-
4
2
88
-5
1258
=11
2.1
0.1
28
20
0.70
-=2
1
=0.04
13
2
227
11
After 1500-ml transfusion
371
47
0.49
42
3.6
1024
38
83
113 - 7
7.6
-25
15
0.04
44
-2
i8
4
4
60
0.9
=0.1
Values are mean - SEM. Statistical analysis by paired t test. All changes were significant (p < 0.05), except for heart rate and
=
88
6
i
systemic vascular resistance.
Abbreviations: EDVI = end-diastolic volume index; PCW = pulmonary capillary wedge pressure; BP = blood pressure;
HR = heart rate; SVR systemicvascular resistance; CI =cardiac index; SWI = stroke work index; EF = ejection fraction;
ESVI = end-systolic volume index; T = estimate of left ventricular wall tension.
VOL 62, No 3, SEPTEMBER 1980
CI RCULATION
538
TABLE 2. Comparison of Preoperative and Intraoperative
Findings in the 15 Patients
Intraoperative volume
loading findings
Preoperative
catheterizationi
Chanige*
Downloaded from http://circ.ahajournals.org/ by guest on July 31, 2017
in
Change*
findings
PCW
in
PCW
LVEDV
(inm
(mm LVEDV Change*
in EF
Hg)
(ml)
Hg)
(ml)
9
143
0.72 0.53 8 13 91 136
14
104
0.61 0.53 5 13 91 135
113
0.69 0.33 3
9 93 170
7
8 46 119
9
100
0.79 0.54 2
0.66 0.53 :3
8
90
6 57 106
0.58 0.44 0
8 52 128
3
145
5
186
0.55
0.53
9 109 147
10
6 125 245
2
184
0.56 0.50 0
124
9
0.37 0.26 1 15 79 160
4 70 142
9
132
0.77 0.69 0
5 46 105
100
7
0.81 0.70 0
2
94
8 62 125
0.86 0.63 1
1
1 56 258
149
0.85 0.33 0
6 50 192
5
105
0.86 0.32 3
4
7 85 207
113
0.71 0.44 3
to the change from the corntrol value to the 1500-
Pt EF
1 0.72
2 0.67
3 0.39
4 0.80
5 0.82
6 0.70
7 0.60
8 0.62
9 0.72
10 0.62
11 0.60
12 0.82
13 0.81
14 0.79
15 0.76
*Refers
ml value.
Abbreviations: EF = ejection fraction; LVEDV = left
pulmnonary capilventricular end-diastolic volume; PCW
lary wedge pressure.
The relationship between pulmonary capillary
wedge pressure and end-diastolic volume index is
shown in figure 5. We discerned four patterns: (1) In
eight of the 15 patients, pulmonary capillary wedge
pressure increased progressively with end-diastolic
volume index, with an average slope of approximately
10 mm Hg per 50 ml/m2. (2) In two patients, the slope
was higher (approximately 20 mm Hg per 50 ml/m2).
One of these patients (indicated by filled square) had
an ejection fraction of 0.76 immediately before bypass
and 0.31 immediately after bypass. Evidence of an
acute anterolateral myocardial infarction (ECG,
CPK-M bands, 99mTc pyrophosphate uptake) was
found and was assumed to occur during bypass. The
second patient (indicated by asterisk) had to be placed
00
60
40
S
V
30
(ml/M2) 20
10
i
g
20
=
40
80
00
A
120
100
EDVfI (ml/M2)
e01
greater. The ascending limbs of the stroke work and
end-diastolic volume index curves were steeper.
However, in at least half of these patients a relative
flattening can be appreciated at end-diastolic volumes
exceeding 70 ml/m2.
Figure 3C shows the relationship between the ejection fraction and the end-diastolic volume index. In
nine of the 15 patients, the ejection fraction
monotonically decreased as the end-diastolic volume
index increased. In six patients, an initial increase in
ejection fraction preceded its progressive decrease.
The average decrease in ejection fraction was 30%
(range 9-63%). The average decrease in ejection fraction was 0.30 for each 50 ml/m2 increase in the enddiastolic volume index.
The relationship between cardiac index and ejection
fraction is shown in figure 4. At each level of infusion
(control, 500, 1000, 1250, and 1500 ml), the enddiastolic volume index, pulmonary wedge pressure,
ejection fraction and cardiac index were averaged and
the standard errors were obtained. The scales of enddiastolic volume-index and pulmonary wedge pressure
were chosen such that the normal range of pulmonary
wedge pressure (0-12 mm Hg, or 12 mm Hg) corresponded to the normal range of the end-diastolic
volume index (50-90 ml/m2, or 40 ml/m2). As infusion progressively increased these two measurements,
it also increased cardiac index and decreased ejection
fraction.
30
S
W
I
(gm
40.
30.
sec/M2)20
10.
i
20
.
.
.
00
40
B
1 .00.,
.
.
.
80
120
100
EDVI (mu/M2)
.80
E
F
.00.
.
40..
.20
0I
C
I
I
20
0
40
0
00
I
80
I
I
100
£
1
1
120
EE'VI (ml/M2)
FIGURE 3. (A) Stroke volume and (B) stroke work ventricular function curves and (C) ejection fraction (EF) curves
for 15 patients. SVI = stroke volume index; EDVI = enddiastolic volume index; SWI = stroke work index.
EFFECT OF INCREASING PRELOAD/Mangano et al.
.70 .f
Z
0
3.5
X
I-
0
<
LU
.03 :o
.60
99 0*
FIGURE 4. Relationship of ejection fraction and cardiac index tofilling volume (left)
and filling pressure (right). EDVI = enddiastolic volume index; PCW = pulmonary
capillary wedge pressure.
1
Z0I
-
539
2.5
.50
I-
C-
(0)
LU
EDVI
(m/M2)
PCW (mmHg)
Downloaded from http://circ.ahajournals.org/ by guest on July 31, 2017
back on bypass because of difficulty with the
anastomosis of the left anterior descending coronary
artery graft. After this, the volume infusion protocol
was performed. (3) In two patients, the pulmonary
wedge pressure did not change significantly (less than
3 mm Hg) until an end-diastolic volume index of 55
mI/M2 was reached. (4) In three patients, pulmonary
capillary wedge pressure increased by 5 mm Hg or less
over a large (> 50 ml/m2) range of end-diastolic
volume index. We could not correlate these findings
with preoperative angiographic ejection fraction, age,
number of vessels bypassed or ischemic time.
Discussion
In this study, ventricular function curves showed
myocardial depression over normal ranges of left ventricular end-diastolic volumes and pulmonary
capillary wedge pressures. In most patients, the ascending limb of the stroke volume curve had a shallow
slope, even over the lower range of end-diastolic
volumes. In approximately half of the patients, the
same was true for the stroke work curves. The stroke
volume and stroke work curves both flattened at enddiastolic volumes greater than 70 ml/m2. Because
these relationships have not been described before,
comparisons with other results cannot be made.
However, data relating ventricular output to filling
pressure in patients without cardiac disease and in
patients with congestive heart failure and acute myocardial infarction are available.4' 5 8-10 These results
suggest that our ventricular function curves are
depressed compared with normal curves, and that our
ventricular outputs are similar to those in patients
with acute myocardial infarction and congestive heart
failure. However, our results diverge from results after
infarction or failure because the ascending limb in our
studies flattened at normal rather than high values of
pulmonary capillary wedge pressure. Several factors
may explain these findings. First, four patients had
ventricular dyssynergy, and one had a depressed ejection fraction (0.39) preoperatively. Second, the effects
of cardioplegic arrest and cardiopulmonary bypass
may have reduced myocardial performance. Comparison with the preoperative catheterization data
(table 2) reveals that at nearly equivalent end-diastolic
volumes, the intraoperative ejection fraction appears
to be decreased. Although comparative afterload data
are not available, it appears that myocardial performance may be reduced after bypass. Third, pericardial
excision before the study may have altered both
diastolic compliance21 and systolic performance.22
Fourth, the pulmonary capillary wedge pressure may
not reflect the left atrial "kick" component to the enddiastolic pressure,23 and therefore may underestimate
the end-diastolic pressure. One factor that probably
did not play a role in terminating the ascent of the ventricular function curve is the anesthetic. Morphine sulfate plus oxygen does not depress contractility at the
doses used in this study.24 25
Ejection fraction decreased with volume loading in
all patients. In nine patients, ejection fraction
progressively decreased as end-diastolic volume increased from control values. In the remaining six
patients, a small initial increase in ejection fraction
preceded its progressive decrease. Although ejection
fraction is affected by changes in preload and
afterload, it is a useful indicator of ventricular systolic
performance and contractility. Studies in animals and
man using ventriculography, echocardiography, and
ultrasonic techniques indicate that ejection fraction
either increases or remains unchanged as end-diastolic
volume increases.2832 However, increases in afterload
1
p
C
2
8
W
(mm Hg)
-
120
EDVI (m1/M2)
FIGURE 5. Relationship between pulmonary capillary
wedge pressure (PCW) and end-diastolic volume index
(ED VI)for 15 patients. Filled square indicates a patient with
intraoperative infarction; asterisk indicates technical
difficulty with anastomosis.
540
CIRCULATION
Downloaded from http://circ.ahajournals.org/ by guest on July 31, 2017
consistently decrease ejection fraction.33 34 In our
study, ejection fraction progressively decreased
despite an increase in end-diastolic volume and a
decrease in systemic vascular resistance.
There are two explanations for these results. First,
myocardial contractility may have decreased with
volume loading. There is evidence that in the control
state the patients were relatively hypovolemic (table
2). Thus, contractility may have been enhanced during
the early volume loading periods. This is difficult to establish because no direct measurements of contractility were made. However, progressive volume
loading increased the rate of rise of the radial artery
pressure tracing in all patients. Although this may be a
weak index of ventricular contractility,35 it weighs
against the hypothesis that contractility progressively
decreased. Second, although systemic vascular
resistance either did not change or decreased with
volume loading, ventricular afterload may have increased as end-diastolic volume increased (Laplace
relationship).36 To illustrate this effect, we estimated
the wall tension at end-systole. The relationships between ejection fraction, systemic vascular resistance,
and T are shown in figure 6. A direct, rather than an
inverse, relationship exists between ejection fraction
and systemic vascular resistance. However, T increased by 63% with volume loading, and its
relationship with ejection fraction is an inverse one.
Thus, changes in T, and not systemic vascular
resistance, may have contributed to the decrease in
ejection fraction with volume loading. The Laplace
effect may play an important role in these patients.
Comparison of the ventricular function curve
results with the ejection fraction curve results (fig. 4)
reveals that increasing filling pressure by volume loading increases ventricular output; however, ejection
fraction decreases and end-systolic volume increases
over the range of normal pulmonary capillary wedge
pressures and end-diastolic volumes. We hypothesize
that the ejection fraction at any end-diastolic volume
is the slope of the line drawn from the origin to a point
on the stroke volume ventricular function curve at that
end-diastolic volume (ejection fraction equals stroke
volume divided by end-diastolic volume). Along the
flat portion of the ventricular function curve, stroke
volume is relatively constant; thus, the ejection fraction (slope) will decrease as end-diastolic volume in-
VOL 62, No 3, SEPTEMBER 1980
creases. In our patients, these curves were flat over the
normal ranges of end-diastolic volume, and the ejection fraction decreased over these ranges. The ejection fraction and ventricular function curve results are
therefore consistent. Based on this interpretation,
ejection fraction can be considered as a characterization of the Starling curve, and as such, is an index of
the effectiveness of the Starling mechanism. As ejection fraction decreases, a comparatively larger increase in end-diastolic volume is needed to effect the
same increase in stroke volume. With further
decreases in ejection fraction, use of the Starling
mechanism to increase cardiac output is less effective.
In our patients, such a situation exists over the normal
ranges of pulmonary wedge pressures and enddiastolic volumes. Though volume loading increased
cardiac output, it resulted in a disproportionately increased end-diastolic volume.
The relationship between pulmonary capillary
wedge pressure and end-diastolic volume was not consistent in this series of patients, as was suggested in
our previous study.'8 It appears that significant
biologic variability exists. A pulmonary capillary
wedge pressure of 4 mm Hg could be associated with
an end-diastolic volume index of 25-1 10 ml/m2. These
inconsistencies may be explained by differences in
pulmonary, atrial and ventricular diastolic compliance. Steep curves occurred in the two patients who
had difficulty during cardiopulmonary bypass (fig. 5,
filled square and asterisk). In contrast, the slopes of
such curves were relatively flat in the three patients
who presented without evidence of preoperative ventricular distension. Pulmonary capillary wedge
pressure was a rather insensitive indicator of enddiastolic volume in these three patients.
In conclusion, our results show that increasing filling pressure by volume loading has associated benefits
and costs. Volume loading from relatively low to normal filling pressures increases ventricular stroke work
and stroke volume, but it also increases T. In addition,
maximizing ventricular output by increasing filling
pressure is associated with a reduction in ejection fraction. Progressive decreases in ejection fraction indicate a decrease in the effectiveness of the Starling
mechanism over normal ranges of pulmonary
capillary wedge pressure and end-diastolic volume.
Based on the results of this study, we recognize the
.70r
.70
i.60 _1
++
.601
(-)
o.50
.50-
500
1000
1500
SVR (dyne-sec/cm5)
0
-
100
200
T (dyne/cm)
300
400
FIGURE 6. Relationship between ejection
fraction and systemic vascular resistance
(SVR) (left) and estimated wall tension (T)
(right).
EFFECT OF INCREASING PRELOAD/Mangano et al.
limitations of the Starling mechanism even over normal ranges of filling pressures in this group of patients, and therefore caution against discontinuation
from cardiopulmonary bypass at high filling pressures.
Acknowledgment
The authors thank Dr. Robert Hickey of the Veterans Administration Hospital, San Francisco, and Dr. Kanu Chatterjee of
the University of California, San Francisco, for advice and encouragement, and Dr. Hal Anger of the Donner Laboratory of
Medical Physics, University of California, Berkeley, for providing
the coaxial cardiac scintillation probe used in these studies.
References
Downloaded from http://circ.ahajournals.org/ by guest on July 31, 2017
1. Frank 0: Zur Dunamik des Herzmuskels. Z Biol 32: 370, 1895.
(translated by Chapman CB, Wasserman E) Am Heart J 58:
282, 467, 1959
2. Patterson SW, Piper H, Starling EH: The regulation of the
heart beat. J Physiol 48: 465, 1914
3. Sarnoff SJ, Berglund E: Ventricular function. I. Starling's law
of the heart studied by means of simultaneous right and left ventricular function curves in the dog. Circulation 9: 706, 1954
4. Russell RO, Rackley CE, Pombo J, Hunt D, Potauin C, Dodge
HT: Effects of increasing left ventricular filling pressure in
patients with acute myocardial infarction. J Clin Invest 49:
1539, 1970
5. Crexells C, Chatterjee K, Forrester JS, Dikshit K, Swan HJC:
Optimal level of filling pressure in the left side of the heart in
acute myocardial infarction. N Engl J Med 289: 1263, 1973
6. Bishop VS, Stone HL, Guyton AC: Cardiac function curves in
conscious dogs. Am J Physiol 207: 677, 1964
7. Frye RL, Braunwald E: Studies on Starling's law of the heart.
I. The circulatory response to acute hypervolemia and its
modification by ganglionic blockade. J Clin Invest 39: 1043,
1960
8. Morton MJ, McAnulty JH, Rahimtoola SH: Ventricular function curve from a single diagnostic left ventriculogram: technique, results and value. Am J Cardiol 41: 711, 1978
9. Ross J, Braunwald E: Studies on Starling's law of the heart. IX.
Effects of impeding venous return on performance of normal
and failing human left ventricle. Circulation 30: 719, 1964
10. Miller RR, Vismara LA, Zelis R, Amsterdam EA, Mason DT:
Clinical use of sodium nitroprusside in chronic ischemic heart
disease. Effects on peripheral vascular resistance and venous
tone and on ventricular volume, pump and mechanical performance. Circulation 51: 328, 1975
11. Weisel RD, Lipton IH, Lyall RN, Baird RJ: Cardiac
metabolism and performance following cold potassium cardioplegia. Circulation 58 (suppl I): 1-217, 1978
12. Wexler JP, Blaufox MD: Radionuclide evaluation of left ventricular function with nonimaging probes. Semin Nucl Med 9:
310, 1979
13. Van Dyke D, Anger HO, Sullivan RW, Vetter WR, Yano Y,
Parker HG. Cardiac evaluation from radioisotope dynamics. J
Nucl Med 13: 585, 1972
14. Steel PP, Van Dyke D, Trow RS, Anger HO, Davies H: A simple and safe bedside method for serial measurement of left ventricular ejection fraction, cardiac output and pulmonary blood
volume. Br Heart J 36: 122, 1974
15. Grach MW, Gottlieb S, Mallon SM, Miale A Jr: A new dualprobe system for the rapid bedside assessment of left ventricular function. J Nucl Med 17: 930, 1976
16. Berndt T, Alderman EL, Wasnich R, Hsieh SC, Van Dyke DC,
Harrison DC: Evaluation of portable radionuclide method for
measurement of left ventricular ejection fraction and cardiac
output. J Nucl Med 16: 289, 1975
541
17. Parker H, Weber P, Van Dyke D, Davies H, Steele P, Sullivan
R: Evaluation of central circulatory dynamics with the
radionuclide angiocardiogram. In Cardiovascular Nuclear
Medicine, edited by Strauss HW, Pitt B, James AE Jr. St.
Louis, CV Mosby Co, 1974
18. Ellis RJ, Mangano DT, Van Dyke DC: Relationship of wedge
pressure to end-diastolic volume in patients undergoing
myocardial revascularization. J Thorac Cardiovasc Surg 78:
605, 1979
19. Ellis RJ, Mangano DT, Van Dyke DC, Ebert PA: Protection of
myocardial function not enhanced by high concentrations of
potassium during cardioplegic arrest. J Thorac Cardiovasc
Surg 78: 698, 1979
20. Mirsky I: Basic terminology and formulae for left ventricular
wall stress. In Cardiac Mechanics: Physiological, Clinical, and
Mathematical Considerations, edited by Mirsky I, Ghista DN,
Sandler H. New York, Wiley, 1974
21. Shirato K, Shabetai R, Bhargava V, Franklin D, Ross J Jr:
Alteration of the left ventricular diastolic pressure-segment
length relation produced by the pericardium: effects of cardiac
distension and afterload reduction in conscious dogs. Circulation 57: 1191, 1978
22. Berglund E, Sarnoff SJ, Issacs JP: Ventricular function. Role
of the pericardium in regulation of cardiovascular hemodynamics. Circ Res 3: 133, 1955
23. Rahimtoola SH, Loeb HS, Ehsani A, Sinno MZ, Chuquimia
R, Lal R, Rosen KM, Gunnar RM: Relationship of pulmonary
artery to left ventricular diastolic pressures in acute myocardial
infarction. Circulation 46: 283, 1972
24. Strauer BE: Contractile response to morphine, piritramide,
meperidine and fentanyl. Anesthesiology 37: 304, 1972
25. Lowenstein E, Hallowell P, Levine FH, Daggett WM, Austen
WG, Laver MB: Cardiovascular response to large doses of intravenous morphine in man. N Engl J Med 281: 1389, 1969
26. Taylor RR, Cingolani HE, McDonald RH: Relationships
between left ventricular volume, ejected fraction, and wall
stress. Am J Physiol 211: 674, 1966
27. Mitchell JH, Wildenthal K, Mullins CB: Geometrical studies of
the left ventricle utilizing biplane cinefluorography. Fed Proc
28: 1334, 1969
28. Krayenbuehl HP, Blussman WD, Turina M, Luthy E: Is the
ejection fraction an index of myocardial contractility? Cardiologia 53: 1, 1968
29. Mahler F, Ross J, O'Rourke RA, Covell JW: Effects of
changes in preload, afterload and inotropic state on ejection and
isovolumic phase measures of contractility in the conscious dog.
Am J Cardiol 35: 626, 1975
30. Holt JP: Effect of plethora and hemorrhage on left ventricular
volume and pressure. Circ Res 5: 273, 1957
31. Quinones MA, Gaasch WH, Cole JS, Alexander JK: Echocardiographic determination of left ventricular stress-velocity
relations in man with reference to the effects of loading and contractility. Circulation 51: 689, 1975
32. Rankin LS, Moos S, Grossman W: Alterations in preload and
ejection phase indices of left ventricular performance. Circulation 51: 910, 1975
33. Tsakiris AG, Vandenberg RA: Variations of left ventricular
end-diastolic pressure, volume, and ejection fraction with
changes in outflow resistance in anesthetized intact dogs. Circ
Res 23: 213, 1968
34. MacGregor DC, Covell JW, Mahler F, Dilley RB, Ross J Jr:
Relations between afterload, stroke volume, and the descending
limb on Starling's curve. Am J Physiol 227: 884, 1974
35. Mason DT: Usefulness and limitations of the rate of rise of intraventricular pressure (dP/dt) in the evaluation of myocardial
contractility in man. Am J Cardiol 23: 516, 1969
36. Burton AC: Importance of shape and size of heart. Am Heart J
54: 801, 1957
The effect of increasing preload on ventricular output and ejection in man. Limitations of
the Frank-Starling Mechanism.
D T Mangano, D C Van Dyke and R J Ellis
Downloaded from http://circ.ahajournals.org/ by guest on July 31, 2017
Circulation. 1980;62:535-541
doi: 10.1161/01.CIR.62.3.535
Circulation is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 1980 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/62/3/535.citation
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/