Download Ventriculoarterial Coupling in Normal and Failing Heart in Humans

483
Ventriculoarterial Coupling in Normal
and Failing Heart in Humans
Hidetsugu Asanoi, Shigetake Sasayama, and Tomoki Kameyama
Downloaded from http://circres.ahajournals.org/ by guest on June 15, 2017
To investigate coupling between the heart and arterial system in normal subjects and cardiac
patients, we determined both the slope of the left ventricular end-systolic pressure-volume
relation (ventricular clastancc) and the slope of the arterial end-systolic pressure-stroke volume
relation (effective arterial elastance) in three groups of subjects: group A, 12 subjects with
ejection fraction of 60% or more; group B, seven patients with ejection fraction of 40-59%; and
group C, nine patients with ejection fraction of less than 40%. We also determined the left
ventricular stroke work, end-systolic potential energy, and the ventricular work efficiency
defined as stroke work per pressure-volume area (stroke work+potential energy). In group A,
ventricular elastance was nearly twice as large as arterial elastance. This is a condition for a
maximal mechanical efficiency. In group B, ventricular elastance was almost equal to arterial
elastance. This is a condition for maximal stroke work from a given end-dlastollc volume. In
group C, ventricular elastance was less than one half of arterial elastance, which resulted in
increased potential energy and decreased work efficiency. Thus, the present study suggests that
ventriculoarterial coupling is normally set toward higher left ventricular work efficiency,
whereas in patients with moderate cardiac dysfunction, ventricular and arterial properties are
so matched as to maximize stroke work at the expense of the work efficiency. Neither the stroke
work nor the work efficiency is near maximum for patients with severe cardiac dysfunction.
{Circulation Research 1989;65:483-493)
T
he ventricle is a generator of hydraulic energy,
which transfers the mechanical energy of
contraction to the blood accumulated in the
ventricular chamber.1'2 The stroke work (SW) and
power output from a given end-diastolic volume
are strongly influenced by the input impedance of
the arterial system. 3 - 5 In physics and engineering,
an energy source and its load are considered
matched when a maximal amount of energy is
transferred from the source to the load. The
matching occurs when the input impedance of the
load equals the output impedance of the energy
source. Using isolated heart preparations, several
investigators confirmed the validity of this matching concept for the coupling of the real ventricle
with simulated arterial loads. 6 - 8 That is, they
verified that the SW was maximized when the
afterloaded impedance was close to the internal
From the Second Department of Internal Medicine, Toyama
Medical and Pharmaceutical University, Toyama, Japan.
Presented at the 36th Scientific Session of the American
College of Cardiology March 9, 1987.
Address for correspondence: Shigetake Sasayama, MD, The
Second Department of Internal Medicine, Toyama Medical and
Pharmaceutical University, 2630 Sugitani, Toyama 930-01, Japan.
Received September 3, 1987; accepted February 7, 1989.
impedance of the ventricle. This does not mean,
however, that the physiological control mechanisms indeed adopt this criterion for optimal coupling between the heart and artery under physiological circumstances in vivo.
Another criterion for optimal coupling between
an energy source and its load is the principle of
economical fuel consumption, or mechanical efficiency. In the case of heart, the mechanical efficiency is defined as the ratio of SW to myocardial
oxygen consumption per beat (MVo?). Several investigators looked into this efficiency of cardiac
pump.1'9-11 They came to a similar theoretical conclusion that the mechanical efficiency from a given
end-diastolic volume becomes maximal when arterial impedance is nearly one half the cardiac output
impedance. Again, it is not known at all whether the
physiological controller of the heart and arterial
system prefers this efficiency-oriented criterion at
rest and the work-maximization criterion in stressful conditions.
In the present study, therefore, we investigated
the resting human's matching of the ventricular
properties quantified by the slope of end-systolic
pressure-volume (P-V) relation (Era) with arterial
load properties expressed by the slope of end-
484
Circulation Research
Vol 65, No 2, August 1989
TABLE 1. Individual Data of Left Ventricniar and Arterial End-Systolic Pressure-Volume Relation
EDVI
ESVI
SV1
2
(ml/m )
EF
(%)
ESP PCW
(mm Hg)
HR
(beats/min)
Vo
E.
(mm Hg/ml/in2)
Ee,
2
rvalue
(ml/m )
0.98
0.90
0.96
0.96
0.91
0.93
0.96
0.97
0.93
0.95
0.98
0.90
0.94
0.03
-10
3.3
2.2
11
5.3
1.6
Group A
1
81
29
53
65
117
5
72
2
86
29
57
66
92
4
55
3
103
38
65
63
125
11
86
4
5
83
63
16
67
81
108
58
22
41
65
107
6
79
32
48
60
71
7
69
28
42
60
94
8
...
...
...
77
66
67
Downloaded from http://circres.ahajournals.org/ by guest on June 15, 2017
8
86
28
59
68
76
9
73
26
47
65
84
10
100
39
61
61
93
11
63
17
47
75
89
12
Mean
70
42
60
73
80
28
28
52
66
94
7
69
SD
13
7
9
6
17
3
10
13
102
56
47
44
145
8
77
14
121
71
50
42
75
7
15
92
43
48
53
108
8
100
67
16
76
35
41
54
98
10
60
17
126
57
69
55
73
65
18
119
64
54
104
47
44
52
44
105
55
57
53
8
11
11
49
6
60
68
18
125
104
26
1
17
NS
NS
NS
<0.05
NS
20
125
85
40
32
109
6
67
21
120
90
30
25
112
18
92
3
61
...
...
. ..
81
64
58
78
-7
3.7
1.5
1.9
3
3.1
1.7
4
4.5
1.8
12
2.0
0.4
0.67
0.30
0.20
0.34
0.76
0.50
0.35
0.37
0.62
0.32
0.51
0.55
0.46
0.17
0.88
0.96
0.97
0.96
0.95
0.97
0.93
0.95
0.03
-3
2.5
3.1
1.24
21
1.4
17
4.0
1.5
2.3
-1
2.8
2.4
-5
1.2
1.1
-9
1.7
3.8
2.5
2.8
2.1
NS
NS
26
9.8
1.9
-9
-10
8
12
4.7
1.6
2.6
3.4
3.0
1.5
6.3
2.2
1.3
6
3.5
-3
2.9
1.8
20
4.7
Group B
19
Mean
SD
p (A vs. B)
Group C
101
NS
22
129
82
47
36
23
202
159
43
21
117
74
26
63
60
24
156
120
36
23
93
16
56
25
81
58
24
29
80
6
71
26
113
79
34
30
80
17
46
27
138
103
35
25
90
14
69
28
206
141
176
106
30
35
14
85
19
67
26
93
15
66
41
39
7
7
16
7
<0.05
NS
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
Mean
SD
p (A vs. Q
p (B vs. C)
13
0.97
0.97
0.95
0.98
0.94
0.91
0.95
0.93
0.92
0.95
0.02
NS
NS
NS
NS
NS
NS
1.1
0.7
1.03
0.58
0.86
0.92
0.94
0.74
0.90
0.21
NS
NS
NS
18
1.7
2.8
38
2.1
3.7
16
1.8
55
0.7
2.5
1.7
37
1.1
2.6
26
2.6
3.4
27
1.5
25
1.0
50
32
0.6
1.5
2.4
2.6
2.9
14
0.7
0.6
<0.05
NS
<0.05
NS
1.63
1.78
1.40
2.48
2.36
1.31
1.60
2.63
4.83
2.56
2.03
<0.05
NS
29
7
15
<0.05
<0.05
1.6
2.7
Based on resting left ventricular ejection fraction, subjects were divided into three groups: group A, ejection fraction £60%; group B,
ejection fraction 40-59%; and group C, ejection fraction s39%. EDVI, end-diastolic volume index; ESVI, end-systolic volume index;
SVI, stroke volume index; EF, ejection fraction; ESP, end-systolic pressure; PCW, pulmonary capillary wedge pressure; HR, heart rate;
Vo and E n , volume intercept and slope of left ventricular end-systolic pressure-volume relation; E ^ slope of arterial end-systolic
pressure-stroke volume relation; SD, standard deviation of the mean; NS, not statistically significant.
systolic pressure-stroke volume (P-SV) relation
under normal and variably depressed cardiac conditions, with special reference to those two matching principles described above. That is, we estimated E^, E t , SW, MVOj, and total mechanical
energy released per contraction from the P-V data
and studied how the relative magnitudes of E ra and
E, are in these subjects with different degrees of
cardiovascular stress.
Subjects and Methods
Subjects
Eight normal subjects (aged 19-62 years; mean,
32.0 years) with no symptoms and signs of cardiac
Asanoi et al Ventriculoarterial Coupling in Humans
485
mmHg
100
0
FIGURE
1.
Simultaneous
Downloaded from http://circres.ahajournals.org/ by guest on June 15, 2017
recordings of left ventricular
echocardiogram and direct
arterial pressure. Left ventricular volume was determined
by the formula of TeichhoLz et
al,12 and ventricular endsystolic pressure was approximated from the arterial
dicrotic pressure. EDD, enddiastolic dimension; ESD, endz systolic dimension; ESP, endsystolic pressure.
L
1 sec
disease, four patients with atypical chest pain, and
16 patients (aged 22-75 years; mean, 56.5 years)
with cardiac dysfunction formed the study group
(Table 1). All patients had supporting clinical, chest
radiographic, and echocardiographic evidences of
impaired left ventricular function. Severity of cardiac disease ranged from class II to class III by New
York Heart Association functional classification,
and the causes of the left ventricular impairment
were dilated cardiomyopathy in 13 subjects and
aortic regurgitation in three subjects. Twodimensional echocardiography or contrast left ventriculography revealed diffuse and uniform impairment of the left ventricular contraction pattern in all
patients. Patients with regional wall-motion abnormalities and mitral regurgitation were excluded in
this study. Coronary arteriography was performed
in four patients with atypical chest pain and in 12
patients with cardiac dysfunction. None of these
patients showed a significant narrowing of major
coronary arteries. In 17 patients, pulmonary capillary wedge pressure was obtained during right heart
catheterization. Diuretics had been used in 10
patients, and isosorbide dinitrate had been used in
three patients before the study. All these medications were withheld for 24 hours with patients under
close observation. Informed consent and ethical
approval were obtained from all subjects.
Subjects were divided into three groups based on
their resting left ventricular ejection fraction (EF)
determined by echocardiography. Group A consisted of 12 subjects with left ventricular EF of 60%
or more. Group B consisted of seven patients with
mild left ventricular dysfunction in whom EF was
40-59%. Group C consisted of nine patients with
more marked left ventricular dysfunction in whom
EF was less than 40%.
Systolic Pressure and Left Ventricular Volume
All patients were in regular sinus rhythm and
were studied in the supine postabsorptive state. A
19-gauge cannula was inserted percutaneously into
a brachial artery and was connected to a straingauge manometer {model P 50, Spectramed, San
Juan, Puerto Rico). After control recordings at rest,
phenylephrine (5 mg/100 ml) was infused intravenously to increase systolic pressure in gradual increments of approximately 20 mm Hg. Subsequently,
adequate recovery time was allowed for peak systolic pressure to return to the baseline level, and
then a sodium nitroprusside infusion (50 mg/500 ml)
was started to decrease systolic pressure by about
20 mm Hg. Thus, the systolic pressure was changed
by about 40 mm Hg during these interventions.
Two-dimensional targeted M-mode echocardiograms of left ventricular cavity were recorded by a
486
Circulation Research Vol 65, No 2, August 1989
Toshiba SHA-60A with a 3-MHz transducer
(Toshiba, Japan) used simultaneously with arterial
pressure (Figure 1). All data were recorded at a
paper speed of 50 mm/sec. The end-diastolic diameter was obtained at the peak of the R wave of the
electrocardiogram, and the end-systolic diameter at
the initial component of the second heart sound.
Left ventricular volume was determined with the
formula of Teichholz et al.12
Left ventricular end-systolic pressure was
approximated from the arterial dicrotic pressure,
which is considered to be caused by aortic valve
closure. It was clearly discernible on pressure
wave form in all patients.
Downloaded from http://circres.ahajournals.org/ by guest on June 15, 2017
Matching Analysis
Ventriculoarterial matching was analyzed in the
framework recently developed by Sunagawa et
ajs, i3,i4 a n c j Burkhoff and Sagawa.11 Namely, the
ventricular output impedance properties were quantified in terms of E a . The arterial input impedance
properties were expressed in terms of the effective
arterial elastance, E,.
Ventricular end-systolic pressure-volume relation. It has been shown that the ventricular endsystolic P-V relation is linear, and the ventricular
contractile properties can be quantified primarily by
its slope, EM, with the aid of its volume axis
intercept, Vo.15"17 To obtain the end-systolic P-V
relation, we plotted more than five different dicrotic
arterial pressures against corresponding left ventricular end-systolic volumes during the pharmacological pressure manipulation in each subject. Then,
linear regression of these pressures on the volumes
was performed to determine EM (Figure 2A).
Arterial end-systolic pressure-stroke volume relation. Given a constant heart rate, arterial endsystolic pressure changes with stroke volume in a
roughly linear relation (Figure 2B). The slope of this
relation is in proportion to the impedance that the
arterial tree offers to the stroke flow. Thus, the
arterial properties can be represented as a first approximation by the slope of the arterial end-systolic P-SV
relation. Sunagawa et al13 called this slope effective
arterial elastance, E,. We superimposed this arterial
end-systolic P-V relation on the ventricular endsystolic P-V relation in the same P-V plane by
transposing the stroke volume axis of the arterial
end-systolic P-SV relation line and letting its origin
fall on the end-diastolic volume point of the ventricular end-systolic P-V relation line (Figure 2Q. This
superposition enables a graphical coupling of the
ventricular output properties with the arterial input
impedance properties. The equilibrium end-systolic
pressure and volume that should exist when the
ventricle is coupled with the arterial system can be
obtained as the intersection between these two endsystolic P-V relation lines (Figure 2C). Conversely, it
can be seen in Figure 2C that if we connect the
equilibrium point (I) with end-diastolic volume, the
slope of this line represents Eo.
Pv
Vo
w
EDV
B
/
/
/
Pa
/
sv
ro
V
EDV
FIGURE 2. The framework of analysis for coupling the
ventricle with the arterial load. The mechanical characteristics of the left ventricle are expressed by the endsystolic pressure {Pvj-volume (V) relation (A). The mechanical characteristics of the arterial system are expressed
by the arterial end-systolic pressure (Pa)-stroke volume
(SV) relation (B). THE arterial Pa-V relation can be
superimposed on the ventricular Pv-V relation in the
same pressure (P)-volume (V) plane. The equilibrium
end-systolic pressure and volume when the ventricle is
coupled with the arterial system are obtained from the
intersection (I) between these two P-V relation lines (Q.
Closed circles, baseline state; open circles, phenylephrine; open triangles, nitroprusside. Ees and Vo, slope and
volume axis intercept of ventricular end-systolic pressurevolume relation; EDV, end-diastolic volume; Ea, slope of
arterial end-systolic pressure-stroke volume relation.
Asanoi et al Ventriculoarterial Coupling in Humans
487
line, end-diastolic P-V relation curve, and the systolic segment of the P-V trajectory.
Ees
Downloaded from http://circres.ahajournals.org/ by guest on June 15, 2017
Vo
ESV
EDV (mt
3. Schematic representation of ventriculoarterial coupling on the pressure-volume plane. The shaded
area represents the stroke work (SW) and the triangular
area shows end-systolic potential energy (PE). Maximum
stroke work (SWmax) for a given end-diastolic volume
can be determined when the slope of the ventricular
end-systolic pressure-volume relation (Ees) is assumed to
be equal to that of the arterial end-systolic pressurestroke volume relation (Ea). Closed circle, baseline state;
open circles, phenylephrine; open triangles, nitroprusside; ESPf calculated end-systolic pressure; Vo, volume
axis intercept of ventricular end-systolic pressure-volume
relation; ESVf calculated end-systolic volume; EDV,
end-diastolic volume.
FIGURE
Stroke Work and Work Efficiency
Ventricular P-V loop area accurately represents
SW of the ventricle. For simplicity, we assumed
that the P-V loop could be regarded as a rectangle
whose height was end-systolic pressure and whose
width was stroke volume. Under the simplifying
assumption, SW was calculated as end-systolic
pressure x (end-diastolic volume -end-systolic volume). End-systolic pressure and end-systolic volume are the graphically determined data points at
which two end-systolic P-V relation lines intersect
with each other (Figure 3). The values of SW, E ej ,
and E, in three groups were surveyed in reference
to the theoretical and experimental analyses by
Sunagawa et al8 that SW was maximized with
E.=E ra . As an index of the efficiency of ventricular
contraction, we studied the ratio of SW to the
systolic P-V area (PVA), which has been defined by
Suga as the total mechanical energy of ejecting
contraction. Piene and Sund studied this ratio in
their analysis of right ventricular coupling with
pulmonary arterial impedance. They called this
ratio pump efficiency. The ratio will be termed work
efficiency in the present paper. The systolic PVA
(the sum of SW and end-systolic potential energy in
Figure 3) is defined by the end-systolic P-V relation
Myocardial Oxygen Consumption
Recently, Suga and his colleagues18-19 have found
that PVA linearly correlates with MVO2 per beat.
This relation is characterized by a slope (a) and an
intercept (b) as MV02=axPVA+b. However, when
basal inotropic state differs among individual cases,
PVA cannot be an alternative of MV02 because of
the difference in intercept b.20-21 Accordingly, we
tried to calculate actual mechanical efficiency by
measuring myocardial oxygen consumption from
thermodilution coronary sinus flow and arterialcoronary sinus oxygen difference. This measurement was available among 12 of 28 patients (four in
group A, three in group B, and five in group C). In
eight subjects, these two measures were obtained
separately, but all measurements of myocardial
oxygen consumption were performed within 2 days
after determination of SW at the same resting heart
rate and blood pressure.
Proximal coronary sinus catheterization was performed with a dual thermistor thermodilution catheter (Webster Laboratories Incorporated, Altadena, California). Thermodilution coronary sinus
blood flow was determined in duplicate with standard technique22; the proximal thermistor was carefully placed in the ostium of the coronary sinus.
Position was confirmed by both fluoroscopic appearance throughout the study and hand injection of
small amounts of contrast medium. Arterial and
coronary sinus blood samples were drawn simultaneously for the determination of oxygen saturation.
Blood oxygen content was calculated as the product
of the percent oxygen saturation and the oxygenhemoglobin binding capacity. Myocardial oxygen
consumption was calculated as the product of the
coronary sinus flow and the arterial-coronary sinus
oxygen difference.
Since SW and myocardial oxygen consumption
are the energy units expressed by mm Hgxml and
ml O2, respectively, these units were converted into
a common unit of energy, the joule (J), with the
following conversions: 1 mm Hgxml = 1.33xlO~4J,
1 ml O2=20J. Mechanical efficiency was expressed
conventionally by (SWx heart rate)/myocardial oxygen consumption. Since Burkhoff and Sagawa11
predicted that the maximal mechanical efficiency of
ventricular contraction from a constant preload is
gained when E ^ E ^ / 2 , we looked at the ratio of Eo
to Ee, with this theoretical conclusion in mind.
Methodological Considerations
Reproducibility for echocardiographic volume.
The intraobserver differences for 23 duplicate measurements of left ventricular volume were 0.3±4.2
ml (r=0.99, p<0.001) for end-diastolic volume and
0.2±2.6 ml (r=0.99,p<0.001) for end-systolic volume, respectively. The interobserver differences
for duplicate studies were 4.1 ±4.8 ml (r=0.99,
488
Circulation Research
Vol 65, No 2, August 1989
ECHO - O.77 AHG»+ia7O
r-0.92 n —73
P<OO1
ANGIO VOLUME (ai)
Downloaded from http://circres.ahajournals.org/ by guest on June 15, 2017
FIGURE 4. Relation between echocardiographic (ECHO)
and contrast ventriculographic (ANGIO) volume in 75
tracings. There is a good correlation between the two
measurements.
/?<0.001) at end diastole and 3.9±3.3 ml (r=0.99,
/?<0.001) at end systole.
The correlation between 73 left ventricular volumes encompassing a wide range of volumes (44
end-diastolic volumes, 29 end-systolic volumes)
determined by echocardiography and contrast ventriculography was 0.92 (/?<0.01). However, the
echocardiographic method slightly underestimated
left ventricular volume (Figure 4), presumably due
to underestimation of the long axis of left ventricle.
Left ventricular end-systolic pressure. To validate the use of dicrotic pressure of the brachial
artery as a measure of left ventricular end-systolic
pressure, we compared left ventricular pressure
obtained by micromanometer-tip catheter and brachial arterial pressure in groups of patients with
different contractile state (six patients with
EF<40% [group 1] and 12 patients with EF^40%
[group 2]). The comparison was also made at
several pressure levels during phenylephrine infusion in the single subject. The dicrotic artery
pressure was lower than the left ventricular endsystolic pressure by 2.6±1.6 mm Hg in group 1
and by 3.7±2.1 mm Hg in group 2. However,
these differences were so slight as to be almost
negligible relative to the pressure changes for the
determination of the ventricular end-systolic P-V
relation. When vascular tone was changed by
phenylephrine in four patients, dicrotic arterial
pressure (y) changed linearly with left ventricular
end-systolic pressure (x) so that _y=l.Qx:-0.6,
r=0.98 QxO.001).
Statistical Analysis
Data are expressed as mean±SD. The statistical
significance of differences in hemodynamic variables among three groups was tested by analysis of
variance, and multiple comparisons were made by
the Bonferroni method. Values oip<5% were considered to represent a statistical significance.
Results
Baseline Cardiac Function
The data are listed for all subjects in Table 1.
There was no significant difference in heart rate and
end-systolic pressure among the three groups. Enddiastolic volume increased with reduction in EF,
but only differences between group A and group C
reached the statistical significance. End-systolic
volume was also greater in group C than in group A
and group B. Stroke volume in group C was significantly less than in group A and group B. Pulmonary capillary wedge pressure was elevated in only
six patients in group C, but this elevation was not
greater than 20 mm Hg, except in one patient.
Ventricular and Arterial Properties
End-systolic pressure was elevated by 20 ±8
mm Hg with phenylephrine and decreased by 18±8
mm Hg with nitroprusside. Alterations in heart rate
during pressure changes were 13 ±8 beats/min. A
linear relation between corresponding end-systolic
pressure and end-systolic volume was observed in
all patients during afterload change (Table 1). Representative P-V data and end-systolic P-V relations
for patients of each group are shown in Figure 5.
Volume axis intercept (Vo) in group C was significantly greater than in group A and group B. Era
significantly differed between group A and group C
but not between group A and group B and between
group B and group C. The effective arterial elastance,
E,, showed an increase with reduction in EF, and
the difference between group A and group C was
statistically significant.
The ratio of Ea to Ea (EJEa) showed a progressive increase as EF decreased. This ratio in subjects in group A was significantly lower than those
with marked ventricular dysfunction (group C),
but the differences between group A and group B
and between group B and group C were not
statistically significant.
Left Ventricular Work
Individual data related to SW and PVA are listed
in Table 2. SW tended to be reduced in group C as
compared with that in group A and group B, but the
differences did not reach statistical significance.
End-systolic potential energy in patients with cardiac dysfunction (group B and group C) was significantly greater than in group A. There were no
significant differences in the PVA among three
groups. The ratio of SW to total mechanical energy
(PVA) was highest in group A and progressively
decreased with the reduction of EF; significant
differences existed among the three groups.
We also calculated the ratio of SW to SW,,,,* for a
given end-diastolic volume, where S W ^ is the
maximal SW obtainable if E, were equal to Era
Asanoi et al Ventriculoarterial Coupling In Humans
r-0.96
Ees • 6.3mmHg /ml /m*
Vo - I 2ml/m 1
Ea
i
""»
ife
Ite
Jbo
Left Ventricular Volume (ml/m 2 )
r.0.96
Ees . I 4 mmHg/hil/m1
Vo -21 ml/m 1
Ea • I SmmHg/ml/m1
£ 100-
£
Downloaded from http://circres.ahajournals.org/ by guest on June 15, 2017
•5
u
50
I0O
150
20O
Left Ventricular Volume (ml/m 2 )
r - 0 98
Ees - 07mmHg/ml/m*
Vo
•55m\/n\
Eo - I 7mmHg/ml/m*
30
100
Left Ventricular
VcJune
ISO
200
(ml/m 2 )
FIGURE 5. Graphs of representative end-systolic pressurevolume relation and end-systolic pressure-stroke volume
relation before (solid line) and after (dashed line) afterload manipulation for subjects of group A (patient 7,
upper panel), of group B (patient 14, middle panel), and
of group C (patient 23, bottom panel). Slopes (Ees,
baseline Ea), volume intercept (Vo) and r value are
shown, Ees in group A is greater than baseline Ea, and
Ees in group B is almost equal to baseline Ea; Ees in
group C is less than baseline Ea. Closed circle, baseline
state; open circles, phenylephrine; open triangles, nitroprusside. Ees, slope of left ventricular end-systolic pressure-volume relation; Ea, slope of arterial end-systolic
pressure-stroke volume relation.
(Figure 3). This ratio was found highest in group B
(98.3%) as expected from the fact that Ea was nearly
equal to Era in this group. No significant difference
in SW/SWnu, was observed among groups.
Left Ventricular Mechanical Efficiency
Table 3 listed the individual data of myocardial
oxygen consumption and mechanical efficiency.
489
Mechanical efficiency tended to be decreased in
group C; the averages were 30 ±3% in group A,
24±6% in group B, and 21 ±6% in group C, but
these changes did not reach statistical significance.
Discussion
To the best of our knowledge, this is the first
study in humans where the coupling between the
ventricular pump and arterial afterload is investigated in terms of end-systolic elastance of the
ventricle and effective input elastance of the arterial
tree. In group A, with normal EF, the ventricle was
in a good contractile state with E^ of 4.5 ±2.0
mm Hg/ml/m2 and EF of 60% or more. We found
that their E, was always set less than E^ and
resulted in a greater work efficiency and a smaller
SW for the given contractility E a and preload than
when Ee, was equal to E,. In those patients whose
ventricles were in a mildly depressed state with Eeg
of 2.5±1.1 mm Hg/ml/m* and EF of 50%, E. was
found to be nearly equal to E o . With this E,/E M
ratio, their ventricles were performing at almost
maximal SW possible at a given preload. In those
patients whose ventricles were severely depressed
with EM of 1.5+0.7 mm Hg/ml/m2 and EF of less
than 40%, the E a /E ra ratio (2.56±2.03) was substantially higher than in the normal group (0.46±0.17)
and the mildly depressed heart group (0.90±0.21).
To our knowledge there has been no clinical evaluation of E./Ej, ratio in normal subjects and patients
with different degrees of ventricular depression.
Potential limitations of the method should be
noted. Minor fluctuations of contractility due to
baroreflex-mediated alterations in sympathetic discharge cannot be excluded. However, Suga et al23
reported that the changes in the slope of endsystolic P-V line either by carotid sinus or aortic
baroreceptor reflexes were only about 13% when
arterial pressure was changed between 100 mm Hg
and 150 mm Hg. Besides, Vatner et al24 found that
in the conscious animals, the baroreflex control of
cardiac contractility was even weaker than in the
anesthetized state. Therefore, the reflex change in
E ej in our study could be negligible. Although
echocardiography offers convenient means to measure cardiac dimension noninvasively, it is not
certain if the end-systolic dimension obtained represents actual end-systolic volume. However, when
the left ventricle contracts in a uniform and symmetric pattern, a close correlation is found between
echocardiographic and angiographic volume measurements in our laboratory and others. Therefore, we chose the patients without any evidence
of regional wall motion abnormalities. Strictly
speaking, SW should be calculated from the P-V
loop as the difference between the systolic work
(output of the ventricle) and the diastolic work
performed in distending the ventricle during diastole (input to the ventricle). However, we disregarded the latter for simplification. This could
cause a larger eiror in some of the patients whose
490
Circulation Research Vol 65, No 2, August 1989
TABLE 2. Individual Data of Pressure-Volume Area and Work Efficiency
ESvi
SVI
1
ESP
SW
PVA
SW/SW^
SW/PVA
2
(mm Hg)
(ml/m )
PE
(mm Hg-ml/m )
(%)
Group A
1
27
54
120
2
28
93
3
38
4
5
6
15
21
32
27
28
25
39
17
27
27
7
58
65
68
42
7
8
9
10
11
Downloaded from http://circres.ahajournals.org/ by guest on June 15, 2017
12
Mean
SD
Group B
13
14
15
16
17
18
19
Mean
55
72
48
42
59
48
61
47
43
53
9
110
108
71
95
76
81
92
88
75
95
18
35
58
41
145
73
108
98
68
75
53
66
41
106
51
11
103
44
60
54
47
49
48
125
118
12
NS
NS
25
NS
20
85
40
110
21
90
30
112
22
23
81
48
42
119
SD
p (A vs. B)
Group C
6,496
5,387
8,105
7,480
4,536
3,408
3,990
4,484
3,888
5,612
4,136
3,225
5,062
1,586
2,116
6,909
3,577
5,177
4,018
5,100
6,996
4,838
5,231
1,311
4,205
1,862
1,458
1,764
2,363
3,286
1,829
2,395
996
11,114
5,439
6,635
5,782
7,463
10,282
6,667
7,626
2,211
NS
<0.05
NS
8,085
6,300
9,580
6,899
7,007
3,279
4,800
5,760
7,268
6,553
1,831
808
794
1,288
1,705
861
713
836
1,175
874
1,056
900
1,094
425
24
160
121
25
57
35
24
26
27
79
34
80
107
31
80
178
28
79
106
35
40
8
92
17
4,400
3,360
5,712
3,066
3,185
1,992
2,720
2,480
2,212
3,236
1,171
<0.05
<0.05
<0.05
<0.05
NS
NS
NS
3,685
2,940
3,868
3,833
3,822
1,287
2,080
3,280
5,056
3,317
1,106
<0.05
NS
NS
28
Mean
SD
p (A vs. C)
p (B vs. C)
73
91
83
8,612
6,195
8,899
8,794
6,241
4,269
4,703
5,320
5,063
6,486
5,192
4,125
6,158
1,739
NS
NS
87
96
71
91
54
85
73
80
85
84
77
87
80
78
82
5
76
98
89
75
62
66
78
69
78
79
94
74
90
92
83
13
99
100
92
99
99
68
68
100
73
97
69
98
5
<0.05
3
NS
54
94
53
81
60
44
98
82
45
83
61
98
57
96
43
80
30
58
50
86
13
10
<0.05
<0.05
NS
NS
In the present study, nine of 28 patients showed negative values of the volume axis intercept. In these patients, end-systolic potential
energy (PE) was also calculated as the area of the triangle. ESV1, calculated end-systolic volume index; SVI, calculated stroke volume
index; ESP, calculated end-systolic pressure; SW, left ventricular stroke work; PVA, left ventricular pressure-volume area (SW+PE);
SW,,,.,, maximum attainable stroke work for a given end-diastolic volume; SD, standard deviation of the mean; NS, not statistically
significant.
ventricular compliance was reduced with elevated
end-diastolic pressure. All patients in the present
study were free from any sign of pulmonary congestion at rest. The average of mean capillary
wedge pressure was 15 mm Hg even in group C
with marked left ventricular dysfunction. Therefore, the errors in the SW estimated could not be
extremely large.
Nine of 28 subjects showed negative values of Vo.
The physiological meaning of the Vo of the endsystolic P-V relation is uncertain. There are several
possible explanations. First, the end-systolic P-V
relation is not linear in the regions of the VQ.25-26 A
second possibility is that the Vo cannot be accurately approximated from the limited end-systolic
P-V data. A third and related possibility is that Vo
Asanoi et al
Ventriculoarterial Coupling in Humans
491
TABLE 3. Individual Data of Myocardial Oxygen Consumption and Mechanical Efficiency
CSF
(ml/min)
A-CSDOj
(ml/100 ml)
129
13.7
10.4
13.1
12.7
12.5
16.2
11.3
21.4
16.1
16.3
24
1.4
MVOj
(ml/min) x (J/min)
SWxHR
(J/min)
SWxHR/MVOj
(%)
Group A
1
118
2
109
3
163
4
127
Mean
SD
323
84
26
227
66
29
428
141
33
323
100
31
325
98
30
4.1
82
32
3
Group B
13
332
9.0
108
18
125
279
75
15
132
29.9
13.9
13.1
19.0
601
14
262
73
27
28
381
Mean
196
11.1
9.9
10.0
85
24
SD
118
1.1
9.5
191
20
6
20
157
18.4
14.7
18.6
14.2
70
19
293
373
64
71
22
19
284
40
14
24
100
150
120
65
7.4
150
44
29
Mean
118
11.7
14.7
12.4
11.8
11.3
12.4
370
21
14.7
294
58
21
38
1.4
4.5
91
15
6
Group C
Downloaded from http://circres.ahajournals.org/ by guest on June 15, 2017
22
23
SD
CSF, coronary sinus flow; A—CSD02, arterial-coronary sinus oxygen difference; MV02. myocardial oxygen consumption; SW, left
ventricular stroke work; HR, heart rate, SD, standard deviation of the mean.
might be affected by changes in afterload used in the
present study.2*
A number of animal studies6-8-13-27-28 have recently
been performed to answer the question: what is the
optimal coupling between the ventricle and arterial
load actually operative in vivo under physiological
and pathological circumstances? Almost all of these
experimental studies produced a similar answer that
physiological states of the ventricle and arterial
input impedance are matched to produce a maximal
SW or power. Wilcken et al27 studied the effects of
sudden changes in arterial input impedance in anesthetized and conscious dogs and found that either
increase or decrease in the impedance reduced the
stroke power and work. Sunagawa et al8 have
theoretically predicted and experimentally validated in isolated physiologically loaded canine heart
that maximization of left ventricular SW occurs
when ventricular contractility measured by Ea and
the simulated arterial input impedance expressed by
E, are matched to equal each other (Figure 3).
Recently, van den Horn et al demonstrated in the
open-chested anesthetized cat that, at given arterial
pressure and cardiac output (therefore, ventricular
power) required by the body, the ventricular properties (contractility) are adjusted to produce the
required power at the peak of its power-flow relation curve. More recently, Myhre et al also showed
that in anesthetized and open-chested dog's heart,
the normal ventricle yielded maximum SW against
normal arterial afterload, but during acute depres-
sion of left ventricular performance, the working
point of the ventricle shifted from the top of the
dome-shaped SW-stroke volume relation curve onto
the left limb. These conclusions on "physiological"
matching disagree with the present findings in normal subjects. Rather, we found the ventriculoarterial coupling to be set for a maximal SW in group
B patients with mildly depressed heart function.
This discrepancy may be explained by remembering
that, in these studies (except part of Wilcken's
study27), the animals were anesthetized and openchested and that, in many of them, the heart was
excised. Under the circumstances, the ventricular
contractility could be significantly less than physiological. At the same time, the physiological values
of arterial input impedance taken from the literature
might be from animals anesthetized and with reflexly
augmented vascular tone.
There are other studies,1-10-11 experimental and
theoretical, which reached conclusions quite consistent with the present findings. First, Elzinga and
Westerhof10 showed in isolated cat heart preparation that the maxima of mean external power,
oxygen consumption, and mechanical efficiency of
ventricular contraction were achieved at different
operating points on their ventricular pump function
curve. From the P-V data obtained in the isolated
cat ventricle, Piene and Sund1 calculated the pump
efficiency as we did in the present study and found
that right ventricular pump efficiency became maximum when pulmonary impedance values loaded on
492
Circulation Research Vol 65, No 2, August 1989
Downloaded from http://circres.ahajournals.org/ by guest on June 15, 2017
the ventricle were in the physiological range. Finally,
Burkhoff and Sagawa11 theoretically analyzed the
arterial load that would produce maximal SW and
maximal mechanical efficiency of ventricular contraction. That is, they showed that the coupling
condition for maximal SW is E , = E a whereas that
for maximal efficiency is E,=E ra /2. They further
speculated that the earlier concept on the physiological coupling between the ventricular pump and
arterial afterload might have been misguided by the
nonphysiologically low E a value and high total
peripheral resistance value determined in the surgically compromised state.
These findings were surprisingly similar to the
present findings. In the present study, the normal
coupling condition in group A was nearly 'Em=~Etx/2,
as predicted by Burkhoff and Sagawa11 to achieve
maximal mechanical efficiency. In moderately
depressed heart, the ventricle and arterial system
were matched to maximize SW (E,=E M ). These
hearts generate a greater potential energy with
resultant reduction in work efficiency of the left
ventricle as compared with normal hearts. Work
efficiency (SW/PVA) is a monotonically decreasing
function of E./E*,: SW/PVA=l/[l+(Ea/Ee,)/2].11 On
the other hand, the relation of MVO2 to PVA is
crucially influenced by the inotropic state (Era).
Suga and coworkers- have demonstrated the
linear relation between MV02 and PVA. This relation has a nonzero positive intercept for PVA=0
(MVo2=axPVA+b, b>0). They also showed that
when the inotropic state is enhanced, the MV02PVA relation shifts upward (increase in b) and when
the inotropic state is depressed, the relation shifts
downward (decrease in b ) . - Consequently, with
the depression of contractile state PVA/MV02
becomes increased for a given PVA. In moderately
depressed heart, mechanical efficiency did not
decrease to the extent as SW/PVA did, presumably
due to an increase in PVA/MV02. In contrast,
severely depressed hearts, as shown in group C
patients, could no longer maintain SW properly and
result in the mismatch in terms of mechanical
efficiency probably due to the substantial fall in
work efficiency.
The nature of the error detector for ventriculoarterial coupling is unclear. It cannot be explained by
the receptors that give out an output proportional to
mechanical efficiency or stroke power as in the
regulation of mean arterial pressure and flow. Therefore, the efficiency or power matching might be
regarded as a coincidence that resulted from cardiovascular adjustment to achieve physiological arterial pressure and flow.
We tried to extend experimental observations
on optimal ventriculoarterial coupling to humans.
The present study delineated the distinctive aspect
between normal and variably depressed left ventricle in terms of the matching concept. The results
of our study suggest that this conceptual framework for quantifying ventriculoarterial interaction
has great potential usefulness to help us gain
insight into the relevance of adaptational changes
in congestive heart failure.
Acknowledgments
We are deeply indebted to Prof. Kiichi Sagawa,
The Johns Hopkins University, for his continuous
encouragement and valuable suggestions. We
acknowledge the secretarial assistance of Masami
Kosugi and Rumiko Sakakibara.
References
1. Piene H, Sund T: Does normal pulmonary impedance constitute the optimal load for the right ventricle? AmJPhysiol
1982;242:H154-H160
2. Yamakoshi K: Interaction between heart as a pump and
artery as a load. Jpn Ore J 1985;49:195-205
3. Elzinga G, Westerhof N: Pressure andflowgenerated by the
left ventricle against different impedances. Ore Res 1973;
32:178-186
4. Milnor WR: Arterial impedance as ventricular afterload.
Ore Res 1975;36:565-570
5. Maughan WL, Sunagawa K, Burkhoff D, Sagawa K: Effect
of arterial impedance changes on the end-systolic pressurevolume relation. On: Res 1984;54:595-602
6. Piene H, Sund T: Flow and power output of right ventricle
facing load with variable input impedance. Am J Physiol
1979;237:H125-H130
7. Van den Horn GJ, Westerhof N, Elzinga G: Optimal power
generation by the left ventricle: A study in the anesthetized
open thorax cat. Ore Res 1985;56:252-261
8. Sunagawa K, Maughan WL, Sagawa K: Optimal arterial
resistance for the maximal stroke work studied in isolated
canine left ventricle. Ore Res 1985;56:586-595
9. Suga H, Igarashi Y, Yamada O, Goto Y: Mechanical efficiency of the left ventricle as a function of preload, afterload,
and contractility. Heart Vessels 1985;l:3-8
10. Elzinga G, Westerhof N: Pump function of the feline left
heart: Changes with heart rate and its bearing on the energy
balance. Cardipvasc Res 1980;14:81-92
11. Burkhoff D, Sagawa K: Ventricular efficiency predicted by
an analytical model. Am J Physiol 1986;250:R1021-R1027
12. Teichbolz LE, Kreulen T, Herman MV, Gorlin R: Problems
in echocardiographic volume determinations: Echocardicgraphic-angiographJc correlations in the presence and absence
of asynergy. Am J Cardiol 1976;37:7-11
13. Sunagawa K, Maughan WL, Burkhoff D, Sagawa K: Left
ventricular interaction with arterial load studied in isolated
canine heart. AmJPhysiol 1983;245:H773-H780
14. Sunagawa K, Sagawa K, Maughan WL: Ventricular interaction with the loading system. Ann Biomcd Eng 1984;
12:163-189
15. Suga H, Sagawa K, Shoukas AA: Load independence of the
instantaneous pressure-volume ratio of the canine left ventricle and effects of cpincphrine and heart rate on the ratio.
Ore Res 1973;32:314-322
16. Suga H, Sagawa K: Instantaneous pressure-volume relationships and their ratio in the excised, supported canine left
ventricle. Ore Res 1974;35:117-126
17. Sagawa K, Suga H, Shoukas AA, Bakalar KM: End-systolic
pressure/volume ratio: A new index of ventricular contractility. Am J Cardiol 1977;40:748-753
18. Suga H: Total mechanical energy of a ventricle model and
cardiac oxygen consumption. Am J Physiol 1979;
236:H498-H505
19. Suga H, Hayashi T, Shirahata M, Suehiro S, Hisano R:
Regression of cardiac oxygen consumption on ventricular
pressure-volume area in dog. Am J Physiol 1981;
240:H320-H325
20. Suga H, Hisano R, Goto Y, Yamada O, Igarashi Y: Effect of
positive inotropic agents on the relation between oxygen
Asanoi et al
21.
22.
23.
24.
consumption and systolic pressure volume area in canine left
ventricle. Ore Res 1983;53:306-318
Suga H, Goto Y, Yasumura Y, Nozawa T, Futaki S, Tanaka
N, Uenishi M: O2 consumption of dog heart under decreased
coronary perfusion and propranolol. Am J Physiol 1988;
254:H292-H303
Ganz W, Tamura K, Marcus HS, Donoso R, Yoshida S,
Swan HJC: Measurement of coronary sinus blood flow by
continuous thermodilution in man. Circulation 1971;
44:181-195
Suga H, Sagawa K, Kostiuk DP: Control of ventricular
contractility assessed by pressure-volume ratio, E^,. Cardiovasc Res 1976;10:582-592
Vatner SF, Higgins CB, Franklin D, Braunwald E: Extent of
carotid sinus regulation of the myocardial contractile state in
conscious dogs. / Clin Invest 1972^1:995-1008
Ventricoloarterial Coupling in Humans
493
25. Sunagawa K, Maughan WL, Friesinger G, Guzman PA,
Cheng M, Sagawa K: Effects of coronary arterial pressure
on left ventricular end-systolic pressure-volume relations of
isolated canine heart. Ore Res 1982;50:727-734
26. Maughan WL, Sunagawa K, Burkhoff D, Sagawa K: Effect
of arterial impedance changes on the end-systolic pressurevolume relation. Ore Res 1984;54:595-602
27. Wilcken DEL, Charlier AA, Hoffman JIE, Guz A: Effects of
alterations in aortic impedance on the performance of the
ventricles. Ore Res 1964;14:283-293
28. Myhre ESP, Johansen A, Bjomstad J, Piene H: The effect of
contractility and preload on matching between the canine
left ventricle and afterload. Circulation 1986;73:161-171
KEYWORDS • ventricular end-systolic pressure-volume relation
• arterial elastance • ventricular stroke work • ventricular
mechanical efficiency • chronic cardiac failure
Downloaded from http://circres.ahajournals.org/ by guest on June 15, 2017
Ventriculoarterial coupling in normal and failing heart in humans.
H Asanoi, S Sasayama and T Kameyama
Downloaded from http://circres.ahajournals.org/ by guest on June 15, 2017
Circ Res. 1989;65:483-493
doi: 10.1161/01.RES.65.2.483
Circulation Research is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 1989 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/65/2/483
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/