(Circulation. 1999;99:1027-1033.)
© 1999 American Heart Association, Inc.
Clinical Investigation and Reports |
From the University of Michigan and Veterans Affairs Medical Centers, Ann Arbor.
| Abstract |
|---|
|
|
|---|
Methods and ResultsForty-five patients with chronic aortic regurgitation (mean age 50±14 years) were studied using a micromanometer LV catheter to obtain LV pressures and radionuclide ventriculography to obtain LV volumes during multiple loading conditions and right atrial pacing. These 45 patients were subgrouped according to their LV contractility (Ees) and ejection fraction values. Group I consisted of 24 patients with a normal Ees. Group IIa consisted of 10 patients with impaired Ees values (Ees <1.00 mm Hg/mL) but normal LV ejection fractions; Group IIb consisted of 11 patients with impaired contractility and reduced LV ejection fractions. The left ventricular-arterial coupling ratio, Ees/Ea, where Ea was calculated by dividing the LV end-systolic pressure by LV stroke volume, averaged 1.60±0.91 in Group I. It decreased to 0.91±0.27 in Group IIa (P<0.05 versus Group I), and it decreased further in Group IIb to 0.43±0.24 (P<0.001 versus Groups I and IIa). The LV ejection fractions were inversely related to the Ea values in both the normal and impaired contractility groups (r=-0.48, P<0.05 and r=-0.56, P<0.01, respectively), although the slopes of these relationships differed (P<0.05). The average LV work was maximal in Group IIa when the left ventricular-arterial coupling ratio was near 1.0 because of a significant decrease in total arterial elastance (P<0.01 versus Group I). In contrast, the decrease in the left ventricular-arterial coupling ratio in Group IIb was caused by an increase in total arterial elastance, effectively double loading the LV, contributing to a decrease in LV pump efficiency (P<0.01 versus Group IIa and P<0.001 versus Group I).
ConclusionsVascular adaptation may be heterogeneous in patients with chronic aortic regurgitation. In some, total arterial elastance decreases to maximize LV work and maintain LV performance, whereas in others, it increases, thereby double loading the LV, contributing to afterload excess and a deterioration in LV performance that is most prominent in those with impaired contractility.
Key Words: regurgitation ventricles contractility arteries elasticity
| Introduction |
|---|
|
|
|---|
Using this concept of total arterial elastance, several studies have analyzed vascular adaptation in animals and in a variety of pathophysiologic conditions in humans.10 11 12 13 14 15 16 However, no study has used this conceptual framework to assess whether vascular adaptation occurs in patients with chronic aortic regurgitation. The application of this conceptual framework may provide insight into the pathophysiologic processes that occur beyond LV chamber and myocardial adaptation in response to aortic regurgitation; this application may also shed light on potential mechanisms of deterioration in LV performance and response to therapeutic interventions. Accordingly, the purpose of this investigation was to use the concept of total arterial elastance to examine the arterial system in patients with aortic regurgitation to test the hypothesis that vascular adaptation occurs in this disease process to maintain LV performance.
| Methods |
|---|
|
|
|---|
Protocol
All medications were stopped 24 to 48 hours before cardiac
catheterization. After a diagnostic cardiac
catheterization documented normal coronary
anatomy, each patient entered the protocol. A bipolar pacing
catheter was placed in the right atrium to maintain a constant heart
rate throughout the protocol. Through the right femoral artery, a
precalibrated, micromanometer LV catheter (Millar
Instruments) was positioned to measure LV pressures. In vivo red blood
cell labeling with 30 mCi of 99mTc was performed
for gated equilibrium radionuclide ventriculography. Then,
simultaneous micromanometer LV
pressures and radionuclide ventriculograms were acquired under control
conditions and during multiple, additional LV loading conditions
produced by steady-state infusions of methoxamine or
nitroprusside with heart rate held constant by right atrial pacing. The
methoxamine infusion was adjusted to achieve a variable
increase in LV systolic pressure of 20 to 50 mm Hg, and
the nitroprusside infusion was adjusted to achieve a variable
decrease in LV systolic pressure of 20 to 40 mm Hg. A
stable hemodynamic condition was considered present
when the micromanometer LV systolic
pressure varied
10 mm Hg.
Hemodynamics
The LV pressure waveforms were acquired at 100 mm/s
paper speed and included an electrocardiographic lead and
micromanometer LV pressures on 40 and 200
mm Hg scales. These hemodynamic measurements were
recorded for 10 to 20 cardiac cycles at the beginning, middle, and
end of each radionuclide acquisition. Then, these LV pressure waveforms
were averaged to obtain an average LV pressure waveform to match with
the corresponding radionuclide LV volume data for each loading
condition. The LV pressure waveforms were digitized using a Calcomp
9100 inductance digitizing surface (resolution 0.2 mm) beginning
at the peak of the R-wave of the simultaneously
recorded ECG. The program developed in this laboratory yields
instantaneous LV pressures at a variable sampling
frequency.17 Interpolation of the LV pressure data was
conducted to guarantee isochronicity of the LV pressure values with
the mid-point of each radionuclide frame throughout the cardiac
cycle.
Radionuclide Ventriculography
Gated equilibrium radionuclide ventriculograms were
obtained for 30 ms frames throughout the cardiac cycle for 250 cardiac
cycles. During the mid-point of each radionuclide acquisition, a 2 mL
blood sample was drawn and later counted for 2 minutes. The time delay
was recorded for decay correction. At the completion of the
protocol, a distance measurement was obtained for attenuation
correction. Absolute radionuclide LV volumes were calculated
frame-by-frame from background-subtracted hand-drawn region-of-interest
LV count data that were standardized for frame duration, number
of cardiac cycles acquired, decay-corrected blood sample counts,
and attenuation.18 19
Data Analysis
Radionuclide LV ejection fraction was calculated as LV stroke
volume divided by end-diastolic volume. In addition, we
also calculated LV regurgitant index. Right ventricular
(RV) stroke counts were obtained using a modification of the method
described by Madahhi and co-workers,20 which we have used
in other investigations.21 22 23 We used it in this
investigation to calculate LV regurgitant index (RI) as:
![]() |
The corresponding micromanometer LV pressures and radionuclide LV volumes for each loading condition were plotted to obtain multiple pressure-volume loops for each patient. The maximal pressure-volume ratio for each pressure-volume loop was subjected to linear regression analysis to obtain a slope (Ees) reflecting LV chamber elastance, a relatively load-independent index of LV contractility.24 25 26 To obtain total arterial elastance (Ea), LV end-systolic pressure (Pes) was divided by radionuclide LV stroke volume. The LV end-systolic pressure (Pes) was defined as the LV pressure at the maximum pressure-volume ratio during the baseline hemodynamic condition. The left ventricular-arterial coupling ratio was then defined as the ratio of Ees to Ea.
Two important assumptions were inherent in these calculations. First, consistent with the data from Sunagawa and colleagues7 8 concerning isolated left heart preparations and with the results from the theoretic studies of Burkhoff and Sugawa,9 we assumed that maximal LV work would occur when Ees equaled Ea. Second, we also assumed that Ea reflected total arterial load, as shown in the studies of Latham and colleagues,11 in normotensive and hypertensive nonhuman primates and by Kelly and co-workers,13 in normotensive and hypertensive patients.
To assess LV mechanical (pump) efficiency, LV stroke work was obtained by calibrated planimetry of the pressure-volume loop. The result was then multiplied by 0.0136 to convert from mm Hg/mL to grams per meter. The LV pressure-volume area was obtained by calibrated planimetry of the area enclosed by Ees, the diastolic curve, and the systolic portion of each pressure-volume loop.7 The ratio of external work to the pressure-volume area is reflective of LV pump efficiency, that is, the efficiency of converting the total energy available to the LV into external work. It is important to recognize that, although the LV pressure-volume area has a linear relationship with myocardial oxygen consumption,9 this relationship has a variable MVO2 axis offset as a result of basal metabolism and the energy costs of excitation-contraction coupling, which are not reflected in the pressure-volume area. Thus, LV pump efficiency cannot be assumed to be synonymous with myocardial efficiency.
Statistical Analysis
All data are expressed as the mean±SD. The patients with
aortic regurgitation were subgrouped according to
whether or not their LV contractility
(Ees) and ejection fraction values were normal or
abnormal. Data from a previous group of control subjects was used to
establish normal Ees values
1.00
mm Hg/mL, whereas LV ejection fraction was considered to be normal
when it was
0.45 by radionuclide ventriculography.27 28
Accordingly, all Group I patients had normal Ees
values, whereas all patients in Group II had impaired LV
contractility. Group II patients were further
subdivided into those with normal LV ejection fractions (IIa) and those
with abnormal LV ejection fractions (IIb). Data were compared between
these 3 subgroups using ANOVA. Then, t tests with a
Bonferroni correction were used to identify specific differences
between subgroups. Continuous variables were compared using linear
regression analysis. P
0.05 was needed to identify
significance.
| Results |
|---|
|
|
|---|
|
Left Ventricular Chamber Performance
The LV contractility index,
Ees, averaged 2.15±1.49 mm Hg/mL in
Group I, which, by subgroup definition, exceeded the average value in
Group II (0.62±0.23 mm Hg/mL in IIa and 0.47±0.27
mm Hg/mL in IIb).
Left Ventricular-Arterial Coupling
Relations
The left ventricular-arterial coupling
relationship varied in these patients with aortic
regurgitation. The LV end-systolic pressures
(Pes) averaged 146±28 mm Hg in Group I and
did not differ significantly from those in Groups IIa and IIb (134±32
and 151±39 mm Hg, respectively). In contrast, the average LV
stroke volume of 118±47 mL in Group I increased in Group IIa to
204±64 mL (P<0.001 versus Group I); and, then, it declined
in Group IIb to 167±101 mL (P<0.01 versus Group I and
P=0.05 versus Group IIa). Consequently, the total
arterial elastance, Ea, which was
determined by dividing Pes by the LV stroke
volume, also differed between the 3 subgroups (Figure 1
). In Group I, Ea
averaged 1.50±0.79 mm Hg/mL, it decreased to 0.70 32±0.23
mm Hg/mL in Group IIa (P<0.01 versus Group I), but it
increased in Group IIb to 1.36±0.98 mm Hg/mL (P<0.05
versus Group IIa). Thus, commensurate with the reduction in
Ees, there was a reduction in
Ea in Group IIa. Despite a further modest
reduction in Ees in Group IIb, there was an
increase in total arterial elastance in Group IIb. Thus,
the ratio of LV chamber elastance to total arterial
elastance, Ees/Ea, averaged
1.60±0.91 in Group I. It approached 1.00 in Group IIa averaging
0.91±0.27 (P<0.05 versus Group I), and it decreased
further in Group IIb to 0.43±0.24 (P<0.001 versus Groups I
and IIa).
|
The effects of these differences in left
ventricular-arterial coupling were
consistent with the conceptual framework of this kind of
analysis.7 8 9 Left ventricular work in
Group I averaged 145±58 g/m, and it increased in Group IIa to 242±103
g/m (P<0.001 versus Group I). With the continued impairment
of LV contractility and increase in total
arterial elastance in Group IIb, there was a decrease in LV
work to 173±105 g/m. Consequently, the average LV pump efficiency
values, represented by the ratio of LV work to
pressure-volume area, demonstrated a progressive decline from
0.61±0.12 in Group I to 0.50±0.11 in Group IIa
(P<0.05 versus Group I) and further to 0.45±0.23 in Group
IIb (P<0.01 versus Group IIa and P<0.001 versus
Group I). These data are also illustrated in Figure 1
. The ratio
of LV work to pressure-volume area, a measure of LV pump efficiency,
and the left ventricular-arterial coupling
ratio, Ees/Ea, had a
curvilinear relationship over a wide range of coupling ratios (Figure 2
). It is evident that, in patients with
aortic regurgitation, LV pump efficiency falls
dramatically as the left ventricular-arterial
coupling ratio declines below 1.0, which is represented
primarily by our patients in Group IIb. The relationship between LV
ejection fraction and left ventricular-arterial
coupling was similar, that is, as the
Ees/Ea ratio fell below
1.0, LV ejection fraction declined sharply.
|
To further examine this relationship, the LV ejection fractions were
compared with the total arterial elastance values; and no
significant relationship was observed (y=-0.036x + 0.56,
r=-0.22, P=0.14) between these variables in
the total population. However, when the aortic
regurgitation patients were divided into those with and
without normal contractility, significant relationships
between the LV ejection fraction and Ea values
were evident in each of these 2 groups (Figure 3
). In the normal contractile group
(Group I), there was an inverse relationship between the LV ejection
fraction and Ea values (y=-0.049x + 0.66,
r=-0.48, P<0.05). In the impaired contractile
group (Group II), a similar relationship was established (y=-0.10x +
0.54, r=-0.56, P<0.01), but the slope of this
relationship was more steep (P<0.05). Thus, there was
evidence that LV performance in each contractile group was
dependent on arterial load across the full continuum of
total arterial elastance values.
|
To further examine these interactions, normalized LV work values were
also plotted against the left
ventricular-arterial coupling ratios (Figure 4
). The solid line represents the
theoretic relationship between normalized LV work and the left
ventricular-arterial coupling
ratio.9 Accordingly, maximum LV work would theoretically
occur when the LV chamber elastance and total arterial
elastance values are coupled at an
Ees/Ea ratio of 1.0. The
normalized LV work values in control subjects over a wide range of LV
loading conditions fall on this theoretic relationship in the higher
left ventricular-arterial coupling ratios
>1.0.29 The aortic regurgitation patients
in Group I, who had normal LV contractility and
ejection fraction values, also fell in the higher range of left
ventricular-arterial coupling ratios
consistent with LV operating between maximal work and
efficiency similar to a normal heart. However, in Group IIa, there was
a decline in the left ventricular-arterial
coupling ratio to
1.0 to maximize LV work in line with the
theoretic framework. As the
Ees/Ea ratio deteriorated
in Group IIb, falling below a coupling ratio of 1.0, normalized LV work
also deteriorated.
|
| Discussion |
|---|
|
|
|---|
The left ventricular-arterial coupling concept proposed by Sunagawa and colleagues7 8 provides a convenient theoretical framework within which LV performance can be evaluated through the interaction of the LV with the arterial system. This framework predicts that the transfer of energy from the LV to the arterial system, that is, the performance of external work, is maximal when these 2 elastances are equal.7 9 Thus, by standardizing LV work to the maximal theoretic value, the effects of aortic regurgitation on LV work and pump efficiency can be examined over a range of contractile states and coupling ratios. The data in this investigation suggest that the left ventricular-arterial coupling ratio remains unaffected and energy transfer from the LV to the arterial remains high, but not maximal, with development of aortic regurgitation similar to that in a normal human heart. However, with progression to LV contractile dysfunction, the left ventricular-arterial coupling ratio may decline to maximize LV work and maintain LV performance. Part of this adaptation is a reduction in total arterial elastance, which offsets the decline in LV chamber performance. This is consistent with prior investigations examining the individual components of arterial load in animals and humans.30 31 In experimental aortic regurgitation,30 it has been reported that increased circumferential distensibility of the aorta results in decreased characteristic impedance. In humans with chronic aortic regurgitation,31 it has been reported that vascular compliance may either be increased or decreased, depending on the calculation methodology; vascular resistance is usually less than normal, although standard methods of calculating vascular resistance in aortic regurgitation may obscure this beneficial vascular response. Thus, vascular adaptation in aortic regurgitation may contribute to preservation of LV performance.
These data are consistent with previous observations in normal human subjects reported from this laboratory,29 but they contrast with data from patients with dilated congestive cardiomyopathy studied by Asanoi and colleagues12 and others.16 In hypertensive, cardiomyopathic, or mitral regurgitation patients, there is a relative increase in total arterial elastance as LV chamber performance deteriorates with these disease processes; this may contribute to worsening LV dysfunction.12 13 16 33 Nevertheless, although there may be a decrease in total arterial elastance in some patients with aortic regurgitation, in others, the arterial system may not be capable of this kind of adaptation. In these patients, double loading of the LV may cause deterioration in LV performance, manifested by a reduction in LV work and pump efficiency and, thereby, contributing to the state of afterload excess. This contrasts with the effects of double loading of the LV in patients with aortic stenosis. Carroll and colleagues33 found that double loading of the LV with aortic stenosis and systemic hypertension did not affect LV performance in any substantive manner. Thus, in contrast to other disease processes, the arterial system may undergo a heterogeneous adaptation in aortic regurgitation.
This may also have significant implications for understanding the development of symptoms and resting LV dysfunction requiring surgical intervention5 6 and the benefits of vasodilator therapy.34 35 36 37 In the asymptomatic patient with aortic regurgitation, the development of symptoms or resting LV dysfunction occurs in approximately 4% of patients per year.5 6 Predictors of progression to surgery include large LV volumes, a relatively low LV ejection fraction, and an elevated end-systolic stress.6 Because these parameters were found predominantly in our Group II patients, these data suggest that maladaptation of the arterial system may represent one possible hemodynamic mechanism for the development of clinical indicators for surgery in these patients. Burkhoff and Sagawa9 have also demonstrated within this theoretical framework that in the depressed contractile state, a heart is more susceptible to an increase in total arterial elastance. Therefore, an increase in Ea in patients with aortic regurgitation and contractile dysfunction would predict a greater reduction in LV performance. In our patient population, there was a steeper, more inverse relationship between LV ejection fraction and Ea in the patients with impaired contractility compared with those with normal contractility manifesting their sensitivity to afterload excess in line with theoretic predictions.9 Reversal of this kind of vascular maladaptation may explain the observations of Bonow and colleagues,28 who described improvement in LV dysfunction after surgical intervention when surgery was performed in a timely fashion. In addition, the long-term benefit observed with vasodilator therapy36 in asymptomatic patients with moderate-to-severe aortic regurgitation may be related to optimization of total arterial elastance through a reduction in vascular resistance,38 which may, in a parallel fashion, prevent or delay maladaptation of the arterial system.
One potential limitation of this approach is that total arterial elastance incorporates vascular resistance, vascular compliance, and characteristic impedance into a single measure of total arterial load. One benefit of this approach, however, is the ability to couple total arterial elastance to LV chamber elastance, because both are measured in the pressure-volume plane.7 8 9 This concept provides a better appreciation of the interaction of the LV and arterial system in aortic regurgitation. Further, it has been recognized that changes in total arterial elastance reflect corresponding and proportionate changes in the 3 specific parameters that characterize the arterial system.11 13 Another potential benefit to this approach is that it obviates the question of how resistance and compliance should be calculated in aortic regurgitation.31 39 Nevertheless, the specific parameters that are altered, and to what extent, cannot be ascertained from this approach. Another potential limitation is that we could not totally exclude residual vasodilator drug effects on arterial elastance. Although we discontinued all medications 24 to 48 hours before the study, some residual effects may have persisted. Nevertheless, these residual effects should have been minimized and were present in only 11 of the 45 patients studied.
In conclusion, the data in this investigation suggest that adaptation of the arterial system is important in some patients with chronic aortic regurgitation to maintain LV work and performance by optimizing coupling of the arterial system to the LV, while maladaptation of the arterial system, which leads to double loading of the LV, may occur in others and contribute to afterload excess and the development of LV pump dysfunction requiring surgical intervention.
| Acknowledgments |
|---|
| Footnotes |
|---|
Received June 1, 1998; revision received November 10, 1998; accepted November 18, 1998.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
F. Tarasoutchi, M. Grinberg, G. S. Spina, R. O. Sampaio, L. u. F. Cardoso, E. G. Rossi, P. Pomerantzeff, F. Laurindo, P. L. da Luz, and J. A. F. Ramires Ten-year clinical laboratory follow-up after application of a symptom-based therapeutic strategy to patients with severe chronic aortic regurgitation of predominant rheumatic etiology J. Am. Coll. Cardiol., April 16, 2003; 41(8): 1316 - 1324. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. Segers, N. Stergiopulos, and N. Westerhof Relation of effective arterial elastance to arterial system properties Am J Physiol Heart Circ Physiol, March 1, 2002; 282(3): H1041 - H1046. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Circulation Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 1999 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |