Effects of β-Adrenergic Blocking Therapy on Left Ventricular Diastolic Relaxation Properties in Patients With Dilated Cardiomyopathy
Background—The hemodynamic mechanism for the improvement in left ventricle (LV) end-diastolic pressure in cardiomyopathy patients treated with β-adrenergic blocking agents is controversial. We hypothesized that the salutary effect of this kind of therapy on LV end-diastolic pressure would be indicative of an improvement in late, passive diastolic relaxation properties.
Methods and Results—We studied 14 cardiomyopathy patients in normal sinus rhythm with no arteriographic evidence of coronary artery disease and an LV ejection fraction of ≤40% by radionuclide angiography both before and after 6 months of metoprolol therapy with simultaneous micromanometry and biplane cineventriculography. Four comparable patients who were not treated with metoprolol were studied in a similar fashion and served as control subjects. In those receiving metoprolol, LV end-diastolic pressure decreased (P=0.001). The isovolumic relaxation index, τln, shortened (P=0.03). In a similar fashion, the LV chamber stiffness constant, κ, decreased (P=0.02), LV volume elastance improved (P=0.04), and the myocardial stiffness constant, κe, decreased (P=0.02). A multiple regression analysis revealed that the decrease in LV end-diastolic pressure was indicative of significant improvements in τln and κe with the relationship: LV end-diastolic pressure=−4.73+0.27 τln+0.54 κe (r=0.81, P<0.0001). These LV diastolic relaxation properties did not change or worsened in the control cardiomyopathy patients.
Conclusions—We conclude that the decrease in LV end-diastolic pressure in cardiomyopathy patients treated with metoprolol is an indicator of improvement in LV diastolic properties resulting from more complete myocardial relaxation.
In 1975, Waagstein and associates1 reported their initial observations with metoprolol therapy in cardiomyopathy patients. An apex echocardiogram from that report demonstrated a strikingly prominent A wave and a correspondingly elevated left ventricular (LV) end-diastolic pressure. These diastolic abnormalities resolved after the judicious administration of metoprolol. Subsequent hemodynamic studies have reported a decrease in heart rate and LV end-diastolic pressure and an increase in LV ejection fraction and systolic arterial pressure with this kind of therapy.2 3 The frequent observation of a decrease in LV end-diastolic pressure in these studies suggests that this kind of therapy may have a beneficial effect on LV relaxation properties. However, in contrast to the frequent observation of improvement in LV systolic performance,2 3 4 the documentation of a consistent pattern of improvement in LV relaxation properties has been more elusive. Some data suggest that active, early isovolumic relaxation is improved with β-adrenergic blocking therapy, but the effect on late, passive relaxation properties remains unresolved.2 5 6 Accordingly, the purpose of this investigation was to test the hypothesis that the salutary effect of β-adrenergic blocking therapy on LV end-diastolic pressure in cardiomyopathy patients is indicative of an improvement in late, passive relaxation properties.
Eighteen patients (16 men and 2 women) with clinically stable NYHA class II to III symptoms and a mean age of 52±8 years were studied. Each patient enrolled in the study met the following criteria: (1) normal sinus rhythm, (2) no arteriographic evidence of coronary artery disease, and (3) a radionuclide LV ejection fraction of ≤40%. In addition, all patients were on a stable medical regimen of digoxin, diuretics, and an ACE inhibitor for ≥2 months before beginning the protocol. Informed consent was obtained from each participant on forms approved by the Human Studies Committees at the University of Michigan and Veterans Affairs Medical Centers, Ann Arbor, Mich.
Before each cardiac catheterization, all medications, including the study medication at the second procedure, were withheld for 24 hours. All 18 patients underwent a diagnostic right and left heart catheterization, including arteriography. If no arteriographic evidence of coronary artery disease was demonstrable, a micromanometer catheter was placed to measure both LV and aortic pressures simultaneously with a biplane cineventriculogram. Metoprolol was then initiated at 12.5 mg QHS in 14 of these 18 patients and was increased every 2 weeks as tolerated to a maximal dose of 100 mg BID. The remaining 4 cardiomyopathy patients, who did not receive metoprolol, served as control subjects. After ≈6 months, including at least 2 months of a stable dose of metoprolol, all 18 patients completed a repeat left heart catheterization.
The LV pressure waveforms were digitized at a variable sampling frequency and interpolated to correspond with the midpoint of each cineventriculographic frame pair by use of algorithms developed in our laboratory.7 LV systolic pressure was defined as the maximum LV pressure, and LV end-diastolic pressure was defined as the pressure at the Z point immediately after the A wave on the LV pressure waveform. Frame-by-frame LV volumes were calculated from biplane cineventriculographic image pairs by use of a modified Simpson’s rule algorithm previously validated against human heart casts in our laboratory.8 Then, LV pressure-volume plots were constructed for the entire cardiac cycle for both the initial and end-of-study data (Figure 1⇓).
The cineventriculographic LV end-diastolic and end-systolic volumes were defined as the maximum and minimum values on the cineventriculographic LV volume curve. Then, LV ejection fraction was calculated by dividing stroke volume by end-diastolic volume.
LV wall thickness was estimated throughout the cardiac cycle from the equation of Hugenholtz and colleagues,9 and LV mass was calculated by the method of Rackley and colleagues.10 Circumferential stresses were calculated by the method of Mirsky.11
The method in our laboratory of analyzing the micromanometer LV pressures also provides for a calculation of the first derivative of LV pressure (dP/dt) at a variable sampling frequency. Thus, (+)dP/dtmax was defined as the maximum value after a 40 mm Hg/s increase in dP/dt but before aortic valve opening. Because (+)dP/dtmax is preload dependent,12 we normalized (+)dP/dtmax to LV end-diastolic volume to incorporate this volume dependence and used the result as an approximate measure of LV contractility.
These approaches of acquiring and processing the LV pressure waveform and cineventriculographic images were also used for calculating several active, early isovolumic relaxation indices and late, passive relaxation properties. The (−)dP/dtmin was defined as the maximum negative value of dP/dt after aortic valve closure but before mitral valve opening. Because (−)dP/dtmin may be dependent on the corresponding LV pressure, it was normalized by this pressure, resulting in (−)dP/dtmin/P. Similarly, the time constant of LV pressure decay, Tln, was initially calculated by the method of Weiss and colleagues.13 Because it has been suggested that this relationship may not be monoexponential, we also calculated the time constraint, Td, using the method proposed by Raff and Glantz.14 The correlation coefficients for all time constant calculations exceeded –0.90.
Similarly, we calculated several indices of late, passive relaxation using the corresponding micromanometer LV pressures, cineventriculographic LV volumes, and calculated LV stress and strain data from minimal LV pressure to the peak of the A wave.15 16 To calculate late, passive chamber stiffness, the nonlinear LV pressure and volume data were fitted to the equation dP/dV=κP+b, where the slope, κ, the late, passive chamber stiffness constant, is a linear function of dP/dV versus P.15 Because comparisons of late, passive chamber stiffness between hearts of different sizes may be problematic, we also calculated LV volume elastance, κ1, using the equation (dP/dV)/V=κ1P+b, which allows the relationship between pressure and volume during diastole to be compared at similar pressures after normalization for differences in heart size.16
Finally, the late, passive myocardial stiffness constant was calculated over the same range of LV stresses and wall thicknesses by the equation dς/dε=κeς+c, where ς represents circumferential stress and ε represents the corresponding strain values.17 Strain, ε, was obtained from the construct of Nakano and colleagues.18 Thus, κe, the late, passive myocardial stiffness constant, represents the linear relationship between dς/dε and ς.
By calculating both the late, passive chamber stiffness and myocardial stiffness constants, we hypothesized that we would be able to differentiate between the effects of ventricular interaction and pericardial restraint and the contribution of alterations in myocardial stiffness to the chamber stiffness constant and thus the relative impact of these factors on LV end-diastolic pressure.
Paired t tests were used to compare the continuous data obtained from the cardiomyopathy patients at baseline and after 6 months of metoprolol therapy. The data from the 4 cardiomyopathy patients who served as control subjects were compared in a similar fashion. Although these data are shown to contrast with those from the cardiomyopathy patients treated with metoprolol, no statistical analyses were performed between these data sets because of the small number of control cardiomyopathy patients. Pearson correlation matrices were used to establish the univariate correlates with LV end-diastolic pressure and ejection fraction. In addition, multiple regression analyses were used to define the independent predictors of LV end-diastolic pressure and ejection fraction from among the univariate predictors. A significant difference or relationship was established when the probability of rejecting the null hypothesis was ≥0.05.
The average dose of metoprolol was 145±70 mg/d in 2 divided doses.
The LV end-diastolic pressure decreased (P=0.001) and the LV systolic pressure increased (P=0.03) with metoprolol (Figure 2⇓). These LV pressures were unchanged in the control cardiomyopathy patients.
LV Volumes, Ejection Fraction, and Mass
The LV end-diastolic volume decreased (P=0.02), as did the end-systolic volume (P=0.01). Consequently, LV ejection fraction increased (P=0.009). The LV mass was unaffected by metoprolol. These LV size and performance measures did not change in the control cardiomyopathy patients (Table⇑).
When the LV (+)dP/dtmax was corrected for the effects of metoprolol on LV end-diastolic volume, this contractile index increased (P=0.007). There was no change in contractility in the control cardiomyopathy patients (Table⇑).
LV Early, Active Isovolumic and Late, Passive Diastolic Relaxation Properties
The LV (−)dP/dtmin corrected for the corresponding LV pressure improved, but this was not significant (Table⇑). In contrast, τln improved (P=0.03), whereas τd improved only modestly. These data are illustrated in Figure 3⇓.
As shown in Figure 4⇓, the late, passive LV chamber stiffness constant decreased (P=0.02), whereas LV volume elastance improved (P=0.04). Finally, the myocardial stiffness constant decreased (P=0.02).
These early, active and late, passive diastolic relaxation indices either did not change or worsened, as indicated by an increase in the chamber stiffness constant (P=0.04), in the control cardiomyopathy patients.
When a Pearson correlation matrix was developed with LV end-diastolic pressure as the independent variable, several interrelationships were identified. The LV end-diastolic pressures correlated with the values for τln (r=0.58, P=0.001), volume elastance (r=0.43, P=0.02), and κe (r=0.55, P=0.003). The τln values also correlated with those for τd (r=0.58, P=0.006) and (−)dP/dtmin/P (r=0.52, P=0.005). Similarly, the myocardial stiffness constants correlated with the chamber stiffness constants (r=0.62, P<0.0001) and volume elastance values (r=0.65, P<0.0001). Consequently, when a multiple regression analysis was run with LV end-diastolic pressure as the independent variable, 1 early, active and 1 late, passive diastolic relaxation index were identified as related to LV end-diastolic pressure with the relationship: LV end-diastolic pressure=−4.73+0.27 τln+0.54 κe (r=0.81, P<0.0001).
When a Pearson correlation matrix was developed with LV ejection fraction used as the independent variable and LV contractility and early, active and late, passive diastolic relaxation indices as dependent variables, several interrelationships were also identified. The LV ejection fractions correlated with the (+)dP/dtmax/EDV (r=0.81, P<0.0001), τln (r=−0.53, P=0.05), and volume elastance values (r=−0.72, P=0.004). However, when a multiple regression analysis was performed with LV ejection fraction as the independent variable, only the contractile index was incorporated into the relationship: LV ejection fraction=17.69+4.25 (+)dP/dt/EDV (r=0.81, P<0.0001).
The data from the present investigation support the hypothesis that the decrease in LV end-diastolic pressure observed with metoprolol in cardiomyopathy patients is an indicator of significant improvements in early, active isovolumic relaxation and late, passive myocardial relaxation. The specific variables, which in a multivariate analysis demonstrated a relationship with LV end-diastolic pressure, were τln and κe. This was because other measures of early, active isovolumic relaxation were strongly related to τln, and both late, passive chamber stiffness and volume elastance were strongly dependent on κe. This further confirms that the improvements in chamber stiffness and volume elastance were determined by improvements in myocardial stiffness. Finally, increases in LV ejection fraction were strongly related to improvements in contractility. Thus, there appeared to be strong relationships between the effects of metoprolol on LV end-diastolic pressure and early, active isovolumic relaxation and late, passive myocardial relaxation, which were in addition to that between LV ejection fraction and contractility.
Previous investigations have demonstrated that β-adrenergic blocking therapy has a consistent beneficial effect on LV ejection fraction in cardiomyopathy patients19 with a more variable and controversial effect on diastolic relaxation properties.2 5 6 20 Waagstein and colleagues1 initially reported improvement in LV end-diastolic pressure with metoprolol. Subsequently, the MDC Study Group evaluated LV diastolic filling using transmitral Doppler echocardiography.6 They reported that metoprolol resulted in a significant improvement in early LV diastolic deceleration times in cardiomyopathy patients. The maximum improvement in deceleration times occurred within 3 months of initiation of therapy, whereas LV systolic performance continued to improve for up to 12 months. These observations in a large group of cardiomyopathy patients are consistent with an improvement in diastolic relaxation properties, but the exact mechanism for this improvement was not addressed.
Investigators2 3 4 have also reported a consistent decrease in LV end-diastolic pressure, whereas improvements in LV diastolic relaxation properties have been reported to be more variable.2 20 The consistent observation in previous studies in which diastolic relaxation properties have been investigated and in the present investigation is the improvement in early, active isovolumic relaxation with β-adrenergic blocking therapy. However, Eichhorn and colleagues did not report a significant improvement in late, passive relaxation properties on bucindolol2 or metoprolol.20 This is important because, in a retrospective analysis, they concluded21 that cardiomyopathy patients with a higher LV end-diastolic pressure, among other indicators, were more likely to experience a hemodynamic improvement with metoprolol. Our observation that cardiomyopathy patients who have the most markedly increased LV end-diastolic pressures and late, passive diastolic relaxation constants have the greatest improvement with metoprolol is consistent with this concept. Therefore, it is appropriate to conclude that the hemodynamic benefits of β-adrenergic blocking therapy accrue in both diastole and systole in these kinds of patients, although they may be temporally disconnected.
The hemodynamic mechanism behind the decrease in LV end-diastolic pressure is more clearly elucidated by the data in the present investigation. In cardiomyopathy patients with LV dilatation and significant systolic and diastolic dysfunction, one must consider the potential effects of ventricular interaction and pericardial restraint on LV end-diastolic pressure.22 23 We did not see a parallel downward displacement of the diastolic pressure-volume relationship indicative of removal of these restraining effects. In addition, if LV end-diastolic volumes had diminished disproportionately, thereby removing these potential confounding effects on LV end-diastolic pressure, we would not have anticipated such a strong relationship between the LV end-diastolic pressures and the late, passive myocardial stiffness constants. Further support for this hypothesis was provided in a study of the acute hemodynamic response to intravenous metoprolol.24 In that study, metoprolol produced both negative inotropic and lusitropic effects without affecting LV end-diastolic pressure. By calculating both the late, passive chamber stiffness and myocardial stiffness constants and defining their relationship with LV end-diastolic pressure, we were able to determine for the first time whether late, passive myocardial relaxation had a significant effect on chamber stiffness and, as a result, on LV end-diastolic pressure, independent of other confounding variables, such as the effects of altered venous return on ventricular interaction and pericardial restraint. The data in this investigation also did not support a shift along an existing diastolic pressure-volume relationship. Therefore, it is reasonable to conclude that late, passive myocardial relaxation was more complete with metoprolol.
In the present investigation, we demonstrated a consistent improvement in LV contractility along with early, active isovolumic and late, passive myocardial relaxation. Although they appeared to be interrelated in a univariate manner, it is hard to explain this interaction without suggesting a common pathophysiological mechanism that may be inherent to each of these processes. A potential unifying pathophysiological mechanism to explain these hemodynamic effects of metoprolol may be its effect on abnormal calcium homeostasis in the cardiomyopathic myocyte. It has been suggested25 that abnormalities in myocyte calcium handling may contribute to myocardial dysfunction in the failing human heart. These abnormalities25 26 27 28 in the cardiomyopathic myocyte include an increase in diastolic calcium levels, a reduction in calcium release from the sarcoplasmic reticulum on calcium stimulation, and delayed reuptake of calcium by the sarcoplasmic reticulum. One could hypothesize that the hemodynamic correlates of these 3 perturbations in calcium homeostasis are increased myocardial stiffness, decreased contractility, and delayed isovolumic relaxation, respectively. High diastolic calcium levels may lead to persistent tension development in diastole by way of continuous formation of active actomyosin complexes.27 Alterations in calcium release from the sarcoplasmic reticulum may lead to variable effects on tension development.28 29 Altered calcium reuptake into the sarcoplasmic reticulum may effectively delay early, active isovolumic relaxation. The observations of a reduction in sarcoplasmic reticulum calcium ATPase29 and altered genomic expression of sarcoplasmic reticulum proteins and receptors30 31 may be the fundamental underlying mechanism for these observations. This is supported by recent data demonstrating the important interplay of sarcoplasmic reticulum calcium ATPase and the Na+-Ca2+ exchanger in determining diastolic force in isolated muscle strips from patients with end-stage heart failure.32 The nearly universal increase in LV systolic pressure and decrease in end-diastolic pressure, coupled with an improvement in LV ejection fraction2 20 ; our observations and those of others2 6 of an improvement in early, active isovolumic relaxation; and the additional finding in this investigation of improvement in late, passive diastolic relaxation properties, particularly myocardial stiffness, all suggest that β-adrenergic blocking therapy in these patients may beneficially affect calcium homeostasis.
There are potential limitations to this investigation that should be considered. First, the small size of our control cardiomyopathy group, who were not treated with metoprolol, precluded statistical comparisons. Nevertheless, either no change or a worsening in all LV systolic and diastolic relaxation indices was noted, which contrasts sharply with the observed beneficial effects observed in the LV systolic and diastolic relaxation indices in our cardiomyopathy patients treated with metoprolol. Second, the contributions of coronary vascular engorgement and abnormalities in coronary blood flow to diastolic dysfunction were not evaluated. Although the effects of vascular engorgement may be small, the contribution of abnormal coronary blood flow to impaired calcium handling and thus impaired LV chamber and myocardial stiffness cannot be ignored.33 This may also have been beneficially affected by metoprolol and have contributed to rectification of the underlying pathophysiological process.
In conclusion, metoprolol improves both early, active isovolumic and late, passive diastolic relaxation properties. Thus, it is reasonable to assume that LV end-diastolic pressure represents a marker of the degree to which diastolic properties of the LV are altered and amenable to β-adrenergic blocking therapy. Our investigation provides support for the contention that β-adrenergic blocking therapy in cardiomyopathy patients is associated with significant improvements not only in LV systolic performance but also in both early, active and late, passive diastolic relaxation properties resulting from more complete myocardial relaxation.
This study was supported by grant NIH RO1-HL-36450 from the National Heart, Lung, and Blood Institute, Bethesda, Md; the Department of Veterans Affairs, Washington, DC; and the Thomas Sine Memorial Research Fund, Ann Arbor, Mich.
Reprint requests to Mark R. Starling, MD, Professor of Internal Medicine, Cardiology Section, 111A, 2215 Fuller Rd, Ann Arbor, MI 48105.
- Received December 1, 1998.
- Revision received May 20, 1999.
- Accepted June 2, 1999.
- Copyright © 1999 by American Heart Association
Waagstein F, Hjalmarson A, Varnauskas E, Wallentin I. Effect of chronic β-adrenergic receptor blockade in congestive cardiomyopathy. Br Heart J. 1975;37:1022–1036.
Eichhorn EJ, Bedotto JB, Malloy CR, Hatfield BA, Deitchman D, Brown M, Willard JE, Grayburn PA. Effect of β-adrenergic blockade on myocardial function and energetics in congestive heart failure: improvements in hemodynamic, contractile, and diastolic performance with bucindolol. Circulation. 1990;82:473–483.
Waagstein F, Caidahl K, Wallentin I, Bergh CH, Hjalmarson A. Long-term β-blockade in dilated cardiomyopathy: effects of short- and long-term metoprolol treatment followed by withdrawal and readministration of metoprolol. Circulation. 1989;80:551–563.
Bristow MR, Gilbert EM, Abraham WT, Adams KF, Fowler MB, Hersberger RE, Kubo SH, Narahara KA, Ingersoll H, Krueger S, Young S, Shusterman N. Carvedilol produces dose-related improvements in left ventricular function and survival in subjects with chronic heart failure. Circulation. 1996;94:2807–2816.
Andersson B, Caidahl K, di Lenarda A, Warren SE, Goss F, Waldenstrom A, Persson S, Wallentin I, Hjalmarson A, Waagstein F. Changes in early and late diastolic filling patterns induced by long-term adrenergic β-blockade in patients with idiopathic dilated cardiomyopathy. Circulation. 1996;94:673–682.
Starling MR, Montgomery DG, Mancini GBJ, Walsh RA. Load independence of the rate of isovolumic relaxation in man. Circulation. 1987;76:1274–1281.
Rackley CE, Dodge HT, Coble YD, Hay RE. A method for determining left ventricular mass in man. Circulation. 1964;19:666–671.
Mirsky I. Left ventricular stresses in the intact human heart. Biophys J. 1969;9:189–194.
Weiss JL, Frederiksen JW, Weisfeldt ML. Hemodynamic determinants of the time-course of fall in canine left ventricular pressure. J Clin Invest. 1976;58:751–760.
Raff GL, Glantz SA. Volume loading slows left ventricular isovolumic relaxation rate: evidence of load-dependent relaxation in the intact dog heart. Circ Res. 1981;48:813–824.
Mirsky I. Assessment of diastolic function: suggested and future considerations. Circulation. 1969;1984:836–841.
Gaasch WM. Passive elastic properties of the left ventricle. In: Gaasch WH, LeWinter MM, eds. Left Ventricular Diastolic Dysfunction and Heart Failure. Malvern, Pa: Lea & Febiger; 1994:143–149.
Peterson KL, Tsuji J, Johnson A, DiDonna J, LeWinter M. Diastolic left ventricular pressure-volume and stress-strain relations in patients with valvular aortic stenosis and left ventricular hypertrophy. Circulation. 1978;58:77–89.
Nakano K, Sugawara M, Ishihara K, Kanazawa S, Corin WJ, Denslow S, Biederman RWW, Carabello BA. Myocardial stiffness derived from end-systolic wall stress and logarithm of reciprocal of wall thickness: contractility index independent of ventricular size. Circulation. 1990;82:1352–1361.
Eichhorn EJ, Heesch CM, Barnett JH, Alvarez LG, Fass SM, Grayburn PA, Hatfield BA, Marcoux LG, Malloy CR. Effect of metoprolol on myocardial function and energetics in patients with non-ischemic dilated cardiomyopathy: a randomized, double-blind, placebo-controlled study. J Am Coll Cardiol. 1994;24:1310–1320.
Ross J Jr. Acute displacement of the diastolic pressure-volume curve of the left ventricle: role of the pericardium and the right ventricle. Circulation. 1979;1:32–37.
Hess OM, Bhargava V, Ross J Jr, Shabetai R. The role of the pericardium in interactions between the cardiac chambers. Am Heart J. 1983;1377–1383.
Gwathmey JK, Copelas L, MacKinnon R, Schoen FJ, Feldman MD, Grossman W, Morgan JP. Abnormal intracellular calcium handling in myocardium from patients with end-stage heart failure. Circ Res. 1987;61:70–76.
Beuckelmann DJ, Nabauer M, Erdmann E. Intracellular calcium handling in isolated ventricular myocytes from patients with terminal heart failure. Circulation. 1992;85:1046–1055.
Apstein CS, Morgan JP. Cellular mechanisms underlying left ventricular diastolic dysfunction. In: Gaasch WH, LeWinter MM, eds. Left Ventricular Diastolic Dysfunction and Heart Failure. Malvern, Pa: Lea & Febiger; 1994:3–24.
Pieske B, Kretschmann B, Meyer M, Holubarsch C, Weirich J, Posival H, Minami K, Just H, Hasenfuss G. Alterations in intracellular calcium handling associated with the inverse force-frequency relation in human dilated cardiomyopathy. Circulation. 1995;92:1169–1178.
Hasenfuss G, Reinecke H, Studer R, Meyer M, Pieske B, Holtz J, Holubarsch C, Posival H, Just H, Drexler H. Relation between myocardial function and expression of sarcoplasmic reticulum Ca2+-ATPase in failing and nonfailing human myocardium. Circ Res. 1994;75:434–442.
Mercadier JJ, Lompre AM, Duc P, Boheler KR, Fraysse JB, Wisnewsky C, Allen PD, Komajda M, Schwartz K. Altered sarcoplasmic reticulum Ca2+-ATPase gene expression in the human ventricle during end-stage heart failure. J Clin Invest. 1990;85:305–309.
Arai M, Alpert NR, MacLennan DH, Barton P, Periasamy M. Alterations in sarcoplasmic reticulum gene expression in human heart failure: a possible mechanism for alterations in systolic and diastolic properties of the failing myocardium. Circ Res. 1993;72:463–469.
Hasenfuss G, Schillinger W, Lehnart SE, Preuss M, Pieske B, Maier LS, Prestle J, Minami K, Just H. Relationship between Na+-Ca2+-exchanger protein levels and diastolic function of failing human myocardium. Circulation. 1999;99:641–648.
Shintani M, Glantz SA. The left ventricular diastolic pressure-volume relation, relaxation, and filling. In: Gaasch WH, LeWinter MM, eds. Left Ventricular Diastolic Dysfunction and Heart Failure. Malvern, Pa: Lea & Febiger; 1994:57–88.