Author + information
- Received December 14, 2000
- Revision received April 19, 2001
- Accepted April 27, 2001
- Published online August 1, 2001.
- Ilan S Wittstein, MDa,
- David A Kass, MDa,
- Peter H Pak, MD, FACCa,
- W.Lowell Maughan, MDa,
- Barry Fetics, MSEa and
- Joshua M Hare, MD, FACCa,* ()
- ↵*Reprint requests and correspondence:
Dr. Joshua M. Hare, The Johns Hopkins Hospital, 600 N. Wolfe Street, Carnegie 568, Baltimore, Maryland 21287
This study tested the hypothesis that angiotensin-converting enzyme (ACE) inhibitors attenuate beta-adrenergic contractility in patients with idiopathic dilated cardiomyopathy (DCM) through nitric oxide (NO) myocardial signaling.
The ACE inhibitors increase bradykinin, an agonist of NO synthase (NOS). Nitric oxide inhibits beta-adrenergic myocardial contractility in patients with heart failure.
The study patients were given the angiotensin-1 (AT-1) receptor antagonist losartan for one week. The hemodynamic responses to intravenous dobutamine were determined before and during intracoronary infusion of enalaprilat (0.2 mg/min) with and without the NOS inhibitor NG-monomethyl-L-arginine (L-NMMA, 5 mg/min).
In patients with DCM (n = 8), dobutamine increased the peak rate of rise of left ventricular pressure (+dP/dt) by 49 ± 8% (p < 0.001) and ventricular elastance (Ees) by 53 ± 16% (p < 0.03). Co-infusion with enalaprilat decreased +dP/dt to 26 ± 12% and Eesto −2 ± 17% above baseline (p < 0.05), and this anti-adrenergic effect was reversed by L-NMMA co-infusion (p < 0.05 vs. enalaprilat). In addition, intracoronary enalaprilat reduced left ventricular end-diastolic pressure (LVEDP), but not left ventricular end-diastolic volume, consistent with increased left ventricular distensibility. Infusion with L-NMMA before enalaprilat in patients with DCM (n = 5) prevented the reduction in +dP/dt, Eesand LVEDP. In patients with normal left ventricular function (n = 5), enalaprilat did not inhibit contractility or reduce LVEDP during dobutamine infusion.
Enalaprilat attenuates beta-adrenergic contractility and enhances left ventricular distensibility in patients with DCM, but not in subjects with normal left ventricular function. This response is NO modulated and occurs in the presence of angiotensin receptor blockade. These findings may have important clinical and pharmacologic implications for the use of ACE inhibitors, AT-1 receptor antagonists and their combination in the treatment of heart failure.
Angiotensin-converting enzyme (ACE) inhibitors, a principal therapy for patients with left ventricular dysfunction, have been shown to reduce mortality and improve symptoms in several large, prospective, randomized clinical trials (1). Angiotensin-converting enzyme, a dipeptidyl carboxypeptidase identical to kininase II, catalyzes the conversion of angiotensin I to angiotensin II, and also degrades bradykinin (2). Although the beneficial effects of ACE inhibitors in the treatment of heart failure have been attributed to numerous favorable neurohormonal and hemodynamic consequences of angiotensin II reduction, the pharmacologic contribution of bradykinin to the action of this drug class remains unclear.
One way in which bradykinin may contribute to the cardiac effects of ACE inhibition is through B2receptor-mediated activation of nitric oxide synthase (NOS). Nitric oxide (NO) attenuates the positive inotropic response to beta-adrenergic stimulation in humans (3)and in animal models (4,5), and this effect appears to be augmented in the setting of left ventricular dysfunction (6). The current study tested the hypothesis that ACE inhibitors attenuate beta-adrenergic contractility through myocardial NO signaling, likely due to inhibition of bradykinin degradation. Because NOS activity is increased in the failing ventricle, we further hypothesized that ACE inhibition would attenuate beta-adrenergic contractility to a greater extent in patients with heart failure than in control subjects with normal left ventricular function.
We studied 13 patients (10 men and 3 women; mean ± SD age 45 ± 3 years) with nonischemic dilated cardiomyopathy (DCM). All patients had ejection fractions <40% (by echocardiography or radionuclide ventriculography) and left ventricular end-diastolic dimensions >55 mm, and none had coronary stenoses >50%. Patients were in New York Heart Association functional class II (n = 4), III (n = 8) or IV (n = 1) heart failure and were receiving digitalis (n = 8), diuretics (n = 11), ACE inhibitors (n = 5), angiotensin receptor blockers (n = 4) and beta-blockers (n = 1) before study enrollment.
The control group consisted of five patients (2 men and 3 women; mean ± SD age 44 ± 4 years) referred for evaluation of atypical chest pain. None of these patients had coronary disease, and all had normal left ventricular function (ejection fraction >50% and left ventricular end-diastolic dimension <55 mm) and normal right heart filling pressures.
All study patients and control subjects were treated with the angiotensin-II receptor antagonist, type 1 (AT-1) losartan (50 mg/day) for one week before the study. Patients already taking ACE inhibitors were withdrawn from them and converted to the angiotensin receptor blocker. This was done so that the acute administration of ACE inhibitors during the study would occur in the presence of AT-1 receptor blockade, thereby principally eliciting signaling pathways not associated with this receptor. This dose of losartan results in near maximal inhibition of the pressor response to exogenous angiotensin I and angiotensin II in humans (7,8), and losartan has similar pharmacokinetic properties in healthy control subjects and in patients with heart failure (9). After diagnostic coronary angiography and ventriculography, a 7F, high-fidelity, micromanometer-tipped conductance catheter (Millar Instruments, model no. SSD-769, Houston, Texas) was placed in the left ventricular apex for simultaneous pressure and volume measurements. Femoral artery pressure was monitored through a 7F side-arm sheath (Cordis Laboratories, Miami, Florida). The electrocardiogram, femoral artery pressure, left ventricular pressure and volume and peak rate of rise of left ventricular pressure (+dP/dt) were digitally recorded. Each measurement represents the mean value of at least 30 consecutive sinus beats. Hemodynamic analysis was performed using custom-designed Matlab (Natick, Massachusetts) software.
Indexes of myocardial systolic and diastolic performance were derived from signal-averaged steady state pressure–volume data. To eliminate the requirement for graded inferior vena cava balloon occlusion, we used a single-beat estimation of the end-systolic pressure-volume relationship and slope (ventricular elastance [Ees]). This method derives a value for Eesbased on steady state pressure-volume data and has been validated in both animals and humans, including patients with cardiomyopathy (10). Preload was indexed by left ventricular end-diastolic volume (LVEDV) and pressure (LVEDP), and afterload was indexed by systemic vascular resistance and effective arterial elastance (Ea; ratio of left ventricular systolic pressure to stroke volume) (11).
Left ventricular diastolic relaxation rates were assessed by the time constant tau, derived from high-fidelity, micromanometer pressure recordings. Both the Weiss formula (tau ln) and the hybrid-logistic method (tau l), which is less dependent on LVEDP, were used to derive tau (12).
Drug infusion protocols
Digitalis, diuretics and nitrates were withheld from subjects for 12 h before the study. Heparinized saline was infused intracoronarily at a rate of 2 ml/min. Heart rate was maintained constant by atrial pacing at 20 to 30 beats/min above the baseline heart rate. Once baseline +dP/dt was recorded, dobutamine (Lilly, Indianapolis, Indiana) was infused through the right femoral vein for the remainder of the protocol and titrated to increase +dP/dt by >30% above baseline (7.9 ± 1.0 μg/kg body weight per min in patients with DCM and 2.6 ± 0.2 μg/kg per min in normal control subjects, p < 0.01). Dobutamine was considered to be at steady state when three consecutive measurements of +dP/dt were stable over a period of 10 min. Using a Harvard pump, enalaprilat and NG-monomethyl-L-arginine (L-NMMA) were directly infused into the left main coronary artery. For group I (with DCM, n = 8) and group III (with normal left ventricular function, n = 5), enalaprilat (0.2 mg/min) was infused intracoronarily for 15 min, followed immediately by co-infusion with L-NMMA (5 mg/min [20 μmol/min]) for 15 min. To determine whether L-NMMA could prevent the effects of enalaprilat, the order of intracoronary infusions was reversed in group II (with DCM, n = 5). Hemodynamic measurements were recorded every 5 min during drug infusion. Eesand +dP/dt were recorded from all patients in groups I and III. In group II, +dP/dt was recorded in all five patients, but Eescould only be measured in three patients during baseline conditions and in two patients during drug infusion.
All data are presented as the mean value ± SEM. Baseline characteristics between the three groups of subjects were compared using one-way analysis of variance or the Student ttest, with the Bonferroni correction. Changes within a single group with successive drug infusions were analyzed by paired two-way analysis of variance, using an identification term for each patient. Post hoc testing was performed using the Student-Newman-Keuls test.
Baseline hemodynamic data
Baseline hemodynamic measurements for patients with DCM (groups I and II, n = 13) and patients with normal left ventricular function (group III, n = 5) are shown in Table 1. Patients with DCM had a decreased ejection fraction of 21 ± 2% (range 12% to 36%), compared with group III, with an ejection fraction of 61 ± 7% (range 50% to 81%; p < 0.001). Patients with DCM also had elevated preload (LVEDP and LVEDV) and afterload (systemic vascular resistance), compared with control subjects. There were no differences in baseline hemodynamic variables between patients in group I and those in group II (Table 1).
Effect of intracoronary enalaprilat on beta-adrenergic contractility
Figure 1depicts representative steady state pressure-volume loops and Eesfrom each group of patients. The positive inotropic effect of dobutamine was evident as an increase in the slope of Eesin all three groups (Fig. 1 and 2). ⇓Dobutamine increased Eesby 53 ± 16% in group I (p < 0.03) and 56 ± 19% in group III (p < 0.05) (Fig. 2). Ventricular elastance, although measured in only two of the five patients in group II, increased with dobutamine in both patients (63 ± 1%). In addition, dobutamine increased +dP/dt by 48 ± 8% in group I (p < 0.001), 33 ± 10% in group II (p < 0.05) and 64 ± 19% in group III (p < 0.04) (Fig. 2). The percent increases in +dP/dt and Eesin response to dobutamine were not different between the three groups.
During dobutamine infusion, intracoronary enalaprilat decreased +dP/dt in all eight patients in group I, resulting in a reduction of ∼50% in the +dP/dt response from 49 ± 18% to 26 ± 12% relative to baseline (p < 0.05) (Fig. 2A). The attenuation of beta-adrenergic contractility indexed by Eeswas even greater, with a near total elimination of the augmented Eesresponse, from 53 ± 16% to −2 ± 12% relative to baseline (p < 0.05) (Fig. 2A). In group II, pretreatment with intracoronary L-NMMA in the presence of dobutamine caused a small augmentation in +dP/dt (36 ± 10% vs. 32 ± 10%, p < 0.05). Subsequent co-infusion with intracoronary enalaprilat in this group decreased +dP/dt to a level comparable to that of dobutamine alone (Table 2, Fig. 2B). Similarly, enalaprilat did not attenuate beta-adrenergic contractility in patients with normal left ventricular function (Table 2, Fig. 2C).
Effect of intracoronary l-NMMA on the antiadrenergic response to enalaprilat
In group I, L-NMMA co-infusion reversed enalaprilat’s attenuation of beta-adrenergic contractility, whether assessed by +dP/dt or Ees. Co-infusion of L-NMMA increased +dP/dt from 1,294 ± 230 to 1,484 ± 263 mm Hg/s (p < 0.05) and increased Eesfrom 0.86 ± 0.21 to 1.25 ± 0.28 mm Hg/ml (p < 0.05) (Table 2, Fig. 2A). In patients with normal left ventricular function (group III) in whom enalaprilat had no effect on beta-adrenergic contractility, L-NMMA co-infusion similarly had no effect on +dP/dt or Ees(Table 2, Fig. 2C).
Effect of intracoronary enalaprilat on afterload and systemic blood pressure
Table 2summarizes the effects of enalaprilat and L-NMMA on afterload and systemic blood pressure in the presence of dobutamine. In all groups, heart rate was maintained constant with atrial pacing. In group I (with DCM), intracoronary enalaprilat did not influence afterload (Eaand systemic vascular resistance). It did, however, reduce both left ventricular systolic pressure and mean arterial pressure in a manner reversible with L-NMMA co-infusion. When the order of intracoronary drugs was reversed in group II, L-NMMA increased left ventricular systolic pressure and mean arterial pressure and prevented their reduction in response to subsequent intracoronary enalaprilat. In patients with normal left ventricular function, intracoronary enalaprilat had no effect on indexes of load or systemic pressure.
Effect of intracoronary enalaprilat on diastole
Figure 3depicts representative pressure-volume data obtained during diastole. In patients with DCM (group I), intracoronary enalaprilat reduced LVEDP without affecting LVEDV (Table 2). This downward shift in the end-diastolic pressure–volume relationship was reversed (group I) and prevented (group II) by L-NMMA. A similar downward shift in the end-diastolic pressure-volume relationship in response to enalaprilat was not observed in control subjects (group III). Active diastolic relaxation, assessed using two formulations of tau, was decreased by dobutamine in all three groups (p < 0.05), but was not affected by subsequent infusion of enalaprilat or L-NMMA in any group (Table 2).
Effect of intracoronary enalaprilat on stroke work
To assess the integrated hemodynamic impact of enalaprilat on beta-adrenergic responsiveness, we examined left ventricular stroke work. Dobutamine increased left ventricular stroke work by 51 ± 15% (from 5,931 ± 1,136 to 8,448 ± 1,402 mm Hg · ml, p < 0.05) in patients with DCM (group I). Subsequent infusion with intracoronary enalaprilat (7,957 ± 1,655 mm Hg · ml) and L-NMMA (7,706 ± 1,573 mm Hg · ml) did not influence this increase in left ventricular stroke work. Similar responses were observed in groups II and III.
This study demonstrates that ACE inhibitors attenuate beta-adrenergic contractility and increase left ventricular distensibility in patients with idiopathic DCM. These effects are modulated by myocardial NO production, are observed in the presence of AT-1 receptor blockade and occur preferentially in patients with left ventricular systolic dysfunction. These previously unappreciated properties of ACE inhibitors may contribute to their therapeutic effects in the treatment of heart failure and should be considered when comparing them with other inhibitors of the renin-angiotensin-aldosterone pathway (13).
Effect of ACE inhibition on beta-adrenergic contractility
The rationale for performing this study in the presence of systemic dobutamine was based on observations from human (3), experimental animal (4)and in vitro studies (14)that NO inhibits beta-adrenergic myocardial contractility. The ability of L-NMMA to both reverse and prevent the negative inotropic response to enalaprilat strongly supports our hypothesis that the inhibitory effect on beta-adrenergic contractility is NO-mediated. These findings are consistent with previous observations that ACE inhibitors increase cardiac NO levels in both ischemic myocardium (15,16)and coronary microvessels from failing explanted human hearts (17).
The effects of ACE inhibition on myocardial contractility have been difficult to study because of the vasodilatory properties that alter loading conditions and increase cardiac output. To circumvent this response, we infused enalaprilat intracoronarily to achieve higher drug concentrations in the coronary circulation, thus maximizing direct cardiac effects and limiting systemic responses. Previous studies with intracoronary enalaprilat infusion in humans with DCM have reported either negative (18)or neutral (19)inotropic effects. The current study differs in that it examined the cardiac effects of ACE inhibition during beta-adrenergic stimulation and AT-1 receptor blockade.
In group I (with DCM), enalaprilat reduced left ventricular systolic pressure and mean arterial pressure. The conclusion that these hemodynamic changes were secondary to an attenuation of myocardial contractility is supported not only by the use of the load-independent index of contractility Ees, but also by the fact that enalaprilat did not alter preload (LVEDV) or afterload (Eaand systemic vascular resistance). The ability of L-NMMA to both reverse and prevent the effects of enalaprilat supports our hypothesis that ACE inhibitors attenuate beta-adrenergic contractility through myocardial NO signaling pathways.
Effect of ACE inhibition on diastole and chamber compliance
Importantly, enalaprilat reduced LVEDP without affecting LVEDV. This suggests that, in addition to influencing contractility, enalaprilat also enhanced left ventricular chamber compliance. In support of this, enalaprilat caused a downward shift in the end-diastolic pressure-volume relationship in patients with DCM (group I). A similar effect has been previously observed with intracoronary administration of sodium nitroprusside (20). The ability of L-NMMA to both reverse (group I) and prevent (group II) this downward shift in the end-diastolic pressure-volume relationship supports the hypothesis that enalaprilat’s effect on left ventricular compliance is NO-mediated (Fig. 3). Interestingly, neither enalaprilat nor L-NMMA influenced tau, suggesting that the ACE inhibitor and/or NO-mediated diastolic effects occur preferentially on late diastolic compliance rather than on early (active) relaxation in subjects with heart failure.
Myocardial NO signaling in heart failure
Nitric oxide–mediated attenuation of beta-adrenergic contractility may contribute to the cardioprotective effects of ACE inhibition. Beta-adrenergic agonists increase mortality (21), whereas receptor antagonists improve survival (22,23)in patients with left ventricular dysfunction. Thus, ACE inhibitors may exert post-receptor beta-blockade by increasing cardiac NO, which, in addition to attenuating contractility, may also decrease myocardial oxygen consumption (24,25).
Our hypothesis that ACE inhibition has a greater NO-mediated inhibitory effect in patients with heart failure than in control subjects was based on observations that myocardial NOS activity is elevated in this disease state (26,27). As a functional correlate, we previously reported that inhibition of NO synthesis potentiates the response to beta-adrenergic stimulation in patients with left ventricular dysfunction, but not in control subjects (6). The present findings raise the issue of the specific isoform being upregulated in failing myocardium. An enhanced NO response due to ACE inhibition suggests an augmented agonist-stimulated pathway (e.g., bradykinin), rather than agonist-independent NO signaling (e.g., due to NOS-2 activity). In this regard, elevations of both NOS-2 and NOS-3 have been reported in failing myocardium (28,29), and the present results suggest that increased NOS-3 may have functional significance. Alternatively, as we have recently shown in an animal model, cardiac NO pathway activity may also be regulated independently of NOS isoform upregulation (5).
There are several limitations that warrant mention. First, our observations, made in the acute setting, may not be applicable to the long-term ACE inhibitor administration that likely results in other compensatory mechanisms. For example, Maisel et al. (30)demonstrated the reversal of beta-adrenergic receptor downregulation after administration of captopril for two weeks. Second, it remains possible that the systemic effects of intracoronary infusions may have influenced loading conditions and contributed to our findings. Our study design attempted to circumvent this problem by employing an intracoronary infusion technique to minimize systemic vascular effects (31)and by using a load-independent measurement of contractility (Ees). Finally, to prevent changes in coronary blood flow from affecting dobutamine concentrations artifactually, dobutamine was administered systemically to avoid any potential decreases in its concentration due to enalaprilat or any increases due to L-NMMA.
Our study demonstrates that ACE inhibitors attenuate beta-adrenergic contractility and increase left ventricular distensibility in patients with idiopathic DCM. These responses result from NO pathway activity and occur during AT-1 receptor blockade. These previously unappreciated pharmacologic properties of ACE inhibitors may contribute to the clinical profile of this class of drugs in the treatment of patients with heart failure and left ventricular dysfunction.
The authors acknowledge the assistance and excellent technical support of Marian Mulholland, RN, BSN, and the staff of the Cardiac Catheterization Laboratory at the Johns Hopkins Hospital.
☆ This work was supported by grant HL-03238 (to Dr. Hare) from the National Heart, Lung, and Blood Institute, National Institutes of Health, Bethesda, Maryland, and by a medical school grant from Merck, Inc. Dr. Hare is the recipient of a Paul Beeson Physician Faculty Scholars in Aging Research Award.
- angiotensin-converting enzyme
- angiotensin-II receptor, type 1
- dilated cardiomyopathy
- peak rate of rise of left ventricular pressure
- arterial elastance
- ventricular elastance
- left ventricular end-diastolic pressure
- left ventricular end-diastolic volume
- nitric oxide
- nitric oxide synthase
- Received December 14, 2000.
- Revision received April 19, 2001.
- Accepted April 27, 2001.
- American College of Cardiology
- Hare J.M,
- Loh E,
- Creager M.A,
- Colucci W.S
- Keaney J.F Jr.,
- Hare J.M,
- Kelly R.A,
- Loscalzo J,
- Smith T.W,
- Colucci W.S
- Hare J.M,
- Lofthouse R.A,
- Juang G.J,
- et al.
- Hare J.M,
- Givertz M.M,
- Creager M.A,
- Colucci W.S
- Brunner H.R,
- Christen Y,
- Munafo A,
- Lee R.J,
- Waeber B,
- Nussberger J
- Senzaki H,
- Chen C.H,
- Kass D.A
- Kelly R.P,
- Ting C.T,
- Yang T.M,
- et al.
- Senzaki H,
- Fetics B,
- Chen C.H,
- Kass D.A
- Balligand J.-L,
- Kelly R.A,
- Marsden P.A,
- Smith T.W,
- Michel T
- Kitakaze M,
- Minamino T,
- Node K,
- et al.
- Foult J.M,
- Tavolaro O,
- Antony I,
- Nitenberg A
- Friedrich S.P,
- Lorell B.H,
- Rousseau M.F,
- et al.
- Paulus W.J,
- Vantrimpont P.J,
- Shah A.M
- Zhang X,
- Xie Y.W,
- Nasjletti A,
- Xu X,
- Wolin M,
- Hintze T.H
- Shinke T,
- Takaoka H,
- Takeuchi M,
- et al.
- Haywood G.A,
- Tsao P.S,
- von der Leyen H.E,
- et al.
- Heymes C,
- Vanderheyden M,
- Bronzwaer J.G,
- Shah A.M,
- Paulus W.J
- Drexler H,
- Kästner S,
- Strobel A,
- Studer R,
- Brodde O.E,
- Hasenfuss G
- Maisel A.S,
- Phillips C,
- Michel M.C,
- Ziegler M.G,
- Carter S.M