Acute Endothelin A Receptor Blockade Causes Selective Pulmonary Vasodilation in Patients With Chronic Heart Failure
Background—Elevated plasma endothelin-1 (ET-1) levels in patients with chronic heart failure correlate with pulmonary artery pressures and pulmonary vascular resistance. ETA receptors on vascular smooth muscle cells mediate pulmonary vascular contraction and hypertrophy. We determined the acute hemodynamic effects of sitaxsentan, a selective ETA receptor antagonist, in patients with chronic stable heart failure receiving conventional therapy.
Methods and Results—This multicenter, double-blind, placebo-controlled trial enrolled 48 patients with chronic New York Heart Association functional class III or IV heart failure (mean left ventricular ejection fraction 21±1%) treated with ACE inhibitors and diuretics. Patients with a baseline pulmonary capillary wedge pressure ≥15 mm Hg and a cardiac index ≤2.5 L · min−1 · m−2 were randomized to 1 of 3 doses (1.5, 3.0, or 6.0 mg/kg) of sitaxsentan or placebo as an intravenous infusion over 15 minutes. Hemodynamic responses were assessed by catheterization of the right side of the heart for 6 hours. Sitaxsentan decreased pulmonary artery systolic pressure, pulmonary vascular resistance, mean pulmonary artery pressure, and right atrial pressure (P≤0.001, 0.003, 0.017, and 0.031, respectively) but had no effect on heart rate, mean arterial pressure, pulmonary capillary wedge pressure, cardiac index, or systemic vascular resistance. Plasma ET-1 levels were elevated at baseline and decreased with sitaxsentan.
Conclusions—In patients with moderate to severe heart failure receiving conventional therapy, acute ETA receptor blockade caused selective pulmonary vasodilation associated with a reduction in plasma ET-1. Sitaxsentan may be of value in the treatment of patients with pulmonary hypertension secondary to chronic heart failure.
Plasma endothelin-1 (ET-1) levels are elevated in patients with heart failure1 and have prognostic significance.2 ET-1 levels correlate strongly with pulmonary artery pressures and pulmonary vascular resistance (PVR),1 3 which suggests that ET-1 may be a mediator of reactive pulmonary hypertension in the setting of left ventricular (LV) failure. The actions of ET-1 are mediated by 2 receptor subtypes, designated ETA and ETB. ETA receptors located on vascular smooth muscle cells mediate vasoconstriction.4 ETB receptors located primarily on vascular endothelial cells mediate NO-dependent vasodilation,5 although some ETB receptors located on vascular smooth muscle cells may mediate vasoconstriction.6 ETB receptors also play a role in ET-1 clearance.7
Understanding of the actions of ET-1 in heart failure has been aided by the development of ET receptor antagonists. Acute intravenous8 and short-term oral9 administration of bosentan, a nonselective (ETA/ETB) receptor antagonist, caused systemic and pulmonary vasodilation in patients with severe heart failure. Because ETB receptors are coupled to endothelium-dependent vasodilation, it might be anticipated that ETA-selective antagonists would be more effective vasodilators than nonselective antagonists. However, there has been relatively little experience with ETA-selective antagonists in patients with heart failure, save for a small, uncontrolled study of the intravenous administration of the ETA-selective antagonist BQ-123.10
Sitaxsentan is a nonpeptide, small organic molecule that binds competitively to the human ETA receptor with high affinity (KI=0.43 nmol/L). Sitaxsentan is 6000-fold more selective for the human ETA receptor than for the human ETB receptor.11 Preclinical studies have shown that sitaxsentan exerts potent vasodilator effects. In rats with hypoxia-induced pulmonary hypertension, sitaxsentan attenuated the increase in pulmonary artery pressure but had no effect on heart rate (HR) or mean arterial pressure (MAP) (Tommy Brock, PhD, unpublished data, 1997). These preliminary data suggest that sitaxsentan may exert a specific action on the pulmonary circulation. We undertook a double-blind, placebo-controlled, randomized study with sitaxsentan to determine the acute hemodynamic effects of ETA receptor blockade in patients with stable chronic heart failure.
The study population consisted of 48 adults, aged ≥21 years, with chronic New York Heart Association (NYHA) class III or IV heart failure and an LV ejection fraction ≤35%. All patients were receiving a diuretic and ACE inhibitor (or had failed attempts to start an ACE inhibitor). Digoxin, amlodipine or nifedipine, and metoprolol or carvedilol were allowed. All patients had a baseline pulmonary capillary wedge pressure (PCWP) ≥15 mm Hg and a cardiac index (CI) ≤2.5 L · min−1 · m−2. Exclusion criteria included myocardial infarction within 8 weeks; unstable angina or angina-limited exercise; heart failure due to uncorrected valvular stenosis, obstructive cardiomyopathy, pericardial disease, amyloidosis, active myocarditis, or a malfunctioning artificial heart valve; symptomatic chronic obstructive pulmonary disease or asthma; history of symptomatic ventricular tachycardia, ventricular fibrillation, or sudden death unless treated with an implantable defibrillator; history of second- or third-degree A-V block unless treated with a pacemaker; HR >115 bpm; and systolic blood pressure <90 mm Hg or >200 mm Hg. The study protocol was approved by the institutional review committee at each study site, and all patients provided written informed consent.
Randomization and Blinding
A single computer-generated randomization code was used for all sites and was coordinated by an independent randomization center. The first 24 patients were randomized in a 1:1:1 ratio to 1 of 3 treatment groups: sitaxsentan 1.5 mg/kg (dose level 1), sitaxsentan 3.0 mg/kg (dose level 2), or placebo (5% dextrose in water). Based on the safety results of the first 24 subjects enrolled and an interim analysis of the pharmacodynamic effects of the 1.5 and 3.0 mg/kg doses, the protocol was amended to eliminate dose level 1 and add a third dose of sitaxsentan (6.0 mg/kg; dose level 3). The subsequent 24 patients were randomized in a 1:1:1 ratio to: sitaxsentan 3.0 mg/kg, sitaxsentan 6.0 mg/kg, or placebo. Study drug was prepared by a research pharmacist and infused in a blinded fashion over 15 minutes into a peripheral vein by site personnel who were not associated with the subject’s clinical management.
After insertion of a Swan-Ganz catheter and after ≥1 hour of stabilization, 2 sets of baseline hemodynamic measurements were obtained. When CI and PCWP values differing by <15% were obtained on subsequent measurements, the study drug was infused over 15 minutes, and the following hemodynamic measurements were obtained at 15 and 30 minutes and at 1, 2, 3, 4, and 6 hours after completion of drug infusion: HR, systemic arterial pressure (systolic and diastolic), pulmonary arterial pressure (mean systolic and diastolic), PCWP, cardiac output (by thermodilution or Fick method), and mean right atrial pressure (RAP). Mean systemic arterial pressure, systemic vascular resistance (SVR), PVR, CI, and LV stroke work index were derived according to standard formulas.
All medications with possible hemodynamic effects, including vasodilators, ACE inhibitors, β-blockers, calcium channel blockers, diuretics, digoxin, and antiarrhythmics, were withheld beginning at midnight on the day of the study and throughout the 6-hour blinded evaluation period. Sitaxsentan sodium (Texas Biotechnology Corporation) was supplied as 250 mg of a sterile, yellow lyophilized cake in a clear 20-mL vial. Sitaxsentan was reconstituted with 10 or 20 mL of sterile water to yield a final concentration of 25 or 12.5 mg/mL, respectively, and the infusion dose was calculated based on the patient’s weight and randomized dose level. Patients were limited to light liquid meals no less than 1 hour before baseline measurements and after hemodynamic readings were obtained at hour 4.
Blood was obtained for general chemistry, complete blood count with differential, and international normalized ratio (INR) at baseline and after completion of the study. Plasma ET-1, norepinephrine, and tumor necrosis factor-α (TNF-α) levels were determined from blood drawn at baseline and at 3 and 6 hours after study-drug infusion. ET-1 levels were measured by radioimmunoassay with a commercial kit (Phoenix Pharmaceuticals). TNF-α levels were measured by ELISA with a commercial kit (R&D Systems). Norepinephrine levels were measured by high-performance liquid chromatography with electrochemical detection.
To determine the comparability of the 4 treatment groups for demographic and baseline variables, categorical variables were analyzed via Fisher’s exact test and continuous variables by 1-way ANOVA with effects for treatment. Primary efficacy analyses were performed on all enrolled patients according to intention to treat, and missing data were replaced by the last-observation-carried-forward principle. For each time point, differences among the treatment groups in mean change from baseline were determined via 1-way ANOVA with effects for treatment. In addition, a repeated-measures model was used on the changes from baseline to determine the effects of treatment, time, and the interaction of treatment and time. Safety was assessed by comparing the treatments with respect to the occurrence of adverse experiences and the changes from baseline in laboratory values and ECG measurements. All analyses were performed with the SAS statistical package (SAS Institute), and data are presented as mean±SEM, with a value of P<0.05 considered statistically significant.
There were no significant differences among the treatment groups with respect to age, sex, LV ejection fraction, NYHA class, or resting hemodynamics (Table 1⇓). Eighty-three percent of patients were male, and 77% were in NYHA functional class III. The majority of patients were taking an ACE inhibitor (92%), diuretic (96%), and digoxin (88%). Resting hemodynamics revealed moderate-to-severe heart failure with elevated right- and left-heart filling pressures, reduced CI, and increased SVR. Pulmonary hypertension was present with a mean pulmonary artery systolic pressure (PASP) of 60±2 mm Hg and a mean PVR of 349±33 dyne · s · cm−5.
Hemodynamic Effects of Sitaxsentan
The hemodynamic effects of sitaxsentan are shown in Table 2⇓⇓. Table 3⇓ summarizes the overall treatment effect of sitaxsentan on hemodynamic parameters as analyzed by intention to treat across all doses (versus placebo). Compared with placebo, sitaxsentan caused significant reductions in PASP, PVR, mean pulmonary artery pressure (MPAP), and RAP (P≤0.001, 0.003, 0.017, and 0.031, respectively) (Figure 1⇓).
The decrease in PASP was significant by 30 minutes, maximal at 2 to 3 hours, and persisted for ≥6 hours after the infusion (Figure 1A⇑). The decrease in PASP was similar at all 3 doses tested. At 2 hours, PASP decreased by 13%, 10%, and 11% in the sitaxsentan 1.5-, 3.0-, and 6.0-mg/kg groups, respectively, and increased by 4% in the placebo group. Likewise, at 2 hours, PVR decreased by 38%, 34%, and 20% in the sitaxsentan 1.5-, 3.0-, and 6.0-mg/kg groups, respectively, and increased by 20% in the placebo group (Figure 1B⇑). In patients receiving sitaxsentan, baseline PVR predicted the magnitude of the decrease in PVR at 2 hours (Figure 2⇓). At 2 hours, RAP decreased by 13%, 23%, and 17% in the sitaxsentan 1.5-, 3.0-, and 6.0-mg/kg groups, respectively, and increased by 11% in the placebo group.
In contrast to the effects on pulmonary and right-heart hemodynamics, sitaxsentan exerted no discernible effect on HR, MAP, PCWP, or SVR (Tables 2⇑ and 3⇑). There were weak trends for CI to increase (P=0.202) and MAP to decrease (P=0.258).
Plasma ET-1, Norepinephrine, and TNF-α Levels
In the study population as a whole, baseline ET-1 levels were markedly elevated, with a mean value of 15.4±1.5 pg/mL. There were no between-group differences. Baseline ET-1 levels did not correlate with baseline PVR (r=0.05, P=0.75) or the reduction in PVR (2 hours) with sitaxsentan (r=0.26, P=0.19). Of note, plasma ET-1 levels decreased by 18±8% and 21±8% at 3 and 6 hours, respectively, in patients receiving sitaxsentan but did not change in the placebo group (Figure 3⇓). Plasma norepinephrine (680±49 pg/mL) and TNF-α (3.3±0.3 pg/mL) levels were elevated at baseline but did not change in any of the treatment groups (data not shown).
Overall, 46% of all patients experienced an adverse event (Table 4⇑). However, except for an increase in the frequency of gastrointestinal events reported with sitaxsentan (2 in the 6.0-mg/kg dose group; P=0.05), there were no significant differences among treatment groups with respect to adverse events, including congestive heart failure, hypotension, bradycardia, and ventricular arrhythmias. Sitaxsentan had no effect on the complete blood count, serum blood urea nitrogen or creatinine, liver-function tests, or INR over the 6-hour study period.
The major new finding of this study is that acute infusion of the ETA-selective antagonist sitaxsentan decreased PVR in patients with chronic stable heart failure but exerted little or no effect on systemic vascular tone. This selective pulmonary vasodilator action was apparent within 30 minutes, persisted for ≥6 hours after drug infusion, and was maximal at the lowest dose tested. Of note, the hemodynamic effect of sitaxsentan was associated with a decrease in plasma ET-1 levels.
This pattern of hemodynamic effect appears to differ from that observed with acute administration of the nonselective ETA/ETB receptor antagonist bosentan. Kiowksi et al8 randomized 24 patients with NYHA class III heart failure to receive bosentan or placebo by intravenous infusion. Bosentan decreased MAP, MPAP, PCWP, PVR, and SVR and increased CI. Similar hemodynamic effects were seen after 2 weeks of oral bosentan therapy.9 Cowburn et al10 infused the ETA-selective antagonist BQ-123 for 60 minutes in 8 patients with stable chronic heart failure. In that uncontrolled study, BQ-123 caused modest decreases in SVR (12%) and PVR (14%), and a small increase in CI.
It is possible that the differing hemodynamic effects of sitaxsentan and bosentan reflect their differences in receptor selectivity. It is also possible that they are due to an inherent difference in the study populations or protocols. Regardless of the reason, it can be concluded that under the conditions of the present study, acute infusion of sitaxsentan exerted a selective pulmonary vasodilator effect. Such an action is consistent with the demonstration that ET-1 causes concentration-dependent contraction of pulmonary arteries and veins in vitro12 and increases PVR in vivo.13 Both of these actions are mediated predominantly by the ETA receptor.14 Stimulation of ETB receptors in the pulmonary vasculature generally attenuates ETA-mediated vasoconstriction,15 although pulmonary vasoconstrictor effects have also been reported.16
The extent to which sitaxsentan decreased PVR was strongly correlated with the baseline PVR. This observation may explain the lack of a significant pulmonary vasodilator effect with BQ-123, because in the study by Cowburn et al,10 patients had only mild pulmonary hypertension with a mean PVR of ≈180 dyne · s · cm−5. By contrast, our patients had moderate-to-severe pulmonary hypertension with a mean PVR of 349 dyne · s · cm−5. In patients with heart failure, inhaled NO also exerts a selective pulmonary vasodilator action, apparently owing to its route of administration.17 Interestingly, we found that the pulmonary vasodilator effect of inhaled NO also correlated best with the baseline PVR.17 However, in contrast to inhaled NO, sitaxsentan did not cause an increase in PCWP.
It is surprising that sitaxsentan did not decrease SVR. Infusion of another ETA-selective antagonist, BQ-123, into the brachial artery of patients with heart failure caused vasodilation.18 Although it is possible that higher doses of sitaxsentan would have decreased SVR, the pulmonary vasodilator effects that we observed appeared to be maximal at the lowest dose tested and, if anything, may have diminished at higher doses. One possible explanation for these findings is that a component of the elevated SVR in patients with heart failure is due to vasoconstrictor ETB receptors located on systemic resistance vessels. In support of this thesis, the intra-arterial infusion of sarafotoxin S6c, an ETB receptor agonist, causes an enhanced vasoconstrictor response in patients with heart failure.18 Likewise, intravenous infusion of the selective ETB receptor agonist ET-3 resulted in constriction of systemic but not pulmonary vasculature in patients with LV failure.19 An enhanced vasoconstrictor response to ETB stimulation in heart failure might reflect desensitization of ETA receptors in the systemic vasculature.20
Despite a reduction in PVR, sitaxsentan did not significantly increase CI. There are several possible explanations. As mentioned, we did not observe a decrease in SVR, and in the absence of a positive inotropic effect, sitaxsentan would not be expected to increase CI. Second, we did not observe an increase in LV preload. Indeed, even if PCWP had increased with sitaxsentan, we might not have observed an increase in CI with selective pulmonary vasodilation. In a prior study using the selective pulmonary vasodilator inhaled NO, we demonstrated small but significant decreases in CI and stroke volume index associated with a 23% increase in PCWP.17 Another potential explanation for the lack of increase in CI is that sitaxsentan exerted a negative inotropic effect. ETA receptors are expressed in human myocardium21 and mediate the positive inotropic effect of ET-1.22
Consistent with previous reports,1 we found increased plasma ET-1 levels in patients with moderate-to-severe heart failure. However, baseline ET-1 levels did not correlate with pulmonary artery pressures or PVR, or with the change in PVR after ETA blockade. In prior studies8 9 with the nonselective ET receptor antagonist bosentan, ET-1 levels increased within 2 to 3 hours of administration. The increase in plasma ET-1 with bosentan is thought to be due to displacement of ET-1 from receptor binding sites and/or decreased clearance by ETB receptors.7 In striking contrast, sitaxsentan caused a decrease in ET-1 levels. The mechanism for this decrease in plasma ET-1 is not clear. Given the important role of the pulmonary vasculature in contributing ET-1 to the systemic circulation, it is possible that it reflects reduced ET-1 pulmonary vascular production due to improved pulmonary hemodynamics. Another possibility is that selective antagonism of the ETA receptor allows more ET-1 to be available for clearance via the ETB receptor.
An important pathophysiological implication of these findings is that reactive pulmonary hypertension in patients with heart failure is mediated, at least in part, by ETA receptors. The ratio of ETA to ETB receptors on human resistance and conduit pulmonary arteries is ≈10:1.23 In animal models of pulmonary hypertension, ETA receptors mediate vasoconstriction and smooth muscle cell proliferation.24 ETB receptors found primarily on vascular endothelial cells exert opposing actions with regard to both vascular tone5 and hypertrophy.25 Thus, selective antagonism of the ETA receptor would be expected to cause pulmonary vasodilation, whereas nonselective ETA/ETB blockade might lead to less vasodilation.
A clinical implication of these findings is that ETA-selective blockade may be of particular value in patients with reactive pulmonary hypertension in the setting of heart failure. The apparent relative lack of systemic vasodilation would not support a primary role for sitaxsentan for the purpose of LV afterload reduction. However, there is evidence from animal studies that ET-1 antagonists may reduce LV remodeling.26 Because this antiremodeling action may be mediated directly at the myocardial level,27 the relative lack of systemic vasodilation with sitaxsentan may be of little consequence in terms of its long-term cardioprotective benefits and might actually facilitate its use as an adjunctive agent in patients with already low SVR due to vasodilator or ACE inhibitor therapy.
Participating investigators are listed alphabetically with their corresponding location: W. Colucci, D. Gauthier, M. Givertz: Boston University Medical Center, Boston, Mass; S. Gottlieb, J. Marshall: University of Maryland, Baltimore, Md; M. Fracyon, J. Hare: Johns Hopkins Medical Center, Baltimore, Md; P. Binkley, C. Leier, L. Newton: Ohio State University, Columbus, Ohio; T. LeJemtel, P. Levato: Albert Einstein College of Medicine, Bronx, NY; E. Galbraith, B. Lewis: Loyola University Medical Center, Maywood, Ill; K. Craig, E. Loh: University of Pennsylvania Hospital, Philadelphia, Pa; J. Nicklas, M. Waidley: University of Michigan Hospital, Ann Arbor, Mich; L. Keane, M. Slawsky: Boston VA Medical Center, Boston, Mass.
Guest Editor for this article was Kirk L. Peterson, MD, University of California at San Diego.
↵1 See Appendix for a complete list of the study investigators.
- Received August 20, 1999.
- Revision received December 15, 1999.
- Accepted February 1, 2000.
- Copyright © 2000 by American Heart Association
Cody RJ, Haas GJ, Binkley PF, et al. Plasma endothelin correlates with the extent of pulmonary hypertension in patients with chronic congestive heart failure. Circulation. 1992;85:504–509.
Pousset F, Isnard R, Lechat P, et al. Prognostic value of plasma endothelin-1 in patients with chronic heart failure. Eur Heart J. 1997;18:254–258.
Bialecki RA, Fisher CS, Murdoch WW, et al. Functional comparison of endothelin receptors in human and rat pulmonary artery smooth muscle. Am J Physiol. 1997;272:L211–L218.
de Nucci G, Thomas R, D’Orleans-Juste P, et al. Pressor effects of circulating endothelin are limited by its removal in the pulmonary circulation and by the release of prostacyclin and endothelium-derived relaxing factor. Proc Natl Acad Sci U S A. 1988;85:9797–9800.
Haynes WG, Strachan FE, Webb DJ. Endothelin ETA and ETB receptors cause vasoconstriction of human resistance and capacitance vessels in vivo. Circulation. 1995;92:357–363.
Dupuis J, Goresky CA, Fournier A. Pulmonary clearance of circulating endothelin-1 in dogs in vivo: exclusive role of ETB receptors. J Appl Physiol. 1996;81:1510–1515.
Sutsch G, Kiowski W, Yan X-W, et al. Short-term oral endothelin-receptor antagonist therapy in conventionally treated patients with symptomatic severe chronic heart failure. Circulation. 1998;98:2262–2268.
Zellers TM, McCormick J, Wu Y. Interaction among ET-1, endothelium-derived nitric oxide, and prostacyclin in pulmonary arteries and veins. Am J Physiol. 1994;267:H139–H147.
Lippton HL, Hauth TA, Summer WR, et al. Endothelin produces pulmonary vasoconstriction and systemic vasodilation. J Appl Physiol. 1989;66:1008–1012.
Loh E, Stamler JS, Hare JM, et al. Cardiovascular effects of inhaled nitric oxide in patients with left ventricular dysfunction. Circulation. 1994;90:2780–2785.
Love MP, Haynes WG, Gray GA, et al. Vasodilator effects of endothelin-converting enzyme inhibition and endothelin ETA receptor blockade in chronic heart failure patients treated with ACE inhibitors. Circulation. 1996;94:2131–2137.
Calderone A, Rouleau JL, de Champlain J, et al. Regulation of the endothelin-1 transmembrane signaling pathway: the potential role of agonist-induced desensitization in the coronary artery of the rapid ventricular pacing-overdrive dog model of heart failure. J Mol Cell Cardiol. 1993;25:895–903.
Ponicke K, Vogelsang M, Heinroth M, et al. Endothelin receptors in the failing and nonfailing human heart. Circulation. 1998;97:744–751.
Fukuroda T, Kobayashi M, Ozaki S, et al. Endothelin receptor subtypes in human versus rabbit pulmonary arteries. J Appl Physiol. 1994;76:1976–1982.
Zamora MA, Dempsey EC, Walchak SJ, et al. BQ123, an ETA receptor antagonist, inhibits endothelin-1-mediated proliferation of human pulmonary artery smooth muscle cells. Am J Respir Cell Mol Biol. 1993;9:429–433.
Mulder P, Richard V, Derumeaux G, et al. Role of endogenous endothelin in chronic heart failure: effect of long-term treatment with an endothelin antagonist on survival, hemodynamics, and cardiac remodeling. Circulation. 1997;96:1976–1982.