- Email: [email protected]
Comparison of vasopeptidase inhibitor, omapatrilat, and lisinopril on exercise tolerance and morbidity in patients with heart failure: IMPRESS randomised trial
ARTICLES
Articles
Comparison of vasopeptidase inhibitor, omapatrilat, and lisinopril on exercise tolerance and morbidity in patients with heart failure: IMPRESS randomised trial Jean L Rouleau, Marc A Pfeffer, Duncan J Stewart, Debra Isaac, Francois Sestier, Edmund K Kerut, Charles B Porter, Guy Proulx, Chunlin Qian, Alan J Block, for the IMPRESS investigators
Summary Background We aimed to assess in patients with congestive heart failure whether dual inhibition of neutral endopeptidase and angiotensin-converting enzyme (ACE) with the vasopeptidase inhibitor omapatrilat is better than ACE inhibition alone with lisinopril on functional capacity and clinical outcome. Methods We did a prospective, randomised, double-blind, parallel trial of 573 patients with New York Heart Association (NYHA) class II–IV congestive heart failure, left-ventricular ejection fraction of 40% or less, and receiving an ACE inhibitor. Patients were randomly assigned omapatrilat at a daily target dose of 40 mg (n=289) or lisinopril at a daily target dose of 20 mg (n=284) for 24 weeks. The primary endpoint was improvement in maximum exercise treadmill test (ETT) at week 12. Secondary endpoints included death and comorbid events indicative of worsening heart failure. Findings Week 12 ETT increased similarly in the omapatrilat and lisinopril groups (24 vs 31 s, p=0·45). The two drugs were fairly well tolerated, but there were fewer cardiovascularsystem serious adverse events in the omapatrilat group than in the lisinopril group (20 [7%] vs 34 [12%], p=0·04). There was a suggestive trend in favour of omapatrilat on the combined endpoint of death or admission for worsening heart failure (p=0·052; hazard ratio 0·53 [95% CI 0·27–1·02]) and a significant benefit of omapatrilat in the composite of death, admission, or discontinuation of study treatment for worsening heart failure (p=0·035; 0·52 [0·28–0·96]). Omapatrilat improved NYHA class more than lisinopril in patients who had NYHA class III and IV (p=0·035), but not if patients with NYHA class II were included. Interpretation Our findings suggest that omapatrilat could have some advantages over lisinopril in the treatment of patients with congestive heart failure. Thus use of vasopeptidase inhibitors could constitute a potentially Division of Cardiology, Toronto General Hospital, University of Toronto, Eaton North 13-312, Toronto M5G 2C4, ON, Canada (J L Rouleau MD); Division of Cardiology, Brigham and Women Hospital, Havard Medical School, Boston, MA, USA (M A Pfeffer MD); Division of Cardiology, St Michael’s Hospital, University of Toronto (D Stewart MD); Department of Cardiology, Foothills General Hospital, University of Calgary, Calgary (D Issac MD); Department of Cardiology, Centre Hospitalier de l’Université de Montréal-Pavillon Notre-Dame, Montreal (F Sestier MD); Heart Clinic of Louisiana, Marrero LA, USA (E K Kerut MD); Mid-America Cardiology Associates, Kansas City, MO, USA (C B Porter MD); Division of Cardiology, Montreal Heart Institute, University of Montreal (G Proulx MD); and Pharmaceutical Research Institute, Bristol-Myers Squibb, Princeton, NJ, USA (C Qian PhD, A J Block PhD) Correspondence to: Jean L Rouleau (e-mail: [email protected])
THE LANCET • Vol 356 • August 19, 2000
important treatment for further improving the prognosis and well being of patients with this disorder. Lancet 2000; 356: 615–20
Introduction Congestive heart failure is characterised by chronic overactivation of sodium-retaining and water-retaining neurohormones.1 Overactivation of these vasoconstrictor neurohumoral systems also leads to excessive cellular growth, cardiac fibrosis, and cellular toxicity.1 Attenuating the effects of these neurohumoral systems has been one of the most successful strategies in lowering the morbidity and mortality of patients with congestive heart failure.2,3 Although the use of angiotensin-converting-enzyme (ACE) inhibitors and -blockers has been beneficial in this respect, the outlook of these patients remains poor. Therefore, new strategies to improve outlook need to be developed.2,3 Human beings have developed several endogenous systems for countering the effects of overactivation of vasoconstrictor neurohormones.1 One mechanism is endogenous vasodilator systems, which include natriuretic peptides, nitric oxide, and prostaglandins.1 These peptides vasodilate and promote diuresis and natriuresis and lessen cellular growth.4–7 In congestive heart failure, endogenous vasodilator systems are activated to try to compensate for chronic activation of vasoconstrictor sodium and waterretaining neurohormones. The vasopeptidase inhibitors, a new class of pharmaceutical agents, have been shown to heighten activity of endogenous vasodilator systems and reduce production of the vasoconstrictor angiotensin II.8 Vasopeptidase inhibitors inhibit the activity of neutral endopeptidase, an enzyme that metabolises endogenous vasodilator peptides, such as natriuretic peptides (atrial, brain, and calcium-activated neutral protease), adrenomedullin, and bradykinin.8 Because these inhibitors better redress the imbalance between endogenous vasoconstrictor and vasodilator substances in congestive heart failure than does ACE inhibition alone, they could be more useful in treatment of patients with the disorder. We did a randomised double-blind trial to compare the effects of omapatrilat with those of lisinopril on exercise tolerance in patients with congestive heart failure. We also assessed side-effects of the treatments, effects on death rate, and comorbid events for worsening heart failure.
Methods Omapatrilat Omapatrilat (BMS-186716) is the first of a new class of cardiovascular agents termed vasopeptidase inhibitors. Omapatrilat is an orally active, long acting, selective, competitive inhibitor of neutral endopeptidase (NEP; enkephalinase, neprilysin, EC 3.4.24.11) and angiotensin converting enzyme (ACE; EC 3.4.15.1) with similar K1
615
For personal use only. Not to be reproduced without permission of The Lancet.
ARTICLES
H
S O HS N H
N O
COOH
Figure 1: Chemical structure of omapatrilat Ki values, neutral endopeptidase 8·9 nM, ACE 6·0 nM.
values for both NEP and ACE (figure 1). As a result, omapatrilat potentiates multiple endogenous vasodilatory peptides including natriuretic peptides (atrial, brain, and calcium activated neutral protease), bradykinin, and adrenomedullin, while also inhibiting the generation of the vasoconstrictive peptide, angiotensin II. The chemical name is [4S-{4␣(R*),7␣10a]}-octahydro-4-[2-mercapto1-oxo-3-phenypropyl)amino]-5-oxo=7H-pyrido[2,1b][1,3]-thiazepine-7-carboxylic acid; its structure is shown in figure 1. Omapatrilat has an oral bioavailability of about 30%, it may be given with or without food, its protein binding is 80%, and it has a very large volume of distribution (1800 L) suggesting tissue penetration. Time to maximum concentration of an oral dose is about 2 h. Its effective half-life is 14–19 h. This drug is metabolised in the liver. Omapatrilat forms disulphide bonds with endogenous thiols, and is extensively metabolised via S-methylation, amide hydrolysis, S-oxidation, and glucuronidation. There are no substantial concentrations of active metabolites of omapatrilat in plasma. About 80% of an intravenous radioactive dose and 64% of an oral radioactive dose were recovered in urine, with less than 1% excreted as unchanged drug (data available). Based on biotransformation data in vitro and in vivo, cytochrome P450 enzymes do not seem to be involved in the metabolism of omapatrilat. In-vitro studies show that omapatrilat does not inhibit P450 isoenzmes CYP-1A2, CYP-3A4, CYP-2C9, CYP-2C19, and CYP-2D6. Omapatrilat has no known interaction with other medications. Study population We did the study in 113 centres in the USA and Canada. The study was approved by central or local-institution review boards. Patients gave written informed consent. Patients were aged at least 18 years, with stable (>3 months) symptomatic heart failure (NYHA II–IV), decreased left-ventricular ejection fraction, and receiving a stable (⭓4 weeks) dose of ACE inhibitors. Patients were eligible if they had a seated systolic blood pressure of 90 mm Hg or higher at randomisation, but no uncontrolled hypertension or history of acute coronary events or revascularisation procedures within 3 months. Patients with serum potassium of less than 3·5 or more than 5·3 mmol/L or creatinine of more than 221 mol/L were excluded as were patients with transaminases more than twice the upper limit of normal, leucocytes less than 3·0⫻109/L, neutrophils of less than 1·5⫻109/L, or platelets of less than 120⫻109/L. -blockers were allowed only if they had been used for at least 6 months, calcium-channel blockers were allowed only for control of atrial fibrillation, and patients who had been
616
included in previous randomised trials of omapatrilat were excluded. Angiotensin-receptor antagonists, vasodilators other than nitrates, and inotropes other than digoxin were prohibited during the study. Finally, inability to give informed consent, or any other medical disorder judged by the investigator to limit optimum participation in the study, were exclusion criteria. Randomisation and study treatment Maximum-exertion-exercise tolerance tests (ETT) were undertaken with a modified Naughton protocol 21–27 h after administration of study and concomitant cardiovascular drugs. ETTs had to be limited by either dyspnoea or fatigue. After initial qualification and a screening ETT of less than 12 min, patients entered a placebo lead-in period, in which they did a qualification ETT that had to be 2–12 min There were no time constraints on the baseline ETT, and it was done within 9 days of the qualifying study. Patients who met all entrance criteria stopped ACE-inhibitor treatment and were randomised the next day to active treatment with 10 mg omapatrilat titrated to the target dose of 40 mg once daily, or 5 mg lisinopril once daily, titrated to 20 mg daily. Forced titration to target or maximum tolerated dose was used at weekly intervals for up to 3 weeks. We based the dose of omapatrilat on a previous long-term haemodynamic study that showed significant reduction of pulmonary capillarywedge pressure with 40 mg omapatrilat.9 Administration of concomitant cardiovascular treatment was required on the mornings of titration visits, but prohibited on the mornings of other visits, including those to assess exercise time. Assessment of patients and endpoints Clinical assessments were done weekly until week 3, and then at weeks 6, 12, 18, and 24. Full laboratory assessments were done at baseline, and at weeks 12 and 24, and abbreviated laboratory panels were obtained at all other visits after randomisation. The primary endpoint was change in exercise duration from baseline to week 12. Secondary endpoints were: the combined endpoint of death and admission for worsening heart failure, and the combined endpoint of death and comorbidity for worsening heart failure (admission, discontinuation of study treatment, emergency-room visit of clinical need for supplemental diuretic). Additional endpoints included NYHA functional class, physician and patient global assessment of heart-failure status at 24 weeks. Safety assessment included occurrence of treatmentemergent adverse events, as well as clinical laboratory assessment of raised concentrations of serum creatinine (>1·5 times baseline and >0·8 times limit of normal), blood urea nitrogen (>2 times baseline and higher than upper limit of normal), potassium, and liver-function tests. A serious adverse event was defined as any of: fatal or life threatening, permanently disabling prolonging time spent in hospital, an important medical event, congenital anomaly, cancer, or overdose. We used equilibrium gated radionuclide ventriculography in a subset of 75 patients to measure left ventricular and right ventricular volumes and ejection fractions at baseline and 24 weeks. Data were analysed by a core laboratory at the New England Medical Centre.10 116 patients participated in the neurohormone substudy. Before the morning dose of the study drugs, a venous cannula was inserted into the patient’s antecubital vein. After 30 min, during which the patients remained recumbent, about 28 mL of blood was withdrawn,
THE LANCET • Vol 356 • August 19, 2000
For personal use only. Not to be reproduced without permission of The Lancet.
ARTICLES
centrifuged at 4ºC, and the plasma stored at ⫺80ºC before analysis. Plasma atrial natriuretic peptide, norepinephrine, angiotensin II, and endothelin 1 were analysed11 by a core laboratory at the Montreal Heart Institute (Quebec, Canada). We measured plasma neurohormones at baseline, and week 12 and 24. Statistical analysis We estimated that 532 participants (266 per group) would be needed to provide 80% power at ␣=0·05 to detect a 30 s difference at week 12 between the two treatment groups (SD 120 s). Patients who stopped taking study drugs were asked to return to do ETTs at weeks 12 and 24, if medically appropriate, so that the primary test for difference could be done by intention to treat. We did a secondary analysis to test for change in exercise performance from baseline to weeks 12 and 24 by ANOVA, with terms for baseline exercise time and treatment. Time-to-event analyses for mortality and comorbidity were based on Cox’s proportional hazard regression with treatment as the only covariant. We assessed treatment difference in NYHA functional class at week 24 with the Cochran-MantelHaenszel method (ANOVA statistic, rank scores) controlling for baseline class (the last observation was used for patients who withdrew). We measured treatment difference in neurohumoral concentrations at 12 weeks and 24 weeks with a two-tailed student t test before and after log transformation, which results in near normal distribution for neurohumoral concentrations.12
Results
Characteristic
Omapatrilat (n=289)
Lisinopril (n=284)
227/62
224/60
Demography Male/female Ethnic origin White Black Other Mean (SD) age (years) NYHA class II III IV
237 26 26 64·3 (10·7)
240 24 20 63·6 (10·0)
186 100 3
175 107 2
Heart failure cause Ischaemic Idiopathic dilated cardiomyopathy Valvular Hypertension Other/not reported
186 (64%) 73 (25%) 10 (3%) 10 (3%) 10 (3%)
191 (67%) 65 (23%) 9 (3%) 12 (4%) 7 (3%)
Measurements (mean [SD]) Baseline ETT (s) Left-ventricular ejection fraction (%) Standing systolic blood pressure (mm Hg) Standing diastolic blood pressure (mm Hg) Standing heart rate (beats/min) Serum potassium (mmol/L) Serum creatinine (mol/L) Blood urea nitrogen (mmol/L urea)
511 (159) 28·4 (7·5) 126·6 (19·7) 76·2 (10·6) 79·6 (12·9) 4·42 (0·39) 103·4 (30·94) 7·71 (3·9)
500 (145) 27·8 (7·5) 125·7 (17·8) 75·8 (10·1) 79·5 (14·8) 4·42 (0·39) 101·7 (27·4) 7·4 (3·1)
Concomitant treatment ACE inhibitors Angiotensin-II-receptor antagonist† Diuretics Digitalis preparations Long-acting nitrates -blockers Calcium-channel blockers
286 (99%) 2 (1%) 231 (80%) 187 (65%) 79 (27%) 83 (29%) 6 (2%)
280 (99%) 2 (1%) 231 (81%) 197 (69%) 90 (32%) 87 (31%) 12 (4%)
†Protocol violation.
716 patients were enrolled over 10 months at 113 study sites in the USA and Canada (figure 2). The first patient was enrolled on Sept 29, 1997, and the last completed the final assessment on Feb 1, 1999. The two treatment groups were similar for baseline and demographic variables (table 1). ACE inhibitors were required as prestudy treatment, and the doses used in the omapatrilat and the lisinopril group was similar (16 mg [12] vs 16 mg [11]). Angiotensin-IIreceptor antagonists had been inadvertently used in four patients (two in each group). Other concomitant treatment was similar in the two study groups; diuretics in 80%, digoxin in 67%, and -blockers in 30% of the study population. At the end of the 3-week titration period, the 716 patients enrolled 143 ineligible or withdrew 573 randomised
289 assigned omapatrilat
284 assigned lisinopril
24 withdrew 21 had adverse events 3 other reasons 7 died
258 completed treatment
24 withdrew 16 had adverse events 8 other reasons 10 died
250 completed treatment
Table 1: Demography and baseline conditions
proportion of patients receiving 40 mg omapatrilat was 88%, 20 mg was 8%, and 10 mg was 4%; and 20 mg lisinopril was 94%, 10 mg was 5%, and 5 mg was 1%. The proportions remained similar throughout the study. Exercise duration At 12 weeks, there were similar and small increments in exercise duration in the two treatment groups. Adjusted mean change from baseline was 24 s (SE 6) for the omapatrilat group (n=274) and 31 s (6) for the lisinopril group (265, p=0·45). The week-24 adjusted change from baseline was 40 s for patients given omapatrilat and 48 s for those receiving lisinopril (p=0·5). Safety and tolerability The two treatments were well tolerated. Serious adverse events are summarised in table 2. There were fewer patients in the omapatrilat group than in the lisinopril group with at least one of these events. Most serious adverse events arose in the cardiovascular system, and were less frequent in the omapatrilat group than the lisinopril group (20 [7%] vs 34 Adverse event
Omapatrilat (n=289)
Lisinopril (n=284)
p
Cardiovascular Heart failure Ischaemic Arrhythmic Others Nervous system Respiratory Gastrointestinal Others Total events Total patients with at least one event
20 (7%) 2 (10%) 7 (35%) 13 (4%) 5 (2%) 6 (2%) 6 (2%) 5 (2%) 21 (7%) 72 (25%) 44 (15%)
34 (12%) 9 (26%) 12 (4%) 14 (5%) 6 (2%) 8 (3%) 9 (3%) 2 (0·7%) 25 (9%) 93 (33%) 60 (21%)
0·04 NS NS NS NS NS NS NS NS NS NS
NS=non-significant.
Figure 2: Trial profile
THE LANCET • Vol 356 • August 19, 2000
Table 2: Serious adverse events
617
For personal use only. Not to be reproduced without permission of The Lancet.
ARTICLES Variable
Omapatrilat
Lisinopril
p
Blood urea nitrogen Creatinine Potassium Asparate aminotransferase Alanine aminotransferase
7 (2·5%) 5 (1·8%) 6 (2·1%) 2 (0·7%) 3 (1·1%)
19 (6·8%) 17 (6·1%) 10 (3·6%) 2 (0·7%) 2 (0·7%)
0·016 0·009 NS NS NS
NS=non-significant.
Table 3: Incidence of significantly elevated prespecified plasma analyses
[12%], p=0·04). Other serious adverse events arose with a similar frequency in both groups. Plasma concentrations of creatinine and blood urea nitrogen (as prespecified) were higher in patients given lisinopril than in those given omapatrilat, but this difference was not significant (table 3). Serum potassium and liver enzyme concentrations did not differ significantly between groups. Table 4 shows the treatment-emergent adverse events. Dizziness (p=0·0001) and vision disturbance (p=0·039) were more frequent in patients given omapatrilat than lisinopril. These events were generally mild, of short duration, and not dose-related—discontinuation of treatment because of these events was infrequent. Patients with systolic blood pressure of less than 120 mm Hg and on omapatrilat seemed at higher risk of hypotension, perhaps because of a more acute effect of omapatrilat on systolic blood pressure. Despite this finding, syncope as a serious adverse event (omapatrilat 0·3% vs lisinopril 1·8%) or an adverse event (1·7% vs 3·2%) occurred more frequently with lisinopril. Diarrhoea was more frequent in patients given omapatrilat (p=0·007), and patients older than 65 years seemed predisposed to this event. The frequency of cough was similar in the two treatment groups, despite the patients being given ACE-inhibitors before randomisation. Tracheobronchitis was seen more frequently with omapatrilat (p=0·038). One patient receiving lisinopril had angioedema compared with none on omapatrilat. Clinical outcome All the individual clinical endpoints related to survival or any comorbid event for worsening heart failure favoured omapatrilat (table 5). There were slightly fewer deaths in Adverse event
Omapatrilat (n=289)
Lisinopril (n=284)
Dizziness Oedema Fatigue Dyspnoea Heart failure Diarrhoea Upper-respiratory infection Musculoskeletal pain Chest pain Cough Hypotension Nausea/vomiting Weight gain Headache Abnormal chest sound Extra heart sound Angina pectoris Influenza Vision disturbance Cardiac murmur Disturbance rhythm ventricular Tracheobronchitis Abdominal pain Weakness Constipation Dyspepsia/heartburn
94 (33%) 60 (21%) 59 (20%) 46 (16%) 38 (13%) 34 (12%) 34 (12%) 32 (11%) 31 (11%) 30 (10%) 29 (10%) 29 (10%) 29 (10%) 28 (10%) 27 (9%) 23 (8%) 21 (7%) 20 (7%) 18 (6%) 17 (6%) 16 (6%) 15 (5%) 13 (5%) 13 (5%) 12 (4%) 12 (4%)
52 (18%) 54 (19%) 53 (19%) 51 (18%) 59 (21%)* 15 (5%)* 35 (12%) 26 (9%) 40 (14%) 31 (11%) 17 (6%) 17 (6%) 31 (11%) 25 (9%) 30 (11%) 31 (11%) 21 (7%) 18 (6%) 7 (3%)* 28 (10%) 9 (3%) 5 (2%)* 11 (4%) 7 (3%) 6 (2%) 6 (2%)
*p<0·05.
Table 4: Most commonly arising treatment-emergent adverse events
618
Death from any cause Admission Study-drug withdrawal Emergency room* visit Supplemental diuretic use Composite: death or admission Composite: death, admission or study treatment discontinued Composite, any event
Omapatrilat Lisinopril Hazard ratio (n=289) (n=284) (95% CI)
p
7 (2%) 8 (3%) 4 (1%) 4 (1%) 45 (16%) 14 (5%) 16 (6%)
10 (4%) 18 (6%) 6 (2%) 8 (3%) 53 (19%) 25 (9%) 29 (10%)
NS NS NS NS NS 0·052 0·035
54 (19%)
68 (24%) 0·75 (0·52–1·07) 0·113
·· ·· ·· ·· ·· 0·52 (0·27–1·02) 0·52 (0·28–0·96)
*Accident and emergency department visits not resulting in admission. NS=nonsignificant.
Table 5: Death and comorbid events of worsening heart failure
patients give lisinopril than in those given omapatrilat (seven vs ten). Less than half the admissions for heart failure, excluding emergency-room visits that did not result in admission, arose in patients on omapatrilat than in those on lisinopril (eight [3%] vs 18 [6%]), The two treatment groups did not differ and any comorbid event of worsening congestive heart failure. The Kaplan-Meier curve for the clinically important composite of first occurrence of death or admission for worsening heart failure separated early, favouring omapatrilat (hazard ratio 0·52 [95% CI 0·27–1·02], p=0·052; figure 3). The cumulative event curves for death, admission, or discontinuation of study treatment were, however, significantly different between groups with the curves separating between days 30 and 60 and continuing to diverge throughout the 24-week study (0·52 [0·28–0·96], p=0·035; figure 3). NYHA functional class and other assessments of heartfailure status At week 24 (or at the observation of patients who did not complete the study) omapatrilat treatment had led to more improvements and less worsening of NYHA class than lisinopril treatment (p=0·059). In patients with NYHA functional class III or IV at baseline, this difference was significant (figure 4; p=0·035). Physicians’ and patients’ global assessments of heart-failure status at the end of study showed that between 64% and 67% of all participants had some degree of improvement and 3–4% had worsened. No consistent treatment benefit could be distinguished with these assessments. Because an overall objective of the treatment of patients with congestive heart failure is to prevent clinical events and improve sense of wellbeing, we assessed a composite endpoint that includes both. For this endpoint, the patient’s condition was deemed to have worsened if they died, were admitted to hospital, or they stopped study treatment for worsening congestive heart failure, increase in NYHA class, or a worsening global assessment. Improvement was judged by decrease in NYHA class or a moderate or striking improvement in global assessment. Omapatrilat had a more beneficial effect on this composite endpoint than lisinopril (p=0·003); 130 (45%) patients improved, 125 (43%) remained unchanged, and 34 (12%) worsened compared with 97 (34%), 136 (48%) and 51 (18%). Radionuclide ventriculography Ventricular volumes and ejection fractions generally remained stable and did not differ between treatment groups. The mean change in left-ventricular end-diastolic volume, which was the primary variable, was less than 1 mL/m2 in each group. Left-ventricular ejection fraction increased about one unit in each group. Left ventricular and right ventricular volumes and ejection fractions differed little between the two groups.
THE LANCET • Vol 356 • August 19, 2000
For personal use only. Not to be reproduced without permission of The Lancet.
ARTICLES
Death, admission for heart failure, or discontinuation of treatment
50
0·15
40 Patients (%)
Lisinopril Omapatrilat 0·10
0·05
30
p < 0·04
20 10 0
Proportion with event
0 0 Number at risk Lisinopril 284 Omapatril 289
30
60 90 120 150 Time from randomisation (days)
274 281
267 280
255 276
247 271
242 265
180 35 34
Death or admission for heart failure 0·15 Lisinopril Omapatrilat 0·10
0·05
0 0 Number at risk Lisinopril 284 Omapatril 289
30 275 283
60 90 120 150 Time from randomisation (days) 268 282
256 278
248 273
243 267
180 35 34
Figure 3: Death and congestive heart failure comorbidity Top: Kaplan-Meier estimates of the combined endpoints of death or admission for heart failure. Bottom: death or admission for heart failure or discontinuation of treatment for worsening heart failure.
Plasma neurohormones The effects of the two drugs on neurohumoral activation differed only slightly (table 6). Atrial natriuretic peptide differed the most, with a decrease over time with lisinopril, but an increase over time with omapatrilat. Plasma norepinephrine increased over time with lisinopril but not with omapatrilat and was significantly different between groups at 12 weeks, but only when data from natural log transformation were compared. Plasma angiotensin II concentrations were similar in the two groups but tended to decrease with lisinopril and to increase with omapatrilat. The effect of both drugs on plasma endothelin-1 was similar.
Discussion In this study of patients with stable heart failure, the vasopeptidase inhibitor omapatrilat did not improve exercise tolerance compared with the ACE inhibitor lisinopril. Exercise tolerance in patients with congestive heart failure is the result of the convergence of numerous factors, some peripheral and some cardiac.13 Interventions that improve exercise tolerance are not necessarily associated with improved survival. Interventions that improve survival, such as -blockers and ACE inhibitors, do not improve exercise tolerance, or increase tolerance less than do interventions such as pure vasodilators or positive inotropic agents, which have less-beneficial or harmful effects on survival.14–17 NYHA functional class is decided by many combined factors, and, therefore, the individual reasons for the greater improvement with omapatrilat in
THE LANCET • Vol 356 • August 19, 2000
Improved Worsened Improved Worsened Omapatrilat Lisinopril Figure 4: Change in NYHA functional class Change in NYHA functional class in patients with NYHA class III or IV heart failure from baseline to final visit.
patients with NYHA class III and IV congestive heart failure are difficult to isolate. Nevertheless mechanisms are probably related to the improved ability of omapatrilat to reduce preload and afterload,9 and its renal effects, which would together be expected to reduce pulmonary congestion and improve cardiac output more than ACE inhibitors. Because the reduction in the need for supplemental diuretics with omapatrilat was small and almost entirely confined to patients receiving -blockers, a diuretic effect seems unlikely to be the main mechanism by which omapatrilat influenced morbid events. Although all individual clinical endpoints related to survival or any comorbid event for worsening heart failure were improved with omapatrilat, the number of patients studied was small and the length of follow-up short, making our results encouraging but not definitive. This statement is particularly true since multiple comparisons, as we did, lower the strength of borderline significant findings. Nevertheless, our results are consistent with studies in cardiomyopathic hamsters that show longer survival with omapatrilat than with captopril,18 and studies in the dog pacing-overdrive heart-failure model, in which omapatrilat improved cardiac and renal function.19,20 Our study supports the idea that increase of endogenous vasodilator and diuretic substances in patients with congestive heart failure could be beneficial and constitutes a potentially important treatment goal for further improving the outlook and wellbeing of patients with the disorder. Indeed, natriuretic peptides, nitric oxide, and prostaglandins have cardiovascular and renal effects that are in direct opposition to those of vasoconstrictor Neurohormone
Lisinopril
Omapatrilat
Norepinephrine (g/L) Baseline 12 weeks 24 weeks
0·368 (0·049) 0·477 (0·094)* 0·521 (0·115)*
0·296 (0·02) 0·336 (0·027)* 0·352 (0·028)
Atrial natriuretic peptides (g/L) Baseline 12 weeks 24 weeks
0·039 (0·004) 0·032 (0·002)* 0·029 (0·003)
0·037 (0·004) 0·039 (0·004)* 0·043 (0·006)†‡
Angiotensin II (g/L) Baseline 12 weeks 24 weeks
0·022 (0·002) 0·017 (0·002)* 0·019 (0·002)
0·024 (0·003) 0·025 (0·004) 0·027 (0·004)
Endothelin (g/L) Baseline 12 weeks 24 weeks
0·002 (0·0002) 0·003 (0·0003) 0·003 (0·0002)
0·002 (0·0002) 0·003 (0·0003) 0·003 (0·0003)
*p<0·05 from baseline. †p<0·05 vs lisinopril using actual norminative values and ‡ values from natural log transformation (p<0·05 for norepinephrine and p=0·053 for atrial natriuetic peptide).
Table 6: Changes in neurohormones over time (mean [SE])
619
For personal use only. Not to be reproduced without permission of The Lancet.
ARTICLES
neurohormones and would thus be expected to be cardioprotective and renoprotective by attenuating the effects of chronic overactivation of vasoconstrictor neurohormones.4–7 In a subset of 19 patients, pulse-wave velocity was lowered more with omapatrilat than with lisinopril, suggesting reduced conduit-vessel stiffness as a potential difference contributing to clinical outcome.21 The improvement associated with the use of omapatrilat did not seem the result of changes in ventricular function that could be identified by radionuclide ventriculography. There were, however, some differences in circulating neurohormones that could help to explain its beneficial effects. Although these changes were of marginal significance, because measurements were taken at the trough of dosing, overall neurohumoral effects of omapatrilat might have been underestimated. Omapatrilat reduced plasma norepinephrine and raised plasma atrial natriuretic peptides more than lisinopril. Since natriuretic peptides reduce adrenergic tone,22 the effects of omapatrilat on natriuretic peptides could lead to the decrease in plasma norepinephrine. Surprisingly, plasma angiotensin II concentrations tended to be higher in patients given omapatrilat, suggesting that greater ACE inhibition with omapatrilat did not lead to the differences in outcome found in this study. Most patients given omapatrilat or lisinopril tolerated the treatment well and titration to the target dose was possible by 3 weeks. Patients given omapatrilat and lisinopril had a similar incidence of cough, which is consistent with previous clinical studies and compatible with an overwhelming role of ACE compared with neuropeptidase in the metabolism of bradykinin in the lung. Indeed, the lung is by far the organ with the greatest concentration of ACE. The finding of fewer patients given omapatrilat with more renal deterioration than those given lisinopril, is compatible with a protective effect of natriuretic peptides on glomerular filtration rate.7,8,20 Our study suggests that the vasopeptidase inhibitor omapatrilat is well tolerated and could have some advantages over the ACE inhibitor lisinopril in the treatment of patients with congestive heart failure. We are undertaking a large-scale phase III study (4420 patients) to compare the effects of omapatrilat and enalapril in patients with heart failure (OVERTURE trial). Acknowledgments Principal Investigator—J L Rouleau. Members of the IMPRESS study group—USA—George F Leatherman, M Tonkon, M Rotman, T Hack, E Kasper, H Hartley, M Klein, J Smith, L Baruch, T LeJemtel, R Gianfagna, M Wilson, B Iteld, J Glode, D Kereiakes, P Phillips, M Dibner-Dunlap, Karl T Weber, R Graf, C East, C Yancy Jr, Francis Menapace, K Klancke, N Vijay, D Hager, A Niederman, C Pepine, T Parker, R Siegel, W J Henry III, Gary V Heller, P Chang, M Lambert, G Torre, P Kirlin, K Adams, C Porter, R Palac, J Plehn, U Elkayam, K Nademanee, N Bittar, J Farnham, K Kerut, L Conway, G Lamas, R Steingart, J Kostis, W Smith, M Klapholz, S Chrysant, U Thadani, J Walker, G Garibian, D Wood, J Felicetta, K Smith, D Forman, S Mehta, G Timmis, R Bellinger, M Drehobl, B Greenberg, L Yellen, R Wright, D Stagaman, M Vicario, S Goldman, W Leimbach, S Ellaham, J Singh, L Petrovich. Canada—R Leader, D Isaac, P T S Ma, R Roux, R S McKelvie, S Kouz, J O Parker, M F Matangi, M J Arnold, J Lenis, D Gossard, D C Phaneuf, F Sestier, J L Rouleau, D Rupka, A Hess, J Imrie, R Bhargava, S J Smith, P Talbot Jr, V A Gebhardt, J Lopez, R Orchard, M B Khouri, T Davies, J Bedard, S Lepage, G Proulx, Y Pesant, B A Sussex, C Fortin, G Moe, J D Parker, V Bernstein, J F Bonet, T E Cuddy, A Morris. Coordinating committee—M Pfeffer (US-chair), D Stewart (Canada-chair); F Sestier, D Isaac, S Smith (Canadian regional coordinators), M Dibner, J Kostis, C Yancy, R Wright (US regional coordinators), J L Rouleau (norepirephrine). Neurohormone analytical centres—C Hall (ANP/BNP), P Sirois (angiotensin), P Cernacek (endothelin-1), N Poitras (central coordinator), J L Rouleau (norepinephrine).
620
Core radionuclide ventriculography laboratory—New England Medical Centre M Konstan, J Udelson, D Kinan (technician). Coordinating Centre—Bristol-Myers Squibb—A Block, P Chew, L Harvey (Study Directors), D Aarons, D Costagliola, G Cucinotta, D Johnson (clinical scientists), W Cooper, C Qian (statisticians), G Samuel (data coordination).
Contributors C Qian was the statistical consultant. All other investigators wrote and designed the study and helped to interpret and write the paper.
References 1 2
3 4
5 6
7 8 9
10
11
12
13 14
15
16
17
18
19
20
21
22
Rouleau JL. The neurohumoral hypothesis and the treatment of heart failure. Can J Cardiol 1996; 12 (suppl F): 3–8. The SOLVD Investigators. Effect of enalapril on survival in patients with reduced left ventricular ejection fractions and congestive heart failure. N Engl J Med 1991; 325: 293–302. The Cardiac Insufficiency Bisoprolol Study II (CIBIS-II): a randomised trial. Lancet 1999; 353: 9–13. Gohlke P, Linz W, Schölkens BA, et al. Angiotensin-converting enzyme inhibition improves cardiac function. Hypertension 1994; 23: 411–18. Remme WJ. Bradykinin-mediated cardiovascular protective actions of ACE inhibitors. Drugs 1997; 54 (suppl 5): 59–70. Naruse M, Takeyama Y, Tanabe A, et al. Atrial and brain natriuretic peptides in cardiovascular diseases. Hypertension 1994; 23 (suppl I): 231–34. Levin ER, Gardner DG, Samson WK. Natriuretic peptides. N Engl J Med 1998; 339: 321–28. Burnett JC Jr. Vasopeptidase inhibition: a new concept in blood pressure management. J Hypertension 1999; 17 (suppl 1): S38–43. Ikram H, McClean DR, Mehta S, et al. Long-term beneficial hemodynamic and neurohormonal effects of vasopeptidase inhibition with omapatrilat in heart failure. J Am Coll Cardiol 1999; 32: 2 (suppl A): 185A. Konstam MA, Kronenberg MW, Rousseau MF, et al. Effects of the angiotensin converting enzyme inhibitor enalapril on the long-term progression of left ventricular dilatation in patients with asymptomatic systolic dysfunction. SOLVD (Studies of Left Ventricular Dysfunction) Investigators. Circulation 1993; 88: 2277–83. Rouleau JL, de Champlain J, Klein M, et al. Activation of neurohumoral systems in postinfarction left ventricular dysfunction. J Am Coll Cardiol 1993; 22: 390–98. Benedict CR, Johnstone DE, Weiner DH, et al. Relation of neurohumoral activation to clinicial variables and degree of ventricular dysfunction: a report from registry of SOLVD. J Am Coll Cardiol 1994; 23: 1410–20. Dickstein K. Negative exercise trials in heart failure: result or design? Eur Heart J 1998; 19: 1747–49. Packer M, Bristow WR, Cohn JN, et al. The effect of carvedilol on morbidity and mortality in patients with chronic heart failure: US carvedilol health failure study group. N Engl J Med 1996; 334: 1349–55. Cohn J, Johnson G, Ziesche S, et al. A comparison of enalapril with hydralazine-isosorbide dinitrate in the treatment of chronic congestive heart failure. N Engl J Med 1991; 325: 303–10. DiBianco R, Shabetai R, Kostuk W, et al. A comparison of oral milrinone, digoxin, and their combination in the treatment of patients with chronic heart failure. N Engl J Med 1989; 320: 677–83. Packer M, Carver JR, Rodeheffer RJ, et al. Effect of oral milrinone on mortality in severe chronic heart failure. N Engl J Med 1991; 325: 1468–75. Trippodo NC, Fox M, Monticello TM, Panchal BC, Asaad MM. Vasopeptidase inhibition with omapatrilat improves cardiac geometry and survival in cardiomyopathic hamsters more than does ACE inhibition with captopril. J Cardiovasc Pharmacol 1999; 34: 782–90. Thomas CV, McDaniel GM, Holzgrefe HH, et al. Chronic dual inhibition of angiotensin-converting enzyme and neutral endopeptidase during the development of left ventricular dysfunction in dogs. J Cardiovasc Pharmacol 1998; 32: 902–12. Cavero PG, Margulies KB, Winaver J, Seymour AA, Delaney NG, Brunett JC Jr. Cardiorenal actions of neutral endopeptidase inhibition in experimental congestive heart failure. Circulation 1990; 82: 196–201. Mitchell GF, Block AJ, Hartley HL, Tardif JC, Rouleau JL. The vasopeptidase inhibitor, omapatrilat, has a favorable, pressureindependent effect on conduit vessel stiffness in patients with congestive heart failure. Circulation 1999; 100 (suppl 1): I–646. Abramson BL, Ando S, Notarius CF, Rongen GA, Floras JS. Effect of atrial natriuretic peptide on muscle sympathetic activity and its reflex control in human heart failure. Circulaton 1999; 99: 1810–15.
THE LANCET • Vol 356 • August 19, 2000
For personal use only. Not to be reproduced without permission of The Lancet.
Articles
Comparison of vasopeptidase inhibitor, omapatrilat, and lisinopril on exercise tolerance and morbidity in patients with heart failure: IMPRESS randomised trial Jean L Rouleau, Marc A Pfeffer, Duncan J Stewart, Debra Isaac, Francois Sestier, Edmund K Kerut, Charles B Porter, Guy Proulx, Chunlin Qian, Alan J Block, for the IMPRESS investigators
Summary Background We aimed to assess in patients with congestive heart failure whether dual inhibition of neutral endopeptidase and angiotensin-converting enzyme (ACE) with the vasopeptidase inhibitor omapatrilat is better than ACE inhibition alone with lisinopril on functional capacity and clinical outcome. Methods We did a prospective, randomised, double-blind, parallel trial of 573 patients with New York Heart Association (NYHA) class II–IV congestive heart failure, left-ventricular ejection fraction of 40% or less, and receiving an ACE inhibitor. Patients were randomly assigned omapatrilat at a daily target dose of 40 mg (n=289) or lisinopril at a daily target dose of 20 mg (n=284) for 24 weeks. The primary endpoint was improvement in maximum exercise treadmill test (ETT) at week 12. Secondary endpoints included death and comorbid events indicative of worsening heart failure. Findings Week 12 ETT increased similarly in the omapatrilat and lisinopril groups (24 vs 31 s, p=0·45). The two drugs were fairly well tolerated, but there were fewer cardiovascularsystem serious adverse events in the omapatrilat group than in the lisinopril group (20 [7%] vs 34 [12%], p=0·04). There was a suggestive trend in favour of omapatrilat on the combined endpoint of death or admission for worsening heart failure (p=0·052; hazard ratio 0·53 [95% CI 0·27–1·02]) and a significant benefit of omapatrilat in the composite of death, admission, or discontinuation of study treatment for worsening heart failure (p=0·035; 0·52 [0·28–0·96]). Omapatrilat improved NYHA class more than lisinopril in patients who had NYHA class III and IV (p=0·035), but not if patients with NYHA class II were included. Interpretation Our findings suggest that omapatrilat could have some advantages over lisinopril in the treatment of patients with congestive heart failure. Thus use of vasopeptidase inhibitors could constitute a potentially Division of Cardiology, Toronto General Hospital, University of Toronto, Eaton North 13-312, Toronto M5G 2C4, ON, Canada (J L Rouleau MD); Division of Cardiology, Brigham and Women Hospital, Havard Medical School, Boston, MA, USA (M A Pfeffer MD); Division of Cardiology, St Michael’s Hospital, University of Toronto (D Stewart MD); Department of Cardiology, Foothills General Hospital, University of Calgary, Calgary (D Issac MD); Department of Cardiology, Centre Hospitalier de l’Université de Montréal-Pavillon Notre-Dame, Montreal (F Sestier MD); Heart Clinic of Louisiana, Marrero LA, USA (E K Kerut MD); Mid-America Cardiology Associates, Kansas City, MO, USA (C B Porter MD); Division of Cardiology, Montreal Heart Institute, University of Montreal (G Proulx MD); and Pharmaceutical Research Institute, Bristol-Myers Squibb, Princeton, NJ, USA (C Qian PhD, A J Block PhD) Correspondence to: Jean L Rouleau (e-mail: [email protected])
THE LANCET • Vol 356 • August 19, 2000
important treatment for further improving the prognosis and well being of patients with this disorder. Lancet 2000; 356: 615–20
Introduction Congestive heart failure is characterised by chronic overactivation of sodium-retaining and water-retaining neurohormones.1 Overactivation of these vasoconstrictor neurohumoral systems also leads to excessive cellular growth, cardiac fibrosis, and cellular toxicity.1 Attenuating the effects of these neurohumoral systems has been one of the most successful strategies in lowering the morbidity and mortality of patients with congestive heart failure.2,3 Although the use of angiotensin-converting-enzyme (ACE) inhibitors and -blockers has been beneficial in this respect, the outlook of these patients remains poor. Therefore, new strategies to improve outlook need to be developed.2,3 Human beings have developed several endogenous systems for countering the effects of overactivation of vasoconstrictor neurohormones.1 One mechanism is endogenous vasodilator systems, which include natriuretic peptides, nitric oxide, and prostaglandins.1 These peptides vasodilate and promote diuresis and natriuresis and lessen cellular growth.4–7 In congestive heart failure, endogenous vasodilator systems are activated to try to compensate for chronic activation of vasoconstrictor sodium and waterretaining neurohormones. The vasopeptidase inhibitors, a new class of pharmaceutical agents, have been shown to heighten activity of endogenous vasodilator systems and reduce production of the vasoconstrictor angiotensin II.8 Vasopeptidase inhibitors inhibit the activity of neutral endopeptidase, an enzyme that metabolises endogenous vasodilator peptides, such as natriuretic peptides (atrial, brain, and calcium-activated neutral protease), adrenomedullin, and bradykinin.8 Because these inhibitors better redress the imbalance between endogenous vasoconstrictor and vasodilator substances in congestive heart failure than does ACE inhibition alone, they could be more useful in treatment of patients with the disorder. We did a randomised double-blind trial to compare the effects of omapatrilat with those of lisinopril on exercise tolerance in patients with congestive heart failure. We also assessed side-effects of the treatments, effects on death rate, and comorbid events for worsening heart failure.
Methods Omapatrilat Omapatrilat (BMS-186716) is the first of a new class of cardiovascular agents termed vasopeptidase inhibitors. Omapatrilat is an orally active, long acting, selective, competitive inhibitor of neutral endopeptidase (NEP; enkephalinase, neprilysin, EC 3.4.24.11) and angiotensin converting enzyme (ACE; EC 3.4.15.1) with similar K1
615
For personal use only. Not to be reproduced without permission of The Lancet.
ARTICLES
H
S O HS N H
N O
COOH
Figure 1: Chemical structure of omapatrilat Ki values, neutral endopeptidase 8·9 nM, ACE 6·0 nM.
values for both NEP and ACE (figure 1). As a result, omapatrilat potentiates multiple endogenous vasodilatory peptides including natriuretic peptides (atrial, brain, and calcium activated neutral protease), bradykinin, and adrenomedullin, while also inhibiting the generation of the vasoconstrictive peptide, angiotensin II. The chemical name is [4S-{4␣(R*),7␣10a]}-octahydro-4-[2-mercapto1-oxo-3-phenypropyl)amino]-5-oxo=7H-pyrido[2,1b][1,3]-thiazepine-7-carboxylic acid; its structure is shown in figure 1. Omapatrilat has an oral bioavailability of about 30%, it may be given with or without food, its protein binding is 80%, and it has a very large volume of distribution (1800 L) suggesting tissue penetration. Time to maximum concentration of an oral dose is about 2 h. Its effective half-life is 14–19 h. This drug is metabolised in the liver. Omapatrilat forms disulphide bonds with endogenous thiols, and is extensively metabolised via S-methylation, amide hydrolysis, S-oxidation, and glucuronidation. There are no substantial concentrations of active metabolites of omapatrilat in plasma. About 80% of an intravenous radioactive dose and 64% of an oral radioactive dose were recovered in urine, with less than 1% excreted as unchanged drug (data available). Based on biotransformation data in vitro and in vivo, cytochrome P450 enzymes do not seem to be involved in the metabolism of omapatrilat. In-vitro studies show that omapatrilat does not inhibit P450 isoenzmes CYP-1A2, CYP-3A4, CYP-2C9, CYP-2C19, and CYP-2D6. Omapatrilat has no known interaction with other medications. Study population We did the study in 113 centres in the USA and Canada. The study was approved by central or local-institution review boards. Patients gave written informed consent. Patients were aged at least 18 years, with stable (>3 months) symptomatic heart failure (NYHA II–IV), decreased left-ventricular ejection fraction, and receiving a stable (⭓4 weeks) dose of ACE inhibitors. Patients were eligible if they had a seated systolic blood pressure of 90 mm Hg or higher at randomisation, but no uncontrolled hypertension or history of acute coronary events or revascularisation procedures within 3 months. Patients with serum potassium of less than 3·5 or more than 5·3 mmol/L or creatinine of more than 221 mol/L were excluded as were patients with transaminases more than twice the upper limit of normal, leucocytes less than 3·0⫻109/L, neutrophils of less than 1·5⫻109/L, or platelets of less than 120⫻109/L. -blockers were allowed only if they had been used for at least 6 months, calcium-channel blockers were allowed only for control of atrial fibrillation, and patients who had been
616
included in previous randomised trials of omapatrilat were excluded. Angiotensin-receptor antagonists, vasodilators other than nitrates, and inotropes other than digoxin were prohibited during the study. Finally, inability to give informed consent, or any other medical disorder judged by the investigator to limit optimum participation in the study, were exclusion criteria. Randomisation and study treatment Maximum-exertion-exercise tolerance tests (ETT) were undertaken with a modified Naughton protocol 21–27 h after administration of study and concomitant cardiovascular drugs. ETTs had to be limited by either dyspnoea or fatigue. After initial qualification and a screening ETT of less than 12 min, patients entered a placebo lead-in period, in which they did a qualification ETT that had to be 2–12 min There were no time constraints on the baseline ETT, and it was done within 9 days of the qualifying study. Patients who met all entrance criteria stopped ACE-inhibitor treatment and were randomised the next day to active treatment with 10 mg omapatrilat titrated to the target dose of 40 mg once daily, or 5 mg lisinopril once daily, titrated to 20 mg daily. Forced titration to target or maximum tolerated dose was used at weekly intervals for up to 3 weeks. We based the dose of omapatrilat on a previous long-term haemodynamic study that showed significant reduction of pulmonary capillarywedge pressure with 40 mg omapatrilat.9 Administration of concomitant cardiovascular treatment was required on the mornings of titration visits, but prohibited on the mornings of other visits, including those to assess exercise time. Assessment of patients and endpoints Clinical assessments were done weekly until week 3, and then at weeks 6, 12, 18, and 24. Full laboratory assessments were done at baseline, and at weeks 12 and 24, and abbreviated laboratory panels were obtained at all other visits after randomisation. The primary endpoint was change in exercise duration from baseline to week 12. Secondary endpoints were: the combined endpoint of death and admission for worsening heart failure, and the combined endpoint of death and comorbidity for worsening heart failure (admission, discontinuation of study treatment, emergency-room visit of clinical need for supplemental diuretic). Additional endpoints included NYHA functional class, physician and patient global assessment of heart-failure status at 24 weeks. Safety assessment included occurrence of treatmentemergent adverse events, as well as clinical laboratory assessment of raised concentrations of serum creatinine (>1·5 times baseline and >0·8 times limit of normal), blood urea nitrogen (>2 times baseline and higher than upper limit of normal), potassium, and liver-function tests. A serious adverse event was defined as any of: fatal or life threatening, permanently disabling prolonging time spent in hospital, an important medical event, congenital anomaly, cancer, or overdose. We used equilibrium gated radionuclide ventriculography in a subset of 75 patients to measure left ventricular and right ventricular volumes and ejection fractions at baseline and 24 weeks. Data were analysed by a core laboratory at the New England Medical Centre.10 116 patients participated in the neurohormone substudy. Before the morning dose of the study drugs, a venous cannula was inserted into the patient’s antecubital vein. After 30 min, during which the patients remained recumbent, about 28 mL of blood was withdrawn,
THE LANCET • Vol 356 • August 19, 2000
For personal use only. Not to be reproduced without permission of The Lancet.
ARTICLES
centrifuged at 4ºC, and the plasma stored at ⫺80ºC before analysis. Plasma atrial natriuretic peptide, norepinephrine, angiotensin II, and endothelin 1 were analysed11 by a core laboratory at the Montreal Heart Institute (Quebec, Canada). We measured plasma neurohormones at baseline, and week 12 and 24. Statistical analysis We estimated that 532 participants (266 per group) would be needed to provide 80% power at ␣=0·05 to detect a 30 s difference at week 12 between the two treatment groups (SD 120 s). Patients who stopped taking study drugs were asked to return to do ETTs at weeks 12 and 24, if medically appropriate, so that the primary test for difference could be done by intention to treat. We did a secondary analysis to test for change in exercise performance from baseline to weeks 12 and 24 by ANOVA, with terms for baseline exercise time and treatment. Time-to-event analyses for mortality and comorbidity were based on Cox’s proportional hazard regression with treatment as the only covariant. We assessed treatment difference in NYHA functional class at week 24 with the Cochran-MantelHaenszel method (ANOVA statistic, rank scores) controlling for baseline class (the last observation was used for patients who withdrew). We measured treatment difference in neurohumoral concentrations at 12 weeks and 24 weeks with a two-tailed student t test before and after log transformation, which results in near normal distribution for neurohumoral concentrations.12
Results
Characteristic
Omapatrilat (n=289)
Lisinopril (n=284)
227/62
224/60
Demography Male/female Ethnic origin White Black Other Mean (SD) age (years) NYHA class II III IV
237 26 26 64·3 (10·7)
240 24 20 63·6 (10·0)
186 100 3
175 107 2
Heart failure cause Ischaemic Idiopathic dilated cardiomyopathy Valvular Hypertension Other/not reported
186 (64%) 73 (25%) 10 (3%) 10 (3%) 10 (3%)
191 (67%) 65 (23%) 9 (3%) 12 (4%) 7 (3%)
Measurements (mean [SD]) Baseline ETT (s) Left-ventricular ejection fraction (%) Standing systolic blood pressure (mm Hg) Standing diastolic blood pressure (mm Hg) Standing heart rate (beats/min) Serum potassium (mmol/L) Serum creatinine (mol/L) Blood urea nitrogen (mmol/L urea)
511 (159) 28·4 (7·5) 126·6 (19·7) 76·2 (10·6) 79·6 (12·9) 4·42 (0·39) 103·4 (30·94) 7·71 (3·9)
500 (145) 27·8 (7·5) 125·7 (17·8) 75·8 (10·1) 79·5 (14·8) 4·42 (0·39) 101·7 (27·4) 7·4 (3·1)
Concomitant treatment ACE inhibitors Angiotensin-II-receptor antagonist† Diuretics Digitalis preparations Long-acting nitrates -blockers Calcium-channel blockers
286 (99%) 2 (1%) 231 (80%) 187 (65%) 79 (27%) 83 (29%) 6 (2%)
280 (99%) 2 (1%) 231 (81%) 197 (69%) 90 (32%) 87 (31%) 12 (4%)
†Protocol violation.
716 patients were enrolled over 10 months at 113 study sites in the USA and Canada (figure 2). The first patient was enrolled on Sept 29, 1997, and the last completed the final assessment on Feb 1, 1999. The two treatment groups were similar for baseline and demographic variables (table 1). ACE inhibitors were required as prestudy treatment, and the doses used in the omapatrilat and the lisinopril group was similar (16 mg [12] vs 16 mg [11]). Angiotensin-IIreceptor antagonists had been inadvertently used in four patients (two in each group). Other concomitant treatment was similar in the two study groups; diuretics in 80%, digoxin in 67%, and -blockers in 30% of the study population. At the end of the 3-week titration period, the 716 patients enrolled 143 ineligible or withdrew 573 randomised
289 assigned omapatrilat
284 assigned lisinopril
24 withdrew 21 had adverse events 3 other reasons 7 died
258 completed treatment
24 withdrew 16 had adverse events 8 other reasons 10 died
250 completed treatment
Table 1: Demography and baseline conditions
proportion of patients receiving 40 mg omapatrilat was 88%, 20 mg was 8%, and 10 mg was 4%; and 20 mg lisinopril was 94%, 10 mg was 5%, and 5 mg was 1%. The proportions remained similar throughout the study. Exercise duration At 12 weeks, there were similar and small increments in exercise duration in the two treatment groups. Adjusted mean change from baseline was 24 s (SE 6) for the omapatrilat group (n=274) and 31 s (6) for the lisinopril group (265, p=0·45). The week-24 adjusted change from baseline was 40 s for patients given omapatrilat and 48 s for those receiving lisinopril (p=0·5). Safety and tolerability The two treatments were well tolerated. Serious adverse events are summarised in table 2. There were fewer patients in the omapatrilat group than in the lisinopril group with at least one of these events. Most serious adverse events arose in the cardiovascular system, and were less frequent in the omapatrilat group than the lisinopril group (20 [7%] vs 34 Adverse event
Omapatrilat (n=289)
Lisinopril (n=284)
p
Cardiovascular Heart failure Ischaemic Arrhythmic Others Nervous system Respiratory Gastrointestinal Others Total events Total patients with at least one event
20 (7%) 2 (10%) 7 (35%) 13 (4%) 5 (2%) 6 (2%) 6 (2%) 5 (2%) 21 (7%) 72 (25%) 44 (15%)
34 (12%) 9 (26%) 12 (4%) 14 (5%) 6 (2%) 8 (3%) 9 (3%) 2 (0·7%) 25 (9%) 93 (33%) 60 (21%)
0·04 NS NS NS NS NS NS NS NS NS NS
NS=non-significant.
Figure 2: Trial profile
THE LANCET • Vol 356 • August 19, 2000
Table 2: Serious adverse events
617
For personal use only. Not to be reproduced without permission of The Lancet.
ARTICLES Variable
Omapatrilat
Lisinopril
p
Blood urea nitrogen Creatinine Potassium Asparate aminotransferase Alanine aminotransferase
7 (2·5%) 5 (1·8%) 6 (2·1%) 2 (0·7%) 3 (1·1%)
19 (6·8%) 17 (6·1%) 10 (3·6%) 2 (0·7%) 2 (0·7%)
0·016 0·009 NS NS NS
NS=non-significant.
Table 3: Incidence of significantly elevated prespecified plasma analyses
[12%], p=0·04). Other serious adverse events arose with a similar frequency in both groups. Plasma concentrations of creatinine and blood urea nitrogen (as prespecified) were higher in patients given lisinopril than in those given omapatrilat, but this difference was not significant (table 3). Serum potassium and liver enzyme concentrations did not differ significantly between groups. Table 4 shows the treatment-emergent adverse events. Dizziness (p=0·0001) and vision disturbance (p=0·039) were more frequent in patients given omapatrilat than lisinopril. These events were generally mild, of short duration, and not dose-related—discontinuation of treatment because of these events was infrequent. Patients with systolic blood pressure of less than 120 mm Hg and on omapatrilat seemed at higher risk of hypotension, perhaps because of a more acute effect of omapatrilat on systolic blood pressure. Despite this finding, syncope as a serious adverse event (omapatrilat 0·3% vs lisinopril 1·8%) or an adverse event (1·7% vs 3·2%) occurred more frequently with lisinopril. Diarrhoea was more frequent in patients given omapatrilat (p=0·007), and patients older than 65 years seemed predisposed to this event. The frequency of cough was similar in the two treatment groups, despite the patients being given ACE-inhibitors before randomisation. Tracheobronchitis was seen more frequently with omapatrilat (p=0·038). One patient receiving lisinopril had angioedema compared with none on omapatrilat. Clinical outcome All the individual clinical endpoints related to survival or any comorbid event for worsening heart failure favoured omapatrilat (table 5). There were slightly fewer deaths in Adverse event
Omapatrilat (n=289)
Lisinopril (n=284)
Dizziness Oedema Fatigue Dyspnoea Heart failure Diarrhoea Upper-respiratory infection Musculoskeletal pain Chest pain Cough Hypotension Nausea/vomiting Weight gain Headache Abnormal chest sound Extra heart sound Angina pectoris Influenza Vision disturbance Cardiac murmur Disturbance rhythm ventricular Tracheobronchitis Abdominal pain Weakness Constipation Dyspepsia/heartburn
94 (33%) 60 (21%) 59 (20%) 46 (16%) 38 (13%) 34 (12%) 34 (12%) 32 (11%) 31 (11%) 30 (10%) 29 (10%) 29 (10%) 29 (10%) 28 (10%) 27 (9%) 23 (8%) 21 (7%) 20 (7%) 18 (6%) 17 (6%) 16 (6%) 15 (5%) 13 (5%) 13 (5%) 12 (4%) 12 (4%)
52 (18%) 54 (19%) 53 (19%) 51 (18%) 59 (21%)* 15 (5%)* 35 (12%) 26 (9%) 40 (14%) 31 (11%) 17 (6%) 17 (6%) 31 (11%) 25 (9%) 30 (11%) 31 (11%) 21 (7%) 18 (6%) 7 (3%)* 28 (10%) 9 (3%) 5 (2%)* 11 (4%) 7 (3%) 6 (2%) 6 (2%)
*p<0·05.
Table 4: Most commonly arising treatment-emergent adverse events
618
Death from any cause Admission Study-drug withdrawal Emergency room* visit Supplemental diuretic use Composite: death or admission Composite: death, admission or study treatment discontinued Composite, any event
Omapatrilat Lisinopril Hazard ratio (n=289) (n=284) (95% CI)
p
7 (2%) 8 (3%) 4 (1%) 4 (1%) 45 (16%) 14 (5%) 16 (6%)
10 (4%) 18 (6%) 6 (2%) 8 (3%) 53 (19%) 25 (9%) 29 (10%)
NS NS NS NS NS 0·052 0·035
54 (19%)
68 (24%) 0·75 (0·52–1·07) 0·113
·· ·· ·· ·· ·· 0·52 (0·27–1·02) 0·52 (0·28–0·96)
*Accident and emergency department visits not resulting in admission. NS=nonsignificant.
Table 5: Death and comorbid events of worsening heart failure
patients give lisinopril than in those given omapatrilat (seven vs ten). Less than half the admissions for heart failure, excluding emergency-room visits that did not result in admission, arose in patients on omapatrilat than in those on lisinopril (eight [3%] vs 18 [6%]), The two treatment groups did not differ and any comorbid event of worsening congestive heart failure. The Kaplan-Meier curve for the clinically important composite of first occurrence of death or admission for worsening heart failure separated early, favouring omapatrilat (hazard ratio 0·52 [95% CI 0·27–1·02], p=0·052; figure 3). The cumulative event curves for death, admission, or discontinuation of study treatment were, however, significantly different between groups with the curves separating between days 30 and 60 and continuing to diverge throughout the 24-week study (0·52 [0·28–0·96], p=0·035; figure 3). NYHA functional class and other assessments of heartfailure status At week 24 (or at the observation of patients who did not complete the study) omapatrilat treatment had led to more improvements and less worsening of NYHA class than lisinopril treatment (p=0·059). In patients with NYHA functional class III or IV at baseline, this difference was significant (figure 4; p=0·035). Physicians’ and patients’ global assessments of heart-failure status at the end of study showed that between 64% and 67% of all participants had some degree of improvement and 3–4% had worsened. No consistent treatment benefit could be distinguished with these assessments. Because an overall objective of the treatment of patients with congestive heart failure is to prevent clinical events and improve sense of wellbeing, we assessed a composite endpoint that includes both. For this endpoint, the patient’s condition was deemed to have worsened if they died, were admitted to hospital, or they stopped study treatment for worsening congestive heart failure, increase in NYHA class, or a worsening global assessment. Improvement was judged by decrease in NYHA class or a moderate or striking improvement in global assessment. Omapatrilat had a more beneficial effect on this composite endpoint than lisinopril (p=0·003); 130 (45%) patients improved, 125 (43%) remained unchanged, and 34 (12%) worsened compared with 97 (34%), 136 (48%) and 51 (18%). Radionuclide ventriculography Ventricular volumes and ejection fractions generally remained stable and did not differ between treatment groups. The mean change in left-ventricular end-diastolic volume, which was the primary variable, was less than 1 mL/m2 in each group. Left-ventricular ejection fraction increased about one unit in each group. Left ventricular and right ventricular volumes and ejection fractions differed little between the two groups.
THE LANCET • Vol 356 • August 19, 2000
For personal use only. Not to be reproduced without permission of The Lancet.
ARTICLES
Death, admission for heart failure, or discontinuation of treatment
50
0·15
40 Patients (%)
Lisinopril Omapatrilat 0·10
0·05
30
p < 0·04
20 10 0
Proportion with event
0 0 Number at risk Lisinopril 284 Omapatril 289
30
60 90 120 150 Time from randomisation (days)
274 281
267 280
255 276
247 271
242 265
180 35 34
Death or admission for heart failure 0·15 Lisinopril Omapatrilat 0·10
0·05
0 0 Number at risk Lisinopril 284 Omapatril 289
30 275 283
60 90 120 150 Time from randomisation (days) 268 282
256 278
248 273
243 267
180 35 34
Figure 3: Death and congestive heart failure comorbidity Top: Kaplan-Meier estimates of the combined endpoints of death or admission for heart failure. Bottom: death or admission for heart failure or discontinuation of treatment for worsening heart failure.
Plasma neurohormones The effects of the two drugs on neurohumoral activation differed only slightly (table 6). Atrial natriuretic peptide differed the most, with a decrease over time with lisinopril, but an increase over time with omapatrilat. Plasma norepinephrine increased over time with lisinopril but not with omapatrilat and was significantly different between groups at 12 weeks, but only when data from natural log transformation were compared. Plasma angiotensin II concentrations were similar in the two groups but tended to decrease with lisinopril and to increase with omapatrilat. The effect of both drugs on plasma endothelin-1 was similar.
Discussion In this study of patients with stable heart failure, the vasopeptidase inhibitor omapatrilat did not improve exercise tolerance compared with the ACE inhibitor lisinopril. Exercise tolerance in patients with congestive heart failure is the result of the convergence of numerous factors, some peripheral and some cardiac.13 Interventions that improve exercise tolerance are not necessarily associated with improved survival. Interventions that improve survival, such as -blockers and ACE inhibitors, do not improve exercise tolerance, or increase tolerance less than do interventions such as pure vasodilators or positive inotropic agents, which have less-beneficial or harmful effects on survival.14–17 NYHA functional class is decided by many combined factors, and, therefore, the individual reasons for the greater improvement with omapatrilat in
THE LANCET • Vol 356 • August 19, 2000
Improved Worsened Improved Worsened Omapatrilat Lisinopril Figure 4: Change in NYHA functional class Change in NYHA functional class in patients with NYHA class III or IV heart failure from baseline to final visit.
patients with NYHA class III and IV congestive heart failure are difficult to isolate. Nevertheless mechanisms are probably related to the improved ability of omapatrilat to reduce preload and afterload,9 and its renal effects, which would together be expected to reduce pulmonary congestion and improve cardiac output more than ACE inhibitors. Because the reduction in the need for supplemental diuretics with omapatrilat was small and almost entirely confined to patients receiving -blockers, a diuretic effect seems unlikely to be the main mechanism by which omapatrilat influenced morbid events. Although all individual clinical endpoints related to survival or any comorbid event for worsening heart failure were improved with omapatrilat, the number of patients studied was small and the length of follow-up short, making our results encouraging but not definitive. This statement is particularly true since multiple comparisons, as we did, lower the strength of borderline significant findings. Nevertheless, our results are consistent with studies in cardiomyopathic hamsters that show longer survival with omapatrilat than with captopril,18 and studies in the dog pacing-overdrive heart-failure model, in which omapatrilat improved cardiac and renal function.19,20 Our study supports the idea that increase of endogenous vasodilator and diuretic substances in patients with congestive heart failure could be beneficial and constitutes a potentially important treatment goal for further improving the outlook and wellbeing of patients with the disorder. Indeed, natriuretic peptides, nitric oxide, and prostaglandins have cardiovascular and renal effects that are in direct opposition to those of vasoconstrictor Neurohormone
Lisinopril
Omapatrilat
Norepinephrine (g/L) Baseline 12 weeks 24 weeks
0·368 (0·049) 0·477 (0·094)* 0·521 (0·115)*
0·296 (0·02) 0·336 (0·027)* 0·352 (0·028)
Atrial natriuretic peptides (g/L) Baseline 12 weeks 24 weeks
0·039 (0·004) 0·032 (0·002)* 0·029 (0·003)
0·037 (0·004) 0·039 (0·004)* 0·043 (0·006)†‡
Angiotensin II (g/L) Baseline 12 weeks 24 weeks
0·022 (0·002) 0·017 (0·002)* 0·019 (0·002)
0·024 (0·003) 0·025 (0·004) 0·027 (0·004)
Endothelin (g/L) Baseline 12 weeks 24 weeks
0·002 (0·0002) 0·003 (0·0003) 0·003 (0·0002)
0·002 (0·0002) 0·003 (0·0003) 0·003 (0·0003)
*p<0·05 from baseline. †p<0·05 vs lisinopril using actual norminative values and ‡ values from natural log transformation (p<0·05 for norepinephrine and p=0·053 for atrial natriuetic peptide).
Table 6: Changes in neurohormones over time (mean [SE])
619
For personal use only. Not to be reproduced without permission of The Lancet.
ARTICLES
neurohormones and would thus be expected to be cardioprotective and renoprotective by attenuating the effects of chronic overactivation of vasoconstrictor neurohormones.4–7 In a subset of 19 patients, pulse-wave velocity was lowered more with omapatrilat than with lisinopril, suggesting reduced conduit-vessel stiffness as a potential difference contributing to clinical outcome.21 The improvement associated with the use of omapatrilat did not seem the result of changes in ventricular function that could be identified by radionuclide ventriculography. There were, however, some differences in circulating neurohormones that could help to explain its beneficial effects. Although these changes were of marginal significance, because measurements were taken at the trough of dosing, overall neurohumoral effects of omapatrilat might have been underestimated. Omapatrilat reduced plasma norepinephrine and raised plasma atrial natriuretic peptides more than lisinopril. Since natriuretic peptides reduce adrenergic tone,22 the effects of omapatrilat on natriuretic peptides could lead to the decrease in plasma norepinephrine. Surprisingly, plasma angiotensin II concentrations tended to be higher in patients given omapatrilat, suggesting that greater ACE inhibition with omapatrilat did not lead to the differences in outcome found in this study. Most patients given omapatrilat or lisinopril tolerated the treatment well and titration to the target dose was possible by 3 weeks. Patients given omapatrilat and lisinopril had a similar incidence of cough, which is consistent with previous clinical studies and compatible with an overwhelming role of ACE compared with neuropeptidase in the metabolism of bradykinin in the lung. Indeed, the lung is by far the organ with the greatest concentration of ACE. The finding of fewer patients given omapatrilat with more renal deterioration than those given lisinopril, is compatible with a protective effect of natriuretic peptides on glomerular filtration rate.7,8,20 Our study suggests that the vasopeptidase inhibitor omapatrilat is well tolerated and could have some advantages over the ACE inhibitor lisinopril in the treatment of patients with congestive heart failure. We are undertaking a large-scale phase III study (4420 patients) to compare the effects of omapatrilat and enalapril in patients with heart failure (OVERTURE trial). Acknowledgments Principal Investigator—J L Rouleau. Members of the IMPRESS study group—USA—George F Leatherman, M Tonkon, M Rotman, T Hack, E Kasper, H Hartley, M Klein, J Smith, L Baruch, T LeJemtel, R Gianfagna, M Wilson, B Iteld, J Glode, D Kereiakes, P Phillips, M Dibner-Dunlap, Karl T Weber, R Graf, C East, C Yancy Jr, Francis Menapace, K Klancke, N Vijay, D Hager, A Niederman, C Pepine, T Parker, R Siegel, W J Henry III, Gary V Heller, P Chang, M Lambert, G Torre, P Kirlin, K Adams, C Porter, R Palac, J Plehn, U Elkayam, K Nademanee, N Bittar, J Farnham, K Kerut, L Conway, G Lamas, R Steingart, J Kostis, W Smith, M Klapholz, S Chrysant, U Thadani, J Walker, G Garibian, D Wood, J Felicetta, K Smith, D Forman, S Mehta, G Timmis, R Bellinger, M Drehobl, B Greenberg, L Yellen, R Wright, D Stagaman, M Vicario, S Goldman, W Leimbach, S Ellaham, J Singh, L Petrovich. Canada—R Leader, D Isaac, P T S Ma, R Roux, R S McKelvie, S Kouz, J O Parker, M F Matangi, M J Arnold, J Lenis, D Gossard, D C Phaneuf, F Sestier, J L Rouleau, D Rupka, A Hess, J Imrie, R Bhargava, S J Smith, P Talbot Jr, V A Gebhardt, J Lopez, R Orchard, M B Khouri, T Davies, J Bedard, S Lepage, G Proulx, Y Pesant, B A Sussex, C Fortin, G Moe, J D Parker, V Bernstein, J F Bonet, T E Cuddy, A Morris. Coordinating committee—M Pfeffer (US-chair), D Stewart (Canada-chair); F Sestier, D Isaac, S Smith (Canadian regional coordinators), M Dibner, J Kostis, C Yancy, R Wright (US regional coordinators), J L Rouleau (norepirephrine). Neurohormone analytical centres—C Hall (ANP/BNP), P Sirois (angiotensin), P Cernacek (endothelin-1), N Poitras (central coordinator), J L Rouleau (norepinephrine).
620
Core radionuclide ventriculography laboratory—New England Medical Centre M Konstan, J Udelson, D Kinan (technician). Coordinating Centre—Bristol-Myers Squibb—A Block, P Chew, L Harvey (Study Directors), D Aarons, D Costagliola, G Cucinotta, D Johnson (clinical scientists), W Cooper, C Qian (statisticians), G Samuel (data coordination).
Contributors C Qian was the statistical consultant. All other investigators wrote and designed the study and helped to interpret and write the paper.
References 1 2
3 4
5 6
7 8 9
10
11
12
13 14
15
16
17
18
19
20
21
22
Rouleau JL. The neurohumoral hypothesis and the treatment of heart failure. Can J Cardiol 1996; 12 (suppl F): 3–8. The SOLVD Investigators. Effect of enalapril on survival in patients with reduced left ventricular ejection fractions and congestive heart failure. N Engl J Med 1991; 325: 293–302. The Cardiac Insufficiency Bisoprolol Study II (CIBIS-II): a randomised trial. Lancet 1999; 353: 9–13. Gohlke P, Linz W, Schölkens BA, et al. Angiotensin-converting enzyme inhibition improves cardiac function. Hypertension 1994; 23: 411–18. Remme WJ. Bradykinin-mediated cardiovascular protective actions of ACE inhibitors. Drugs 1997; 54 (suppl 5): 59–70. Naruse M, Takeyama Y, Tanabe A, et al. Atrial and brain natriuretic peptides in cardiovascular diseases. Hypertension 1994; 23 (suppl I): 231–34. Levin ER, Gardner DG, Samson WK. Natriuretic peptides. N Engl J Med 1998; 339: 321–28. Burnett JC Jr. Vasopeptidase inhibition: a new concept in blood pressure management. J Hypertension 1999; 17 (suppl 1): S38–43. Ikram H, McClean DR, Mehta S, et al. Long-term beneficial hemodynamic and neurohormonal effects of vasopeptidase inhibition with omapatrilat in heart failure. J Am Coll Cardiol 1999; 32: 2 (suppl A): 185A. Konstam MA, Kronenberg MW, Rousseau MF, et al. Effects of the angiotensin converting enzyme inhibitor enalapril on the long-term progression of left ventricular dilatation in patients with asymptomatic systolic dysfunction. SOLVD (Studies of Left Ventricular Dysfunction) Investigators. Circulation 1993; 88: 2277–83. Rouleau JL, de Champlain J, Klein M, et al. Activation of neurohumoral systems in postinfarction left ventricular dysfunction. J Am Coll Cardiol 1993; 22: 390–98. Benedict CR, Johnstone DE, Weiner DH, et al. Relation of neurohumoral activation to clinicial variables and degree of ventricular dysfunction: a report from registry of SOLVD. J Am Coll Cardiol 1994; 23: 1410–20. Dickstein K. Negative exercise trials in heart failure: result or design? Eur Heart J 1998; 19: 1747–49. Packer M, Bristow WR, Cohn JN, et al. The effect of carvedilol on morbidity and mortality in patients with chronic heart failure: US carvedilol health failure study group. N Engl J Med 1996; 334: 1349–55. Cohn J, Johnson G, Ziesche S, et al. A comparison of enalapril with hydralazine-isosorbide dinitrate in the treatment of chronic congestive heart failure. N Engl J Med 1991; 325: 303–10. DiBianco R, Shabetai R, Kostuk W, et al. A comparison of oral milrinone, digoxin, and their combination in the treatment of patients with chronic heart failure. N Engl J Med 1989; 320: 677–83. Packer M, Carver JR, Rodeheffer RJ, et al. Effect of oral milrinone on mortality in severe chronic heart failure. N Engl J Med 1991; 325: 1468–75. Trippodo NC, Fox M, Monticello TM, Panchal BC, Asaad MM. Vasopeptidase inhibition with omapatrilat improves cardiac geometry and survival in cardiomyopathic hamsters more than does ACE inhibition with captopril. J Cardiovasc Pharmacol 1999; 34: 782–90. Thomas CV, McDaniel GM, Holzgrefe HH, et al. Chronic dual inhibition of angiotensin-converting enzyme and neutral endopeptidase during the development of left ventricular dysfunction in dogs. J Cardiovasc Pharmacol 1998; 32: 902–12. Cavero PG, Margulies KB, Winaver J, Seymour AA, Delaney NG, Brunett JC Jr. Cardiorenal actions of neutral endopeptidase inhibition in experimental congestive heart failure. Circulation 1990; 82: 196–201. Mitchell GF, Block AJ, Hartley HL, Tardif JC, Rouleau JL. The vasopeptidase inhibitor, omapatrilat, has a favorable, pressureindependent effect on conduit vessel stiffness in patients with congestive heart failure. Circulation 1999; 100 (suppl 1): I–646. Abramson BL, Ando S, Notarius CF, Rongen GA, Floras JS. Effect of atrial natriuretic peptide on muscle sympathetic activity and its reflex control in human heart failure. Circulaton 1999; 99: 1810–15.
THE LANCET • Vol 356 • August 19, 2000
For personal use only. Not to be reproduced without permission of The Lancet.