Statistics from Altmetric.com
Two landmark clinical trials, CONSENSUS I and SOLVD-T, have shown, unequivocally, that angiotensin converting enzyme (ACE) inhibitors reduce all cause mortality in patients with chronic heart failure (CHF) and underlying left ventricular systolic dysfunction.1 2These and other trials have also confirmed that ACE inhibitors reduce morbidity, as manifest by hospital admission, in patients with CHF.3
A number of other key randomised, controlled, trials have also shown that ACE inhibitors reduce the risk of all cause mortality and major clinical events (sudden death, reinfarction, heart failure) after myocardial infarction.w1–3 These benefits are most clearly seen in patients with left ventricular systolic dysfunction or clinical evidence of heart failure.w4
The ATLAS study has shown that higher doses of ACE inhibitor give greater morbidity/mortality benefits.w5
Recently, another drug known to block a component of the renin-angiotensin-aldosterone system (RAAS), spironolactone (an aldosterone antagonist), has also been shown to reduce mortality and morbidity in CHF, even when added to an ACE inhibitor (this was demonstrated in RALES).w6
The question arises, therefore, as to what the role of the newest agents available for RAAS inhibition, the angiotensin II receptor antagonists or blockers (ARBs), might be in CHF and acute myocardial infarction?
ARB–ACE inhibitor comparison studies
Though, logically, the first question to ask of ARBs might be whether these new drugs are better than placebo, the first comparison actually made in a large scale trial was with an ACE inhibitor. The first of these—the ELITE-1 trial—addressed tolerability, whereas the hypothesis of the larger ELITE-2 trial was that losartan would be more efficacious than captopril. The approach of the ELITE trials was based on the belief that: (1) ARBs are more effective inhibitors of the RAAS than ACE inhibitors; and (2) bradykinin, the breakdown of which is blocked by ACE inhibitors, is directly or indirectly responsible for cough and possibly other adverse effects of these agents.4There is some scientific basis for the view that ARBs might be more efficacious than ACE inhibitors at blocking the RAAS. If it is accepted that ACE inhibitors bring about benefit through reducing the actions of angiotensin II, then the recent demonstration that angiotensin II generating pathways that bypass ACE exist in human myocardium and arteries is of some significance (fig 1).w7–9 Clearly, these observations suggest that ARBs offer a potentially more effective means of inhibiting the actions of angiotensin II. Arguably, the ATLAS and RALES trials also support the view that more intense inhibition of the RAAS might be better.w5 w6 This hypothesis, however, needed to be tested in one or more definitive morbidity/mortality trials—for example, ELITE-2 (see below). Selective blockade of the angiotensin (AT) II receptor subtype, with hyperstimulation of the unblocked AT2 receptor by displaced angiotensin II, is also more attractive, theoretically, than reduced stimulation of both receptor subtypes with an ACE inhibitor (fig 1). This is because the AT2R may mediate biological actions which are the opposite of those that follow AT1R activation (and, hence, potentially favourable in CHF). How have these hypotheses stood the test of clinical trials?
Tolerability: SPICE and ELITE-1
If ARBs are no more efficacious than ACE inhibitors, but are better tolerated, then there may be potentially substantial public health benefit to be gained. Though we do not yet know if ARBs are as efficacious as ACE inhibitors (see below), we do know ARBs are better tolerated. ARBs do not cause cough and do not seem to cause more of any other adverse effect than ACE inhibitors. Certainly, the SPICE trial found that candesartan was well tolerated by patients deemed to be ACE inhibitor intolerant by their physician.5 In the ELITE-1 study 20.8% of captopril treated patients discontinued treatment because of an adverse event (excluding death) compared to 12.2% of losartan treated patients (p ⩽ 0.002).4 6 The more recent ELITE-2 study supports this finding, reporting that 14.5% of patients withdrew from captopril because of adverse effects, compared to 9.4% from losartan (p < 0.001).7 8
Efficacy: ELITE-1, RESOLVD, and ELITE-2
The ELITE-1 study reported the surprising finding that patients with CHF treated with losartan had a lower mortality than patients treated with captopril.4 6 This trial was not designed to test this hypothesis and was too small to prove, with confidence, that ARBs have superior efficacy to ACE inhibitors. The RESOLVD study, comparing candesartan cilexetil (candesartan) to enalapril did not support the findings of ELITE-1.9 10
Recently, a large, prospective, properly powered, study has compared an ACE inhibitor (captopril) to an ARB (losartan).7 8ELITE-2 was designed to test the hypothesis that losartan was more efficacious than captopril (tables 1, 2, and 3).7 8 This hypothesis was not proven— that is, losartan was not superior to captopril (table 4). ELITE-2 cannot, strictly speaking, answer any more questions than the one asked. It cannot tell us whether losartan is “as good as” (equivalent to) captopril or, at least, no worse than (not inferior) to captopril. These terms have strict statistical and regulatory definitions that the design and results of ELITE-2 do not fulfil.w10 w11 Indeed, ELITE-2 cannot even tell us if losartan is superior to placebo.w12
ARB–ACE inhibitor combination studies
The second question, addressed by a major clinical trial, was whether ARB–ACE inhibitor combination therapy is better than ACE inhibitor monotherapy. The hypothesis underlying this approach is very different from that underlying the ELITE studies. While the belief that ARBs are better inhibitors of the RAAS is shared by both hypotheses, the combination therapy approach takes the view that bradykinin is “good” rather than “bad”.11 This is because bradykinin may enhance the production of nitric oxide and possibly other vasoactive mediators, such as vasodilator prostanoids in vascular and other tissues.w13–15 Bradykinin may also stimulate tissue plasminogen activator release from the endothelium and favourably influence coagulation/fibrinolysis balance.w16Consequently, combination ARB–ACE inhibitor therapy may give optimum RAAS inhibition and the putative benefits of bradykinin accumulation, through inhibition of its breakdown.
A number of relevant “mechanistic” and “pilot” clinical studies preceded what was expected to be the first definitive large scale trial, Val-HeFT, exploring this hypothesis.11
Small scale studies testing combination therapy
There is conflicting clinical evidence that some of the effect of ACE inhibitors may be caused by the blocking effect these drugs have on bradykinin breakdown. Two recent studies in hypertensive individuals and healthy subjects, using a selective bradykinin inhibitor, supports such an action, though another in heart failure does not.w13–15 These mechanistic findings are supported by some clinical observations. Dunselman reported that substitution of telmisartan (10–80 mg once daily) for enalapril (10 mg twice daily) in patients with heart failure led to an increase in blood pressure (the REPLACE study), thus supporting the possibility that ACE inhibitors have an additional hypotensive mechanism of action compared to ARBs.12 Baruch and colleagues studied the immediate and four week haemodynamic and neurohumoral effects of placebo, valsartan 80 mg twice daily or valsartan 160 mg twice daily added to conventional treatment (including an ACE inhibitor) in patients with CHF.13 Compared to placebo, high dose valsartan reduced pulmonary capillary wedge pressure and systolic blood pressure acutely and after one month's treatment. Valsartan 80 mg and 160 mg twice daily significantly reduced aldosterone at four weeks. Not all studies have supported these findings, however, and there remains the nagging doubt that similar benefits might be obtained by using a bigger dose of ACE inhibitor, rather than adding an ARB.w17 w18 However, Hamroff and colleagues reported a small but impressive six month randomised trial in which patients with moderately severe CHF were randomised to placebo or losartan 50 mg once daily.14 All were receiving full conventional treatment including an ACE inhibitor given in an adequate dose (for example, the mean daily dose of enalapril was 32 mg). The primary end points were exercise capacity and New York Heart Association (NYHA) functional class. Both improved significantly and losartan was well tolerated. In the RESOLVD pilot study the combination of enalapril and candesartan had significantly more favourable effects on the left ventricular remodelling than either monotherapy.9 10 Clinical outcome was not, however, better in the candesartan–enalapril combination group. The hypothesis that combination therapy is the optimum had, therefore, to be tested in a large scale morbidity–mortality trial.
- AIRE Acute Infarction Ramipril Efficacy
- ATLAS Assessment of Treatment with Lisinopril And Survival
- CHARM Candesartan in Heart failure: Assessment of Reduction in Mortality and Morbidity
- CONSENSUS-1 Co-operative North Scandanavian Enalapril Survival Study
- ELITE Evaluation of Losartan In The Elderly
- HOPE Heart Outcomes Prevention Evaluation
- OPTIMAAL Optimal Therapy in Myocardial infarction with the Angiotensin II Antagonist Losartan
- PEP-CHF Perindopril for Elderly People with Chronic Heart Failure
- RALES Randomised Aldactone Evaluation Study
- REPLACE Replacement of angiotensin converting enzyme inhibition
- RESOLVD Randomized Evaluation of Strategies for Left Ventricular Dysfunction
- SAVE Survival and Ventricular Enlargement
- SOLVD-T Treatment arm of the Studies Of Left Ventricular Dysfunction
- SPICE Study of Patients Intolerant of Converting Enzyme Inhibitors
- STRETCH Symptom, Tolerability, Response to Exercise Trial of Candesartan cilexetil in Heart failure
- TRACE Trandolapril Cardiac Evaluation
- Val-HeFT Valsartan Heart Failure Trial
- V-HeFT Vasodilator Heart Failure Trial
- VALIANT Valsartan In Acute myocardial infarction
The key features of the design of Val-HeFT trial are shown in tables 1 and 2.11 The demographic characteristics and preliminary results of Val-HeFT were presented by J Cohn at the 73rd scientific sessions of the American Heart Association (AHA) in New Orleans, 15 November 2000, and all of the subsequent information reported in this review has been obtained from that presentation. The principal hypothesis tested by Val-HeFT was that adding valsartan to conventional treatment (including an ACE inhibitor and β blocker, where appropriate) would improve clinical outcome. The co-primary end points were: (1) mortality (all cause); and (2) mortality or morbidity (where morbidity included hospitalisation for CHF, resuscitated sudden death, and administration of intravenous inotropic or vasodilator treatment for CHF for ⩾ 4 hours). Secondary end points included change in NYHA functional class, signs and symptoms of CHF, left ventricular ejection fraction, and quality of life. Val-HeFT randomised 5010 patients, the clinical characteristics of whom are summarised in table 3.11 15 The average follow up time was approximately 1.9 years.
The principal results of the Val-HeFT trial, as presented at the AHA, are shown in table 5. Valsartan did not reduce mortality but did significantly reduce the combined morbidity/mortality end point by approximately 13% (p = 0.009). This effect was principally caused by a substantial 27% reduction in CHF hospitalisation (table 5, p = 0.00001). There were similarly impressive and significant improvements in the other secondary outcomes presented. At face value, therefore, Val-HeFT would appear to be a “positive” trial, with significant improvements in pre-specified co-primary and secondary end points. Unfortunately, however, the story may not be that simple. This is because Cohn went on to present detailed subgroup analyses which appeared to raise important questions about the overall findings of Val-HeFT.
Firstly, outcomes in the small minority (7%) of patients not taking on ACE inhibitor at baseline were compared to those in patients taking an ACE inhibitor. The former group had an approximately 45% reduction in mortality/morbidity compared to a 12% reduction in the latter. In other words, this analysis raises the possibility that most of the benefit in the overall trial can be explained by a particularly large effect in patients not receiving an ACE inhibitor. To complicate matters further, patients receiving a β blocker at baseline (about 35%) were compared to those not taking a β blocker. The hazard ratio for the mortality/morbidity end point in patients taking a β blocker was 1.15 (that is, there was a trend for such patients to do worse on valsartan) compared to 0.78 in those not on a β blocker. In other words, this and some further analysis raised the possibility that “triple neurohumoral blockade” (ACE inhibitor, β blocker, ARB) has no advantage over double blockade and may even be disadvantageous. It must be emphasised that subgroup analysis of this type is fraught with danger, can be very misleading, and should only be regarded as hypothesis generating.w19 Unfortunately, however, the β blocker subgroup analysis has attracted much attention because β blockers, along with ACE inhibitors, are now regarded as mandatory treatment for CHF. The issue is further confounded by an apparently similar β blocker interaction in ELITE-2 (captopril treated patients did better than losartan treated patients if receiving β blocker treatment at baseline).8
Where then do we stand with ARBs following both ELITE-2 and Val-HeFT? I would have to conclude that the picture still remains unclear. From a purist perspective Val-HeFT does suggest that adding an ARB to conventional treatment reduces morbidity (CHF hospitalisation). How much attention one should pay to subgroup analyses is very debatable. Before reaching any firm conclusion we must at least wait for full publication of the Val-HeFT results. It must be reiterated that only preliminary data are available at the time of writing. Analysis of the neurohumoral and left ventricular remodelling data from this study should give additional insight into the issues raised above. We will also have a more complete picture when the CHARM programme completes.16 This three study programme is now ideally poised to address some of the remaining uncertainty about ARBs.16 In particular, CHARM includes a study in patients not taking an ACE inhibitor (study 0003 or “alternative CHARM”—see below), and more patients in the ACE-I + ARB combination arm of CHARM (study 0006 or “added CHARM”) are receiving a β blocker at baseline than in Val-HeFT (55% v 35%) andall are taking an ACE inhibitor, allowing further exploration of any potential ARB–β blocker interaction.
Clinical pharmacology of ARBs versus ACE inhibitors
Angiotensin II exerts its known biological actions via the angiotensin II type 1 receptor (AT1R). The angiotensin receptor blockers (ARBs) currently available for clinical use selectively antagonise the action of angiotensin II at the AT1R
The role of the angiotensin II type 2 receptor (AT2R) is uncertain. It may, when stimulated, increase the production of nitric oxide and other vasodilator/anti mitotic substances. The AT2R is hyperstimulated during selective AT1R blockade in vivo
Angiotensin II may be generated through the action of ACE and, probably, chymase on angiotensin I. There may be additional “non-ACE” pathways for angiotensin II production. AT1R blockers (but not ACE inhibitors) antagonise the effects of angiotensin II generated by “non ACE” pathways
ACE is also known as kininase II which degrades bradykinin (that is, bradykinin accumulates during ACE inhibition). Bradykinin may, directly or indirectly, cause some of the adverse effects of ACE inhibitors (for example, cough) but may also have beneficial vasodilator, growth, and fibrinolytic actions.
ARB–placebo comparison studies
Remarkably, it has been generally assumed that ARBs are an effective treatment for CHF (that is, superior to placebo) even though there are very few data to support this assumption. Clearly, the widely held opinion that ARBs are efficacious in CHF is based on the view that RAAS inhibition is beneficial, a perception based on the belief that ACE inhibitors and spironolactone exert their effect in this way. While almost certainly true, at least in part, even the most apparently obvious hypotheses should be tested in medicine.
Small studies comparing ARBs to placebo in CHF
ARBs, like ACE inhibitors, have favourable acute and chronic neurohumoral and haemodynamic actions in CHF.10 13-17 w17 w18 w20 There are, however, remarkably few data showing any clinical benefit of ARBs over placebo. A recent trial, STRETCH, has shown that one of these agents, candesartan cilexetil, can improve exercise tolerance in CHF in a dose dependent manner, compared to placebo.18 This has not, however, been a consistent finding in all studies (there are two fairly large, unpublished, exercise studies showing no benefit of losartan over placebo).w21 Meta-analyses of relatively small trials with losartan and candesartan have, however, suggested that ARBs might improve clinical outcomes when compared to placebo, though an imputed placebo analysis of ELITE-2 (using SOLVD-T as the historical control) gives only weak support to these meta-analyses.19 w12 w21 Adding the data from the Val-HeFT subgroup not treated with an ACE inhibitor gives more support to the view that ARBs are more efficacious than placebo (table 6). However, no prospective, randomised, placebo controlled trial has, to date, tested the hypothesis that ARBs are superior to placebo in terms of morbidity/mortality or mortality end points in CHF.
A wider role for RAAS inhibition in CHF?
The emergence of a new class of drug for RAAS inhibition also presents the opportunity to test additional questions not formally tested in previous trials with ACE inhibitors. One pressing issue in clinical cardiology is the treatment of CHF in patients with preserved left ventricular systolic function, who make up perhaps a third of all patients with CHF and who also have an increased morbidity and mortality.w22 w23 Though left ventricular ejection fraction (LVEF) measurements were not required for entry into either the CONSENSUS-1 or V-HeFT studies, these trials are generally considered to have recruited patients with left ventricular systolic dysfunction.1 w24 w25
Major ARB clinical trials in CHF
ELITE-2 compared losartan to captopril. The primary end point in this trial was all cause mortality. Mortality was not reduced by losartan compared to captopril
Val-HeFT compared valsartan to placebo, added to full conventional treatment including, in most patients, an ACE inhibitor. These were two co-primary end points—all cause mortality and morbidity/mortality. While valsartan had no effect on the former, the latter was reduced by 13% (mainly because of a 27% reduction in heart failure hospitalisation). Subgroup analysis suggested a particularly large benefit in the minority of patients not taking ACE inhibitors at baseline. More controversially, there was also the suggestion that adding valsartan to background treatment with both an ACE inhibitor and β blocker increased risk
The CHARM programme is still underway and comprises three component trials. One is comparing candesartan to placebo in patients intolerant of an ACE inhibitor, and the second is making the same comparison in patients taking an ACE inhibitor. The third is comparing these two treatments in patients with heart failure and preserved left ventricular systolic function. The primary end point in each individual trial is cardiovascular death or heart failure hospitalisation and all cause mortality in the overall programme
I-PRESERVE is a trial still in the planning stage which intends to compare irbesartan to placebo in patients with heart failure, preserved left ventricular systolic function, and an increased plasma BNP concentration.
It is possible that RAAS inhibition might also be of benefit in patients with CHF and preserved left ventricular systolic function. These patients are treated with diuretics which may be expected to cause RAAS activation.w26 Many are hypertensive, diabetic, and have left ventricular hypertrophy, comorbidities that might be expected to respond favourably to RAAS inhibition, especially in the light of the recently reported HOPE study.w27 w28 Once more, of course, this is a hypothesis that needs to be tested in an appropriately designed clinical trial. Two trials are underway. One with the ACE inhibitor perindopril (PEP-CHF)w29 and one which is a component trial of the CHARM programme (study 0007 or “preserved CHARM”).16 Another trial, with irbesartan, is at an advanced stage of planning. This trial, to be known as I-PRESERVE, will enrol patients with current signs/symptoms of heart failure (or an admission to hospital primarily because of heart failure within the last three months), an LVEF ⩾ 0.45, and a raised plasma brain natriuretic peptide (BNP) concentration.
ARB myocardial infarction trials
Two trials are underway in patients with acute myocardial infarction. These are OPTIMAAL and VALIANT which are outlined in table7. OPTIMAAL is similar to the ELITE trials in comparing losartan to captopril.20 21 The patients randomised are high risk survivors broadly similar, but not identical, to those recruited into the seminal ACE inhibitor post-myocardial infarction trials (SAVE, AIRE, TRACE).w1–3 OPTIMAAL is sufficiently large to have a 95% power of showing a 20% relative reduction in the risk of death with losartan compared to captopril. VALIANT has a more complex design with three treatment groups (captopril, valsartan, and their combination) and is powered not just to test for superiority but also for non-inferiority (table 7).22
As a consequence, the patient entry criteria must exactly mirror those of the reference trials SAVE,w1 AIRE,w2 and TRACEw3. Among the questions VALIANT can address is the one of whether ARBs have similar efficacy to ACE inhibitors (non-inferiority) but are better tolerated. VALIANT will, of course, also show whether combination ACE inhibitor–ARB treatment is superior to ACE inhibitor monotherapy in the post-myocardial infarction setting.
Conclusions and clinical recommendations
It seems reasonable to recommend use of an angiotensin receptor blocker as an alternative to an ACE inhibitor in the patient truly intolerant of the inhibitor. An angiotensin receptor blocker may also be added to an ACE inhibitor to improve symptoms and reduce the risk of hospital admission with worsening heart failure. At present angiotensin receptor blockers are only indicated where there is ACE inhibitor or β blocker intolerance.
Angiotensin receptor blockers may be used as an alternative to an ACE inhibitor in the patient truly intolerant of an ACE inhibitor.
Angiotensin receptor blockers may also be used in addition to an ACE inhibitor when a patient is intolerant of a β blocker.
“Triple neurohumoral blockade” with an ACE inhibitor, β blocker, and angiotensin receptor blocker is not, presently, recommended.
Angiotensin receptor blockers for chronic heart failure and acute myocardial infarction - JJV McMurray. Heart 2001;86:97-103.
References for web version only
- Pfeffer MA, Braunwald E, Moye LA, et al. Effect of captopril on mortality and morbidity in patients with left ventricular dysfunction after myocardial infarction. Results of the survival and ventricular enlargement trial. The SAVE investigators. N Engl J Med 1992;327:669- 77.
- The AIRE Study Investigators. Effect of ramipril on mortality and morbidity of survivors of acute myocardial infarction with clinical evidence of heart failure. The acute infarction ramipril efficacy (AIRE) study investigators. Lancet 1993;342:821- 8.
- Kober L, Torp-Pedersen C, Carlsen JE, et al. A clinical trial of the angiotensin converting enzyme inhibitor trandolapril in patients with left ventricular dysfunction after myocardial infarction. N Engl J Med 1995;333:1670- 6.
- Flather MD, Yusuf S, Kober L, et al. Long-term ACE-inhibitor therapy in patients with heart failure or left-ventricular dysfunction: a systematic overview of data from individual patients. ACE-inhibitor myocardial infarction collaborative group. Lancet 2000;355:1575- 81.
- Packer M, Poole-Wilson PA, Armstrong PW, et al. Comparative effects of low and high doses of the angiotensin-converting enzyme inhibitor, lisinopril, on morbidity and mortality in chronic heart failure. Circulation 1999;341:709- 17.
- Pitt B, Zannad F, Remme WJ, et al. The effect of spironolactone on morbidity and mortality in patients with severe heart failure. Randomized aldactone evaluation study. N Engl J Med 1999;100:2312- 18.
- Wolny A, Clozel JP, Rein J, et al. Functional and biochemical analysis of angiotensin II forming pathways in the human heart. Circulation Res 1997;80:219- 27.
- Voors AA, Pinto YM, Buikema H, et al. Dual pathway for angiotensin II formation in human internal mammary arteries. Br J Pharmacol 1998;125:1028- 32.
- Petrie MC, Padmanabhan N, McDonald JE, et al. Angiotensin converting enzyme (ACE) and non-ACE dependent angiotensin II generation in resistance arteries from patients with heart failure and coronary heart disease. J Am Coll Cardiol 2001;37:
- Jones B, Jarvis P, Lewis JA, et al. Trials to assess equivalence: the importance of rigorous methods. BMJ 1996;313:36- 9.
- Siegel JP. Equivalence and non-inferiority trials Am Heart J 2000;139:166- 70.
- Berry C, Norrie J, McMurray J. Are Angiotensin II receptor blockers more efficacious than placebo in heart failure? Implications of ELITE-2. Am J Cardiol 2001;87:606- 7.
- Gainer JV, Morrow JD, Loveland A, et al. Effect of bradykinin-receptor blockade on the response to angiotensin-converting enzyme inhibitor in normotensive and hypertensive subjects. N Engl J Med 1998;339:1285�92.
- Squire IB, O�Kane KP, Anderson N, et al. Bradykinin B(2) receptor antagonism attenuates blood pressure response to acute angiotensin-converting enzyme inhibition in normal men. Hypertension 2000;36:132- 6.
- Davie AP, Dargie HJ, McMurray JJ. Role of bradykinin in the vasodilator effects of losartan and enalapril in patients with heart failure. Circulation 1999;100:268- 73.
- Labinjoh C, Newby DE, Dawson P, et al. Fibrinolytic actions of intra-arterial angiotensin II and bradykinin in vivo in man.Cardiovasc Res 2000;47:707- 14.
- Tonkon M, Awan N, Niazi, et al. A study of the efficacy and safety of irbesartan in combination with conventional therapy, including ACE inhibitors, in heart failure. Irbesartan heart failure group. Int J Clin Pract 2000;54:11- 4, 16- 8.
- Jorde UP, Ennezat PV, Lisker J, et al. Maximally recommended doses of angiotensin-converting enzyme (ACE) inhibitors do not completely prevent ACE-mediated formation of angiotensin II in chronic heart failure. Circulation 2000;101:844- 6.
- Horton R. From star signs to trial guidelines. Lancet 2000;355:1033- 4.
- Parker AB, Azevedo ER, Baird MG, et al. ARCTIC: assessment of haemodynamic response in patients with congestive heart failure to telmisartan: a multicentre dose-ranging study in Canada. Am Heart J 1999;138:843- 8.
- Sharma D, Buyse M, Pitt B, et al. The losartan heart failure mortality meta-analysis study group. Meta-analysis of observed mortality data from all controlled, double-blind, multiple-dose studies of losartan in heart failure. Am J Cardiol 2000;85:187- 92.
- Senni M, Tribouilloy CM, Rodeheffer RJ, et al. Congestive heart failure in the community � a study of all incident cases in Olmsted County, Minnesota in 1991. Circulation 1998;98:2282- 9.
- Vasan RS, Larson MG, Benjamin EJ, et al. Congestive heart failure in subjects with normal versus reduced left ventricular ejection fraction � prevalence and mortality in a population-based cohort. J Am Coll Cardiol 1999;33:1948- 55.
- Cohn JN, Archibald DG, Ziesche S, et al. Effect of vasodilator therapy on mortality in chronic congestive heart failure. N Engl J Med 1986;314:1547- 52.
- Cohn JN, 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.
- Benedict CR, Weiner DH, Johnstone DE, et al. Comparative neurohormonal responses in patients with preserved and impaired left ventricular ejection fraction � Results of the studies of left ventricular dysfunction (SOLVD) Registry. J Am Col Cardiol 1993;212:A146- 53.
- Mindlen F, Nordaby R, Ruiz M, et al. The HOPE (heart outcomes prevention evaluation) study: the design of a large, simple randomized trial of an angiotensin converting enzyme inhibitor (ramipril) and vitamin E in patients at high risk of cardiovascular events. Can J Cardiol 1996;12:127- 37.
- Yusuf S, Sleight P, Pogue J, et al. Effects of an angiotensin-converting-enzyme inhibitor, ramipril, on cardiovascular events in high-risk patients. The heart outcomes prevention evaluation study investigators. N Engl J Med 2000;342:145- 53.
- Cleland JG, Tendera M, Adamus J, et al. Perindopril for elderly people with chronic heart failure: the PEP-CHF study. The PEP investigators. Eur J Heart Failure 1999;1:211- 7.