Author + information
- Received April 9, 2004
- Revision received July 19, 2004
- Accepted July 28, 2004
- Published online November 2, 2004.
- Christian N. Gring, MD* ( and )
- Gary S. Francis, MD, FACC
- ↵*Reprint requests and correspondence:
Dr. Christian N. Gring, c/o Dr. Brian Griffin, Cleveland Clinic Foundation, Desk F-15, 9500 Euclid Avenue, Cleveland, Ohio 44195
Multiple trials over the past several years have examined indications for angiotensin receptor blockers (ARBs) in the treatment of left ventricular dysfunction, both acutely after myocardial infarction and in chronic heart failure. Yet despite these data, there is still confusion regarding the efficacy of ARBs as monotherapy in these patient populations, as well as the specific indications for combination ARB/angiotensin-converting enzyme (ACE) inhibitor therapy. We examine the key differences among the trials—including the ACE inhibitor dose, the ARB and its dose, blood pressure reduction, and patient populations—to present our perspective on ARB use, alone or in combination with ACE inhibitors, in patients with chronic heart failure and post-myocardial infarction left ventricular dysfunction. We conclude that ACE inhibitors remain the first-line therapy for left ventricular dysfunction. Angiotensin receptor blockers should be reserved for monotherapy in ACE intolerant patients and for combination therapy in symptomatic class II/III patients with chronic heart failure.
Angiotensin-converting enzyme (ACE) inhibitors are a cornerstone of therapy in patients with left ventricular (LV) systolic dysfunction, both immediately after myocardial infarction and long-termfor heart failure. In patients with postmyocardial LV dysfunction, randomized, placebo-controlled trials have consistently shown a substantial mortality benefit of ACE inhibitor therapy. A systematic review of these trials, whose mean duration of therapy was 31 months, reported a 5.7% absolute risk reduction in mortality with ACE inhibitors (1). In chronic LV dysfunction, both Studies Of Left Ventricular Dysfunction (SOLVD) (2) and Cooperative North Scandinavian Enalapril Survival study (CONSENSUS) (3) showed significant survival benefits with ACE inhibitor therapy. Thus, ACE inhibitors have become the primary antagonist of the renin-angiotensin-aldosterone system (RAAS) in the treatment of patients with heart failure.
Because of the impressive record of ACE inhibitors, RAAS antagonism has become a target for additional pharmacologic therapies—most notably with angiotensin receptor blockers (ARBs). The rationale for this class of drugs is diverse, and includes prevention of “ACE-escape” (4,5) (which may portend a worse prognosis ), the potential for synergism when used in combination with ACE inhibitors (7), more specific angiotensin II blockade via AT1receptors, and the preservation of (theoretical) benefits derived from unopposed AT2receptor agonism (8). The ACE-escape is defined as the gradual elevation of serum angiotensin II and aldosterone levels, despite ongoing RAAS inhibition with ACE inhibitors (9–11). Unopposed AT2agonism may be a benefit of ARB therapy because, although serum angiotensin II activates both AT1and AT2receptors, ARBs only block the AT1receptor. The AT1receptors mediate a host of potentially deleterious cardiovascular effects, including cell proliferation, vasoconstriction, aldosterone secretion, and sodium reabsorption (8). AT2receptors are not completely understood, and there are no clinical data to prove that their unopposed activation is beneficial, but they are generally believed to counter the AT1response and lead to antiproliferative, antigrowth, and vasodilatory effects (8,12). Thus ARBs, either as monotherapy or in combination with ACE inhibitors, have become a logical target of investigation for use in patients with heart failure.
Table 1lists the key heart failure trials that have examined the use of ARBs compared with, or in combination with, ACE inhibitors (13–18). Tables 2 and 3⇓outline the trial designs, treatment strategies, and major results. The Assessment of Treatment with Lisinopril and Survival (ATLAS) trial (19) did not include ARBs, but is included for reference and is discussed later. These trials have generated uncertainty about how ARBs should be used to treat patients with heart failure, because they have not consistently shown benefit in patients with LV dysfunction. For example, the Valsartan Heart Failure Trial (Val-HeFT) and the Candesartan in Heart Failure: Assessment of Reduction in Mortality and Morbidity-Added (CHARM-Added) trial showed a reduction in cardiovascular morbidity and/or mortality with combination therapy compared with ACE inhibitor monotherapy, while Valsartan in Acute Myocardial Infarction Trial (VALIANT) did not; the VALIANT trial and Losartan Heart Failure Survival study (ELITE-II) showed no difference between ACE inhibitors and ARBs, while Optimal Trial in Myocardial Infarction with the Angiotensin II Antagonist Losartan (OPTIMAAL) suggested that ARBs were inferior to ACE inhibitors.
These results, however, are not as inconsistent as they initially appear. The ARB trials have many similarities, but differ in their enrollment criteria, treatment strategies, and end points. We need to understand how these differences may have contributed to the disparate results and define more precisely which subgroups of patients with heart failure are likely to benefit most from ARB therapy. Specifically, we believe that an examination of the patient populations studied, ARB selection and dose, and blood pressure reduction help to put the various trial results in perspective. The trials also differ in ACE inhibitor dosing but, as discussed below, this fact may be less relevant. Ultimately, we must clarify the subgroups of patients who should be treated with ARBs, the doses that should be used, and the combination of additional medications that seem to offer the most benefit.
Chronic heart failure or postinfarct LV dysfunction?
Perhaps the most striking difference among trials is the patient populations studied (Table 2). The CHARM-Added, Val-HeFT, and ELITE-II trials all had similar patient profiles: a predominance of class II to III patients with chronic heart failure, mean ejection fractions of 27% to 31%, and an approximate 60% incidence of ischemic cardiomyopathy. The CHARM trial patients were likely sicker, because all class II enrollees had been recently hospitalized with congestive symptoms. In contrast, the patients enrolled in the VALIANT and OPTIMAAL trials had LV dysfunction in the setting of a recent myocardial infarction, a milder degree of heart failure (26% to 32% of patients were Killip class I), and a mean ejection fraction of at least 35%. Finally, by definition, patients had ischemic heart disease.
The significance of these differences is not entirely clear, as the patient populations did share a key demographic—the majority had depressed ejection fractions and heart failure symptoms due to ischemic heart disease. In addition, the overall event rates in the trials were similar. Annualized total mortality (9% to 10%) and cardiovascular mortality (8% to 9%) were comparable in the CHARM-Added and VALIANT trials; the Val-HeFT trial reported a total mortality of 9% as well, but did not examine a cardiovascular mortality end point. As expected, annualized hospitalizations for heart failure in the CHARM-Added and Val-HeFT trials (both 8% to 9%) were higher than in the VALIANT (∼6%) trial, whereas the VALIANT trial's myocardial infarction rate (2.9%) exceeded that of the CHARM-Added trial (1.6%). The event rates argue that there is substantial overlap in the disease processes and suggest that the patient populations homogenize over time.
Nevertheless, at the time of enrollment, patients were at different points in time along a lengthy disease continuum; it is quite possible that patients with chronic heart failure have different morbidities and mortality, as well as different responses to therapies, than those with acute myocardial infarctions. Moreover, one cannot equate acute postmyocardial infarction LV dysfunction, which is often transient, with established chronic heart failure. Finally, the myocardial tissue RAAS may be quite different between these patient groups, and may evolve as patients progress from recent myocardial infarction to worsening classes of chronic heart failure (20).
The CHARM-Added studies and Val-HeFT provide compelling evidence that ARBs have a role in treating patients with chronic heart failure, particularly in ACE-intolerant patients (CHARM-Alternativetrial), and patients with relatively advanced heart failure. The data for treating postmyocardial infarction LV dysfunction with ARBs are not as robust. In the VALIANT trial, ARB monotherapy was equivalent to ACE inhibitor therapy—again supporting ARB use in ACE intolerant patients—but combination therapy was not beneficial. In the OPTIMAAL trial, the ARB was inferior. The variation in patient populations studied likely explains some of the trials' discordant results, and argues that ARB therapy works best in more symptomatic patients with long-term heart failure. The discord between the OPTIMAAL and VALIANT trials and the fact that the ELITE-II trial found a neutral effect of ARBs versus ACE inhibitors in patients with chronic failure suggest that additional differences in the trials may be impacting their results.
Are all ARBs equal?
The specific ARB used and its dose may provide a second explanation for the disparate results in the trials. Clearly there are pharmacokinetic and pharmacodynamic differences among ARBs, which may be clinically relevant. Both candesartan and valsartan are dose-dependent inhibitors of the AT1receptor, with respective receptor affinities that are approximately 80 and 10 times that of losartan (7). Importantly, valsartan's bioavailability drops by 40% if it is taken with food (7). Losartan is a relatively weaker receptor blocker than either valsartan or candesartan, but its active metabolite is capable of an insurmountable receptor blockade (7). Clinically, these differences may have importance. Head-to-head blood pressure studies have shown that valsartan 160 mg is superior to losartan 100 mg daily, and that candesartan at doses of 8 mg and 16 mg daily is superior to losartan at 50 mg and 100 mg daily (7).
The CHARM studies and the Val-HeFT trial both used relatively potent ARBs in high doses, and demonstrated reductions in hospital admissions for heart failure; the CHARM-Added trial showed a cardiovascular mortality reduction. Conversely, the ELITE-II and OPTIMAAL trials both used low-dose losartan versus high-dose captopril, and neither study showed a benefit of ARB therapy over ACE inhibitor. These data suggest that 50 mg daily of losartan may be too low. It is difficult to draw conclusions about the relative strength of candesartan 32 mg versus valsartan 160 mg twice a day.
Does the ACE inhibitor dose matter?
A third difference in the trials was variation in the ACE inhibitor and dose used. Although the protocols of the CHARM and Val-HeFT trials allowed the use of multiple ACE inhibitors, all of the trials included patients receiving captopril. Thus, one might make some assumptions about whether ACE dose impacted the trial results. In the CHARM-Added and Val-HeFT trials, the two largest trials that showed ARB benefit in patients with chronic heart failure, the average dose of captopril was substantially lower than in the VALIANT or OPTIMAAL trials, which found no incremental benefit of ARB therapy compared to, or in combination with, ACE inhibitors (Table 2). The ATLAS study, however, provides compelling data that higher-dose ACE inhibitor is superior to lower-dose therapy. In this study, which compared low-dose versus high-dose lisinopril in patients with chronic heart failure, the investigators reported a 10% reduction in cardiovascular mortality (p = 0.07) and a highly significant 24% reduction in hospitalization for heart failure in the higher-dose group. There was no difference in the primary end point of all-cause mortality. In similar patient populations, the CHARM-Added and Val-HeFT trials both reported comparable results with combination ARB and ACE inhibitor therapy: highly significant reductions in heart failure hospitalizations but no effect on total mortality. It should be noted, however, that the CHARM-Added trial's reduction in cardiovascular mortality was significant. The ATLAS study results might suggest that ARBs showed benefit in the CHARM-Added and Val-HeFT trials because patients were receiving submaximal doses of ACE inhibitor, and that maximizing the ACE dose would have yielded similar results.
But two points argue that an insufficient ACE inhibitor dose is not the whole answer. First, on a pathophysiologic basis, one would expect that higher doses of ACE inhibitor would suppress angiotensin II levels more completely than lower doses, thereby marginalizing any benefit of ARBs. Multiple studies have failed to demonstrate this dose-response relationship in patients with heart failure; rather, angiotensin II levels can vary markedly with any given ACE inhibitor dose (6,10,11). These data suggest that higher doses of ACE inhibitor do not predictably alter angiotensin II levels, at least in the serum. Thus, if higher doses of ACE inhibitor do confer additional benefits, they might be mediated through other mechanisms, including alterations in myocardial tissue ACE activity or extra-neurohormonal pathways.
Second, a substantial number of patients in both the Val-HeFT and CHARM-Added trials received relatively high doses of ACE inhibitors, and still showed benefit with the addition of an ARB. Although captopril doses in these studies were moderate compared with postmyocardial infarction heart failure studies, the respective mean doses of enalapril were 17 mg/day and 17.2 mg/day, which were comparable with the mean dose in the SOLVD trial (16.6 mg/day). Effectively, these patients were on evidence-based doses of enalapril and still benefited with ARB therapy. Moreover, in the CHARM-Added trial, patients were divided into subgroups based on whether they were on high-dose, “target” daily doses of ACE inhibitor—for example, captopril 150 mg, or enalapril 20 mg. These patients actually had a greater benefit with the addition of candesartan than those on lower, non–evidence-based ACE doses. The ATLAS study argues that ACE inhibitor dose is important in the treatment of patients with chronic heart failure, and that underdosed patients have poorer outcomes. But it is difficult to argue that patients in the CHARM and Val-HeFT trials were underdosed with ACE inhibitors; we would assert that the variations in ACE inhibitor dosing across the trials did not substantially influence their results.
The importance of blood pressure reduction
A final possible explanation for the trials' different outcomes is that the results reflect variations in blood pressure response to therapy. Hypertension has long been recognized as a risk factor for congestive heart failure and ischemic heart disease. Trials of essential hypertension, which typically have excluded patients with heart failure, have estimated that a 3 mm Hg reduction in systolic blood pressure will decrease the risk of developing heart failure by 10% to 20% (21). A recent meta-analysis has estimated that a long-term 5 mm Hg decline in diastolic blood pressure would decrease ischemic heart disease mortality by 30%, and even a 2-point reduction in systolic blood pressure would yield a 7% decrement in ischemic heart disease mortality (22).
The VALIANT, OPTIMAAL, and ELITE II trials did not demonstrate ARB superiority, and none of these studies had substantial blood pressure differences among the treatment arms (Table 3). Conversely, the CHARM-Added, Val-HeFT, and ATLAS trials had significant reductions in blood pressure—4.4 to 5.2 mm Hg systolic and 1.3 to 3.9 mm Hg diastolic. The CHARM-Added and ATLAS trials reported decreases in cardiovascular mortality of 17% and 10%, respectively—reductions that could be largely explained on the basis of improved blood pressure alone. The consistent parallels between blood pressure reduction and clinical outcome argue for a possible causative relationship.
We now have multiple large, rigorously performed studies that have investigated potential indications for ARBs in patients with heart failure. We believe several conclusions can be drawn. First, ACE inhibitors, at evidence-based doses, continue to be first-line drugs in the treatment of heart failure; no other RAAS antagonist has proven to be superior. Second, patients with heart failure who are intolerant of ACE inhibitors should be treated with ARBs. In ACE-intolerant patients with postmyocardial infarction LV dysfunction, valsartan 320 mg is indicated. In patients with chronic heart failure, candesartan 32 mg daily is clearly beneficial. Additionally, subgroup analysis from the Val-HeFT trial suggests valsartan 320 mg daily is valuable monotherapy in this population as well (23). The third conclusion is that the ARB dose is likely very important—a point that the ongoing Heart failure End point evaluation with the Angiotensin II Antagonist Losartan (HEAAL) study, which is comparing 50 mg versus 150 mg daily of losartan in ACE-intolerant patients with heart failure, is addressing. Finally, combined therapy with ACE inhibitor and ARB is without merit in acute, postinfarct LV dysfunction.
Assessing the role of combined ACE/ARB therapy in patients with chronic LV dysfunction remains challenging because the data are incomplete. Asymptomatic and minimally symptomatic patients have not been adequately studied. In symptomatic patients with heart failure, the Val-HeFT and CHARM-Added trials found that combined therapy reduces hospitalizations for heart failure, but not overall mortality. But the VALIANT trial showed no mortality or morbidity benefit of combination therapy, despite the fact that a significant portion of the VALIANT trial patients, several months postinfarct, must have resembled the Val-HeFT or CHARM trial patients. Although tissue RAAS activity might initially be quite different between patient groups in the studies, it presumably evolves as patients progress to symptomatic heart failure. This evolution is poorly characterized; neither its implications for ARB therapy nor its clinical relevance is entirely known. If tissue RAAS activity is a critical variable, we must determine how soon after an infarct ARB therapy should be added. We must also consider the possibility that the benefits seen with combination therapy might not be due specifically to ARB therapy, but perhaps to blood pressure reduction or other unrecognized differences between patient populations.
Currently, the most compelling indication for combination therapy is in patients who remain symptomatic despite standard therapy including target doses of ACE inhibitors and beta-blockers. In the CHARM-Added trial, which provides the best evidence for combined therapy, 73% of patients were class III, and all class II patients had at least one hospitalization for heart failure in the six months before enrollment. Likewise, in subgroup analysis of the Val-HeFT trial, only class III and IV patients experienced significant benefit with the addition of valsartan. Finally, a critical question that still needs further clarification is whether an aldosterone antagonist or ARB should be added first to the medical regimens of highly symptomatic patients.
Polypharmacy has become essential in the management of heart failure, but it exposes patients to increased costs and risks of drug-drug interactions. Thus, it is paramount that we define clearly which patient groups will benefit most from therapy. Angiotensin receptor blockers are excellent therapeutic substitutes for ACE inhibitors, but they are costly: evidence-based doses of captopril (50 mg three times a day) or enalapril (10 mg twice a day) cost $10 to $15 per month, whereas the monthly target ARB doses in the CHARM and VALIANT trials cost approximately $50 and $100, respectively (24). From a public health perspective, therefore, ACE inhibitors are preferable. For the treatment of chronic heart failure, combination therapy should be reserved for increasingly symptomatic patients because the incremental value of ARBs when added to ACE inhibitors and beta-blockers remains somewhat speculative in class I and II patients. Even less clear are the risks and benefits of triple RAAS therapy with ACE inhibitors, ARBs, and aldosterone antagonists; we recommend that such combinations be avoided until their safety and efficacy are better established.
Dr. Francis serves as a consultant for Novartis and on the Data Safety and Monitoring Boardfor the ongoing HEAAL trial.
- Abbreviations and acronyms
- angiotensin-converting enzyme
- angiotensin receptor blocker
- Assessment of Treatment with Lisinopril And Survival trial
- left ventricle/ventricular
- renin-angiotensin-aldosterone system
- Studies Of Left Ventricular Dysfunction
- Received April 9, 2004.
- Revision received July 19, 2004.
- Accepted July 28, 2004.
- American College of Cardiology Foundation
- Swedberg K.,
- Kjekshus J.,
- Snapinn S.,
- for the CONSENSUS Investigators
- Petrie M.C.,
- Padmanabhan N.,
- McDonald J.E.,
- Hillier C.,
- Connell J.M.C.,
- McMurray J.J.V.
- Urata H.,
- Healy B.,
- Stewart R.W.,
- Bumpus M.,
- Husain A.
- Roig E.,
- Perez-Villa F.,
- Morales M,
- et al.
- Struthers A.D.
- Tang W.H.W.,
- Vagelos R.H.,
- Yee Y,
- et al.
- MacFadyen R.J.,
- Lee A.F.C.,
- Morton J.J.,
- Pringle S.D.,
- Struthers A.D.
- Nakajima M.,
- Hutchinson H.G.,
- Fujinaga M,
- et al.
- Packer M.,
- Poole-Wilson P.A.,
- Armstrong P.W,
- et al.
- Hirsch A.T.,
- Talsness C.E.,
- Schunkert H.,
- Paul M.,
- Dzau V.J.
- The ALLHAT Officers and Coordinators for the ALLHAT Collaborative Research Group
- Maggioni A.P.,
- Anand I.,
- Gottlieb S.O.,
- Latini R.,
- Tognoni G.,
- Cohn J.N.
- ↵Available at: http://www.drugstore.com/pharmacy/drugindex/.Accessed May 2004.