Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
International Journal of Cardiology 93 (2004) 105 – 111 www.elsevier.com/locate/ijcard Review Meta-analyses of mortality and morbidity effects of an angiotensin receptor blocker in patients with chronic heart failure already receiving an ACE inhibitor (alone or with a h-blocker) Konstantinos Dimopoulos a,b,*, Tushar V. Salukhe a,b, Andrew J.S. Coats a,c, Jamil Mayet a,d, Massimo Piepoli a,e, Darrel P. Francis a,d b a National Heart and Lung Institute, London, UK Heart Failure Unit, Royal Brompton Hospital, Sydney St., London SW3 6NP, UK c Sydney University Medical School, Sydney, Australia d St. Mary’s Hospital, London, UK e Ospedale Civile di Piacenza, Italy Received 1 October 2003; accepted 9 October 2003 Abstract Background: While treatment with either angiotensin-converting enzyme inhibitors (ACEi) and angiotensin receptor blockers (ARBs) is clearly superior to placebo in the treatment of heart failure patients, controversy still surrounds the effects of ARBs in patients already receiving an ACEi. Even more controversial is the wisdom of administering ARBs in patients already on an ACEi and h-blocker. Methods: We present meta-analyses of the available randomised controlled trials to date (October 2003) of angiotensin II receptor antagonists versus placebo in patients with symptomatic chronic heart failure in which both groups received ACEi. The two largest eligible trials were CHARMAdded and Val-HeFT. We examined two endpoints: mortality and a combined endpoint of mortality and morbidity. Results: In the first metaanalysis, covering all patients regardless of h-blocker use, we found a significant reduction in the combined endpoint (odds ratio [OR] = 0.89; 95% confidence interval [CI] 0.81 – 0.98), but no significant reduction in mortality itself (OR = 0.97; CI: 0.87 – 1.08). In the second metaanalysis, covering patients concomitantly on h-blockers, we found no significant effect on mortality (OR = 1.08; CI: 0.90 – 1.29) or on the combined endpoint (OR = 0.94; CI: 0.82 – 1.10). In the third meta-analysis, covering patients not on concomitant h-blockers, there is clear evidence of a reduction in the combined endpoint (OR = 0.83; CI: 0.73 – 0.94), but not on mortality (OR = 0.93; CI: 0.81 – 1.06). Conclusion: There is now good evidence for the use of ARBs to prevent events in patients with heart failure on ACEi who are not suitable for h-blockers. D 2004 Published by Elsevier Ireland Ltd. Keywords: Meta-analysis; Heart failure; Angiotensin receptor II antagonists 1. Introduction Controversy surrounds the use of angiotensin receptor blockers (ARBs) in patients with chronic heart failure who are already receiving angiotensin converting enzyme inhibitors (ACEi). Theoretically, ARBs can antagonise angiotensin II produced by enzymes other than ACE and therefore could potentially further suppress the effects of overactivation of the renin – angiotensin– aldosterone system. * Corresponding author. Heart Failure Unit, Royal Brompton Hospital, Sydney St., London SW3 6NP, UK. Tel.: +44-780-051-0072; fax: +44-208429-2312. E-mail address: [email protected] (K. Dimopoulos). In particular, in patients already receiving both a hblocker and an ACEi, it is especially controversial whether the effect of adding an ARB is harmful or beneficial. Data is now available from large randomised controlled trials that should help to evaluate whether the theoretical benefit of ARBs in patients already taking ACEi translates into reduced clinical events. From two large studies (ValHeFT, CHARM), there are also data stratifying the patients according to whether they were receiving hblockers. Quantitative synthesis by meta-analysis is the most reliable way of putting together results of large randomised controlled trials. We performed a meta-analysis of trials of ARBs versus placebo in patients receiving ACEi for chronic 0167-5273/$ - see front matter D 2004 Published by Elsevier Ireland Ltd. doi:10.1016/j.ijcard.2003.10.001 IJCA-05331; No of Pages 7 2 Study Population ARBS ACEi Control Primary endpoint Hamroff [3] NYHA III – IV losartan placebo peak VO2, NYHA class RESOLVD [4] NYHA II – IV, 6 MWD < 500 m, EF < 40% candesartan various (enalapril, captopril, fosinopril, lisinopril) enalapril placebo vs. candesartan vs. enalapril Val-HeFT[5,27] II – IV, EF < 40% LVEDD>2.9 cm/m2 valsartan various (enalapril, lisinopril, quinazil, ramipril, captopril) placebo CHARM-Added [6] NYHA III – IV, NYHA II plus hospitalization last 6 months candesartan various (enalapril, lisinopril, captopril, ramipril, etc.) placebo 6-min walk test, EF, LV volumes, neurohormones, QOL, clinical events mortality/mortality and morbidity (hospitalization, SCD with resuscitation, IV vasodilators or inotropes >4 h) mortality or hospitalization Overall Total number of patients Patients in ARB arm Mean follow-up (months) Mortality (%) 6 Mortality per year of follow-up (%) 33 16 6 12.1 441 332 47 7.4 1.9 4644 2326 23 19.5 10.1 2548 1276 41 25.4 7.4 7666 3950 31 21.6 8.6 ACEi=ACE inhibitor; ARBs=angiotensin receptor blocker; EF=ejection fraction; LV=left ventricle; LVEDD=left ventricular end diastolic diameter; NYHA=New York Heart Association; VO2=Oxygen consumption; 6MWD=6-min walk distance; QOL=Quality of Life; SCD=sudden cardiac death. K. Dimopoulos et al. / International Journal of Cardiology 93 (2004) 105–111 Table 1 Randomised controlled trials involving ARBs and ACEi therapy versus placebo plus ACEi, mean treatment duration >6 months, primary outcome mortality, or mortality – morbidity K. Dimopoulos et al. / International Journal of Cardiology 93 (2004) 105–111 3 2.3. Statistical analysis heart failure. We also performed separate meta-analyses of subgroups of patients according to whether they were receiving concomitant beta-blockade. For each of the three meta-analyses (all patients; patients on h-blockers; patients not on h-blockers), we calculated the effects of added ARB on (i) mortality and (ii) a combined endpoint of mortality and morbidity. The odds ratio (OR) method was used to combine the individual study effect sizes into a single overall effect size, expressed as odds ratio and 95% confidence interval (CI).[1] A fixed effect model was used initially, but if there was evidence of heterogeneity by the Cochran Q method [2] ( p < 0.05 by v2), we moved on to the random effects model. 2. Methods We assessed the available randomised controlled trials to date (October 2003) of ARBs versus placebo in patients with symptomatic chronic heart failure in which both groups received ACEi. Our primary endpoint was mortality. Secondarily, we evaluated the effect on the combined endpoint of mortality and morbidity. Since the trials used different definitions of this combined endpoint, we decided that the only unbiased solution was to use the prespecified combined endpoint within the individual trials. 3. Results 3.1. Trial characteristics 2.1. Presence of concomitant b-blocker therapy A total of four randomised controlled studies met the entrance criteria, with a total of 7666 patients (Table 1). All were double-blind, randomised placebocontrolled studies in which comparison between combined ARBs plus ACEi therapy was compared to ACEi alone. Three types of ARBs were being tested in these studies: losartan, valsartan and candesartan. As for the ACEi used as background therapy, in only one trial was the ACEi constant (enalapril in the RESOLVD trial). Patients in the remaining studies were on treatment with various open-label ACEi. In one study (RESOLVD), we considered the ACEi alone and ACEi plus ARB arms only. All-cause mortality was reported in all studies. Combined endpoint expressed as mortality or morbidity was reported in all but one study [3]. Mortality and morbidity characteristics of the subgroups with and without concomitant h-blockers were reported in two studies (Val-HeFT and CHARMAdded). In the RESOLVD and Val-HeFT studies, not all patients were treated with an ACEi, and only the subgroups that were are included in this meta-analysis (Table 2). We performed one analysis (a) pooling all patients irrespective of whether they were on concomitant h-blockade. Where studies reported separate results for patients with and without concomitant h-blockade, we were able to include those studies in additional meta-analyses of (b) patients on concomitant ACEi and h-blocker and (c) patients on concomitant ACEi without h-blocker. 2.2. Study selection A systematic MEDLINE search was performed using key search words: angiotensin receptor antagonist (or blocker), heart failure, angiotensin converting enzyme (or ACE) inhibitor and randomised controlled trial. We specifically identified (i) double-blinded randomised controlled studies of combined ARB and ACEi therapy versus ACEi therapy alone in patients with symptomatic heart failure (NYHA II – IV); (ii) studies with mortality and morbidity as a primary outcome; and (iii) with duration of follow-up of at least 6 months. Data were extracted by two readers independently, and a third reviewer was used in the eventuality of disagreement between the two. Table 2 Baseline demographic data and clinical characteristics Study Mean age (years) Male (%) NYHA Class II/III/IV (mean) Mean LVEF (%) Ischemic/idiopathic etiology (%) % on digoxin % on diuretic % on h-blocker Hamroff RESOLVD Val-HeFT CHARM -Added Overall 61 63 63 64 49 86 80 79 NA 63:35:2 (2.3) 62:36:2 (2.3) 24:73:3 (2.7) 26 28 27 28 30:70 71:18 57:31 62:26 97 68 67 58 100 85 85.5 90 6 15 35 55 63.2 79.8 49.3:48.3:2.3 (2.5) 25.6 59.6:28.8 64.5 NYHA = New York Heart Association, LVEF = left ventricular ejection fraction, NA = not available. 87.0 40.5 4 K. Dimopoulos et al. / International Journal of Cardiology 93 (2004) 105–111 Fig. 1. Effect of ARBs on mortality and the combined endpoint mortality – morbidity in patients with heart failure taking ACE inhibitors, regardless of h-blocker therapy. The OR and 95% confidence interval are shown for each study and overall. 3.2. ARB versus no ARB in patients taking ACEi (regardless of b-blocker therapy) A total of 3950 patients were randomised to combined ARB and ACEi therapy, whereas 3716 patients were randomised to ACEi plus placebo. Mean age was 63.2 years and 79.8% were men. Mean NYHA class was 2.5, and 59.6% had heart failure of ischemic aetiology. Only 40.5% were on h-blocker therapy. All-cause mortality was not significantly affected by ARB treatment (OR = 0.97; 95% CI: 0.87 –1.08) (Fig. 1). The combined endpoint showed subtle evidence of benefit from ARB treatment in the fixed effects model (OR = 0.89; 95% CI: 0.81 –0.98). There was no evidence of heterogeneity with either endpoint. Fig. 3. Effect of ARBs on mortality and the combined endpoint of mortality – morbidity in patients with heart failure taking ACEi but not hblockers. The odds ratio and 95% confidence interval are shown for each study and overall. 3.3. ARB versus no ARB in patients taking ACEi and b-blocker A total of 3163 patients were on h-blocker therapy, of which 1569 were randomised to ARB and 1594 to no ARB in the Val-HeFT and CHARM-Added trials. Mortality in the two groups was 23.3% and 24.1%, respectively. No significant difference in mortality was seen between the two groups by the fixed effects model (OR = 1.08; 95% CI: 0.90 –1.29, Fig. 2). Because of evidence of heterogeneity [v2 = 6.8, p < 0.05], the random effects model was also calculated, which confirmed no significant treatment effect (OR = 1.12; 95% CI: 0.69– 1.81). Combined mortality and morbidity was also not significantly different (OR = 0.94; 95% CI: 0.80– 1.10) by the fixed effects model. Because of evidence of heterogeneity [v2 = 9.2, p < 0.05], the random effects model was also calculated, and this confirmed no significant treatment effect (OR = 0.95; 95% CI: 0.59– 1.54 ). 3.4. ARB versus no ARB in patients taking ACEi but not b-blocker A total of 4029 patients were in the non-h-blockers subgroups of the Val-HeFT and CHARM-Added trials. Mortality was not significantly different between the ARB and non-ARB arms (OR = 0.93; 95% CI: 0.81 – 1.06, fixed model; no evidence of heterogeneity). The combined endpoint of mortality and morbidity was significantly reduced in the ARB arm (OR = 0.83; 95% CI: 0.73– 0.94, fixed model; no evidence of heterogeneity) (Fig. 3). Fig. 2. Effect of ARBs on mortality and the combined endpoint of mortality – morbidity in patients with heart failure taking ACE inhibitors and h-blockers. The odds ratio and 95% confidence interval are shown for each study and overall. 4. Discussion In these meta-analyses, we have found that ARBs have a beneficial effect in patients who are taking ACEi, as long as K. Dimopoulos et al. / International Journal of Cardiology 93 (2004) 105–111 they are not taking h-blockers. The reduction in the combined endpoint is not accompanied by reduction in mortality, suggesting that the benefit might be largely through reduction in hospitalization. The deleterious effects of neurohormonal activation in the context of chronic heart failure have been the dominant treatment targets of recent decades. High levels of catecholamines, renin and aldosterone have negative effects on the cardiovascular system producing progressive impairment of left ventricular structure and function as well as vascular damage.[5,7 – 11] ACEi are clearly beneficial for prognosis, even though doubts remain on optimal dosage and interactions with other drugs.[12 13] However, ACEi alone, even if at full recommended doses, does not provide complete inhibition of the renin – angiotensin converting enzyme system (RAS), thus permitting angiotensin to exert its negative systemic effects. [14 – 17]. ARBs, on the other hand, have been shown to be safe and have similar efficacy to ACEi in heart failure. [18 –21] A combined approach using both ACEi and ARBs could provide a more effective blockade of the RAS than either drug alone. [22] Moreover, the beneficial effects of ACEi and ARBs may go beyond inhibition of the RAS [23 24]. ACEi increase kinin levels since ACE is also a potent kininase. ARBs may enhance the activation of ATII receptors by inducing circulating levels of angiotensin, thus triggering kinin generation and nitric oxide production. [23 24]. 4.1. Studies of ARB on a background of ACEi The first clinical trial to compare ACEi treatment to the combination ACEi plus ARBs in patients with heart failure was published by Hamroff et al. Even though the study was underpowered for survival analysis, a clear beneficial effect on the primary endpoints of exercise capacity and functional class was seen after a mean observation time of 6 months, suggesting an additive effect of ACEi and ARBs [3]. The RESOLVD pilot study showed a beneficial effect of the combination strategy on blood pressure, left ventricular dimensions and neurohormonal factors (renin, epinephrine, N-terminal pro-atrial natriuretic peptide). [4] The ACEi plus ARB arm also showed the greatest increase in renin and greatest decrease in aldosterone, suggesting a more effective blockage of the RAS. No significant differences were seen in mortality and morbidity, leaving doubts on the long-term survival benefits of the combination therapy. It is nevertheless noteworthy that the ACEi arm consisted of only 109 patients and a significantly lower proportion of patients took h-blockers in the ARB arm than in the non-ARB arm, possibly affecting survival analysis in this study. The Val-HeFT Study compared valsartan to placebo in heart failure patients, the vast majority of which were on ACEi. [5] Even though a large population of patients was included and mean observation period was nearly 2 years, a significant effect of the combination therapy was only 5 seen on morbidity expressed as hospitalization, whereas mortality was similar in the two groups. An unexpected increase in mortality emerged during post hoc analysis in the subgroup of patients on triple therapy (ARB, ACEi and h-blocker): 129 versus 97 deaths, HR 1.42, 95% CI: 1.09– 1.85. Doubts were raised on whether this was a true harmful effect of excessive neurohormonal blockade by the triple therapy or a manifestation of the random play of chance. The CHARM-Added Study was recently published, comparing candesartan to placebo in a large population of heart failure patients already on ACEi. It was large and had a long mean observation period of more than 3 years and was thus potentially capable of answering the questions that Val-HeFT had raised. There was a significant decrease in the primary endpoint of death or hospitalization for worsening heart failure in the ARB plus ACEi arm, which had not been seen in Val-HeFT. In mortality, however, there was no significant difference. Subgroup analysis showed that the addition of h-blockers to the combination ACEi-ARB as well as administration of the maximum recommended doses of ACEi were the two factors that conferred the maximum benefit. Paradoxically, therefore, the CHARM-Added results favoured triple therapy (ACEi-ARBS-h-blocker), while ValHeFT had suggested that such combination could be harmful. The present meta-analysis was largely dependent on these two discordant studies since, as the largest two trials ever on this question, they were responsible for the vast majority of events. 4.2. Quantitative synthesis by meta-analysis Across the trials as a whole, all-cause mortality is not significantly affected by ARB. There is a borderline reduction in combined endpoint, which is significant by the fixed effect model, but nonsignificant in the random effects model, which is more appropriate in a meta-analysis with this degree of heterogeneity. Our conclusion is therefore that at present, there is not good endpoint-based evidence for routine administration of ARB across all patients receiving ACEi for heart failure. Within the subgroup taking h-blockers, there is no evidence of reduced mortality nor of reduce combined endpoint. The apparent increase in mortality seen in ValHeFT is neutralised by the nonsignificant trend to reduce mortality in CHARM-Added. In a mirror-image situation, the reduced combine endpoint in CHARM-Added is neutralised by a nonsignificant trend to increased combined endpoint in Val-HeFT. We conclude that there are no grounds to recommend routine ARBs receiving both ACEi and h-blockers. Those patients not taking h-blockers, however, had clear evidence of reduction, the combined endpoint of morbidity and mortality when ARB was added to a background of ACEi, although mortality itself was not significantly re- 6 K. Dimopoulos et al. / International Journal of Cardiology 93 (2004) 105–111 duced. We can therefore conclude that for patients not taking h-blockers, adding an ARB reduces the clinical events, largely (or entirely) by reducing hospital admissions/morbidity. Our meta-analysis thus furthers the previous meta-analysis by Jong et al. [25], since data newly available (CHARM-Added) demonstrate a significant benefit of double treatment on the combined endpoint of mortality and morbidity. 4.3. Possible mechanisms Why does adding ARBs to ACEi only seem to help those patients not on h-blockers? It could be a consequence of an interaction between AT1 receptors and h-adrenergic receptors.[26] It has been shown that selective AT1 receptor blockade inhibits downstream signalling of h-adrenergic receptors, thus producing a h-blocker effect. It may be that the benefit derives mainly from the h-blocking properties of ARBs. Why was there such a contrast between the results of ValHeFT and CHARM-Added in the subset of patients on hblockers? Differences in baseline characteristics of these two large trials that could explain the difference. First, being a more recent study, CHARM-Added had a significantly higher proportion of patients on h-blockers than Val-HeFT (Table 1). Second, the vast majority of patients in the ValHeFT were in NYHA class II, whereas the majority of patients in CHARM-Added were in class III. The sicker CHARM-Added patients may have had more neurohormonal activation and therefore more opportunity to benefit from a more complete neurohormonal blockade provided by the triple therapy. Our meta-analysis combines the two populations, producing a more homogeneous distribution between the two main functional classes (II and III, 49.3% and 48.3%, respectively) of chronic heart failure. Perhaps triple therapy may in the future be proved beneficial in the more impaired patients. More studies would be required to clarify this question. 4.4. Study limitations The key limitation was the small number of studies available on combined ACEi and ARBs treatment. Moreover, there was evidence of heterogeneity in some of the analyses. Nevertheless, the overall population was large enough to permit analysis and identify certain key results, and recalculations (as presented in Results) using a randomeffects model in those analyses showing heterogeneity did not affect the conclusions. Secondly, three different ARBs were used in the studies included in this meta-analysis, and only one used a uniform ACEi, whereas the rest used several ACEi, as long as treatment was optimal. We are therefore indirectly considering ACEi to have a class effect. While this viewpoint would be widely held by clinicians to be valid, there is no proof of a uniform class effect. Thirdly, the proportion of patients on h-blockers varied depending on the age of the study, from 6% in the Hamroff et al., to 55.4% in the most recent CHARM-Added. Fourthly, the prespecified combined mortality and morbidity endpoint differed between trials. Specifically, the morbidity element varied significantly. As detailed data to standardise morbidity was not available, the only unbiased solution was to use the prespecified combined endpoint definition. Finally, as pointed out above, the studies differed in their distributions of functional classes of the patients. CHARMAdded seems to have included patients with greater degree of impairment than Val-HeFT and SOLVD. However, this adds breadth to the range of patients studied. 4.5. Conclusions There is now satisfactory evidence that patients with heart failure who are taking ACEi but not h-blockers would benefit from routine administration of an ARB. The evidence supporting this is a significant reduction in combined morbidity – mortality, without evidence of reduced mortality. For those already taking both ACEi and h-blockers, there is not convincing evidence of benefit nor of harm from giving additional ARBs. We are in equipoise for both mortality and combined morbidity –mortality. Beta-blockade can no longer be considered a contraindication for combined ACEi-ARB therapy. Acknowledgements Dr. Dimopoulos is supported by the European Society of Cardiology; Dr. Salukhe is supported by the Robert Luff Fellowship; Dr. Francis was supported by the British Heart Foundation. References [1] Hedges LV, Olkin I. Statistical methods for meta-analysis. San Diego: Academic Press; 1985. Chapter 5. [2] Cochran WG. The combination of estimates from different experiments. Biometrics 1954;10:101 – 29. [3] Hamroff G, Katz SD, Mancini D, et al. Addition of angiotensin II receptor blockade to maximal angiotensin-converting enzyme inhibition improves exercise capacity in patients with severe congestive heart failure. Circulation 1999;99:990 – 2. [4] McKelvie RS, Yusuf S, Pericak D, et al. Comparison of candesartan, enalapril, and their combination in congestive heart failure: randomized evaluation of strategies for left ventricular dysfunction (RESOLVD) pilot study. The RESOLVD Pilot Study Investigators. Circulation 1999;100:1056 – 64. [5] Cohn JN, Tognoni G. A randomized trial of the angiotensin-receptor blocker valsartan in chronic heart failure. N Engl J Med 2001;345: 1667 – 75. [6] McMurray JJ, Ostergren J, Swedberg K. Effects of candesartan in patients with chronic heart failure and reduced left-ventricular systolic K. Dimopoulos et al. / International Journal of Cardiology 93 (2004) 105–111 [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] function taking angiotensin-converting-enzyme inhibitors: the CHARM-Added trial. Lancet 2003;362:767 – 71. Dzau VJ. Tissue renin – angiotensin system in myocardial hypertrophy and failure. Arch Intern Med 1993;153:937 – 42. CIBIS-II Investigators and Committees. The Cardiac Insufficiency Bisoprolol Study II (CIBIS-II): a randomised trial. Lancet 1999; 353:9 – 13. Packer M, Bristow MR, Cohn JN, et al. The effect of carvedilol on morbidity and mortality in patients with chronic heart failure. N Engl J Med 1996;334:1349 – 55. MERIT-HF Study Group. Effect of metoprolol CR/XL in chronic heart failure: metoprolol CR/XL randomised intervention trial in congestive heart failure (MERIT-HF). Lancet 1999;353:2001 – 7. Pitt B, Zannad F, Remme WJ, et al. The effect of spironolactone in morbidity and mortality in patients with severe heart failure. N Engl J Med 1999;341:709 – 17. Williams SG, Cooke GA, Wright DJ, Tan LB. Disparate results of ACE inhibitor dosage on exercise capacity in heart failure: a reappraisal of vasodilator therapy and study design. Int J Cardiol 2001; 77:239 – 45. Harjai KJ, Solis S, Prasad A, Loupe J. Use of aspirin in conjunction with angiotensin-converting enzyme inhibitors does not worsen longterm survival in heart failure. Int J Cardiol 2003;88:207 – 14. Rousseau MF, Konstam MA, Benedict CR, et al. Progression of left ventricular dysfunction secondary to coronary artery disease, sustained neurohormonal activation and effects of ibopamine therapy during long-term therapy with angiotensin-converting enzyme inhibitor. Am J Cardiol 1994;73:488 – 93. Francis GS, Cohn JN, Johnson G, Rector TS, Goldman S, Simon A. the V-HeFT VA Cooperative Studies Group. Plasma norepinephrine, plasma renin activity, and congestive heart failure: relations to survival and the effects of therapy in V-HeFT II. Circulation 1993;87(Suppl VI):VI-40 – 8. Jorde UP, Ennezat PV, Lisker J, Suryadevara V, Infeld J, Cukon S, 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. 7 [17] St. John Sutton M, Pfeffer MA, Moye L, et al. Cardio-vascular death and left ventricular remodeling two years after myocardial infarction: baseline predictors and impact of long-term use of captopril: information from the Survival and Ventricular Enlargement (SAVE) trial. Circulation 1997;96:3294 – 9. [18] Willenheimer R, Helmers C, Pantev E, Rydberg E, Lofdahl P, Gordon A. Safety and efficacy of valsartan versus enalapril in heart failure patients. Int J Cardiol 2002;85:261 – 70. [19] Pitt B, Poole-Wilson PA, et al. Effect of Losartan compared with captopril on mortality in patients with symptomatic heart failure: randomised trial—the Losartan Heart Failure Survival Study ELITE II. Lancet 2000;355:1582 – 7. [20] Sharma D, Buyse M, Pitt B, Rucinska EJ. Meta-analysis of observed mortality data from all-controlled, double-blind, multiple-dose studies of losartan in heart failure. Losartan Heart Failure Mortality Metaanalysis Study Group. Am J Cardiol 2000;85:187 – 92. [21] Coats AJ. Angiotensin receptor blockers—finally the evidence is coming in: IDNT and RENAAL. Int J Cardiol 2001;79:99 – 102. [22] Azizi M, Chatellier G, Guyene TT, et al. Additive effects of combined angiotensin-converting enzyme inhibition and angiotensin II antagonism on blood pressure and renin release in sodium-depleted normotensives. Circulation 1995;92:825 – 34. [23] Wollert KC, Drexler H. The renin – angiotensin system and experimental heart failure. Cardiovasc Res 1999;43:838 – 49. [24] Yu G, Liang X, Xie X, Su M, Zhao S. Diverse effects of chronic treatment with losartan, fosinopril, and amlodipine on apoptosis, angiotensin II in the left ventricle of hypertensive rats. Int J Cardiol 2001;81:123 – 9. [25] Jong P, Demers C, McKelvie RS, Liu PP. Angiotensin receptor blockers in heart failure: meta-analysis of randomized controlled trials. Am Coll Cardiol 2002;39:463 – 70. [26] Barki-Harrington L, Luttrell LM, Rockman HA. Dual inhibition of {beta}-adrenergic and angiotensin II receptors by a single antagonist: a functional role for receptor – receptor interaction in vivo. Circulation 2003;108:r79 – 86. [27] J.N. Cohn,Val-HeFT efficacy subgroup analyses (presentation). US Food and Drug Administration Web site. Available at www.fda.gov/ ohrms/dockets/ac/01/slides/3793s1_03_efficacy%20 subgroups.pdf.