© 2000 European Society of Cardiology
Update of ELITE-II, BEST, CHAMP, and IMPRESS clinical trials in heart failure
The University of Hull, Department of Cardiology, Academic Unit of Cardiology, Castle Hill Hospital Kingston upon Hull, HU16 5JQ, UK
* Corresponding author. Tel.: +44-1482-624087; fax: +44-1482-624085.
 |
Abstract |
|---|
 Top Abstract 1. Evaluation of losartan...2. Beta-blocker evaluation...3. Combination hemotherapy and...4. Inhibition of metaloprotease...References |
|---|
The ELITE-II, BEST and CHAMP Trials were reported for the first time at the American Heart Association in November 1999. These trials provide valuable new information to guide clinical practice in the management of heart failure and of myocardial infarction, although none mandate a major change from current clinical practice. The IMPRESS trial of the vasopeptidase inhibitor, omapatrilat, indicated a promising new treatment for the management of heart failure.
Received December 13, 1999; Revised December 17, 1999; Accepted December 17, 1999
|
1. Evaluation of losartan in the elderly (ELITE II) [1] |
|---|
|
TopAbstract1. Evaluation of losartan...2. Beta-blocker evaluation...3. Combination hemotherapy and...4. Inhibition of metaloprotease...References |
|---|
ELITE II, a sequel to the ELITE study [2], was a multicentre double-blind, randomised, parallel group study comparing losartan and captopril in patients with symptomatic heart failure (HF). The first ELITE study was designed to investigate the effects of these agents on renal function. It showed no difference in the effect of losartan and captopril on renal dysfunction over 48 weeks, but did show a 46% (P<0.05) reduction in all-cause mortality and a 64% reduction in sudden death, although this was based on only a small number (49) of deaths. Also, mortality was only pre-specified as an outcome after patient recruitment was complete and, although it was pre-specified, it was not identified as an outcome of primary or secondary interest.
ELITE II was designed to properly address the question of whether losartan was superior to captopril in reducing all-cause mortality. The secondary endpoint was sudden death and/or resuscitated cardiac arrest. Other endpoints were the combined outcome of all-cause mortality and all-cause hospitalisation, safety and tolerability. The study was designed with 90% power to detect a 25% difference in all-cause mortality. The study was event-driven (expecting 510 deaths over 1.5 years).
Patients with NYHA II–IV heart failure who were aged >60 years and who had an ejection fraction <40% were eligible to participate. Investigators were discouraged from recruiting patients who had been treated previously with ACE inhibitors or angiotensin receptor antagonists.
1.1. Results
Patients (n=1574) were randomised to Captopril 50 mg, three times daily and 1578 to Losartan 50 mg daily. The clinical characteristics of the groups were well matched. The outcome for the primary end-point and other key clinical outcomes are shown in Table 1.
|
There was no significant difference in mortality or the combined end-points of (a) sudden cardiac death or resuscitated cardiac arrest; or (b) all-cause mortality or all-cause hospitalisation (time to first event) over a median follow-up time of 1.5 years, although trends were consistently in favour of captopril. However, as in ELITE, losartan was significantly better tolerated than captopril with 14.5% of patients withdrawing from captopril compared to only 9.4% on losartan (P<0.001). Full interpretation of these results awaits the formal publication of the trial, however, a number of comments can be made at this stage.
Firstly, the population studied was at only moderate risk for patients with heart failure, with a crude annualised mortality of approximately 10%. The combined risk of death or hospitalisation during a median follow-up of 1.5 years was <50%. The relatively good prognosis probably reflects the inclusion of many NYHA class II patients, the rather liberal ejection fraction cut-off as well as the efficacy of treatment (35% of patients were on a beta-blocker). However, the SOLVD-treatment study [3] also included mainly patients with NYHA class II heart failure and was able to show an effect in this patients group, demonstrating that effective treatments can be shown to modify prognosis in this population. Low-risk alone does not explain the lack of difference.
Could ELITE-II have been a falsely neutral result that underestimated the true benefits of losartan? This is a possibility. It is difficult to be sure that equally effective doses of losartan and captopril were being used. A higher dose of losartan could have been more effective even though haemodynamic studies do not support such a view [4]. The ELITE-II study does not exclude the possibility that losartan is superior to captopril and that the inability to show superiority occurred by chance, although it is unlikely that the benefit could be as large as the hypothesised 25% reduction in mortality.
Could ELITE-II have been a falsely neutral result that underestimated the true benefit of ACE inhibition? There are several reasons to suspect that this could be so. Firstly, more patients withdrew from captopril than losartan. These patients could not benefit from ACE inhibition and this would be expected to disadvantage the ACE inhibitor group. In clinical practice this means that in ACE inhibitor tolerant patients ACE inhibitors might perform better than the ELITE-II study suggests. Secondly, many physicians believe (without conclusive evidence) that long-acting ACE inhibitors may be superior to captopril. Thirdly, it is possible that compliance with captopril or its matching placebo (taken three times daily) was inferior to that with losartan or its matching placebo (once daily). This could also bias the study against captopril.
The ELITE-II study was not powered for equivalence and does not prove that losartan and captopril are equally effective. Consequently, losartan cannot be considered an alternative to an ACE inhibitor with improved tolerability given the present information. The RESOLVD study also suggested that an angiotensin receptor blocker, candesartan, was not superior, and possibly inferior, to an ACE inhibitor, enalapril [5]. We are not aware of any other large study comparing ACE inhibitors and angiotensin receptor blockers for patients with heart failure and left ventricular systolic dysfunction. A number of studies are comparing ACE inhibitors with angiotensin receptor blockers in patients with post-infarction left ventricular dysfunction and/or heart failure. The combined power of these studies may allow the question of equivalence between ACE inhibitors and angiotensin receptor blockers to be resolved.
Another question that ELITE-II raises is whether losartan is superior to placebo for the treatment of heart failure. Some studies have suggested that angiotensin receptor blockers improve symptoms, reduce hospitalisation for worsening heart failure and improve prognosis while other studies have failed to demonstrate benefit for these same outcomes [6]. Overall, the data suggest that angiotensin receptor blockers are probably superior to placebo. Conclusive evidence to determine whether this is the case should come from the CHARM study [7].
A number of studies are investigating the possibility that angiotensin receptor blockers give benefit in addition to those achieved by ACE inhibitors alone. A number of small, single-centre studies have suggested a benefit [6]. However, the RESOLVD study showed mixed results with a trend to increased mortality with the combination but a greater effect on left ventricular remodelling [5].
In conclusion, the ELITE-II study has only partially clarified the role of angiotensin receptor blockers for the management of heart failure. For the moment, the current recommendation only to use angiotensin receptor blockers when the patient cannot tolerate an ACE inhibitors stands, although this recommendation also awaits conclusive supportive evidence.
|
2. Beta-blocker evaluation survival trial (BEST) [8] |
|---|
|
TopAbstract1. Evaluation of losartan...2. Beta-blocker evaluation...3. Combination hemotherapy and...4. Inhibition of metaloprotease...References |
|---|
The BEST study was a double-blind, placebo-controlled trial testing the hypothesis that the addition of the beta-blocker bucindolol to standard medical therapy reduced all-cause mortality in patients with severe (NYHA Class III–IV) heart failure and left ventricular systolic dysfunction. Secondary end-points included cardiovascular mortality, hospitalisation, death or heart transplantation, left ventricular ejection fraction at 3 and 12 months, myocardial infarction, quality of life scores and changes in the requirement for concomitant heart failure therapy.
Patients were randomised to treatment stratified according to the presence or absence of CHD, LVEF, gender and race. The study was conducted predominantly in US Veterans’ Hospitals and therefore included relatively few women and a relatively large number of African-Americans. Bucindolol was uptitrated over 6–8 weeks from 3 mg to 50 mg twice daily (if weight <75 kg) or 100 mg (if weight >75 kg). On 29 July 1999, the study was prematurely terminated on the recommendation of the Data and Safety Monitoring Board that felt that study continuation was unwarranted in the light of the results of the trials of a number of beta-blockers including the US Carvedilol Trial [9], CIBIS-II (bisoprolol) [10] and MERIT (metoprolol) [11].
2.1. Results
Twenty-three percent of the patients were African-Americans and 22% were women. Ninety-two percent were in NYHA Class III and 8% in Class IV. For patients randomised to bucindolol, 92% were on some dose of bucindolol at the end of titration and 77% were taking bucindolol by the end of the study.
Bucindolol significantly reduced systemic adrenergic activity as reflected by changes in plasma norepinephrine. Although the mean plasma norepinephrine levels were slightly higher in the bucindolol group at baseline, there was an average reduction of 70 pg/ml at 3 months, and 18 pg/ml at 12 months with bucindolol (P=0.0001). In the bucindolol group left ventricular ejection fraction (LVEF) improved by an average of 5.5% (absolute) at 3 months, and 7.3% (absolute) by 12 months (P=0.0001). Therefore, compared to placebo, bucindolol produced the expected reduction in systemic adrenergic activity and increased LVEF.
Mortality was reduced by 10% (relative) with bucindolol compared to placebo (Z value=1.60, P=0.109 unadjusted; total events 856). Although this was a beneficial trend favouring bucindolol, it did not reach statistical significance. Bucindolol reduced cardiovascular deaths by 12.5%, a statistically significant reduction, and this benefit seemed equally distributed among pump failure deaths, sudden deaths or deaths due to myocardial infarction. The number of non-cardiovascular deaths was similar in both groups. All cause hospitalisation tended to decrease with bucindolol but the change was not significant (64% placebo vs. 61% bucindolol). However, the number of patients hospitalised for a heart failure-related illness was significantly reduced by 16.7% in the bucindolol group (42% on placebo vs. 35% on bucindolol; P=0.001). The endpoint of death/transplantation was reduced by 10.3% with bucindolol, but this difference was not significantly different (35% on placebo vs. 31.6% on bucindolol).
Subgroup analysis showed a heterogeneous response. Patients with NYHA Class III heart failure or left ventricular ejection fraction >20% showed improved survival with bucindolol. In contrast, patients with NYHA Class IV heart failure or ejection fraction <20% did not appear to benefit. Effects were similar in patients with or without coronary disease. There was a statistically significant survival benefit in the non-African-American subgroup. By contrast, the African-American subgroup had a 17% excess mortality (not significantly different from placebo on subgroup analysis) suggesting a lack of benefit, and a possible harmful effect. Men tended to benefit more than women.
Although the BEST study did not prove its primary hypothesis the study showed a number of positive trends such that adding this study to that of other beta-blockers strengthens the overall hypothesis that beta-blockers reduce mortality in heart failure. However inclusion of the BEST data will reduce the calculated magnitude of benefit with this class of agents.
The results of BEST raise a number of issues. Had BEST been conducted in a predominantly Caucasian population with NYHA II/III heart failure, as was the the case for MERIT and CIBIS-II, it probably would have achieved its primary outcome. The question of racial differences in the response to beta-blockers is important. Differences in compliance, withdrawal for adverse events, an intrinsic lack of response and the play of chance will all have to be considered as explanations. BEST also calls into question the wisdom of treating very advanced heart failure with beta-blockers. Fortunately, a large study of carvedilol (COPERNICUS) is being conducted in this population, which will provide more information [12]. Probably, the most important question that BEST raises is whether there are real differences in efficacy among the class of beta-blocking agents for the treatment of heart failure. It is unlikely that meta-analysis comparing metoprolol, bisoprolol, carvedilol and bucindolol will show significant heterogeneity of effect. Previous meta-analyses of trials of heart failure and for myocardial infarction that have grouped beta-blocking agents by their pharmacological properties suggested that non-selective agents without intrinsic sympathomimetic activity (ISA) may be superior to other agents in this class but the analyses are far from conclusive [13–16]. While bucindolol is certainly non-selective, uncertainty exists about the extent of its ISA in humans. If bucindolol does have appreciable ISA in human tissues this could account for lesser benefit with this agent. Thus, BEST sets the stage for another very important study, COMET [12], that is a head-to-head comparison between metoprolol and carvedilol.
|
3. Combination hemotherapy and mortality prevention (CHAMP) |
|---|
|
TopAbstract1. Evaluation of losartan...2. Beta-blocker evaluation...3. Combination hemotherapy and...4. Inhibition of metaloprotease...References |
|---|
CHAMP was a randomised, open-label, multi-centre trial comparing long-term treatment with a combination of aspirin (ASA) and warfarin compared to aspirin alone in survivors of acute myocardial. The aim was to detect a 15% reduction in long-term, all-cause mortality with the combination compared to aspirin monotherapy. Patients were randomised within 14 days of an AMI to either ASA 162 mg/daily (chewable therapy) or ASA 81 mg/day plus warfarin titrated to an INR of 1.5–2.5 IU. The lowest dose of aspirin used in any other substantial placebo-controlled, long-term post-infarction trial was 300 mg/day [17], making the results of the present study potentially difficult to interpret. The WARIS study [18] used a target INR of 2.8–4.8. The primary end-point was all cause mortality, secondary endpoints included non-fatal myocardial infarction, non-fatal stroke, vascular mortality and major haemorrhage.
3.1. Results
Patients (n=20 036) were screened and 5059 patients entered (25%) the study. All-cause mortality was approximately 17% overall (no difference between groups), non-fatal stroke approximately 3% overall (no difference between groups) and non-fatal MI approximately 14% overall (no difference between groups) (Tables 2 and 3).
|
|
3.2. Conclusions
There was no benefit with the addition of low-dose warfarin to low-dose aspirin on mortality or morbidity in patients who survived an AMI but the combination was associated with an increased risk of haemorrhage, although rates of haemorrhage were relatively low at least in younger patients. The risks of haemorrhage increased markedly with age in both groups.
The CHAMP study did not compare warfarin with aspirin. It may well be that the combination of warfarin and aspirin is inferior to warfarin alone. Also, in the CHAMP study most patients did not achieve the target INR. This probably reflected fears about the combination of the use of warfarin and aspirin. Further studies directly comparing these two drugs, especially in high risk groups, who are suspected not to benefit from aspirin, such as those with heart failure, are warranted.
|
4. Inhibition of metaloprotease by BMS186716 in a randomised exercise and symptoms study in subjects with heart failure (The IMPRESS trial) |
|---|
|
TopAbstract1. Evaluation of losartan...2. Beta-blocker evaluation...3. Combination hemotherapy and...4. Inhibition of metaloprotease...References |
|---|
Omapatrilat is a vasopeptidase inhibitor, a class of agents designed to inhibit both the angiotensin converting enzyme and neutral endopeptidase for use in hypertension and heart failure. Vasopeptidase inhibitors may be expected not only to reduce the production of angiotensin II but also to inhibit the degradation of bradykinin and natriuretic peptides. Thus, these agents may have all the benefits of ACE inhibitors but also increase natriuresis, reduce vascular tone and have direct effects on tissues, including inhibition of vascular smooth muscle proliferation. However, they may also increase plasma concentrations of endothelin, a potentially adverse effect, because endothelins are also substrates for neutral endopeptidase. Omapatrilat is a particularly potent ACE inhibitor and this must be taken into account when assessing its effects.
The IMPRESS trial randomised 573 patients with NYHA class II–IV heart failure to omapatrilat 40 mg/day (n=289) or lisinopril 20 mg/day (n=284). The mean LVEF was 28% and 63% were in NYHA class II. All patients were on ACE inhibitors prior to randomisation and 30% were on beta-blockers. Follow-up was for 24 weeks. Pre-determined end-points were exercise time, NYHA class and a combined end-point of mortality or worsening heart failure leading to hospitalisation or discontinuation of study medication.
4.1. Results
Exercise capacity increased equally in both groups. As all patients were already on an ACE inhibitor at baseline this probably reflects a placebo effect. NYHA class tended to improve more on omapatrilat although the effect was only significant on subset analysis of NYHA class III/IV patients. Sixteen patients randomised to omapatrilat and 29 randomised to lisinopril died or had worsening heart failure as defined above (risk ratio 0.52; P<0.04). Fewer patients had adverse outcomes such as marked worsening of renal function on omapatrilat.
4.2. Conclusions
Omapatrilat is the first in a class of promising new agents for the treatment of heart failure. However, definitive studies are required. One is underway that will recruit 4400 patients with heart failure, the OVERTURE study, comparing omapatrilat and an ACE inhibitor. However, such studies will not address all the questions.
Omapatrilat is a very effective ACE inhibitor. The ATLAS study [19] shows that dose is a determinant of the efficacy of ACE inhibitors. Studies with neutral endopeptidase inhibitors alone or in combination with ACE inhibitors [20–25] have shown mixed results with the largest study reported so far [20] showing no obvious benefit. Novel study designs will need to be adopted in order to show that simply increasing the dose of the ACE inhibitor does not replicate the effects of omapatrilat. As with angiotensin receptor blockers, doctors should not prescribe because the hypotheses are attractive but rather because evidence of benefit exists.
|
References |
|---|
|
TopAbstract1. Evaluation of losartan...2. Beta-blocker evaluation...3. Combination hemotherapy and...4. Inhibition of metaloprotease...References |
|---|
- Pitt B., Poole-Wilson P.A., Segal R., et al. Effects of losartan versus captopril on mortality in patients with symptomatic heart failure: rationale, design and baseline characteristics of patients in the Losartan Heart Failure Survival Study — ELITE II. J Card Fail (1999) 5:146–154.[CrossRef][Web of Science][Medline]
- Pitt B., Segal R., Martinez F.A., et al. Randomised trial of losartan versus captopril in patients over 65 with heart failure (Evaluation of losartan in the elderly study, ELITE). Lancet (1997) 349:747–752.[CrossRef][Web of Science][Medline]
- Yusuf S. Effect of enalapril on survival in patients with reduced left ventricular ejection fractions and congestive heart failure. New Engl J Med (1991) 325:293–302.[Abstract]
- Crozier I., Ikram H., Awan N., et al. Losartan in heart failure: hemodynamic effects and tolerability. Circulation (1995) 91:691–697.
[Abstract/Free Full Text] - McElvie R.S., 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. Circulation (1999) 100:1056–1064.
[Abstract/Free Full Text] - Cleland J.G.F., Alamgir F. Angiotensin II receptor blockers for heart failure: the current state of play. In: Heart failure in clinical practice—McMurray J.J.V., Cleland J.G.F., eds. (1999) London: Dunitz. 197–224.
- Swedberg K., Pfeffer M., Granger C., et al. Candesartan in heart failure — assessment of reduction in mortality and morbidity (CHARM): rationale and design. J Card Fail (1999) 5:276–282.[CrossRef][Medline]
- The BEST Steering Committee. Design of the beta-blocker evaluation survival trial (BEST). Am J Cardiol (1995) 75:1220–1223.[CrossRef][Web of Science][Medline]
- Packer M., Bristow M.R., Cohn J.N., et al. The effect of carvedilol on morbidity and mortality in patients with chronic heart failure. New Engl J Med (1996) 334:1349–1355.
[Abstract/Free Full Text] - CIBIS-II Investigators and Committee. The Cardiac insufficiency Bisoprolol Study II (CIBIS-II): a randomised trial. Lancet (1999) 353:9–13.[CrossRef][Web of Science][Medline]
- 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) 333:2001–2007.
- McGowan J., Murphy R., Cleland J.G.F. Carvedilol for heart failure; clinical trials in progress. Heart Fail Rev (1999) 4:89–95.[CrossRef]
- Lechat P., Packer M., Chalon S., et al. Clinical effects of
-adrenergic blockade in chronic heart failure. A meta-analysis of double-blind placebo-controlled, randomised trials. Circulation (1998) 98:1184–1191.[Abstract/Free Full Text] - Heidenreich P.A., Lee T.T., Massie B.M. Effect of beta-blockade on mortality in patients with heart failure: a meta-analysis of randomised trials. J Am Coll Cardiol (1997) 30:27–34.[Abstract]
- Cleland J.G.F., Freemantle N., McGowan J., et al. The evidence for beta-blockers equals or surpasses that for ACE inhibitors in heart failure. Br Med J (1999) 318:824–825.
[Free Full Text] - Freemantle N., Cleland J.G.F., et al. What is the place of beta-blockade in secondary prevention after myocardial infarction in the late 1990’s. Br Med J (1999) 318:1730–1737.
[Abstract/Free Full Text] - Cleland J.G.F., Bulpitt C.J., Falk R.H., et al. Is aspirin safe for patients with heart failure? Br Heart J (1995) 74:215–219.
[Free Full Text] - Smith P., Arnesen H., Holme I. The effect of warfarin on mortality and reinfarction after myocardial infarction. New Engl J Med (1990) 323:147–151.[Abstract]
- Packer M., Poole Wilson P.A., Armstrong P.W., 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) 100:2312–2318.
[Abstract/Free Full Text] - Cleland J.G.F., Swedberg K. on behalf of the International ecadotril multi-centre dose-ranging study investigators. Lack of efficacy of neutral endopeptidase inhibitor ecadotril in heart failure. Lancet (1998) 351:1657–1658.[Web of Science][Medline]
- Cleland JGF, Armstrong PW, Horowitz JD et al. Baseline characteristics of patients recruited into the assessment of treatment with lisinopril and survival study (ATLAS). Eur J Heart Failure 1999:1.
- Motwani J.G., Lang C.C., Allen M.J., et al. Dose-ranging effects of candoxatril on elimination of exogenous atrial natriuretic peptide in chronic heart failure. Clin Pharmacol Ther (1993) 54:661–669.[Web of Science][Medline]
- Good J.M., Peters M., Wilkins M. Renal response to candoxatrilat in patients with heart failure. J Am Coll Cardiol (1995) 25:1273–1281.[Abstract]
- Cleland JGF. Chronic hemodynamic effects of candoxatril: a multi-centre dose ranging study. J Am Coll Cardiol 1997;29 Suppl A:169A(Abstract).
- Newby DE, McDonagh T, Currie PF et al. Candoxatril improves exercise capacity in patients with chronic heart failure receiving angiotensin converting enzyme inhibition. J Am Coll Cardiol 1997;29 Suppl A:169A(Abstract).
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  C. H.M. van Jaarsveld, A. V. Ranchor, G. I.J.M. Kempen, J. C. Coyne, D. J. van Veldhuisen, and R. Sanderman Epidemiology of heart failure in a community-based study of subjects aged >=57 years: Incidence and long-term survival Eur J Heart Fail, January 1, 2006; 8(1): 23 - 30. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. D.R. Thackray, K. K.A. Witte, A. Khand, A. Dunn, A. L. Clark, and J. G.F. Cleland Clinical trials update: highlights of the scientific sessions of the American Heart Association year 2000: Val HeFT, COPERNICUS, MERIT, CIBIS-II, BEST, AMIOVIRT, V-MAC, BREATHE, HEAT, MIRACL, FLORIDA, VIVA and the first human cardiac skeletal muscle myoblast transfer for heart failure Eur J Heart Fail, January 1, 2001; 3(1): 117 - 124. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Witte, S. Thackray, A. L. Clark, M. Cooklin, and J. G.F. Cleland Clinical trials update: IMPROVEMENT-HF, COPERNICUS, MUSTIC, ASPECT-II, APRICOT and HEART Eur J Heart Fail, December 1, 2000; 2(4): 455 - 460. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Mcmurray Beta-blockers, ventricular arrhythmias, and sudden death in heart failure: not as simple as it seems Eur. Heart J., August 1, 2000; 21(15): 1214 - 1215. [PDF]
|
|
|
|
|
|
S. Thackray, K. Witte, A. L. Clark, and J. G.F. Cleland Clinical trials update: OPTIME-CHF, PRAISE-2, ALL-HAT Eur J Heart Fail, June 1, 2000; 2(2): 209 - 212. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Heart failure drugs: whats new? DTB, April 1, 2000; 38(4): 25 - 27. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Willenbrock, S. Philipp, V. Mitrovic, and R. Dietz Neurohumoral blockade in CHF management Journal of Renin-Angiotensin-Aldosterone System, March 1, 2000; 1(1_suppl): 24 - 30. [Abstract] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||