Congestive Heart Failure
Meta-analysis of observed mortality data from all-controlled, double-blind, multiple-dose studies of losartan in heart failure

https://doi.org/10.1016/S0002-9149(99)00646-3Get rights and content

Abstract

Clinical studies of heart failure utilizing losartan, an angiotensin-II receptor antagonist, found that this drug is well tolerated and demonstrates hemodynamic, neurohormonal, and symptomatic improvement. To assess all-cause mortality in heart failure patients treated with losartan, a meta-analysis including 1,896 patients was performed on 6 controlled, double-blind, multiple-dose studies, regardless of sample size or duration of follow-up. A combination of logarithmic (log) odds ratios with a continuity correction was utilized for the meta-analysis. Treatment groups were comparable with regard to demographic characteristics, heart failure characteristics, and concomitant cardiovascular therapies. Concomitant use of open-label angiotensin-converting enzyme (ACE) inhibitors was not allowed in any study. The mean left ventricular ejection fraction obtained in individual studies ranged from 23% to 31%. Seven hundred forty patients were randomized to control therapy and 1,154 patients were randomized to losartan therapy. There were 36 deaths (3.12%) in the losartan groups compared with 47 in the control groups (6.35%) during the double-blind periods. The odds of dying in the losartan groups were 0.51 times (0.31 to 0.81) that of dying in the control groups (p = 0.004). In this analysis, treatment with losartan provided a beneficial effect upon survival. However, because the number of deaths in these studies is relatively small and the follow-up relatively short, a large confirmatory study is needed to assess the mortality benefit of losartan compared with an ACE inhibitor.

Section snippets

Selection of studies

A total of 8 clinical studies in the losartan heart failure program were available. Two of the 8 clinical studies in heart failure patients did not qualify for inclusion into the meta-analysis. One was a pharmacokinetic study, and the other was a single-dose study. No patients died during the course of these 2 studies. The remaining 6 studies were multicenter, randomized, double-blind studies, and had parallel treatment groups (Table I). These 6 studies were included in the meta-analysis,

Results

Baseline demographics and heart failure characteristics of patients enrolled in the losartan heart failure studies are presented in Table II. More men than women participated in all studies. The mean age of patients was ∼60 years, except for the ELITE trial, where age was >70 years. The studies were dominated by white patients, whereas the percentage of Hispanics was approximately 30% in 1 trial (Phase III-Int’l). More than half of all patients had coronary (ischemic) heart disease as the

Discussion

The meta-analysis method based on a combination of log ORs with a continuity correction was selected for this analysis. The OR is an approximation of the relative risk when event rates are low. To evaluate the robustness of the meta-analysis results, the data were also reanalyzed by Peto’s modification of the Mantel-Haenszel method.14 A significant mortality benefit was once again observed in favor of losartan and similar estimates of the overall ORs were observed. However, because the Peto

Acknowledgements

The authors would like to express their gratitude to all of the clinical investigators associated with the lorsartan heart failure studies, and Cindy White for designing the figures.

References (26)

  • I. Crozier et al.

    Losartan in heart failure. Hemodynamic effects and tolerability

    Circulation

    (1995)
  • C.S. Sweet et al.

    Losartan in heart failurepreclinical experiences and initial clinical outcomes

    Europ Heart

    (1994)
  • Lang RM, Yellen LG, McKelvie RS, Vaughn D, Ney DE, Makris L, Chang P, on behalf of the Losartan Pilot Exercise Study...
  • Cited by (79)

    View all citing articles on Scopus

    This study was supported by Merck Research Laboratories, West Point, Pennsylvania.

    A list of the study group investigators appears in the .

    View full text