Sixteen RCTs (14,857 participants) were included. Seven studies tested the oral beta-blocker carvedilol (3,847 participants), 5 studies tested metoprolol (4,875 participants), and 2 studies each tested bisoprolol (3,288 participants) and bucindolol (2,847 participants).
Overall, compared with placebo, there was a 22% reduction in all-cause mortality (95% CI: 16, 28, P<0.001) with beta-blocker treatment and a 24% (95% CI: 20, 29, P<0.001) reduction in hospitalisations for heart failure. Significant heterogeneity was observed for all-cause mortality (P=0.035). This heterogeneity disappeared when one trial of bucindolol was excluded. This study and a further study of the same drug were then omitted from the subgroup analyses.
The benefit of beta-blocker treatment on mortality and hospitalisations for heart failure was similar for the drugs metoprolol, bisoprolol and carvedilol. Mortality was reduced by 31% (95% CI: 17, 42) with metoprolol, 29% (95% CI: 17, 40) with bisoprolol and 37% (95% CI: 24, 47) with carvedilol. Hospitalisations for heart failure were reduced by 28% (95% CI: 19, 36) with metoprolol, 32% (95% CI: 21, 42) with bisoprolol and 29% (95% CI: 18, 39) with carvedilol.
The effectiveness of beta-blockers in reducing mortality and hospitalisations for heart failure was similar for different grades of severity of heart failure, according to both the NYHA (4 studies) and LVEF (4 studies). Beta-blockers reduced mortality by 32% (95% CI: 21, 41) in NYHA class III disease, by 32% (95% CI: 19, 43) in NYHA class IV disease, by 34% (95% CI: 20, 45) in those with LVEF of 25% or greater, and by 29% (95% CI: 19, 38) in those with LVEF less than 25%.
When examining the IPD from one trial, similar results were found in that there were no significant differences in the number of deaths or hospitalisations between the different levels of heart failure severity (further details were provided in the paper).