Twenty-two randomised controlled trials with 10,135 patients were included.
The combined odds ratio (OR) for total mortality was 0.65 (95% confidence interval, CI: 0.53, 0.80). Assuming a mortality rate among placebo recipients of 12%, 3.8 lives are saved per 100 patients treated (95% CI: 2.1, 5.3) in the first year of treatment. The probability density curve showed a 99% probability that beta-blockers save at least 2 lives per 100 patients, and an 85% probability that 3 or more lives are saved per 100 patients. Selective agents (predominantly metoprolol and bisoprolol) and nonselective agents (predominantly carvedilol) were associated with reduced mortality: the ORs were 0.67 (95% CI: 0.57, 0.79) and OR 0.52 (95% CI: 0.28, 0.89), respectively.
The OR for hospitalisations for heart failure was 0.64 (95% CI: 0.53, 0.79). Recent trials showed the weighted average of placebo recipients who were hospitalised in the first year of follow-up was 14%. Given this assumption, beta-blockers reduce hospitalisations by 4 patients per 100 treated (95% CI: 2.4, 5.6).
When the results of an unpublished trial of bucindolol (n=2,708) were incorporated, the accumulative mortality OR was 0.72 (95% CI: 0.61, 0.84). Funnel plots did not suggest any publication bias. Adding further (hypothetical trials) involving 2,000 patients with mortality rates of 20 and 10% in the beta-blocker and placebo groups, respectively, beta-blockers would still reduce the OR by 16% (OR 0.84, 95% CI: 0.74, 0.94).