Seventeen studies with a total of 3,039 participants (1,723 randomised to beta-blockers and 1,316 randomised to control) were included.
Total mortality.
Beta-blockade treatment was associated with a significant reduction in mortality with a summary OR of 0.69 (95% confidence interval, CI: 0.54, 0.88). The summary rate-difference between the control and treatment arms in the 17 studies was 2.9 deaths per 100 patients treated (95% CI: 22, 84). This suggests a number-needed-to-treat of 35 over 9 months to prevent one death. The chi-squared test for homogeneity was 0.9.
Cardiac and sudden death mortality.
Beta-blockade treatment was associated with a significant reduction in cardiac mortality with a summary OR of 0.68 (95% CI: 0.53, 0.89). The summary rate-difference between the control and treatment arms in the 14 studies was 3.0 deaths per 100 patients treated (95% CI: 1.0, 4.8). This corresponded to one cardiac death prevented for every 33 treated (95% CI: 21, 82). For sudden cardiac death, the summary OR was 0.84 (95% CI: 0.59, 1.2) (15 trials). For non-sudden cardiac death, the summary OR was 0.58 (95% CI: 0.40, 0.83).
Ischaemic versus non-ischaemic cardiomyopathy.
The combined OR for 7 trials of participants with ischaemic cardiomyopathy (n=1,387) was 0.69 (95% CI: 0.49, 0.98). This was not significantly different from the combined OR for 9 trials of participants with non-ischaemic cardiomyopathy (n=1,436) which was 0.69 (95% CI: 0.47, 0.99).
Carvedilol versus other beta-blockers.
The combined OR was 0.54 (95% CI: 0.36, 0.81) for trials of carvedilol and 0.82 (95% CI: 0.60, 1.12) for trials of other beta-blockers. This difference was not statistically significant (p=0.1).