OBJECTIVE: To evaluate the effect of beta-blockers on mortality and morbidity, and to provide an appraisal of the reliability of the available data. DATA SOURCES: MEDLINE search for trials of beta-blockers for congestive heart failure (CHF). STUDY SELECTION: All randomized trials of beta-blockers versus placebo, or greater than one month's duration, in patients with CHF. Eighteen published trials involving 2986 patients were selected. DATA EXTRACTION: Independently by two authors. DATA SYNTHESIS: The Yusuf-Peto method for combining data was used. Data were available on mortality in 2841 patients (95%), on hospitalization for heart failure in 1514 (51%) and on heart transplantation in 2330 (79%). There was a lower rate of death in the active treatment group (131 of 1606) [8.2] versus 155 of 1235 [12.6%]; OR = 72; 99% CI 0.51 to 1.00), a lower rate of hospitalization for heart failure (137 of 756 [18.1%] versus 218 of 758 [28.7%]; OR = 0.54; 99% CI 0.39 to 0.74) and a trend towards a lower proportion of patients receiving heart transplantation (15 of 1354 [l.1%] versus 26 of 976 [2.7%]; OR = 0.45; 99% CI 0.20 to 1.03). Ventricular function improved; however, there was no effect on exercise duration. Although the effects on mortality were nominally statistically significant, the use of formal methods of interim monitoring adapted for meta-analyses suggests that substantially more patients still need to be studied in large scales trials to provide reliable and conclusive evidence. CONCLUSIONS: While the available data on the use of beta-blockers in CHF appear to be promising, they are neither complete nor robust. The routine use of beta-blockers in patients with heart failure should wait the results of ongoing studies.
OBJECTIVE: To evaluate the effect of beta-blockers on mortality and morbidity, and to provide an appraisal of the reliability of the available data. DATA SOURCES: MEDLINE search for trials of beta-blockers for congestive heart failure (CHF). STUDY SELECTION: All randomized trials of beta-blockers versus placebo, or greater than one month's duration, in patients with CHF. Eighteen published trials involving 2986 patients were selected. DATA EXTRACTION: Independently by two authors. DATA SYNTHESIS: The Yusuf-Peto method for combining data was used. Data were available on mortality in 2841 patients (95%), on hospitalization for heart failure in 1514 (51%) and on heart transplantation in 2330 (79%). There was a lower rate of death in the active treatment group (131 of 1606) [8.2] versus 155 of 1235 [12.6%]; OR = 72; 99% CI 0.51 to 1.00), a lower rate of hospitalization for heart failure (137 of 756 [18.1%] versus 218 of 758 [28.7%]; OR = 0.54; 99% CI 0.39 to 0.74) and a trend towards a lower proportion of patients receiving heart transplantation (15 of 1354 [l.1%] versus 26 of 976 [2.7%]; OR = 0.45; 99% CI 0.20 to 1.03). Ventricular function improved; however, there was no effect on exercise duration. Although the effects on mortality were nominally statistically significant, the use of formal methods of interim monitoring adapted for meta-analyses suggests that substantially more patients still need to be studied in large scales trials to provide reliable and conclusive evidence. CONCLUSIONS: While the available data on the use of beta-blockers in CHF appear to be promising, they are neither complete nor robust. The routine use of beta-blockers in patients with heart failure should wait the results of ongoing studies.