Send to

Choose Destination
J Am Coll Cardiol. 2011 Aug 23;58(9):915-22. doi: 10.1016/j.jacc.2011.03.057.

Influence of global region on outcomes in heart failure β-blocker trials.

Author information

Department of Medicine, Division of Cardiology, Division of Clinical Pharmacology, Duke University Medical Center, Durham, North Carolina, USA.



We sought to describe the United States and the rest of the world (ROW) outcomes from the major β-blocker heart failure (HF) trials.


HF trials have demonstrated differences in outcomes by geographic region.


Randomized, double-blind, placebo-controlled studies that evaluated β-blockers in HF patients, had a primary endpoint of mortality, and enrolled U.S. patients were included. Relative risk (RR) was calculated for patients enrolled in the United States and ROW. Meta-analysis of the combined mortality rates was performed using the Cochran-Mantel-Haenszel statistic, stratified by study.


A total of 8,988 patients were enrolled in the MERIT-HF (Metoprolol Controlled-Release Randomized Intervention Trial in Heart Failure), COPERNICUS (Carvedilol Prospective Randomized Cumulative Survival trial), and BEST (β-Blocker Evaluation of Survival Trial) combined; 4,198 (46.7%) were from the United States. In the U.S. cohort, the RR reduction for each β-blocker was of smaller magnitude than in the overall cohort and no longer significant, whereas in the ROW subgroup, the mortality benefit for β-blockade persisted. In the pooled analysis (n = 11,635), the RR of death was reduced by 23% (p < 0.001) with β-blockade compared with placebo. In contrast, the mortality reduction associated with β-blockade in the U.S. cohort was small and not statistically significant (RR: 0.92, 95% confidence interval [CI]: 0.82 to 1.02, p = 0.11). The survival benefit persisted in the ROW cohort (RR: 0.64, 95% CI: 0.56 to 0.72, p < 0.001).


Among patients enrolled in the United States, β-blockade was associated with a lower magnitude of survival benefit, whereas the ROW response was similar to the total study population. This geographic difference in treatment response may be a reflection of population differences, genetics, cultural or social differences in disease management, or low power and statistical chance.

[Indexed for MEDLINE]
Free full text

Supplemental Content

Full text links

Icon for Elsevier Science
Loading ...
Support Center