In order to bring you the best possible user experience, this site uses Javascript. If you are seeing this message, it is likely that the Javascript option in your browser is disabled. For optimal viewing of this site, please ensure that Javascript is enabled for your browser.
Did you know that your browser is out of date? To get the best experience using our website we recommend that you upgrade to a newer version. Learn more.

We use cookies to optimise the design of this website and make continuous improvement. By continuing your visit, you consent to the use of cookies. Learn more

BB-meta-HF: Efficacy of beta-blockers in heart failure patients with atrial fibrillation: Individual patient meta-analysis from the Beta-blockers in Heart Failure Collaborative Group

ESC Congress 2014 - Hot Line report

Cardiovascular Pharmacology and Pharmacotherapy



By Dipak Kotecha, (Birmingham, United Kingdom)
View Discussant report
Open the Presentation
Watch the Webcast

List of Authors:

Dipak Kotecha, Jane Holmes, Henry Krum, Douglas G Altman, Luis Manzano, John GF Cleland, Gregory YH Lip, Andrew JS Coats, Bert Andersson, Paulus Kirchhof, Thomas G von Lueder, Hans Wedel, Giuseppe Rosano, Marcelo C Shibata, Alan Rigby and Marcus D Flather, on behalf of the Beta-Blockers in Heart Failure Collaborative Group.


Atrial fibrillation (AF) and heart failure (HF) frequently coexist, causing substantial cardiovascular (CV) morbidity and mortality. Beta-blockers are a Class1A indication in symptomatic HF with reduced ejection fraction, however their efficacy in patients with concomitant AF is uncertain.
We extracted individual patient data from 10 randomised controlled trials comparing beta-blockers versus placebo in HF. The presence of sinus rhythm or AF was determined from the baseline electrocardiogram. Outcome data were meta-analysed using adjusted Cox proportional hazards regression and are presented as hazard ratios (HR) with confidence intervals.
A total of 18,254 participants were assessed, of which 13,946 (76.4%) were in sinus rhythm and 3,066 (16.8%) in AF. Crude death rates over a mean follow-up of 1.5 years (SD 1.1) were 16.0% in sinus rhythm and 20.7% in AF. Beta-blocker therapy in patients with sinus rhythm led to a significant reduction in all-cause mortality (HR 0.73, 0.67-0.80; p<0.001) and CV-hospitalisation (HR 0.78, 0.73-0.83; p<0.001), as well as CV-death, HF-hospitalisation, the composite of all-cause mortality/CV-hospitalisation and the composite of CV-death/HF-hospitalisation. In contrast, there was no apparent effect from beta-blockers on any outcome in patients with AF, including mortality (HR 0.97, 0.83-1.14; p=0.73) and CV-hospitalisation (HR 0.91, 0.79-1.04; p=0.15). All outcomes demonstrated significant interaction p-values for beta-blocker efficacy according to baseline rhythm. The lack of efficacy for mortality was seen across all AF sub-groups. AF patients had more frequent CV-hospitalisation and longer length of stay (11.9 versus 9.7 days in sinus rhythm).
Patients with HF and concomitant AF have higher rates of death and hospitalisation. In contrast to the beneficial effects observed for those with sinus rhythm, beta-blocker therapy has no or minimal effect on mortality or cardiovascular hospitalisation in HF patients with reduced ejection fraction and AF. Our results dispute the preferential use of beta-blockers compared to other rate-control medications and highlight the need for further trials in this common and increasingly important group of patients.
Registration: PROSPERO CRD42014010012; NCT00832442. 


By Lars Ryden, FESC (Stockholm, Sweden)
See Presenter abstract
Open the presentation
Watch the Webcast


The combination of heart failure and atrial fibrillation (AF) is frequent and has a negative prognostic implication. Heart failure guidelines recommend prescription of renal angiotensive system blockers combined with beta-blockade not only for symptomatic relief but also to increase longevity. Such treatment has been considered important for patients with sinus rhythm (SR) and AF. The present meta-analysis of major clinical trials of beta-blockers in heart failure due to compromised left ventricular function was conducted to provide answer to the question if such treatment is effective in the presence of AF.
The meta-analysis
The present meta-analysis was conducted as an individual patient data report collecting data bases comprising 18,254 participants in major randomised controlled trials (RCT) of beta-blockers in heart failure from the MDC trial 1993 to SENIORS 2005. The proportion of patients in SR was 76% while 17% had AF. The latter group had a somewhat higher risk profile, but statistical corrections took this into account. Mortality in patients with AF was 21% compared to 16% among patients in SR. The 27% decrease in mortality in patients with SR was significant [HR 0,73; 95 % CI 0,67-0,80;p<0,001] but there was no mortality impact in patients with AF [HR of 0,97; 95 % CI 0,83-0,14; p=0,73]. Sensitivity analyses were performed for different subgroups, the various studies, beta-blocker dosages etc. demonstrating that this outcome was consistent with no impact of beta-blockers on mortality in patients in AF and a significant reduction among those in SR.
The authors conclude that beta-blocker therapy in heart failure patients with AF appeared safe but suggest that beta-blockers should no longer be considered standard therapy to improve prognosis in patients AF.
This meta-analysis is of the highest possible standard with a cautious collection of study material including not only reports from the trials included but also access to the original databases. It also seems as if the authors have made great efforts to collect information by looking at individual patient data in order to harmonise the report. Moreover the analysis comprises a sizeable patient material with few missing data giving a high power to the observations made.
It should be noted that this study addresses patients with heart failure due to compromised left ventricular function and cannot be seen as informative for patients with symptoms of heart failure but with preserved left ventricular function. Moreover, and as always with meta-analyses of trials, the patient material may be more or less representative for patients seen in all day practice. A proportion of patients with AF of 17 % can be compared to data derived from the Swedish Heart Failure Registry comprising a majority of Swedish patients with heart failure in which almost 40 % of the patient had AF. It may therefore be that patients included in RCT are to some extent selected.
AF increases by age and the median age in the meta-analysis is for patients in SR 64 and for those in AF 69 years. Many RCT of beta-blockers have an age limit and it is therefore difficult to know whether elderly patients with heart failure react in the same way as regards beta-blocker treatment as younger patients even in the presence of one study, SENIORS, intentionally recruiting older patients. Another question is whether studies conducted 1993 (MDC) can be compared to trials performed ten years later (SENIORS). Clinical science has improved by time and it may be that the demands on study details increased.
Nevertheless one must be impressed by the careful way this meta-analysis was conducted. In the lack of RCT comparing beta-blockers in heart failure patients with AF the present findings should be seriously considered. They call for a definite answer to the important question and this can only be given by a RCT. Since patient AF comprises a fairly large group of heart failure patients definite information may have important implications. Beta-blockers do not only had beneficial but also side effects. Why expose patients to a treatment that does not impact their prognosis? Why invest in the cost for beta-blockers in patients who do not need them?
We can probably not expect a RCT on the initiative of pharmaceutical companies selling beta-blockers. In my opinion it is a societal responsibility to seriously to consider conducting such trial on a representative patient population. Such trial can give a definitive answer to an important question on the benefit of beta-blockers in heart failure patients with AF, but until then we have to accept strong indications from a well-conducted meta-analysis.




Clinical Trial Update Hot Line: Stable CAD and atrial fibrillation

The content of this article reflects the personal opinion of the author/s and is not necessarily the official position of the European Society of Cardiology.