Canadian Journal of Cardiology 30 (2014) 898e903

Systematic Review/Meta-analysis
Comparable Benet of b-Blocker Therapy in Heart
Failure Across Regions of the World: Meta-analysis
of Randomized Clinical Trials
Saurav Chatterjee, MD,a Jacob A. Udell, MD, MPH,b Partha Sardar, MD,c Edgar Lichstein, MD,d
and John J. Ryan, MBBChe
a
Department of Cardiovascular Diseases, St Lukes-Roosevelt Hospital of the Mount Sinai Health System, New York, New York, USA
b
Womens College Research Institute and Cardiovascular Division, Department of Medicine, Womens College Hospital, University of Toronto, Toronto, Ontario, Canada
c
Department of Cardiology, Texas Tech University Health Sciences Center, El Paso, Texas, USA
d
Department of Medicine, Maimonides Medical Center and SUNY Downstate, Brooklyn, New York, USA
e
Division of Cardiovascular Medicine, University of Utah, Salt Lake City, Utah, USA

See editorial by de Denus et al., pages 858-860 of this issue.

ABSTRACT 
RESUM 
E
Background: There is a concern about geographical region heteroge- Introduction : Lhe  te
roge
 ne
ite
 entre les re gions ge
ographiques
neity regarding clinical benet of b-blocker (BB) therapy in heart failure demeure une pre occupation sur le plan des avantages cliniques du
with reduced ejection fraction (HFrEF). This study sought to compare traitement par b-bloquants (BB) lors dinsufsance cardiaque fraction
benets of BB use within randomized controlled trials (RCTs) that jection re
de duite (ICFER). Cette e tude visait comparer les avantages
enrolled patients with HFrEF from North America (NA) compared with de lutilisation des BB dans le cadre dessais cliniques ale atoires (ECA)
other regions of the world (ROW). mene s auprs de patients de lAme rique du Nord (NAM) et dautres
Methods: We conducted a meta-analysis using MEDLINE, EMBASE, gions du monde (RDM) qui souffraient dune ICFER.
re
Cochrane Library, Web of Science, and Scopus (inceptions-December Me thodes : Nous avons mene  une me ta-analyse des ECA sur les BB
2012) of BB RCTs stratied according to NA vs ROW. The primary end s selon NAM vs RDM partir de MEDLINE, dEMBASE, de la
stratie
point was all-cause mortality and secondary end points were cardio- Bibliothque Cochrane, du Web of Science et de Scopus (de leur
vascular death, sudden death, death due to pump failure, and premature ation de
cre cembre 2012). Le critre de jugement principal e tait la

b-Blockers (BBs) reduce mortality in patients with heart
failure (HF) with reduced ejection fraction (HFrEF) and this
has been postulated to be a class effect.1 Several registries and
trials that included HFrEF patients have reported differences
in short-term mortality risk across continents or geographic
regions.2-7 Reasons for regional heterogeneity in HF outcomes
are complex, but might be caused in part by differences in
utilization of, and responses to, BB therapy.7,8
A previous meta-analysis of the efcacy of BB therapy in
patients with HFrEF assessed regional differences in clinical
outcomes within 4 large randomized controlled trials
(RCTs).7 The most marked reduction in mortality was found
Received for publication February 21, 2014. Accepted March 12, 2014.
Corresponding author: John J. Ryan, Division of Cardiovascular
Medicine, University of Utah Health Science Center, 30 North 1900 East,
Room 4A100, Salt Lake City, Utah 84132, USA. Tel.: 1-801-585-2341;
fax: 1-801-587-5874.
E-mail: john.ryan@hsc.utah.edu
See page 902 for disclosure information.
in trials that enrolled patients only from outside North
America (NA). Whether potential differences in efcacy are a
result of regional differences in patient risk or medication
response remains unclear.9-12 We performed a further meta-
analysis to test the hypothesis that differences in BB efcacy
according to region for all-cause mortality and other cardio-
vascular outcomes are present in most of the high-quality
RCTs of HFrEF patients.
Methods
Study search and data sources
A systematic literature search, for, and selection of, ran-
domized trials of BB therapy in HFrEF patients within major
biomedical literature databases was conducted as described in
the Data Sources and Searches section of the Supplementary
Material. Studies were considered suitable for inclusion if
they reported on randomized trials, compared BBs vs placebo/
comparators in HF, enrolled at least 90% of the population

0828-282X/$ - see front matter 2014 Canadian Cardiovascular Society. Published by Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.cjca.2014.03.012

Downloaded from ClinicalKey.com at Texas Tech University of Health Science Center November 20, 2016.
For personal use only. No other uses without permission. Copyright 2016. Elsevier Inc. All rights reserved.
Chatterjee et al. 899
Meta-analysis of b-Blockers in HFrEF by Region

drug discontinuation. Summary odds ratios (ORs) and 95% condence mortalite toutes causes confondues et les critres de jugement se-
intervals (CIs) for each outcome were calculated with interaction terms condaires e taient la mortalite lie
e aux maladies cardiovasculaires, la
for region. Two-sided P values were calculated with P < 0.05 consid- mort subite, la mortalite lie
e linsufsance cardiaque et linterruption
ered signicant. precoce des me dicaments. Les ratios dincidence approche s (RIA)
Results: The analysis included 16 RCTs with 14,452 patients; 7 trials sommaires et les intervalles de conance (IC) 95 % de chacun des
were conducted in NA and 9 trials in ROW with follow-up durations of 3- sultats e
re taient calcules partir des termes dinteraction des regions.
58 months. All-cause mortality was consistently reduced in NA (OR, Les valeurs P bilatres e taient calculees partir de P < 0,05, qui est
0.82; 95% CI, 0.71-0.96; P 0.01) and ROW (OR, 0.76; 95% CI, 0.69- considere
 comme signicatif.
0.84; P < 0.001; P-interaction 0.40). Overall and according to region, Resultats : Lanalyse incluait 14 452 patients de 16 ECA; 7 essais
all secondary end points including premature drug discontinuation were alise
re s en NAM et 9 essais dans les RDM pour lesquels les dure es de
also less with BB therapy (P-interactions all  0.10). taient de 3 58 mois. La mortalite
suivi e  toutes causes confondues
Conclusions: For the regions represented in the included trials, there is tait syste
e matiquement re duite en NAM (RIA, 0,82; IC 95 %, 0,71-
no evidence to suggest that geographic region is a signicant moder- 0,96; P 0,01) et dans les RDM (RIA, 0,76; IC 95 %, 0,69-0,84;
ator of clinical outcomes with BB therapy in HFrEF patients. P < 0,001; interaction P 0,40). Dans lensemble, et selon la re gion,
les critres de jugement secondaires incluant linterruption pre coce
des me dicaments e taient egalement moindres avec le traitement par
BB (interactions P, toutes  0,10).
Conclusions : Pour les re gions repre sente
es dans les essais inclus,
aucune donne e ne montre que la re gion geographique est un mo-
derateur important des re sultats cliniques lors de traitement par BB
chez les patients souffrant dICFER.

from either continental NA (United States and Canada) or proportion of drug discontinuation. Region was categorized as
other regions of the world (ROW; Europe, Australia, and either NA or ROW if at least 90% of the recruited trial
New Zealand), or stratied results according to geographic population originated from that location. Study quality and
region in a manner that allowed comparison of NA vs ROW, risk of bias was appraised using Cochrane criteria as described
and reported the primary outcome of interest. Studies were in the Data Synthesis and Analysis section of the
excluded if they were not RCTs, had <100 patients, or had a Supplementary Material, and results were reported according
follow-up duration of <3 months. to the Preferred Reporting of Items for Systematic Reviews
and Meta-Analyses statement.28,29
Data extraction and quality assessment
Statistical analysis
Baseline characteristics were extracted stratied according
to geographic region as described in the Data Extraction, and Event rates, or proportion achieving an event, were
Quality Assessment section of the Supplementary Material and analyzed per treatment arm from individual studies stratied
summarized in Table 1. The primary outcome was all-cause by region with good interrater agreement (k 0.88). Sum-
mortality. Secondary outcomes included cardiovascular mary odds ratios (ORs) and 95% condence intervals (CIs)
death, sudden cardiac death, pump failure death, and for each outcome were calculated comparing BB with placebo/
Table 1. Key features of included studies
Experimental Control Age, Male Ischemic
Trial-location Patients, n (% AA) group drug group drug years sex, % HF, % Baseline LVEF, % Follow-up, months
ANZ13-ROW 415 (NR) Carvedilol Placebo 67 80 88 29 19
BEST14-NA 2708 (24) Bucindolol Placebo 60 77 58 23 36
Bristow et al.15-NA 139 (NR) Bucindolol Placebo 56 62 29 26 3
CARMEN16-ROW 381 (0) Carvedilol Enalapril 62 77 68 <40 22
CIBIS17-ROW 641 (NR) Bisoprolol Placebo 60 83 56 25 23
CIBIS II18-ROW 2647 (NR) Bisoprolol Placebo 61 81 50 28 16
CIBIS III19-ROW 1010 (NR) Bisoprolol Enalapril 72 66 61 29 24
COMET20-ROW 3029 (1) Carvedilol Metoprolol 62 79 51 26 58
COPERNICUS21,* 2289 (NR) Carvedilol Placebo 63 79 67 20 10
ENECA22-ROW 260 (0.75) Nebivolol Placebo 72 70 60 25 12
MERIT-HF 200023 and 200424,* 3991 (6) Metoprolol Placebo 64 77 65 28 12
Packer et al.25-NA 1094 (NR) Carvedilol Placebo 58 77 48 23 12
SENIORS26-ROW 2128 (NR) Nebivolol Placebo 76 72 69 36 21
Sturm et al.27-ROW 100 (NR) Atenolol Placebo 51 87 31 17 24
AA, African-American; ANZ, Australia/New Zealand; BEST, Beta-Blocker Evaluation of Survival Trial; CARMEN, Carvedilol ACE-Inhibitor Remodelling
Mild CHF Evaluation; CIBIS, Cardiac Insufciency Bisoprolol Study; COMET, Carvedilol or Metoprolol European Trial; COPERNICUS, Carvedilol
Prospective Randomized Cumulative Survival; ENECA, Efcacy of Nebivolol in the Treatment of Elderly Patients With Chronic Heart Failure as Add-on Therapy
to ACE Inhibitors or Angiotensin II Receptor Blockers, Diuretics, and/or Digitalis; HF, heart failure; LVEF, left ventricular ejection fraction; MERIT-HF,
Metoprolol CR/XL Randomized Intervention Trial in Congestive Heart Failure; NA, North America; NR, not reported; ROW, other regions of the world;
SENIORS, Study of the Effects of Nebivolol Intervention on Outcomes and Rehospitalization in Seniors With Heart Failure.
* Trials that included NA and ROW populations, and also reported region-specic outcomes to enable inclusion in this study.

Downloaded from ClinicalKey.com at Texas Tech University of Health Science Center November 20, 2016.
For personal use only. No other uses without permission. Copyright 2016. Elsevier Inc. All rights reserved.
900 Canadian Journal of Cardiology
Volume 30 2014

Figure 1. Search strategy. BB, b-blocker; RCT, randomized controlled trial.

control use.30 We assessed extent of small study effects and
publication bias by visual inspection of funnel plots and the
Harbord regression modication of the Egger test.31 Interac-
tion terms for treatment effect by geographic region were
further introduced into the statistical models as described in
the Data Synthesis and Analysis section of the Supplementary
Material.32 For trials with overlapping populations, we
included the largest trial for the primary analysis, and sepa-
rately included other individual trials to test differential effects
on primary outcome (Trial Sequential Analysis section of the
Supplementary Material). If unpublished or missing data were
unavailable from corresponding authors, missing data were
imputed using the validated metamiss program/macro in
Stata 11SE (StataCorp LP, College Station, TX).33 Finally,
several sensitivity analyses, including trial sequential analysis,
for testing validity and robustness of our results were per-
formed as described in the Trial Sequential Analysis section of
the Supplementary Material.34-37 Two-sided P values were
calculated with P < 0.05 considered signicant.
Results
After screening 720 citations, a total of 16 RCTs for the
qualitative synthesis (12 RCTs for the quantitative analysis) were
included (ROW 9; NA 7; Fig. 1; N 14,452), treated with
either atenolol, bisoprolol, bucindolol, carvedilol, metoprolol, or
nebivolol, vs placebo/standard therapy.13-20,22,25-27,38-41 One
trial was conducted in Australia and New Zealand,13 8 in
Europe,16-20,22,26,27 and 7 in NA.14,15,25,38-41 Baseline charac-
teristics of NA and ROW studies are summarized in Table 1. In
addition to standard HF therapy, control groups received pla-
cebo in all but 3 trials conducted in ROW in which the control
was either an angiotensin-converting enzyme inhibitor,16,19 or a
BB, specically a comparison of carvedilol vs metoprolol
tartrate.20
There was a signicant mortality reduction with BB use vs
any comparator (primarily placebo) in NA (OR, 0.82; 95% CI,
0.71-0.96; P 0.01; I2 74%) and ROW (OR, 0.76; 95%
CI, 0.69-0.84; P < 0.001; I2 0; Fig. 2 and Table 2). There
was no signicant effect of modication by region for all-cause
mortality (P-interaction 0.40). No publication bias was
noted within individual regions by visual inspection of funnel
plots or modied Egger test (P 0.27 for NA and P 0.31 for
ROW, respectively), nor signicant between-region heteroge-
neity (overall I2 0%). Overall and by region, BBs also reduced
cardiovascular mortality, sudden cardiac death, and pump
failure death (Table 2, Supplemental Figs. S1-S5). There were
no regional differences detected for drug discontinuation.

Downloaded from ClinicalKey.com at Texas Tech University of Health Science Center November 20, 2016.
For personal use only. No other uses without permission. Copyright 2016. Elsevier Inc. All rights reserved.
Chatterjee et al. 901
Meta-analysis of b-Blockers in HFrEF by Region

Figure 2. Pairwise meta-analysis of b-blockers and effect on all-cause mortality in HF with reduced ejection fraction according to geographic region.
BB, b-blocker; CI, condence interval; CV, cardiovascular; HF, heart failure; M-H, Mantel-Haenszel; ROW, other regions of the world; US, United
States.

Sensitivity analyses results consistently supported the primary
ndings, including analyses that reclassied Canadian patients
as part of the ROW group and United States patients as only
comprising the NA group, as described in the Trial Sequential
Analysis section of the Supplementary Material14,15
(Supplemental Figs. S6-S11). Trials were generally graded as
low risk of bias (Supplemental Figs. S11).
The cumulative Z curve crossed the trial sequential
monitoring boundary for mortality benet for the NA and the
ROW groups, conrming evidence of a 20% decrease in odds
of all-cause mortality with BB therapy in HFrEF, regardless of
geographic location (trial sequential analysis adjusted 95% CI
for NA was 0.66-0.92 [OR, 0.78], and that for ROW was
0.68-0.85 [OR, 0.76]) (Supplemental Figs. S12).
Discussion
The present meta-analysis, which included the totality of
available data from RCTs enrolling >100 patients with
HFrEF, and for the regions represented by the included trials,
this study demonstrates that region of the world had no sig-
nicant effect on the therapeutic benet of use of BBs. The
strengths of our study include its systematic inclusion of all
applicable trials, with a resultant large sample size and
consequently high statistical power. In addition, we conrmed
no regional differences in additional relevant cardiovascular
outcomes, specically cardiovascular mortality, sudden cardiac
death, death due to pump failure, and drug discontinuation.
There are several potential explanations for the observation
of a consistent benet of BB therapy in HFrEF patients across
regions. The patient groups in our ROW population were
derived largely from Europe, Australia, and New Zea-
landdregions that have a demographic and socioeconomic
prole comparable with NAdsuggesting sociodemographic
differences between regions are unlikely substantial enough to
be associated with variation in BB efcacy. Enrollment of
patients of African heritage might differ by region, which
might contribute to observed differences in regional BB
therapy analysis in light of reported differences in African-
American (AA) patients response to BB use.42 However,
our metaregression sensitivity analysis suggested no signicant
association between the proportion of AA patients in a trial
and mortality risk or benet, although this analysis might have
limited power because of low sample sizes of AA patients

Table 2. Clinical outcomes with BB therapy in RCTs in North America and ROW
Odds ratio (95% CI)
Interaction P (for geography
Clinical outcome assessed RCTs contributing to outcome NA ROW as moderator)
All-cause mortality NA 3, ROW 9 (N 14,452) 0.82 (0.71-0.96) 0.76 (0.69-0.84) 0.40
Cardiovascular mortality NA 3, ROW 8 (N 13,542) 0.78 (0.66-0.92) 0.76 (0.68-0.85) 0.81
Sudden cardiac death NA 3, ROW 6 (N 12,901) 0.84 (0.68-1.03) 0.73 (0.62-0.84) 0.28
Pump failure mortality NA 3, ROW 5 (N 10,132) 0.74 (0.58-0.95) 0.83 (0.69-1.00) 0.49
Drug discontinuation NA 3, ROW 9 (N 14,552) 1.03 (0.66-1.61) 1.00 (0.91-1.09) 0.87
BB, b-blocker; CI, condence interval; N, number of participants in pooled analysis; NA, North America; RCT, randomized controlled trial; ROW, other
regions of the world.

Downloaded from ClinicalKey.com at Texas Tech University of Health Science Center November 20, 2016.
For personal use only. No other uses without permission. Copyright 2016. Elsevier Inc. All rights reserved.
902
studied. We demonstrated no signicant regional differences
in drug discontinuation, suggesting differences in drug toler-
ability or adherence is also not substantial enough to result in
regional differences in BB effect.43 Thus, in the context of our
systematic study selection and sample size, our results are
discrepant from a previous meta-analysis that suggested a
difference in clinical outcomes with BB therapy between NA
compared with ROW, likely because of previous selection
bias.7
Study limitations
Our region subgroup categorization, and resultant inter-
action tests for treatment effect, was a post hoc analysis. In
addition, some of the included trials did not specify all-cause
mortality as the primary outcome. We also pooled results
from trials that examined different types of BBs, with varying
b-1 and b-2 receptor afnity, conducted over different time
intervals with varying degrees of background therapy. Finally,
differences in baseline characteristics of each trial population
might have inuenced our results in addition to unmeasured
confounders. Nevertheless, our results were robust, highly
statistically signicant, and consistent across different types of
cardiovascular outcomes. Nonavailability of genetic and
biomarker data limit our ability to comment on their associ-
ations and effects on clinical outcomes associated with BB use
in HFrEF. Also, patient-level data were not available for our
analysis. The analysis, and interpretations therein are limited
by the specic regions represented in the included trials
(namely, NA, Europe, and Australia New Zealand)dand
these ndings might not be generalizable in other regions not
represented in the current analysis.
Conclusions
BB therapy is associated with a comparable magnitude of
survival benet, and reduction in other important cardiovas-
cular outcomes, in trials of HF patients enrolled in NA and
ROW. Our data reafrm benets of BB therapy in HFrEF
patients consistently across NA and ROW, for the regions
represented in the included trials.
Acknowledgements
The authors thank Dr Michael Bristow, for providing
unpublished data15 for analysis and interpretation.
Prospective registration: PROSPERO 2013:CRD4201300
4647 (http://www.crd.york.ac.uk/PROSPERO/display_record.
asp?IDCRD42013004647).
Disclosures
The authors have no conicts of interest to disclose.
References
1. Chatterjee S, Biondi-Zoccai G, Abbate A, et al. Benets of b blockers in
patients with heart failure and reduced ejection fraction: network meta-
analysis [erratum in 2013;346:f596]. BMJ 2013;346:f55.
2. Cleland JG, Swedberg K, Follath F, et al. The EuroHeart Failure survey
programmee a survey on the quality of care among patients with heart
failure in Europe. Part 1: patient characteristics and diagnosis. Eur Heart
J 2003;24:442-63.
Downloaded from ClinicalKey.com at Texas Tech University of Health Science Center November 20, 2016.
For personal use only. No other uses without permission. Copyright 2016. Elsevier Inc. All rights reserved.
Canadian Journal of Cardiology
Volume 30 2014
3. Komajda M, Follath F, Swedberg K, et al. The Euro Heart Failure Survey
programmeea survey on the quality of care among patients with heart
failure in Europe. Part 2: treatment. Eur Heart J 2003;24:464-74.
4. Adams KF Jr, Fonarow GC, Emerman CL, et al. Characteristics and
outcomes of patients hospitalized for heart failure in the United States:
rationale, design, and preliminary observations from the rst 100,000
cases in the Acute Decompensated Heart Failure National Registry
(ADHERE). Am Heart J 2005;149:209-16.
5. Fonarow GC, Abraham WT, Albert NM, et al. Prospective evaluation of
beta-blocker use at the time of hospital discharge as a heart failure per-
formance measure: results from OPTIMIZE-HF. J Card Fail 2007;13:
722-31.
6. Blair JE, Zannad F, Konstam MA, et al. Continental differences in
clinical characteristics, management, and outcomes in patients hospital-
ized with worsening heart failure results from the EVEREST (Efcacy of
Vasopressin Antagonism in Heart Failure: Outcome Study with Tol-
vaptan) program. J Am Coll Cardiol 2008;52:1640-8.
7. OConnor CM, Fiuzat M, Swedberg K, et al. Inuence of global region
on outcomes in heart failure b-blocker trials. J Am Coll Cardiol 2011;58:
915-22.
8. Krumholz HM, Radford MJ, Wang Y, et al. National use and effec-
tiveness of beta-blockers for the treatment of elderly patients after acute
myocardial infarction: National Cooperative Cardiovascular Project.
JAMA 1998;280:623-9.
9. Van de Werf F, Topol EJ, Lee KL, et al. Variations in patient manage-
ment and outcomes for acute myocardial infarction in the United States
and other countries. Results from the GUSTO trial. Global Utilization of
Streptokinase and Tissue Plasminogen Activator for Occluded Coronary
Arteries. JAMA 1995;273:1586-91.
10. Giugliano RP, Llevadot J, Wilcox RG, et al. Geographic variation in
patient and hospital characteristics, management, and clinical outcomes
in ST-elevation myocardial infarction treated with brinolysis. Results
from InTIME-II. Eur Heart J 2001;22:1702-15.
11. Wallentin L, Becker RC, Budaj A, et al. Ticagrelor versus clopidogrel in
patients with acute coronary syndromes. N Engl J Med 2009;361:
1045-57.
12. Mentz RJ, Kaski JC, Dan GA, et al. Implications of geographical varia-
tion on clinical outcomes of cardiovascular trials. Am Heart J 2012;164:
303-12.
13. Australia/New Zealand Heart Failure Research Collaborative Group.
Randomised, placebo-controlled trial of carvedilol in patients with
congestive heart failure due to ischaemic heart disease. Lancet 1997;349:
375-80.
14. Beta-Blocker Evaluation of Survival Trial Investigators. A trial of the
beta-blocker bucindolol in patients with advanced chronic heart failure.
N Engl J Med 2001;344:1659-67.
15. Bristow MR, OConnell JB, Gilbert EM, et al. Dose-response of chronic
beta-blocker treatment in heart failure from either idiopathic dilated or
ischemic cardiomyopathy. Circulation 1994;89:1632-42.
16. Komajda M, Lutiger B, Madeira H, et al. Tolerability of carvedilol and
ACE-inhibition in mild heart failure: results of CARMEN (Carvedilol
ACE-Inhibitor Remodelling Mild CHF EvaluatioN). Eur J Heart Fail
2004;6:467-75.
17. CIBIS Investigators and Committees. A randomized trial of beta-
blockade in heart failure: the Cardiac Insufciency Bisoprolol Study
(CIBIS). Circulation 1994;90:1765-73.
18. The Cardiac Insufciency Bisoprolol Study II (CIBIS-II): a randomised
trial. Lancet 1999;353:9-13.
Chatterjee et al.
Meta-analysis of b-Blockers in HFrEF by Region
19. Willenheimer R, van Veldhuisen DJ, Silke B, et al. Effect on survival and
hospitalization of initiating treatment for chronic heart failure with
bisoprolol followed by enalapril, as compared with the opposite sequence:
results of the randomized Cardiac Insufciency Bisoprolol Study (CIBIS)
III. Circulation 2005;112:2426-35.
20. Poole-Wilson PA, Swedberg K, Cleland JG, et al. Comparison of car-
vedilol and metoprolol on clinical outcomes in patients with chronic
heart failure in the Carvedilol Or Metoprolol European Trial (COMET):
randomised controlled trial. Lancet 2003;362:7-13.
21. Packer M, Coats AJ, Fowler MB, et al. Effect of carvedilol on survival in
severe chronic heart failure. N Engl J Med 2001;344:1651-8.
22. Edes I, Gasior Z, Wita K. Effects of nebivolol on left ventricular function
in elderly patients with chronic heart failure: results of the ENECA study.
Eur J Heart Fail 2005;7:631-9.
23. Hjalmarson A, Goldstein S, Fagerberg B, et al. Effects of controlled-
release metoprolol on total mortality, hospitalizations, and well-being
in patients with heart failure: the Metoprolol CR/XL Randomized
Intervention Trial in congestive heart failure (MERIT-HF). JAMA
2000;283:1295-302.
24. Deedwania PC, Gottlieb S, Ghali JK, et al. Efcacy, safety and tolera-
bility of beta-adrenergic blockade with metoprolol CR/XL in elderly
patients with heart failure. Eur Heart J 2004;25:1300-9.
25. Packer M, Bristow MR, Cohn JN, et al. The effect of carvedilol on
morbidity and mortality in patients with chronic heart failure. N Engl J
Med 1996;334:1349-55.
26. Flather MD, Shibata MC, Coats AJ, et al. Randomized trial to determine
the effect of nebivolol on mortality and cardiovascular hospital admission
in elderly patients with heart failure (SENIORS). Eur Heart J 2005;26:
215-25.
27. Sturm B, Pacher R, Strametz-Juranek J, et al. Effect of beta 1 blockade
with atenolol on progression of heart failure in patients pretreated with
high-dose enalapril. Eur J Heart Fail 2000;2:407-12.
28. Higgins JP, Green S, eds. Cochrane handbook for systematic reviews of
interventions version 5.1.0 [updated March 2011]. Available at: www.
cochrane-handbook.org. Accessed April 2, 2013.
29. Liberati A, Altman DG, Tetzlaff J, et al. The PRISMA statement for
reporting systematic reviews and meta-analyses of studies that evaluate
healthcare interventions: explanation and elaboration. BMJ 2009;339:
b2700.
30. DerSimonian R, Laird N. Meta-analysis in clinical trials. Control Clin
Trials 1986;7:177-88.
31. Harbord RM, Egger M, Sterne JA. A modied test for small-study effects
in meta-analyses of controlled trials with binary endpoints. Stat Med
2006;25:3443-57.
Downloaded from ClinicalKey.com at Texas Tech University of Health Science Center November 20, 2016.
For personal use only. No other uses without permission. Copyright 2016. Elsevier Inc. All rights reserved.
903
32. Altman DG, Bland JM. Interaction revisited: the difference between two
estimates. BMJ 2003;326:219.
33. Higgins JP, White IR, Wood AM. Imputation methods for missing
outcome data in meta-analysis of clinical trials. Clin Trials 2008;5:
225-39.
34. Wetterslev J, Thorlund K, Brok J, Gluud C. Trial sequential analysis may
establish when rm evidence is reached in cumulative meta-analysis.
J Clin Epidemiol 2008;61:64-75.
35. Thorlund K, Imberger G, Walsh M, et al. The number of patients and
events required to limit the risk of overestimation of intervention effects
in meta-analysisea simulation study. PLoS One 2011;6:e25491.
36. Wetterslev J, Thorlund K, Brok J, Gluud C. Estimating required infor-
mation size by quantifying diversity in random-effects model meta-
analyses. BMC Med Res Methodol 2009;9:86.
37. OBrien PC, Fleming TR. A multiple testing procedure for clinical trials.
Biometrics 1979;35:549-56.
38. Cohn JN, Fowler MB, Bristow MR, et al. Safety and efcacy of carvedilol
in severe heart failure. J Card Fail 1997;3:173-9.
39. Colucci WS, Packer M, Bristow MR, et al. Carvedilol inhibits clinical
progression in patients with mild symptoms of heart failure. Circulation
1996;94:2800-6.
40. Bristow MR, Gilbert EM, Abraham WT, et al. Carvedilol produces dose-
related improvements in left ventricular function and survival in subjects
with chronic heart failure. Circulation 1996;94:2807-16.
41. Packer M, Colucci WS, Sackner-Bernstein JD, et al. Double-blind,
placebo-controlled study of the effects of carvedilol in patients with
moderate to severe heart failure: the PRECISE Trial. Circulation
1996;94:2793-9.
42. Lanfear DE, Hrobowski TN, Peterson EL, et al. Association of b-blocker
exposure with outcomes in heart failure differs between African American
and white patients. Circ Heart Fail 2012;5:202-8.
43. Simpson SH, Eurich DT, Majumdar SR, et al. A meta-analysis of the
association between adherence to drug therapy and mortality. BMJ
2006;333:15.
Supplementary Material
To access the supplementary material accompanying this
article, visit the online version of the Canadian Journal of
Cardiology at www.onlinecjc.ca and at http://dx.doi.org/10.
1016/j.cjca.2014.03.012.