CLINICAL RESEARCH STUDY

Clinical Outcomes with b-Blockers for MyocardialInfarction: A Meta-analysis of Randomized Trials

Sripal Bangalore, MD, MHA,a Harikrishna Makani, MD,b Martha Radford, MD,a Kamia Thakur, MD,a Bora Toklu, MD,c

Stuart D. Katz, MD,a James J. DiNicolantonio, PharmD,d,e P.J. Devereaux, MD, PhD,f Karen P. Alexander, MD,g

Jorn Wetterslev, MD, PhD,h Franz H. Messerli, MDbaNew York University School of Medicine, New York, NY; bSt. Luke’s Roosevelt Hospital, Mt. Sinai School of Medicine, New York, NY;cVirginia Commonwealth University, Richmond; dMid America Heart Institute, St. Luke’s Hospital, Kansas City, Mo; eWegmans Pharmacy,Ithaca, NY; fPopulation Health Research Institute, Hamilton, Ont., Canada; gDuke Clinical Research Institute, Durham, NC; hTheCopenhagen Trial Unit, Copenhagen University Hospital, Copenhagen, Denmark.

Funding: NonConflicts of I

accepting honorarpersonal financialfrom industry to suto participate in mBased on study qugrants from Abboheim, Bristol-Myehas also participate

0002-9343/$ -seehttp://dx.doi.org/1

ABSTRACT

BACKGROUND: Debate exists about the efficacy of b-blockers in myocardial infarction and their requiredduration of usage in contemporary practice.METHODS: We conducted a MEDLINE/EMBASE/CENTRAL search for randomized trials evaluatingb-blockers in myocardial infarction enrolling at least 100 patients. The primary outcome was all-causemortality. Analysis was performed stratifying trials into reperfusion-era (> 50% undergoing reperfusionor receiving aspirin/statin) or pre-reperfusion-era trials.RESULTS: Sixty trials with 102,003 patients satisfied the inclusion criteria. In the acute myocardial infarctiontrials, a significant interaction (Pinteraction ¼ .02) was noted such that b-blockers reduced mortality in the pre-reperfusion (incident rate ratio [IRR] 0.86; 95% confidence interval [CI], 0.79-0.94) but not in the reperfu-sion era (IRR 0.98; 95% CI, 0.92-1.05). In the pre-reperfusion era, b-blockers reduced cardiovascular mortality(IRR 0.87; 95% CI, 0.78-0.98), myocardial infarction (IRR 0.78; 95% CI, 0.62-0.97), and angina (IRR 0.88;95% CI, 0.82-0.95), with no difference for other outcomes. In the reperfusion era, b-blockers reduced myocardialinfarction (IRR 0.72; 95% CI, 0.62-0.83) (number needed to treat to benefit [NNTB] ¼ 209) and angina (IRR0.80; 95% CI, 0.65-0.98) (NNTB¼ 26) at the expense of increase in heart failure (IRR 1.10; 95% CI, 1.05-1.16)(number needed to treat to harm [NNTH] ¼ 79), cardiogenic shock (IRR 1.29; 95% CI, 1.18-1.41) (NNTH ¼90), and drug discontinuation (IRR 1.64; 95% CI, 1.55-1.73), with no benefit for other outcomes. Benefits forrecurrent myocardial infarction and angina in the reperfusion era appeared to be short term (30 days).CONCLUSIONS: In contemporary practice of treatment of myocardial infarction, b-blockers have no mor-tality benefit but reduce recurrent myocardial infarction and angina (short-term) at the expense of increasein heart failure, cardiogenic shock, and drug discontinuation. The guideline authors should reconsider thestrength of recommendations for b-blockers post myocardial infarction.� 2014 Elsevier Inc. All rights reserved. � The American Journal of Medicine (2014) 127, 939-953

KEYWORDS: b-blockers; Myocardial infarction; Outcomes; Reperfusion

e.nterest: PJD is part of a group that has a policy of notiums or other payments from industry for their owngain. They do accept honorariums or other paymentspport research endeavors and for reimbursement of costseetings such as scientific or advisory committee meetings.estions he originated and grants he wrote, he has receivedtt Diagnostics, Astra Zeneca, Bayer, Boehringer Ingel-rs Squibb, Covidien, Stryker, and Roche Diagnostics. Hed in an advisory board meeting for GlaxoSmithKline and

an expert panel meeting for Astra Zeneca. The remaining authors havenothing to disclose.

Authorship: All authors had access to the data and played a role inwriting this manuscript.

Requests for reprints should be addressed to Sripal Bangalore, MD,MHA, Cardiac Catheterization Laboratory, Cardiovascular OutcomesGroup, Cardiovascular Clinical Research Center, New York UniversitySchool of Medicine, The Leon H. Charney Division of Cardiology, NewYork, NY 10016.

E-mail address: sripalbangalore@gmail.com

front matter � 2014 Elsevier Inc. All rights reserved.0.1016/j.amjmed.2014.05.032

940 The American Journal of Medicine, Vol 127, No 10, October 2014

For more than a quarter of a century, b-blockers have been acornerstone in the treatment of patients with myocardialinfarction. The American College of Cardiology Foundation/American Heart Association (ACCF/AHA) ST-elevationmyocardial infarction guideline gives a class I recommen-dation for oral b-blockers within the first 24 hours in patients

CLINICAL SIGNIFICANCE

� In the treatment of patients withmyocardial infarction, b-blockers re-duced mortality in the pre-reperfusionbut not in the reperfusion era, wherethere was reduction (short-term) inmyocardial infarction and angina, butincrease in heart failure, cardiogenicshock, and drug discontinuation.

� The benefit for recurrent myocardialinfarction and angina reduction by b-blockade in the reperfusion era appearedto be short term (30 days).

with myocardial infarction, and aclass IIa indication for intravenousb-blockers for patients who arehypertensive or having ongoingischemia.1 Not surprisingly, theCenters for Medicare and MedicaidServices, the National Committeefor Quality Assurance, the NationalQuality Forum, and the Joint Com-mission on Accreditation of Health-care Organizations have adoptedb-blocker use at discharge postmyocardial infarction as a qualityindicator.

However, many of the data tosupport the use of b-blockers inmyocardial infarction predate reper-fusion and contemporary medicaltherapy with statins and antiplatelet

agents.2-4 Recent data have called into question the role of b-blockers in myocardial infarction.5-8 Moreover, there hasbeen longstanding controversy over the required duration oftreatment post myocardial infarction, with the ACCF/AHAguidelines recommending a minimum of 3 years,9 while theEuropean Society of Cardiology guidelines recommendlong-term therapy only in patients with left ventricularsystolic dysfunction.10

Our objectives were to evaluate: 1) the impact of con-temporary treatment (reperfusion/aspirin/statin) status onthe association of b-blocker use and outcomes in pa-tients with myocardial infarction; 2) the role of earlyintravenous b-blocker; and 3) the required duration ofb-blocker use.

METHODS

Study Design And Eligibility CriteriaWe performed a systematic search (using PUBMED,EMBASE, Cochrane Central Register of Controlled Trials[CENTRAL], and Google Scholar), without language restric-tion, for randomized trials using theMedical Subject Headingsterms “b-blockers” and the names of individual b-blockers,and “myocardial infarction,” until February 2013 (Week 1).

Inclusion criteria were trials comparing b-blockers withcontrols (placebo/no treatment/other active treatment) in pa-tients withmyocardial infarction enrolling at least 100 patients.

Exclusion criteria were: 1) trials comparing 2 differentb-blockers; and 2) postmyocardial infarction heart failure/left ventricular systolic dysfunction trials such as the Car-vedilol Post-Infarct Survival Control in LV Dysfunction

(CAPRICORN) trial,11 as b-blockers have been proven tobe efficacious in such cohorts.12

Trial Selection and Assessment of Risk of BiasTwo authors (KT, SB) independently reviewed trial eligi-bility and assessed risk of bias using the Cochrane Collab-

oration criteria based on thefollowing components: sequencegeneration of allocation; allocationconcealment; blinding of partici-pants, staff, and outcome assessors;incomplete outcome data; selectiveoutcome reporting; and other sour-ces of bias.13 Trials with high orunclear risk of bias for the first 3criteria were considered as highrisk of bias trials and the rest astrials with lower risk of bias.

OutcomesThe primary outcome was all-causemortality. Secondary outcomeswere cardiovascular mortality, sud-den death, recurrent myocardial

infarction, angina pectoris, heart failure, cardiogenic shock,stroke, and drug discontinuation. In trials that reported long-term outcomes beyond the randomized treatment phase,only the outcomes associated with randomized treatmentphase were extracted.

Data Extraction and SynthesisStudies have shown that the mortality rate after a myocardialinfarction falls steeply and progressively from the onset ofpain to the end of the first 48 hours.14 Therefore, trials wereclassified as acute myocardial infarction trials (randomizedwithin 48 hours of symptom onset) or postmyocardial infarc-tion trials (randomized > 48 hours of symptoms). In addition,trials were classified as reperfusion-era trials if > 50% ofpatients received reperfusion either with thrombolytics orwith revascularization or aspirin/statin. Otherwise, theywere considered to be pre-reperfusion-era trials.

Statistical AnalysisWe performed an intention-to-treat meta-analysis in linewith recommendations from the Cochrane Collaborationand the Preferred Reporting Items for Systematic Reviewsand Meta-Analyses (PRISMA) Statement15,16 using standardsoftware (STATA 9.0, StataCorp, College Station, Tex). Theanalysis used the incident rate of outcomes per 100 person-months to obtain the log incident rate ratios (IRR) of onetreatment relative to another treatment.

Analysis was performed for the acute myocardial in-farction and postmyocardial infarction cohorts separatelyafter stratifying trials based on the reperfusion-era status. Forthe primary outcome, the difference between the 2 strata

Figure 1 Study selection.

Bangalore et al b-Blockers for Myocardial Infarction 941

(pre-reperfusion vs reperfusion) was tested by a test forinteraction,17 with Pinteraction < .10 considered significantand indicated a treatment effect that differed considerably

Figure 2 b-blockers vs controls for the omyocardial infarction trials. Analysis stratifiedinterval; COMMIT¼ Clopidogrel and MetoprolEsmolol Myocardial Ischemia Trial; ICSG¼ ThIRR ¼ incident rate ratio; ISIS-1 ¼ First Internorative Group; MEMO ¼ Metoprolol-MorphinEffect of Metroprolol in Cardioprotection DurMIAMI ¼ Metoprolol in Acute Myocardial Infafor the Limitation of Infarct Size; RIMA ¼ RiStudy; TIMI ¼ Thrombolysis in Myocardial IStudy Group.

between the 2 strata. If the test for interaction for the primaryoutcome was significant, all other outcomes were interpretedseparately for the 2 strata. In addition, further analysis wasperformed categorizing trials by early initial intravenousdose vs no initial intravenous b-blocker dose to test for theeffect of intravenous b-blocker on outcomes. Finally, a seriesof landmark analyses (at 30 days post myocardial infarction,between 30-day and 1-year and > 1-year landmark timepoints) were performed to evaluate the duration of benefit ofb-blocker. Patients who died were censored at the beginningof each landmark analysis, that is, for the 30-days to 1-yearanalysis, patients who died within 30 days were excluded.

Trial Sequential Analysis. Trial sequential analysis (TSAver 0.9 Beta)18 anticipating a 10% relative risk reduction wasperformed on the primary outcome. Themethodology has beendescribed previously19,20 and is similar to interim analyses in atrial,wheremonitoring boundaries are used to decidewhether atrial could be terminated early for efficacy or for futility.

Sensitivity Analysis. Various sensitivity analyses were per-formed to test the robustness of the results. Analysis wasperformed: 1) combining acute myocardial infarction and

utcome of all-cause mortality in acuteby reperfusion status. CI ¼ confidenceol in Myocardial Infarction Trial; EMIT¼e International Collaborative Study Group;ational Study of Infarct Survival Collab-e Study Group; METOCARD-CNIC ¼ing an Acute Myocardial Infarction trial;rction; MILIS ¼ Multicenter Investigationmodellamento Infarcto Miocardico Acutonfarction; UKCSG ¼ UK Collaborative

Table 1 Baseline Characteristics of Included Trials

Trial YearSampleSize Cohort b-Blocker Control

TreatmentDuration

Revascularized(%) Quality

Ahlmark et al21 1974 162 Post MI Alprenolol Placebo 2 years NR 1Amsterdam Metoprolol Trial22 1983 584 Post MI Metoprolol Placebo 1 year NR 1Andersen et al23 1979 480 AMI Alprenolol Placebo 1 year NR 1APSI24 1997 607 Post MI Acebutolol Placebo 6 years NR 2Australian and SwedishPindolol Study25

1983 529 Post MI Pindolol Placebo 2 years NR 1

Baber et al26 1980 720 Post MI Propranolol Placebo 9 months NR 1Balcon et al27 1966 114 AMI Propranolol Placebo 28 days NR 1Barber et al28 1967 107 AMI Propranolol Placebo 1 month NR 1Barber et al29 1976 298 AMI Practolol Placebo 2 years NR 1Basu et al30 1997 151 AMI Carvedilol Placebo 6 months 95 % Streptokinase,

7% tPA1

BHAT31 1982 3837 Post MI Propranolol Placebo 2 years 9% CABG 2Briant & Norris32 1970 119 AMI Alprenolol Placebo Hospital NR 2Clausen et al33 1967 130 AMI Propranolol Placebo 14 days NR 1COMMIT5 2005 45,852 AMI Metoprolol Placebo 1 month 54.5% lytics 2CPRG34 1981 313 Post MI Oxprenolol Placebo 56 days NR 1EIS35 1984 1741 Post MI Oxprenolol Placebo 1 year NR 1EMIT36 2002 108 AMI Esmolol Placebo 6 weeks 64.5%lytics

42.5% PCI2

Federman et al37 1984 101 AMI Timolol Placebo 28 days NR 1Fuccella et al38 1968 220 AMI Oxprenolol Placebo 3 weeks NR 1Gardtman et al39 1999 262 AMI Metoprolol Placebo 1 month 22.5%Lytics 7% PTCA

(52% of patientswith MI)

1

Hjalmarson et al40 1981 1395 AMI Metoprolol Placebo 3 months NR 1Heber et al41 1987 166 AMI Labetalol Control 5 days NR 1Herlitz et al3 1988 1395 AMI Metoprolol Placebo 5 months NR 1ICSG42 1984 144 AMI Timolol Placebo Hospital stay NR 1ISIS-12 1986 16027 AMI Atenolol Control 7 days NR 2JBCMI43 2004 1090 Post MI b-blockers CCB 1.2 years 82.8% (6.3% lytics;

76.5% PCI)2

Julian et al44 1982 1456 Post MI Sotalol Placebo 1 year NR 2LIT45 1987 2395 Post MI Metoprolol Placebo 1 year 0% 1Lombardo et al46 1979 260 AMI Oxprenolol Placebo 21 days NR 1Mazur et al47 1984 204 Post MI Propranolol Placebo 1.5 years NR 1METOCARD-CNIC48 2013 270 AMI Metoprolol Controls 1 day 95% PCI 2MEMO49 1999 265 AMI Metoprolol Morphine 6 months 54% lytics 1MIAMI50 1985 5778 AMI Metoprolol Placebo 15 days NR 2MILIS51 1984 269 AMI Propranolol Placebo 9 days NR 1Multicenter Trial52 1966 195 AMI Propranolol Placebo 28 days NR 1Multicenter International53 1975 3038 Post MI Practolol Placebo 3 years NR 2Nakagomi et al54 2011 120 Post MI Atenolol Benidipine 3 years 92.5% 2Norris et al55 1984 735 AMI Propranolol Control Hospital stay NR 2Norris et al56 1968 454 AMI Propranolol Placebo 3 weeks NR 2Norwegian57 1983 1884 Post MI Timolol Placebo 33 months NR 1Norwegian MulticenterPropranolol trial58

1982 560 Post MI Propranolol Placebo 1 year NR 2

Owensby & O’Rourke59 1985 100 AMI Pindolol Placebo Hospital stay NR 2RIMA60 1999 149 AMI Metoprolol Captopril 6 months NR 1Roque et al61 1987 200 AMI Timolol Placebo 2 years NR 1Rossi et al62 1983 182 AMI Atenolol Control Hospital stay NR 1Salathia et al4 1985 800 AMI Metoprolol Placebo 1 year NR 1Schwartz et al63 1992 973 Post MI Oxprenolol Placebo 4 years NR 2Snow et al64 1966 107 AMI Propranolol Control 14 days NR 1Snow et al22 1980 143 AMI Practolol Control Hospital stay NR 1

942 The American Journal of Medicine, Vol 127, No 10, October 2014

Table 1 Continued

Trial YearSampleSize Cohort b-Blocker Control

TreatmentDuration

Revascularized(%) Quality

Stockholm Metoprolol Trial65 1988 301 Post MI Metoprolol Placebo 3 years NR 1Taylor et al66 1982 1103 Post MI Oxprenolol Placebo 4 years NR 2TIMI IIB 1991 1434 AMI Immediate

metoprololDeferredMetoprolol

6 days 100% lytics 1

Thompson et al67 1979 143 AMI Practolol Placebo 5 days NR 1Galcerá-Tomás et al68 2001 121 AMI Atenolol Captopril Hospital stay 91.5% 2UKCSG22 1984 108 AMI Timolol Placebo Hospital stay NR 1Van de Werf et al69 1993 194 AMI Atenolol Placebo Hospital stay 100% lytics 1Wilcox et al70 1980 315 AMI Oxprenolol Placebo 6 weeks NR 2Wilcox et al71 1980 261 AMI Propranolol Placebo 1 year NR 2Wilcox et al71 1980 256 AMI Atenolol Placebo 1 year NR 2Wilhelmsson et al72 1974 230 Post MI Atenolol Placebo 2 years NR 1Yusuf et al73 1983 477 AMI Atenolol Control 10 days NR 2

AMI ¼ acute myocardial infarction; APSI ¼ Acebutolol et Prévention Secondaire de l’Infarctus; BHAT ¼ Beta Blocker Heart Attack Trial; COMMIT ¼Clopidogrel and Metoprolol in Myocardial Infarction Trial; CPRG ¼ Coronary Prevention Research Group; EIS ¼ European Infarction Study; EMIT ¼ EsmololMyocardial Ischemia Trial; ICSG ¼ The International Collaborative Study Group; ISIS -1 ¼ First International Study of Infarct Survival Collaborative Group;JCBMI ¼ The Japanese beta Blockers and Calcium Antagonists Myocardial Infarction; LIT ¼ Lopressor Intervention Trial Research Group; MEMO ¼ Metoprolol-Morphine Study Group; METOCARD-CNIC ¼ Effect of Metroprolol in Cardioprotection During an Acute Myocardial Infarction trial; MIAMI ¼ Metoprolol in AcuteMyocardial Infarction; MILIS ¼ Multicenter Investigation for the Limitation of Infarct Size; PCI ¼ percutaneous coronary intervention; Post MI ¼ postmyocardial infarction; RIMA ¼ Rimodellamento Infarcto Miocardico Acuto Study; TIMI ¼ Thrombolysis in Myocardial Infarction; tPA ¼ tissue plasminogenactivator; UKCSG ¼ UK Collaborative Study Group.

Figure 3 b-blockers vs controls for the outcome of myocardial infarction in acutemyocardial infarction trials. Analysis stratified by reperfusion status. CI ¼ confidenceinterval; COMMIT ¼ Clopidogrel and Metoprolol in Myocardial Infarction Trial;EMIT ¼ Esmolol Myocardial Ischemia Trial; ICSG ¼ The International CollaborativeStudy Group; IRR ¼ incident rate ratio; ISIS-1 ¼ First International Study of InfarctSurvival Collaborative Group; MEMO ¼ Metoprolol-Morphine Study Group;METOCARD-CNIC ¼ Effect of Metroprolol in Cardioprotection During an AcuteMyocardial Infarction trial; MIAMI ¼ Metoprolol in Acute Myocardial Infarction;MILIS ¼ Multicenter Investigation for the Limitation of Infarct Size; RIMA ¼ Rimo-dellamento Infarcto Miocardico Acuto Study; TIMI ¼ Thrombolysis in MyocardialInfarction; UKCSG ¼ UK Collaborative Study Group.

Bangalore et al b-Blockers for Myocardial Infarction 943

Figure 4 b-blockers vs controls for the outcome of angina pectoris in acute myocardialinfarction trials. Analysis stratified by reperfusion status. CI ¼ confidence interval;COMMIT ¼ Clopidogrel and Metoprolol in Myocardial Infarction Trial; EMIT ¼Esmolol Myocardial Ischemia Trial; ICSG ¼ The International Collaborative StudyGroup; IRR ¼ incident rate ratio; ISIS-1 ¼ First International Study of Infarct SurvivalCollaborative Group; MEMO ¼ Metoprolol-Morphine Study Group; METOCARD-CNIC ¼ Effect of Metroprolol in Cardioprotection During an Acute MyocardialInfarction trial; MIAMI ¼ Metoprolol in Acute Myocardial Infarction; MILIS ¼Multicenter Investigation for the Limitation of Infarct Size; RIMA ¼ RimodellamentoInfarcto Miocardico Acuto Study; TIMI ¼ Thrombolysis in Myocardial Infarction;UKCSG ¼ UK Collaborative Study Group.

944 The American Journal of Medicine, Vol 127, No 10, October 2014

postmyocardial infarction trials; 2) excluding trials thatcompared b-blockers with active comparator; 3) using tradi-tional meta-analysis with counts rather than patient-months;4) restricting analyses to trial enrolling � 400 patients; 5)excluding ClOpidogrel and Metoprolol in MyocardialInfarction Trial (COMMIT); and 6) based on the qualityassessment of the trials. In addition, a meta-regression anal-ysis was performed to evaluate the relationship of percentageof patients with reperfusion in each trial on the risk ratio of b-blockers vs controls for mortality.

RESULTS

Trial SelectionWe identified 60 trials that enrolled 102,003 patientswhowerefollowed up for a mean of 10 months (range: in-hospital to 4years), with 640,891 patient-months of follow-up (Figure 1).Fourteen trials (20,418 patients) provided data on > 1-yearfollow-up. Forty trials were considered as acute myocardialinfarction trials and the rest (n ¼ 20) were postmyocardialinfarction trials (Table 1).21-73

Reperfusion-Era Status and OutcomesA majority of the trials (n ¼ 48; 31,479 patients) were in thepre-reperfusion era, with only 12 trials in the reperfusion

era (48,806 patients). The pre-reperfusion-era trials were mainlyhigh risk for bias trials (36/48 trials), whereas this proportionwas somewhat lower in the reperfusion-era trials (6/12 trials).

In the acute myocardial infarction trials, a significantinteraction (Pinteraction ¼ .02) was noted with reperfusion statussuch that b-blockers reduced mortality in the pre-reperfusionera (IRR 0.86; 95% CI, 0.79-0.94) but not in the reperfusionera (IRR 0.98; 95% CI, 0.92-1.05) (Figure 2).

In the pre-reperfusion era, b-blockers were associated withreductions in cardiovascular mortality (IRR 0.87; 95% CI,0.78-0.98), myocardial infarction (IRR 0.78; 95% CI, 0.62-0.97) (Figure 3), and angina (IRR 0.88; 95% CI, 0.82-0.95)(Figure 4), with no difference for sudden death (IRR 0.77;95% CI, 0.56-1.05), heart failure (Figure 5), cardiogenicshock (Figure 6), or stroke (IRR 2.96; 95% CI, 0.47-18.81).In the reperfusion era, b-blockers were associated withreductions in myocardial infarction (IRR 0.72; 95% CI,0.62-0.83) (number needed to treat to benefit [NNTB] ¼209) (Figure 3) and angina (IRR 0.80; 95% CI, 0.65-0.98)(NNTB ¼ 26) (Figure 4) at the expense of an increase inheart failure (IRR 1.10; 95% CI, 1.05-1.16) (number neededto treat to harm [NNTH] ¼ 79) (Figure 5), cardiogenic shock(IRR 1.29; 95% CI, 1.18-1.41) (NNTH ¼ 90) (Figure 6),and drug discontinuation (IRR 1.64; 95% CI, 1.55-1.73)(Figure 7), with no impact on cardiovascular mortality

Figure 5 b-blockers vs controls for the outcome of heart failure in acute myocardialinfarction trials. Analysis stratified by reperfusion status. CI ¼ confidence interval;COMMIT ¼ Clopidogrel and Metoprolol in Myocardial Infarction Trial; EMIT ¼Esmolol Myocardial Ischemia Trial; ICSG ¼ The International Collaborative StudyGroup; IRR ¼ incident rate ratio; ISIS-1 ¼ First International Study of Infarct SurvivalCollaborative Group; MEMO ¼ Metoprolol-Morphine Study Group; METOCARD-CNIC ¼ Effect of Metroprolol in Cardioprotection During an Acute MyocardialInfarction trial; MIAMI ¼ Metoprolol in Acute Myocardial Infarction; MILIS ¼Multicenter Investigation for the Limitation of Infarct Size; RIMA ¼ RimodellamentoInfarcto Miocardico Acuto Study; TIMI ¼ Thrombolysis in Myocardial Infarction;UKCSG ¼ UK Collaborative Study Group.

Bangalore et al b-Blockers for Myocardial Infarction 945

(IRR 1.00; 95% CI, 0.91-1.09), sudden death (IRR 0.94;95% CI, 0.86-1.01), or stroke (IRR 1.09; 95% CI, 0.91-1.30).

Results in the postmyocardial infarction trials werelargely similar (Figures 8-11).

Intravenous b-Blocker and OutcomesIn the pre-reperfusion-era trials, a significant interaction wasobserved (Pinteraction ¼ .09) such that the benefit for all-causemortality was driven by trials where early intravenousb-blocker (IRR 0.83; 95% CI, 0.75-0.92) was administered,but not in trials where b-blockers were administered orally(IRR 0.99; 95% CI, 0.83-1.19). Similarly, early intravenousb-blocker was associated with benefit for cardiovascularmortality (IRR 0.88; 95% CI, 0.78-0.99), sudden death(IRR 0.59; 95% CI, 0.38-0.91), myocardial infarction (IRR0.78; 95% CI, 0.62-0.98), and angina pectoris (IRR 0.88;95% CI, 0.82-0.95), with no difference in heart failure (IRR1.07; 95% CI, 0.97-1.18) and cardiogenic shock (IRR 1.06;95% CI, 0.89-1.27). In the reperfusion era, early intravenousb-blocker was associated with reduction in myocardialinfarction (IRR 0.72; 95% CI, 0.62-0.84) and angina pectoris

(IRR 0.80; 95% CI, 0.65-0.99), an increase in heart failure(IRR 1.10; 95% CI, 1.05-1.16) and cardiogenic shock (IRR1.29; 95% CI, 1.18-1.41), and no impact on mortality (IRR0.98; 95% CI, 0.92-1.05), cardiovascular mortality, suddendeath, and stroke.

Landmark Analysis: Required Duration ofb-Blockers UsageIn the pre-reperfusion era, b-blockers were associated withsignificant benefit at 30 days (for all-cause mortality,cardiovascular mortality, and angina), between 30 daysand 1 year (for all-cause mortality, cardiovascular mor-tality, sudden death, and myocardial infarction), and even forevents > 1 year (for all-cause mortality and sudden death)(Table 2). However, in the reperfusion era, b-blockers wereassociated with no benefit at most time points exceptmyocardial infarction and angina at 30 days, a significantincrease in heart failure, cardiogenic shock and drugdiscontinuation at 30 days, and an increase in heart failureand drug discontinuation between 30 days and 1 year(Table 2).

Figure 6 b-blockers vs controls for the outcome of cardiogenic shock in acutemyocardial infarction trials. Analysis stratified by reperfusion status. CI ¼ confidenceinterval; COMMIT ¼ Clopidogrel and Metoprolol in Myocardial Infarction Trial;EMIT ¼ Esmolol Myocardial Ischemia Trial; ICSG ¼ The International CollaborativeStudy Group; IRR ¼ incident rate ratio; ISIS-1 ¼ First International Study of InfarctSurvival Collaborative Group; MEMO ¼ Metoprolol-Morphine Study Group;METOCARD-CNIC ¼ Effect of Metroprolol in Cardioprotection During an AcuteMyocardial Infarction trial; MIAMI ¼ Metoprolol in Acute Myocardial Infarction;MILIS ¼ Multicenter Investigation for the Limitation of Infarct Size; RIMA ¼ Rimo-dellamento Infarcto Miocardico Acuto Study; TIMI ¼ Thrombolysis in MyocardialInfarction; UKCSG ¼ UK Collaborative Study Group.

946 The American Journal of Medicine, Vol 127, No 10, October 2014

Trial Sequential AnalysisThe cumulative Z-curve crosses the futility boundary,showing with confidence the lack of even a 10% reductionin the risk of mortality with b-blocker when compared withcontrols in the reperfusion era (Figure 12).

Sensitivity AnalysisVarious sensitivity analyses outlined in the methodsyielded largely similar results (data available on request).In addition, there was no benefit of b-blockers for mor-tality in the reperfusion era even after exclusion of theCOMMIT trial (IRR 0.76; 95% CI, 0.48-1.21; P ¼ .25).Furthermore, the beneficial effect of b-blockers for mor-tality in the acute myocardial infarction cohort was drivenby trials with high risk for bias (low-quality trials) (IRR0.82; 95% CI, 0.72-0.94; P ¼ .005), whereas no benefitwas observed in trials with low risk for bias (high-qualitytrials) (IRR 0.96; 95% CI, 0.91-1.02; P ¼ .18). In themeta-regression analysis, the beneficial effect of b-blockers on mortality diminished with increasing per-centage of patients with reperfusion therapy (P ¼ .056)(Figure 13).

DISCUSSIONIn patients with a myocardial infarction, a significant inter-action of reperfusion-era status on the association ofb-blocker and outcomes was seen such that while b-blockerswere associated with reduction in events, including mortalityin the pre-reperfusion era (driven by trials where early in-travenous b-blockers were administered), the benefits werereduced in the reperfusion era with reductions in myocardialinfarction and angina (short-term only) at the expense ofincreases in heart failure, cardiogenic shock, and drugdiscontinuation with no mortality benefit. The results wereconsistent in several sensitivity analyses performed to assessthe robustness of the results.

Efficacy of b-Blockers in the Reperfusion EraWhy is there a lack of efficacy of b-blockers in the reper-fusion era? Some of the considerations are the following:Are the negative results in the reperfusion-era trials due tolack of power to show a difference? Has the underlyingsubstrate changed due to reperfusion/contemporary medicaltherapy?

For the acute myocardial infarction trials, the pre-reperfusion strata with a sample size of 31,479 patients

Figure 7 b-blockers vs controls and drug discontinuation in acute myocardialinfarction trials. Analysis stratified by reperfusion status. CI ¼ confidence interval;COMMIT ¼ Clopidogrel and Metoprolol in Myocardial Infarction Trial; EMIT ¼ EsmololMyocardial Ischemia Trial; ICSG ¼ The International Collaborative Study Group; IRR ¼incident rate ratio; ISIS-1 ¼ First International Study of Infarct Survival CollaborativeGroup; MEMO ¼ Metoprolol-Morphine Study Group; METOCARD-CNIC ¼ Effect ofMetroprolol in Cardioprotection During an Acute Myocardial Infarction trial; MIAMI ¼Metoprolol in Acute Myocardial Infarction; MILIS ¼ Multicenter Investigation for theLimitation of Infarct Size; RIMA ¼ Rimodellamento Infarcto Miocardico Acuto Study;TIMI ¼ Thrombolysis in Myocardial Infarction; UKCSG ¼ UK Collaborative StudyGroup.

Bangalore et al b-Blockers for Myocardial Infarction 947

had a power of 92% to detect a hazard ratio of 0.95 forbenefit and 1.05 for harm. However, the reperfusion stratawith a sample size of 48,806 patients had a greater power of99% to detect the same hazard ratio. Thus, the power todetect a difference was, if anything, better for the reperfusionstrata. Moreover, the TSA showed that for the reperfusion-era trials, there is firm evidence to rule out even a 10%reduction in mortality with b-blockers.

In the First International Study of Infarct Survival (ISIS-1) trial, only 5% of patients were on an antiplatelet agent atdischarge, none received reperfusion, but atenolol (vs con-trols) resulted in a reduction in vascular death.2 On thecontrary, in COMMIT, all patients received aspirin, 50%received dual antiplatelet therapy, two-thirds were on anangiotensin-converting enzyme inhibitor, and 54% receivedfibrinolytics. In COMMIT, metoprolol was not superior toplacebo for both the co-primary endpoints of 30-day mor-tality and 30-day death/myocardial infarction or cardiacarrest, despite almost 3 times the sample size and greaterpower than that of the ISIS-1 trial.5 The difference thereforeappears to be both reperfusion and aggressive contemporarymedical therapy. Reperfusion and contemporary medicaltherapy modify the underlying substrate in patients with a

myocardial infarction. In the pre-reperfusion era, lack ofreperfusion and contemporary medical therapy likely resultedin extensive myocardial scarring, providing a substrate forre-entrant circuits and fatal ventricular arrhythmias. b-blockers are beneficial in this setting by preventing suddendeath, which was the major cause of mortality in the pre-reperfusion era. In the reperfusion era, prompt reperfusionreduces the likelihood of extensive scar formation. More-over, contemporary medical and device therapies are alsoefficacious at reducing the risk of arrhythmic deaths, therebyfurther reducing the impact of b-blockers.74 Conceivably,b-blockers, due to their negative inotropic effects, mayreduce myocardial contractility, which in the setting ofstunned myocardium during a myocardial infarction couldlead to heart failure and cardiogenic shock. While in the pre-reperfusion era the risk of heart failure and cardiogenicshock was likely outweighed by the benefits of preventingventricular arrhythmias and sudden death, in the reperfusionera the riskebenefit ratio no longer seems to be favorable.

A number of trials have shown that reperfusion therapy,aspirin, or statin reduces infarct size.75-78 In addition, bothstreptokinase and aspirin prevent cardiac arrest, includingdeath, due to ventricular fibrillation in the ISIS-2 trial.79

Figure 8 b-blockers vs controls for the outcome of all-cause mortality in post-myocardial infarction trials. Analysis stratified by reperfusion status. APSI ¼ Acebutololet Prévention Secondaire de l’Infarctus; BHAT ¼ Beta-Blocker Heart Attack Trial; CI ¼confidence interval; CPRG ¼ Coronary Prevention Research Group; EIS ¼ EuropeanInfarction Study; IRR ¼ incident rate ratio; JCBMI ¼ The Japanese beta Blockers andCalcium Antagonists Myocardial Infarction; LIT ¼ Lopressor Intervention TrialResearch Group.

Figure 9 b-blockers vs controls for the outcome of myocardial infarction in postmyocardial infarction trials. Analysis stratified by reperfusion status. APSI ¼ Acebutololet Prévention Secondaire de l’Infarctus; BHAT ¼ Beta-Blocker Heart Attack Trial; CI ¼confidence interval; CPRG ¼ Coronary Prevention Research Group; EIS ¼ EuropeanInfarction Study; IRR ¼ incident rate ratio; JCBMI ¼ The Japanese beta Blockers andCalcium Antagonists Myocardial Infarction; LIT ¼ Lopressor Intervention TrialResearch Group.

948 The American Journal of Medicine, Vol 127, No 10, October 2014

Figure 10 b-blockers vs controls for the outcome of heart failure in post myocardialinfarction trials. Analysis stratified by reperfusion status. APSI ¼ Acebutolol et Pré-vention Secondaire de l’Infarctus; BHAT ¼ Beta-Blocker Heart Attack Trial; CI ¼confidence interval; CPRG ¼ Coronary Prevention Research Group; EIS ¼ EuropeanInfarction Study; IRR ¼ incident rate ratio; JCBMI ¼ The Japanese beta Blockers andCalcium Antagonists Myocardial Infarction; LIT ¼ Lopressor Intervention TrialResearch Group.

Figure 11 b-blockers vs controls and drug discontinuation in postmyocardial infarc-tion trials. Analysis stratified by reperfusion status. APSI ¼ Acebutolol et PréventionSecondaire de l’Infarctus; BHAT ¼ Beta-Blocker Heart Attack Trial; CI ¼ confidenceinterval; CPRG ¼ Coronary Prevention Research Group; EIS ¼ European InfarctionStudy; IRR ¼ incident rate ratio; JCBMI ¼ The Japanese beta Blockers and CalciumAntagonists Myocardial Infarction; LIT ¼ Lopressor Intervention Trial Research Group.

Bangalore et al b-Blockers for Myocardial Infarction 949

Table2

Land

markAn

alyses:b-Blockers

vsControls(From

Fixed-effect

Mod

el)

Death

CVDe

ath

Sudd

enDe

ath

MI

Angina

Stroke

HeartFailu

reCardiogenic

Shock

Withd

rawal

Events

at30

days

Pre-reperfusion

0.87

(0.79,

0.96

)0.86

(0.77,

0.96

)0.82

(0.59,

1.13

)0.81

(0.63,1.04

)0.89

(0.83,

0.95

)2.96

(0.47,

18.81)

1.06

(0.97,

1.16

)1.03

(0.87,

1.21

)1.11

(1.00,

1.23

)Reperfusionera

0.98

(0.92,

1.05

)1.00

(0.91,1.10

)0.94

(0.86,

1.01

)0.72

(0.62,

0.84

)0.81

(0.66,

1.00

)1.09

(0.91,

1.30

)1.10

(1.05,

1.16

)1.29

(1.18,

1.41

)1.64

(1.55,

1.73

)Events

between

30days

and1year

Pre-reperfusion

0.79

(0.71,

0.88

)0.84

(0.71,

1.00

)0.61

(0.49,

0.76

)0.77

(0.64,

0.91

)0.94

(0.75,

1.18

)1.54

(0.60,

3.95

)1.07

(0.91,

1.27

)1.88

(0.51,

6.96

)1.16

(1.03,

1.30

)Reperfusionera

1.50

(0.53,

4.21

)1.50

(0.53,

4.21

)NA

0.71

(0.23,

2.25

)1.03

(0.72,

1.48

)4.00

(0.45,

35.79)

3.83

(1.56,

9.41

)NA

1.49

(1.01,

2.19

)Events

>1year

Pre-reperfusion

0.81

(0.66,

0.98

)0.73

(0.48,

1.11

)0.64

(0.43,

0.97

)0.81

(0.62,

1.06

)NA

0.20

(0.01,

4.20

)0.25

(0.03,

2.25

)NA

1.00

(0.65,

1.54

)Reperfusionera

NANA

NANA

NANA

NANA

NA

Figure 12 Trial Sequential Analysis using fixed-effect meta-analysis in the reperfusion era. The required information of49,990 patients is based on an anticipated intervention effect of10% relative risk reduction, a control event proportion of 7.36%(estimated from the cumulated comparator event proportion),absence of heterogeneity (diversity ¼ 0%), and a ¼ 0.05 andb ¼ 0.10.

950 The American Journal of Medicine, Vol 127, No 10, October 2014

There is, thus, ample evidence to suggest that the underlyingsubstrate is altered by the use of these therapies in patientswith myocardial infarction.

Clinical ImplicationsBased on the above data, it may be reasonable to concludethat in patients who develop extensive scars (patients withdelayed presentation and large myocardial infarction) andtherefore are prone to develop heart failure or ventriculararrhythmias, b-blockers will remain highly efficacious inpreventing events, as has been shown in numerous heartfailure trials,12,80,81 and in preventing ventricular arrhyth-mias and sudden death. One may be tempted to conclude

Figure 13 Meta-regression analysis of the relationship ofpercentage of patients with reperfusion therapy on the risk ratioof mortality with b-blockers.

Bangalore et al b-Blockers for Myocardial Infarction 951

from the pre-reperfusion-era trials that b-blockers will alsobe efficacious in patients with myocardial infarction treatedconservatively (that is, no reperfusion). However, in theCOMMIT trial,5 there was no benefit of b-blockers formortality in patients who did or did not receive fibrinolytictherapy, likely underscoring the role of contemporarymedical therapy in patients who are treated conservatively.In addition, one may consider b-blockers short term (30days) after a myocardial infarction to reduce the risk ofrecurrent myocardial infarction and angina, but this has tobe weighed against the potential harm of heart failure andcardiogenic shock.

Study LimitationsThe results in the reperfusion era are driven by theCOMMIT trial. However, in the sensitivity analysisexcluding COMMIT, there was still no benefit of b-blockersfor mortality in the reperfusion era. The categorization ofpre-reperfusion vs reperfusion era was not done based oncalendar years, as there was wide variability in the use ofmedication and reperfusion. Moreover, our results wereconsistent in the sensitivity analysis where percentage ofreperfusion was considered for each trial as a continuousvariable in the meta-regression analysis rather than artificialcategorization into pre-reperfusion vs reperfusion era. Wewere unable to separate out the effect of reperfusion frommodern medical therapy given the limitations of a trial-levelmeta-analysis. Moreover, although a significant benefit wasnoted for b-blockers in the pre-reperfusion era, most of thetrials were high risk for bias.

CONCLUSIONSIn this analysis of b-blockers in myocardial infarction,a significant interaction of reperfusion-era status on theassociation of b-blocker and outcomes was seen, in thatb-blocker reduced the risk of events, including mortality inthe pre-reperfusion-era trial, but not in the reperfusion-eratrials. In patients undergoing contemporary treatment, datasupports use of b-blockers short term (30 days) to reducerecurrent myocardial infarction and angina, but this hasto be weighed at the expense of increase in heart failure,cardiogenic shock, and drug discontinuation, with no mor-tality benefit. Guidelines should reconsider the current rec-ommendations for b-blockers for myocardial infarction,especially in patients undergoing contemporary treatment.

References1. O’Gara PT, Kushner FG, Ascheim DD, et al. 2013 ACCF/AHA

guideline for the management of ST-elevation myocardial infarction:executive summary: a report of the American College of CardiologyFoundation/American Heart Association Task Force on PracticeGuidelines. Circulation. 2013;127(4):529-555.

2. Randomised trial of intravenous atenolol among 16 027 cases of sus-pected acute myocardial infarction: ISIS-1. First International Study ofInfarct Survival Collaborative Group. Lancet. 1986;2(8498):57-66.

3. Herlitz J, Hjalmarson A, Swedberg K, et al. Effects on mortality duringfive years after early intervention with metoprolol in suspected acutemyocardial infarction. Acta Med Scand. 1988;223(3):227-231.

4. Salathia KS, Barber JM, McIlmoyle EL, et al. Very early interventionwith metoprolol in suspected acute myocardial infarction. Eur Heart J.1985;6(3):190-198.

5. Chen ZM, Pan HC, Chen YP, et al. Early intravenous then oralmetoprolol in 45,852 patients with acute myocardial infarction: rand-omised placebo-controlled trial. Lancet. 2005;366(9497):1622-1632.

6. Bangalore S, Steg G, Deedwania P, et al. beta-Blocker use and clinicaloutcomes in stable outpatients with and without coronary artery disease.JAMA. 2012;308(13):1340-1349.

7. Danchin N, Laurent S. Coronary artery disease. Are beta-blockers trulyhelpful in patients with CAD? Nat Rev Cardiol. 2013;10(1):11-12.

8. Ozasa N, Morimoto T, Bao B, et al. beta-Blocker use in patients afterpercutaneous coronary interventions: one size fits all? Worse outcomesin patients without myocardial infarction or heart failure. Int J Cardiol.2012;76(8):1889-1894.

9. Smith SC Jr, Benjamin EJ, Bonow RO, et al. AHA/ACCF SecondaryPrevention and Risk Reduction Therapy for Patients with Coronary andother Atherosclerotic Vascular Disease: 2011 update: a guideline fromthe American Heart Association and American College of CardiologyFoundation. Circulation. 2011;124(22):2458-2473.

10. Hamm CW, Bassand JP, Agewall S, et al. ESC Guidelines for themanagement of acute coronary syndromes in patients presentingwithout persistent ST-segment elevation: the Task Force for the man-agement of acute coronary syndromes (ACS) in patients presentingwithout persistent ST-segment elevation of the European Society ofCardiology (ESC). Eur Heart J. 2011;32(23):2999-3054.

11. Dargie HJ. Effect of carvedilol on outcome after myocardial infarctionin patients with left-ventricular dysfunction: the CAPRICORN rando-mised trial. Lancet. 2001;357(9266):1385-1390.

12. Effect of metoprolol CR/XL in chronic heart failure: MetoprololCR/XL Randomised Intervention Trial in Congestive Heart Failure(MERIT-HF). Lancet. 1999;353(9169):2001-2007.

13. Higgins JPT, Altman DG. Assessing risk of bias in included studies. In:Higgins JPT, Green S, eds. Cochrane Handbook for Systematic Re-views of Interventions, version 5.0.0. Oxford, UK: The CochraneCollaboration; 2008.

14. Mittra B. Potassium, glucose, and insulin in treatment of myocardialinfarction. Lancet. 1965;2(7413):607-609.

15. Moher D, Cook DJ, Eastwood S, et al. Improving the quality of reportsof meta-analyses of randomised controlled trials: the QUOROMstatement. Quality of Reporting of Meta-analyses. Lancet. 1999;354(9193):1896-1900.

16. Higgins J, Green S. Cochrane Handbook for Systematic Reviews ofInterventions Version, version 5.0.0. Oxford, UK: The CochraneCollaboration; 2008.

17. Altman DG, Bland JM. Interaction revisited: the difference betweentwo estimates. BMJ. 2003;326(7382):219.

18. Thorlund K, Engstrøm J, Wetterslev J, Brok J, Imberger G, Gluud C.User Manual for Trial Sequential Analysis (TSA). Copenhagen, Denmark:Centre for Clinical Intervention Research: Available at: http://www.ctu.dk/tsa/files/tsa_manual.pdf; 2011:Accessed June 27, 2014.

19. Wetterslev J, Thorlund K, Brok J, Gluud C. Trial sequential analysismay establish when firm evidence is reached in cumulative meta-analysis. J Clin Epidemiol. 2008;61(1):64-75.

20. Wetterslev J, Thorlund K, Brok J, Gluud C. Estimating required in-formation size by quantifying diversity in random-effects model meta-analyses. BMC Med Res Methodol. 2009;9:86.

21. Ahlmark G, Saetre H, Korsgren M. Letter: Reduction of sudden deathsafter myocardial infarction. Lancet. 1974;2(7896):1563.

22. Yusuf S, Peto R, Lewis J, et al. Beta blockade during and aftermyocardial infarction: an overview of the randomized trials. ProgCardiovasc Dis. 1985;27(5):335-371.

23. Andersen MP, Bechsgaard P, Frederiksen J, et al. Effect of alprenololon mortality among patients with definite or suspected acute myocar-dial infarction. Preliminary results. Lancet. 1979;2(8148):865-868.

952 The American Journal of Medicine, Vol 127, No 10, October 2014

24. Cucherat M, Boissel JP, Leizorovicz A. Persistent reduction of mortalityfor five years after one year of acebutolol treatment initiated during acutemyocardial infarction. The APSI Investigators. Acebutolol et PreventionSecondaire de l’Infarctus. Am J Cardiol. 1997;79(5):587-589.

25. The effect of pindolol on the two years mortality after complicatedmyocardial infarction. Eur Heart J. 1983;4(6):367-375.

26. Baber NS, Evans DW, Howitt G, et al. Multicentre post-infarction trialof propranolol in 49 hospitals in the United Kingdom, Italy, andYugoslavia. Br Heart J. 1980;44(1):96-100.

27. Balcon R, Jewitt DE, Davies JP, Oram S. A controlled trial of pro-pranolol in acute myocardial infarction. Lancet. 1966;2(7470):918-920.

28. Barber JM, Murphy FM, Merrett JD. Clinical trial of propranolol inacute myocardial infarction. Ulster Med J. 1967;36(2):127-130.

29. Barber JM, Boyle DM, Chaturvedi NC, et al. Practolol in acutemyocardial infarction. Acta Med Scand Suppl. 1976;587:213-219.

30. Basu S, Senior R, Raval U, et al. Beneficial effects of intravenous andoral carvedilol treatment in acute myocardial infarction. A placebo-controlled, randomized trial. Circulation. 1997;96(1):183-191.

31. A randomized trial of propranolol in patients with acute myocardialinfarction. I. Mortality results. JAMA. 1982;247(12):1707-1714.

32. Briant RB, Norris RM. Alprenolol in acute myocardial infarction:double-blind trial. N Z Med J. 1970;71(454):135-138.

33. Clausen J, Felsby M, Jorgensen FS, et al. Absence of prophylacticeffect of propranolol in myocardial infarction. Lancet. 1966;2(7470):920-924.

34. An early intervention secondary prevention study with oxprenololfollowing myocardial infarction. Eur Heart J. 1981;2(5):389-393.

35. European Infarction Study (E.I.S.). A secondary prevention study withslow release oxprenolol after myocardial infarction: morbidity andmortality. Eur Heart J. 1984;5(3):189-202.

36. Mitchell RG, Stoddard MF, Ben-Yehuda O, et al. Esmolol in acuteischemic syndromes. Am Heart J. 2002;144(5):E9.

37. Federman J, Pitt A, Tonkin A, et al. Australian trial of intravenous and oraltimolol in acute myocardial infarction. Circulation. 1984:70(Part 2):57th.

38. Fuccella LM. Trasicor: review of the pharmacology and clinical results.S Afr Med J. 1969;Dec 6:Suppl:7-14.

39. Gardtman M, Dellborg M, Brunnhage C, et al. Effect of intravenousmetoprolol before hospital admission on chest pain in suspected acutemyocardial infarction. Am Heart J. 1999;137(5):821-829.

40. Hjalmarson A, Elmfeldt D, Herlitz J, et al. Effect on mortality ofmetoprolol in acute myocardial infarction. A double-blind randomisedtrial. Lancet. 1981;2(8251):823-827.

41. Heber ME, Rosenthal E, Thomas N, et al. Effect of labetalol on indicesof myocardial necrosis in patients with suspected acute infarction. EurHeart J. 1987;8(1):11-18.

42. Reduction of infarct size by the early use of intravenous timolol inacute myocardial infarction. International Collaborative Study Group.Am J Cardiol. 1984;54(11):14E-15E.

43. Japanese beta-blockers and Calcium Antagonists Myocardial Infarc-tion (JBCMI) Investigators. Comparison of the effects of betablockers and calcium antagonists on cardiovascular events after acutemyocardial infarction in Japanese subjects. Am J Cardiol. 2004;93(8):969-973.

44. Julian DG, Prescott RJ, Jackson FS, Szekely P. Controlled trial ofsotalol for one year after myocardial infarction. Lancet. 1982;1(8282):1142-1147.

45. The Lopressor Intervention Trial: multicentre study of metoprolol insurvivors of acute myocardial infarction. Lopressor Intervention TrialResearch Group. Eur Heart J. 1987;8(10):1056-1064.

46. Lombardo M, Selvini A, Motolese M, et al., eds. Beta-blockingtreatment in 440 cases of acute myocardial infarction: a study withoxprenolol. Proceedings of Florence International Meeting onMyocardial Infarction. Amsterdam: Excerpta Medica; 1979.

47. Mazur N, Kulginskaya I, Ivanova L, et al. Results of long-term pro-pranolol treatment in myocardial infarction survivors with advancedgrades of ventricular extrasystoles. Cor et vasa. 1984;26(4):241.

48. Ibanez B, Macaya C, Sánchez-Brunete V, et al. Effect of early meto-prolol on infarct size in ST-segment elevation myocardial infarction

patients undergoing primary percutaneous coronary intervention: theEffect of Metroprolol in Cardioprotection During an Acute MyocardialInfarction (METOCARD-CNIC) Trial. Circulation. 2013;128(14):1495-1503.

49. Everts B, Karlson B, Abdon NJ, et al. A comparison of metoprolol andmorphine in the treatment of chest pain in patients with suspected acutemyocardial infarctionethe MEMO study. J Intern Med. 1999;245(2):133-141.

50. Group MTR. Metoprolol in acute myocardial infarction (MIAMI).A randomised placebo-controlled international trial. Eur Heart J. 1985;6(3):199-226.

51. Roberts R, Croft C, Gold HK, et al. Effect of propranolol on myocardial-infarct size in a randomized blinded multicenter trial. N Engl J Med.1984;311(4):218-225.

52. Propranolol in acute myocardial infarction. A multicentre trial. Lancet.1966;2(7479):1435-1438.

53. Improvement in prognosis of myocardial infarction by long-term beta-adrenoreceptor blockade using practolol. A multicentre internationalstudy. Br Med J. 1975;3(5986):735-740.

54. Nakagomi A, Kodani E, Takano H, et al. Secondary preventive effectsof a calcium antagonist for ischemic heart attack: randomized parallelcomparison with beta-blockers. Circ J. 2011;75(7):1696-1705.

55. Norris RM, Barnaby PF, Brown MA, et al. Prevention of ventricularfibrillation during acute myocardial infarction by intravenous propranolol.Lancet. 1984;2(8408):883-886.

56. Norris RM, Caughey DE, Scott PJ. Trial of propranolol in acutemyocardial infarction. Br Med J. 1968;2(5602):398-400.

57. Pedersen TR. The Norwegian Multicenter Study of Timolol afterMyocardial Infarction. Circulation. 1983;67(6 Pt 2):I49-I53.

58. Hansteen V, Moinichen E, Lorentsen E, et al. One year’s treatmentwith propranolol after myocardial infarction: preliminary report ofNorwegian multicentre trial. Br Med J (Clin Res Ed). 1982;284(6310):155-160.

59. Owensby DA, O’Rourke MF. Failure of intravenous pindolol to reducethe hemodynamic determinants of myocardial oxygen demand orenzymatically determined infarct size in acute myocardial infarction.Aust N Z J Med. 1985;15(6):704-711.

60. Coletta C, Ricci R, Ceci V, et al. Effects of early treatment withcaptopril and metoprolol singly or together on six-month mortality andmorbidity after acute myocardial infarction. Results of the RIMA(Rimodellamento Infarto Miocardico Acuto) study. The RIMA re-searchers. G Ital Cardiol. 1999;29(2):115-124; discussion 125e129.

61. Roque F, Amuchastegui LM, Lopez Morillos MA, et al. Beneficialeffects of timolol on infarct size and late ventricular tachycardiain patients with acute myocardial infarction. Circulation. 1987;76(3):610-617.

62. Rossi PR, Yusuf S, Ramsdale D, et al. Reduction of ventricular ar-rhythmias by early intravenous atenolol in suspected acute myocardialinfarction. Br Med J (Clin Res Ed). 1983;286(6364):506-510.

63. Schwartz PJ, Motolese M, Pollavini G, et al. Prevention of suddencardiac death after a first myocardial infarction by pharmacologic orsurgical antiadrenergic interventions. J Cardiovasc Electrophysiol. 1992;3(1):2-16.

64. Snow P. Treatment of acute myocardial infarction with propranolol.Am J Cardiol. 1966;18(3):458-462.

65. Olsson G. Thromboatherosclerotic complications in hypertensives:results of the Stockholm Metoprolol (secondary prevention) Trial. AmHeart J. 1988;116(1):334-338.

66. Taylor S, Silke B, Ebbutt A, et al. A long-term prevention study withoxprenolol in coronary heart disease. N Engl J Med. 1982;307(21):1293-1301.

67. Thompson PL, Fletcher EE, Katavatis V. Enzymatic indices of myocar-dial necrosis: influence on short-and long-term prognosis after myocardialinfarction. Circulation. 1979;59(1):113-119.

68. Galcerá-Tomás J, Castillo-Soria FJ, Villegas-García M, et al. Effects ofearly use of atenolol or captopril on infarct size and ventricular volumeA double-blind comparison in patients with anterior acute myocardialinfarction. Circulation. 2001;103(6):813-819.

Bangalore et al b-Blockers for Myocardial Infarction 953

69. Van De Werf F, Janssens L, Brzostek T, et al. Short-term effects ofearly intravenous treatment with a beta-adrenergic blocking agent or aspecific bradycardiac agent in patients with acute mycardial infarctionreceiving thrombolytic therapy. J Am Coll Cardiol. 1993;22(2):407-416.

70. Wilcox R, Hampton J, Rowley J, et al. Randomised placebo-controlledtrial comparing oxprenolol with disopyramide phosphate in immediatetreatment of suspected myocardial infarction. Lancet. 1980;316(8198):765-769.

71. Wilcox R, Roland J, Banks D, et al. Randomised trial comparingpropranolol with atenolol in immediate treatment of suspected myo-cardial infarction. Br Med J. 1980;280(6218):885-888.

72. Wilhelmsson C, Vedin JA, Wilhelmsen L, et al. Reduction of suddendeaths after myocardial infarction by treatment with alprenolol. Pre-liminary results. Lancet. 1974;2(7890):1157-1160.

73. Yusuf S, Sleight P, Rossi P, et al. Reduction in infarct size, arrhythmiasand chest pain by early intravenous beta blockade in suspected acutemyocardial infarction. Circulation. 1983;67(6 Pt 2):I32.

74. Exner DV, Reiffel JA, Epstein AE, et al. Beta-blocker use and survivalin patients with ventricular fibrillation or symptomatic ventriculartachycardia: the Antiarrhythmics Versus Implantable Defibrillators(AVID) trial. J Am Coll Cardiol. 1999;34(2):325-333.

75. Simoons ML, Serruys PW, van den Brand M, et al. Early thrombolysisin acute myocardial infarction: limitation of infarct size and improvedsurvival. J Am Coll Cardiol. 1986;7(4):717-728.

76. A prospective trial of intravenous streptokinase in acute myocar-dial infarction (I.S.A.M.).Mortality,morbidity, and infarct size at 21days.The I.S.A.M. Study Group. N Engl J Med. 1986;314(23):1465-1471.

77. Col NF, Yarzbski J, Gore JM, et al. Does aspirin consumption affectthe presentation or severity of acute myocardial infarction? Arch InternMed. 1995;155(13):1386-1389.

78. Wolfrum S, Grimm M, Heidbreder M, et al. Acute reduction ofmyocardial infarct size by a hydroxymethyl glutaryl coenzyme Areductase inhibitor is mediated by endothelial nitric oxide synthase.J Cardiovasc Pharmacol. 2003;41(3):474-480.

79. Randomised trial of intravenous streptokinase, oral aspirin, both, orneither among 17,187 cases of suspected acute myocardial infarction:ISIS-2. ISIS-2 (Second International Study of Infarct Survival)Collaborative Group. Lancet. 1988;2(8607):349-360.

80. The Cardiac Insufficiency Bisoprolol Study II (CIBIS-II): a randomisedtrial. Lancet. 1999;353(9146):9-13.

81. Heidenreich PA, Lee TT, Massie BM. Effect of beta-blockade onmortality in patients with heart failure: a meta-analysis of randomizedclinical trials. J Am Coll Cardiol. 1997;30(1):27-34.