JOURNAL OF THE AMERICAN COLLEGE OF CARDIOLOGY VOL. 67, NO.

23, 2016

2016 BY THE AMERICAN COLLEGE OF CARDIOLOGY FOUNDATION ISSN 0735-1097/$36.00

PUBLISHED BY ELSEVIER http://dx.doi.org/10.1016/j.jacc.2016.03.522

ORIGINAL INVESTIGATIONS

Early Intravenous Beta-Blockers in

Patients With ST-Segment Elevation
Myocardial Infarction Before Primary
Percutaneous Coronary Intervention
Vincent Roolvink, MD,a Borja Ibez, MD, PHD,b,c Jan Paul Ottervanger, MD, PHD,a Gonzalo Pizarro, MD,b,d
Niels van Royen, MD, PHD,e Alonso Mateos, MD,f Jan-Henk E. Dambrink, MD, PHD,a Noemi Escalera, BPT,b
Erik Lipsic, MD, PHD,g Agustn Albarran, MD, PHD,h,i Antonio Fernndez-Ortiz, MD, PHD,h,j
Francisco Fernndez-Avils, MD, PHD,h,k Javier Goicolea, MD, PHD,h,l Javier Botas, MD, PHD,h,m Wouter Remkes, MD,a
Victoria Hernandez-Jaras, PHARMD,f Elvin Kedhi, MD, PHD,a Jos L. Zamorano, MD, PHD,h,n
Felipe Navarro, MD, PHD,c,h Fernando Alfonso, MD, PHD,h,o Alberto Garca-Lled, MD, PHD,h,p
Joaquin Alonso, MD, PHD,h,q Maarten van Leeuwen, MD,e Robin Nijveldt, MD, PHD,e Sonja Postma, PHD,r
Evelien Kolkman, MSC,r Marcel Gosselink, MD, PHD,a Bart de Smet, MD, PHD,s Saman Rasoul, MD, PHD,t
Jan J. Piek, MD, PHD,u Valentin Fuster, MD, PHD,b,v Arnoud W.J. van t Hof, MD, PHD,a
on behalf of the EARLY-BAMI Investigators

ABSTRACT

BACKGROUND The impact of intravenous (IV) beta-blockers before primary percutaneous coronary intervention (PPCI)
on infarct size and clinical outcomes is not well established.

OBJECTIVES This study sought to conduct the rst double-blind, placebo-controlled international multicenter study
testing the effect of early IV beta-blockers before PPCI in a general ST-segment elevation myocardial infarction (STEMI)
population.

METHODS STEMI patients presenting <12 h from symptom onset in Killip class I to II without atrioventricular block were
randomized 1:1 to IV metoprolol (2
5-mg bolus) or matched placebo before PPCI. Primary endpoint was myocardial
infarct size as assessed by cardiac magnetic resonance imaging (CMR) at 30 days. Secondary endpoints were enzymatic
infarct size and incidence of ventricular arrhythmias. Safety endpoints included symptomatic bradycardia, symptomatic
hypotension, and cardiogenic shock.

RESULTS A total of 683 patients (mean age 62  12 years; 75% male) were randomized to metoprolol (n 336) or
placebo (n 346). CMR was performed in 342 patients (54.8%). Infarct size (percent of left ventricle [LV]) by CMR did not
differ between the metoprolol (15.3  11.0%) and placebo groups (14.9  11.5%; p 0.616). Peak and area under the
creatine kinase curve did not differ between both groups. LV ejection fraction by CMR was 51.0  10.9% in the metoprolol
group and 51.6  10.8% in the placebo group (p 0.68). The incidence of malignant arrhythmias was 3.6% in the
metoprolol group versus 6.9% in placebo (p 0.050). The incidence of adverse events was not different between groups.

Listen to this manuscripts

audio summary by
JACC Editor-in-Chief
Dr. Valentin Fuster.
CONCLUSIONS In a nonrestricted STEMI population, early intravenous metoprolol before PPCI was not associated with
a reduction in infarct size. Metoprolol reduced the incidence of malignant arrhythmias in the acute phase and was not
associated with an increase in adverse events. (Early-Beta blocker Administration before reperfusion primary PCI in
patients with ST-elevation Myocardial Infarction [EARLY-BAMI]; EudraCT no: 2010-023394-19)
(J Am Coll Cardiol 2016;67:270515) 2016 by the American College of Cardiology Foundation.
2706 Roolvink et al. JACC VOL. 67, NO. 23, 2016

The EARLY-BAMI Trial JUNE 14, 2016:270515

D Infarction) trial (N 220) showed that IV metoprolol

ABBREVIATIONS espite advances in the care for
AND ACRONYMS patients with ST-segment elevation
myocardial infarction
BP = blood pressure
mortality in these patients remains relatively
CK = creatine kinase
high, especially in an all-comer population
CMR = cardiac magnetic
(1). Although early diagnosis and treatment
resonance imaging
have improved outcome of these patients,
IV = intravenous
additional interventions early after onset of
LV = left ventricular
ischemia might further enhance outcomes.
LVEF = left ventricular
The clinical guidelines recommend treatment
ejection fraction
with beta-blockers for STEMI patients (2),
MACE = major adverse
cardiac events
although evidence of mortality reduction
with beta-blockers after reperfusion therapy
MI = myocardial infarction
is limited (24).
PPCI = primary percutaneous
coronary intervention Whether administration before reperfu-
STEMI = ST-segment elevation
sion improves clinical outcome or reduces
myocardial infarction myocardial infarct size is less clear. Experi-
TIMI = Thrombolysis In mental studies presented conicting results
Myocardial Infarction as to whether beta-blockers decrease extent
of myocardial necrosis (46). In clinical studies in
STEMI, the effect of early b-blockade was mostly
studied in the pre-reperfusion era, with inconclusive
results (710). In the era of thrombolysis, 2 random-
ized controlled trials testing the effect of beta-
blockers in STEMI showed no reduction in mortality
with b -blockade (11,12). In patients treated by primary
percutaneous coronary intervention (PPCI), only
2 randomized controlled trials studied the effect
of early beta-blocker treatment. In 1 small study
(N 96), Hanada et al. (13) observed that continuous
intravenous (IV) landiolol immediately after PPCI was
associated with improved left ventricular (LV) func-
tion. The METOCARD-CNIC (Effect of Metoprolol
in Cardioprotection During an Acute Myocardial
administered before PPCI reduced infarct size and
(STEMI), preserved LV function (1416). In the latter trial,
however, only patients with an anterior STEMI were
included, and the trial was neither blinded nor
placebo controlled.
SEE PAGE 2716
We present the results of the EARLY-BAMI (Early
Beta-blocker Administration before primary PCI in
patients with ST-elevation Myocardial Infarction)
trial, the rst double-blinded, placebo-controlled
multicenter international study assessing the effect
of early IV beta-blocker therapy before PPCI in a less
restricted STEMI population (17).
METHODS
The primary objective of the EARLY-BAMI trial was to
assess the effect of early, pre-hospital, pre-reperfu-
sion administration of IV metoprolol on myocardial
infarct size in patients with STEMI eligible for PPCI.
The study design has previously been published (17).
The EARLY-BAMI trial was a double-blind, placebo-
controlled randomized clinical trial involving a total
of 5 PCI centers and 3 ambulance services in the
Netherlands, and 9 PCI centers and 1 ambulance ser-
vice in Spain. All centers had longstanding experience
in the pre-hospital diagnosis, triage, and treatment of
STEMI patients in the ambulance and were part of a
STEMI network. The study was approved by the
medical ethical committees of the participating hos-
pitals. Trial funding came from a research grant of the
Dutch Heart Foundation (Utrecht, the Netherlands,

From the aDepartment of Cardiology, Isala Hospital, Zwolle, the Netherlands; bDepartment of Cardiology, Centro Nacional de
Investigaciones Cardiovasculares Carlos III (CNIC), Madrid, Spain; cDepartment of Cardiology, IIS-Fundacion Jimenez Daz Hos-
pital, Madrid, Spain; dDepartment of Cardiology, Hospital Universitario Quirn, Universidad Europea de Madrid & Hospital Ruber-
Quirnsalud, Madrid, Spain; eDepartment of Cardiology, VU University Medical Center, Amsterdam, the Netherlands; fDepartment
of Cardiology, Servicio de Urgencia Medica de Madrid (SUMMA 112), Madrid, Spain; gDepartment of Cardiology, University Medical
Center Groningen, Groningen, the Netherlands; hDepartment of Cardiology, Codigo Infarto, Madrid, Spain; iDepartment of Car-
diology, Hospital 12 de Octubre, Madrid, Spain; jDepartment of Cardiology, Hospital Clnico San Carlos, Madrid, Spain; kDepartment
of Cardiology, Hospital Gregorio Maran, Madrid, Spain; lDepartment of Cardiology, Hospital Puerta de Hierro, Madrid, Spain;
m
Department of Cardiology, Hospital Fundacin Alcorcn, Madrid, Spain; nDepartment of Cardiology, Hospital Ramn y Cajal,
Madrid, Spain; oDepartment of Cardiology, Hospital de La Princesa, Madrid, Spain; pDepartment of Cardiology, Hospital Prncipe de
Asturias, Alcal de Henares, Madrid, Spain; qDepartment of Cardiology, Hospital de Getafe, Madrid, Spain; rDiagram, Diagnostic
Research and Management, Zwolle, the Netherlands; sDepartment of Cardiology, Meander Medisch Centrum, Amersfoort, the
Netherlands; tDepartment of Cardiology, Maastricht University Medical Center, Maastricht, the Netherlands; uDepartment of
Cardiology, Academic Medical Center, Amsterdam, the Netherlands; and vThe Zena and Michael A Wiener Cardiovascular Institute,
Icahn School of Medicine at Mount Sinai, New York, New York. The Early-BAMI trial was an investigator-initiated, noncommercial
trial. The trial was funded by a research grant from the Dutch Heart Foundation (Utrecht, the Netherlands) (no. 2010B125) and an
unrestricted grant of Medtronic Inc. (Heerlen, the Netherlands). QMass software version MR 7.6 was partially supported by a
scientic collaboration between CNIC and Medis. Dr. Botas has been a consultant for Terumo. Dr. van t Hof has received speakers
fees from AstraZeneca, Iroko, and Daiichi-Sankyo; has received nonpersonal grants from Medtronic and Daiichi-Sankyo to his
research institution. All other authors have reported that they have no relationships relevant to the contents of this paper to
disclose. Christopher Granger, MD, served as Guest Editor for this paper.

Manuscript received March 17, 2016; revised manuscript received March 24, 2016, accepted March 24, 2016.
JACC VOL. 67, NO. 23, 2016
JUNE 14, 2016:270515
no. 2010B125) and an unrestricted grant by Medtronic
Inc. (Heerlen, the Netherlands), which was used for
additional analyses.
STUDY PROTOCOL. Patients >18 years of age with
symptoms of acute STEMI for >30 min but <12 h,
plus ST-segment elevation >1 mV in 2 adjacent
electrocardiogram leads or new left bundle branch
block were eligible for enrollment. The diagnosis
of STEMI was made by the ambulance medical
personnel. Electrocardiogram transmission to a their treating physician. Follow-up included outpa-
physician at the PCI center was allowed to conrm
the diagnosis. After the in-ambulance diagnosis of
STEMI, medical treatment proceeded per current
guidelines: administration of 500 mg of IV aspirin,
600 mg of clopidogrel or 180 mg of ticagrelor orally,
and 5,000 international units of IV unfractionated
heparin. Exclusion criteria were Killip class III and IV,
systolic blood pressure (BP) <100 mm Hg, heart
rate <60 beats/min, type II and III atrioventricular
block, history of previous myocardial infarction (MI),
known asthma bronchiale, pacemaker or implanted
cardioverter-debrillator (no cardiac magnetic reso-
nance imaging [CMR] possible), pregnancy or breast-
feeding, or inability to provide informed consent. If
patients fullled the inclusion/exclusion criteria,
verbal informed consent was obtained.
The trained ambulance paramedic completed the
administration/enrollment procedure. After informed
consent, a blinded study medication box was opened.
This box contained 2 vials of metoprolol 5 mg or
matching placebo and labeled with a number that
corresponded with the randomization list. Randomi-
zation took place without stratication and in blocks
of 4. The rst bolus of study medication was given in
the ambulance, the second bolus was given at the
catheterization laboratory pre-PCI but only if systolic
BP was >100 mm Hg and heart rate >60 beats/min.
Given that the COMMIT CC2 study (Clopidogrel and
Metoprolol in Myocardial Infarction Trial/Second
Chinese Cardiac Study) (12) showed that 15 mg of
metoprolol administered in a short interval was
associated with a slight increase in cardiogenic shock
(although this was restricted to Killip III patients), the
reference ethics committee suggested reducing
the dose to 10 mg separated into two 5-mg boluses:
the rst 5-mg bolus during ambulance transit and the
second 5-mg bolus on arrival at the catheterization
laboratory (i.e., immediately before initiating PCI).
The METOCARD-CNIC trial results (using 15 mg of
metoprolol) were not known at the time of the study
design. Patients participating in the trial were treated
during hospital admission and thereafter according to
current guidelines. During PCI, the use of thrombus
Roolvink et al. 2707
The EARLY-BAMI Trial
aspiration and glycoprotein IIb/IIIa inhibitors was left
to operator discretion. Stenting was performed with a
second-generation drug-eluting stent. After PCI, pa-
tients received detailed written study information
detailing the protocol and the CMR follow-up,
whereafter written informed consent was obtained.
Plans were for all patients to receive oral metoprolol
within 12 h post-PCI, according to current guidelines,
during hospitalization. At discharge, all patients
received oral metoprolol at a dose recommended by
tient clinic visits to obtain follow-up data.
The primary endpoint was myocardial infarct size
(percent infarcted myocardium, percentage of LV) as
measured by CMR at 30 days (10 days). The sec-
ondary efcacy endpoints were peak creatine kinase
(CK), peak CK-MB, troponin at 24 h, the CK and CK-
MB area under the curve during the rst 24 h, resid-
ual ST-segment deviation 1 h after PCI/coronary
angiogram, ventricular arrhythmias requiring de-
brillation during transportation and hospitalization,
and major adverse cardiac event (MACE) rate, dened
as cardiac death, nonfatal reinfarction, or target
vessel revascularization at 30 days. The secondary
safety endpoints included symptomatic bradycardia,
symptomatic hypotension, and cardiogenic shock.
The following subgroups were prespecied for anal-
ysis: anterior versus nonanterior infarctions, patients
presenting <6 h after symptom onset versus patients
presenting $6 h, and occluded (Thrombolysis
In Myocardial Infarction [TIMI] ow grade 0 and
1) infarct-related vessel at time of PCI versus open
(TIMI ow grade 2 and 3) infarct-related vessel.
STATISTICAL METHODS. Initially, the study was
designed and initiated with enzymatic infarct size as
the primary endpoint. However, after inclusion of 164
patients, the primary endpoint was changed into a
CMR-based endpoint (at July 3, 2013), which was
approved by the steering committee and the medical
ethics committee. The original primary outcome
measure was enzymatic infarct size assessed by car-
diac troponin T, and required 770 patients, based on
the assumption that pre-hospital administration of
2
5-mg metoprolol IV would give a relative 20%
reduction in infarct size (alpha 0.05; power: 80%;
mean troponin T: 3.34 ng/l; SD 3.30). This original
primary outcome became a secondary outcome after
the protocol change. This change in primary outcome
was mostly made to reduce the necessary sample size
and accommodate a time limitation on funding. Also,
infarct size could be studied more precisely with CMR.
The change occurred while the investigators were still
entirely blinded to trial results and without any
interim analysis performed. The sample size was than
2708 Roolvink et al. JACC VOL. 67, NO. 23, 2016

The EARLY-BAMI Trial JUNE 14, 2016:270515

F I G U R E 1 Patient Flow

Included
N=684

No MRI No MRI
44 Inclusion before amendment 1 patient randomization 47 Inclusion before amendment
1 CABG unknown 3 CABG
25 Claustrophobia 15 Claustrophobia
11 Comorbidity 14 Comorbidity
3 Distance Metoprolol placebo 5 Distance
1 ICD N=336 N=347 2 ICD
9 No infarction criteria (enzymes) 9 No infarction criteria (enzymes)
4 Previous MI 6 Previous MI
14 No written informed consent 15 No written informed consent
15 Out of window 20 Out of window
8 Died 4 Died
17 Refused 18 Refused
7 Unable to plan appointment Underwent cardiac MRI Underwent cardiac MRI
10 Unable to plan appointment
8 Other N=169 N=173
6 Other

MRI analyzed
MRI analyzed
N=169
N=162
Excluded (n = 4)
Excluded (n = 7)
1 Previous MI
3 Acquisition problem
2 Acquisition problem
4 No infarction criteria (enzymes)
1 No infarction criteria (enzymes)

Reasons as to why patients did not undergo cardiac magnetic resonance imaging (CMR) are shown for both groups. CABG coronary artery bypass graft surgery;
ICD implantable cardioverter-debrillator; MI myocardial infarction; MRI magnetic resonance imaging.

determined for the revised primary endpoint by a
power analysis with reasonable clinical and statistical
assumptions. With an expected infarct size of 28% (18)
in a population under standard treatment (no
beta-blocker pre-PCI), we considered a reduction in
infarct size from 28% to 23.5% clinically relevant.
Assuming an SD of the MI size measured by CMR equal
to 10% (19), the power analysis indicated a total sam-
ple size of 326 patients (163 subjects in each group) was
needed to achieve 80% power with signicance level
of 0.05 to detect a difference in infarct size.
Patients who died after completing the CMR study
were included in the primary outcome analysis. Pa-
tients who died before performance of the CMR were
not included in the primary outcome analysis; how-
ever, they were included in the secondary outcome
analysis because death within 1 year is a secondary
endpoint. Statistical analysis was performed with
SAS, version 9.3 (SAS Institute, Cary, North Carolina)
and SPSS Statistics for Windows, version 22.0 (IBM
Corp, Armonk, New York). Continuous data were
expressed as mean  SD, and categorical data as
percentage, unless otherwise denoted. The analysis
of variance and the chi-square tests were appropri-
ately used for continuous and categorical variables,
respectively. For quantitative variables, data were
expressed as mean  SD, and median with rst and
third quartiles. Non-normal data were compared by
nonparametric methods (Wilcoxon rank sum test) and
normal data by parametric methods. Categorical data
were expressed as percentages and compared by chi-
square test (or Fisher exact test when appropriate).
For all analyses, statistical signicance was assumed
when the 2-tailed p < 0.05.
CMR ANALYSIS. All CMR studies were performed
blinded to treatment allocation and according to a
centralized protocol. Dedicated sequences evaluating
cardiac function, myocardial edema, myocardial
perfusion, and myocardial necrosis/brosis were per-
formed. All CMR studies were stored and further
analyzed in a central core laboratory at the Centro
Nacional de Intervestigaciones Cardiovasculares
Carlos III in Madrid, Spain. CMR study analyses were
performed in a random manner by expert observers
blinded to treatment allocation. Quantication was
performed on a separate workstation using a dedicated
software package (QMass MR 7.6, Medis, Leiden, the
Netherlands). In all CMR studies, the following infor-
mation was determined: LV volumes, LV function,
and myocardial delayed enhancement. Myocardial
JACC VOL. 67, NO. 23, 2016 Roolvink et al. 2709
JUNE 14, 2016:270515 The EARLY-BAMI Trial

necrosis was dened by the extent of abnormal

T A B L E 1 Baseline Characteristics
delayed enhancement. All measurements were
expressed as percentage of the total LV myocardial Metoprolol Placebo Total
(n 336) (n 347) (N 684) p Value
volume; the absolute MI size was also quantied in
Age, yrs 62.39  12.42 62.46  12.58 62.42  12.49 0.882
grams. All results were given in absolute numbers and
Female 84/336 (25.0) 88/347 (25.4) 172/683 (25.2) 0.914
indexed by patients body surface. Differences be- Length, cm 174.3  10.22 175.2  9.59 174.8  9.91 0.327
tween the 2 groups were estimated by multivariable Weight, kg 82.82  16.39 84.71  16.04 83.78  16.23 0.102
linear regression adjusted for the participating hospi- BMI, kg/m2 27.14  4.45 27.40  4.11 27.27  4.28 0.246
tal center and the stratication variables. Diabetes 48/335 (14.3) 62/347 (17.9) 110/682 (16.1) 0.209
Previous hypertension 135/335 (40.3) 133/344 (38.7) 268/679 (39.5) 0.695
Beta-blocker use as home 54/298 (18.1) 60/308 (19.5) 114/606 (18.8) 0.669
RESULTS medication
Anterior location 154/312 (49.4) 166/318 (52.2) 320/630 (50.8) 0.677
Between February 2012 and November 2015, 684 First medical contact 0.595
patients were enrolled in 14 participating hospitals Referring hospital 19/335 (5.7) 16/347 (4.6) 35/682 (5.1)
and 4 ambulance services in the Netherlands and PCI center 8/335 (2.4) 12/347 (3.5) 20/682 (2.9)
Ambulance 308/335 (91.9) 319/347 (91.9) 627/682 (91.9)
Spain. In 1 case, the box with the study medication
Time from onset complaints
was lost during transportation, and this patient was
First medical contact, min 135.5  231.9 147.9  212.5 141.7  222.3 0.880
excluded from the analysis. The remaining 683
Early presenters (within 6 h) 288/307 (93.8) 277/310 (89.4) 565/617 (91.6) 0.046
patients were randomized to metoprolol (n 336) or Admission, min 195.5  262.5 201.6  262.1 198.6  262.1 0.755
placebo (n 347) before PPCI (Figure 1). Mean age was (n 307) (n 318) (n 625)
62  12 years, and 75% were male (Table 1 presents Hemodynamics at admission
baseline characteristics). Although time from onset of Systolic BP, mm Hg 136.4  22.91 138.7  26.43 137.6  24.75 0.384

complaints until rst medical contact was comparable Diastolic BP, mm Hg 82.25  14.67 82.83  16.16 82.54  15.43 0.702
Heart rate, beats/min 74.35  13.71 78.68  15.69 76.53  14.89 <0.001
between the 2 groups, early presenters (<6 h after
Discharge medication
complaint onset) were more often present in the
ACE inhibitors 215/333 (64.6) 209/341 (61.3) 424/674 (62.9) 0.379
metoprolol group. At inclusion (before randomiza-
Angiotensin II blockers 14/333 (4.2) 23/341 (6.7) 37/674 (5.5) 0.148
tion), mean BP on admission was comparable be- beta-blocker 260/333 (78.1) 249/341 (73.0) 249/341 (73.0) 0.127
tween the 2 groups, whereas mean heart rate was Angiographic ndings
not signicantly lower in the metoprolol group 1-vessel disease 175/330 (53.0) 201/339 (59.3) 376/669 (56.2)
(78.6 beats/min vs. 80.5 beats/min; p 0.09). A total 2-vessel disease 100/330 (30.3) 71/339 (20.9) 171/669 (25.6)

of 20 patients (2.9%) were enrolled at the emergency 3-vessel disease 39/330 (11.8) 46/339 (13.6) 85/669 (12.7)
Primary PCI 306/315 (97.1) 306/322 (95.0) 612/637 (96.1) 0.164
department of the PCI center, mainly because patient
Additional PCI during 16/334 (4.8) 15/345 (4.3) 31/679 (4.6) 0.782
transport to the PCI center was very short. In these admission
patients, the rst study bolus was given as soon as CABG during admission 12/335 (3.6) 24/345 (7.0) 36/680 (5.3) 0.049
possible at the emergency department with the sec-
Values are mean  SD or n/N (%).
ond bolus at catheterization laboratory arrival.
ACE angiotensin-converting enzyme; BMI body mass index; BP blood pressure; CABG coronary artery
Of the 336 patients allocated to pre-reperfusion bypass graft surgery; PCI percutaneous coronary intervention.

metoprolol, 81.1% also received the second bolus

compared with 86% in the placebo group (p 0.08).
BP before the second bolus (administered at the
catheterization laboratory) was not different between
groups, whereas heart rate was lower in the meto-
prolol group. Approximately 99% of patients in both
groups underwent coronary angiography. One-vessel
disease was observed in 53% of the metoprolol
group compared with 59% of the placebo group
(p 0.06). Similar rates of initial TIMI ow grade 0 or
1 were demonstrated in the metoprolol group versus
the placebo group (62% vs. 60%; p 0.61) and PPCI
performed in (93% vs. 92%; p 0.66); the latter was
successful in 97% in both groups (p 0.70). Oral
metoprolol was initiated within 24 h in 78% of
the metoprolol group and 73% of the placebo group
(p 0.13).
ENDPOINTS. CMR was performed in 342 of the 520
patients (66%) who were included after the change of
the primary endpoint (67% in each group; p 0.77)
(Online Appendix). In the 342 patients who had CMR
performed, no major differences in baseline charac-
teristics between the treatment groups were present.
In the CMR sample size calculation, we accounted for a
20% dropout rate, but the actual dropout rate was
higher (34%). So, to meet the needed CMR number,
patient recruitment was continued until 326 analyzed
CMRs were reached. The main reasons given for not
performing CMR included claustrophobia, planning
out of the time window of 1 month  10 days, and pa-
tient refusal due to transfer to a different hospital
(Figure 1).
2710 Roolvink et al. JACC VOL. 67, NO. 23, 2016

The EARLY-BAMI Trial JUNE 14, 2016:270515

administration of IV metoprolol did not improve LV

T A B L E 2 CMR Imaging*
ejection fraction (LVEF) on CMR (50.97  10.93% vs.
Metoprolol Placebo Total 51.65  10.83% in the placebo group).
Outcome (n 336) (n 347) (N 684) p Value
The peak CK, CK-MB, and troponin levels at 24 h
CMR performed 169/305 (55.4) 173/319 (54.2) 342/624 (54.8) 0.768
were available in 591 (86.4%) patients (Table 3). Peak
CMR analyzed 162/169 (95.9) 169/174 (97.1) 331/343 (96.5) 0.523
Delayed enhancement infarction, % of LV 0.616 CK was 2,102  2,029 U/l in the metoprolol group
Mean  SD 15.3  11.0 14.1  11.5 15.1  11.2 versus 2,072  2,018 U/l in the placebo group
Median (IQR) 13.4 (6.421.3) 13.3 (5.621.3) 13.4 (5.921.3) (p 0.88). Mean single troponin T measured at 24 h
Min-max 0.0044.1 0.0049.9 0.0049.9 of hospitalization was 3,711  3,587 ng/l in the meto-
(n 159) (n 167) (n 326) prolol group versus 3,166  3,998 ng/l in the placebo
LVEF, % 0.683
group (p 0.1). Results of enzymatic infarct size are
Mean  SD 51.0  10.9 51.7  10.8 51.2  10.9
summarized in Figure 2 (area under the curve).
Median (IQR) 53.0 (44.159.3) 53.3 (45.358.6) 53.5 (44.959.3)
Min-max 21.069.6 15.075.8 15.075.8
MACE rates at 30 days were 6.2% (n 19) in the
(n 162) (n 169) (n 331) metoprolol group versus 6.9% (n 22) in the placebo
LVEDV, ml 0.398 group (p 0.72) (Table 4).
Mean  SD 183.9  52.4 184.2  40.1 184.0  46.4 Pre-reperfusion administration of IV metoprolol
Median (IQR) 177.9 (149.1209.1) 181.8 (157.5212.0) 180.0 (153.7209.9) did not change the incidence of the prespecied sec-
Min-max 83.6469.7 85.4292.8 83.6469.7
ondary safety endpoints. There were 16 (4.8%) safety
(n 162) (n 169) (n 331)
events in the metoprolol group versus 11 (3.2%) in the
LVESV, ml 0.651
placebo group (p 0.271). Bradycardia was observed
Mean  SD 93.3  46.1 90.5  32.91 91.9  39.9
Median (IQR) 82.7 (64.0108.7) 86.3 (65.8106.1) 85.4 (64.8108.1) in more metoprolol patients, but the difference was
Min-max 26.0359.5 27.8187.0 26.0359.5 not signicant (4.2% vs. 2.6%; p 0.25) (Table 5).
(n 162) (n 169) (n 331) There was, however, a borderline signicant reduc-
LV mass from function, g 0.893 tion in the occurrence of malignant arrhythmias in the
Mean  SD 96.4  25.2 96.5  23.1 96.5  24.1 acute phase in the metoprolol group, 12 patients
Median (IQR) 96.4 (77.2110.7) 94.6 (80.6110.1) 94.8 (79.5110.6)
(3.6%) versus 24 patients (6.9%; p 0.050).
Min-max 48.4195.4 48.2187.8 48.2195.4
(n 162) (n 169) (n 331) PRESPECIFIED SUBGROUP ANALYSIS. Infarct size
LV mass from delayed enhancement, g 0.782 1 month post-STEMI did not differ in patients pre-
Mean  SD 104.6  29.0 103.1  25.7 103.9  27.5 senting with an anterior infarction between the
Median (IQR) 100.6 (85.1122.5) 101.3 (84.1119.0) 100.9 (84.3121.4) metoprolol (18.8  12.2%) and placebo groups (19.3 
Min-max 50.6244.0 46.7189.6 46.7244.0
12.7%; p 0.33) (Figure 3). Similarly, there were no
(n 159) (n 167) (n 326)
signicant differences seen between patients in the
Poor quality delayed enhancement images 0.914
metoprolol and placebo groups in regard to infarct
No 150/162 (92.6) 157/169 (92.9) 307/331 (92.7)
Yes 12/162 (7.4) 12/169 (7.1) 24/331 (7.3) size in patients with a nonanterior infarction, early or
Edema 0.398 late presentation after symptom onset, or infarct size
Absence 10/86 (11.6) 11/102 (10.8) 21/188 (11.2) in patients with an occluded vessel (TIMI ow grade
Small zone 17/86 (19.8) 16/102 (15.7) 33/188 (17.6) 0/1) at time of coronary angiography (Figure 3).
Extended zone 45/86 (52.3) 48/102 (47.1) 93/188 (49.5)
Black (hemorrhage) 14/86 (16.3) 27/102 (26.5) 41/188 (21.8) DISCUSSION
core
Localization 0.742
In this double-blind, randomized controlled trial of
No evidence of 12/162 (7.4) 8/170 (4.7) 20/332 (6.0)
infarction patients with STEMI undergoing PPCI, pre-reperfusion
Anterior or septal 71/162 (43.8) 75/170 (44.1) 146/332 (44.0) administration of up to 10 mg of IV metoprolol was
Inferior 47/162 (29.0) 57/170 (33.5) 104/332 (31.3) safe, but had no effect on infarct size or LVEF.
Lateral or inferolateral 24/162 (14.8) 24/170 (14.1) 48/332 (14.5)
Beta-blockers have multiple actions on the heart.
Typical of myocarditis 8/162 (4.9) 6/170 (3.5) 14/332 (4.2)
Blockade of b1 receptors results in slowing of heart
Values are n/N (%) unless otherwise indicated. *1 month  10 days after randomization. rate, reduction in myocardial contractility, and
CMR cardiac magnetic resonance; IQR interquartile range; LV left ventricle; LVEDV left ventricular lowering of systemic BP. In the context of acute MI,
end-diastolic volume; LVEF left ventricular ejection fraction; LVESV left ventricular end-systolic volume.
which represents a state of reduced oxygen supply to
the affected portion of the heart, these effects may be
The primary endpoint, mean infarct size (percent benecial, because they result in reduced myocardial
delayed enhancement of LV), was 15.29  10.97% in workload and oxygen demand. Furthermore, beta-
the metropolol group versus 14.91  11.52% in the blocker therapy decreases the incidence of life-
placebo group (p 0.616) (Table 2). Pre-reperfusion threatening arrhythmias, reinfarction, and recurrent
JACC VOL. 67, NO. 23, 2016 Roolvink et al. 2711
JUNE 14, 2016:270515 The EARLY-BAMI Trial

ischemia, preventing LV remodeling (11,1315,1821).

T A B L E 3 Secondary Endpoints
These agents have been shown to be benecial,
resulting in mortality reductions in patients with Metoprolol Placebo Total
Outcome (n 336) (n 347) (N 684) p Value
reduced LV function and when administered post-MI.
Maximal CK-MB, U/l 0.750
However, there is debate whether pre-reperfusion
Mean  SD 224  212 213  195 218  203
administration of IV beta-blockers may reduce Median (IQR) 148 (65327) 167 (57300) 155 (62317)
reperfusion injury compared with post-reperfusion Min-max 9998 10943 9998
administration (22). (n 207) (n 204) (n 411)
The MIAMI (Metoprolol in Acute Myocardial Maximal CK-MB, mg/l 0.181

Infarction) trial tested the effect of pre-reperfusion Mean  SD 276  243 232  254 254  248
Median (IQR) 188 (97450) 133 (49282) 152 (60440)
metoprolol (3
5 mg IV) versus placebo in STEMI
Min-max 3880 0969 0969
patients (N 5,778) treated by thrombolysis and
(n 52) (n 54) (n 106)
demonstrated no effect of metoprolol (11). In the Maximal CK (total), U/l 0.880
COMMIT CC2 trial with 45,825 patients, IV metoprolol Mean  SD 2,102  2,029 2,072  2,018 2,087  2,022
3
5 mg followed by oral administration up to 4 Median (IQR) 1,370 (5383,050) 1,411 (4662,980) 1,376 (5043,050)
weeks did not improve survival in STEMI patients Min-max 479,857 308,769 309,857

(12). However, this was mainly caused by a higher (n 298) (n 293) (n 591)
24-h hs-cTn T, ng/l 0.103
incidence of cardiogenic shock in patients treated by
Mean  SD 3,711  3,587 3,166  3,998 3,451  3,790
early beta-blocker therapy, possibly due to inclusion
Median (IQR) 2,530 1,994 2,224
of patients with heart failure. In the current era of (1,2005,450) (962.53,800) (1,0594,800)
PPCI, the METOCARD-CNIC trial showed reduced Min-max 22.4019,480 1.2531,700 1.2531,700
infarct size and increased LVEF in STEMI patients (n 114) (n 104) (n 218)

without signs of heart failure treated with early IV 24-h cTn I, mg/l 0.948
Mean  SD 42.37  39.85 53.04  56.49 48.26  49.19
metoprolol (14,23). However, this study had a rela-
Median (IQR) 32.60 (13.3058.99) 32.83 (9.2874.71) 32.60 (9.7058.99)
tively small sample size (N 270), was not blinded
Min-max 0.65136.5 0.07177.1 0.07177.1
or placebo controlled, and included a select (n 13) (n 16) (n 29)
patient group (anterior STEMI presenting <6 h from
symptom onset). CK creatine kinase; cTn cardiac troponin; hs high sensitivity; other abbreviations as in Table 2.

Our study included all patients with STEMI,

with a double-blinded, placebo-controlled design.
Our results do not conrm the effect observed in
the METOCARD-CNIC trial. One possible explana-
tion could be that the METOCARD-CNIC trial included
only anterior infarctions, and the average infarct
size (infarcted myocardium, percent LV) in the
METOCARD-CNIC trial was 21.2% in patients treated
with IV metoprolol compared with 15.3% in our
study. These differences might support this theory:
the smaller the infarction, the less likely an addi-
tional treatment effect could be demonstrated.
However, the subgroup with anterior infarction
also had no benet of early b-blockade. In the
METOCARD-CNIC trial, the dose of metoprolol was
higher, up to 3 times 5 mg (15-mg target dose),
whereas in our trial only 2 times 5 mg (10-mg target
dose) was given.
Another explanation might be that 18.8% of pa-
tients in our trial were on long-term beta-blocker
treatment before admission; this, however, was an
exclusion criteria in the METOCARD-CNIC trial. Also,
CMR timing can be an inuence. In the METOCARD
trial, CMRs were performed at 5 to 7 days and at
6 months. Data from the rst CMR showed an
improvement in LVEF in the metoprolol group and a
signicant reduction in infarct size. Follow-up CMR
data at 6 months showed an even more signicant
difference in LVEF favoring metoprolol, but with no
remaining signicant difference in infarct size. In the
metoprolol groups, the CMR infarct size as percent of
LV was 15.7  9.6% in the METOCARD-CNIC trial at
6 months versus 15.3  11.0% in the EARLY-BAMI trial
at 1 month. Another potential reason for the different
effect observed between these trials pertains to when
metoprolol was administered. A recent METOCARD-
CNIC subanalysis showed that the timing of meto-
prolol administration was a critical factor accounting
for its infarct-limiting effect: only patients receiving
IV metoprolol long before reperfusion had a reduction
in infarct size (24). Similar to control patients,
METOCARD-CNIC trial patients receiving IV meto-
prolol close to PPCI had signicantly larger in-
farctions than those receiving IV metoprolol long
before reperfusion. In the present trial, the second
5-mg bolus (to complete the 10-mg target dose) was
administered per protocol immediately before cath-
eterization (median time from bolus and reperfusion:
14 min). The rst 5 mg of IV metoprolol might be
insufcient to attain cardioprotection; in fact, BP
2712 Roolvink et al. JACC VOL. 67, NO. 23, 2016

The EARLY-BAMI Trial JUNE 14, 2016:270515

F I G U R E 2 Enzymatic Infarction Size

A 3000 B 2500
p value = 0.88 P value = 0.6416

2500
2000 Placebo
Metoprolol
Infarct Size (Peak CK, U/L)

Infarct Size (CK Release)

2000
1500

1500

1000
1000

500
500

0 0
Metoprolol Placebo 0 3 6 9 12 18 24 36 48 72
Time After Admission (Hrs)

There was no signicant difference in infarct size estimated by (A) peak creatine kinase (CK) or (B) area under the CK curve.

after the rst bolus of medication (i.e., before the reduction of acute malignant arrhythmias seen in
second bolus) was no different between metoprolol this trial encourage the performance of additional
and placebo arms in this trial, supporting the low- larger trials.
dosing hypothesis. In most randomized trials, in which infarct size or
Based on these data, and the conicting results LVEF was measured with CMR, CMR was performed
with the METOCARD-CNIC trial, additional random- at 1 month (25,26). This was the main reason why we
ized trials are needed to clarify whether early chose 1 month in our trial. All participating PCI cen-
beta-blocker treatment has any effect in these pa- ters and ambulance services had a longstanding
tients. We advocate that future studies test the car- experience in pre-hospital triage and treatment of
dioprotective effects of IV metoprolol in STEMI STEMI patients. Regional differences in systems of
patients using a target dose of 15 mg and administer care in which prehospital drug administration in the
medication immediately after STEMI diagnosis to ambulance differ (Europe vs. United States) can lead
allow a maximum on-board metoprolol time before to different application of these study results to daily
reperfusion. Given the reduced observed infarct size practice.
with a trial of our sample size, the reduction in STUDY LIMITATIONS. During the course of the trial,
infarct size by metoprolol should be at least 3.5% the primary endpoint was changed from enzymatic
to demonstrate a signicant difference between infarct size to infarct size measured by CMR, to
the groups. The safety prole, low cost, and the reduce the necessary sample size and because infarct

T A B L E 4 30-day Adverse Cardiac Events T A B L E 5 Safety Endpoints

Metoprolol Placebo Total Metoprolol Placebo Total

Outcome (n 336) (n 347) (N 684) p Value Outcome (n 336) (n 347) (N 684) p Value
MACE 19/307 (6.2) 22/319 (6.9) 41/626 (6.5) 0.721 Severe bradycardia 5/334 (1.5) 2/345 (0.6) 7/679 (1.0) 0.279
Cardiac mortality 7/307 (2.3) 7/319 (2.2) 14/626 (2.2) 0.942 Severe hypotension 9/310 (2.9) 18/326 (5.5) 27/636 (4.2) 0.102
MI 3/307 (1.0) 2/319 (0.6) 5/626 (0.8) 0.681 Cardiogenic shock 2/334 (0.6) 1/345 (0.3) 3/679 (0.4) 0.618
TVR 12/307 (3.9) 15/319 (4.7) 27/626 (4.3) 0.625 Ventricular 12/335 (3.6) 24/346 (6.9) 36/681 (5.3) 0.050
arrhythmias in
Values are n/N (%). acute phase
MACE major adverse cardiac events; MI myocardial infarction; TVR target vessel
revascularization. Values are n/N (%).
JACC VOL. 67, NO. 23, 2016 Roolvink et al. 2713
JUNE 14, 2016:270515 The EARLY-BAMI Trial

F I G U R E 3 Effect of Metoprolol on Delayed Enhancement Infarction

Metoprolol Placebo Metoprolol Placebo Difference in Delayed Enhancement Infarct (95% CI) P for interaction

All patients 159 167 15.29 10.97 14.91 11.52 -0.38 (-2.84 - 2.07)

Location of MI

Anterior infarction 78 88 18.81 12.20 19.30 12.65 0.49 (-3.33 - 4.31) 0.33

Non anterior infarction 75 71 12.24 8.01 10.41 7.78 -1.83 (-4.42 - 0.75)

Presentation

Early 135 129 15.11 10.69 14.56 11.09 -0.55 (-3.19 - 2.09) 0.72

Late 9 12 20.87 16.29 18.45 14.76 -2.43 (-16.7 - 11.81)

Initial TIMI flow

Initial 0,1 98 97 17.78 10.81 18.12 11.81 0.35 (-2.85 - 3.54) 0.74

Initial 2,3 43 49 9.59 9.01 9.03 8.55 -0.56 (-4.20 - 3.08)

-10 -5 0 5 10
Favors Placebo Favors Metoprolol

Compared with placebo, metoprolol exhibited no signicant effects on delayed enhancement infarction (percent of left ventricle) in pre-specied subgroups.
CI condence interval; MI myocardial infarction; TIMI Thrombolysis In Myocardial Infarction; other abbreviations as in Figure 1.

size could be studied more precisely with CMR. The
results from enzymatic infarct size analysis, howev-
er, were completely in line with the results from the
primary CMR endpoint.
The trial was powered with a reduction of infarct
size from 28% to 23.5%. The smaller-than-estimated
infarct size in this trial (15.1%) could affect the
neutral effect seen (the smaller the infarct size,
the less probability a difference could be found).
However, in the larger anterior infarctions, there
was no observed effect. Also, patients who died
before CMR was performed, meaning they probably
had larger infarctions, may have caused a selection
of patients with smaller infarctions undergoing
CMR.
Although we dened several subanalyses, these
analyses should be interpreted with caution, since the
included number of patients in several subgroups
were small. Also, we could not blind physicians and
nurses to heart rate and BP; however, CMR analyses

CENTRAL ILLUSTRATION Balancing the Results of the EARLY-BAMI Trial

Early Beta-blocker Administration before primary PCI in patients with ST-elevation Myocardial Infarction (EARLY-BAMI) trial

336 patients: 10 mg metoprolol Significant improvement with metoprolol?

Baseline characteristics

Infarct size 15.3% vs. 14.9% placebo

683 Peak and area under creatine kinase (CK) curve
patients

Left ventricular ejection fraction 51% vs. 51.6%

Incidence of adverse events

346 patients: placebo Incidence of malignant arrhythmias 3.6% vs. 6.9%

Roolvink, V. et al. J Am Coll Cardiol. 2016;67(23):270515.

2714 Roolvink et al. JACC VOL. 67, NO. 23, 2016

The EARLY-BAMI Trial JUNE 14, 2016:270515

(primary endpoint) were blinded for both heart rate

and study medication. REPRINT REQUESTS AND CORRESPONDENCE: Dr.
Arnoud W.J. van t Hof, Department of Cardiology,
CONCLUSIONS Isala Hospital, Dr. van Heesweg 2, 8025 AB, Zwolle,
the Netherlands. E-mail: v.r.c.derks@isala.nl.
Early pre-reperfusion administration of IV metopro-
lol, at a dose of 10 mg (2
5 mg), had no benecial PERSPECTIVES
effect on infarct size in patients with STEMI treated
by PPCI (Central Illustration). COMPETENCY IN PATIENT CARE: In patients with
acute STEMI symptoms of <12 h duration given
ACKNOWLEDGMENTS The authors are grateful for
intravenous metoprolol before primary angioplasty,
the commitment of all the ambulance personnel in
infarct size measured by CMR was not signicantly
the Netherlands (RAV IJsselland, Witte kruis, UMCG
different than in those given a placebo.
ambulancezorg) and in Spain at SUMMA 112. The
recruitment of patients in Spain was performed
TRANSLATIONAL OUTLOOK: The conicting
within the Codigo Infarto Network. Dr. Ibez, Dr.
results of available studies require additional large
Pizarro, Ms. Escalera, and Dr. Fernndez-Avils are
randomized trials to clarify whether intravenous beta-
members of the Red de Investigacin Cardiovascular-
blocker treatment before angioplasty is benecial for
RIC (RETIC# RD12/0042/0054, RETIC#RD12/0042). All
patients with STEMI.
coinvestigators in the different hospitals have been
capital for the rigorous conduct of this trial.

REFERENCES

1. Fokkema ML, James SK, Albertsson P, et al.
Population trends in percutaneous coronary
intervention: 20-year results from the SCAAR
(Swedish Coronary Angiography and Angioplasty
Registry). J Am Coll Cardiol 2013;61:122230.
2. Van de Werf F, Bax J, Betriu A, et al. Manage-
ment of acute myocardial infarction in patients
presenting with persistent ST-segment elevation:
the Task Force on the Management of ST-Segment
Elevation Acute Myocardial Infarction of the Eu-
ropean Society of Cardiology. Eur Heart J 2008;
29:290945.
3. Dargie HJ. Effect of carvedilol on outcome after
myocardial infarction in patients with left-
ventricular dysfunction: the CAPRICORN rando-
mised trial. Lancet 2001;357:138590.
4. Feuerstein GZ, Yue TL, Cheng HY, Ruffolo RR Jr.
Myocardial protection by the novel vasodilating
beta-blocker, carvedilol: potential relevance of anti-
oxidant activity. J Hypertens Suppl 1993;11:S418.
5. Hammerman H, Kloner RA, Briggs LL,
Braunwald E. Enhancement of salvage of reper-
fused myocardium by early beta-adrenergic
blockade (timolol). J Am Coll Cardiol 1984;3:
143843.
6. Rasmussen MM, Reimer KA, Kloner RA,
Jennings RB. Infarct size reduction by propranolol
before and after coronary ligation in dogs. Circu-
lation 1977;56:7948.
7. Sederholm, et al., for The International Collab-
orative Study Group. Reduction of infarct size with
the early use of timolol in acute myocardial
infarction. N Engl J Med 1984;310:915.
8. Peter T, Heng MK, Singh BN, et al. Failure of
high doses of propranolol to reduce experimental
myocardial ischemic damage. Circulation 1978;57:
53440.
9. Rude RE, Buja LM, Willerson JT. Propranolol in
acute myocardial infarction: the MILIS experience.
Am J Cardiol 1986;57:38F42F.
10. Yusuf S, Sleight P, Rossi P, et al. Reduction in
infarct size, arrhythmias and chest pain by early
intravenous beta blockade in suspected acute
myocardial infarction. Circulation 1983;67:I3241.
11. The MIAMI Trial Research Group. Metoprolol in
acute myocardial infarction (MIAMI). A randomised
placebo-controlled international trial. Eur Heart J
1985;6:199226.
12. Herlitz J, Waagstein F, Lindqvist J,
Swedberg K, Hjalmarson A. Effect of metoprolol
on the prognosis for patients with suspected acute
myocardial infarction and indirect signs of
congestive heart failure (a subgroup analysis of
the Goteborg Metoprolol Trial). Am J Cardiol
1997;80:40J4J.
13. Hanada K, Higuma T, Nishizaki F, et al. Ran-
domized study on the efcacy and safety of
landiolol, an ultra-short-acting beta1-adrenergic
blocker, in patients with acute myocardial infarc-
tion undergoing primary percutaneous coronary
intervention. Circ J 2012;76:43945.
14. Ibanez B, Macaya C, Sanchez-Brunete V, et al.
Effect of Early Metoprolol on Infarct Size in ST-
Segment Elevation Myocardial Infarction Patients
Undergoing Primary PCI: the METOCARD-CNIC
trial. Circulation 2013;128:1495503.
15. Ibanez B, Fuster V, Macaya C, et al.
Study design for the effect of METOprolol in
CARDioproteCtioN during an acute myocardial
InfarCtion (METOCARD-CNIC): a randomized,
controlled parallel-group, observer-blinded clin-
ical trial of early pre-reperfusion metoprolol
administration in ST-segment elevation myocar-
dial infarction. Am Heart J 2012;164:47380.e5.
16. Roolvink V, Vant Hof AW, van der Horst IC.
The sooner, the better: early beta-blocker
administration in patients with ST-elevation
myocardial infarction. Crit Care Med 2013;41:
15668.
17. Roolvink V, Rasoul S, Ottervanger JP, et al.
Rationale and design of a double-blind, multi-
center, randomized, placebo-controlled clinical
trial of early administration of intravenous beta-
blockers in patients with ST-elevation myocardial
infarction before primary percutaneous coronary
intervention: EARLY beta-blocker administration
before primary PCI in patients with ST-elevation
myocardial infarction trial. Am Heart J 2014;168:
6616.
18. Wu E, Ortiz JT, Tejedor P, et al. Infarct size by
contrast enhanced cardiac magnetic resonance is a
stronger predictor of outcomes than left ventric-
ular ejection fraction or end-systolic volume index:
prospective cohort study. Heart 2008;94:7306.
19. Atar D, Petzelbauer P, Schwitter J, et al. Effect
of intravenous FX06 as an adjunct to primary
percutaneous coronary intervention for acute ST-
segment elevation myocardial infarction results
of the F.I.R.E. (Efcacy of FX06 in the Prevention
of Myocardial Reperfusion Injury) trial. J Am Coll
Cardiol 2009;53:7209.
20. Hirschl MM, Wollmann CG, Erhart F, et al.
Benet of immediate beta-blocker therapy on
mortality in patients with ST-segment elevation
myocardial infarction. Crit Care Med 2013;41:
1396404.
21. First International Study of Infarct Survival
Collaborative Group. Randomised trial of intrave-
nous atenolol among 16,027 cases of suspected
acute myocardial infarction: ISIS-1. Lancet 1986;2:
5766.
JACC VOL. 67, NO. 23, 2016
JUNE 14, 2016:270515
22. Ibanez B, Heusch G, Ovize M, Van de Werf F.
Evolving therapies for myocardial ischemia/
reperfusion injury. J Am Coll Cardiol 2015;65:
145471.
23. Pizarro G, Fernandez-Friera L, Fuster V, et al.
Long term benet of early pre-reperfusion
metoprolol administration in patients with acute
myocardial infarction: results from the
METOCARD-CNIC trial. J Am Coll Cardiol 2014;63:
235662.
24. Garca-Ruiz JM, Fernndez-Jimnez R, Garca-
Alvarez A, et al. Impact of the timing of meto-
prolol administration during ongoing STEMI
Roolvink et al. 2715
The EARLY-BAMI Trial
on infarct size and ventricular function: a infarction: rationale and study design of a pro-
METOCARD-CNIC trial substudy with an experi- spective, randomized, clinical trial (HEBE III). Am
mental validation. J Am Coll Cardiol 2016;67: Heart J 2008;155:81722.
2093104.
25. Eppinga RN, Hartman MH, van Veldhuisen DJ,
et al. Effect of metformin treatment on lipoprotein
KEY WORDS cardiac magnetic resonance,
subfractions in non-diabetic patients with acute
ejection fraction, infarct size, metoprolol
myocardial infarction: a Glycometabolic Interven-
tion as Adjunct to Primary Coronary Intervention in
ST Elevation Myocardial Infarction (GIPS-III) Trial.
PloS One 2016;11:e0145719.
A PP END IX For a supplemental gure and
26. Belonje AM, Voors AA, van Gilst WH, et al. table, please see the online version of this
Effects of erythropoietin after an acute myocardial article.