Original research 1
Complete revascularization for patients with ST-segment
elevation myocardial infarction and multivessel coronary
artery disease: a meta-analysis of randomized trials
Gani Bajraktaria,b,c, Haki Jasharia,b, Pranvera Ibrahimia,b, Fernando Alfonsod,
Fisnik Jasharia,c, Gjin Ndrepepae, Shpend Elezic and Michael Y. Heneina,f
Introduction Despite the recent findings in randomized
clinical trials (RCTs) with limited sample sizes and the
updates in clinical guidelines, the current available data for
the complete revascularization (CR) in hemodynamically
stable patients with ST-segment elevation myocardial
infarction (STEMI) at the time of primary percutaneous
coronary intervention (PCI) are still contradictory.
Aim The aim of this meta-analysis of the existing RCTs was
to assess the efficacy of the CR versus revascularization of
infarct-related artery (IRA) only during primary PCI in
patients with STEMI and multivessel disease (MVD).
Patients and methods We searched PubMed, MEDLINE,
Embase, Scopus, Google Scholar, Cochrane Central
Register of Controlled Trials (CENTRAL), and ClinicalTrials.
gov databases aiming to find RCTs for patients with STEMI
and MVD which compared CR with IRA-only. Random effect
risk ratios (RRs) were calculated for efficacy and safety
outcomes.
Results Ten RCTs with 3291 patients were included. The
median follow-up duration was 17.5 months. Major adverse
cardiac events (RR = 0.57; 0.43–0.76; P < 0.0001), cardiac
mortality (RR = 0.52; 0.31–0.87; P = 0.014), and repeat
revascularization (RR = 0.50; 0.30–0.84; P = 0.009) were
lower in CR compared with IRA-only strategies. However,
there was no significant difference in the risk of all-cause
mortality, recurrent nonfatal myocardial infarction, stroke,
major bleeding events, and contrast-induced nephropathy.
Introduction
Approximately half of patients with ST-segment eleva-
tion myocardial infarction (STEMI) undergoing primary
percutaneous coronary interventions (PCI) have multi-
vessel disease (MVD) [1], and the outcome of these
patients after primary PCI is worse compared with those
with one-vessel coronary artery disease [2,3]. The current
standard treatment for hemodynamically stable patients
with STEMI and MVD is PCI with stent implantation of
the infarct-related coronary artery (IRA) aiming to relieve
the obstruction of the culprit vessel and establish
Supplemental Digital Content is available for this article. Direct URL citations
appear in the printed text and are provided in the HTML and PDF versions of this
article on the journal’s website, www.coronary-artery.com.
0954-6928 Copyright © 2018 Wolters Kluwer Health, Inc. All rights reserved.
Copyright r 2018 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Conclusion For patients with STEMI and MVD undergoing
primary PCI, the current evidence suggests that the risk of
major adverse cardiac events, repeat revascularization,
and cardiac death is reduced by CR. However, the risk for
all-cause mortality and PCI-related complications is not
different from the isolated culprit lesion-only treatment.
Although these findings support the cardiac mortality and
safety benefit of CR in stable STEMI, further large trials are
required to provide better guidance for optimum
management of such patients. Coron Artery Dis 00:000–000
Copyright © 2018 Wolters Kluwer Health, Inc. All rights
reserved.
Coronary Artery Disease 2018, 00:000–000
Keywords: complete revascularization, coronary artery disease,
infarct-related artery-only revascularization, multivessel disease,
ST-segment elevation myocardial infarction
a
Department of Public Health and Clinical Medicine, Umeå University, Umeå,
Sweden, bClinic of Cardiology, University Clinical Centre of Kosova, cDepartment
of Internal Medicine, Medical Faculty, University of Prishtina, Prishtina, Republic of
Kosovo, dCardiac Department, La Princesa University Hospital, Institute of Health
Research, IIS-IP, University Autonoma of Madrid, Madrid, Spain, eDepartment of
Adult Cardiology, German Heart Centre Munich, Technical University of Munich,
Germany and fSt George University, London, UK
Correspondence to Gani Bajraktari, MD, FESC, FACC, Department of Public
Health and Clinical Medicine, Umeå University, Umeå, Sweden
Tel: + 46 907 850 000; fax: + 46 90 13 76 33; e-mail: gani.bajraktari@umu.se
Received 8 October 2017 Revised 26 November 2017
Accepted 27 November 2017
thrombolysis in myocardial infarction blood flow grade of
3 in the IRA to achieve myocardial reperfusion [4,5].
However, the early randomized clinical trials (RCTs)
[6–8] influenced the American College of Cardiology/
American Heart Association and European Society of
Cardiology to update their guidelines recommendation
class from III to IIb for the PCI of the non-IRA if the
patient is hemodynamically stable [9,10].
The recently published COMPARE-ACUTE trial
revealed that fractional flow reserve (FFR)-guided
revascularization of non-IRA arteries in patients with
STEMI and MVD decreased the composite cardiovas-
cular adverse effects after primary PCI compared with
the culprit artery-only treatment, offering new evidence
DOI: 10.1097/MCA.0000000000000602
 2 Coronary Artery Disease 2018, Vol 00 No 00

for revascularization of all significant stenosis in these Eligibility criteria

patients [11]. However, despite these findings, these Selected studies had to fulfill the following criteria: (i)
trials are limited by limited sample sizes. Moreover, the studies with hemodynamically stable patients with
current evidence from non-RCTs for complete revascu- STEMI, (ii) randomized design comparing CR versus
larization (CR) in hemodynamically stable patients with IRA-only revascularization, and (iii) data for outcome at
STEMI and MVD at the time of primary PCI remains follow-up. None of the included studies in this meta-
contradictory. Therefore, we conducted this meta- analysis included hemodynamically unstable patients
analysis of the randomized clinical studies to assess the complicated with shock.
efficacy of the CR compared with the revascularization of
IRA-only during primary PCI in patients with STEMI Data extraction
and MVD. Included RCTs were double checked by two authors
(H.J. and P.I.) to ensure appropriate data inclusion.
Patients’ demographics, sample size, clinical data of
interest as well as number of events with respect to
Patients and methods clinical outcomes were extracted. One study had two
Data sources randomized intervention arms regarding CR: CR per-
PubMed, MEDLINE, Embase, Scopus, Google Scholar, formed simultaneously at the time of culprit artery
Cochrane Central Register of Controlled Trials revascularization and CR performed as a staged revas-
(CENTRAL), and ClinicalTrials.gov databases were cularization. For this study, these two arms were com-
bined into one CR arm [13].
searched using combinations of the following key terms:
‘percutaneous coronary intervention’, ‘myocardial infarc-
Outcomes and definitions
tion’, ‘multivessel’, ‘multi-vessel’, ‘culprit’, ‘target vessel
The primary outcomes tested were major adverse cardiac
revascularization’, ‘infarct related artery revasculariza-
events (MACE), all-cause mortality, cardiac mortality,
tion’, ‘non-culprit’, and ‘complete revascularization’. All
nonfatal myocardial infarction (MI), revascularization,
randomized controlled trials published until 1 May 2017 stroke, contrast-induced nephropathy, and major bleed-
were identified. The reference lists of the retrieved ing (Table 1).
articles were manually searched for additional potential
articles. No language restriction was applied. The search CR was defined as revascularization of nonculprit lesions
algorithm is shown in Fig. 1. in patients with STEMI, either during the same proce-
dure or staged during index hospitalization or after dis-
The meta-analysis was performed according to the charge. Culprit-only revascularization was defined as
Preferred Reporting Items for Systematic Reviews and revascularization of IRA-only at the time of PCI.
Meta-Analyses statement [12].
Statistical analysis
All statistical analyses were performed using Comprehensive
Fig. 1 Meta-Analysis (version 3; Biostat Inc., Englewood, New
Potentially relevant papers form search
Jersey, USA). Random effects models presented as risk
N = 561 ratios (RRs) with 95% confidence interval were used.
Statistical heterogeneity for each outcome was assessed by
Limited to "clinical trials"
N = 87 Cochrane’s Q statistics. I2 statistic values of less than 25%,
Unrelated to study objective 25–50%, and more than 50% were considered as low,
N = 73 moderate, and high heterogeneity, respectively. Egger’s test
Potentially relevant papers for was used to assess the risk for publication bias. The CR
inclusion in Meta analysis
N =14 group was considered as the experimental group; hence, a
Registries N = 1
RR less than 1.0 favors that group. The influence of follow-
up period on the outcomes was assessed using meta-
Non Randomized N = 2
Reference list search regression. P value less than 0.05 was considered statisti-
N=2
Compared only index vs. cally significant.
staged complete
revascularization
N=2
Results
NSTEMI N = 1 The initial search identified 561 studies (after exclusion
Clinical trials included in Meta analysis of duplicates and triplicates). However, through the
N =10
selection process, we ended up with 10 RCTs [6–8,10,
PRISMA study selection flowchart. NSTEMI, non-ST-segment elevation 13–18] that fulfilled the aforementioned inclusion cri-
myocardial infarction. teria. Two studies used the same population of patients,
hence the study with the longest follow-up was included

Copyright r 2018 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Copyright r 2018 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Table 1 Definition of major adverse cardiac events, reinfarction, and multivessel disease
References MACE definition
Smits et al. [11] All-cause mortality, nonfatal MI, any
(COMPARE- revascularization, and
ACUTE) cerebrovascular events
Henriques et al. [14] Cardiac death, myocardial infarction,
(EXPLORE) and CABG.
Hamza and Elgendy [15] Composite of all-cause mortality,
recurrent MI, and ischemia-driven
revascularization
Hlinomaz [16] Composite of all-cause mortality,
(PRAGUE 13) recurrent MI, and stroke
Engstrøm et al. [8] All-cause mortality, recurrent MI, and
(DANAMI- ischemia-driven revascularization
3–PRIMULTI)
Gershlick et al. [7] All-cause mortality, recurrent MI,
(CvLPRIT) heart failure hospital admission,
and repeat revascularization
Wald et al. [6] (PRAMI) Cardiac death, recurrent MI, and
refractory angina
Ghani et al. [17] All-cause mortality, recurrent MI, and
urgent revascularization
Politi et al. [13] All-cause mortality, recurrent MI,
hospitalization for acute coronary
syndrome, and revascularization
Di Mario et al. [18] All-cause mortality, recurrent MI, and
(HELP AMI) repeat revascularization
CABG, coronary artery bypass grafting; CK, creatinine kinase; CR, complete revascularization; IRA, infarct-related artery; LBBB, left bundle branch block; MACE, major adverse cardiac events; MI, myocardial infarction;
PCI, percutaneous coronary intervention; QCA, quantitative coronary angiography; ULN, upper limit normal.
Reinfarction definition
NA
According to the Third Universal Definition of
Myocardial Infarction Criteria
NA
NA
Typical chest pain was accompanied by a
substantial rise in troponins, development
of new Q-waves on the ECG, or both
Hospital admission, or be diagnosed in
hospital, with one or more of the following:
Type 1: recurrent angina symptoms or new
ECG changes occurring before PCI or
<48 h from PCI that is compatible with re-
MI associated with an elevation of CK-MB,
troponin, or total CK beyond ULN and 20%
or more above the previous value.
Type 4a: CK-MB or total CK > 3 times the
ULN within 48 h following PCI with either
an appropriate clinical presentation or new
ischemic ECG changes (ST-segment
depression or ST-segment elevation or
development of new pathological Q-waves/
LBBB).
Type 4b: MI associated with stent thrombosis
as documented by angiography or at
autopsy and fulfilling the criteria of
spontaneous MI (type 1).
Symptoms of cardiac ischemia and a troponin
level > 99th centile. Within 14 days after
randomization: as above + new ECG
evidence of ST-segment elevation or LBBB
and angiographic evidence of coronary
artery occlusion.
New Q-waves on the ECG or a new CK and
CK-MB rise above the ULN. This included
periprocedural infarctions in the invasive
treatment arm.
NA
NA
Contrast-induced
Multivessel disease definition Major bleeding event nephropathy
IRA plus non-IRA or their major side NA NA
branches of at least 2.0 mm
diameter show ≥ 50% stenosis
by QCA or visual assessment
IRA plus at least 70% stenosis in Bleeding according to GUSTO criteria NA
non-IRA
IRA plus at least 80% stenosis in NA NA
non-IRA
NA NA NA
IRA plus > 50% stenosis in one or Bleeding requiring transfusion or surgery Contrast-induced
more non-IRA nephropathy (>50%
rise in plasma
creatinine)
IRA plus at least one non-infarct Major bleeding defined as the cumulative NA
artery with at least one lesion occurrence of intracranial or intraocular
> 70% single view/50% in two bleeding, hemorrhage at the vascular
views access site requiring intervention, a
reduction in hemoglobin levels of at least
5 g/dl, reoperation for bleeding or
transfusion of a blood product (≥2 U),
bleeding causing substantial
hypotension requiring the use of
inotropic agents. All other bleeding
events were considered as minor (i.e.
epistaxis, blood traces in the stool etc.).
CR for patients with STEMI Bajraktari et al. 3
IRA plus one or more non- NA NA
IRA > 50% stenosis
One or more stenoses of ≥ 50% (in NA NA
at least one view visually or by
QCA) in at least two major
epicardial coronary arteries
> 70% stenosis of at least two NA NA
epicardial coronary arteries or
their major branches
IRA plus at least one to three NA NA
lesions in major nonculprit related
artery
Copyright r 2018 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.

4 Coronary Artery Disease 2018, Vol 00 No 00

Table 2 General characteristics of the studies
Male EF at EF at Timing of
Number of (CR/ Age (CR/ Diabetes Hypertension Smoking Previous MI admission discharge staged
patients (CR/ IRA- IRA-only) (CR/IRA- (CR/IRA-only) (CR/IRA- (CR/IRA- Hypercholesterolemia (CR/ (CR/IRA-only) (CR/IRA-only) procedure of Follow-up
References IRA-only) (n) only) (%) (years) only) (%) (%) only) (%) only) (%) IRA-only) (%) (%) (%) CR (months)

Smits et al. [11] 295/590 79/76 62/61 15/16 46/48 41/49 7.5/8.1 32/30 – – Within 3 days 12
(COMPARE-
ACUTE)
Henriques et al. [14] 148/154 89/82 60/60 15/16 40/45 52/49 13/16 35/34 – 44/45 Within 7 days 4
(EXPLORE)
Hamza and Elgendy 50/50 82/86 56/52 100/100 26/36 72/78 10/6 48/42 46/47 – Within 3 days 6
[15]
Hlinomaz [16] 106/108 NA NA – – – – – – 49 3–40 days 38
(PRAGUE 13)
Engstrøm et al. [8] 314/313 80/81 64/63 9/13 41/47 51/48 5/9 – – 50/50 Within 2 days 27
(DANAMI-
3–PRIMULTI)
Gershlick et al. [7] 150/146 85/77 64/65 13/14 37/36 34/37 4.8/3.6 28/24 46/45 – Within 3 days 12
(CvLPRIT)
Wald et al. [6] 234/231 76/81 62/62 15/21 40/40 50/45 8/7 – – – At the same 23
(PRAMI) procedure
Ghani et al. [17] 79/40 80/81 62/61 6.3/5.0 26/43 44/48 6.3/4.9 15/30 – – Within 36
3 weeks
Politi et al. [13] 130/84 77/78 65/67 23/24 57/60 67/81 – – 45/45 47/45 NA 30
Di Mario et al. [18] 52/17 88/85 64/65 12/41 37/59 – – 41/53 48/49 – NA 12
(HELP AMI)

CR, complete revascularization; EF, ejection fraction; IRA, infarct-related artery; MI, myocardial infarction.
 CR for patients with STEMI Bajraktari et al. 5

[17,19]. A total number of 3291 patients were reported:
1558 in the CR group and 1733 in the IRA-only revas-
cularization group. The duration of the follow-up ranged
from 4 to 38 months (median: 17.5 months). General
characteristics of the studies are presented in Table 2.
Major adverse cardiac events
MACE was reported in all studies but its definition dif-
fered among studies (Supplementary Table 1,
Supplemental digital content 1, http://links.lww.com/MCA/
A182). The RR with 95% confidence interval of MACE
was significantly lower in the CR group compared with
the IRA-only group (0.574; 0.429–0.768; P < 0.0001)
(Fig. 2). There was high heterogeneity of the included
studies (I2 = 62.3%), but there was no publication bias
(P = 0.36).
Recurrent nonfatal myocardial infarction
Recurrent MI was reported in all studies. RR for recur-
rent MI was not different between the two groups
(RR = 0.744; 0.473–1.170; P = 0.2; Fig. 5). Heterogeneity
across the included studies was moderate (I2 = 34.6%),
and no publication bias was found (P = 0.58).
Repeat revascularization
Nine of the 10 studies reported repeat revascularization.
There was a significantly lower risk for repeat revascu-
larization in the CR group compared with the IRA
revascularization-only group (RR = 0.506; 0.305–0.840;
P = 0.009; Fig. 6). Heterogeneity across the included
studies was high (I2 = 82.9%), and there was no publica-
tion bias among them (P = 0.84).

All-cause mortality Stroke

All-cause mortality was reported in eight of 10 studies. No
significant difference was found between the two groups
with respect to the risk of all-cause mortality (RR = 0.751;
0.529–1.066; P = 0.109; Fig. 3). Heterogeneity between
the studies was very low (I2 = 0.0%), and Egger’s test
found no evidence for publication bias (P = 0.92).
Six studies reported stroke events. There was a similar
risk for stroke between the two groups (RR = 0.647;
0.226–1.849; P = 0.41; Fig. 7). There was a low hetero-
geneity across the studies (I2 = 0.66%) and the Egger’s
test-based publication bias was just above the level of
significance (P = 0.06).

Cardiac mortality
Seven of the 10 studies reported cardiac mortality.
Significantly lower RR was found with CR (RR = 0.524;
0.314–0.875; P = 0.014; Fig. 4) compared with IRA-only
group. Heterogeneity across the included studies was low
(I2 = 0.44%), and there was no publication bias (Egger’s
test P = 0.09).
Major bleeding events
Four of the 10 studies reported major bleeding events; the
risk of which was similar among the two groups
(RR = 0.621; 0.279–1.380; P = 0.24; Fig. 8). Heterogeneity
across the studies was low (I2 = 0.0%), and there was no
publication bias (P = 0.69).

Fig. 2

Major adverse cardiac events

Study name Statistics for each study Events /Total Risk ratio and 95% CI
Risk Lower Upper
ratio limit limit Z-Value p-Value Complete IRA-only
COMPARE-ACUTE 0.380 0.249 0.581 -4.477 0.000 23 / 295 121 /590
EXPLORE 2.081 0.640 6.765 1.219 0.223 8/148 4/154
Hamza et al 0.250 0.075 0.832 -2.259 0.024 3/50 12/50
PRAGUE-13 1.155 0.609 2.191 0.440 0.660 17/106 15/108
DANAMI-3-PRIMULTI 0.586 0.410 0.839 -2.924 0.003 40/314 68/313
CvLPRIT 0.471 0.266 0.835 -2.576 0.010 15/150 31/146
PRAMI 0.391 0.244 0.627 -3.902 0.000 21/234 53/231
Ghani et al. 1.013 0.604 1.698 0.048 0.962 28/79 14/40
Politi et al. 0.431 0.291 0.637 -4.215 0.000 28/130 42/84
HELP AMI 0.599 0.261 1.375 -1.208 0.227 11/52 6/17
0.574 0.429 0.768 -3.744 0.000

0.1 0.2 0.5 1 2 5 10

Favours Complete Favours IRA-only
Heterogeneity: Tau2 = 0.124, Q-value = 23.9; df (Q) = 9 (p=0.004); I2 = 62.3 Revascularization Revascularization

Risk ratios of major adverse cardiovascular events with complete revascularization versus culprit-only revascularization. Pooled effects showed
significant difference in the incidence of major adverse cardiovascular events at follow-up between the two groups. CI, confidence interval;
IRA, infarct-related artery.

Copyright r 2018 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
 6 Coronary Artery Disease 2018, Vol 00 No 00

Fig. 3

All-cause mortality
Study name Statistics for each study Events/Total Risk ratio and 95% CI

Risk Lower Upper

ratio limit limit Z-Value p-Value Complete IRA-only
COMPARE-ACUTE 0.800 0.253 2.529 -0.380 0.704 4/295 10/590
Hamza et al 0.250 0.029 2.159 -1.260 0.208 1/50 4/50
PRAGUE-13 0.873 0.303 2.513 -0.251 0.802 6/106 7/108
DANAMI-3-PRIMULTI 1.359 0.634 2.913 0.789 0.430 15/314 11/313
CvLPRIT 0.389 0.125 1.214 -1.626 0.104 4/150 10/146
PRAMI 0.740 0.358 1.530 -0.811 0.417 12/234 16/231
Ghani et al. 4.613 0.254 83.610 1.034 0.301 4/79 0/40
Politi et al. 0.497 0.228 1.081 -1.762 0.078 10/130 13/84
HELP AMI 1.019 0.043 23.911 0.012 0.991 1/52 0/17
0.751 0.529 1.066 -1.603 0.109

0.1 0.2 0.5 1 2 5 10

Favours Complete Favours IRA-only

Heterogeneity: Tau2 = 0.000, Q-value = 7.32; df (Q) = 8 (p = 0.502); I2 = 0.00
Revascularization Revascularization

Risk ratios of all-cause mortality with complete revascularization versus culprit-only revascularization. Pooled effects showed nonsignificant difference
in the incidence of all-cause mortality at follow-up between the two groups. CI, confidence interval; IRA, infarct-related artery.

Fig. 4

Cardiac death
Study name Statistics for each study Events/Total Risk ratioand 95% CI

Risk Lower Upper

ratio limit limit Z-Value p-Value Complete IRA-only

COMPARE-ACUTE 1.000 0.252 3.970 0.000 1.000 3/295 6/590

EXPLORE 9.362 0.508 172.393 1.505 0.132 4/148 0/154

DANAMI-3-PRIMULTI 0.554 0.188 1.634 -1.071 0.284 5/314 9/313

CvLPRIT 0.278 0.059 1.317 -1.613 0.107 2/150 7/146

PRAMI 0.395 0.126 1.241 -1.590 0.112 4/234 10/231

Politi et al. 0.388 0.146 1.027 -1.906 0.057 6/130 10/84

HELP AMI 1.019 0.043 23.911 0.012 0.991 1/52 0/17

0.524 0.314 0.875 -2.469 0.014

0.1 0.2 0.5 1 2 5 10

Favours Complete Favours IRA-only

Heterogeneity: Tau2 = 0.002, Q-value = 6.02; df (Q) = 6 (p = 0.420); I2 = 0.44 Revascularization Revascularization

Risk ratios of cardiac death with complete revascularization versus culprit-only revascularization. Pooled effects showed significant difference in the
incidence of cardiac death at follow-up between the two groups. CI, confidence interval; IRA, infarct-related artery.

Contrast-induced nephropathy revascularization strategy and any of the outcomes ana-

Three of the 10 studies reported contrast-induced
nephropathy. There was no significant difference
between the two groups with respect to the occurrence of
contrast-induced nephropathy (RR = 1.061; 0.445–2.531;
P = 0.893; Fig. 9). Heterogeneity across the studies was
low (I2 = 0.0%), and the publication bias was not sig-
nificant (P = 0.42).
Influence analysis
The influence analysis showed that no single study sig-
nificantly altered the direction of association between the
lyzed. When the EXPLORE trial, which included only
patients who had chronic total occlusion (CTO) in non-
IRA arteries, was excluded, the risk for cardiac death was
further reduced in patients treated with CR versus those
treated with IRA-only revascularization strategy
(RR = 0.478; 0.284–0.803; P = 0.005) (Fig. 10).
Analysis in fractional flow reserve-guided patients
When we pooled only studies that compared FFR-guided
CR versus IRA-only strategy, the risk for repeat revascu-
larization was reduced (RR = 0.480; 0.242–0.951; P = 0.035)

Copyright r 2018 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
 CR for patients with STEMI Bajraktari et al. 7

Fig. 5

Myocardial infarction
Study name Statistics for each study Events/Total Risk ratio and 95% CI

Risk Lower Upper

ratio limit limit Z-Value p-Value Complete IRA-only
COMPARE-ACUTE 0.500 0.221 1.131 -1.664 0.096 7/295 28/590
EXPLORE 1.734 0.422 7.128 0.763 0.445 5/148 3/154
Hamza et al 0.500 0.047 5.339 -0.574 0.566 1/50 2/50
PRAGUE-13 1.401 0.587 3.345 0.759 0.448 11/106 8/108
DANAMI-3-PRIMULTI 0.935 0.470 1.857 -0.193 0.847 15/314 16/313
CvLPRIT 0.487 0.091 2.617 -0.839 0.401 2/150 4/146
PRAMI 0.346 0.149 0.801 -2.476 0.013 7/234 20/231
Ghani et al. 14.863 0.909 242.933 1.893 0.058 14/79 0/40
Politi et al. 0.554 0.193 1.591 -1.097 0.272 6/130 7/84
HELP AMI 0.327 0.022 4.949 -0.806 0.420 1/52 1/17
0.744 0.473 1.170 -1.281 0.200

0.1 0.2 0.5 1 2 5 10

Favours Complete Favours IRA-only

Heterogeneity: Tau2 = 0.158, Q-value = 13.34; df (Q) = 9 (p = 0.14); I2 = 34.6 Revascularization Revascularization

Risk ratios of nonfatal myocardial infarction with complete revascularization versus culprit-only revascularization. Pooled effects showed nonsignificant
difference in the incidence of myocardial infarction at follow-up between the two groups. CI, confidence interval; IRA, infarct-related artery.

Fig. 6

Repeat revascularization
Study name Statistics for each study Events/Total Risk ratio and 95% CI

Risk Lower Upper

ratio limit limit Z-Value p-Value Complete IRA-only
EXPLORE 2.029 1.244 3.311 2.833 0.005 39/148 20/154
Hamza et al 0.167 0.021 1.335 -1.688 0.091 1/50 6/50
COMPARE-ACUTE 0.350 0.216 0.565 -4.285 0.000 18/295 103/590
CvLPRIT 0.487 0.215 1.102 -1.727 0.084 8/150 16 /146
HELP AMI 0.490 0.204 1.177 -1.594 0.111 9/52 6 /17
PRAMI 0.343 0.200 0.589 -3.887 0.000 16/234 46/231
DANAMI-3-PRIMULTI 0.326 0.193 0.551 -4.188 0.000 17/314 52/313
Politi et al. 0.323 0.181 0.577 -3.819 0.000 14/130 28/84
Ghani et al. 0.976 0.580 1.645 -0.089 0.929 27/79 14/40
0.506 0.305 0.840 -2.631 0.009

0.1 0.2 0.5 1 2 5 10

Heterogeneity: Tau2 = 0.463, Q-value = 46.78; df (Q) = 8 (p = 0.00); I2 = 82.9 Favours Complete Favours IRA-only
Revascularization Revascularization

Risk ratios of repeat revascularization with complete revascularization versus culprit-only revascularization. Pooled effects showed significant
difference in the incidence of repeat revascularization at follow-up between the two groups. CI, confidence interval; IRA, infarct-related artery.

in patients who underwent CR compared with those who
underwent IRA-only strategy, but recurrent MI
(RR = 0.977; 0.344–2.773; P = 0.965) and all-cause mortality
(RR = 1.232; 0.662–2.292; P = 0.509) were not reduced
between groups. Furthermore, the meta-analysis of the two
FFR studies [8,11] that reported the cardiac death did not
show significant benefit from CR on this outcome
(RR = 0.694; 0.296–1.625; P = 0.4) (Fig. 11).
Follow-up time effect on the outcomes
Meta-regression found that MACE was affected by the
follow-up period when fixed effect model was used
(P = 0.02), but it becomes insignificant using the random
effect model (P = 0.2). Moreover, repeat revascularization
was found to be significantly affected with the fixed
effect model (P = 0.02) but not when random effect
model was used (P = 0.60) (Fig. 12). All-cause mortality,

Copyright r 2018 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
 8 Coronary Artery Disease 2018, Vol 00 No 00

Fig. 7

Stroke
Study name Statistics for each study Events/Total Risk ratio and 95% CI

Risk Lower Upper

ratio limit limit Z-Value p-Value Complete IRA-only
COMPARE-ACUTE 0.222 0.012 4.107 -1.011 0.312 0/295 4/590

EXPLORE 0.208 0.010 4.298 -1.016 0.310 0/148 2/154

Hamza et al 0.333 0.014 7.991 -0.678 0.498 0/50 1/50

PRAGUE-13 0.146 0.008 2.784 -1.280 0.201 0/106 3/108

DANAMI-3-PRIMULTI 3.987 0.448 35.474 1.240 0.215 4/314 1/313

CvLPRIT 0.973 0.139 6.819 -0.027 0.978 2/150 2/146

0.647 0.226 1.849 -0.813 0.416

0.1 0.2 0.5 1 2 5 10

2 2 Favours Complete Favours IRA-only

Heterogeneity: Tau = 0.012, Q-value = 5.03; df (Q) = 5 (p = 0.41); I = 0.66
Revascularization Revascularization

Risk ratios of stroke with complete revascularization versus culprit-only revascularization. Pooled effects showed nonsignificant difference in the
incidence of stroke at follow-up between the two groups. CI, confidence interval; IRA, infarct-related artery.

Fig. 8

Major bleeding events

Study name Statistics for each study Events/Total Risk ratio and 95% CI

Risk Lower Upper

ratio limit limit Z-Value p-Value Complete IRA-only

COMPARE-ACUTE 0.750 0.200 2.806 -0.427 0.669 3/295 8/590

EXPLORE 3.121 0.128 76.004 0.699 0.485 1/148 0/154

DANAMI-3-PRIMULTI 0.249 0.028 2.217 -1.246 0.213 1/314 4/313

CvLPRIT 0.556 0.166 1.860 -0.952 0.341 4/150 7/146

0.621 0.279 1.380 -1.169 0.242

0.1 0.2 0.5 1 2 5 10

Favours Complete Favours IRA-only

Heterogeneity: Tau2 = 0.000, Q-value = 1.76; df (Q) = 3 (p = 0.623); I2 = 0.00 Revascularization Revascularization

Risk ratios of major bleeding events with complete revascularization versus culprit-only revascularization. Pooled effects showed nonsignificant
difference in the incidence of major bleeding events at follow-up between the two groups. CI, confidence interval; IRA, infarct-related artery.

cardiac mortality, and MI were not affected by either (iii) CR is safe regarding procedure-related stroke,
model (P = 0.29, 0.19, and 0.17, respectively). contrast-induced nephropathy, and major bleeding.

The evidence on which the current American College of

Discussion
This meta-analysis of RCTs compared the efficacy and
safety of CR versus IRA-only PCI in hemodynamically
stable patients with STEMI. The main findings can be
summarized as follows: (i) CR in hemodynamically stable
patients with STEMI and MVD is associated with a
significant reduction (42%) in the risk for MACE over a
median of 17.5 months (range: 4–38 months). This ben-
efit is derived from a significant reduction in the need for
repeat revascularization (43%) and from a significant
reduction in cardiac deaths (31%). (ii) CR failed to show
benefit in reducing all-cause mortality and nonfatal MI,
compared with IRA-only revascularization strategy.
Cardiology/American Heart Association and European
Society of Cardiology guidelines on CR of patients with
STEMI and MVD are based is poor, especially in hemo-
dynamically stable patients, with a weak recommendation
class (IIb) [9,10]. This is owing to the balanced advantages
(e.g. the high incidence of MVD in patients with STEMI
[1] and poor outcome [2], generalized plaque instability,
adverse events owing to lack of treatment of other lesions
[20], reduced frequency of future procedures, optimum
anticoagulant therapy, associated bleeding events arising
from future procedures, and decreased costs [21]) and dis-
advantages (e.g. prolonged intervention duration, increased

Copyright r 2018 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
 CR for patients with STEMI Bajraktari et al. 9

Fig. 9

Contrast induced nephropathy

Study name Statistics for each study Events/Total Risk ratio and 95% CI

Risk Lower Upper

ratio limit limit Z-Value p-Value Complete IRA-only

Hamza etal 3.000 0.323 27.871 0.966 0.334 3/50 1/50

DANAMI-3-PRIMULTI 0.854 0.290 2.514 -0.286 0.775 6/314 7/313

CvLPRIT 0.973 0.139 6.819 -0.027 0.978 2/150 2/146

1.061 0.445 2.531 0.134 0.893

0.1 0.2 0.5 1 2 5 10

Favours Complete Favours IRA-only

Heterogeneity: Tau2 = 0.00, Q-value = 0.998; df (Q) = 2 (p = 0.607); I2 = 0.000 Revascularization Revascularization

Risk ratios of contrast-induced nephropathy with complete revascularization versus culprit-only revascularization. Pooled effects showed
nonsignificant difference in the incidence contrast-induced nephropathy at follow-up between the two groups. CI, confidence interval; IRA, infarct-
related artery.

Fig. 10

MACE All-cause mortality

Favours Complete Favours IRA-only Favours Complete Favours IRA-only

Revascularization Revascularization Revascularization Revascularization

Cardiac death Repeat revascularization

Favours Complete Favours IRA-only

Revascularization Revascularization
Favours Complete Favours IRA-only
Revascularization Revascularization

Myocardial infarction

Favours Complete Favours IRA-only

Revascularization Revascularization

Risk ratios of efficacy outcomes with complete revascularization versus culprit-only revascularization without the EXPLORE trial. CI, confidence
interval; IRA, infarct-related artery; MACE, major adverse cardiovascular events.

Copyright r 2018 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
 10 Coronary Artery Disease 2018, Vol 00 No 00
Fig. 11
MACE
Favours Complete
Revascularization
Cardiac death
Favours Complete
Revascularization
Myocardial infarction
Favours Complete
Revascularization
Risk ratios of efficacy outcomes with complete revascularization versus culprit-only revascularization of the fractional flow reserve-guided studies
alone. CI, confidence interval; IRA, infarct-related artery; MACE, major adverse cardiovascular events.
radiation doses, increased contrast volumes with associated
risks of contrast-induced nephropathy, the need for high
dose of anticoagulants, and overestimation of the nonculprit
stenosis severity owing to the hyperadrenergic state and
associated vasoconstriction [22]). However, the results of
this meta-analysis provide significant support for the CR
compared with the single culprit lesion-only revasculariza-
tion strategy. The primary endpoint (MACE) was sig-
nificantly lower in patients who received CR compared
with IRA-only PCI group, consistent with the results of
most of the individual included studies [6–8,13,15] and
previous meta-analyses [23–25]. This finding is further
strengthened by the recently published COMPARE-
ACUTE randomized trial [11], in contrast to previous
RCTs which reported conflicting results [14,16–18].
The MACE risk reduction in patients who underwent
CR for STEMI with MVD is mainly affected by the
reduction of the need of repeat revascularization com-
pared with the IRA-only group. This also explains the
higher rate of cardiac death in patients in the IRA-only
group, who were discharged after the primary PCI of the
culprit coronary lesion with significant stenosis in other
arteries, and some of these deaths might have been
caused by vulnerable plaques. On the contrary, the sen-
sitivity analysis showed a higher risk of cardiac death
Copyright r 2018 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
All-cause mortality
Favours IRA-only Favours Complete Favours IRA-only
Revascularization Revascularization Revascularization
Repeat revascularization
Favours IRA-only
Favours Complete Favours IRA-only
Revascularization
Revascularization Revascularization
Favours IRA-only
Revascularization
(P = 0.005) in patients with CR compared with those with
IRA-only treatment, after exclusion of the EXPLORE
trial [13] which included only patients with CTO. These
findings may suggest that the benefit from CR in patients
with STEMI with MVD is more profound in non-CTO
patients.
Another issue closely related to the CR is the use of FFR.
The three studies [8,11,17] that compared FFR-guided
CR versus IRA-only strategy showed that the risk for
repeat revascularization was lower in patients with FFR-
guided CR. However, this significance was the same in
all included studies. Furthermore, the meta-analysis of
the two FFR studies [8,11] that reported the cardiac
death did not show significant benefit from CR on this
outcome (P =0.4). These results may suggest that treat-
ment using FFR-guided CR has no additional advantage
over CR alone in STEMI and MVD.
The risk for all-cause mortality and MI was not sig-
nificant between groups despite significantly lower
mortality and nonfatal MI in the CR group. The rest of
the clinical outcomes, including stroke, major bleeding,
and contrast-induced nephropathy were not different
between the two groups. This finding is reassuring with
respect to the safety of the CR strategy in stable STEMI
with MVD.
 CR for patients with STEMI Bajraktari et al. 11

Fig. 12

Meta regression of log-risk ratio of major adverse cardiac events on follow-up time – (a) random effects model, (b) fixed effects model; and of repeat
revascularization on follow-up time – (c) random effects model, (d) fixed effects model.

Few meta-analyses that included RCTs and non-
randomized studies which compared CR versus IRA-only
treatment have already been published. These meta-
analyses used different inclusion criteria, and the results
were different [3,23,24,26]. Moreover, the only published
meta-analysis that included the most recent and largest
RCT, COMPARE-ACUTE, did not include patients
with CTO in nonculprit arteries [27]. The results of our
meta-analysis differ from those of the recent one by
Vaidya et al. [27] as we found less difference between
groups regarding cardiac death and repeat revasculariza-
tion. In addition, we assessed the risk of stroke between
study groups, and the influence of CTO and FFR-guided
PCI on the results of CR.
The time from the primary PCI to the staged procedure
of CR was different in the included RCTs and was at the
investigator’s discretion. We consider that future RCTs
will establish the best time for the treatment of non-
culprit significant artery stenosis. Moreover, there is no
convincing evidence to support CR in STEMI compli-
cated by cardiogenic shock [28].
Conclusion
For patients with STEMI and MVD undergoing primary
PCI, current meta-analysis suggests that the risk of
MACE, repeat revascularization, and cardiac death is
reduced by CR. However, the risk for all-cause mortality
and PCI-related complications is not different from the
IRA-only revascularization strategy. Although these
findings support the cardiac mortality and safety benefits
of CR in stable patients with STEMI, further RCTs
powered for clinical outcomes of interest are required to
provide better guidance for optimum management of
such patients.
Acknowledgements
Conflicts of interest
There are no conflicts of interest.

Copyright r 2018 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
 12 Coronary Artery Disease 2018, Vol 00 No 00
References
1 Cardarelli F, Bellasi A, Ou FS, Shaw LJ, Veledar E, Roe MT, et al. Combined
impact of age and estimated glomerular filtration rate on in-hospital mortality
after percutaneous coronary intervention for acute myocardial infarction
(from the American College of Cardiology National Cardiovascular Data
Registry). Am J Cardiol 2009; 103:766–771.
2 Sorajja P, Gersh BJ, Cox DA, McLaughlin MG, Zimetbaum P, Costantini C,
et al. Impact of multivessel disease on reperfusion success and clinical
outcomes in patients undergoing primary percutaneous coronary intervention
for acute myocardial infarction. Eur Heart J 2007; 28:1709–1716.
3 Witberg G, Kornowski R. Current perspectives on revascularization in
multivessel ST elevation myocardial infarction. Coron Artery Dis 2017;
28:498–506.
4 Task Force on the Management of ST-Segment Elevation Acute Myocardial
Infarction of the European Society of Cardiology (ESC), Steg PG, James SK,
Atar D, Badano LP, Blömstrom-Lundqvist C, et al. ESC guidelines for the
management of acute myocardial infarction in patients presenting with ST-
segment elevation. Eur Heart J 2012; 33:2569–2619.
5 O’Gara PT, Kushner FG, Ascheim DD, Casey DE, Chung MK, de Lemos JA,
et al. 2013 ACCF/AHA guideline for the management of ST-elevation
myocardial infarction: executive summary: a report of the American College
of Cardiology Foundation/American Heart Association Task Force on
Practice Guidelines. Circulation 2013; 127:529–555.
6 Wald DS, Morris JK, Wald NJ, Chase AJ, Edwards RJ, Hughes LO, et al.
Randomized trial of preventive angioplasty in myocardial infarction. N Engl J
Med 2013; 369:1115–1123.
7 Gershlick AH, Khan JN, Kelly DJ, Greenwood JP, Sasikaran T, Curzen N,
et al. Randomized trial of complete versus lesion-only revascularization in
patients undergoing primary percutaneous coronary intervention for STEMI
and multivessel disease: the CvLPRIT trial. J Am Coll Cardiol 2015;
65:963–972.
8 Engstrøm T, Kelbæk H, Helqvist S, Høfsten DE, Kløvgaard L, Holmvang L,
et al. Complete revascularisation versus treatment of the culprit lesion only in
patients with ST-segment elevation myocardial infarction and multivessel
disease (DANAMI-3–PRIMULTI): an open-label, randomised controlled trial.
Lancet 2015; 386:665–671.
9 Levine GN, Bates ER, Blankenship JC, Bailey SR, Bittl JA, Cercek B, et al.
2015 ACC/AHA/SCAI focused update on primary percutaneous coronary
intervention for patients with st-elevation myocardial infarction: an update of
the 2011 ACCF/AHA/SCAI guideline for percutaneous coronary intervention
and the 2013 ACCF/AHA Guideline for the management of ST-elevation
myocardial infarction. J Am Coll Cardiol 2016; 67:1235–1250.
10 Authors/Task Force Members, Windecker S, Kolh P, Alfonso F, Collet JP,
Cremer J, et al. 2014 ESC/EACTS Guidelines on myocardial
revascularization: the Task Force on Myocardial Revascularization of the
European Society of Cardiology (ESC) and the European Association for
Cardio-Thoracic Surgery (EACTS)Developed with the special contribution of
the European Association of Percutaneous Cardiovascular Interventions
(EAPCI). Eur Heart J 2014; 35:2541–2619.
11 Smits PC, Abdel-Wahab M, Neumann FJ, Boxma-de Klerk BM, Lunde K,
Schotborgh CE, et al. Fractional flow reserve-guided multivessel angioplasty
in myocardial infarction. N Engl J Med 2017; 376:1234–1244.
12 Moher D, Liberati A, Tetzlaff J, Altman DG. PRISMA Group. Preferred
reporting items for systematic reviews and meta-analyses: the PRISMA
statement. J Clin Epidemiol 2009; 62:1006–1012.
13 Politi L, Sgura F, Rossi R, Monopoli D, Guerri E, Leuzzi C, et al. A randomised
trial of target-vessel versus multi-vessel revascularisation in ST-elevation
myocardial infarction: major adverse cardiac events during long-term follow-
up. Heart 2010; 96:662–667.
14 Henriques JPS, Hoebers LP, Råmunddal T, Laanmets P, Eriksen E, Bax M,
et al. Percutaneous intervention for concurrent chronic total occlusions in
Copyright r 2018 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
patients with STEMI: the EXPLORE Trial. J Am Coll Cardiol 2016;
68:1622–1632.
15 Hamza M, Elgendy IY. A randomized trial of complete versus culprit-only
revascularization during primary percutaneous coronary intervention in
diabetic patients with acute ST elevation myocardial infarction and multi
vessel disease. J Interv Cardiol 2016; 29:241–247.
16 Hlinomaz O Multivessel coronary disease diagnosed at the time of primary
PCI for STEMI: complete revascularization versus conservative strategy
(PRAGUE 13 trial). Paper Presented at EuroPCR; 19 May 2015; Paris,
France.
17 Ghani A, Dambrink JHE, van’t Hof AWJ, Ottervanger JP, Gosselink ATM,
Hoorntje JCA. Treatment of non-culprit lesions detected during primary PCI:
long-term follow-up of a randomised clinical trial. Neth Heart J 2012;
20:347–353.
18 Di Mario C, Mara S, Flavio A, Imad S, Antonio M, Anna P, et al. Single vs
multivessel treatment during primary angioplasty: results of the multicentre
randomised HEpacoat for cuLPrit or multivessel stenting for Acute
Myocardial Infarction (HELP AMI) Study. Int J Cardiovasc Intervent 2004; 6:
128–133.
19 Dambrink JHE, Debrauwere JP, van’t Hof AWJ, Ottervanger JP,
Gosselink ATM, Hoorntje JCA, et al. Non-culprit lesions detected during
primary PCI: treat invasively or follow the guidelines? EuroIntervention 2010;
5:968–975.
20 Cavender MA, Milford-Beland S, Roe MT, Peterson ED, Weintraub WS,
Rao SV. Prevalence, predictors, and in-hospital outcomes of non-infarct
artery intervention during primary percutaneous coronary intervention for ST-
segment elevation myocardial infarction (from the National Cardiovascular
Data Registry). Am J Cardiol 2009; 104:507–513.
21 Moussa I, Reimers B, Moses J, di Mario C, di Francesco L, Ferraro M, et al.
Long-term angiographic and clinical outcome of patients undergoing
multivessel coronary stenting. Circulation 1997; 96:3873–3879.
22 Corpus RA, House JA, Marso SP, Grantham JA, Huber KC, Laster SB, et al.
Multivessel percutaneous coronary intervention in patients with multivessel
disease and acute myocardial infarction. Am Heart J 2004; 148:493–500.
23 Anantha Narayanan M, Reddy YN, Sundaram V, Reddy YN, Baskaran J,
Agnihotri K, et al. What is the optimal approach to a non-culprit stenosis after
ST-elevation myocardial infarction – conservative therapy or upfront
revascularization? An updated meta-analysis of randomized trials. Int J
Cardiol 2016; 216:18–24.
24 Elgendy IY, Mahmoud AN, Kumbhani DJ, Bhatt DL, Bavry AA. Complete or
culprit-only revascularization for patients with multivessel coronary artery
disease undergoing percutaneous coronary intervention: a pairwise and
network meta-analysis of randomized trials. JACC Cardiovasc Interv 2017;
10:315–324.
25 Spencer FA, Sekercioglu N, Prasad M, Lopes LC, Guyatt GH. Culprit vessel
versus immediate complete revascularization in patients with ST-segment
myocardial infarction – a systematic review. Am Heart J 2015 Dec;
170:1133–1139.
26 Shah R, Berzingi C, Mumtaz M, Jasper JB, Goswami R, Morsy MS, et al.
Meta-analysis comparing complete revascularization versus infarct-related
only strategies for patients with ST-segment elevation myocardial infarction
and multivessel coronary artery disease. Am J Cardiol 2016;
118:1466–1472.
27 Vaidya SR, Qamar A, Arora S, Devarapally SR, Kondur A, Kaul P. Culprit
versus multivessel coronary intervention in ST-segment elevation myocardial
infarction: a meta-analysis of randomized trials. Coron Artery Dis 2017. doi:
10.1097/MCA.0000000000000578.
28 Thiele H, Akin I, Sandri M, Fuernau G, de Waha S, Meyer-Saraei R, et al. PCI
strategies in patients with acute myocardial infarction and cardiogenic shock.
N Engl J Med 2017; 377:2419–2432.