International Journal of Cardiology 245 (2017) 27–34
Contents lists available at ScienceDirect
International Journal of Cardiology
j ourna l homepage: www.e lsev ie r .com/ locate / i j ca rdGlobal geographical variations in ST-segment elevation myocardial
infarction management and post-discharge mortality☆,☆☆Xavier Rosselló a, Yong Huo b, Stuart Pocock a, Frans Van de Werf c, Chee Tang Chin d, Nicolas Danchin e,
Stephen W.-L. Lee f, Jesús Medina g, Ana Vega g, Héctor Bueno h,i,j,⁎
a London School of Hygiene and Tropical Medicine, London, UK
b Department of Cardiology, Peking University First Hospital, Beijing, China
c University Hospitals Leuven, Belgium
d National Heart Centre Singapore, Singapore
e Hôpital Européen Georges Pompidou & René Descartes University, Paris, France
f Queen Mary Hospital, Hong Kong SAR, China
g Medical Evidence and Observational Research, Global Medical Affairs, AstraZeneca, Madrid, Spain
h Centro Nacional de Investigaciones Cardiovasculares (CNIC), Spain
i Instituto de investigación i+12 and Cardiology Department, Hospital Universitario 12 de Octubre, Spain
j Universidad Complutense de Madrid, Spain☆ All authors take responsibility for all aspects of the reli
the data presented and their discussed interpretation.
☆☆ Funding: EPICOR and EPICOR Asia are funded by Astr
tional study, no drugs were supplied or funded.
⁎ Corresponding author at: Centro Nacional de Inv
(CNIC), Melchor Fernandez Almagro, 3, 28029 Madrid, Sp
E-mail address: hbueno@cnic.es (H. Bueno).
http://dx.doi.org/10.1016/j.ijcard.2017.07.039
0167-5273/© 2017 The Authors. Published by Elsevier Irea b s t r a c ta r t i c l e i n f oArticle history:
Received 12 May 2017
Received in revised form 15 June 2017
Accepted 12 July 2017
Available online 15 July 2017Background: There is a shortage of information on regional variations in ST-segment elevation myocardial
infarction (STEMI) management and prognosis at a global level. We aimed to compare patient profiles,
in-hospital management and post-discharge mortality across several world regions.
Methods: In total, 11,559 patients with STEMI were enrolled in two prospective studies of acute coronary
syndrome survivors: EPICOR (4943 patients from 555 hospitals in 20 countries in Europe and Latin America
recruited between September 2010 and March 2011) and EPICOR Asia (6616 patients from 218 hospitals in
eight Asian countries recruited between June 2011 and May 2012). Comparisons were performed by eight
pre-defined regions: Northern Europe (NE), Southern Europe (SE), Eastern Europe (EE), Latin America (LA),
China (CN), India (IN), Southeast Asia (SA), and South Korea/Hong Kong/Singapore (KS).
Results: Reperfusion therapy rates ranged between 53.9% (IN) and 81.2% (SE), primary percutaneous coronary
intervention (PCI) between 24.8% (IN) and 65.6% (NE) and fibrinolysis between 8.1% (CN) and 34.2% (SA).
Median time to primary PCI (h) ranged from 3.9 (NE) to 20.9 (IN) and to fibrinolysis from 2.4 (SE) to 6.3 (IN).
Two-year mortality ranged between 2.5% in NE and 7.4% in LA. Regional variations in mortality persisted after
adjustment for reperfusion therapy and known prognostic factors.
Conclusions: Among patients with STEMI, there is a wide regional variation in clinical profiles, hospital care and
mortality. Substantial room for improvement remains at a global level for increasing reperfusion rates, reducing
delays and post-discharge mortality in patients with STEMI.
© 2017 The Authors. Published by Elsevier Ireland Ltd. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).Keywords:
ST-segment elevation myocardial infarction
Early management
Reperfusion therapy
Regional variation
Postdischarge mortality
Risk1. Introduction
Acutemyocardial infarction is one of the leading causes of morbidity
andmortalityworldwide [1]. A number of evidence-based interventions
have been adopted in recent decades, leading to widespread improve-
ment in prognosis [2]. However, the burden of cardiovascular riskability and freedom frombias of
aZeneca. Being a non-interven-
estigaciones Cardiovasculares
ain.
land Ltd. This is an open access articlfactors and the incidence of acute myocardial infarction are increasing
disproportionately in some geographical regions, and remain a major
global health concern [3,4].
Despite standard, evidence-based therapy advocated by international
guidelines, marked geographical disparities in clinical management and
mortality have been reported. Randomized clinical trials (RCT) [5,6]
and registries [7–10] have shown considerable variation in practice
across countries, highlighting a large gap between guideline recommen-
dations and received patient care.
Data from EPICOR (long-tErm follow uP of antithrombotic manage-
ment patterns In acute CORonary syndrome patients) and EPICOR Asia
cohorts have shown an unexplained variation in mortality [11] and
identified geographical region as an independent predictor of 2-yeare under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
28 X. Rosselló et al. / International Journal of Cardiology 245 (2017) 27–34mortality [12]. The aim of the present study is to describe regional
variation in patient profile, hospital management, including evidence-
based management, and 2-year mortality of patients with ST-segment
elevation myocardial infarction (STEMI), using combined data from
the EPICOR and EPICOR Asia studies.
2. Methods
EPICOR (NCT01171404) and EPICOR Asia (NCT01361386) are
prospective, international, observational, real-world practice cohort
studies comprising consecutive patients hospitalized for an acute coro-
nary syndromewithin 24 h (EPICOR) or 48 h (EPICORAsia) of symptom
onset, who survived to hospital discharge.
Theprotocol and case record formwere almost identical for both stud-
ies, and their rationale and designs have been described elsewhere [13,
14]. Informed consent was obtained from each patient and the study pro-
tocol conforms to the ethical guidelines of the 1975 Declaration of Helsin-
ki. The current study is a pre-defined secondary objective of both studies.
Results for STEMI patients were analyzed according to eight pre-
specified regions: Southern Europe (France, Greece, Italy and Spain),
Northern Europe (Belgium, Denmark, Finland, Germany, Luxembourg,
the Netherlands, Norway and UK), Eastern Europe (Poland, Romania,
Slovenia and Turkey), Latin America (Argentina, Brazil, Mexico and
Venezuela) [12], China, India, Southeast Asia (Malaysia, Vietnam and
Thailand) and South Korea/Hong Kong/Singapore. Supplementary
Table S1 lists the participating regions and the number of patients
enrolled per country.
2.1. Statistical analysis
Continuous parameters were mostly presented as mean (standard
deviation) and regions were compared using one-way analysis ofFig. 1.Geographical distribution of EPICOR study. Patients were distributed across the following
1145), Latin America (n=1066), China (n=3961), India (n=1482), Southeast Asia (n=751
India; KS, South Korea/Hong Kong/Singapore; LA, Latin America; NE, Northern Europe; SA, Souvariance. Ordinal and highly skewed distributed variables were pre-
sented bymedian and inter-quartile range, and regions were compared
using the non-parametric Kruskal–Wallis test. Categorical data were
expressed as percentages and compared between regions using Chi-
square tests. Overall survival curves were obtained using the Kaplan–
Meier method and compared using the log-rank test. We used Poisson
regression models to estimate 2-year mortality rates in each region,
adjusted for 16 known predictors of 2-year mortality in the overall
study population. These predictors were previously obtained from a
risk scoring system developed using forward stepwise Cox regression
to estimate 2-year mortality risk in acute coronary syndrome patients
and with well reported goodness-of-fit, discriminatory power and
internal validation [12]. Those predictors were (ranked by predictive
strength) age, low ejection fraction, no coronary revascularization or
thrombolysis, elevated serum creatinine, poor EQ-5D score, low
hemoglobin, previous cardiac disease, previous chronic obstructive
pulmonary disease, elevated blood glucose, on diuretics at discharge,
male sex, lower education level, on aldosterone inhibitor at discharge,
body mass index, in-hospital cardiac complications and Killip class
[12]. All p-values were two-sided and values of b0.05 were considered
statistically significant. All statistical analyses were performed using
STATA software, version 13.1 (Stata Corp, College Station, TX, USA).
Results are reported according to the research reporting guidelines for
observational studies (STROBE guidelines) [15].
3. Results
In total, 11,559 patients with STEMI were enrolled between
September 2010 and March 2011 from 555 hospitals in 20 countries
across Europe and Latin America (EPICOR, n = 4943), and between
June 2011 and May 2012 from 218 hospitals across eight countries
and regions in Asia (EPICOR Asia, n = 6616). Fig. 1 shows theregions: Southern Europe (n=1124), Northern Europe (n=1608), Eastern Europe (n=
) and South Korea, Hong Kong and Singapore (n=422). CN, China; EE, Eastern Europe; IN,
theast Asia; SE, Southern Europe.
Table 1
Patient and hospital characteristics.
All patients Southern
Europe
Northern
Europe
Eastern
Europe
Latin
America
China SK/HK/Singapore India Southeast
Asia
n, (%) 11,559 (100) 1124 (9.7) 1608 (13.9) 1145 (9.9) 1066 (9.2) 3961 (34.3) 422 (3.7) 1482 (12.8) 751 (6.5)
Patient characteristics at baseline
Age, mean (SD) 58.9 (11.9) 61.6 (12.7) 59.6 (11.9) 57.7 (11.7) 58.7 (11.8) 59.4 (11.7) 59.1 (12.2) 55.7 (11.2) 59.2 (12.1)
Age ≥ 75, n (%) 1260 (10.9) 221 (19.7) 192 (11.9) 99 (8.7) 108 (10.1) 428 (10.8) 52 (12.3) 74 (5.0) 86 (11.5)
Male, n (%) 9370 (81.1) 917 (81.6) 1238 (77.0) 923 (80.6) 846 (79.4) 3231 (81.6) 358 (84.8) 1259 (85.0) 598 (79.6)
Hypertension, n (%) 5431 (47.5) 565 (50.8) 667 (42.2) 552 (48.9) 625 (59.5) 1950 (49.7) 207 (49.2) 564 (38.6) 301 (40.5)
Hypercholesterolemia, n (%) 2849 (25.8) 521 (47.3) 648 (41.9) 369 (34.0) 402 (41.4) 551 (14.5) 90 (21.6) 84 (6.0) 184 (25.0)
Diabetes mellitus, n (%) 2332 (20.5) 227 (20.5) 192 (12.1) 198 (17.7) 276 (26.4) 794 (20.3) 105 (24.9) 421 (28.8) 119 (16.1)
Smoking status, n (%) 3598 (33.3) 322 (30.2) 410 (27.0) 324 (29.8) 322 (32.0) 1214 (31.7) 135 (32.5) 682 (59.1) 189 (25.8)
Prior cardiovascular disease, n (%) 2194 (19.3) 213 (19.1) 325 (20.4) 264 (23.2) 240 (22.8) 820 (20.9) 63 (15.0) 161 (11.5) 108 (14.5)
Clinical status on admission
Creatinine ≥1.2 mg/dL, n (%) 2469 (22.5) 160 (15.2) 298 (19.3) 176 (15.6) 195 (20.6) 944 (24.1) 124 (29.6) 341 (28.1) 231 (30.8)
Blood glucose N160 g/dL, n (%) 2730 (27.0) 269 (26.5) 296 (22.3) 290 (27.0) 264 (28.7) 956 (25.0) 166 (45.5) 334 (34.4) 155 (24.5)
Hemoglobin b13 g/dL, n (%) 2796 (25.6) 208 (19.8) 287 (18.8) 201 (17.7) 196 (21.3) 1048 (26.7) 78 (18.6) 482 (39.6) 296 (39.7)
Killip class
I, n (%) 8714 (80.4) 923 (87.2) 1159 (90.0) 960 (85.6) 890 (86.7) 3027 (77.0) 296 (71.5) 910 (70.8) 549 (76.9)
II, n (%) 1386 (12.8) 96 (9.1) 84 (6.5) 123 (11.0) 94 (9.2) 638 (16.2) 61 (14.7) 221 (17.2) 69 (9.7)
III–IV, n (%) 738 (6.4) 39 (3.5) 45 (2.8) 39 (3.4) 43 (4.0) 264 (6.7) 57 (13.5) 155 (10.5) 96 (12.8)
Hospital characteristics⁎
Coronary/intensive care unit, n (%) 11,371 (98.4) 1102 (98.0) 1530 (95.2) 1142 (99.7) 1039 (97.5) 3933 (99.3) 422 (100) 1482 (100) 721 (96.0)
Cath lab, n (%) 10,510 (90.9) 965 (85.9) 1159 (72.1) 1000 (87.3) 903 (84.7) 3961 (100.0) 417 (98.8) 1482 (100) 623 (83.0)
If yes, 24/7 PCI program, n (%) 9497 (82.2) 888 (79.0) 828 (51.5) 939 (82.0) 752 (70.5) 3756 (94.8) 395 (93.6) 1441 (97.2) 498 (66.3)
Cardiac surgery, n (%) 9100 (78.7) 508 (45.2) 675 (42.0) 740 (64.6) 922 (86.5) 3782 (95.5) 417 (98.8) 1449 (97.8) 607 (80.8)
p-Value was b0.001 in all comparisons between regions.
Bold text depicts the highest value of each row; italic text illustrates the lowest value.
Cath, catheterization; HK, Hong Kong; PCI, percutaneous coronary intervention; SD, standard deviation; SK, South Korea.
⁎ Results are for number of patients, not number of hospitals (i.e., large hospitals contribute more patients).
29X. Rosselló et al. / International Journal of Cardiology 245 (2017) 27–34distribution of patients across countries and regions: Southern Europe
(n= 1124), Northern Europe (n= 1608), Eastern Europe (n= 1145),
Latin America (n=1066), China (n=3961), India (n=1482), Southeast
Asia (n = 751) and South Korea/Hong Kong/Singapore (n = 422).
Supplementary Table S1 illustrates each country contribution within
regions.
3.1. Patient profile
Significant differences were observed in patient profile across
regions (Table 1). Mean age ranged between 55.7 years (India) and
61.6 years (Southern Europe), diabetes mellitus (DM) between 12.1%
(Northern Europe) and 28.8% (India) and smoking between 25.8%
(Southeast Asia) and 59.1% (India). The percentage of male patients
was above 75% in all regions. India showed a marked difference in
cardiovascular risk factors compared with other regions, with theTable 2
Use of resources.
All patients Southern
Europe
Northern
Europe
Eas
Eu
Pre-hospital care
Pre-hospital ECG, n (%) 4883
(42.4)
463
(41.2)
945
(58.8)
452
(39
Time from first-symptom to ECG (in h),
median (IQR)
3.0
(1.3, 5.4)
1.7
(0.9, 4.0)
1.9
(1.0, 4.3)
2.5
(1.
In-hospital care
Length of hospital stay,
median in days (IQR)
6
(4,10)
6
(4, 8)
5
(3, 7)
5
(4,
LVEF recorded, n (%) 8615
(74.5)
1023
(91.0)
1174
(73.0)
978
(85
LVEF b40%, n (%) 1193
(10.3)
105
(9.3)
176
(11.0)
152
(13
p-Value was b0.001 in all comparisons between regions.
Bold text depicts the highest value of each row; italic text illustrates the lowest value.
ECG, electrocardiogram; HK, Hong Kong; IQR, interquartile range; LVEF, left ventricular ejectiolowest percentage of older patients, those with hypertension, hyper-
cholesterolemia and prior cardiovascular disease but the highest
percentage of male patients, DM and cigarette smoking.
Other patient characteristics at admission, including blood test
results and Killip class, are also illustrated in Table 1. Of note, elevated
serum creatinine and Killip class II to IV were more common in all
Asian regions compared with other regions.3.2. Hospital profile
There were major regional variations in the type of hospitals
recruiting patients. While almost all hospitals had a coronary/intensive
care unit, only 51.5% of patients in Northern Europe were attended in
hospitals with 24/7 percutaneous coronary intervention (PCI) program,
in contrast to 97.2% of patients in India. The percentage of patientstern
rope
Latin
America
China SK/HK/Singapore India Southeast
Asia
.5)
263
(24.7)
2059
(52.0)
92
(21.8)
355
(24.0)
254
(33.8)
1, 6.0)
3.0
(1.5, 7.3)
4.0
(1.5, 11.4)
2.1
(1.1, 5.4)
6.1
(2.7, 15.5)
2.5
(1.2, 5.4)
8)
5
(4, 8)
10
(7, 13)
5
(4, 7)
4
(3, 5)
4
(3, 6)
.4)
706
(66.2)
3007
(75.9)
358
(84.8)
777
(52.4)
592
(78.8)
.3)
138
(13.0)
197
(5.0)
63
(14.9)
254
(17.1)
108
(14.4)
n fraction; SK, South Korea.
30 X. Rosselló et al. / International Journal of Cardiology 245 (2017) 27–34attending hospitals with on-site cardiac surgery was higher in Asia
compared with Europe and Latin America.3.3. Pre-hospital and in-hospital management
Pre-hospital electrocardiogram(ECG)wasmore frequently recorded
in Northern Europe (58.8%) and China (52.0%) compared with other
regions, in particular, South Korea/Hong Kong/Singapore (21.8%).
Time to ECG was significantly longer in India compared with all other
regions (Table 2). Mean length of stay ranged between 4 (India and
Southeast Asia) and 10 days (China). During hospitalization, left-
ventricular ejection fraction (LVEF) assessment also varied widely
among countries, with marked differences in patients with significant
left ventricular systolic dysfunction (Table 2).Fig. 2. Type of reperfusion therapy by region. Panel A depicts type of reperfusion; panel B shows
interquartile range) to fibrinolysis. IQR, interquartile range; PPCI, primary percutaneous coronThe type of reperfusion and delays to primary percutaneous
coronary intervention (PPCI) and fibrinolysis are illustrated in Fig. 2
and Supplementary Table S2. Reperfusion therapy rates ranged
between 53.9% (India) and 81.2% (Southern Europe). Specifically, PPCI,
according to the investigators' reports, ranged between 24.8% (India)
and 65.6% (Northern Europe) and fibrinolysis between 8.1% (China)
and 34.2% (Southeast Asia). The median time to PPCI (defined as time
from symptom onset to PCI, in h) ranged from 3.9 h (Northern
Europe) to 20.9 h (India) and to fibrinolysis from 2.4 h (Southern
Europe) and 6.3 h (India). However, in those patients who underwent
PPCI with recorded timings (62.3% of them), percentages of PCI within
the first 12 h ranged from 86.6% (Eastern Europe) to 37.7% (India)
(Supplementary Table S2).
There were major variations across regions in regard to the key
medications prescribed at discharge, as shown in Table 3. Dualmedian time (and its interquartile range) to PPCI; panel C illustratesmedian time (and its
ary intervention.
Table 3
Medications used at discharge.
All patients Southern Europe Northern Europe Eastern Europe Latin America China SK/HK/Singapore India Southeast Asia
Medication at discharge
Antithrombotics
DAPT, n (%) 10,574
(91.5)
1040
(92.5)
1489
(92.6)
1048
(91.5)
914
(85.7)
3775
(95.3)
396
(93.8)
1206
(81.4)
706
(94.0)
Aspirin + clopidogrel 9771
(84.5)
891
(79.3)
1180
(73.4)
1014
(88.6)
866
(81.2)
3773
(95.3)
381
(90.3)
961
(64.8)
705
(93.9)
Aspirin + prasugrel 766
(6.6)
146
(13.0)
308
(19.2)
11
(1.0)
48
(4.5)
0
(0.0)
15
(3.6)
238
(16.1)
0
(0.0)
SAPT, n (%) 615
(5.3)
39
(3.5)
46
(2.9)
47
(4.1)
96
(9.0)
141
(3.6)
16
(3.8)
205
(13.8)
25
(3.3)
Anticoagulant, n (%) 300
(2.6)
43
(3.8)
73
(4.5)
50
(4.4)
49
(4.6)
14
(0.4)
10
(2.4)
43
(2.9)
18
(2.4)
Statin, n (%) 10,513
(91.0)
1070
(95.4)
1536
(95.6)
1068
(93.3)
984
(93.4)
3694
(93.5)
355
(84.3)
1129
(77.4)
677
(90.2)
Beta-blockers, n (%) 8894
(76.9)
995
(88.5)
1489
(92.6)
1052
(91.9)
847
(79.5)
2867
(72.4)
351
(83.2)
864
(58.3)
429
(57.1)
ACEI/ARB, n (%) 8239
(71.3)
897
(79.8)
1261
(78.4)
987
(86.2)
844
(79.2)
2598
(65.6)
341
(80.8)
822
(55.5)
489
(65.1)
MRA, n (%) 1184
(10.2)
99
(8.8)
105
(6.5)
165
(14.4)
159
(14.9)
473
(11.9)
23
(5.5)
104
(7.0)
56
(7.5)
p-Value was b0.001 in all comparisons between regions.
Bold text depicts the highest value of each row; italic text illustrates the lowest value.
Anticoagulants were defined as taking warfarin or dabigatran regardless of other medication. DAPT was defined as aspirin plus another oral antiplatelet agent, such as clopidogrel,
prasugrel or ticlopidine. SAPT was defined as taking aspirin alone or ≥1 other antiplatelet agent that did not include aspirin.
ACEI, angiotensin-converting enzyme inhibitor; ARB, angiotensin receptor blocker; DAPT, dual antiplatelet therapy; HK, Hong Kong; MRA, mineralocorticoid receptor antagonist; SAPT,
single antiplatelet therapy; SK, South Korea.
31X. Rosselló et al. / International Journal of Cardiology 245 (2017) 27–34antiplatelet therapy (DAPT) and statins showed relatively high pre-
scription rates inmost regions, above 90%. Cilostazol was used as an an-
tiplatelet treatment only in 1.8% of patients in Asia, with relatively high
rates of use in South Korea/Hong Kong/Singapore (12.1%).
Large variation in the use of beta-blockers and the use of either
angiotensin-converting enzyme (ACE) inhibitors or angiotensin receptor
blockers (ARB) was found, particularly between Eastern Europe (91.9%
and 86.2%, respectively) and India (58.3% and 55.5%, respectively). The
use of mineralocorticoid receptor antagonist (MRA) was particularly in-
frequent in South Korea/Hong Kong/Singapore (5.5%), despite having
one of the highest percentages of patients with LVEF b40% (14.9%).
3.4. Post-discharge mortality
Two-year post-discharge mortality rates (cumulative incidence)
ranged between 2.5% in Northern Europe and 7.4% in Latin America
(Fig. 3A). Fig. 3B and Supplementary Table S3 illustrate that these
regional variations in mortality persisted after adjustment for 16
established prognostic factors, with South Korea/Hong Kong/Singapore
being the region with the lowest adjusted 2-year mortality rate (2.7%,
95% confidence interval [CI]: 1.1–4.4) and Latin America showing the
highest (8.1%, 95% CI: 6.2–10), compared with other regions.
4. Discussion
Our descriptive analysis shows that, in spite of the dissemination of
similar recommendations by practice guidelines worldwide, large
regional variations in clinical practice patterns for patients with STEMI
can still be found. We also observed significant differences across
regions in post-discharge mortality risk. Overall, the results of our
study suggest that there is substantial room for improvement in the
management of STEMI and for improving patient outcomes following
hospital discharge globally. However, different targets for intervention
may be needed for different regions.
4.1. Major findings
Unsurprisingly, patient characteristics on admission differed
between regions, reflecting regional differences in cardiovascular riskat a population level. While Southern Europe had the highest percent-
age of patients aged ≥75 years, the percentage in India was four-fold
less, but with a markedly elevated proportion of patients with DM and
currently smoking. EPICOR was designed to enrol a representative
sample of hospitals at a national level [13]. However, hospitals in Asia
were, overall, better equipped than those participating in European
and Latin American regions. This might have had an influence on the
rate of PCI procedures and potentially on patient outcomes [16].
Early reperfusionwith PPCI or fibrinolytic therapy is a cornerstone in
the treatment of STEMI [17,18]. However, up to a third of patients did
not receive reperfusion therapy, with significant geographical differ-
ences between regions as well as in the preferred type of reperfusion
therapy. Large inequalities in the use of reperfusion therapy for patients
with STEMI have previously been reported in randomized trials and
registries [6,7,10,19,20]. Large differences in the use of thrombolytic
therapy (highest in India, Latin America and Southeast Asia) may be
explained by differences in healthcare system organization. The low
usage of reperfusion therapy and, in particular, long delays in India are
of concern. Geography, weather, local resources and the organization
of regional health system networking may all contribute [21]. Regard-
less of the long delays experienced in India, there is still substantial
variation in the median times to PPCI across the other regions that
need to be analyzed and corrected. These differences should not be
interpreted from an economic perspective only, i.e., regional income
level, but by other system features that may work as opportunities for
improvement, such as the level of coordination within networks,
availability and preparation for thrombolysis administration when
needed, availability of adequately prepared transport systems for the
local geography and weather, traffic issues and preferential pathways
or health education at a population level. Without having a global and
multidisciplinary approach tackling all factors involved in STEMI
networks, it will be very difficult to reach evidence-based time frames
needed for efficient PPCI programs at a global level [22]. Thus, this
represents a sizeable health issue worldwide, with vast implications in
both mortality and morbidity.
In patients with STEMI, clinical guidelines recommend a variety of
medications at discharge that are considered quality indicators:
antithrombotics, statins, beta-blockers, ACE inhibitors or ARBs, and
MRAs [23]. The gap between evidence-based guideline recommendations
Fig. 3.Mortality by region. Panel A depicts unadjusted mortality; panel B illustrates adjusted mortality for 16 key factors.
32 X. Rosselló et al. / International Journal of Cardiology 245 (2017) 27–34and current practice is evident in our study, althoughwe found a reason-
able proportion of patients were discharged on DAPT and statins, above
90% in most regions. In the Global Registry of Acute Coronary Events
(GRACE), the use of aspirin, statins and ACE inhibitors varied only mod-
estly by geographical region compared with other treatments [16]. Rea-
sons for underuse need to be clarified, although the Prospective Urban
Rural Epidemiology (PURE) study has implicated availability and afford-
ability ofmedications, patients' attitudes and knowledge towards preven-
tative medicine, access to healthcare providers and prescribing patterns
[24].
The large variation in regional mortality is a matter of concern.
Whether these differences reflect differences in patient characteristics,
quality of care or, to some extent, patient selection is not clear. It is
known that STEMI outcomes vary considerably worldwide, with
differences in prognosis both within [8] and between countries [25].
Consistent with our results, Latin America has already been described
as a region with a higher adjusted mortality risk compared with other
regions such as Western Europe, Eastern Europe and North America[5]. In RCTs, discrepancies in treatment effect across regions [26] can
partially be explained by differences in patient cohorts or management
strategies. However, our data reveal an unexplained difference even
after adjustment by confounding factors. This disparity may be partly
explained by differences in the following four factors. Firstly, we
recorded differences in patients' cardiovascular risk factors, which
may be explained by different ethnic and regional backgrounds,
encompassing both genetic and cultural contexts [3]. Secondly, differ-
ences in individual and regional socioeconomic status, either within
[27] or between countries [28], are linked to mortality following
STEMI. Thirdly, performance of national or regional healthcare systems
(e.g., pre-hospital care access, STEMI networks, and access to medica-
tions) as well as hospital characteristics (e.g., cathetherization facilities,
24/7 PCI programs, and number of beds) may also lead to health
disparities [29]. Finally, there may be differences in local standard
clinical practice in terms of the dosage and duration of post-discharge
medication, approach to up-titration and changes in medication during
the follow-up period, and patient adherence to drug regimens.
33X. Rosselló et al. / International Journal of Cardiology 245 (2017) 27–34India deserves special attention in both the assessment of patient
management and outcomes. Despite having the most PPCI-capable
centres, it has the lowest rate of PPCI and the longest delays.
Likewise, its adjusted post-discharge mortality was lower than in Latin
America despite having a lower compliance to evidence-based
treatment, as represented by both reperfusion and post-dischargemed-
ication. Several reasons may explain these characteristics or special
features.
First, the long distances and the local healthcare systemmay explain
the time delays in reperfusion [30]. Second, Indian patients were
younger and showed a much higher prevalence of current smoking
than in any other region. Third, a potential cohort survival effect given
the nature of our inclusion criteria – the sickest patients may have
died in-hospital and thus not had been eligible for our cohort. Of note,
as important as assessing the causes of India's singularity is seeking
for potential ground for outcome improvement. We highly encourage
those initiatives aimed to narrow the disparities in patientmanagement
and outcomes between regions, such as the one recently published
demonstrating that a hub-and-spoke model in South India improved
STEMI care through greater use of PCI [31].
4.2. Strengths and limitations
Several considerations are important in interpreting our findings.
Despite being adjusted by 16 known predictors of 2-year mortality,
this study has limitations inherent to an analysis of observational
data: differences on mortality across regions due to a potential patient
selection bias cannot be completely ruled out. While both studies
were designed to recruit representative patients from representative
centres in each country, we cannot rule out bias in regard to the type
of hospitals participating, which may not necessarily represent clinical
practice in that geographical area. Hence, caution must be expressed
in extrapolating our results to non-EPICOR recruiting centres. Moreover,
regional grouping is needed for simplicity but is a somewhat artificial
construct. Since there are many different approaches to forming
geographical groupings,we considered countries' similarities in practice
patterns in addition to geographical neighbourhood, but this cannot
capture potential within-country and within-region variability. Finally,
our study lacks an assessment of other regions such asAfrica, theMiddle
East and North America.
A major strength of our study is that we assessed a comprehensive
set of patient and hospital characteristics, practice patterns and adjusted
2-year mortality across an extended range of regions. The main
evidence-based therapies for patients with STEMI, advocated by clinical
guidelines, were compared across territories in our real-world clinical
practice registry. Other strengths include a large sample size and a
broad worldwide representativeness of our findings.
4.3. Future implications
EPICOR has provided a unique opportunity to investigate regional
variations. This study should be interpreted as a preliminary step in
mapping those gaps and needs that should be addressed worldwide.
There is a global lack of continuous national quality of care and outcome
registries for acute myocardial infarction, such as those present in
the UK (NICOR/MINAP) and Sweden (SWEDEHEART/RIKS-HIA) [8].
National efforts are needed to improve the care and outcomes for
patients with STEMI using national registries, including those in China
[4], the USA [32], and other countries.
Our findings have several important research, policy and clinical
implications. Firstly, theymay help in the interpretation of geographical
variations in outcomes and treatment effect observed in RCTs. Secondly,
they can help focus quality improvement initiatives in individual
countries and regions by targeting specific proven therapies that are
underused. Lastly, although we lacked data on timing from symptom
onset to PCI in 37.8% of STEMI patients, our available data suggest thatdelays to reperfusion are still too frequent. Reducing time to reperfusion
is one of the goals to further improve outcomes, with room for
additional enhancements in healthcare provision that should translate
into clinical benefits.
4.4. Conclusions
Substantial geographical variation exists worldwide with regard to
patient profile, practice patterns and post-discharge prognosis in
patients with STEMI. Therapies with a proven benefit for STEMI are
underused despite strong evidence and guideline recommendations.
Our results may help to guide clinicians, researchers and policymakers
in the commitment to reduce these disparities as well as to close the
gap between evidence-based guideline recommendations for STEMI
and current regional practice patterns.
Acknowledgements
Editorial support was provided by Prime (Knutsford, Cheshire, UK),
funded by AstraZeneca. Dr. J.P.S. Sawhney reviewed the draft and
suggested some references.
Conflicts of interest
H.B. has received advisory/consulting fees fromAbbott, AstraZeneca,
Bayer, Bristol-Myers Squibb, Daiichi-Sankyo, Eli Lilly, Novartis, Pfizer,
Sanofi, and Servier, and grants from AstraZeneca. X.R. has nothing to
disclose. S.P. has received research and statistical consulting honoraria
from AstraZeneca. F.V.W. has received consulting fees and research
grants from Boehringer Ingelheim and Merck, and consulting fees from
Roche, Sanofi-Aventis, AstraZeneca and The Medicines Company. C.T.C.
has received research support from Eli Lilly and honoraria from
Medtronic, and has been a consultant or advisory board member for
AstraZeneca. N.D. has received consulting or speaking fees from Amgen,
AstraZeneca, Bayer, Bristol-Myers Squibb, Boehringer Ingelheim,
GlaxoSmithKline, MSD-Schering Plough, Novartis, Pfizer, Roche,
Sanofi-Aventis, Servier, Takeda, and The Medicines Company. S.W.-L.L.
has nothing to disclose. J.M. and A.V. are employees of AstraZeneca.
Y.H. has nothing to disclose.
Appendix A. Supplementary data
Supplementary data to this article can be found online at http://dx.
doi.org/10.1016/j.ijcard.2017.07.039.
References
[1] Global Burden of Disease Study 2013 Collaborators, Global, regional, and national
incidence, prevalence, and years lived with disability for 301 acute and chronic
diseases and injuries in 188 countries, 1990–2013: a systematic analysis for the
Global Burden of Disease Study 2013, Lancet 386 (2015) 743–800.
[2] R.W. Yeh, S.L. Normand, Y.Wang, C.D. Barr, F. Dominici, Geographic disparities in the
incidence and outcomes of hospitalized myocardial infarction: does a rising tide lift
all boats? Circ. Cardiovasc. Qual. Outcomes 5 (2012) 197–204.
[3] S. Yusuf, S. Reddy, S. Ounpuu, S. Anand, Global burden of cardiovascular diseases:
part II: variations in cardiovascular disease by specific ethnic groups and geographic
regions and prevention strategies, Circulation 104 (2001) 2855–2864.
[4] J. Li, X. Li, Q. Wang, S. Hu, Y. Wang, F.A. Masoudi, et al., ST-segment elevation
myocardial infarction in China from 2001 to 2011 (the China PEACE-Retrospective
Acute Myocardial Infarction Study): a retrospective analysis of hospital data, Lancet
385 (2015) 441–451.
[5] R.P. Giugliano, J. Llevadot, R.G. Wilcox, E.P. Gurfinkel, C.H. McCabe, A. Charlesworth,
et al., Geographic variation in patient and hospital characteristics, management, and
clinical outcomes in ST-elevation myocardial infarction treated with fibrinolysis.
Results from InTIME-II, Eur. Heart J. 22 (2001) 1702–1715.
[6] S.D. Reed, J.J.McMurray, E.J. Velazquez, K.A. Schulman, R.M. Califf, L. Kober, et al., Geo-
graphic variation in the treatment of acute myocardial infarction in the VALsartan In
Acute myocardial iNfarcTion (VALIANT) trial, Am. Heart J. 152 (2006) 500–508.
[7] K.A. Eagle, S.G. Goodman, A. Avezum, A. Budaj, C.M. Sullivan, J. Lopez-Sendon, Prac-
tice variation and missed opportunities for reperfusion in ST-segment-elevation
myocardial infarction: findings from the Global Registry of Acute Coronary Events
(GRACE), Lancet 359 (2002) 373–377.
34 X. Rosselló et al. / International Journal of Cardiology 245 (2017) 27–34[8] S.C. Chung, J. Sundstrom, C.P. Gale, S. James, J. Deanfield, L. Wallentin, et al., Compar-
ison of hospital variation in acute myocardial infarction care and outcome between
Sweden and United Kingdom: population based cohort study using nationwide clin-
ical registries, BMJ 351 (2015) h3913.
[9] G.T. O'Connor, H.B. Quinton, N.D. Traven, L.D. Ramunno, T.A. Dodds, T.A. Marciniak,
et al., Geographic variation in the treatment of acute myocardial infarction: the Co-
operative Cardiovascular Project, JAMA 281 (1999) 627–633.
[10] H. Bueno, P. Sinnaeve, L. Annemans, N. Danchin, M. Licour, J. Medina, et al., Oppor-
tunities for improvement in anti-thrombotic therapy and other strategies for the
management of acute coronary syndromes: insights from EPICOR, an international
study of current practice patterns, Eur. Heart J. Acute Cardiovasc. Care 5 (2016)
3–12.
[11] S. Pocock, H. Bueno, M. Licour, J. Medina, L. Zhang, L. Annemans, et al., Predictors of
one-year mortality at hospital discharge after acute coronary syndromes: a new risk
score from the EPICOR (long-tErm follow uP of antithrombotic management pat-
terns In acute CORonary syndrome patients) study, Eur. Heart J. Acute Cardiovasc.
Care 4 (2015) 509–517.
[12] S. Pocock, Y. Huo, F. Van deWerf, S. Newsome, C.T. Chin, A.M. Vega, et al., Predicting
two-year mortality from discharge after acute coronary syndrome: an
internationally-based risk score, Eur. Heart J. Acute Cardiovasc. Care (2017) In press.
[13] H. Bueno, N. Danchin, M. Tafalla, C. Bernaud, L. Annemans, F. Van de Werf, EPICOR
(long-tErm follow-up of antithrombotic management Patterns In acute CORonary
syndrome patients) study: rationale, design, and baseline characteristics, Am.
Heart J. 165 (2013) 8–14.
[14] Y. Huo, S.W. Lee, J.P. Sawhney, H.S. Kim, R. Krittayaphong, V.T. Nhan, et al., Rationale,
design, and baseline characteristics of the EPICOR Asia study (Long-tErm follow-uP
of antithrombotic management patterns In Acute CORonary Syndrome patients in
Asia), Clin. Cardiol. 38 (2015) 511–519.
[15] E. von Elm, D.G. Altman, M. Egger, S.J. Pocock, P.C. Gotzsche, J.P. Vandenbroucke, The
Strengthening the Reporting of Observational Studies in Epidemiology (STROBE)
statement: guidelines for reporting observational studies, Ann. Intern. Med. 147
(2007) 573–577.
[16] K.A. Fox, S.G. Goodman, F.A. Anderson Jr., C.B. Granger, M. Moscucci, M.D. Flather,
et al., From guidelines to clinical practice: the impact of hospital and geographical
characteristics on temporal trends in the management of acute coronary syn-
dromes. The Global Registry of Acute Coronary Events (GRACE), Eur. Heart J. 24
(2003) 1414–1424.
[17] P.G. Steg, S.K. James, D. Atar, L.P. Badano, C. Blomstrom-Lundqvist, M.A. Borger, et al.,
ESC guidelines for the management of acute myocardial infarction in patients pre-
senting with ST-segment elevation, Eur. Heart J. 33 (2012) 2569–2619.
[18] P.T. O'Gara, F.G. Kushner, D.D. Ascheim, D.E. Casey Jr., M.K. Chung, J.A. de Lemos,
et al., 2013 ACCF/AHA guideline for the management of ST-elevation myocardial in-
farction: a report of the American College of Cardiology Foundation/American Heart
Association Task Force on practice guidelines, J. Am. Coll. Cardiol. 61 (2013)
e78–140.[19] K.G. Laut, C.P. Gale, A.B. Pedersen, K.A. Fox, T.L. Lash, S.D. Kristensen, Persistent geo-
graphical disparities in the use of primary percutaneous coronary intervention in
120 European regions: exploring the variation, EuroIntervention 9 (2013) 469–476.
[20] S.D. Kristensen, K.G. Laut, J. Fajadet, Z. Kaifoszova, P. Kala, C. Di Mario, et al., Reper-
fusion therapy for ST elevation acute myocardial infarction 2010/2011: current sta-
tus in 37 ESC countries, Eur. Heart J. 35 (2014) 1957–1970.
[21] T. Alexander, STEMI in India, A newly developed system towards a practical and eq-
uitable ST-segment elevation myocardial infarction (STEMI) system of care
connecting rural India, Eur. Heart J. 37 (2016) 2449–2453.
[22] A.H. Gershlick, A.P. Banning, A. Myat, F.W. Verheugt, B.J. Gersh, Reperfusion therapy
for STEMI: is there still a role for thrombolysis in the era of primary percutaneous
coronary intervention? Lancet 382 (2013) 624–632.
[23] F. Schiele, C.P. Gale, E. Bonnefoy, F. Capuano, M.J. Claeys, N. Danchin, et al., Quality
indicators for acute myocardial infarction: a position paper of the Acute Cardiovas-
cular Care Association, Eur. Heart J. Acute Cardiovasc. Care 6 (2017) 34–59.
[24] R. Khatib, M. McKee, H. Shannon, C. Chow, S. Rangarajan, K. Teo, et al., Availability
and affordability of cardiovascular disease medicines and their effect on use in
high-income, middle-income, and low-income countries: an analysis of the PURE
study data, Lancet 387 (2016) 61–69.
[25] S.C. Chung, R. Gedeborg, O. Nicholas, S. James, A. Jeppsson, C. Wolfe, et al., Acute
myocardial infarction: a comparison of short-term survival in national outcome reg-
istries in Sweden and the UK, Lancet 383 (2014) 1305–1312.
[26] K.W. Mahaffey, D.M.Wojdyla, K. Carroll, R.C. Becker, R.F. Storey, D.J. Angiolillo, et al.,
Ticagrelor compared with clopidogrel by geographic region in the Platelet Inhibition
and Patient Outcomes (PLATO) trial, Circulation 124 (2011) 544–554.
[27] E.S. Spatz, A.L. Beckman, Y. Wang, N.R. Desai, H.M. Krumholz, Geographic variation
in trends and disparities in acutemyocardial infarction hospitalization andmortality
by income levels, 1999–2013, JAMA Cardiol. 1 (2016) 255–265.
[28] S. Yusuf, S. Islam, C.K. Chow, S. Rangarajan, G. Dagenais, R. Diaz, et al., Use of second-
ary prevention drugs for cardiovascular disease in the community in high-income,
middle-income, and low-income countries (the PURE Study): a prospective epide-
miological survey, Lancet 378 (2011) 1231–1243.
[29] E.M. Bucholz, N.M. Butala, S. Ma, S.T. Normand, H.M. Krumholz, Life expectancy after
myocardial infarction, according to hospital performance, N. Engl. J. Med. 375
(2016) 1332–1342.
[30] STEMI in India, Eur. Heart J. 37 (2016) 2449–2453.
[31] T. Alexander, A.S. Mullasari, G. Joseph, K. Kannan, G. Veerasekar, S.M. Victor, et al., A
system of care for patients with ST-segment elevation myocardial infarction in
India: the Tamil Nadu-ST-Segment Elevation Myocardial Infarction Program, JAMA
Cardiol. 2 (2017) 498–505.
[32] E.D. Peterson, M.T. Roe, J.S. Rumsfeld, R.E. Shaw, R.G. Brindis, G.C. Fonarow, et al., A
call to ACTION (Acute Coronary Treatment and Intervention Outcomes Network): a
national effort to promote timely clinical feedback and support continuous quality
improvement for acute myocardial infarction, Circ. Cardiovasc. Qual. Outcomes 2
(2009) 491–499.