Author + information
- Received January 24, 2019
- Revision received June 21, 2019
- Accepted June 27, 2019
- Published online September 9, 2019.
- Héctor Bueno, MD, PhDa,b,c,∗ (, )@CNIC_CARDIO,
- Xavier Rossello, MD, PhDa,d,e@RosselloXavier,
- Stuart J. Pocock, PhDd,
- Frans Van de Werf, MD, PhDf,
- Chee Tang Chin, MDg,
- Nicolas Danchin, MD, PhDh,
- Stephen W.-L. Lee, MD, PhDi,
- Jesús Medina, PhDj and
- Yong Huo, MD, PhDk
- aCentro Nacional de Investigaciones Cardiovasculares (CNIC), Madrid, Spain
- bInstituto de Investigación i+12 and Cardiology Department, Hospital Universitario 12 de Octubre, Madrid, Spain
- cFacultad de Medicina, Universidad Complutense de Madrid, Madrid, Spain
- dLondon School of Hygiene and Tropical Medicine, London, United Kingdom
- eCIBER de Enfermedades CardioVasculares, Madrid, Spain
- fDepartment of Cardiovascular Sciences, University of Leuven, Leuven, Belgium
- gNational Heart Centre Singapore, Singapore
- hHôpital Européen Georges Pompidou & René Descartes University, Paris, France
- iQueen Mary Hospital, Hong Kong SAR, China
- jGlobal Medical Affairs, AstraZeneca, Madrid, Spain
- kBeijing University First Hospital, Beijing, China
- ↵∗Address for correspondence:
Dr. Héctor Bueno, Centro Nacional de Investigaciones Cardiovasculares (CNIC), Melchor Fernandez Almagro, 3, 28029-Madrid, Spain.
Background The relationship between in-hospital coronary revascularization rate (CRR) and post-discharge mortality rates in survivors of non–ST-segment elevation acute coronary syndrome (NSTE-ACS) at a system level is unclear.
Objectives The purpose of this study was to evaluate CRR and 2-year post-discharge mortality rate (2YMR) in NSTE-ACS.
Methods CRR and 2YMR were analyzed by hospital rate of CRR (in deciles), by country, and by world region in 11,931 patients with NSTE-ACS who survived to discharge and were enrolled in the EPICOR (long-tErm follow uP of antithrombotic management patterns In acute CORonary syndrome patients) and EPICOR Asia: twin multinational, observational, prospective cohort studies.
Results Significant differences in patient baseline characteristics, medical therapies, CRR, and 2YMR were found. Mean CRR ranged from 0.0% to 96.8% in the first and tenth decile, respectively (p < 0.001); from 12.3% in Romania to 92.4% in Slovenia (p < 0.001); and from 53.9% in South East Asia (SEAsia) to 90.4% in South Korea–Singapore–Hong Kong. 2YMR varied significantly between hospital deciles of CRR (3.6% in tenth decile vs. 9.2% in first decile; p < 0.001), countries (lowest 1.5% in Slovenia, highest 19.4% in Malaysia; p < 0.001), and regions (lowest 3.8% in South Korea–Singapore–Hong Kong, highest 11.7% in SEAsia; p < 0.001). Poisson regression models, adjusted for 15 mortality predictors, showed a significant inverse association between CRR and 2YMR for hospitals (r = −0.90; p < 0.001), countries (r = −0.65; p < 0.001), and regions (r = −0.87; p = 0.005).
Conclusions Higher CRRs at the hospital, country, and world region levels are strongly associated with higher post-discharge survival, suggesting CRR as a marker of higher system quality.
Patients with non-ST–segment elevation acute coronary syndrome (NSTE-ACS) are, in general, at high risk of recurrent coronary events and mortality (1). Guidelines recommend an early invasive strategy—that is, coronary angiography and coronary revascularization when feasible—for a majority of these patients, in particular those who are at high or intermediate risk (2,3). Although there has been progress in recent years (4), there is still significant variation between hospitals, countries, and regions in coronary revascularization rate (CRR) for these patients (1,5). Further, the long-term prognostic implications of differences in clinical practice regarding early coronary revascularization for NSTE-ACS patients are largely unknown. The present work was, therefore, designed to assess whether the rate of coronary revascularization performed during index admission in patients with NSTE-ACS, measured at a health care system level (hospital, country, region) is independently associated with post-discharge survival. This represents a post hoc analysis of the EPICOR (long-tErm follow uP of antithrombotic management patterns In acute CORonary syndrome patients) and EPICOR Asia studies.
All patients with a final diagnosis of NSTE-ACS enrolled in 2 twin prospective studies, EPICOR (NCT01171404, 5,625 patients from 555 hospitals in 20 countries across Europe and Latin America between September 2010 and March 2011) and EPICOR Asia (NCT01361386, 6,306 patients from 219 hospitals from 8 Asian countries between June 2011 and May 2012) (6,7), were considered.
National coordinators in each country were responsible for site selection according to the national proportion of patients treated at hospitals with and without invasive cardiac intervention facilities, with the intention of achieving a fair representation of real-life practice at a country level.
NSTE-ACS comprised non–ST-segment elevation myocardial infarction (NSTEMI) (chest pain/discomfort, lack of persistent ST-segment elevation, left bundle branch block or intraventricular conduction disturbances, and elevation of cardiac biomarkers [creatine kinase-isoenzyme MB or troponins] with at least 1 value above the 99th percentile of the upper reference limit) and unstable angina (clinical presentation consistent with rest angina or on minimal exercise, with transient ST-T changes in the electrocardiogram and no significant increase in biomarkers of necrosis, but objective evidence of ischemia by noninvasive imaging, or significant coronary stenosis on angiography). Coronary revascularization was defined as the performance of any kind of percutaneous coronary intervention (PCI) or coronary artery bypass grafting (CABG) performed during index hospitalization for NSTE-ACS (6).
Patient characteristics, pre-hospital care, hospital features, in-hospital care (including medical therapy), invasive and noninvasive interventions, coronary angiography results, and coronary revascularization, were collected. Follow-up was prospectively performed through periodically scheduled, structured telephone calls assessing medical therapy continuation, the occurrence of cardiovascular events, and mortality ≤2 years after discharge (6).
The relationship between CRR during index hospitalization and 2-year post-discharge all-cause mortality rate (2YMR) was assessed at a system level using 3 different approaches: individual hospital level, country level, and region level. The 743 participating hospitals that recruited NSTE-ACS patients were divided according to the decile of in-hospital CRR, and these were correlated with 2YMRs. Across the 28 participating countries, the national CRR were correlated with the national 2YMRs. Finally, 8 geographical regions were defined: Southern Europe (France, Greece, Italy, and Spain), Northern Europe (Belgium, Denmark, Finland, Germany, Luxembourg, the Netherlands, Norway, and the United Kingdom [UK]), Eastern Europe (Poland, Romania, Slovenia, and Turkey), Latin America (Argentina, Brazil, Mexico, and Venezuela), China, India, South East Asia (Malaysia, Vietnam, and Thailand), and South Korea–Singapore–Hong Kong (1,8), and regional CRR were correlated with average regional 2YMRs.
Continuous parameters are mostly presented as mean ± SD and groups. Categorical data are expressed as percentages and compared across deciles, countries, and regions using chi-square tests.
Poisson regression models were used to estimate 2YMR according to the revascularization status, adjusted for 15 known predictors of 2-year mortality in the overall study population. These predictors were previously obtained from a risk scoring system using Cox regression to estimate 2YMR in individual patients with acute coronary syndrome and with well-reported goodness-of-fit, discriminatory power, and internal validation (9). Predictors, ranked by predictive strength, were: age, low left ventricular ejection fraction, elevated serum creatinine at admission, poor EuroQol-5 dimensions score, low hemoglobin at admission, previous cardiac disease, previous chronic obstructive pulmonary disease, elevated blood glucose at admission, prescription of diuretic agents at discharge, male sex, lower educational level, prescription of an aldosterone inhibitor at discharge, low body mass index, in-hospital cardiac complications, and Killip class. After obtaining the patient-level adjusted 2YMR, the data were collapsed to obtain the mean adjusted 2YMR for each level of analysis (i.e., by hospital deciles, countries, and regions). We then plotted them against their corresponding percentage of revascularization, and estimated the Pearson's correlation coefficient, weighted by group size, to determine the strength of the association between 2YMR and CRR. All probability values were 2-sided, and p values of <0.05 were considered statistically significant. All statistical analyses were performed using STATA software, version 13.1 (Stata Corp, College Station, Texas).
The study cohort comprised 11,931 patients recruited in 743 hospitals from 28 countries of 8 world regions. The mean age was 62.8 ± 11.6 years, 70.5% male, with high prevalence of risk factors: 7,530 (63.7%) hypertensive, 4,746 (42.5%) currently smoking, 4,242 (37%) with hypercholesterolemia, and 3,234 (27.4%) with diabetes mellitus; and with high prevalence of prior cardiovascular disease (5,039, 43%). A majority of patients were admitted in Killip class I (6,173, 82.9%), 3,602 (31.9%) presented with hemoglobin <13 mg/dl, 2,815 (24.6%) with creatinine ≥1.2 mg/dl, and 2,135 (20.3%) with blood glucose >160 mg/dl. There were substantial differences in patient baseline profiles and medical management across hospital deciles of CRR (Table 1, Online Table 1), countries (data not shown), and regions (Online Table 2). The in-hospital CRRs were 0% in hospitals belonging to the first decile, 56.8% for hospitals in the fifth decile, and 96.8% in hospitals of the tenth decile (Table 1). Compared with other hospitals, those from the lowest decile of CRR were less frequently tertiary/university centers (26%), and less often had catheterization laboratories (31%) or participated in a 24/7 primary PCI network (70%).
CRRs also differed among countries, ranging from 12.3% in Romania to 92.4% in Slovenia (p < 0.001) (Online Table 3), and regions, ranging from 53.9% in South East Asia to 90.4% in South Korea–Singapore–Hong Kong (p < 0.001). Complete 2-year follow-up was achieved in 10,275 patients (86.1%). The mean follow-up time for the rest of the patients (n = 1,656, 13.9%) was 385 ± 222 days. Two-year post-discharge mortality rates also varied significantly between hospital deciles (3.6% for decile 10, 9.2% for decile 1; p < 0.001), countries (lowest 1.5% in Slovenia, highest 19.4% in Malaysia; p < 0.001), and regions (lowest 3.8% in South Korea–Singapore–Hong Kong, highest 11.7% in South East Asia; p < 0.001).
Using Poisson regression models that adjust for the 15 previously identified independent predictors of post-discharge mortality (Figure 1) (9), we showed a significant difference in the estimated 2YMR between revascularized and nonrevascularized patients: 5.2% (95% confidence interval [CI]: 4.5% to 5.8%) versus 9.5% (95% CI: 8.5% to 10.4%), respectively (p < 0.001). Moreover, a significant correlation exists between average in-hospital CRR and adjusted 2YMR at hospital level (r = −0.90; p < 0.001), country level (r = −0.65; p < 0.001), and region level (r = −0.87, p = 0.005) (Central Illustration).
This large, prospective, international cohort study of patients who were discharged alive from hospital after NSTE-ACS shows that the rate of coronary revascularization performed during index hospitalization is inversely associated with 2-year covariate-adjusted mortality rates measured at a hospital, national, or regional level. Our findings align well with and support guideline recommendations for coronary revascularization if significant coronary artery disease is present in patients presenting with NSTE-ACS (2,3). They also support the suggestion of considering early invasive strategy as a hospital quality indicator for patients with NSTEMI (10).
We, and others, have previously shown that there are marked regional differences in the management and prognosis of patients with ST-segment elevation myocardial infarction (STEMI) (1) and NSTE-ACS (5,8,11–13), which cannot be explained by baseline patient characteristics alone. In this study, we show that important differences in revascularization rates exist not only at a regional but at a country and hospital level, and that these differences are associated with long-term survival at any level of analysis. This consistent association is relevant, as it suggests that patients admitted to facilities offering early coronary revascularization have better long-term outcomes. However, this does not necessarily mean that the mortality benefit is explained only by coronary revascularization itself. Other aspects of care associated with providing early intensive care in these hospitals may also play a role.
A number of other studies, mostly observational, support the hypothesis of a survival benefit associated with coronary revascularization in the early phase of NSTE-ACS. A recent study showed a significant risk-adjusted reduction in 6-month mortality in the United Kingdom during the last decade, following a marked increase in the use of coronary angiography and revascularization (4). We, and others, had previously described that NSTE-ACS patients undergoing an invasive strategy and coronary revascularization, either PCI or CABG, showed improved survival compared with medically treated patients (14–16). A recent analysis of the ARIC (Atherosclerosis Risk in Communities) surveillance study showed that early PCI (<24 h after symptom onset) in hospitalized NSTEMI patients was associated with improved 28-day survival, especially in high-risk patients, but not at 1 year (17). Although it used a wider time frame (1 month) for revascularization, a large Canadian cohort study assessing the relationship between coronary revascularization and 1-year prognosis showed consistent results, with a lower risk of death or cardiovascular hospitalization among those undergoing revascularization (18).
There are, however, alternative explanations to a direct beneficial effect of coronary revascularization. The first is selection bias. It is known that patients undergoing an invasive strategy and receiving coronary revascularization are usually younger, show less comorbidity, and are at lower risk than medically managed patients (14,19,20). Although this is true, the comprehensive adjustment of our model and the consistency of the effect found at any level of CRR (i.e., hospital deciles) suggest that the effect cannot fully be explained by differences in baseline characteristics. One study comparing the differences in management and outcomes of acute myocardial infarction between all U.K. and Swedish hospitals reported that coronary revascularization in patients with NSTEMI was associated with a lower 1-year mortality risk difference between hospitals in the highest and lowest revascularization quarter when adjusted for patient case mix (9%; 95% CI: −1% to 14% in Sweden, and 14%; 95% CI: 8% to 20% in the United Kingdom) (21). A second alternative explanation is that the differences observed in revascularization rates may be a reflection of differences in the quality of care seen from a wider perspective provided by the different health care systems or hospitals. In fact, our study shows significant differences in the use of dual antiplatelet therapy, beta-blockers, or statins across hospitals from different deciles of CRR, as well as between countries and regions. The use of these treatments is a quality indicator for the management of myocardial infarction (10), which has been validated as a prognostic factor as well (10,22). However, although there is some degree of association between quality indicators for the management of myocardial infarction, this association is, at best, modest (22).
At the hospital level, despite the important differences between hospitals within different deciles of CRR (i.e., those in the first decile less frequently tertiary centers, less often with catheterization laboratories, or participating in STEMI networks), we cannot say that hospitals with the lowest CRRs were either spoke or hub hospitals, as there was a large variation in their characteristics, ranging from hospitals without catheterization laboratories to hospitals with a catheterization laboratory and 24/7 PCI programs. In fact, although Northern Europe participated with the lowest proportion of hospitals with on-site catheterization laboratory, they had the highest levels of CRR, indicating a highly efficient referral network. On the contrary, a number of countries, mostly from Asia, participated only with hospitals with a catheterization laboratory facility but showed much lower rates of coronary angiography or revascularization (8). The fact that the hospitals with zero CRRs were the ones enrolling the lowest number of patients (mean 8 patients; 53% recruited ≤5 patients) may in part explain this finding.
A number of limitations should be considered in this study. First, it is important to highlight that a causal relationship between initial coronary revascularization and post-discharge mortality risk cannot be concluded from these findings. Despite the thorough risk adjustment, its proven value as a quality indicator, and the reasoning as a wider marker of system quality, the causal effect of coronary revascularization remains hypothetical. Second, we cannot rule out the possibility that differences in post-discharge mortality may be explained in part by differences in hospital mortality (i.e., initial survival of higher-risk patients may translate into higher post-discharge mortality). Also, mortality related to elective post-discharge revascularization procedures is not considered, which may be particularly relevant for CABG. However, interestingly, the regions with highest baseline risk did not show the highest post-discharge mortality rate. Third, the lack of information on the cause of death precludes the inference of a potential mechanism for a survival advantage. Fourth, while both registries were designed to recruit representative patients from representative centers in each country, we cannot rule out a bias in regard to the type of hospitals recruiting patients. Hence, some caution is needed to extrapolate our findings regarding geographic differences, because these centers might not be fully representative of the whole country. Finally, despite its wide representation, this is not a global study as important regions are not included. However, it is likely that similar findings would have been found had more regions been included in the study.
Higher average rates of coronary revascularization performed during index hospitalization in patients presenting with NSTE-ACS, measured at a hospital, country, and world region level, are independently associated with lower adjusted 2-year mortality rates.
COMPETENCY IN SYSTEMS-BASED PRACTICE: In patients with NSTE-ACS, higher rates of coronary revascularization are associated with lower mortality after hospital discharge.
TRANSLATIONAL OUTLOOK: Future studies should evaluate the rate of in-hospital coronary revascularization as a quality measure for management of patients with NSTE-ACS.
Editorial assistance was provided by Carl V. Felton, PhD, Paragon (Knutsford, Cheshire, United Kingdom), funded by AstraZeneca.
The EPICOR and EPICOR Asia studies were funded by AstraZeneca. The CNIC is supported by the Spanish Ministry of Economy and Competitiveness (MINECO) and the Pro-CNIC Foundation, and is a Severo Ochoa Center of Excellence (MINECO award SEV-2015-0505). Dr. Bueno has received research funding from the Instituto de Salud Carlos III, Spain (PIE16/00021 and PI17/01799), AstraZeneca, Bristol-Myers Squibb, Janssen, and Novartis; has received consulting fees from AstraZeneca, Bayer, Bristol-Myers Squibb-Pfizer, and Novartis; and has received speaker fees or support for attending scientific meetings from AstraZeneca, Bayer, Bristol-Myers Squibb-Pfizer, Ferrer, Novartis, and MEDSCAPE-the heart.org. Dr. Rossello has received support from the SEC-CNIC CARDIOJOVEN fellowship program. Dr. Pocock has received research funding from AstraZeneca. Dr. Van de Werf has received consulting fees and research grants from Boehringer Ingelheim, Merck, and Sanofi; and has received consulting and speaker fees from Boehringer Ingelheim, Roche, Sanofi, AstraZeneca, and The Medicines Company. Dr. Chin has received consulting or speaker fees from AstraZeneca, Merck, Servier, and Medtronic. Dr. Danchin has received consulting or speaker fees from Amgen, AstraZeneca, Bayer, Bristol-Myers Squibb, Boehringer Ingelheim, Merck Sharp & Dohme, Novo Nordisk, Pfizer, Sanofi, and Servier. Dr. Medina is an employee of AstraZeneca. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose. Deepak Bhatt, MD, served as Guest Editor-in-Chief for this paper.
Listen to this manuscript's audio summary by Editor-in-Chief Dr. Valentin Fuster on JACC.org.
- Abbreviations and Acronyms
- 2-year post-discharge mortality rate
- coronary artery bypass grafting
- coronary revascularization rate
- non–ST-segment elevation acute coronary syndrome
- non–ST-segment elevation myocardial infarction
- percutaneous coronary intervention
- ST-segment elevation myocardial infarction
- Received January 24, 2019.
- Revision received June 21, 2019.
- Accepted June 27, 2019.
- 2019 American College of Cardiology Foundation
- Rossello X.,
- Huo Y.,
- Pocock S.,
- et al.
- Roffi M.,
- Patrono C.,
- Collet J.P.,
- et al.
- Amsterdam E.A.,
- Wenger N.K.,
- Brindis R.G.,
- et al.
- Hall M.,
- Dondo T.B.,
- Yan A.T.,
- et al.
- Huo Y.,
- Lee S.W.,
- Sawhney J.P.,
- et al.
- Bueno H.,
- Rossello X.,
- Pocock S.,
- et al.
- Pocock S.J.,
- Huo Y.,
- Van de Werf F.,
- et al.
- Schiele F.,
- Gale C.P.,
- Bonnefoy E.,
- et al.
- Bueno H.,
- Pocock S.,
- Medina J.,
- et al.
- Wijeysundera H.C.,
- Sidhu M.S.,
- Bennell M.C.,
- et al.
- Chin C.T.,
- Ong T.K.,
- Krittayaphong R.,
- et al.
- Arora S.,
- Matsushita K.,
- Qamar A.,
- Stacey R.B.,
- Caughey M.C.
- Bainey K.R.,
- Kaul P.,
- Armstrong P.W.,
- et al.
- Yang Q.,
- Wang Y.,
- Liu J.,
- et al.
- Heras M.,
- Bueno H.,
- Bardaji A.,
- Fernandez-Ortiz A.,
- Marti H.,
- Marrugat J.
- Chung S.C.,
- Sundstrom J.,
- Gale C.P.,
- et al.
- Bebb O.,
- Hall M.,
- Fox K.A.A.,
- et al.