Journal of the American Heart Association ORIGINAL RESEARCH Trends in Short- and Long-Term ST-Segment–Elevation Myocardial Infarction Prognosis Over 3 Decades: A Mediterranean Population-Based ST-Segment–Elevation Myocardial Infarction Registry Cosme García-García , MD, PhD; Teresa Oliveras, MD†; Jordi Serra, PhD; Joan Vila, ScM; Ferran Rueda , MD; German Cediel , PhD; Carlos Labata, MD; Marc Ferrer, MD; Xavier Carrillo, MD, PhD; Irene R. Dégano, MD, PhD; Oriol De Diego , MD; Nabil El Ouaddi, MD; Santiago Montero , MD; Josepa Mauri, MD, PhD; Roberto Elosua , MD, PhD; Josep Lupón , MD, PhD; Antoni Bayes-Genis , MD, PhD; on behalf of the Ruti-STEMI Investigators* BACKGROUND: Coronary artery disease remains a major cause of death despite better outcomes of ST-segment–elevation myocardial infarction (STEMI). We aimed to analyze data from the Ruti-STEMI registry of in-hospital, 28-day, and 1-year events in patients with STEMI over the past 3 decades in Catalonia, Spain, to assess trends in STEMI prognosis. METHODS AND RESULTS: Between February 1989 and December 2017, a total of 7589 patients with STEMI were admitted con- secutively. Patients were grouped into 5 periods: 1989 to 1994 (period 1), 1995 to 1999 (period 2), 2000 to 2004 (period 3), 2005 to 2009 (period 4), and 2010 to 2017 (period 5). We used Cox regression to compare 28-day and 1-year STEMI mortality and in-hospital complication trends across these periods. Mean patient age was 61.6±12.6 years, and 79.3% were men. The 28-day all-cause mortality declined from period 1 to period 5 (10.4% versus 6.0%; P<0.001), with a 40% reduction after multi- variable adjustment (hazard ratio [HR], 0.6; 95% CI, 0.46–0.80; P<0.001). One-year all-cause mortality declined from period 1 to period 5 (11.7% versus 9.0%; P=0.001), with a 24% reduction after multivariable adjustment (HR, 0.76; 95% CI, 0.60–0.98; P=0.036). A significant temporal reduction was observed for in-hospital complications including postinfarct angina (−78%), ventricular tachycardia (−57%), right ventricular dysfunction (−48%), atrioventricular block (−45%), pericarditis (−63%), and free wall rupture (−53%). Primary ventricular fibrillation showed no significant downslope trend. CONCLUSIONS: In-hospital STEMI complications and 28-day and 1-year mortality rates have dropped markedly in the past 30 years. Reducing ischemia-driven primary ventricular fibrillation remains a major challenge. Key Words: prognosis ■ ST-segment–elevation myocardial infarction ■ STEMI complications ■ STEMI mortality Correspondence to: Cosme García-García, MD, PhD, Heart Institute. Hospital Universitari Germans Trias i Pujol, Carretera de Canyet s/n 08916. Badalona (Barcelona), Spain. E-mail: cosmecg7@gmail.com *A complete list of the Ruti-STEMI Investigators can be found in the Appendix at the end of the article. † Dr Teresa Oliveras is currently located at the Department of Medicine, Autonomous University of Barcelona. For Sources of Funding and Disclosures, see page 10. © 2020 The Authors. Published on behalf of the American Heart Association, Inc., by Wiley. This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made. JAHA is available at: www.ahajournals.org/journal/jaha J Am Heart Assoc. 2020;9:e017159. DOI: 10.1161/JAHA.120.0171591 García-García et al Trends in STEMI Prognosis Over 3 Decades century, before percutaneous reperfusion was widely CLINICAL PERSPECTIVE implemented.7,9,10 Some studies have used data for nonconsecutive patients enrolled in cross-sec- What Is New? tional registries with defined inclusion and exclusion • The Ruti-STEMI is a population-based registry criteria.11 that included consecutive patients with ST- Accordingly, the aim of this study using data from segment–elevation myocardial infarction in a the Ruti-STEMI registry was to conduct a popula- Mediterranean cohort over the past 3 decades. tion-based analysis of trends in 28-day and 1-year • Most in-hospital ST-segment–elevation myo- mortality and in-hospital complications over the past cardial infarction complications have decreased 3 decades. Our goal was to provide a contempo- by 50%. rary real-life perspective of STEMI prognosis in a • Twenty-eight-day case fatality declined 40% Mediterranean region. mainly because of the reduction in early acute- phase mortality in both anterior- and inferior-wall ST-segment–elevation myocardial infarction; 1-year all-cause mortality also decreased 25% METHODS from 1989 to 2017, though in the past decade Study Population mortality rates have remained quite stable, The Ruti-STEMI registry is a prospective population- mainly driven by cardiogenic shock. based registry maintained from February 1989 to What Are the Clinical Implications? December 2017 and including all consecutive pa- • Furthermore, there are still few targets to im- tients with STEMI serving a stable and well-defined prove: Ischemia-driven primary ventricular fibril- geographic area of ≈850 000 inhabitants in the lation rates have not substantially changed over northern metro area of Barcelona in Catalonia, Spain time. (Figure 1). During the 30-year period, the healthcare • Some mechanical complications (ventricular physical structure has remained stable, with only 1 septal rupture and papillary muscle rupture) also university hospital with an intensive cardiac care unit remain without changes in the past 3 decades. (ICCU) and 4 community hospitals that refer patients with STEMI to the ICCU (Figure 1). Several organi- zational changes have occurred during the registry period. Until the year 2000, reperfusion therapy was mainly performed with fibrinolysis; between 2000 Nonstandard Abbreviations and Acronyms and 2009, primary percutaneous coronary interven- tion (PCI) was performed only during working hours. ICCU intensive cardiac care unit A major change in June 2009 was the establishment of the Codi IAM STEMI network, intended as a rep- erfusion network that prioritizes primary PCI for all T he first description of myocardial infarction is at- patients with STEMI 24 hours a day, 7 days a week. tributed to James B. Herrick in 1912.1 The century The setup of the Codi IAM network, including the that followed has seen major changes in medicine, territorial organization and available resources, has and myocardial infarction has been at the forefront. been described previously.13,14 Galvez-Montón et al2 divided the history of myocardial Definitions of myocardial infarction and standard of infarction into three eras: an initial era of mainly clinical care were based on current guidelines available during observation and description of pathophysiology (1912– the study life span.15–18 STEMI was defined as ST- 1950s); a second era centered on coronary care units, segment elevation of ≥1 mm in at least 2 contiguous the discovery of new drugs, and the development of leads (in V2–V3 ≥2 mm was required) in any location in surgical alternatives (1960s–1980s); and the most re- the index or qualifying ECG. cent era of interventional cardiology, focused on inva- In our present analysis, patients were stratified sive cardiology and devices (1980s to present). by years of admission into 5 periods: 1989 to 1994 ST-segment–elevation myocardial infarction (STEMI) (period 1), 1995 to 1999 (period 2), 2000 to 2004 management has undergone an impressive transfor- (period 3), 2005 to 2009 (period 4), and 2010 to mation in recent times because of the establishment 2017 (period 5). of regional and national reperfusion networks and use All study procedures were in accordance with of newer evidence-based drugs.3–12 Some studies the ethical standards outlined in the Declaration of used only in-hospital3 or short-term outcomes,5,7,11,12 Helsinki. Patients provided written consent for use while others tracked the evolution of acute coronary of their clinical data for research purposes. The data syndrome outcomes in the last quarter of the 20th that support the findings of this study are available J Am Heart Assoc. 2020;9:e017159. DOI: 10.1161/JAHA.120.0171592 García-García et al Trends in STEMI Prognosis Over 3 Decades Figure 1. Map of the University and community hospitals. Geographical distribution of the Ruti-STEMI population-based registry, in northern Barcelona metro area. from the corresponding author upon reasonable class. The definitions of these complications were request. standardized and remained stable during the study period. Outcomes The aim of the study was to analyze in-hospital STEMI Statistical Analysis complications and trends in 28-day and 1-year case Categorical variables are expressed with frequency fatality over the past 3 decades. The primary end and percentages and continuous variables as points were 28-day and 1-year all-cause mortality. mean±SD. Statistical differences between groups Mortality was curated from patient health records or were compared using the chi-squared and Student by direct phone contact with patients or relatives and t test or analysis of variance including linear trend double-verified by the Catalan and Spanish health sys- analysis. Departures from normality were evaluated tem databases. by normal QQ plots. Multivariate analysis was per- Secondary end points included changes in formed with logistic regression or proportional Cox in-hospital STEMI complications during the 5 stud- regression models (backward conditional stepwise ied periods: angina, reinfarction, primary ventricu- method), with the following covariates: age, sex, car- lar fibrillation (VF) and tachycardia, atrioventricular diovascular risk factors, previous acute myocardial block, atrial fibrillation/flutter, ventricular septum or infarction, peripheral artery disease, Killip–Kimball papillary muscle or free wall rupture, pericarditis, right class, and reperfusion therapies. Assumption of ventricular dysfunction, and maximum Killip–Kimball linearity of continuous variables (logistic regression J Am Heart Assoc. 2020;9:e017159. DOI: 10.1161/JAHA.120.0171593 García-García et al Trends in STEMI Prognosis Over 3 Decades and Cox) and proportionality (Cox) was tested. Trend or cardiogenic shock) showed a 45% reduction from curves were graphically fitted using polynomial re- period 1 to period 5 (15.3% versus 8.5%, respectively; gression, as they provide better fits to the nonlinear P<0.001), with 81.2% of patients in Killip–Kimball I data. Period of admission was treated as a continu- in the period 5 group. Ventricular septal rupture and ous measure for trend testing. Probability values papillary muscle rupture, both infrequent mechani- <0.05 from 2-sided tests were considered to indicate cal complications of STEMI, remained unchanged at statistical significance. All analyses were performed <1% throughout the study period. Of note, primary VF using the software Statistics SPSS 21 (IBM, Armonk, was the only prevalent acute-phase complication that NY) and STATA version 13.0 (StataCorp, College showed no significant downslope trend over the past Station, TX). 3 decades (7.6% in period 1 versus 6.8% in period 5, respectively; P=0.114). The remarkable reductions in STEMI complications RESULTS observed during the study period allowed for a signifi- A total of 7589 consecutive patients with STEMI were cant reduction in ICCU length of stay, from 5.0 days in included in the Ruti-STEMI registry. The mean age was period 1 to 2.4 days in period 5 (P<0.001). Early mor- 61.6±12.6 years, and 79.3% were men. Relative to the tality, defined as occurring in the ICCU, was cut by half year of admission, patients were grouped into the 5 during the study period, from 8.9% to 4.2% (percent- periods defined above: 1989 to 1994, n=1337; 1995 to age change, −4.7% [95% CI, −2.9% to −6.3%]), 11.2% 1999, n=960; 2000 to 2004, n=1059; 2005 to 2009, versus 4.7% for anterior-wall acute myocardial infarc- n=1535; and 2010 to 2017, n=2698. Baseline demo- tion and 8.9% versus 3.8% for inferior-wall infarcts; all graphic and clinical characteristics of each group are P for trend <0.001) (Figure 3). shown in Table 1. Prevalence in dyslipidemia and hy- Short-term 28-day STEMI mortality showed a pro- pertension increased over the 3 decades, but periph- gressive downslope trend over the past 3 decades, eral disease and previous acute coronary syndrome from 10.4% in period 1 to 6.0% in period 5 (P<0.001) showed a significant declining trend. The use of rep- (Table 2; Figure 4A). After multivariable adjustment erfusion therapy increased over time, with a relative by age, sex, hypertension, diabetes mellitus, dyslip- change of 114%, the use of primary PCI tripled between idemia, peripheral disease, previous acute myocar- period 3 and period 5, while fibrinolysis evolved to neg- dial infarction, Killip–Kimball class, and reperfusion ligible levels. Use of coronary angiography at any time therapy, 28-day case-fatality rates declined by 40% during the index admission increased to 97% in 2017. over time (hazard ratio [HR], 0.60; 95% CI, 0.46– The use of evidence-based treatments during the 0.80; P<0.001; Table 3). Figure 5 illustrates 28-day first 48 hours from admission increased gradually mortality by study period stratified according to max- over the 3-decade period. Early use of β-blockers in- imum Killip–Kimball class. Mortality benefits over creased 59%, use of angiotensin-converting enzyme time were significant across all Killip–Kimball strata, inhibitors or angiotensin receptor blockers increased and especially so in patients with Killip–Kimball class 167%, and statins use increased from 1.6% to 88.0%. IV, despite 59% and 50% of all 28-day deaths were Likewise, use of antithrombotic medications during attributable to patients with Killip–Kimball class IV in the first 48 hours of admission changed markedly, periods 4 and 5, respectively. This benefit was mainly with increasing early use of aspirin (6.6% relative in- observed in the early acute phase, during ICCU ad- crease) and clopidogrel, which use was 2.5-fold higher mission (Figure 6). from period 3 to period 5; in period 5, ticagrelor and Long-term 1-year all-cause STEMI mortality also prasugrel were incorporated. Moreover, the use of showed a downslope trend over the past 3 decades, low-molecular-weight heparin and glycoprotein IIb/IIIa from 11.7% in 1989–1994 to 9.0% in 2010–2017 inhibitors peaked in period 4. (P<0.001) (Table 2; Figure 4B). After multivariable adjust- In-hospital STEMI complications were prospectively ment by age, sex, Killip–Kimball class, hypertension, registered during these 3 decades (Table 2), and most diabetes mellitus, dyslipidemia, peripheral disease, declined significantly. Postinfarct angina decreased previous acute myocardial infarction, and reperfusion 78%; sustained ventricular tachycardia 57%; atrio- therapy, 1-year all-cause mortality decreased by 24% ventricular block 45%, free wall rupture 153%, acute- over time (HR, 0.76; 95% CI, 0.60–0.98; P=0.036; phase atrial fibrillation/flutter 27%; pericarditis 63%, Table 3). and right ventricular dysfunction 48%. Trends in most relevant in-hospital complications over the 5 periods are shown in Figure 2. Acute heart failure complicating STEMI also DISCUSSION showed a downslope trend over time. Patients with In this report, we provide outcomes data from a large maximum Killip–Kimball class III-IV (pulmonary edema population-based registry of patients with STEMI J Am Heart Assoc. 2020;9:e017159. DOI: 10.1161/JAHA.120.0171594 García-García et al Trends in STEMI Prognosis Over 3 Decades Table 1. Demographic Characteristics, Medical Therapies, and Management of Patients With STEMI Among the 5 Periods Period 1 Period 2 Period 3 Period 4 Period 5 1989–1994 1995–1999 2000–2004 2005–2009 2010–2017 Characteristics* (N=1337) (N=960) (N=1059) (N=1535) (N=2698) P for Trend Age, y, mean (SD) 61.9 (13.8) 61.8 (12.4) 62.2 (11.8) 62.5 (13.1) 62.1 (12.9) 0.481 Men, % 77.0 77.9 80.4 78.1 76.9 <0.001 Smoker, % 46.2 43.2 38.1 39.8 45.0 <0.001 Hypertension, % 45.8 46.9 49.5 52.3 54.2 <0.001 Dyslipidemia, % 39.5 39.8 50.2 56.6 58.4 <0.001 Diabetes mellitus, % 25.1 27.5 31.1 27.3 24.5 <0.001 Peripheral disease, % 9.9 10.0 11.6 12.7 7.1 0.005 Stroke, % … … … … 2.3 … Kidney disease, % … … … … 4.8 … Previous AMI 19.7 17.6 17.5 15.8 12.0 <0.001 Killip–Kimball class* <0.001 I 73.4 74.7 77.1 79.4 81.2 II 10.8 10.1 9.8 8.8 10.4 III 8.0 8.5 7.6 5.6 2.5 IV 7.3 6.4 5.5 5.8 6.0 AMI location <0.001 Anterior wall 42.6 41.7 46.5 49.3 45.2 Inferior wall 57.4 58.3 53.5 50.3 54.8 Medications Aspirin, % 91.7 92.9 93.2 96.6 97.8 <0.001 Clopidogrel, % … … 25.4 78.6 89.3 <0.001 Ticagrelor, % … … … … 2.5 … Prasugrel, % … … … … 12.4 … GPIIb/IIIa inh. % … 0.1 12.2 38.1 25.1 <0.001 Heparin, % 59.3 64.9 71.9 64.5 71.2 <0.001 Low-molecular-weight 3.8 4.4 30.7 44.0 30.7 <0.001 heparin, % β-Blockers, % 51.0 74.2 82.7 83.7 81.3 <0.001 Statins, % 1.6 0.1 26.3 75.6 88.0 <0.001 ACE inhibitor/ARB, % 20.5 26.0 41.0 54.2 54.8 <0.001 Lidocaine, % 22.9 15.8 7.6 6.0 2.4 <0.001 Amiodarone, % 7.9 8.6 7.5 6.9 4.9 <0.001 Reperfusion, % 42.8 54.7 57.7 78.9 91.9 <0.001 Fibrinolysis, % 100 100 66.7 22.1 0.9 Primary PCI, % 0 0 33.3 77.9 99.1 Rescue PCI, % 0 0 0.7 4.6 0.9 Time onset-reperfusion … … 170 (138) 240 (210) 185 (170) <0.001 min, median (IQR) Coronary angiography 3.1 5.2 27.3 70.6 97 <0.001 CABG, % 1.2 0.2 3.0 2.9 1.1 <0.001 PAC, % 5.1 4.8 3.5 14.6 1.5 <0.001 IABP, % … 0.2 1.7 3.0 2.8 <0.001 LVAD, % … … … … 0.9 … ACE indicates angiotensin-converting enzyme; AMI, acute myocardial infarction; ARB, angiotensin receptor blocker; CABG, coronary artery bypass grafting; GPIIB/IIIa inh, glycoprotein IIb/IIIa inhibitors; IABP, intra-aortic balloon pump; IQR, interquartile range; LVAD, left ventricular assist device (Impella CP); PAC, pulmonary artery catether; PCI, percutaneous coronary intervention; and STEMI, ST-segment–elevation myocardial infarction. *Maximum Killip–Kimball class. J Am Heart Assoc. 2020;9:e017159. DOI: 10.1161/JAHA.120.0171595 García-García et al Trends in STEMI Prognosis Over 3 Decades Table 2. In-Hospital Prognosis and Mortality Period 1 Period 2 Period 3 Period 4 Period 5 1989–1994 1995–1999 2000–2004 2005–2009 2010–2017 (N=1337) (N=960) (N=1059) (N=1535) (N=2698) P for Trend Angina, % 9.8 10.3 8.9 6.0 2.1 <0.001 Reinfarction, % 1.3 1.5 2.1 1.7 1.7 0.686 Primary VF, % 7.6 6.9 6.9 6.6 6.8 0.114 VT, % 8.7 7.7 4.4 7.8 3.7 <0.001 AV block, % 9.7 12.5 6.2 5.8 5.3 <0.001 AFib/flutter, % 8.4 11.9 8.5 7.6 6.1 <0.001 VS rupture, % 0.7 0.8 0.9 0.6 0.5 0.609 PM rupture, % 0.3 0.5 0.4 0.7 0.3 0.066 FW rupture, % 1.9 2.1 1.4 0.7 0.9 0.002 Pericarditis, % 7.6 3.6 2.8 2.1 2.8 <0.001 RV dysfunction, % 9.6 12.1 6.9 5.1 5.0 <0.001 ACCU LoS, d 5.0 5.5 4.7 3.3 2.4 <0.001 ACCU mortality, % 8.9 8.1 5.8 3.8 4.2 <0.001 Anterior wall AMI 11.2 11.0 6.1 4.3 4.7 <0.001 Inferior wall AMI 8.9 8.1 5.8 3.4 3.8 <0.001 28-d mortality, % 10.4 9.9 7.3 5.1 6.0 <0.001 1-y mortality, % 11.7 13.4 10.5 8.7 9.0 <0.001 ACCU indicates acute cardiovascular care unit; AFib, atrial fibrillation; AV, atrioventricular; FW, free-wall; LoS, length of stay; PM, papillary muscle; RV, right ventricle; VF, ventricular fibrillation; VS, ventricular septum; and VT, sustained ventricular tachycardia. (N=7589) living in a Mediterranean region in Catalonia was associated mostly with patients without STEMI, from 1989 to 2017. In the Ruti-STEMI registry, we without significant changes among those with STEMI identified a 40% reduction in 28-day and 25% reduc- (odds ratio, 0.93; 95% CI, 0.71–1.20). The European tion in 1-year STEMI mortality over the past 3 dec- Heart Survey, including only STEMI cases,10 revealed ades, together with a 60% decrease in early mortality a 30-day 15% decline in case fatalities from 2000 to (during ICCU hospitalization) with decreases in both 2004, with 6.4% of the mortality decreased in the last anterior (−68%) and inferior (−57%) STEMI mortal- period, similar to our data for periods 4 and 5 (cov- ity. Detailed data during the hospitalization period ering 2005–2017). The FAST-MI (French Registry on allowed for a refined characterization of in-hospital Acute ST-Elevation and Non–ST-Elevation Myocardial complications. Most STEMI-driven in-hospital com- Infarction) registries, covering 1995 to 2010,11 showed plications were reduced by 50% or more over the a 30-day 60% reduction in case fatalities after adjust- past 3 decades, except for primary VF and some ment by patient profile and reperfusion. These authors low-prevalence mechanical complications (papillary point to the increase in evidence-based medical ther- muscle and septal rupture), whose incidence re- apies (antithrombotic agents, statins, β-blockers, and mained unaltered (Figure 7). angiotensin-converting enzyme inhibitors) as one fac- The past 3 decades have witnessed a surge of tor in the improved prognosis for patients with STEMI. new treatments for STEMI and new organizational net- We observed a similar increase in evidence-based works to prioritize percutaneous reperfusion based medical therapies in our study. Nevertheless, the most on the premise that “time is myocardium.” Several relevant decline observed in our series also coincides studies, mainly from Northern Europe and the United with the implementation of primary PCI, initially during States, have already yielded data on the short- and normal working hours (2000–2009) and available 24 long-term impacts of these novel strategies,5,7,11,19,20 hours a day, 7 days a week since 2009. although most included a mix of patients with and Data on long-term prognosis in real-life STEMI reg- without STEMI, representing different clinical enti- istries are scarce. In this study, we observed a 25% ties with different management and outcomes. In the reduction in 1-year adjusted STEMI mortality over the Worcester registry,5 in-hospital mortality fell from 9.6% 3 studied decades. In addition to the above-reported to 6.5% from 2001 to 2011, but >64% of their patients benefits of acute reperfusion and implementation of were non-STEMI. Yeh et al7 reported 30-day mortal- newer drugs and devices, we observed a remarkable ity reduction from 10.5% to 7.8% in the US state of decline in post-STEMI heart failure, a major driver of California during 1999 to 2008; in that registry, benefit adverse outcomes in the long term. In the Ruti-STEMI J Am Heart Assoc. 2020;9:e017159. DOI: 10.1161/JAHA.120.0171596 García-García et al Trends in STEMI Prognosis Over 3 Decades Figure 2. Trends in STEMI in-hospital complications among periods. Angina (top left); atrial fibrillation (top middle); pericarditis (top right); right ventricular (RV) dysfunction (middle left); atrioventricular block (middle right); primary ventricular fibrillation (VF; bottom). Shaded regions represents 95% CI. AV indicates atrioventricular; and STEMI, ST-segment–elevation myocardial infarction. registry, we found a significant increase in Killip– period 1 to period 5. Indeed, cardiogenic shock— Kimball I infarcts (without heart failure) and a 45% re- defined as Killip-Kimball class IV—declined by 18% duction in Killip–Kimball class III to IV STEMIs from over 3 decades, similar to the results of the Worcester registry.21 In our population, cardiogenic shock devel- oped in 6% of patients with STEMI in period 5, much like the findings reported in the Swedish STEMI regis- try.22 These data are in agreement with Bayes-Genis et al,23 who previously reported having proportionally fewer patients with heart failure with reduced ejec- tion fraction and more patients with heart failure and midrange ejection fraction since the implementation of the STEMI Codi IAM network in Catalonia. Current evidence supports much better outcomes for patients with reduced versus midrange ejection fraction. Regarding in-hospital complications, our data are in line with previously reported studies.24,25 As an exam- ple, in our last period, high-grade atrioventricular block Figure 3. Trends in changes in 28-day case fatality related appeared in 5.3% of patients STEMI, similar to the to infarct location between periods. Inferior wall AMI (blue), anterior wall AMI (green), all STEMI (red). 6% reported in the recent European Heart Survey.24 AMI indicates acute myocardial infarction; and STEMI, ST- The widespread use of primary PCI also underlies the segment–elevation myocardial infarction. reduction in some mechanical complications, mainly J Am Heart Assoc. 2020;9:e017159. DOI: 10.1161/JAHA.120.0171597 García-García et al Trends in STEMI Prognosis Over 3 Decades Figure 4. Kaplan–Meier curves among periods. A, 28-day case fatalities and (B) 1-year all-cause mortality among patients with STEMI during the 5 periods: period 1 (dark blue), period 2 (brown), period 3 (green), period 4 (orange) and period 5 (light blue). STEMI indicates ST-segment–elevation myocardial infarction. free-wall rupture.26 Of note, in our series, we found in other population cohorts.28 The widespread use of no reduction in ventricular septal rupture or papillary reperfusion therapies has probably reduced in-hospital muscle rupture, in contrast with findings in a recent primary VF, but a recent report indicated that 75% of study of elderly acute myocardial infarction patients.26 all primary VFs occurred out of hospital.29 The STEMI Ventricular septal rupture still developed in 0.5% of pa- reperfusion network prioritizes emergent attention to tients, similar to the 0.6% reported in a recent Swiss cardiac patients with a prompt transfer, which allows registry.27 more patients to arrive alive at the hospital. Patients There are reasons for optimism but not for com- have more probabilities to arrive alive at the hospital placency in the reported trends. There are (at least) after an out-of-hospital cardiac arrest attributable to 2 issues of concern. On one hand, no reduction in PVF after the onset of primary PCI reperfusion net- mortality was achieved in the past decade, very likely works. In the clinical treatment of patients with STEMI, attributable to acute cardiogenic shock mortality. the current standard is based on the premise that Indeed, cardiogenic shock is the challenge of the pres- primary VF is benign in terms of long-term prognosis ent decade if we wish to improve our performance in if the patient survives to discharge. However, recent acute STEMI. On the other hand, a major finding of registry data challenge this impression, suggesting the Ruti-STEMI registry is the lack of reduction in isch- that primary VF is an independent predictor of short- emia-driven primary VF over the past 3 decades. In and long-term mortality.29 In sum, our data emphasize our registry, primary VF occurred in 7.6% of patients the need to seek novel out-of-the-box strategies for in period 1 and in 6.8% in period 5, without statistical preventing and managing primary VF. Ischemia could differences over time. Similar trends in primary VF and be speculated to be a second hit for primary VF in a malignant arrhythmias have been reported previously myocardium with existing susceptibility from genetic or Table 3. Multivariable Cox Regression Analyses for 28-Day Case Fatality and 1-Year All-Cause Mortality 28-d Case-Fatality 1-y All-Cause Mortality Model 1 Model 2 Model 1 Model 2 HR (95% CI) P Value HR (95% CI) P Value HR (95% CI) P Value HR (95% CI) P Value 1989–1994 1 … 1 … 1 … 1 … 1995–1999 0.92 (0.70–1.21) 0.56 1.08 (0.81–1.43) 0.602 1.07 (0.83–1.36) 0.612 1.21 (0.93–1.57) 0.150 2000–2004 0.68 (0.51–0.91) 0.009 0.81 (0.60–1.09) 0.168 0.82 (0.64–1.06) 0.132 0.89 (0.68–1.17) 0.419 2005–2009 0.46 (0.35–0.61) <0.001 0.51 (0.37–0.69) <0.001 0.66 (0.52–0.84) 0.001 0.70 (0.54–0.92) 0.010 2010–2017 0.55 (0.43–0.69) <0.001 0.61 (0.46–0.80) <0.001 0.69 (0.56–0.86) 0.001 0.76 (0.60–0.98) 0.036 Model 1: Crude 28-day case fatality or 1-year all-cause mortality by period; Model 2: Including as covariates age, sex (women), hypertension, diabetes mellitus, dyslipidemia, peripheral disease, previous AMI, Killip–Kimball class, interaction between age-Killip and reperfusion therapy. AMI indicates acute myocardial infarction; and HR, hazard ratio. J Am Heart Assoc. 2020;9:e017159. DOI: 10.1161/JAHA.120.0171598 García-García et al Trends in STEMI Prognosis Over 3 Decades Figure 5. Twenty-eight-day case fatality among periods depending on maximum Killip class during hospital admission. A, Killip I; B, Killip II; C, Killip III; D, Killip IV. Period 1 (dark blue), period 2 (brown), period 3 (green), period 4 (orange) and period 5 (light blue). inflammatory determinants of arrhythmia, offering po- Study Limitations tential targets for risk stratification. We acknowledge the lack of valid prospective all- comers echocardiography data from the start of the registry, precluding us from providing individual pa- tient ejection fraction. This lack is a major limitation of most, if not all, real-life registries. Second, a historical cohort study covering 3 decades implies changes in some definitions of cardiovascular risk factors (diabe- tes mellitus, hypertension, and hypercholesterolemia), which may have affected their prevalence across pe- riods. Moreover, a better knowledge of coronary risk factors may have also affected treatment, that is, differ- ent management of hypertension or dyslipidemia over time. The definition of myocardial infarction has also changed over time, as have the biomarkers used for myocyte injury. Nevertheless, the essential diagnostic criteria for STEMI remained the presence of symptoms and compatible ECG findings. Injury biomarkers are Figure 6. Early acute-phase mortality relative to maximum not needed to initiate treatment in STEMI. In patients Killip–Kimball class during intensive cardiac care unit admission. Period 1 (dark blue), period 2 (red), period 3 (green), period 4 undergoing primary PCI for STEMI, the type of stent (violet) and period 5 (light blue). (bare metal or drug-eluting stent) is not available in this J Am Heart Assoc. 2020;9:e017159. DOI: 10.1161/JAHA.120.0171599 García-García et al Trends in STEMI Prognosis Over 3 Decades Figure 7. Trends in STEMI outcomes among 3 decades in a Mediterranean population-based registry. Most in-hospital STEMI (ST-segment–elevation myocardial infarction) complications have been reduced above 50%: angina, right ventricular dysfunction, pericarditis, atrioventricular block, ventricular tachycardia (VT), heart failure, and free-wall (FW) rupture, although primary ventricular fibrillation (VF) remains without changes. Twenty-eight-day case fatality declined 40% and 1-year mortality has been reduced 25% among the past 3 decades. K-K indicates Killip–Kimball. registry. Finally, although follow-up data were double- ARTICLE INFORMATION verified through the Catalan and Spanish health sys- Received April 22, 2020; accepted August 31, 2020. tem databases, we cannot exclude potential sources of error in the early stages. If this were the case, the Affiliations number of events in early stages would be even greater From the Heart Institute, Hospital Universitari Germans Trias i Pujol, Badalona, Spain (C.G.-G., T.O., J.S., F.R., G.C., C.L., M.F., X.C., O.D.D., and the reduction over time still higher. N.E.O., S.M., J.M., J.L., A.B.-G.); CIBER Enfermedades Cardiovasculares (CIBERCV), Badalona, Spain (C.G.-G., X.C., I.R.D., R.E., J.L., A.B.-G.); Grup d’Epidemiologia i Genètica Cardiovasculars (EGEC), REGICOR Study Group, CONCLUSIONS IMIM (Institut Hospital del Mar d’Investigacions Mèdiques), Barcelona, Spain (J.V., I.R.D., R.E.); Catalan Health Service, Generalitat de Catalunya, The results of the prospective all-comers Ruti- Barcelona, Spain (J.M.); and Department of Medicine, Autonomous University of Barcelona, Barcelona, Spain (J.L., A.B.-G.). STEMI registry over 3 decades in a well-defined Mediterranean area indicate that 28-day STEMI case Acknowledgments fatality declined by 40%, and 1-year all-cause mor- The authors thank Drs Antoni Curós and Jordi Serra for their vision and per- tality declined by 24% from 1989 to 2017. In-hospital severance in data collection and their invaluable clinical work during these 3 decades. The authors also thank Susanna Tello, Marta Cabañero, and complications also decreased by ≥50%, although Leny Franco (from Institut Hospital del Mar d’Investigacions Mèdiques) and primary VF and some mechanical complications Sandra Rios and Pilar Gomariz (from Germans Trias Hospital) for their roles in showed stable rates over time. Generalization of data management and performing database maintenance. We also express special gratitude to all doctors, residents, and nurses who participated in the reperfusion therapy, with the widespread use of pri- Ruti-STEMI registry over the past 3 decades. mary PCI more recently, and the use of evidence- based medical therapies have been the main drivers Sources of Funding None. underlying the progressive decline in most STEMI complications. Tackling ischemia-driven primary VF Disclosures requires imagination and divergent thinking because None. everything that has been done so far has been es- sentially futile. REFERENCES 1. Herrick JB. Clinical features of sudden obstruction of the coronary ar- APPENDIX teries. JAMA. 1912;250:1757–1765. 2. Gálvez-Montón C, Ordoñez-Llanos J, de Luna AB, Bayes-Genis The RUTI-STEMI Investigators: Maria José Martínez- A. One hundred years of myocardial infarction. Eur Heart J. Membrive, Elisabeth Zamora, Angel Caballero, Cinta 2012;33:2888–2891. 3. Goldberg RJ, Makam RC, Yarzebski J, McManus DD, Lessard D, Gore Llibre, Oriol Rodríguez-Leor, Eduard Fernandez- JM. Decade long trends (2001–2011) in the incidence and hospital death Nofrerias, Omar Abdul-Jawad, Victoria Vilalta. rates associated with the in-hospital development of cardiogenic shock J Am Heart Assoc. 2020;9:e017159. DOI: 10.1161/JAHA.120.01715910 García-García et al Trends in STEMI Prognosis Over 3 Decades after acute myocardial infarction. Circ Cardiovasc Qual Outcomes. F, et al. ESC Guidelines for the management of acute myocardial in- 2016;9:117–125. farction in patients presenting with ST-segment elevation: the task 4. Velagapudi P, Kolte D, Ather K, Khera S, Gupta T, Gordon PC, Aronow force on the management of ST-segment elevation acute myocar- HD, Kirtane AJ, Abbott JD. Temporal trends and factors associated dial infarction of the European Society of Cardiology. Eur Heart J. with prolonged length of stay in patients with ST-elevation myocardial 2012;33:2569–2619. infarction undergoing primary percutaneous coronary intervention. Am 18. Ibañez B, James S, Agewall S, Antunes MJ, Bucciarelli-Ducci C, Bueno J Cardiol. 2018;122:185–191. H, Caforio ALP, Crea F, Goudevenos JA, Halvorsen S, et al; ESC 5. Goldberg RJ, Tisminetzky M, Tran HV, Yarzebski J, Lessard D, Gore JM. Scientific Document Group. 2017 ESC guidelines for the management Decade long trends (2001–2011) in the incidence rates of initial acute of acute myocardial infarction in patients presenting with ST-segment myocardial infarction. Am J Cardiol. 2019;123:206–211. elevation: the Task Force for the management of acute myocardial 6. Ford ES, Ajani UA, Croft JB, Critchley JA, Labarthe DR, Kottke TE, Giles infarction in patients presenting with ST-segment elevation of the WH, Capewell S. Explaining the decrease in U.S. deaths from coronary European Society of Cardiology (ESC). Eur Heart J. 2018;39:119–177. disease, 1980–2000. N Engl J Med. 2007;356:2388–2398. 19. Sala C, Grau M, Masia R, Vila J, Subirana I, Ramos R, Aboal J, Sureda 7. Yeh RW, Sidney S, Chandra M, Sorel M, Selby JV, Go AS. Population A, Brugada R, Marrugat J, et al. Trends in Q-wave acute myocardial in- trends in the incidence and outcomes of acute myocardial infarction. N farction case fatality from 1978 to 2007 and analisys of the effectiveness Engl J Med. 2010;362:2155–2165. of different treatments. Am Heart J. 2011;162:444–450. 8. Goldberg R, Yarzebski J, Lessard D, Gore JM. A two-decade (1975 20. Rosamond WD, Chambless LE, Heiss G, Mosley TH, Coresh J, Whitsel to 1995) long experience in the incidence, in-hospital and long-term E, Wagenknecht L, Ni H, Folsom AR. Twenty-two year trends in inci- case-fatality rates of acute myocardial infarction: a community-wide dence of myocardial infarction, CHD mortality, and case-fatality in four perspective. J Am Coll Cardiol. 1999;33:1533–1539. US communities, 1987 to 2008. Circulation. 2012;125:1848–1857. 9. Gil M, Marrugat J, Sala J, Gil M, Marrugat J, Sala J, Masiá R, Elosua R, 21. Goldberg RJ, Spencer FA, Gore JM, Lessard D, Yarzebski J. Thirty-year Albert X, Pena A, et al. Relationship of therapeutic improvements and trends (1975 to 2005) in the magnitude of, management of, and hospital 28-day case fatality in patients hospitalized with acute myocardial in- death rates associated with cardiogenic shock in patients with acute farction between 1978 and 1993 in the REGICOR study, Gerona, Spain. myocardial infarction: a population-based perspective. Circulation. Circulation. 1999;99:1767–1773. 2009;119:1211–1219. 10. Mandelzweig L, Battler A, Boyko V, Bueno H, Danchin N, Filippatos G, Gitt 22. Redfors B, Angerås O, Råmunddal T, Dworeck C, Haraldsson I, Ioanes A, Hasdai D, Hasin Y, Marrugat J, et al.; Euro Heart Survey Investigators. D, Petursson P, Libungan B, Odenstedt J, Stewart J, et al. 17-year The second Euro Heart Survey on acute coronary syndromes: charac- trends in incidence and prognosis of cardiogenic shock in patients teristics, treatment, and outcome of patients with ACS in Europe and the with acute myocardial infarction in western Sweden. Int J Cardiol. Mediterranean Basin in 2004. Eur Heart J. 2006;27:2285–2293. 2015;185:256–262. 11. Puymirat E, Simon T, Steg PG, Schiele F, Gueret P, Blanchard D, Khalife 23. Bayes-Genis A, García-Garcia C, de Antonio M, Fernandez-Nofrerías E, K, Goldstein P, Cattan S, Vaur L, et al; USIK USIC 2000 Investigators; Domingo M, Zamora E, Moliner P, Lupón J. Impact of a “stent for life” FAST MI Investigators. Association of changes in clinical characteristics initiative on post-ST elevation myocardial infarction heart failure: a 15 and management with improvement in survival among patients with ST- year heart failure clinic experience. ESC Heart Fail. 2018;5:101–105. elevation myocardial infarction. JAMA. 2012;308:998–1006. 24. Singh SM, FitzGerald G, Yan AT, Brieger D, Fox KA, López-Sendón J, 12. Chung SC, Gedeborg R, Nicholas O, James S, Jeppsson A, Wolfe C, Yan RT, Eagle KA, Steg PG, Budaj A, et al. High grade atrioventricular Heuschmann P, Wallentin L, Deanfield J, Timmis A, et al. Acute myocar- block in acute coronary syndromes: insights from the Global Registry of dial infarction: a comparison of short-term survival in national outcome Acute Coronary Events. Eur Heart J. 2015;36:976–983. registries in Sweden and the UK. Lancet. 2014;383:1305–1312. 25. Kundu A, O’Day K, Shaikh AY, Lessard D, Saczynski JS, Yarzebski J, 13. Bosch X, Curós A, Argimon JM, Faixedas M, Figueras J, Jiménez-Fàbrega Darling C, Thabet R, Akhter M, Floyd KC, et al. Relation of atrial fibrilla- FX, Masià R, Mauri J, Tresserras R. Model of primary percutaneous inter- tion in acute myocardial infarction to in-hospital complications and early vention in Catalonia. Rev Esp Cardiol. 2011;11(suppl C):51–60. hospital readmission. Am J Cardiol. 2016;117:1213–1218. 14. Carrillo X, Fernandez-Nofrerias E, Rodriguez-Leor O, Oliveras T, Serra 26. Puerto E, Viana-Tejedor A, Martínez-Sellés M, Domínguez-Pérez L, J, Mauri J, Curos A, Rueda F, García-García C, Tresserras R, et al; Moreno G, Martín-Asenjo R, Bueno H. Temporal trends in mechanical Codi IAM Investigators. Early ST elevation myocardial infarction in complications of acute myocardial infarction in the elderly. J Am Coll non-capable percutaneous coronary intervention centres: in situ fibri- Cardiol. 2018;72:959–966. nolysis vs. percutaneous coronary intervention transfer. Eur Heart J. 27. Lanz J, Wyss D, Räber L, Stortecky S, Hunziker L, Blöchlinger 2016;37:1034–1040. S, Reineke D, Englberger L, Zanchin T, Valgimigli M, et al. 15. Gunnar RM, Bourdillon PD, Dixon DW, Fuster V, Karp RB, Kennedy JW, Mechanical complications in patients with ST-segment elevation Klocke FK, Passamani ER, Pitt B, Rapaport E, et al. ACC/AHA guide- myocardial infarction: a single centre experience. PLoS One. lines for the early management of patients with acute myocardial infarc- 2019;14:e0209502. tion. A report of the American College of Cardiology/American Heart 28. Thompson CA, Yarzebski J, Goldberg RJ, Lessard D, Gore JM, Dalen Association Task Force on assessment of diagnostic and therapeutic JE. Changes over time in the incidence and case-fatality rates of pri- cardiovascular procedures. Circulation. 1990;82:664–707. mary ventricular fibrillation complicating acute myocardial infarction: 16. Van de Werf F, Ardissino D, Betriu A, Cokkinos DV, Falk E, Fox KA, perspectives from the Worcester Heart Attack Study. Am Heart J. Julian D, Lengyel M, Neumann FJ, Ruzyllo W, et al. Management of 2000;139:1014–1022. acute myocardial infarction in patients presenting with ST-segment ele- 29. García-García C, Oliveras T, Rueda F, Pérez-Fernández S, Ferrer M, vation. Task Force on the management of acute myocardial infarction of Serra J, Labata C, Vila J, Carrillo X, Rodríguez-Leor O, et al. Primary the European Society of Cardiology. Eur Heart J. 2003;24:28–66. ventricular fibrillation in the primary percutaneous coronary intervention 17. Steg PG, James SK, Atar D, Badano LP, Blömstrom-Lundqvist C, ST-segment elevation myocardial infarction era (from the “Codi IAM” Borger MA, Di Mario C, Dickstein K, Ducrocq G, Fernandez-Aviles Multicenter Registry). Am J Cardiol. 2018;122:529–536. J Am Heart Assoc. 2020;9:e017159. DOI: 10.1161/JAHA.120.01715911