- Journal List
- ESC Heart Fail
- v.11(4); 2024 Aug
- PMC11287335
As a library, NLM provides access to scientific literature. Inclusion in an NLM database does not imply endorsem*nt of, or agreement with, the contents by NLM or the National Institutes of Health.
Learn more: PMC Disclaimer | PMC Copyright Notice
ESC Heart Fail. 2024 Aug; 11(4): 2344–2353.
Published online 2024 Apr 29. doi:10.1002/ehf2.14790
PMCID: PMC11287335
PMID: 38685603
Satoshi Takeuchi,1,† Satoshi Honda,2,† Kensaku Nishihira,3 Sunao Kojima,4 Misa Takegami,5,6 Yasuhide Asaumi,2 Mike Saji,7 Jun Yamash*ta,8 Kiyoshi Hibi,9 Jun Takahashi,1 Yasuhiko Sakata,2 Morimasa Takayama,7 Tetsuya Sumiyoshi,7 Hisao Ogawa,10 Kazuo Kimura,9 Satoshi Yasuda,
1 and JAMIR Investigators
Author information Article notes Copyright and License information PMC Disclaimer
Associated Data
- Supplementary Materials
Abstract
Aims
The incidence and prognosis of symptomatic heart failure following acute myocardial infarction (AMI) in the primary percutaneous coronary intervention era have rarely been reported in the literature. This study aimed to (i) determine the incidence of heart failure admission among AMI survivors, (ii) compare 1year outcomes between patients with heart failure admission and those without, and (iii) identify the independent risk factors associated with heart failure admission.
Methods and results
The Japan Acute Myocardial Infarction Registry is a prospective multicentre registry from which data on consecutively enrolled patients with AMI from 50 institutions between 2015 and 2017 were obtained. Among the 3411 patients enrolled, 3226 who survived until discharge were included in this study. The primary endpoint was all‐cause mortality. The secondary endpoints were major adverse cardiovascular events (defined as cardiovascular mortality, non‐fatal myocardial infarction, or non‐fatal cerebral infarction) and major bleeding events corresponding to Bleeding Academic Research Consortium Type 3 or 5. Clinical outcomes were compared between the patients who were and were not admitted for heart failure. Over a median follow‐up of 12months, 124 patients (3.8%) were admitted due to heart failure. Independent risk factors for heart failure admission included older age, female sex, Killip class ≥2 on admission, left ventricular ejection fraction <40%, estimated glomerular filtration rate ≤30mL/min/1.73m2, a history of malignancy, and non‐use of angiotensin‐converting enzyme inhibitors at discharge. The cumulative incidence of all‐cause mortality was significantly higher in the heart failure admission group than in the no heart failure admission group (11.3% vs. 2.5%, P<0.001). The rates of major adverse cardiovascular events (16.9% vs. 2.7%, P<0.001) and major bleeding (6.5% vs. 1.6%, P<0.001) were significantly higher in the heart failure admission group. Heart failure admission was associated with a higher risk of all‐cause mortality, even after adjusting for potential confounders (adjusted hazard ratio: 2.41, 95% confidence interval: 1.33–4.39, P=0.004).
Conclusions
Utilizing real‐world data of the contemporary percutaneous coronary intervention era from the Japan Acute Myocardial Infarction Registry database, this study demonstrates that the heart failure admission of AMI survivors was significantly associated with higher all‐cause mortality rates.
Keywords: Acute myocardial infarction, Heart failure, Percutaneous coronary intervention, Registry
Introduction
The in‐hospital mortality rate for acute myocardial infarction (AMI) has decreased in recent years owing to advancements in primary percutaneous coronary intervention (PCI) and improvements in pre‐hospital and post‐hospital care strategies.1, 2 However, as patients with AMI successfully recover from the acute phase, they remain at risk of experiencing subsequent adverse events during the chronic phase, and heart failure (HF) emerges as a significant concern.
The development of HF following hospitalization for myocardial infarction (MI) arises from cardiomyocyte loss and scar formation, which initiates a cascade of chronic neurohumoral activation, including up‐regulation of the renin–angiotensin–aldosterone and sympathetic nervous systems, ultimately leading to ventricular remodelling.3, 4 Ischaemic heart disease is the leading cause of HF, and its increasing prevalence, according to data from the Japanese cohort registry of chronic HF, is a key concern.5 Yet studies addressing post‐AMI HF in the era of primary PCI remain scarce.
In the early days of the primary PCI era in Japan, a registry‐based study on post‐discharge HF revealed a 4.4% HF hospitalization rate within the first year and ~1.0% per year thereafter.6 Nonetheless, over the past decade, the landscape of AMI management underwent significant transformations.7 For instance, newer generation drug‐eluting stents (DESs) replaced bare metal stents and first‐generation DES. The diagnostic criteria have been transitioned from the WHO‐MONICA criteria to a universal definition. Drug therapies, including antiplatelet and lipid‐lowering agents, have also evolved significantly. Hence, it is imperative to investigate the incidence of HF following AMI and its implications for prognosis using contemporary data aligned with the more recent standards in AMI management.
This study therefore aimed to (i) determine the incidence of HF admission among AMI survivors, (ii) compare 1year outcomes between patients with HF admission and those without, and (iii) identify the independent risk factors associated with HF admission using a comprehensive and more recent AMI database from Japan.
Methods
Study population and data source
This study included patients with AMI who survived until discharge and were enrolled in the Japan Acute Myocardial Infarction Registry (JAMIR). The JAMIR is a nationwide, multicentre, and prospective registry designed to enrol patients with AMI in Japan.8, 9 Between December 2015 and May 2017, consecutive patients from 50 participating institutions throughout Japan presenting with spontaneous‐onset AMI were recruited. For this study, AMI was diagnosed on the basis of the universal definition, with allowance of the MONICA criteria according to the institutional setting. All institutes except one used the universal definition for the diagnosis of AMI, and 99% (3371/3411 patients) of study patients were diagnosed by the universal definition. We excluded patients who were admitted to the hospital ≥24h after onset, with no return of spontaneous circulation on admission after out‐of‐hospital cardiopulmonary arrest, or who had AMI as a complication of PCI or coronary artery bypass grafting (CABG). Patient management, including the selection and adjustment of antiplatelet medications, was at the discretion of the attending physicians. A follow‐up study of patients was performed a year after the discharge from index AMI hospitalization based on the medical information available at each study site. The HF admission data during the follow‐up period were collected, and the patients were divided into two groups: patients with HF admission and patients with no HF admission, to evaluate the impact of the HF event on a sequential mortality event. The HF admission was determined by the judgement of the attending physician at each institute in accordance with the HF guidelines.10
This study was conducted in adherence to the ethical guidelines for medical research involving humans as outlined in the Declaration of Helsinki. The research protocol was approved by the Institutional Review Board of the National Cerebral and Cardiovascular Centre (M26‐150‐5) and the local ethics committees or institutional review boards at each study site. Given the observational nature of this registry, informed consent was waived; however, information about the study was made available on a website and at the study sites using an opt‐out approach. This study was also registered in the Japanese UMIN Clinical Trials Registry (UMIN000019479).
Study endpoints
We collected clinical events that occurred from discharge of the index AMI hospitalization until the end of the follow‐up period. The primary endpoint was the occurrence of all‐cause mortality. The secondary endpoints were major adverse cardiovascular events (MACEs) (comprising cardiovascular mortality, non‐fatal MI, and non‐fatal cerebral infarction) and major bleeding events as defined by the Bleeding Academic Research Consortium (BARC) criteria, specifically BARC Type 3 or 5.11 Clinical events were adjudicated by study investigators at each site.
Statistical analysis
Continuous variables were presented as the means±standard deviations (SDs) or the medians with inter‐quartile ranges (IQRs), depending on the data distribution. Categorical variables were expressed as counts and percentages. Continuous variables were compared using the t‐test and Mann–Whitney U test, while dichotomous variables were compared using the χ2 test. For the analysis of HF admission, both univariate and multivariate logistic regression models were utilized to calculate odds ratios (ORs) along with 95% confidence intervals (CIs). Through multivariate analysis, clinically relevant factors were selected as covariates and analysed by the stepwise method. The following covariates were used as stepwise input variables: age ≥75years, female, body mass index (BMI) <25kg/m2, hypertension, diabetes, smoking history, history of coronary artery disease, malignancy, atrial fibrillation, stroke, peripheral artery disease, haemoglobin <11g/dL, estimated glomerular filtration rate (eGFR) <30mL/min/1.73m2, STEMI, Killip Class 2 or higher, anterior wall infarction, multivessel disease, final thrombolysis in myocardial infarction 3 (TIMI3) flow, peak creatine kinase (CK) ≥3000IU/L, left ventricular ejection fraction (LVEF) <40%, major bleeding event during AMI hospitalization, angiotensin‐converting enzyme inhibitor (ACEI) usage, usage of angiotensin receptor blocker (ARB), and use of beta‐receptor inhibitor. The cumulative incidence rates of post‐discharge outcomes in patients experiencing HF admission after their index hospitalization were estimated using the Kaplan–Meier method, and the differences were assessed using the log‐rank test. To evaluate the impact on late mortality events, we conducted a landmark analysis 90days after discharge based on the HF admission within 90days.
Hazard ratios (HRs) and 95% CIs for all‐cause mortality and out‐of‐hospital bleeding events were computed using both univariate and multivariate Cox proportional hazard models. The covariates in these multivariate models were selected based on their clinical significance. The multivariable analysis for all‐cause mortality included covariates such as age (≥75years), sex, Killip Class 2 or higher, eGFR<30mL/min/1.73m2, peak CK levels ≥3000IU/L, and a history of malignancy. For assessing the occurrence of bleeding events beyond the hospital setting, the multivariable analysis incorporated covariates such as age (≥75years), sex, BMI, eGFR<30mL/min/1.73m2, use of prasugrel, use of anticoagulants, history of cerebrovascular disease, history of malignancy, haemoglobin levels at the time of AMI admission, and in‐hospital BARC Type 3 or 5 bleeding. BMI and haemoglobin levels were treated as continuous variables, while the others were treated as categorical variables.
All statistical analyses were performed using JMP Version 16.0.0 (SAS Institute, Cary, NC, USA) and SAS Version 9.4 (SAS Institute, Cary, NC, USA). A P value of <0.05 was considered significant.
Results
Baseline patient characteristics
Among the initial 3411 patients registered in the JAMIR, 185 of those who died during hospitalization were excluded, leaving 3226 patients for analysis (Figure1). Among them, 124 (3.8%) were admitted due to HF during a median follow‐up period of 345days (IQR, 277–386days). The time distribution from initial AMI discharge to HF admission is depicted in Supporting Information, FigureS1, revealing a median time to HF admission of 126days (IQR, 49–222), with a particularly high incidence within 90days after discharge from AMI. The patient characteristics of the HF admission group and no HF admission group are detailed in Table1. The HF admission group included significantly more older patients and women. In addition, this group involved more patients presenting with HF on admission (Killip class ≥2); patients experiencing renal failure, anaemia, and elevated peak CK levels; and patients with reduced LVEF. Additionally, the group exhibited a higher prevalence of comorbidities, including diabetes mellitus, a history of MI, prior PCI, peripheral artery disease, malignancy, or atrial fibrillation, than the no HF admission group. However, the frequency of patients who presented with ST‐segment elevation MI (STEMI) or those arriving at the hospital earlier did not significantly differ between the two groups.
Figure 1
Study flow chart. HF, heart failure; JAMIR, Japan Acute Myocardial Infarction Registry.
Table 1
Patient characteristics
| Overall | HF admission | No HF admission | P value | |
|---|---|---|---|---|
| N=3226 | N=124 | N=3102 | ||
| Age (years) | 67.6±13.2 | 75.1±11.9 | 67.3±13.1 | <0.001 |
| Women | 23.2 | 41.1 | 22.4 | <0.001 |
| BMI (kg/m2) | 23.9±3.9 | 23.3±5.4 | 23.9±3.9 | 0.205 |
| Use of ambulance | 81.9 | 86.3 | 81.7 | 0.195 |
| Time from onset to admission (min) | 140 (65–318) | 155 (61–300) | 139 (65–320) | 0.530 |
| Out‐of‐hospital cardiopulmonary arrest | 2.9 | 3.2 | 2.9 | 0.801 |
| STEMI | 76.5 | 76.6 | 76.5 | 0.983 |
| Killip Class 2 or 3 | 13.7 | 29.8 | 13.1 | <0.001 |
| Killip Class 4 | 6.6 | 15.3 | 6.3 | <0.001 |
| Hypertension | 72.9 | 79.0 | 72.7 | 0.119 |
| Diabetes | 34.6 | 46.8 | 34.1 | 0.004 |
| Dyslipidaemia | 69.2 | 70.2 | 69.1 | 0.805 |
| Previous myocardial infarction | 9.5 | 19.4 | 9.1 | <0.001 |
| Previous PCI | 11.5 | 19.4 | 11.2 | 0.005 |
| Previous CABG | 2.5 | 4.8 | 2.4 | 0.125 |
| Previous cerebrovascular disease | 9.1 | 13.7 | 9.0 | 0.072 |
| Peripheral artery disease | 3.8 | 10.5 | 3.6 | 0.001 |
| Malignancy | 8.1 | 14.5 | 7.8 | 0.007 |
| Atrial fibrillation | 6.2 | 12.9 | 6.0 | 0.002 |
| Current smoking | 40.9 | 24.2 | 41.6 | <0.001 |
| Systolic blood pressure | 140.6±31.4 | 128±36.1 | 141.1±31.1 | <0.001 |
| Heart rate | 78.3±21.0 | 79.7±23.4 | 78.3±20.9 | 0.477 |
| eGFR (mL/min/1.73m2) | 66.2±27.5 | 50±24.3 | 66.8±27.4 | <0.001 |
| Haemoglobin (g/dL) | 13.9±2.2 | 12.8±2.3 | 13.9±2.1 | <0.001 |
| Peak CK (IU/L) | 1465 (528–3209) | 1720 (738–3816) | 1455 (524–3176) | 0.045 |
| Peak CK‐MB (IU/L) | 138 (50–306) | 169 (84–334) | 137 (49–306) | 0.143 |
| LVEF (%) | 52.4±11.9 | 45.5±12.3 | 52.7±11.8 | <0.001 |
| Medication during hospitalization | ||||
| Aspirin | 98.1 | 96.8 | 98.2 | 0.295 |
| Prasugrel | 81.6 | 75.8 | 81.8 | 0.092 |
| Clopidogrel | 17.7 | 19.4 | 17.7 | 0.629 |
| ACE inhibitors or ARBs | 78.1 | 74.2 | 78.2 | 0.286 |
| ACE inhibitors | 53.4 | 39.5 | 54.0 | 0.002 |
| ARBs | 28.4 | 36.3 | 28.1 | 0.048 |
| Beta‐blockers | 66.3 | 69.4 | 66.2 | 0.459 |
| Statins | 90.7 | 85.5 | 90.9 | 0.041 |
| Oral anticoagulants | 13.8 | 29.0 | 13.2 | <0.001 |
| Proton pump inhibitors | 92.1 | 90.3 | 92.2 | 0.447 |
Open in a separate window
ACE, angiotensin‐converting enzyme; ARBs, angiotensin receptor blockers; BMI, body mass index; CABG, coronary artery bypass grafting; CK, creatine kinase; eGFR, estimated glomerular filtration rate; LVEF, left ventricular ejection fraction; PCI, percutaneous coronary intervention; STEMI, ST‐segment elevation myocardial infarction.
Data are presented as the means±standard deviations, medians (inter‐quartile ranges), or percentages.
Regarding medications, the overall usage rates of ACEIs or ARBs, beta‐blockers, and statins were 78.1%, 66.3%, and 90.7%, respectively, in the total population. Except for statins, there were no significant differences in the utilization rates of renin–angiotensin system (RAS) inhibitors and beta‐blockers between the two groups. Notably, the frequency of ACEI use was significantly lower in the HF admission group than in the no HF admission group (39.5% vs. 54.0%, P=0.002), while the utilization of ARBs displayed a contrasting trend (HF admission group: 36.3% vs. no HF admission group: 28.1%, P=0.048).
The angiographic and interventional characteristics are outlined in Table2. The HF admission group exhibited a higher mean number of diseased vessels and had a greater likelihood of receiving mechanical circulatory support, including intra‐aortic balloon pumping. Meanwhile, there were no significant between‐group differences in culprit lesions, final thrombolysis in MI flow grade, door‐to‐balloon time, number of mechanical complications, or primary PCI rates. The overall rate of primary PCI implementation was notably high, reaching 93.4%.
Table 2
Angiographic and interventional characteristics
| Overall | HF admission | No HF admission | P value | |
|---|---|---|---|---|
| N=3226 | N=124 | N=3102 | ||
| Emergent CAG | 97.3 | 96.0 | 97.4 | 0.383 |
| Primary PCI | 93.4 | 93.3 | 93.4 | 0.953 |
| Thrombolysis | 0.6 | 0 | 0.6 | 1.000 |
| Culprit lesion | ||||
| Left main coronary artery | 1.8 | 3.2 | 1.7 | 0.279 |
| Left anterior descending artery | 46.0 | 47.6 | 45.9 | 0.719 |
| Left circumflex artery | 15.2 | 14.5 | 15.3 | 0.824 |
| Right coronary artery | 35.8 | 34.7 | 35.9 | 0.790 |
| None | 1.3 | 0.8 | 1.4 | 1.000 |
| Number of diseased vessels | 0.005 | |||
| 0 | 1.8 | 0.8 | 1.9 | |
| 1 | 56.3 | 42.0 | 56.8 | |
| 2 | 25.4 | 31.9 | 25.2 | |
| 3 | 16.5 | 25.2 | 16.2 | |
| Multivessel disease | 41.9 | 57.1 | 41.3 | <0.001 |
| Mean number of diseased vessels | 1.6±0.8 | 1.8±0.8 | 1.6±0.8 | <0.001 |
| Door‐to‐balloon time (min) | 69 (51–102) | 72 (51–139) | 69 (51–101) | 0.301 |
| Final TIMI flow | 0.126 | |||
| 0 | 2.6 | 5.0 | 2.5 | |
| 1 | 1.3 | 0.8 | 1.3 | |
| 2 | 5.4 | 8.4 | 5.2 | |
| 3 | 90.8 | 85.7 | 91 | |
| Concomitant PCI in non‐culprit lesion | 5.1 | 8.4 | 5.0 | 0.099 |
| Use of IABP | 11.4 | 25.8 | 10.8 | <0.001 |
| Use of VA‐ECMO | 0.9 | 2.4 | 0.8 | 0.098 |
| Use of CABG | 2.8 | 4.8 | 2.7 | 0.158 |
| Length of hospitalization | 12 (8–17) | 17 (12–27) | 12 (8–16) | <0.001 |
| In‐hospital BARC Type 3 or 5 bleeding | 2.3 | 8.1 | 2.0 | <0.001 |
| In‐hospital ischaemic events | 1.0 | 0.8 | 1.0 | 1.000 |
| Non‐fatal myocardial infarction | 0.6 | 0.8 | 0.6 | 0.545 |
| Non‐fatal stroke | 0.4 | 0 | 0.4 | 1.000 |
| Stent thrombosis | 0.4 | 0.8 | 0.4 | 0.376 |
| Mechanical comorbidities | ||||
| Free wall rupture (blowout type) | 0 | 0 | 0 | NA |
| Free wall rupture (oozing type) | 0.4 | 0 | 0.4 | 1.000 |
| Ventricular septum perforation | 0.1 | 0.8 | 0.1 | 0.111 |
| Papillary muscle rupture | 0.1 | 0 | 0.1 | 1.000 |
| RV involvement | 2.8 | 5.7 | 2.7 | 0.086 |
Open in a separate window
BARC, Bleeding Academic Research Consortium; CABG, coronary artery bypass grafting; CAG, coronary angiography; IABP, intra‐aortic balloon pumping; PCI, percutaneous coronary intervention; RV, right ventricular; TIMI, thrombolysis in myocardial infarction; VA‐ECMO, venoarterial extracorporeal membrane oxygenation.
Data are presented as the means±standard deviations, medians (inter‐quartile ranges), or percentages.
Regarding in‐hospital bleeding and ischaemic events, in‐hospital BARC Type 3 or 5 bleeding events were more frequent in the HF admission group. However, there were no significant differences in in‐hospital ischaemic events between the two groups.
Factors associated with heart failure admission after acute myocardial infarction
In the multivariable analysis, the independent predictors of HF admission were age ≥75years (OR: 2.33, 95% CI: 1.44–3.76), female sex (OR: 1.71, 95% CI: 1.05–2.77), Killip class ≥2 (OR: 1.89, 95% CI: 1.16–3.06), LVEF<40% (OR: 2.78, 95% CI: 1.65–4.67), eGFR≤30mL/min/1.73m2 (OR: 2.04, 95% CI: 1.14–3.67), and a history of malignancy (OR: 1.90, 95% CI: 1.03–3.53). Importantly, the use of ACEIs was the sole protective factor (OR: 0.48, 95% CI: 0.30–0.77) (Supporting Information, TableS1).
Unadjusted outcomes of heart failure admission after acute myocardial infarction
The clinical outcomes after discharge are detailed in Table3. The unadjusted Kaplan–Meier curves are shown in Figure2A,B. The all‐cause mortality rate, as the primary endpoint, was higher in the HF admission group than in the no HF admission group (11.3% vs. 2.5%, log‐rank P<0.001). Similarly, the incidence of MACEs was higher in the HF admission group (16.9% vs. 2.7%, log‐rank P<0.001). In the subgroup analysis for the AMI subset, including STEMI or non‐STEMI (NSTEMI), the impact of HF admission on mortality was consistently observed in both the STEMI and NSTEMI groups (STEMI: log‐rank P<0.001; NSTEMI: log‐rank P=0.035). We performed a landmark analysis to evaluate the impact of early post‐discharge HF hospitalization events on mortality events in a more distant period. The landmark analysis revealed that the patients who developed HF admission within 90days had a significantly higher incidence of mortality after 90days (Supporting Information, FigureS2).
Table 3
Clinical outcomes after discharge
| HF admission | No HF admission | P value | |
|---|---|---|---|
| N=124 | N=3102 | ||
| All‐cause mortality | 14 (11.3) | 77 (2.5) | <0.001 |
| MACEs (cardiovascular mortality, non‐fatal myocardial infarction, or non‐fatal cerebral infarction) | 21 (16.9) | 83 (2.7) | <0.001 |
| Cardiovascular mortality | 10 (8.1) | 17 (0.6) | <0.001 |
| Myocardial infarction | 12 (9.7) | 55 (1.8) | <0.001 |
| Cerebral infarction | 3 (2.4) | 28 (0.9) | 0.009 |
Open in a separate window
HF, heart failure; MACEs, major adverse cardiovascular events.
Data are presented as number (%).
Figure 2
Unadjusted Kaplan–Meier curves comparing all‐cause mortality between the heart failure (HF) admission group and the no HF admission group. (A) Overall. (B) ST‐segment elevation myocardial infarction (STEMI) subgroup or non‐ST‐segment elevation myocardial infarction (NSTEMI) subgroup. CI, confidence interval.
Heart failure admission as an independent risk factor for all‐cause mortality
We conducted a multivariable analysis to evaluate whether HF admission independently contributed to the risk of all‐cause mortality. HF admission was associated with a higher risk of all‐cause mortality, even after accounting for potential confounding variables such as age, sex, Killip Class 2 or higher at index hospitalization for AMI, renal function, peak CK level, and history of malignancy (adjusted HR: 2.24, 95% CI: 1.23–4.08, P=0.008) (Table4).
Table 4
Independent predictors of all‐cause mortality beyond the hospital setting after acute myocardial infarction
| Unadjusted | Adjusted | |||
|---|---|---|---|---|
| HR (95% CI) | P value | HR (95% CI) | P value | |
| HF admission | 4.58 (2.59–8.09) | <0.001 | 2.24 (1.23–4.08) | 0.008 |
| Age ≥75years | 3.89 (2.53–5.99) | <0.001 | 2.84 (1.78–4.53) | <0.001 |
| Female sex | 2.22 (1.46–3.38) | <0.001 | 1.40 (0.89–2.18) | 0.144 |
| Killip class ≥2 | 2.42 (1.58–3.71) | <0.001 | 1.67 (1.07–2.63) | 0.025 |
| eGFR≤30mL/min/1.73m2 | 4.84 (2.99–7.84) | <0.001 | 2.83 (1.70–4.72) | <0.001 |
| Peak CK≥3000IU/L | 0.92 (0.57–1.47) | 0.716 | 1.05 (0.64–1.72) | 0.848 |
| Malignancy | 2.85 (1.70–4.78) | <0.001 | 1.68 (0.97–2.91) | 0.065 |
Open in a separate window
CI, confidence interval; CK, creatine kinase; eGFR, estimated glomerular filtration rate; HF, heart failure; HR, hazard ratio.
HF admission: heart failure admission after discharge from the index acute myocardial infraction hospitalization is included as a time‐dependent variable.
Bleeding events beyond the hospital setting associated with heart failure following acute myocardial infarction
Supporting Information, FigureS3 presents the unadjusted Kaplan–Meier curves comparing the incidence of major bleeding events after discharge between the HF admission and no HF admission groups. BARC Type 3 or 5 bleeding events beyond the hospital setting occurred more frequently in the HF admission group than in the no HF admission group (6.5% vs. 1.6%, log‐rank P<0.001). In the multivariable analysis, HF admission was an independent factor associated with major bleeding events beyond the hospital setting. This association remained significant even after adjusting for factors significantly related to bleeding, such as the use of prasugrel, anticoagulation drugs, prior cerebrovascular disease, previous malignant disease, anaemia on admission, and in‐hospital major bleeding events (adjusted HR: 2.37, 95% CI: 1.07–5.23, P=0.033) (Supporting Information, TableS2).
Discussion
The comprehensive analysis of the nationwide prospective AMI registry can be summarized by the following key findings: first, in the contemporary management of AMI, 3.8% of patients experienced hospitalization for HF within the first year after their AMI event. Second, hospitalization for HF was significantly associated with subsequent all‐cause mortality, and this risk persisted even after adjustment for background factors, such as age and sex. Third, the predictors of HF admission following AMI included older age, female sex, higher Killip class on admission, reduced LVEF, renal failure, and a history of malignancy. Fourth, the use of ACEIs had a protective effect against HF, whereas the use of ARBs did not demonstrate a similar protective effect. Finally, there was a potential association between post‐hospital bleeding events and HF admission.
The CREDO‐KYOTO study,6 an AMI cohort study in Japan, had reported the incidence and prognosis of HF after AMI. This multicentre registry in Japan enrolled 3682 consecutive patients with STEMI undergoing PCI within 24h after onset between 2005 and 2007. The 1year HF admission rate was 4.4%. A landmark analysis also demonstrated a significant association between HF admission and the incidence of all‐cause mortality up to 5years thereafter. However, it is important to note that the CREDO‐KYOTO study was conducted in the early primary PCI era. Thus, AMI management was different from that in current practice. The results of the present study are significant as they provide valuable insights into the prognostic impact of HF admission using real‐world data in the context of more recent AMI management practices.
We compared our results with those from the CREDO‐KYOTO registry.6 The CREDO‐KYOTO focused exclusively on patients with STEMI diagnosed using the MONICA criteria, while the JAMIR registry employed the universal definition for diagnosis, wherein 23% of the patients had NSTEMI. We found that the usage rate of DES was significantly increased from 30% to 98%, and the door‐to‐balloon time decreased from 90 to 69min. Additionally, both the utilization rates of statins and beta‐blockers at discharge increased from 57% to 91% and from 44% to 66%, respectively. Despite the implementation of rigorous initial treatment and the enhancement of secondary prevention therapy, the 1year HF admission rate was only slightly decreased between the two studies, 4.4% in 2005–07 (CREDO‐KYOTO registry study) and 3.8% in 2015–17 (JAMIR study).
There are limited reports on the trends in hospitalization for HF after AMI in other regions. Medical data from the United States indicate that the incidence rate of HF in the first year after AMI declined from 16.1% in 1998 to 14.2% in 2010.12 Additionally, the SWEDEHEART registry also reported that the 2year post‐AMI HF admission rate decreased from 15.5% in 2004–05 to 14.4% in 2010–11.13 These results, which show a relatively higher incidence rate of HF admission following AMI in Western countries, may be related to differences in the implementation rate of PCI, which is 65% for the United States (2004–10),14 59% for Sweden (2008),15 and 93% for Japan (2015–17). Regarding the difference in primary PCI rates between these studies, several potential reasons exist. First, the primary PCI rate in Japan is relatively high compared with other countries due to geographical issues, differences in the insurance system, and the large number of available facilities providing PCI. Second, the JAMIR excluded patients who were admitted to the hospital ≥24h after onset or with no return of spontaneous circulation on admission after out‐of‐hospital cardiopulmonary arrest. Therefore, a relatively large number of patients may have been eligible for primary PCI in our study.
Early intervention during HF hospitalization may improve the long‐term prognosis, including all‐cause mortality. Several risk factors for post‐AMI HF have been reported; these include older age, female sex, history of MI, chronic kidney disease, and low LVEF.16, 17 The present findings from the JAMIR support these previous conclusions. ACEI use was the sole protective factor associated with HF in the present analysis. Although there was no significant difference in the overall composite use of RAS inhibitors, univariate analysis revealed less frequent ACEI use in patients without HF than in those with HF. Interestingly, ACEI usage was associated with a significantly lower risk of HF after adjusting for background factors. Large‐scale observational studies and meta‐analyses have consistently shown a higher efficacy of ACEIs than ARBs. The Korea Acute Myocardial Infarction Registry study demonstrated that ACEI use reduced the 3year HF admission rate in AMI patients without hypertension, outperforming ARBs (1.5% vs. 3.6%; adjusted HR: 0.399, 95% CI: 0.294–0.541, P<0.001).18
Another meta‐analysis by the Blood Pressure Lowering Treatment Trialists' Collaboration revealed that unlike ARBs, ACEI use reduced coronary events independent of their antihypertensive effects.19 These differences can be attributed to the pharmacological disparities between ACEIs and ARBs. ACEIs affect the kallikrein–kinin system, increasing nitric oxide and vasodilatory factors, potentially preventing HF. In contrast, ARBs selectively inhibit angiotensin II type 1 receptors, promoting stimulation of the angiotensin II type 2 (AT2) receptor. The role of AT2 receptors remains contentious, with some suggesting detrimental effects on HF. These pharmacological distinctions likely contribute to the superiority of ACEIs in HF treatment, as supported by our findings. Our results endorse the preference for ACEIs over ARBs in post‐AMI patients, in line with guideline recommendations. ACEIs should be the first choice, except in cases of ACEI intolerance.
In contrast, the usage rate of beta‐blockers was not significantly different between the HF admission and no HF admission groups. One possible reason for this could be that the mean LVEF was relatively well preserved (52.4±11.9%). Previous studies demonstrating the efficacy of beta‐blockers in AMI were conducted before the widespread adoption of early reperfusion therapy, and the enrolled patients had LVEF values of ≤40%. Meanwhile, a recent randomized trial and an observational study have shown no clear benefit of beta‐blockers in AMI patients with preserved ejection fraction.20, 21 Further research is needed to verify the usefulness of beta‐blockers for patients with AMI in the primary PCI era.
Since the completion of this study, innovative therapies such as sodium–glucose cotransporter 2 inhibitors (SGLT2Is) and angiotensin receptor–neprilysin inhibitors (ARNIs) have emerged as promising options for HF management. They may prevent adverse remodelling after MI and new HF onset. Evidence suggests the benefits of early SGLT2I use after MI. Paolisso et al. found reduced HF risk with pre‐MI SGLT2I use.22 Von Lewinski et al. demonstrated improved outcomes with early SGLT2I after PCI.23 However, randomized controlled trials validating these findings are limited. James et al. reported metabolic benefits but inconclusive cardiovascular impact with early SGLT2I use after MI.24 Ongoing randomized controlled trials aim to clarify SGLT2I's role in preventing HF and mortality after MI.25
Our additional analysis indicated an association between bleeding events beyond the hospital setting and HF admission. Similarly, a Japanese multicentre PCI registry study identified that a history of HF was a significant predictor of bleeding complications.26 A Danish cohort study also reported a substantial increase in haemorrhagic stroke risk within the first 30days after initial HF diagnosis, with persistent positive associations thereafter (HR for intracerebral haemorrhage: 2.13; HR for sub‐arachnoid haemorrhage: 3.52).27 Recent research has underscored the association between bleeding events and subsequent mortality, emphasizing the importance of managing both ischaemic and bleeding risks during antithrombotic therapy.28, 29 Given the elevated risk of bleeding events in HF patients, interventions aimed at halting HF progression and careful selection of antithrombotic therapy in these patients may be pivotal for mitigating the bleeding risk after AMI.
There are several limitations in the present study. First, the registry lacked data on specific HF treatments, including the use of mineralocorticoid receptor antagonists, serum BNP concentrations, detailed cardiac function data based on echocardiography highlighting right ventricular function and significant valvular pathologies, previous history of HF and non‐ischaemic cardiomyopathy, information about treatment for non‐culprit lesions, and device therapy. Second, our study featured only a 1year follow‐up period, and thus, the long‐term prognosis and follow‐up cardiac function are unknown. Third, the choice between ACEIs and ARBs was decided by the attending physician, introducing inherent bias into the findings. Moreover, we do not have data on ACEI intolerance. Fourth, the study did not include data on the use of novel HF therapies, such as SGLT2I and ARNI, as these medications were not widely used during the study period. Therefore, the study might not have reflected the latest treatment. Fifth, all clinical events were adjudicated by the cardiologist in each facility and not an independent clinical event committee. Sixth, the causal relationship between post‐discharge HF and MACEs or bleeding events remains unclear and warrants further investigation.
In conclusion, the real‐world data from the contemporary AMI registry JAMIR revealed that HF admission was significantly associated with increased mortality. These findings emphasize the urgent need for comprehensive strategies aimed at reducing HF admission rates, thus enhancing the long‐term outcomes of patients with AMI in the primary PCI era.
Conflict of interest
S.Y. reports remuneration for lectures from Takeda, Daiichi‐Sankyo, and Bristol‐Myers Squibb and trust research/joint research funds from Takeda and Daiichi‐Sankyo. M. Takayama reports lecture fees from Daiichi‐Sankyo. H.O. reports lecture fees and research grants from Abbott Japan, Bayer, Daiichi‐Sankyo, Eisai, Kowa, Takeda Pharmaceutical Company, and Teijin. The other authors have no conflicts of interest.
Funding
This work was planned by the Japan Cardiovascular Research Foundation and was funded by Daiichi‐Sankyo.
Supporting information
Figure S1. The distribution of the time from AMI to HF admission. AMI, acute myocardial infarction; HF, heart failure.
Click here to view.(113K, pdf)
Figure S2. Unadjusted cumulative mortality rate using the 90‐days landmark analysis based on the HF admission within 90days. HF, heart failure.
Click here to view.(45K, pdf)
Figure S3. Unadjusted Kaplan–Meier curves comparing bleeding events (BARC type 3 or 5) beyond the hospital setting between the HF admission (blue line) and the no Hf admission (red line) groups. BARC, Bleeding Academic Research Consortium; HF, heart failure; CI, confidence interval.
Click here to view.(12K, pdf)
Table S1. Predictors of HF admission.
Table S2. Independent predictors of out‐hospital major bleeding events after acute myocardial infarction.
Click here to view.(801K, docx)
Acknowledgements
The authors wish to thank all the investigators, clinical research coordinators, and data managers involved in the JAMIR study for their contributions.
Notes
Takeuchi, S., Honda, S., Nishihira, K., Kojima, S., Takegami, M., Asaumi, Y., Saji, M., Yamash*ta, J., Hibi, K., Takahashi, J., Sakata, Y., Takayama, M., Sumiyoshi, T., Ogawa, H., Kimura, K., Yasuda, S., and JAMIR Investigators(2024) Prognostic impact of heart failure admission in survivors of acute myocardial infarction. ESC Heart Failure, 11: 2344–2353. 10.1002/ehf2.14790. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
References
1. Cui Y, Hao K, Takahashi J, Miyata S, Shindo T, Nishimiya K, et al. Age‐specific trends in the incidence and in‐hospital mortality of acute myocardial infarction over 30years in Japan—Report from the Miyagi AMI Registry Study. Circ J2017;81:520‐528. doi: 10.1253/circj.CJ-16-0799 [PubMed] [CrossRef] [Google Scholar]
2. Degano IR, Salomaa V, Veronesi G, Ferrieres J, Kirchberger I, Laks T, et al. Twenty‐five‐year trends in myocardial infarction attack and mortality rates, and case‐fatality, in six European populations. Heart2015;101:1413‐1421. doi: 10.1136/heartjnl-2014-307310 [PubMed] [CrossRef] [Google Scholar]
3. Jenca D, Melenovsky V, Stehlik J, Stanek V, Kettner J, Kautzner J, et al. Heart failure after myocardial infarction: Incidence and predictors. ESC Heart Fail2021;8:222‐237. doi: 10.1002/ehf2.13144 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
4. Curley D, Lavin Plaza B, Shah AM, Botnar RM. Molecular imaging of cardiac remodelling after myocardial infarction. Basic Res Cardiol2018;113:10. doi: 10.1007/s00395-018-0668-z [PMC free article] [PubMed] [CrossRef] [Google Scholar]
5. Shiba N, Nochioka K, Miura M, Kohno H, Shimokawa H, CHART‐2 Investigators. Trend of westernization of etiology and clinical characteristics of heart failure patients in Japan—First report from the CHART‐2 study. Circ J2011;75:823‐833. doi: 10.1253/circj.cj-11-0135 [PubMed] [CrossRef] [Google Scholar]
6. Taniguchi T, Shiomi H, Morimoto T, Watanabe H, Ono K, Shizuta S, et al. Incidence and prognostic impact of heart failure hospitalization during follow‐up after primary percutaneous coronary intervention in ST‐segment elevation myocardial infarction. Am J Cardiol2017;119:1729‐1739. doi: 10.1016/j.amjcard.2017.03.013 [PubMed] [CrossRef] [Google Scholar]
7. Kimura K, Kimura T, Ishihara M, Nakagawa Y, Nakao K, Miyauchi K, et al. JCS 2018 guideline on diagnosis and treatment of acute coronary syndrome. Circ J2019;83:1085‐1196. doi: 10.1253/circj.CJ-19-0133 [PubMed] [CrossRef] [Google Scholar]
8. Honda S, Nishihira K, Kojima S, Takegami M, Asaumi Y, Suzuki M, et al. Rationale, design, and baseline characteristics of the prospective Japan Acute Myocardial Infarction Registry (JAMIR). Cardiovasc Drugs Ther2019;33:97‐103. doi: 10.1007/s10557-018-6839-1 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
9. Yasuda S, Honda S, Takegami M, Nishihira K, Kojima S, Asaumi Y, et al. Contemporary antiplatelet therapy and clinical outcomes of Japanese patients with acute myocardial infarction—Results from the prospective Japan Acute Myocardial Infarction Registry (JAMIR). Circ J2019;83:1633‐1643. doi: 10.1253/circj.CJ-19-0145 [PubMed] [CrossRef] [Google Scholar]
10. JCS Joint Working Group. Guidelines for treatment of acute heart failure (JCS 2011)—Digest version. Circ J2013;77:2157‐2201. doi: 10.1253/circj.cj-66-0068 [PubMed] [CrossRef] [Google Scholar]
11. Mehran R, Rao SV, Bhatt DL, Gibson CM, Caixeta A, Eikelboom J, et al. Standardized bleeding definitions for cardiovascular clinical trials: A consensus report from the Bleeding Academic Research Consortium. Circulation2011;123:2736‐2747. doi: 10.1161/CIRCULATIONAHA.110.009449 [PubMed] [CrossRef] [Google Scholar]
12. Chen J, Hsieh AF, Dharmarajan K, Masoudi FA, Krumholz HM. National trends in heart failure hospitalization after acute myocardial infarction for Medicare beneficiaries: 1998–2010. Circulation2013;128:2577‐2584. doi: 10.1161/CIRCULATIONAHA.113.003668 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
13. Desta L, Jernberg T, Spaak J, Hofman‐Bang C, Persson H. Risk and predictors of readmission for heart failure following a myocardial infarction between 2004 and 2013: A Swedish nationwide observational study. Int J Cardiol2017;248:221‐226. doi: 10.1016/j.ijcard.2017.05.086 [PubMed] [CrossRef] [Google Scholar]
14. Shah RU, Henry TD, Rutten‐Ramos S, Garberich RF, Tighiouart M, Bairey Merz CN. Increasing percutaneous coronary interventions for ST‐segment elevation myocardial infarction in the United States: Progress and opportunity. JACC Cardiovasc Interv2015;8:139‐146. doi: 10.1016/j.jcin.2014.07.017 [PubMed] [CrossRef] [Google Scholar]
15. Gerber Y, Weston SA, Berardi C, McNallan SM, Jiang R, Redfield MM, et al. Contemporary trends in heart failure with reduced and preserved ejection fraction after myocardial infarction: A community study. Am J Epidemiol2013;178:1272‐1280. doi: 10.1093/aje/kwt109 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
16. Lewis EF, Velazquez EJ, Solomon SD, Hellkamp AS, McMurray JJ, Mathias J, et al. Predictors of the first heart failure hospitalization in patients who are stable survivors of myocardial infarction complicated by pulmonary congestion and/or left ventricular dysfunction: A VALIANT study. Eur Heart J2008;29:748‐756. doi: 10.1093/eurheartj/ehn062 [PubMed] [CrossRef] [Google Scholar]
17. Gho J, Postema PG, Conijn M, Bruinsma N, de Jong J, Bezzina CR, et al. Heart failure following STEMI: A contemporary cohort study of incidence and prognostic factors. Open Heart2017;4:e000551. doi: 10.1136/openhrt-2016-000551 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
18. Her AY, Choi BG, Rha SW, Kim YH, Choi CU, Jeong MH. The impact of angiotensin‐converting‐enzyme inhibitors versus angiotensin receptor blockers on 3‐year clinical outcomes in patients with acute myocardial infarction without hypertension. PLoS ONE2020;15:e0242314. doi: 10.1371/journal.pone.0242314 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
19. Blood Pressure Lowering Treatment Trialists' Collaboration, Turnbull F, Neal B, Pfeffer M, Kostis J, Algert C, et al. Blood pressure‐dependent and independent effects of agents that inhibit the renin–angiotensin system. J Hypertens2007;25:951‐958. doi: 10.1097/HJH.0b013e3280bad9b4 [PubMed] [CrossRef] [Google Scholar]
20. Watanabe H, Ozasa N, Morimoto T, Shiomi H, Bingyuan B, Suwa S, et al. Long‐term use of carvedilol in patients with ST‐segment elevation myocardial infarction treated with primary percutaneous coronary intervention. PLoS ONE2018;13:e0199347. doi: 10.1371/journal.pone.0199347 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
21. Joo SJ, Kim SY, Choi JH, Park HK, Beom JW, Lee JG, et al. Effect of beta‐blocker therapy in patients with or without left ventricular systolic dysfunction after acute myocardial infarction. Eur Heart J Cardiovasc Pharmacother2021;7:475‐482. doi: 10.1093/ehjcvp/pvaa029 [PubMed] [CrossRef] [Google Scholar]
22. Paolisso P, Bergamaschi L, Gragnano F, Gallinoro E, Cesaro A, Sardu C, et al. Outcomes in diabetic patients treated with SGLT2‐inhibitors with acute myocardial infarction undergoing PCI: The SGLT2‐I AMI PROTECT registry. Pharmacol Res2023;187:106597. doi: 10.1016/j.phrs.2022.106597 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
23. von Lewinski D, Kolesnik E, Tripolt NJ, Pferschy PN, Benedikt M, Wallner M, et al. Empagliflozin in acute myocardial infarction: The EMMY trial. Eur Heart J2022;43:4421‐4432. doi: 10.1093/eurheartj/ehac494 [PMC free article] [PubMed] [CrossRef] [Google Scholar]
24. James S, Erlinge D, Storey RF, McGuire DK, Belder M, Eriksson N, et al. Dapagliflozin in myocardial infarction without diabetes or heart failure. NEJM Evidence2024;3:EVIDoa2300286. doi: 10.1056/EVIDoa2300286 [PubMed] [CrossRef] [Google Scholar]
25. Harrington J, Udell JA, Jones WS, Anker SD, Bhatt DL, Petrie MC, et al. Empagliflozin in patients post myocardial infarction rationale and design of the EMPACT‐MI trial. Am Heart J2022;253:86‐98. doi: 10.1016/j.ahj.2022.05.010 [PubMed] [CrossRef] [Google Scholar]
26. Numasawa Y, Kohsaka S, Ueda I, Miyata H, Sawano M, Kawamura A, et al. Incidence and predictors of bleeding complications after percutaneous coronary intervention. J Cardiol2017;69:272‐279. doi: 10.1016/j.jjcc.2016.05.003 [PubMed] [CrossRef] [Google Scholar]
27. Adelborg K, Szepligeti S, Sundboll J, Horvath‐Puho E, Henderson VW, Ording A, et al. Risk of stroke in patients with heart failure: A population‐based 30‐year cohort study. Stroke2017;48:1161‐1168. doi: 10.1161/STROKEAHA.116.016022 [PubMed] [CrossRef] [Google Scholar]
28. Nakamura M, Kimura K, Kimura T, Ishihara M, Otsuka F, Kozuma K, et al. JCS 2020 guideline focused update on antithrombotic therapy in patients with coronary artery disease. Circ J2020;84:831‐865. doi: 10.1253/circj.CJ-19-1109 [PubMed] [CrossRef] [Google Scholar]
29. Kazi DS, Leong TK, Chang TI, Solomon MD, Hlatky MA, Go AS. Association of spontaneous bleeding and myocardial infarction with long‐term mortality after percutaneous coronary intervention. J Am Coll Cardiol2015;65:1411‐1420. doi: 10.1016/j.jacc.2015.01.047 [PubMed] [CrossRef] [Google Scholar]
Articles from ESC Heart Failure are provided here courtesy of Wiley
