Abstract
Abstract
Objectives
To assess whether a simple echocardiographic grading system reflecting the severity and stage of myocardial injury can stratify long-term mortality risk among patients with first-time ST-segment elevation myocardial infarction (STEMI).
Design
Multicentre prospective registry-based cohort study.
Setting
Eight hospitals in China.
Participants
Consecutive patients with first-time STEMI between June 2016 and June 2024 who survived to hospital discharge and underwent echocardiography were included. A total of 2708 patients were enrolled and followed up for a median of 5.5 years (IQR: 5.3–6.4).
Outcome measures
The primary outcome was long-term all-cause mortality.
Results
Patients were stratified into four grades: 1392 (51.4%) were classified as Grade 1; 905 (33.4%) as Grade 2; 350 (12.9%) as Grade 3; and 61 (2.3%) as Grade 4. During follow-up, long-term all-cause mortality increased stepwise across increasing echocardiographic grades. In multivariable Cox regression analysis, the baseline grading remained an independent predictor of long-term mortality after adjustment (adjusted HR for Grade 4 vs Grade 1: 3.35, 95% CI 1.34 to 8.38, p<0.01). In a subset of patients with available 90-day echocardiographic data, the stratification yielded comparable group distributions and mortality trends to those observed with baseline data (adjusted HR for Grade 4 vs Grade 1: 6.22, 95% CI 1.69 to 22.82, p<0.01).
Conclusion
Patient grade from the echocardiographic grading system was associated with long-term mortality after first-time STEMI, with the highest risk observed in patients classified as Grade 4. As a simple and objective classification, this grading system may provide a practical way to describe and communicate infarct severity among clinicians and others involved in patient care, as the term ‘myocardial infarction’ alone may inadequately reflect disease severity.
Keywords: Echocardiography, Myocardial infarction, Prognosis, Risk Assessment
STRENGTHS AND LIMITATIONS OF THIS STUDY.
This multicentre cohort study included long-term follow-up of 5 years.
The echocardiographic grading system is based on standard, widely available measurements, making it practical and easy to use in everyday clinical care.
Only patients who survived to hospital discharge and had echocardiographic data were included, which may introduce survivor bias.
As an observational study, residual confounding cannot be excluded despite multivariable adjustment.
Introduction
Contemporary reperfusion therapy has dramatically improved survival rates following ST-elevation myocardial infarction (STEMI), fundamentally transforming the natural history of this condition.1 This therapeutic success has revealed a broad spectrum of clinical outcomes among STEMI survivors, ranging from minimal cardiac impact to severe left ventricular dysfunction with adverse remodelling.
The heterogeneity of post-STEMI outcomes is striking. With timely reperfusion strategies, a substantial proportion of patients experience only transient electrocardiographic changes with minimal biomarker elevation—so-called ‘abortive infarction’—resulting in tissue injury without detectable functional or morphological cardiac abnormalities.2 Conversely, many patients develop progressive left ventricular dysfunction, characterised by reduced left ventricular ejection fraction (LVEF), pathological remodelling and, in severe cases, significant mitral regurgitation (MR).
Despite this well-recognised clinical spectrum, cardiology lacks a standardised, objective grading system for STEMI severity and progression—a striking contrast to other medical disciplines that routinely employ staging systems (hepatic cirrhosis, chronic kidney disease, oncology).3 This absence of systematic classification represents a significant gap in STEMI care, limiting effective communication among healthcare providers, hindering optimal therapeutic decision-making and impeding patient education and rehabilitation planning.
The clinical implications of this unmet need are substantial. A standardised grading system would facilitate more precise risk stratification, enable personalised cardiac rehabilitation programmes, optimise follow-up intensity and improve resource allocation. Furthermore, such a system would enhance occupational health assessments for return-to-work decisions and provide a common language for multidisciplinary care teams.
Current clinical practice predominantly relies on either clinical risk scores or LVEF for risk stratification, but both approaches have important limitations. Clinical tools such as Killip class and the Global Registry of Acute Coronary Events (GRACE) score primarily reflect acute haemodynamic status, symptom severity and short-term risk during the index hospitalisation. In contrast, LVEF is commonly used to estimate longer-term prognosis after myocardial infarction. However, LVEF fails to capture haemodynamic status and may fluctuate significantly in the acute post-infarction period. More critically, LVEF alone does not adequately characterise the progressive pathophysiological changes that define post-STEMI cardiac remodelling—a key determinant of long-term prognosis. Additionally, the clinical interpretation of numerical LVEF changes remains challenging and inconsistent among practitioners.
To address these limitations, we previously developed a cardiac grading system that integrates both functional and morphological parameters based on cardiac magnetic resonance (CMR) imaging.4 This pathophysiology-based classification system stratifies patients into five grades: Grade 0 (no detectable myocardial necrosis), Grade 1 (myocardial necrosis without functional or morphological abnormalities), Grade 2 (necrosis with reduced LVEF), Grade 3 (LVEF reduction with significant cardiac remodelling) and Grade 4 (severe disease with MR plus morphological and functional abnormalities). Our initial validation demonstrated a strong association between this grading system and 90-day major adverse cardiovascular events (MACE).
For broader clinical applicability in settings where CMR is not routinely available,3 we have simplified this classification for use with transthoracic echocardiography by combining Grades 0 and 1 into a unified Grade 1 category (LVEF ≥55%), making the system more practical for widespread implementation while maintaining its prognostic utility.
Using a comprehensive first-time STEMI registry, this study aims to: (1) characterise the contemporary baseline characteristics and distribution of cardiac grading categories among patients with STEMI in a real-world, multicentre population; (2) examine the dynamic evolution of cardiac grading from index hospitalisation to 90-day follow-up, providing insights into the temporal progression of post-STEMI cardiac status; and (3) evaluate the association between an echocardiography-based cardiac grading system and long-term all-cause mortality in this expanded clinical cohort.
Methods
Study design
The STEMI follow-up registry was a prospective, longitudinal, multicentre registry study of patients hospitalised with first-time STEMI in China (ChiCTR-IDR-16007765).
Study population
A total of eight sites were included in the registry, with three academic hospitals located in different regions and five tier-2 district-level hospitals. The registry thus covered both tertiary and secondary hospitals, with long-term follow-up data available for all patients. From June 2016 to June 2024, first-time patients with STEMI aged 18 years or older who survived at discharge were included. The only exclusion criterion was missing echo data within hospitalisation. STEMI was defined as a chest pain lasting ≥30 min together with an ST-segment elevation in two contiguous leads on a standard 12-lead electrocardiogram (≥2 mm in precordial leads and ≥1 mm in the limb leads).5 Because this registry included only patients who survived the index hospitalisation, survivor bias cannot be excluded.
Echocardiography protocol
Transthoracic echocardiography was performed as part of standard clinical care for patients hospitalised with STEMI, in accordance with guideline-recommended practice. Follow-up echocardiography at 90 days was performed at the discretion of the treating physician, based on clinical indication during routine follow-up. Echocardiographic examinations were performed using commercially available ultrasound systems, mostly the Vivid E9 platform (GE Vingmed Ultrasound, Horten, Norway). Images were analysed using commercial software (Echopac, GE Vingmed Ultrasound, Horten, Norway). LVEF and left ventricular end-diastolic volume (LVEDV) were assessed using two-dimensional echocardiography, either by the biplane Simpson’s method (apical two-chamber and four-chamber views) or M-mode from the parasternal long-axis view, depending on the performing echocardiographer and measurements were performed by experienced physicians. MR was categorised as mild (grades 0 and 1), medium (grades 2 and 3) or severe (Grade 4) based on the ratio of the regurgitant jet area to the left atrial area.6 MR grades 3 and 4 were collectively defined as ‘apparent mitral regurgitation’. Echocardiographic readers were not involved in outcome assessment and were unaware of the grading algorithm. All echocardiographic measurements were interpreted by trained physicians and reviewed by experienced cardiologists. Data quality control procedures included systematic review of measurements and verification of outliers before data entry and analysis.
Modified grading system
In the present study, we developed an echocardiography-based four-grade classification system (Grade 1–4), adapted from our previously proposed five-grade CMR-based model (Grade 0–4).4 The new classification is as follows (figure 1A): Grade 1: LVEF ≥55%; Grade 2: Reduced LVEF (30%≤LVEF<55%) without cardiac remodelling; Grade 3: reduced LVEF (30%≤LVEF<55%) and cardiac remodelling; Grade 4: LVEF<30% or medium or severe MR in addition to the Grade-3 criteria. There were two major modifications of the previous grading system. First, Grade 0 and Grade 1 from the original CMR-based system were combined into a single Grade, as the only difference between these two groups of patients was the presence of myocardial necrosis detectable by cardiovascular magnetic resonance imaging (CMR), which cannot be reliably identified by echocardiography. Second, patients with extremely low EF (EF <30%) were included in Grade 4, as these patients were likely to have massive damage after myocardial infarction and exhibited a prognosis similar to those with moderately reduced EF and severe cardiac remodelling. Cardiac remodelling was defined as an increased LVEDV indexed to body surface area (LVEDV/BSA), with the upper limits of the normal range being 61 mL/m² for females and 74 mL/m² for males.7
Figure 1. Study design overview. (A) Illustration of the echocardiography-based cardiac risk grading system. The novel grading system includes four grades assessed by transthoracic echocardiography during hospitalisation: Grade 1: LVEF ≥55%, Grade 2: reduced LVEF (30%≤LVEF < 55%) without cardiac remodelling, Grade 3: reduced LVEF with cardiac remodelling, Grade 4: LVEF <30%, or medium-to-severe MR in addition to Grade 3 criteria. (B) Flowchart of patient selection and follow-up. EF, ejection fraction; LVEF, left ventricular ejection fraction; MI, myocardial infarction; STEMI, ST-segment elevation myocardial infarction.
Events and follow-up
The primary outcome of this study was all-cause mortality. Mortality data were obtained through medical record review and confirmed by telephone follow-up. All follow-up data were collected via the electronic data capture system, as well as in-person or telephone follow-up conducted by physicians and nurses.
Patients were followed up for a median of 5.5 years (IQR, 5.3–6.4 years). Of the entire cohort, 2704 patients (99.9%) had complete follow-up data. A total of 1420 patients (52.4%) underwent echocardiographic evaluation at 90 days, forming a 90-day echocardiographic follow-up subgroup.
Statistical analysis
Continuous variables were expressed as mean±SD or median with IQR. Categorical variables were presented as counts and percentages. Comparisons between the two groups were performed by Student’s t-test for continuous variables and χ² or Fisher’s exact test for categorical variables. Continuous variables among multiple groups were compared using one-way analysis of variance. Correlations between variables were assessed using Spearman’s correlation coefficients. To adjust for baseline characteristics in the prediction of all-cause mortality, logistic regression analysis was conducted, including major potential confounders. Covariates were selected based on clinical relevance and univariable associations, with consideration of the number of outcome events. ORs with 95% CIs were reported. Kaplan-Meier analysis with log-rank testing and Cox proportional hazards models was used to estimate HRs for time-to-event outcomes. All statistical tests were two-sided, and a p value <0.05 was considered statistically significant. All analyses were performed using R software (V.4.3.3; R Foundation for Statistical Computing, Vienna, Austria) within the RStudio environment (V.2024.12.1+563).
Patient and public involvement
None.
Results
Study cohort
The initial study population included all patients in the first-time STEMI registry. Between June 2016 and June 2024, there were 2708 patients included in our cohort study from eight hospitals (figure 1B).
Baseline characteristics
Of the 2708 patients, 1392 were classified as Grade 1 (51.40%), 905 as Grade 2 (33.4%), 350 as Grade 3 (12.9%) and 61 as Grade 4 (2.3%) (figure 2A and table 1). Baseline characteristics are listed in table 1.
Figure 2. Cardiac risk grading and long-term mortality at index hospitalisation. (A) Distribution of cardiac risk grades at initial hospitalisation. Colours represent Grades 1–4, labelled on the right; the y-axis indicates the proportion of patients in each grade. (B) Long-term mortality rates according to cardiac risk grade at baseline. Colours represent Grades 1–4, labelled on the right; the y-axis indicates the percentage of deaths. (C) Kaplan-Meier analysis of long-term survival by cardiac grade at index hospitalisation. Survival curves show progressively lower survival rates from Grades 1 to 4. In multivariable Cox proportional hazards analysis, Grade 4 was associated with the highest risk of all-cause mortality compared with Grade 1 (adjusted HR=3.35; 95% CI 1.34 to 8.38; p<0.0001).
Table 1. Baseline characteristics of the study population, stratified by grade as defined by the post-STEMI cardiac grading system.
| Overall N=2708 |
Grade 1 n=1392 |
Grade 2 n=905 |
Grade 3 n=350 |
Grade 4 n=61 |
P value | |
|---|---|---|---|---|---|---|
| Demographics | ||||||
| Age | 62.09±11.67 | 61.62±11.67 | 62.15±11.99 | 63.17±10.89 | 65.23±10.54 | 0.01 |
| Male | 2304 (85.1) | 1174 (84.3) | 820 (90.6) | 261 (74.6) | 49 (80.3) | <0.01 |
| BSA, m2 | 1.79±0.16 | 1.79±0.16 | 1.81±0.16 | 1.77±0.17 | 1.74±0.17 | <0.01 |
| Medical history | ||||||
| Previous CAD | 68 (2.5) | 36 (2.6) | 17 (1.9) | 13 (3.7) | 2 (3.3) | 0.20 |
| Previous PCI | 20 (0.7) | 7 (0.5) | 5 (0.6) | 8 (2.3) | 0 | <0.01 |
| Previous stroke | 65 (2.4) | 42 (3) | 17 (1.9) | 6 (1.7) | 0 | 0.12 |
| Hypertension | 1528 (56.4) | 779 (56) | 489 (54) | 223 (63.7) | 37 (60.7) | 0.01 |
| Diabetes | 703 (26) | 328 (23.6) | 248 (27.4) | 104 (29.7) | 23 (37.7) | <0.01 |
| Presentation at admission | ||||||
| Heart rate, bpm | 78.33±14.76 | 76.72±14.18 | 79.59±14.68 | 80.15±15.75 | 85.8±17.61 | <0.01 |
| Systolic BP, mm Hg | 126.28±22.09 | 127.77±22.47 | 124.39±21.91 | 126.26±20.73 | 120.7±21.09 | <0.01 |
| Diastolic BP, mm Hg | 75.71±13.32 | 76.13±13.72 | 74.66±12.84 | 77.03±12.62 | 74.31±13.85 | <0.01 |
| Killip class | <0.01 | |||||
| I | 2399 (88.9) | 1261 (90.9) | 798 (88.4) | 297 (85.6) | 43 (70.5) | |
| II | 173 (6.4) | 78 (5.6) | 59 (6.5) | 27 (7.8) | 9 (14.8) | |
| III | 44 (1.6) | 15 (1.1) | 11 (1.2) | 14 (4) | 4 (6.6) | |
| IV | 83 (3.1) | 34 (2.4) | 35 (3.9) | 9 (2.6) | 5 (8.2) | |
| Anterior infarction | 1396 (53.2) | 595 (44.2) | 559 (64.2) | 203 (58.5) | 39 (65) | <0.01 |
| Laboratory tests | ||||||
| LDL-C, mmol/L | 3.03±0.98 | 3.05±0.95 | 3.06±1.02 | 2.95±0.96 | 2.85±0.96 | 0.17 |
| Creatinine, μmol/L | 74 (63–87) | 72 (63–84) | 76.95 (63.08–89) | 74 (60–90) | 79.5 (61.25–97.75) | <0.01 |
| Haemoglobin, g/L | 141.16±18.41 | 140.81±17.87 | 143.73±19.27 | 136.94±17.3 | 135.15±17.74 | <0.01 |
| BNP, ng/L | 65 (20–216) | 47.46 (17.25–159.7) | 81 (23–246.25) | 107.19 (22.56–423) | 368 (100.51–855) | <0.01 |
| Peak troponin I, ng/mL | 14.23 (1.48–34.6) | 10.54 (1.01–30) | 19.31 (3.11–42.4) | 18.71 (2.14–59) | 15.74 (0.93–30) | <0.01 |
| Peak CK, ng/ml | 94.96 (33.95–232.12) | 80 (29.8–190.2) | 115.5 (46.4–288) | 118.84 (28.2–283.25) | 78.15 (11.55–281.92) | <0.01 |
| Reperfusion therapy | ||||||
| PCI | 2594 (95.8) | 1335 (95.9) | 871 (96.2) | 330 (94.3) | 58 (95.1) | 0.48 |
| Medications at discharge | ||||||
| Aspirin/P2Y12 inhibitor | 2648 (99.4) | 1375 (99.8) | 878 (99) | 339 (99.7) | 56 (94.9) | <0.01 |
| Beta-blocker | 2362 (88.7) | 1227 (89) | 779 (87.8) | 312 (91.8) | 44 (74.6) | <0.01 |
| ACEi/ARB | 1892 (71) | 1055 (76.6) | 562 (63.4) | 245 (72.1) | 30 (50.8) | <0.01 |
| Statins | 2515 (94.3) | 1317 (95.4) | 820 (92.3) | 329 (96.8) | 49 (83.1) | <0.01 |
Continuous variables are expressed as mean±SD or median with IQR, as appropriate. Categorical variables are presented as number (percentage). All statistical calculations were performed using available data.
ACEi, ACE inhibitor; ARB, angiotensin receptor blocker; BNP, B-type natriuretic peptide; BP, blood pressure; BSA, body surface area; CAD, coronary artery disease; CK, creatine kinase; LDL-C, low-density lipoprotein cholesterol; PCI, percutaneous coronary intervention; STEMI, ST-segment elevation myocardial infarction.
Compared with patients with lower cardiac grading levels, those with higher cardiac grading levels were older and more likely to have diabetes. Compared with Grade 1–3 patients, patients with Grade 4 had lower systolic blood pressure, higher heart rate and stepwise increase in higher Killip levels. In addition, the prevalence of anterior myocardial infarction was highest among patients with Grade 4. Patients with higher cardiac grades had a stepwise increase in baseline serum creatinine levels and B-type natriuretic peptide (BNP) on laboratory findings (table 1).
At the time of discharge, patients with higher cardiac grading levels were less likely to receive standard guideline-recommended therapies during hospitalisation, including antiplatelets, ACE inhibitor (ACEi)/angiotensin receptor blocker (ARB), beta-blockers and statins (table 1).
Echo characteristics
Patients with higher cardiac grading levels had lower EF as determined by echocardiography (Grade 1–4, 60%, 47%, 45% and 29%, p<0.01), larger LVEDV/BSA (Grade 1–4, 59.15 mL/m2, 58.37 mL/m2, 81.96 mL/ m2 and 82.79 mL/ m2, p<0.01). The proportion of patients with medium or severe MR increased with higher grading levels (online supplemental appendix table 1).
Observed long-term mortality among patients with different cardiac grades
The median follow-up duration was 5.5 years (IQR, 5.3–6.4 years). We examined the long-term mortality of all causes among different cardiac grades, which was compared across differently graded patients. Long-term mortality increased incrementally from Grade 1 to Grade 4 (2.4%, 2.7%, 4.6% and 13.1%, respectively, p<0.01 by Fisher’s exact test; figure 2B and table 2). A similar trend was observed in 90-day restratification with stepwise increase in mortality from Grade 1 to Grade 4 (1.7%, 1.3%, 3.2%, 8.6%, p=0.032 by Fisher’s exact test; Figure 3B and table 3).
Table 2. Logistic regression model of long-term mortality according to post-STEMI cardiac grading at admission.
| Main interest | Long-term mortality | Unadjusted OR (95% CI) | Adjusted OR* (95% CI) | |
|---|---|---|---|---|
| Grade on admission | Grade 1 | 2.4% (33/1392) | / | / |
| Grade 2 | 2.7% (24/905) | 1.12 (0.65 to 1.90) | 1.10 (0.61 to 1.98) | |
| Grade 3 | 4.6% (16/350) | 1.97 (1.05 to 5.57) | 1.43 (0.68 to 2.85) | |
| Grade 4 | 13.1% (8/61) | 6.22 (2.57 to 13.51) | 3.67 (1.27 to 9.17) | |
Adjusted for age, sex, Killip class, reperfusion therapy via percutaneous coronary intervention, anterior infarction, B-type natriuretic peptide levels and creatinine levels.
STEMI, ST-segment elevation myocardial infarction.
Figure 3. Cardiac risk grading and long-term mortality at 90-day follow-up. (A) Distribution of cardiac risk grades at 90-day follow-up. Colours represent Grades 1–4, labelled on the right; the y-axis indicates the proportion of patients in each grade. (B) Long-term mortality rates according to cardiac risk grade at 90-day follow-up. Colours represent Grades 1–4, labelled on the right; the y-axis indicates the percentage of deaths. (C) Kaplan-Meier analysis of long-term survival by cardiac grade at 90-day follow-up. Survival curves show progressively lower survival rates from Grades 1 to 4. In multivariable Cox proportional hazards analysis, Grade 4 was associated with the highest risk of all-cause mortality compared with Grade 1 (adjusted HR=6.22; 95% CI 1.69 to 22.82; p=0.0066).
Table 3. Logistic regression model of long-term mortality according to post-STEMI cardiac grading at 90 days.
| Main interest | Long-term mortality | Unadjusted OR (95% CI) | Adjusted OR* (95% CI) | |
|---|---|---|---|---|
| Grade on 90 days | Grade 1 | 1.7% (16/969) | / | / |
| Grade 2 | 1.3% (3/227) | 0.80 (0.18 to 2.42) | 0.63 (0.14 to 2.08) | |
| Grade 3 | 3.2% (6/189) | 1.95 (0.69 to 4.82) | 2.62 (0.88 to 6.94) | |
| Grade 4 | 8.6% (3/35) | 5.58 (1.25 to 17.84) | 5.95 (1.23 to 21.95) | |
Adjusted for age, sex, Killip class, reperfusion therapy via percutaneous coronary intervention, anterior infarction, B-type natriuretic peptide levels and creatinine levels.
STEMI, ST-segment elevation myocardial infarction.;
Long-term mortality model
To evaluate whether the proposed cardiac grading system is predictive of mortality, we performed further analyses on its association with long-term outcomes. The median follow-up time is 5.5 years. Compared with Grade 1, Grade 2 had similar long-term mortality risks (adjusted OR=1.10, 95% CI 0.61 to 1.98). Grade 3 was associated with increased long-term mortality, although the difference did not reach statistical significance (adjusted OR=1.43, 95% CI 0.68 to 2.85, p=0.32). Grade 4 showed a significantly higher mortality risk (adjusted OR=3.67, 95% CI 1.27 to 9.17, p<0.01) (table 2).
Kaplan-Meier survival analysis showed significant differences in long-term survival across the different cardiac grades. In the multivariable Cox regression model, compared with Grade 1, Grade 4 showed a significantly higher risk (adjusted HR=3.35, 95% CI 1.34 to 8.38, p<0.01; figure 2C).
Subgroup analysis of patients with 90-day follow-up echocardiography
Among 1420 patients who underwent 90-day follow-up echocardiographic examinations, 969 were classified as Grade 1 (68.2%), 227 as Grade 2 (16%), 189 as Grade 3 (13.3%) and 35 as Grade 4 (2.5%) (figure 3A). Of these, 141 (9.9%) patients showed a worsening in cardiac grade, defined as an increase of ≥1 grade from the initial assessment at admission (figure 4A). Notably, only one patient initially classified as Grade 1 progressed to Grade 4 within 90 days (figure 4B).
Figure 4. Changes in cardiac risk grading from hospitalisation to 90-day follow-up. (A) Proportions of patients whose cardiac risk grade improved, remained stable or worsened between baseline and 90-day follow-up. (B) Distribution of 90-day cardiac grades (Grades 1–4) stratified by baseline cardiac grade.
Compared with those who did not undergo 90-day echocardiographic examinations, patients with 90-day follow-up had similar baseline characteristics in terms of gender, comorbidities, peak troponin levels and most echocardiographic parameters. However, they were older, had higher creatine kinase levels, lower EF during hospitalisation and were more likely to receive optimal medical therapy (online supplemental appendix table 2).
90-day cardiac grading and long-term mortality
Among patients who underwent 90-day echocardiographic follow-up, the observed mortality rates were 1.7%, 1.3%, 3.2% and 8.6% across Grades 1–4, respectively (figure 3B and table 3). After adjusting for age, gender, Killip class, reperfusion therapy via percutaneous coronary intervention, anterior infarction, BNP levels and creatinine levels, Grade 4 at 90 days was independently associated with long-term mortality (adjusted OR=5.95, 95% CI 1.23 to 21.95, p<0.01; table 3). Similar results were observed in a multivariable Cox regression model, where Grade 4 had a significantly higher risk of all-cause mortality compared with those with Grade 1 (adjusted HR=6.22, 95% CI 1.69 to 22.82, p<0.01; figure 3C).
Discussion
Among contemporary first-time patients with STEMI, we demonstrated that an echocardiographic grading system was significantly associated with long-term all-cause mortality in a large, multicentre real-world cohort. Extending our previous 90-day study,4 we found that patients with Grade 1–4 had stepwise increasing long-term mortality. After multivariable adjustment, Grade 4 patients had four times the adjusted mortality risk of Grade 1. The mortality difference persisted using repeated cardiac grading 90 days post-STEMI. Beyond its association with clinical outcomes, the proposed grading system is intended to provide a structured description of myocardial injury severity after STEMI using routinely available echocardiographic findings. By integrating functional impairment and structural involvement, the grading system helps characterise disease stage and its evolution over time.
The strong association between Grade 4 and all-cause mortality observed in our study is likely explained by severely impaired systolic function (EF <30%) and extensive cardiac remodelling, as evidenced by enlarged left ventricle and significant MR. EF below 30% has consistently been associated with increased mortality risk and is incorporated in major myocardial infarction guidelines as a high-risk marker.8 Likewise, postinfarction remodelling, particularly left ventricular dilation,9 has been shown to predict long-term outcomes and medium-to-severe MR has emerged as an independent predictor of mortality in post-STEMI patients.10
From a broader clinical perspective, patients with cardiogenic shock or those requiring mechanical circulatory support represent a well-recognised high-risk subgroup after STEMI and have been consistently associated with markedly increased in-hospital mortality in prior studies.11,13 Although these investigations primarily focus on acute-phase outcomes, they underscore the prognostic importance of severe myocardial and systemic derangement during STEMI. In this context, Grade 4 identified in our study reflects pronounced structural and functional myocardial impairment, which may represent a downstream manifestation of similar high-risk pathophysiological processes and is associated with adverse long-term mortality among patients who survive the index hospitalisation.
Left ventricular remodelling is generally considered a long-lasting process that begins after myocardial infarction. However, accumulating evidence suggests that remodelling may develop very early following myocardial injury. Korup et al14 demonstrated that left ventricular dilatation can be detected as early as 3 hours after acute myocardial infarction and tends to stabilise during the initial 6-day period. More recently, prospective echocardiographic studies have further highlighted the prognostic significance of early ventricular dilatation. Early changes in LVEDV after STEMI have been shown to predict subsequent left ventricular remodelling during follow-up9 . Early structural dilatation may in turn lead to functional consequences, including papillary muscle displacement and annular dilatation, ultimately resulting in MR. Consistent with this pathophysiological framework, MR detected within 48 hours after myocardial infarction has been identified as an independent predictor of all-cause mortality.10 Together, these observations suggest that echocardiographic abnormalities detected at index admission reflect an integrated process of early left ventricular remodelling, involving both structural dilatation and its functional sequelae, and therefore offer a feasible opportunity for risk stratification. Among available measures reflecting ventricular structure, LVEDV is a simple, routinely measured and readily available marker of ventricular dilatation, making it particularly practical for assessing early remodelling in real-world clinical settings.
To minimise the influence of baseline left ventricular dilation and MR, which may partly result from reversible conditions such as myocardial stunning, leaflet tethering or transient ischaemia, we repeated the analysis using echocardiographic data obtained at 90 days after myocardial infarction. The consistency of the results supports the robustness and validity of the proposed risk stratification strategy. To further test the robustness of our findings, we performed a sensitivity analysis using an alternative echocardiography-based grading system derived from dilated left ventricular end-systolic volume indexed to BSA (online supplemental appendix tables 3–5). The association between higher grades and long-term mortality remained consistent, supporting the stability and clinical relevance of the proposed grading strategy.
In contrast to Grade 4, which had the highest mortality, Grades 1 through 3 showed a non-significant stepwise difference in long-term mortality. This may be explained by several factors: First, timely reperfusion therapy and optimised postdischarge medical treatment may have contributed to improved survival in patients without extensive myocardial necrosis or remodelling.2 Second, the persistently high mortality in Grade 4 patients could be partly explained by the treatment-risk paradox, whereby patients with the highest clinical risk are less likely to receive or tolerate guideline-recommended therapies.15 Severe haemodynamic instability or organ dysfunction in this group may limit the use of ACEi/ARB, beta-blockers or statins during hospitalisation, which could contribute to their unfavourable outcomes. Third, patients in Grade 1–3 typically exhibit only moderate degrees of ventricular remodelling or functional impairment, which may be insufficient to result in all-cause mortality during the follow-up period. However, these patients may still experience other MACE, such as heart failure or recurrent myocardial infarction.16 Additionally, it is possible that other echocardiographic indicators not included in the current model may better capture the heterogeneity of moderate remodelling.7 It would be expected that a statistically significant stepwise outcome would be observed between Grade 1 and Grade 3 in future studies.
Our observations highlight that myocardial injury and repair following infarction are a dynamic process, rather than a static state. The postinfarction heart undergoes continuous morphological and functional changes, which may evolve significantly over time. In this context, echocardiography, which is a practical and widely accessible imaging modality compared with CMR, serves as an ideal tool to capture these ongoing changes. In light of this, our grading system offers a feasible approach not only for early postinfarction risk stratification but also for tracking structural progression and supporting individualised treatment strategies during follow-up.
Our study showed that, at the index hospitalisation, patients were distributed across Grades 1–4 in progressively decreasing proportions, reflecting real-world observations in the contemporary era of early reperfusion and intensive secondary prevention.17 Most patients sustained only minor myocardial injury, whereas a smaller proportion experienced long-term adverse outcomes. From a clinical perspective, this graded distribution may help inform post-discharge management strategies. Patients classified as Grade 1, who exhibited no apparent structural abnormalities and remained stable during follow-up, may require less intensive surveillance, allowing rational de-escalation of follow-up frequency. In contrast, patients in Grades 2 and 3, despite having intermediate risk, demonstrated a higher likelihood of progression to more advanced disease, suggesting a potential role for closer monitoring, repeat imaging and early optimisation of medical therapy. Notably, Grade 4 patients represented a small but particularly vulnerable subgroup with the highest long-term mortality risk. Identification of this group may prompt intensified follow-up, early referral to specialised heart failure services and timely use of advanced therapies and rehabilitation resources. Importantly, the observation that some Grade 4 patients showed improvement at 90 days suggests that this high-risk state is not invariably irreversible, highlighting a potential window for timely intervention. Further studies are needed to evaluate the relationship between this grading system and heart failure outcomes as well as cardiovascular mortality. Taken together, these findings suggest that the proposed grading system may serve not only as a prognostic tool, but also as a practical framework for clinical communication, follow-up planning and longitudinal assessment of myocardial injury and recovery after STEMI.
Limitations
This study has several limitations. First, as a registry-based observational study, only patients who survived to hospital discharge and had available echocardiographic data were included, which may introduce survivor bias. In addition, although multivariable adjustment was performed, residual confounding inherent to observational studies cannot be fully excluded. Furthermore, due to the real-world, multicentre design, some variability in echocardiographic and laboratory measurements across centres was unavoidable and formal interobserver and intraobserver reproducibility analyses were not performed. Second, this study focused exclusively on all-cause mortality. Other clinically relevant outcomes, such as heart failure or recurrent ischaemia, were not evaluated and may differ across cardiac grades. Future studies are warranted to address these additional endpoints. Third, the rate of 90-day echocardiographic follow-up was limited. This is a common challenge in real-world observational studies,16 18 and the follow-up rate in our cohort was comparable to that reported in previous studies. To address potential bias, we performed sensitivity analyses and found that most baseline clinical characteristics were similar between patients with and without 90-day echocardiographic follow-up (online supplemental appendix table 2), suggesting that the impact of incomplete follow-up on our results is likely limited. Finally, most participating centres were located in Shanghai, which may limit generalisability; however, patient characteristics were consistent with national registry data.19 20
Conclusion
Our echocardiography-based grading system provides an accessible and objective framework for long-term risk stratification after first-time STEMI. By integrating routinely available echocardiographic markers of functional impairment and structural involvement, the grading system describes the stage and extent of post-infarction myocardial injury. Patients classified as Grade 4 exhibited substantially higher long-term mortality, identifying a subgroup that may benefit from closer surveillance and more intensive management. Beyond outcome prediction, this approach offers a practical and intuitive way to support longitudinal assessment, clinical communication and follow-up planning across different care settings.
Supplementary material
Footnotes
Funding: The clinical trial was supported by a research grant from the Shanghai ShenKang Hospital Development Center, Grant No. SHDC2020CR1039B and Shanghai Arrhythmia Research Center Project (Grant No. 2022ZZ01008), the National Natural Science Foundation of China (Grant No. 81900308 to LS) and National Natural Science Foundation of China (Grant No. 82130012).
Prepub: Prepublication history and additional supplemental material for this paper are available online. To view these files, please visit the journal online (https://doi.org/10.1136/bmjopen-2025-111116).
Provenance and peer review: Not commissioned; externally peer reviewed.
Patient consent for publication: Not applicable.
Ethics approval: This study involves human participants and was approved by the Ethics Committee of Shanghai Chest Hospital (in December 2015; approved number 2015-111). Participants gave informed consent to participate in the study before taking part.
Patient and public involvement: Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.
Data availability statement
No data are available.
References
- 1.Vogel B, Claessen BE, Arnold SV, et al. ST-segment elevation myocardial infarction. Nat Rev Dis Primers. 2019;5:39. doi: 10.1038/s41572-019-0090-3. [DOI] [PubMed] [Google Scholar]
- 2.Maleki ND, Van de Werf F, Goldstein P, et al. Aborted myocardial infarction in ST-elevation myocardial infarction: insights from the STrategic Reperfusion Early After Myocardial infarction trial. Heart. 2014;100:1543–9. doi: 10.1136/heartjnl-2014-306023. [DOI] [PubMed] [Google Scholar]
- 3.Kumar A, Connelly K, Vora K, et al. The Canadian Cardiovascular Society Classification of Acute Atherothrombotic Myocardial Infarction Based on Stages of Tissue Injury Severity: An Expert Consensus Statement. Can J Cardiol. 2024;40:1–14. doi: 10.1016/j.cjca.2023.09.020. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.He B, Ge H, Yang F, et al. A Novel Method in the Stratification of Post-Myocardial-Infarction Patients Based on Pathophysiology. PLoS ONE. 2015;10:e0130158. doi: 10.1371/journal.pone.0130158. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Byrne RA, Rossello X, Coughlan J. 2023 ESC Guidelines for the management of acute coronary syndromes. Eur Heart J. 2023;44:3720–826. doi: 10.1093/eurheartj/ehad191. [DOI] [PubMed] [Google Scholar]
- 6.Zoghbi WA, Adams D, Bonow RO, et al. Recommendations for Noninvasive Evaluation of Native Valvular Regurgitation: A Report from the American Society of Echocardiography Developed in Collaboration with the Society for Cardiovascular Magnetic Resonance. J Am Soc Echocardiogr. 2017;30:303–71. doi: 10.1016/j.echo.2017.01.007. [DOI] [PubMed] [Google Scholar]
- 7.Lang RM, Badano LP, Mor-Avi V, et al. Recommendations for cardiac chamber quantification by echocardiography in adults: an update from the American Society of Echocardiography and the European Association of Cardiovascular Imaging. J Am Soc Echocardiogr. 2015;28:1–39. doi: 10.1016/j.echo.2014.10.003. [DOI] [PubMed] [Google Scholar]
- 8.Solomon SD, Zelenkofske S, McMurray JJV, et al. Sudden death in patients with myocardial infarction and left ventricular dysfunction, heart failure, or both. N Engl J Med. 2005;352:2581–8. doi: 10.1056/NEJMoa043938. [DOI] [PubMed] [Google Scholar]
- 9.Yi L, Zhu T, Qu X, et al. Predictive value of early left ventricular end-diastolic volume changes for late left ventricular remodeling after ST-elevation myocardial infarction. Cardiol J. 2024;31:451–60. doi: 10.5603/CJ.99492. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Feinberg MS, Schwammenthal E, Shlizerman L, et al. Prognostic significance of mild mitral regurgitation by color Doppler echocardiography in acute myocardial infarction. Am J Cardiol. 2000;86:903–7. doi: 10.1016/s0002-9149(00)01119-x. [DOI] [PubMed] [Google Scholar]
- 11.Çinar T, Hayiroğlu Mİ, Şeker M, et al. The predictive value of age, creatinine, ejection fraction score for in-hospital mortality in patients with cardiogenic shock. Coron Artery Dis. 2019;30:569–74. doi: 10.1097/MCA.0000000000000776. [DOI] [PubMed] [Google Scholar]
- 12.Hayıroğlu Mİ, Keskin M, Uzun AO, et al. Predictors of In-Hospital Mortality in Patients With ST-Segment Elevation Myocardial Infarction Complicated With Cardiogenic Shock. Heart, Lung and Circulation . 2019;28:237–44. doi: 10.1016/j.hlc.2017.10.023. [DOI] [PubMed] [Google Scholar]
- 13.Tokarek T, Dziewierz A, Plens K, et al. Radial approach reduces mortality in patients with ST-segment elevation myocardial infarction and cardiogenic shock. Pol Arch Intern Med . 2021;131:421–8. doi: 10.20452/pamw.15886. [DOI] [PubMed] [Google Scholar]
- 14.Korup E, Dalsgaard D, Nyvad O, et al. Comparison of Degrees of Left Ventricular Dilation Within Three Hours and Up to Six Days After Onset of First Acute Myocardial Infarction. Am J Cardiol. 1997;80:449–53. doi: 10.1016/S0002-9149(97)00393-7. [DOI] [PubMed] [Google Scholar]
- 15.Roffi M, Mukherjee D. Treatment-risk paradox in acute coronary syndromes. Eur Heart J. 2018;39:3807–9. doi: 10.1093/eurheartj/ehy577. [DOI] [PubMed] [Google Scholar]
- 16.Butler J, Hammonds K, Talha KM, et al. Incident heart failure and recurrent coronary events following acute myocardial infarction. Eur Heart J. 2025;46:1540–50. doi: 10.1093/eurheartj/ehae885. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Rao SV, O’Donoghue ML, Ruel M, et al. 2025 ACC/AHA/ACEP/NAEMSP/SCAI Guideline for the Management of Patients With Acute Coronary Syndromes: A Report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines. Circulation. 2025;151:e771–862. doi: 10.1161/CIR.0000000000001309. [DOI] [PubMed] [Google Scholar]
- 18.Wu WY, Biery DW, Singh A, et al. Recovery of Left Ventricular Systolic Function and Clinical Outcomes in Young Adults With Myocardial Infarction. J Am Coll Cardiol. 2020;75:2804–15. doi: 10.1016/j.jacc.2020.03.074. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Szummer K, Wallentin L, Lindhagen L, et al. Improved outcomes in patients with ST-elevation myocardial infarction during the last 20 years are related to implementation of evidence-based treatments: experiences from the SWEDEHEART registry 1995-2014. Eur Heart J. 2017;38:3056–65. doi: 10.1093/eurheartj/ehx515. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Li S, Gao X, Yang J, et al. Number of standard modifiable risk factors and mortality in patients with first-presentation ST-segment elevation myocardial infarction: insights from China Acute Myocardial Infarction registry. BMC Med. 2022;20:217. doi: 10.1186/s12916-022-02418-w. [DOI] [PMC free article] [PubMed] [Google Scholar]