Abstract
Background
Type 2 myocardial infarction (T2MI) accounts for a substantial share of acute coronary syndromes but remains challenging to diagnose and manage due to its varied presentations and underlying profiles. This study aims to identify key differences and distinct clinical phenotypes in a large T2MI population.
Methods
All consecutive patients with non–ST‐segment–elevation myocardial infarction undergoing coronary angiography with a confirmed T2MI diagnosis between January 1, 2017, and March 31, 2023, were analyzed. Precipitating factors of supply–demand mismatch were identified, and coronary burden was assessed using the Gensini score. Latent class analysis was used to identify clinical phenotypes, and multivariable analyses were performed to determine prognostic predictors. A composite of major adverse cardiovascular events was assessed during follow‐up, along with additional outcomes including cardiovascular death and nonfatal type 2 reinfarction.
Results
Among 774 patients with T2MI, latent class analysis identified 2 phenotypes. Phenotype 1 (31.5%) was younger with a higher prevalence of nonatherosclerotic coronary causes and unknown pathogeneses. Phenotype 2 (68.5%) exhibited greater comorbidity and a higher atherosclerotic burden, reflected by elevated Gensini scores (median, 11 versus 1.5; P<0.001). Over a median follow‐up of 53 months, major adverse cardiovascular events occurred in 49.1% of patients, with a higher rate in phenotype 2 (60.8% versus 23.8%, P<0.001). Predictors of major adverse cardiovascular events included peak cardiac troponin levels for phenotype 1 and age, known cardiovascular disease, chronic obstructive pulmonary disease, peak cardiac troponin levels, and Gensini score for phenotype 2.
Conclusions
This study identified 2 clinical phenotypes in T2MI, highlighting differences in characteristics, precipitating factors, outcomes, and prognostic predictors, emphasizing the potential for phenotype‐driven approaches in diagnosis and management.
Keywords: cardiovascular outcomes, non–ST‐segment–elevation myocardial infarction, type 2 myocardial infarction
Subject Categories: Myocardial Infarction, Coronary Circulation
Nonstandard Abbreviations and Acronyms
- AMIPE
Acute Myocardial Infarction, Prognostic and Therapeutic Evaluation
- cTn
cardiac troponin
- LCA
latent class analysis
- MACE
major adverse cardiovascular event
- P1 T2MI
phenotype 1 type 2 myocardial infarction
- P2 T2MI
phenotype 2 type 2 myocardial infarction
- T1MI
type 1 myocardial infarction
- T2MI
type 2 myocardial infarction
Clinical Perspective.
What Is New?
This is the first large real‐world study to empirically identify 2 distinct clinical phenotypes among patients with type 2 myocardial infarction, based on routine clinical variables.
What Are the Clinical Implications?
The 2 type 2 myocardial infarction phenotypes, 1 younger and healthier and the other older with greater comorbidity and coronary atherosclerotic burden, demonstrate significantly different clinical characteristics, precipitating factors, long‐term outcomes, and prognostic factors.
Despite similar management strategies derived from type 1 myocardial infarction protocols, phenotype‐specific differences suggest that a tailored therapeutic approach based on clinical phenotyping may improve patient outcomes.
Type 2 myocardial infarction (T2MI) constitutes a significant portion of acute coronary syndromes, accounting for up to three‐quarters of all myocardial infarctions in the United States. 1 As outlined in the Fourth Universal Definition of Myocardial Infarction, T2MI results from an imbalance between oxygen supply and demand, occurring in the absence of acute coronary plaque rupture or thrombosis. 2 Although the pathophysiological framework of T2MI is well established, its diagnosis and management remain complex. While distinguishing T2MI from type 1 myocardial infarction (T1MI) is generally feasible when standard definitions are applied, it may still be challenging in the presence of atypical symptoms and absence of ischemic ECG changes. 3 , 4 , 5 , 6 Furthermore, most research and clinical guidelines have focused on T1MI, often neglecting T2MI. As a result, clinical cardiologists frequently struggle with its management due to limited information on therapy and prognosis. 7 Simply applying T1MI management strategies to T2MI is inadequate because of its distinct pathogenetic mechanisms and the lack of supporting evidence. 8
Importantly, T2MI is not a benign condition, as it is associated with poor outcomes. 3 , 9 These outcomes are often comparable with, or worse than, those of T1MI, but they vary widely, making disease progression difficult to predict. 10 , 11 , 12 , 13
Historically, research has primarily focused on the differences between T1MI and T2MI, with limited attention given to the clinical heterogeneity within T2MI itself. Recently, some researchers have proposed the use of clinical phenotypes to better understand the variability of T2MI. 1 , 14
Clinical experience suggests that patients with T2MI often exhibit distinct phenotypic patterns, which may vary according to age, cardiovascular risk factors, precipitating factors, and comorbidities. However, this remains an unverified hypothesis. 15
The aim of this study is to provide clinical evidence of phenotypic differentiation in a large, real‐world cohort of patients with non–ST‐segment–elevation myocardial infarction with T2MI, identify key differences between these clinical phenotypes, and assess whether these phenotypes are associated with varying prognoses and distinct prognostic factors.
METHODS
Study Design and Population
The present study is a prespecified subanalysis of data from the ongoing multicenter observational prospective AMIPE (Acute Myocardial Infarction, Prognostic and Therapeutic Evaluation) registry (NCT03883711), which aims to evaluate the characteristics and outcomes of patients admitted with diagnosis of acute myocardial infarction (AMI).
All consecutive patients with non–ST‐segment–elevation myocardial infarction who underwent coronary angiography (CAG) with a confirmed diagnosis of T2MI from January 1, 2017, to March 31, 2023, and who had at least 1‐year of follow‐up, were included in this study. Exclusion criteria were (1) other AMI subtypes, (2) nonischemic myocardial injury, (3) lack of informed consent, and (4) patients aged <18 years.
The protocol was approved by the institutional review board (Registration No. 600/2018/Oss/AOUBo). This study was conducted in accordance with the Declaration of Helsinki. All patients were informed about their participation in the registry and provided informed consent for the anonymous publication of scientific data.
The data that support the findings of this study are available from the corresponding author upon reasonable request.
Data Collection
Demographic data, baseline clinical characteristics, and information regarding admission, hospital stay, and discharge were prospectively collected for every patient. Upon admission, blood samples were obtained for standard analyses, including complete blood count, C‐reactive protein, electrolytes, creatinine, lipid profile, glucose, and cardiac troponin (cTn) levels. All patients underwent serial ECGs and at least 1 transthoracic echocardiogram during hospitalization. Additional ECG and transthoracic echocardiogram assessments were performed on the basis of clinical indication. In patients with myocardial infarction and nonobstructive coronary arteries without an evident underlying cause, cardiac magnetic resonance was systematically performed to exclude nonischemic mechanisms underlying myocardial injury, as recommended by current guidelines. 8 , 16 , 17 Both admission and discharge medications were recorded, including antiplatelet agents, β blockers, renin–angiotensin–aldosterone system inhibitors, statins, and anticoagulant therapy.
Invasive CAG was performed during the index hospitalization for all patients. Obstructive coronary artery disease (CAD) was defined as the presence of a stenosis with a diameter of ≥50% in at least 1 major epicardial vessel, in accordance with current guidelines. 2 Intracoronary physiology and intravascular imaging were performed if indicated. 8 The severity of coronary atherosclerosis was assessed using the Gensini score. 18
T2MI Definition and Adjudication
Diagnosis of T2MI was independently assessed by 2 clinical cardiologists (D.B. and M.A.) in all patients on the basis of all clinical and instrumental data collected during the hospitalization. In case of disagreement, a third experienced cardiologist (C.P.) was consulted to reach a consensus. According to the Fourth Universal Definition of Myocardial Infarction, patients exhibiting signs or symptoms of myocardial ischemia, where myocardial injury occurred due to an imbalance between oxygen supply and without evidence of acute atherothrombotic plaque disruption, were diagnosed with T2MI. 2 Patients with elevated cTn levels but without signs or symptoms of myocardial ischemia were classified as having nonischemic myocardial injury.
For all patients with a definitive diagnosis of T2MI, precipitating factors for the supply–demand mismatch were identified and treated: sepsis, severe anemia or relevant recent bleeding, tachyarrhythmia, bradyarrhythmia, respiratory failure, hypertension, hypotension/shock, recent noncardiac surgery, acute kidney injury, and nonatherosclerotic coronary artery lesions such as vasospasm, embolism, myocardial bridging, or dissection. While some authors have proposed classifying patients with myocardial infarction due to reduced coronary blood flow from nonatherothrombotic causes separately, 19 others support their inclusion within T2MI according to the Fourth Universal Definition of Myocardial Infarction. 20 In line with this latter view, and given our study objectives, such cases were included in the present analysis. For patients in whom ≥2 causes of T2MI were present, these were categorized as multiple contributing factors. Detailed criteria for each causative factor can be found in the Supplemental Material (Supplemental Methods).
Follow‐Up and Outcomes
Patients were followed from the time of discharge through outpatient visits or telephone contacts using a standard questionnaire. The follow‐up was completed on April 30, 2024. A composite of major adverse cardiovascular events (MACEs), including all‐cause death, nonfatal reinfarction, unplanned revascularization, nonfatal ischemic stroke, and hospitalization for heart failure, was assessed, considering only the first event for calculation. Additional outcomes included cardiovascular death and nonfatal type 2 reinfarction. Further details on outcome definitions are provided in the Supplemental Material (Supplemental Methods).
All follow‐up events were independently adjudicated by 2 clinical cardiologists (A.S. and F.A.) who were blinded to the clinical, laboratory, and angiographic data. In the event of disagreement, a third experienced cardiologist (L.B.) was brought in to achieve consensus.
Statistical Analysis
Continuous variables were summarized as mean±SD or as median and interquartile range, and categorical variables were presented as absolute and relative frequencies. Between‐group differences in continuous variables were tested using the independent‐sample t test or Mann–Whitney U test, as appropriate. Differences in categorical variables were tested using the χ2 test.
We used latent class analysis (LCA) to identify homogeneous clinical phenotypes in the study population. LCA is a statistical method used to identify hidden subgroups or “latent classes” within a population based on patterns of responses to a set of variables. These latent classes are not directly observed but are inferred from the data. LCA is particularly useful for analyzing categorical or ordinal data and can help uncover underlying structures or patterns that might not be apparent from direct observation. 21 Since its first definition in 1996, 22 the use of LCA has increased significantly across various medical disciplines. 23 , 24 , 25
The analysis involves the following steps: selecting the variables to include in the model and justifying their selection and fitting the model to the data. Multiple models, each consisting of k‐classes, are fitted to the data. Typically, the first model consists of a single class (k=1), and sequential models, each adding 1 more class, are fit, selecting the optimal number of classes. The general approach is to choose the model with the fewest classes that best fit the data. The separation of classes should have clinical significance.
In this study, we used age as a continuous covariate along with dichotomous risk factors (current/past smoking, hypercholesterolemia, diabetes, hypertension, family history of CAD, known cardiovascular disease, chronic obstructive pulmonary disease [COPD], and chronic kidney disease) as input of the model. These clinical covariates were selected a priori to include commonly available variables in routine practice, while also taking into account their known associations with adverse outcomes in T2MI.
Then, we first estimated a 1‐class model and then added classes until we identified the best fit model. We examined model fit on the basis of our clinical understanding of the possible number of clinical phenotypes of the disease, on parsimony, ease of interpretation, and the following statistical criteria: (1) the Bayesian information criterion, with lower values indicating better model fit 26 ; (2) the smallest class size, with higher values indicating better fit; and (3) the average latent class posterior probability, with higher values denoting better fit.
After identifying the best model, we assigned each case to a specific clinical phenotype based on its maximum posterior class membership probability.
Outcomes were estimated using Kaplan–Meier curves and compared among the study phenotypes using the log‐rank test. For stratified comparisons based on the presence of obstructive CAD, adjustment for multiple comparisons was performed using the Bonferroni correction. As a sensitivity analysis, we repeated the long‐term outcome comparison after excluding patients with coronary artery spasm, embolism, or dissection to assess the robustness of the prognostic findings across clinical phenotypes.
Cox regression models were used to determine the independent predictors of MACEs in each T2MI clinical phenotype. Variables with a significance level of 0.1 in univariate models were included in multivariable models. For the analysis, we selected the main known predictors of outcome in patients with T2MI, specifically, age; sex; hypertension; dyslipidemia; diabetes; history of cardiovascular disease; COPD; estimated glomerular filtration rate; hemoglobin levels; left ventricular ejection fraction; peak cTn values; coronary atherosclerotic burden as evaluated by the Gensini score; and secondary prevention therapy, including aspirin, dual antiplatelet therapy, statins, β blockers, and angiotensin‐converting enzyme inhibitors or angiotensin receptor blockers. The proportional hazards assumption for the Cox model was tested using Schoenfeld residuals, both globally and for each covariate included in the multivariable models. Influential observations were detected using DFBETA residuals. Linearity in the relationship between continuous covariates and the log hazard was assessed using deviance residuals.
All the statistical analyses were performed using SPSS Statistics version 28.0.1.1 (IBM, Armonk, NY) and Stata version 17 (StataCorp, College Station, TX). The significance level was set to P<0.05.
RESULTS
Study Population
A total of 2633 patients hospitalized for non–ST‐segment–elevation myocardial infarction who underwent CAG were potentially eligible for the study. Among these, 1677 patients were excluded due to a clinical presentation consistent with AMI subtypes other than T2MI, 166 patients due to a nonischemic mechanism of myocardial damage, and 16 patients due to incomplete 1‐year follow‐up.
The final sample consisted of 774 patients with confirmed T2MI and complete 1‐year follow‐up data.
Baseline characteristics of the overall cohort are provided in Tables 1 and 2. The study population included predominantly older people with a mean age of 70.2±12.7 years, with an even sex distribution (47.9% were women), and a high prevalence of cardiovascular risk factors, particularly hypertension. Many had existing cardiovascular diseases, and a significant proportion were treated with aspirin. Most patients had identifiable causes for T2MI, with tachyarrhythmias and hypertension being the most common. CAG revealed that more than half (55.0%) had obstructive CAD, with a smaller group (6.5%) showing nonatherosclerotic coronary lesions. The median Gensini score was 7 (interquartile range, 0–34.75); 34.6% of patients underwent revascularization, predominantly via percutaneous coronary intervention, and discharge medications commonly included aspirin, statins, β blockers, and renin–angiotensin–aldosterone system inhibitors.
Table 1.
Baseline Characteristics of the Overall Cohort Study and Clinical Phenotypes Identified Using Latent Class Analysis
| Total | P1 T2MI | P2 T2MI | P value | |
|---|---|---|---|---|
| N=774 | N=244 | N=530 | ||
| Age,* y | 70.2±12.7 | 57.1±10.6 | 76.2±8.2 | <0.001 |
| Female sex | 371 (47.9) | 101 (41.4) | 270 (50.9) | 0.013 |
| Body mass index, kg/m2 | 26.6±4.8 | 26.4±4.6 | 26.6±4.9 | 0.704 |
| Cardiovascular risk factors | ||||
| Current/past smoking* | 405 (52.3) | 148 (60.7) | 257 (48.5) | 0.002 |
| Hypertension* | 590 (76.2) | 101 (41.4) | 489 (92.3) | <0.001 |
| Dyslipidemia* | 492 (63.6) | 111 (45.5) | 381 (71.9) | <0.001 |
| Type 2 diabetes* | 193 (24.9) | 20 (8.2) | 173 (32.6) | <0.001 |
| Family history of ischemic heart disease* | 125 (16.1) | 85 (34.8) | 40 (7.5) | <0.001 |
| Medical history | ||||
| Known cardiovascular disease* | 278 (35.9) | 35 (14.3) | 243 (45.8) | <0.001 |
| Previous myocardial infarction | 185 (23.9) | 26 (10.7) | 159 (30.0) | <0.001 |
| Previous PCI | 142 (18.3) | 21 (8.6) | 121 (22.8) | <0.001 |
| Previous CABG | 39 (5.0) | 1 (0.4) | 38 (7.2) | <0.001 |
| Previous TIA/stroke | 50 (6.5) | 5 (2.0) | 45 (8.5) | <0.001 |
| PAD | 59 (7.6) | 4 (1.6) | 55 (10.4) | <0.001 |
| COPD* | 118 (15.2) | 9 (3.7) | 109 (20.6) | <0.001 |
| Chronic kidney disease* | 253 (32.7) | 9 (3.7) | 244 (46.0) | <0.001 |
| Medical therapy at admission | ||||
| Aspirin | 311 (40.2) | 46 (18.9) | 265 (50.0) | <0.001 |
| P2Y12 inhibitor | 73 (9.4) | 9 (3.7) | 64 (12.1) | <0.001 |
| Statin | 268 (34.6) | 38 (15.6) | 230 (43.4) | <0.001 |
| β blocker | 348 (45.0) | 52 (21.3) | 296 (55.8) | <0.001 |
| ACEI/ARB | 396 (51.2) | 47 (19.3) | 349 (65.8) | <0.001 |
| Oral anticoagulant | 102 (13.2) | 7 (2.9) | 95 (17.9) | <0.001 |
| Clinical and laboratory characteristics at admission | ||||
| Heart rate, pulse/min | 88.6±28.1 | 84.5±27.0 | 90.5±28.4 | 0.002 |
| Systolic blood pressure, mm Hg | 142.2±30.8 | 140.1±29.0 | 143.2±31.5 | 0.129 |
| Diastolic blood pressure, mm Hg | 80.5±17.4 | 83.0±16.2 | 79.3±17.8 | 0.002 |
| Creatinine, mg/dL | 0.9 (0.8–1.2) | 0.8 (0.7–1.0) | 1.0 (0.8–1.3) | <0.001 |
| Peak cTnI, X URL | 19.7 (6.1–60.7) | 15.7 (6.0–52.6) | 21.1 (6.2–65.5) | 0.445 |
| Total cholesterol, mg/dL | 185.6±50.0 | 200.7±43.0 | 178.6±51.5 | <0.001 |
| Hemoglobin, g/dL | 13.2±2.1 | 14.2±1.8 | 12.8±2.0 | <0.001 |
| C‐reactive protein, mg/dL | 0.4 (0.2–1.2) | 0.3 (0.1–0.7) | 0.5 (0.2–1.6) | <0.001 |
| GRACE score | 143±38.9 | 113±30.6 | 156±34.3 | <0.001 |
| Sinus rhythm at ECG | 632 (81.7) | 227 (93.0) | 405 (76.4) | <0.001 |
| Signs of ischemia at ECG | 403 (52.1) | 109 (44.7) | 294 (55.5) | 0.005 |
| LVEF, % | 54.4±11.1 | 57.1±9.1 | 53.2±11.7 | <0.001 |
| LVEF <50%, n (%) | 194 (25.1) | 37 (15.2) | 157 (29.6) | <0.001 |
Continuous variables are presented as median (interquartile range) or mean±SD according to the frequency distribution, and categorical variables as n (%). ACEI indicates angiotensin converting enzyme inhibitor; ARB, angiotensin receptor blocker; CABG, coronary artery bypass graft; COPD, chronic obstructive pulmonary disease; cTnI, cardiac troponin I; GRACE, Global Registry of Acute Coronary Events; LVEF, left ventricular ejection fraction evaluated by transthoracic echocardiogram; P1 T2MI, phenotype 1 type 2 myocardial infarction; P2 T2MI, phenotype 2 type 2 myocardial infarction; PAD, peripheral artery disease; PCI, percutaneous coronary intervention; TIA, transient ischemic attack; and URL, upper reference limit.
Variables used in latent class analysis.
Table 2.
Angiographic Findings, Procedural Characteristics, and Discharge Details in the Overall Cohort Study and by Clinical Phenotype
| Total | P1 T2MI | P2 T2MI | P value | |
|---|---|---|---|---|
| N=774 | N=244 | N=530 | ||
| Angiographic findings | ||||
| 1‐vessel disease | 152 (19.8) | 45 (18.4) | 108 (20.4) | 0.530 |
| 2‐vessel disease | 95 (12.3) | 24 (9.8) | 71 (13.4) | 0.161 |
| 3‐vessel disease | 120 (15.5) | 19 (7.8) | 101 (19.1) | <0.001 |
| Left main disease only | 12 (1.6) | 2 (0.8) | 10 (1.9) | 0.264 |
| Obstructive CAD | 426 (55.0) | 98 (40.2) | 328 (61.9) | <0.001 |
| Gensini score | 7 (0–34.75) | 1.5 (0–14) | 11 (0–42) | <0.001 |
| Nonatherosclerotic coronary artery lesions | 50 (6.5) | 39 (16.0) | 11 (2.1) | <0.001 |
| Coronary dissection | 24 (3.1) | 20 (8.2) | 4 (0.8) | <0.001 |
| Coronary embolism | 2 (0.3) | 0 (0.0) | 2 (0.4) | 0.337 |
| Coronary vasospasm | 18 (2.3) | 14 (5.7) | 4 (0.8) | <0.001 |
| Myocardial bridging | 6 (0.8) | 5 (2.0) | 1 (0.2) | 0.006 |
| Myocardial revascularization | ||||
| PCI | 257 (33.2) | 74 (30.3) | 183 (34.5) | 0.249 |
| CABG | 18 (2.3) | 2 (0.8) | 16 (3.0) | 0.059 |
| Medical therapy at discharge* | ||||
| Aspirin | 631 (82.4) | 199 (81.6) | 432 (82.8) | 0.684 |
| DAPT | 459 (59.9) | 158 (64.8) | 301 (57.7) | 0.062 |
| Statins | 616 (80.4) | 192 (78.7) | 424 (81.2) | 0.410 |
| β blockers | 631 (82.4) | 195 (79.9) | 436 (83.5) | 0.222 |
| ACEI/ARB | 568 (74.2) | 148 (60.7) | 420 (80.5) | <0.001 |
| Oral anticoagulant | 163 (21.3) | 17 (7.0) | 146 (28.0) | <0.001 |
Continuous variables are presented as median (interquartile range) or mean±SD according to the frequency distribution, and categorical variables as n (%). ACEI indicates angiotensin converting enzyme inhibitor; ARB, angiotensin receptor blocker; CABG, coronary artery bypass graft; CAD, coronary artery disease; DAPT, dual antiplatelet therapy; P1 T2MI, phenotype 1 type 2 myocardial infarction; P2 T2MI, phenotype 2 type 2 myocardial infarction; and PCI, percutaneous coronary intervention.
Data from patients discharged alive only.
Identification of Clinical Phenotypes and Baseline Characteristics
Two clinical phenotypes were identified using LCA, as described in the Methods section (Figure 1). As shown in Table S1, which presents the LCA results for different class models, the Bayesian information criterion suggested a 3‐class model. However, the difference in Bayesian information criterion between the 3‐ and 2‐class solutions was modest, and the average latent class posterior probability was excellent for the 2‐class solution (ie, >0.90). For this reason, and for a practical approach, we chose the 2‐class solution over the 3‐class solution. The baseline characteristics of the 2 identified clinical phenotypes are presented in Tables 1 and 2.
Figure 1. Predicted probability of phenotype assignment on the basis of age and comorbidities.
A, predicted probability of assignment to P1 or P2 T2MI as a function of age: the curves show that starting from about age 64 years, patients are more likely to belong to P2 T2MI; B, predicted probability (and 95% CI) of dichotomous indicators of disease in P1 T2MI; C, predicted probability (and 95% CI) of dichotomous indicators of disease in P2 T2MI. CKD indicates chronic kidney disease; COPD, chronic obstructive pulmonary disease; CV, cardiovascular; P1 T2MI, phenotype 1 type 2 myocardial infarction; and P2, phenotype 2 type 2 myocardial infarction.
Of the 774 patients, 244 (31.5%) were classified in phenotype 1 (P1 T2MI) and 530 (68.5%) in phenotype 2 (P2 T2MI). P1 T2MI comprised significantly younger patients (mean age, 57.1±10.6 versus 76.2±8.2 years) than P2 T2MI and had fewer cardiovascular risk factors, except for smoking, which had a higher prevalence in the first phenotype (60.7% versus 48.5%, P=0.002). P2 T2MI had a markedly greater burden of comorbidities, including a significantly higher prevalence of known cardiovascular disease (14.3% in P1 T2MI versus 45.8% in P2 T2MI, P<0.001). Conversely, in P1 T2MI, the index event was often the first event in patients' cardiovascular histories.
Although peak cTn levels did not significantly differ between the 2 clinical phenotypes, Global Registry of Acute Coronary Events score was higher in P2 T2MI (mean values 156±34.3 versus 113±30.6, P<0.001), whereas left ventricular ejection fraction was higher in P1 T2MI (57.1% versus 53.2%, P<0.001), as were total cholesterol levels (mean values, 200.7±43.0 versus 178.6±51.5 mg/dL, P<0.001).
Comparison of T2MI Pathogeneses and Atherosclerotic Burden Between Clinical Phenotypes
Table S2 presents the precipitating factors of T2MI in the overall population and within each clinical phenotype (Figure 2). Significant differences in precipitating factors were observed between the 2 clinical phenotypes. P2 T2MI experienced multiple causative factors more frequently than P1 T2MI (16.4% versus 10.7%, P=0.035), with higher rates of sepsis (12.1% versus 5.7%, P=0.007), anemia/relevant bleeding (5.1% versus 0.4%, P=0.001), and respiratory failure (16.2% versus 6.6%, P<0.001). Nonatherosclerotic coronary artery lesions were significantly more common in P1 T2MI (16.0% versus 2.1%, P<0.001), which also had a lower prevalence of obstructive CAD compared with P2 T2MI (40.2% versus 61.9%, P<0.001). Consequently, P2 T2MI had a markedly higher atherosclerotic burden, as indicated by higher Gensini (median values, 1.5 [interquartile range, 0–14] versus 11 (interquartile range, 0–42), P<0.001]). Despite these differences, while overall rates of revascularization, whether percutaneous or surgical, did not significantly differ between the clinical phenotypes, among patients with obstructive CAD, those in phenotype 1 were more frequently revascularized than those in phenotype 2. Notably, secondary prevention therapy was substantially intensified in both phenotypes, even among patients who were already receiving similar treatment before admission, and despite the absence of significant coronary stenosis (Tables S3 and S4).
Figure 2. Distribution of pathogeneses of T2MI among clinical phenotypes.
The bar chart illustrates the distribution of different pathogeneses of T2MI across the 2 clinical phenotypes identified using latent class analysis. Each bar represents the proportion (%) of precipitating factors causes in each clinical phenotype, highlighting the variability in pathogenetic factors between the clinical phenotypes. NS=P>0.05; *P<0.05; **P<0.01; ***P<0.001. T2MI indicates type 2 myocardial infarction.
Short‐ and Long‐Term Outcomes and Prognostic Predictors
At 30 days after the T2MI event, no significant differences in overall MACE were observed between the 2 phenotypes (Table S5). However, when examining individual components of the composite outcome, P2 T2MI had higher rates of all‐cause death (2.5% versus 0%, P=0.014) and cardiovascular death (2.1% versus 0%, P=0.023) compared with P1 T2MI.
Table 3 shows the rates of MACEs and other outcomes in the overall cohort and the 2 clinical phenotypes over a median follow‐up of 53 months (interquartile range, 30–71). In the overall population, 380 patients (49.1%) experienced MACEs during follow‐up, with all‐cause death accounting for 25.7% of cases, primarily driven by cardiovascular death (15.8%). Nonfatal reinfarction occurred in 94 patients (12.1%), with more than half of these classified as type 2 reinfarctions (7.1%).
Table 3.
Long‐Term Outcomes in the Overall Cohort Study and by Clinical Phenotype
| Total | P1 T2MI | P2 T2MI | P value | |
|---|---|---|---|---|
| N=774 | N=244 | N=530 | ||
| MACEs | 380 (49.1) | 58 (23.8) | 322 (60.8) | <0.001 |
| All‐cause death | 199 (25.7) | 14 (5.7) | 185 (34.9) | <0.001 |
| Reinfarction | 94 (12.1) | 21 (8.6) | 73 (13.8) | 0.041 |
| Unplanned revascularization | 42 (5.4) | 11 (4.5) | 31 (5.8) | 0.444 |
| Ischemic stroke | 26 (3.4) | 3 (1.2) | 23 (4.3) | 0.026 |
| Heart failure hospitalization | 107 (13.8) | 11 (4.5) | 96 (18.1) | <0.001 |
| Type 2 reinfarction | 55 (7.1) | 10 (4.1) | 45 (8.5) | 0.027 |
| Cardiovascular death | 122 (15.8) | 4 (1.6) | 118 (22.3) | <0.001 |
MACE indicates major adverse cardiovascular event; P1 T2MI, phenotype 1 type 2 myocardial infarction; and P2 T2MI, phenotype 2 type 2 myocardial infarction;.
Regarding the 2 clinical phenotypes, P1 T2MI had a significantly better long‐term prognosis in terms of MACEs compared with P2 T2MI (23.8% versus 60.8%, P<0.001). Specifically, P2 T2MI had higher rates of all‐cause death (34.9% versus 5.7%, P<0.001), cardiovascular death (22.3% versus 1.6%, P<0.001), reinfarction (13.8% versus 8.6%, P=0.041), type 2 reinfarction (8.5% versus 4.1%, P=0.027), ischemic stroke (4.3% versus 1.2%, P=0.026), and heart failure hospitalization (18.1% versus 4.5%, P<0.001). Unplanned revascularization rates did not differ between the 2 clinical phenotypes (5.8% versus 4.5%, P=0.444).
The Kaplan–Meier curves show that patients in P2 T2MI had worse outcomes in terms of MACEs over a 5‐year period compared with those in P1 T2MI (Figure 3A), as well as higher rates of type 2 reinfarction (Figure S1). The Kaplan–Meier curves, stratified by the presence of obstructive CAD, showed that, in P2 T2MI, patients with obstructive CAD had a worse long‐term prognosis compared with those with nonobstructive CAD. In contrast, no significant differences were observed between patients with obstructive and nonobstructive CAD in P1 T2MI (Figure 3B). Similar results were observed in the sensitivity analysis excluding patients with coronary artery spasm, embolism, or dissection, confirming the robustness of the prognostic differences between phenotypes (Table S6 and Figure S2).
Figure 3. Kaplan–Meier estimates of survival from MACEs in the 2 clinical phenotypes (A) and in each clinical phenotype stratified by the presence of obstructive CAD (B).
Kaplan–Meier curves illustrating MACE‐free survival. A, compares survival between the 2 clinical phenotypes identified using latent class analysis, while (B) shows survival in each clinical phenotype stratified according to the presence/absence of obstructive CAD. The curves indicate significant differences in MACE‐free survival between the 2 clinical phenotypes and highlight that the impact of obstructive CAD on survival outcomes is observed only among patients in P2 T2MI. In (B), the Bonferroni‐adjusted significance threshold was set at P<0.025 to account for multiple comparisons. CAD indicates coronary artery disease; MACE, major adverse cardiovascular event; P1 T2MI, phenotype 1 type 2 myocardial infarction; and P2 T2MI, phenotype 2 type 2 myocardial infarction.
In P1 T2MI, a multivariable Cox regression model identified peak cTn values as the only independent predictor of MACEs (adjusted hazard ratio [aHR], 1.13 [95% CI, 1.05–1.21]; P=0.001; Table 4). Conversely, in P2 T2MI, age (aHR, 1.02 [95% CI, 1.01–1.04]; P=0.005), known cardiovascular disease (aHR, 1.45 [95% CI, 1.14–1.87]; P=0.003), COPD (aHR, 1.73 [95% CI, 1.33–2.26]; P<0.001), peak cTn values (aHR, 1.05 [95% CI, 1.01–1.08]; P=0.012), and the Gensini score (aHR, 1.04 [95% CI, 1.01–1.07]; P=0.003) emerged as independent predictors of MACEs (Table 5). The proportional hazards assumption was confirmed for all models using Schoenfeld residuals (Table S7). The exclusion of influential observations did not modify the results in either phenotype. Linearity in the relationship between continuous covariates and the log hazard was supported by the analysis of deviance residuals (Figures S3 and S4).
Table 4.
Univariate and Multivariable Analyses to Identify MACE Predictors in P1 T2MI
| Univariate regression | Multivariable regression | |||||
|---|---|---|---|---|---|---|
| HR | 95% CI | P value | HR | 95% CI | P value | |
| Age, y | 0.99 | 0.97–1.02 | 0.448 | … | … | … |
| Sex, female | 0.95 | 0.56–1.62 | 0.846 | … | … | … |
| Hypertension | 1.11 | 0.65–1.88 | 0.708 | … | … | … |
| Dyslipidemia | 0.66 | 0.38–1.12 | 0.124 | … | … | … |
| Diabetes | 1.97 | 0.88–4.37 | 0.098 | 1.64 | 0.73–3.72 | 0.234 |
| Known cardiovascular disease | 1.61 | 0.88–2.94 | 0.126 | … | … | … |
| COPD | 1.66 | 0.60–4.59 | 0.333 | … | … | … |
| eGFR, mL/min per 1.73 m2 | 1.00 | 0.98–1.01 | 0.776 | … | … | … |
| Hemoglobin, g/dL | 1.03 | 0.88–1.20 | 0.719 | … | … | … |
| LVEF, % | 0.98 | 0.95–1.00 | 0.076 | 0.98 | 0.96–1.01 | 0.303 |
| Peak cTnI, X URL | 1.15 | 1.08–1.23 | <0.001 | 1.13 | 1.05–1.21 | 0.001 |
| Aspirin | 1.16 | 0.59–2.30 | 0.664 | … | … | … |
| DAPT | 1.32 | 0.73–2.38 | 0.353 | … | … | … |
| Statin | 0.90 | 0.49–1.64 | 0.723 | … | … | … |
| β blocker | 1.37 | 0.67–2.80 | 0.384 | … | … | … |
| ACEI/ARB | 1.45 | 0.83–2.54 | 0.190 | … | … | … |
| Gensini score | 1.08 | 1.01–1.14 | 0.016 | 1.05 | 0.98–1.12 | 0.201 |
Variables associated with MACEs in univariate regression (P value<0.1) were included in the multivariable model. ACEI indicates angiotensin‐converting enzyme inhibitor; ARB, angiotensin receptor blocker; COPD, chronic obstructive pulmonary disease; cTnI, cardiac troponin I; DAPT, dual antiplatelet therapy; eGFR, estimated glomerular filtration rate; HR, hazard ratio; LVEF, left ventricular ejection fraction evaluated by transthoracic echocardiogram; MACE, major adverse cardiovascular event; P1 T2MI, phenotype 1 type 2 myocardial infarction; and URL, upper reference limit.
Table 5.
Univariate and Multivariable Analyses to Identify MACE Predictors in P2 T2MI
| Univariate regression | Multivariable regression | |||||
|---|---|---|---|---|---|---|
| HR | 95% CI | P value | HR | 95% CI | P value | |
| Age, y | 1.03 | 1.02–1.05 | <0.001 | 1.02 | 1.01–1.04 | 0.005 |
| Sex, female | 0.81 | 0.65–1.02 | 0.067 | 0.94 | 0.73–1.21 | 0.637 |
| Hypertension | 0.85 | 0.57–1.26 | 0.414 | … | … | … |
| Dyslipidemia | 0.65 | 0.51–0.82 | <0.001 | 0.65 | 0.57–0.82 | 0.016 |
| Diabetes | 1.16 | 0.92–1.47 | 0.199 | … | … | … |
| Known cardiovascular disease | 1.63 | 1.30–2.04 | <0.001 | 1.45 | 1.14–1.87 | 0.003 |
| COPD | 1.75 | 1.36–2.26 | <0.001 | 1.73 | 1.33–2.26 | <0.001 |
| eGFR, mL/min per 1.73 m2 | 0.99 | 0.98–0.99 | <0.001 | 0.99 | 0.98–0.99 | <0.001 |
| Hemoglobin, g/dL | 0.91 | 0.86–0.96 | 0.001 | 0.97 | 0.91–1.03 | 0.295 |
| LVEF, % | 0.98 | 0.97–0.99 | <0.001 | 0.99 | 0.98–1.00 | 0.057 |
| Peak cTnI, X URL | 1.05 | 1.02–1.08 | 0.003 | 1.05 | 1.01–1.08 | 0.012 |
| Aspirin | 0.97 | 0.71–1.32 | 0.851 | … | … | … |
| DAPT | 1.18 | 0.93–1.48 | 0.166 | … | … | … |
| Statin | 0.92 | 0.70–1.21 | 0.550 | … | … | … |
| β blocker | 1.03 | 0.76–1.40 | 0.859 | … | … | … |
| ACEI/ARB | 0.67 | 0.51–0.87 | 0.003 | 0.84 | 0.63–1.13 | 0.246 |
| Gensini score | 1.04 | 1.02–1.07 | <0.001 | 1.04 | 1.01–1.07 | 0.003 |
Variables associated with MACEs in univariate regression (P value <0.1) were included in the multivariable model. ACEI indicates angiotensin‐converting enzyme inhibitor; ARB, angiotensin receptor blocker; COPD, chronic obstructive pulmonary disease; cTnI, cardiac troponin I; DAPT, dual antiplatelet therapy; eGFR, estimated glomerular filtration rate; HR, hazard ratio; LVEF, left ventricular ejection fraction evaluated by transthoracic echocardiogram; MACE, major adverse cardiovascular event; P1 T2MI, phenotype 1 type 2 myocardial infarction; and URL, upper reference limit.
DISCUSSION
To our knowledge, this is the first study to search for empirical evidence of different clinical phenotypes in a large real‐world cohort of patients with T2MI and to explore specific risk factors and long‐term prognosis on the basis of these clinical phenotypes. The main findings are (1) T2MI is associated with a poor long‐term prognosis, that is, high rates of MACEs and frequent type 2 reinfarctions; (2) in the heterogeneous spectrum of patients with T2MI, 2 distinct clinical phenotypes can be identified: relatively young and healthy patients and older patients with high comorbidity rates; (3) the 2 clinical phenotypes exhibit different clinical characteristics, precipitating factors, prognoses, and predictive prognostic factors, despite still sharing common management and therapeutic strategies; and (4) the extent of CAD is an independent prognostic predictor only in older and comorbid patients with T2MI, while it shows no prognostic value in younger and relatively healthier patients with T2MI (Graphical Abstract).
Clinical Characteristics and Prognosis of T2MI
The study cohort before phenotyping aligned with the existing knowledge about patients with T2MI, characterized by a high mean age and significant cardiovascular risk profiles. 13 In our study cohort, 83.7% of patients had identifiable precipitating factors explaining the oxygen supply–demand mismatch; this is promising for treatment, as managing T2MI mainly involves addressing the underlying causes. 8 Literature reports on triggering factor recognition vary widely, ranging from 20% to almost 100%; however, our findings corroborate that the most common T2MI causes include tachyarrhythmias, hypertension, respiratory failure/hypoxia, and sepsis. 11 , 12 , 13 , 27 , 28 , 29
Additionally, the atherosclerotic coronary burden was substantial, with more than half of the patients exhibiting obstructive CAD and nearly 1 in 5 having critical 3‐vessel disease, confirming the existing literature. 1 , 30 Although evidence specifically guiding optimal T2MI management remains limited, current clinical practice often includes the use of therapies traditionally recommended for secondary prevention in atherosclerotic cardiovascular disease. In our cohort, aspirin use more than doubled from admission to discharge (from 40.2% to 82.4%), and 59.8% of patients were discharged on dual antiplatelet therapy. Notably, this rate of dual antiplatelet therapy use was not solely driven by revascularization, as only one third underwent percutaneous coronary intervention. This proportion aligns with prior reports, which describe revascularization rates ranging widely in T2MI, and likely reflects both the heterogeneity of clinical presentations and the selective use of invasive strategies in this complex population. 15 Patients were also likely to start β blockers, renin–angiotensin–aldosterone system inhibitors, and statins during their hospital stay, mirroring treatment strategies commonly used in T1MI and other high‐risk cardiovascular settings.
Our results confirm that the prognosis for T2MI is far from benign: over a median follow‐up of 53 months, half of the patients experienced a MACE. These findings highlight the complexity of T2MI management and suggest that directly applying strategies developed for T1MI may not adequately address the diverse pathophysiological mechanisms and clinical needs of patients with T2MI. Inadequate or misaligned treatment, whether due to underuse of preventive therapies or inappropriate intensification, may contribute to the high rate of adverse outcomes. Furthermore, the cardiovascular death rate in our cohort (15.8%) aligns with existing literature and accounted for more than half of all‐cause deaths (25.7%), 31 , 32 , 33 suggesting that cardiovascular causes might represent a potentially relevant contributor to death in patients with T2MI. Reinfarction was a frequent occurrence among those experiencing MACEs, constituting 24.7% of total MACEs. Notably, more than half (58.5%) of these were type 2 reinfarctions, indicating that T2MI is often a recurrent event in the natural history of this specific disease. 9 , 34
Distinct Clinical Phenotypes in T2MI: Differing Clinical Profiles and Event Triggers
The identification of 2 clinical phenotypes was consistent with everyday clinical observations. In fact, T2MI affects not only older individuals with multiple comorbidities and a higher cardiovascular risk profile (P2 T2MI), but also middle‐aged patients with few or no significant health issues and a lower cardiovascular risk profile (P1 T2MI). All factors used in the model were significantly different between the 2 clinical phenotypes and biologically plausible. Given this clear distinction, most patients with T2MI are likely to align with 1 of these 2 profiles in clinical practice.
Each clinical phenotype displayed a distinct profile of precipitating factors, which could have significant implications for management. P1 T2MI experienced higher rates of nonatherosclerotic coronary lesions, such as dissection, vasospasm, myocardial bridging and embolism, along with cases of unknown pathogenesis. The significant number of patients lacking an identified cause suggests the need for additional investigative methods beyond CAG to better understand the underlying mechanisms of AMI in this phenotype in order to implement the most appropriate treatment. In such cases, the integrated use of coronary optical coherence tomography and cardiac magnetic resonance has proven helpful in identifying potential mechanisms of AMI in the majority of patients. 35 , 36 , 37 , 38
In contrast, P2 T2MI had more easily diagnosable triggers, including higher rates of sepsis, anemia, and respiratory failure, consistent with clinical observations that often reveal multiple precipitating factors in this phenotype. Interestingly, no significant differences were observed between the 2 clinical phenotypes regarding hypertension and tachyarrhythmias as causes of T2MI. This may be explained by the fact that both conditions precipitate ischemic injury through increased myocardial oxygen demand rather than reduced supply, making them equally likely to affect patients across both clinical phenotypes.
Conversely, the 2 clinical phenotypes significantly differed in their atherosclerotic coronary burden, with P2 T2MI exhibiting markedly higher rates of obstructive CAD, critical 3‐vessel disease, and elevated Gensini score. Interestingly, although overall revascularization rates did not differ between clinical phenotypes, patients in P1 T2MI with obstructive CAD were more frequently revascularized compared with those in P2 T2MI. Nonetheless, medical therapy at discharge was broadly similar between the groups. This discrepancy between the varying degrees of CAD severity across clinical phenotypes and the similarity in therapeutic strategies suggests that our understanding and management of T2MI continue to be informed by our experiences with T1MI.
Distinct Risk Profiles and Therapeutic Implications in T2MI Clinical Phenotypes
While prognosis in terms of MACE was significantly better for patients in P1 T2MI compared with P2 T2MI, as expected, notable differences in predictive prognostic factors emerged during multivariable analysis. In P1 T2MI, only peak cTn levels were associated with a higher adjusted hazard ratio for MACEs. Conversely, in P2 T2MI, several prognostic factors were identified, including age, cardiovascular disease, COPD, peak cTn values, and atherosclerotic burden as estimated by Gensini score.
The high proportion of patients in P1 T2MI with absent or unremarkable CAD, combined with a relatively greater incidence of nonatherosclerotic coronary lesions, helps explain why differences in atherosclerotic burden lack meaningful prognostic implications. The relative absence of clear prognostic factors in P1 T2MI may be attributed to the presence of generally healthy patients with minimal cardiac disease and few comorbidities. However, this also underscores the need for further research to uncover the determinants of outcomes in this subgroup. Since P1 T2MI included a larger proportion of patients with undetected triggering causes compared to with P2 T2MI, accurately identifying these factors could provide valuable insights into their prognosis.
In P2 T2MI, age and COPD as prognostic factors may similarly reflect a greater degree of frailty. Known cardiovascular diseases and higher atherosclerotic burden also play a significant role in this clinical phenotype. This observation aligns with the concept of T2MI: when faced with the same triggering conditions, hearts with reduced functional reserve, both mechanical and coronary, are more prone to experience supply–demand mismatches compared with healthier hearts. This is further supported by the fact that these patients are more likely to experience type 2 reinfarctions over time.
Previous studies have shown that CAD is an independent predictor of MACEs in patients with T2MI. 4 , 10 , 28 , 39 , 40 However, our study suggests that the prognostic value of CAD may vary according to the patient's clinical phenotype. In particular, in younger patients with less extensive CAD (P1 T2MI), its impact on outcomes may be attenuated, especially in the short term, whereas in older and more comorbid patients (P2 T2MI) it appears to be more pronounced.
This evidence indicates the need to move beyond a “1‐size‐fits‐all” approach to the management of T2MI. Clinicians should be encouraged to adopt more tailored and individualized therapeutic strategies based on specific patient clinical phenotypes. 1 , 41 , 42 For patients with P1 T2MI, the focus should be on identifying and addressing the inciting cause of T2MI, especially nonatherosclerotic coronary events or other cardiovascular conditions. In contrast, patients with P2 T2MI require a dual approach: recognizing and eliminating the systemic trigger of T2MI, while also treating underlying atherosclerosis. Future studies will be crucial in demonstrating the efficacy of these phenotype‐driven management strategies, potentially leading to improved long‐term outcomes in each clinical phenotype.
Study Limitations
Although this is one of the first studies addressing this topic with a large sample size, some limitations should be considered. First, the study population included only patients who underwent CAG. Opposing trends in T2MI management are worth noting: on the one hand, patients with a higher pretest probability of CAD, typically older and more comorbid, are more likely to undergo CAG. 10 On the other hand, this same clinical phenotype may also be more often managed conservatively, without invasive procedures. As a result, our findings reflect a population of patients with T2MI selected for angiographic evaluation and may not fully capture the clinical spectrum and treatment variability of all patients with T2MI. This underscores the importance of future research comparing the characteristics and outcomes of patients with T2MI who do and do not undergo CAG. 43 Second, although all coronary angiograms consistent with T1MI were excluded during screening and intracoronary imaging was used when clinically indicated, we cannot rule out the possibility that some minor acute plaque disruptions were undetected and thus included in the study. In addition, due to the inherent diagnostic complexity of T2MI, a proportion of patients with acute myocardial injury might have been misclassified as having T2MI, as demonstrated in previous studies, despite formal adjudication according to the Fourth Universal Definition of Myocardial Infarction. 10 , 44 Third, considering the study design, T2MI phenotyping is not applicable to patients with ST‐segment–elevation myocardial infarction. In the ST‐segment–elevation myocardial infarction setting, urgent reperfusion is prioritized regardless of the infarct mechanism. Including patients with ST‐segment–elevation myocardial infarction would have introduced treatment and prognostic heterogeneity compared with patients with non–ST‐segment–elevation myocardial infarction, where management more directly reflects the clinical complexity of T2MI. Finally, despite the statistical robustness of clinical phenotype identification, some overlap between clinical phenotypes exists, as they necessarily represent a simplified model of a complex population. Therefore, our findings should be considered as exploratory, although biologically plausible, and larger studies are needed to validate these results.
CONCLUSIONS
This study, conducted on a large, real‐world cohort of patients with T2MI, is the first to identify 2 distinct clinical phenotypes, each characterized by unique baseline features and pathogeneses. Phenotype 1 included younger, healthier patients with fewer comorbidities and lower coronary atherosclerotic burden; vice versa, Phenotype 2 included older, comorbid patients with higher prevalence of significant CAD. Despite being managed with similar strategies derived from T1MI, these clinical phenotypes exhibit different prognoses and distinct prognostic indicators. Phenotype 1 had more favorable outcomes and peak troponin values as the sole independent predictor, while Phenotype 2 demonstrated worse outcomes and several robust, independent predictors, including a higher coronary atherosclerotic burden. These findings highlight the heterogeneity of the T2MI population and underscore the potential of phenotyping as a promising approach to develop tailored diagnostic and therapeutic strategies.
Disclosures
None.
Supporting information
Data S1. Supplemental Methods
Tables S1–S7
Figures S1–S4
References 45,46
This manuscript was sent to Amgad Mentias, MD, Associate Editor, for review by expert referees, editorial decision, and final disposition.
Supplemental Material is available at https://www.ahajournals.org/doi/suppl/10.1161/JAHA.125.042858
For Disclosures, see page 13.
The abstract of this work was presented at the European Society of Cardiology Congress, August 29 to September 1, 2025, in Madrid, Spain.
References
- 1. Sandoval Y, Jaffe AS. Type 2 myocardial infarction: JACC review topic of the week. J Am Coll Cardiol. 2019;73:1846–1860. doi: 10.1016/j.jacc.2019.02.018 [DOI] [PubMed] [Google Scholar]
- 2. Thygesen K, Alpert JS, Jaffe AS, Chaitman BR, Bax JJ, Morrow DA, White HD; Executive Group on behalf of the Joint European Society of Cardiology/American College of Cardiology/American Heart Association/World Heart Federation Task Force for the Universal Definition of Myocardial I . Fourth universal definition of myocardial infarction (2018). Circulation. 2018;138:e618–e651. doi: 10.1161/CIR.0000000000000617 [DOI] [PubMed] [Google Scholar]
- 3. Januzzi JL, Sandoval Y. The many faces of type 2 myocardial infarction. J Am Coll Cardiol. 2017;70:1569–1572. doi: 10.1016/j.jacc.2017.07.784 [DOI] [PubMed] [Google Scholar]
- 4. Nestelberger T, Boeddinghaus J, Badertscher P, Twerenbold R, Wildi K, Breitenbucher D, Sabti Z, Puelacher C, Rubini Gimenez M, Kozhuharov N, et al. Effect of definition on incidence and prognosis of type 2 myocardial infarction. J Am Coll Cardiol. 2017;70:1558–1568. doi: 10.1016/j.jacc.2017.07.774 [DOI] [PubMed] [Google Scholar]
- 5. Wang G, Zhao N, Zhong S, Li J. A systematic review on the triggers and clinical features of type 2 myocardial infarction. Clin Cardiol. 2019;42:1019–1027. doi: 10.1002/clc.23230 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Gard A, Lindahl B, Batra G, Hadziosmanovic N, Hjort M, Szummer KE, Baron T. Interphysician agreement on subclassification of myocardial infarction. Heart. 2018;104:1284–1291. doi: 10.1136/heartjnl-2017-312409 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Murphy SP, McCarthy CP, Cohen JA, Rehman S, Jones‐O'Connor M, Olshan DS, Singh A, Vaduganathan M, Cui J, Januzzi JL Jr, et al. Application of the GRACE, TIMI, and TARRACO risk scores in type 2 myocardial infarction. J Am Coll Cardiol. 2020;75:344–345. doi: 10.1016/j.jacc.2019.11.004 [DOI] [PubMed] [Google Scholar]
- 8. Byrne RA, Rossello X, Coughlan JJ, Barbato E, Berry C, Chieffo A, Claeys MJ, Dan GA, Dweck MR, Galbraith M, et al. 2023 ESC guidelines for the management of acute coronary syndromes. Eur Heart J. 2023;44:3720–3826. doi: 10.1093/eurheartj/ehad191 [DOI] [PubMed] [Google Scholar]
- 9. Raphael CE, Roger VL, Sandoval Y, Singh M, Bell M, Lerman A, Rihal CS, Gersh BJ, Lewis B, Lennon RJ, et al. Incidence, trends, and outcomes of type 2 myocardial infarction in a community cohort. Circulation. 2020;141:454–463. doi: 10.1161/CIRCULATIONAHA.119.043100 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Chapman AR, Shah ASV, Lee KK, Anand A, Francis O, Adamson P, McAllister DA, Strachan FE, Newby DE, Mills NL. Long‐term outcomes in patients with type 2 myocardial infarction and myocardial injury. Circulation. 2018;137:1236–1245. doi: 10.1161/CIRCULATIONAHA.117.031806 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Etaher A, Gibbs OJ, Saad YM, Frost S, Nguyen TL, Ferguson I, Juergens CP, Chew D, French JK. Type‐II myocardial infarction and chronic myocardial injury rates, invasive management, and 4‐year mortality among consecutive patients undergoing high‐sensitivity troponin T testing in the emergency department. Eur Heart J Qual Care Clin Outcomes. 2020;6:41–48. doi: 10.1093/ehjqcco/qcz019 [DOI] [PubMed] [Google Scholar]
- 12. Singh A, Gupta A, DeFilippis EM, Qamar A, Biery DW, Almarzooq Z, Collins B, Fatima A, Jackson C, Galazka P, et al. Cardiovascular mortality after type 1 and type 2 myocardial infarction in Young adults. J Am Coll Cardiol. 2020;75:1003–1013. doi: 10.1016/j.jacc.2019.12.052 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. McCarthy CP, Kolte D, Kennedy KF, Vaduganathan M, Wasfy JH, Januzzi JL Jr. Patient characteristics and clinical outcomes of type 1 versus type 2 myocardial infarction. J Am Coll Cardiol. 2021;77:848–857. doi: 10.1016/j.jacc.2020.12.034 [DOI] [PubMed] [Google Scholar]
- 14. Santucci A, Cavallini C. Diagnostic, therapeutic and prognostic aspects of type 2 myocardial infarction. G Ital Cardiol (Rome). 2022;23:523–532. doi: 10.1714/3831.38170 [DOI] [PubMed] [Google Scholar]
- 15. Chapman AR, Taggart C, Boeddinghaus J, Mills NL, Fox KAA. Type 2 myocardial infarction: challenges in diagnosis and treatment. Eur Heart J. 2025;46:504–517. doi: 10.1093/eurheartj/ehae803 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Bergamaschi L, Foà A, Paolisso P, Renzulli M, Angeli F, Fabrizio M, Bartoli L, Armillotta M, Sansonetti A, Amicone S, et al. Prognostic role of early cardiac magnetic resonance in myocardial infarction with nonobstructive coronary arteries. JACC Cardiovasc Imaging. 2024;17:149–161. doi: 10.1016/j.jcmg.2023.05.016 [DOI] [PubMed] [Google Scholar]
- 17. Collet JP, Thiele H, Barbato E, Barthelemy O, Bauersachs J, Bhatt DL, Dendale P, Dorobantu M, Edvardsen T, Folliguet T, et al. 2020 ESC guidelines for the management of acute coronary syndromes in patients presenting without persistent ST‐segment elevation. Eur Heart J. 2021;42:1289–1367. doi: 10.1093/eurheartj/ehaa575 [DOI] [PubMed] [Google Scholar]
- 18. Gensini GG. A more meaningful scoring system for determining the severity of coronary heart disease. Am J Cardiol. 1983;51:606. doi: 10.1016/s0002-9149(83)80105-2 [DOI] [PubMed] [Google Scholar]
- 19. de Lemos JA, Newby LK, Mills NL. A proposal for modest revision of the definition of type 1 and type 2 myocardial infarction. Circulation. 2019;140:1773–1775. doi: 10.1161/CIRCULATIONAHA.119.042157 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Sandoval Y, Thygesen K, Jaffe AS. The universal definition of myocardial infarction: present and future. Circulation. 2020;141:1434–1436. doi: 10.1161/CIRCULATIONAHA.120.045708 [DOI] [PubMed] [Google Scholar]
- 21. Sinha P, Calfee CS, Delucchi KL. Practitioner's guide to latent class analysis: methodological considerations and common pitfalls. Crit Care Med. 2021;49:e63–e79. doi: 10.1097/CCM.0000000000004710 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Formann AK, Kohlmann T. Latent class analysis in medical research. Stat Methods Med Res. 1996;5:179–211. doi: 10.1177/096228029600500205 [DOI] [PubMed] [Google Scholar]
- 23. Zhang Y, Liu S, Miao Q, Zhang X, Wei H, Feng S, Li X. The heterogeneity of symptom burden and fear of progression among kidney transplant recipients: a latent class analysis. Psychol Res Behav Manag. 2024;17:1205–1219. doi: 10.2147/PRBM.S454787 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Siroux V, Basagana X, Boudier A, Pin I, Garcia‐Aymerich J, Vesin A, Slama R, Jarvis D, Anto JM, Kauffmann F, et al. Identifying adult asthma phenotypes using a clustering approach. Eur Respir J. 2011;38:310–317. doi: 10.1183/09031936.00120810 [DOI] [PubMed] [Google Scholar]
- 25. Cohen JB, Schrauben SJ, Zhao L, Basso MD, Cvijic ME, Li Z, Yarde M, Wang Z, Bhattacharya PT, Chirinos DA, et al. Clinical Phenogroups in heart failure with preserved ejection fraction: detailed phenotypes, prognosis, and response to spironolactone. JACC Heart Fail. 2020;8:172–184. doi: 10.1016/j.jchf.2019.09.009 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Nylund KL, Asparouhov T, Muthén BO. Deciding on the number of classes in latent class analysis and growth mixture modeling: a Monte Carlo simulation study. Struct Equ Model Multidiscip J. 2007;14:535–569. doi: 10.1080/10705510701575396 [DOI] [Google Scholar]
- 27. Saaby L, Poulsen TS, Hosbond S, Larsen TB, Pyndt Diederichsen AC, Hallas J, Thygesen K, Mickley H. Classification of myocardial infarction: frequency and features of type 2 myocardial infarction. Am J Med. 2013;126:789–797. doi: 10.1016/j.amjmed.2013.02.029 [DOI] [PubMed] [Google Scholar]
- 28. Baron T, Hambraeus K, Sundstrom J, Erlinge D, Jernberg T, Lindahl B; group T‐As . Impact on long‐term mortality of presence of obstructive coronary artery disease and classification of myocardial infarction. Am J Med. 2016;129:398–406. doi: 10.1016/j.amjmed.2015.11.035 [DOI] [PubMed] [Google Scholar]
- 29. Smilowitz NR, Weiss MC, Mauricio R, Mahajan AM, Dugan KE, Devanabanda A, Pulgarin C, Gianos E, Shah B, Sedlis SP, et al. Provoking conditions, management and outcomes of type 2 myocardial infarction and myocardial necrosis. Int J Cardiol. 2016;218:196–201. doi: 10.1016/j.ijcard.2016.05.045 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Bularga A, Hung J, Daghem M, Stewart S, Taggart C, Wereski R, Singh T, Meah MN, Fujisawa T, Ferry AV, et al. Coronary artery and cardiac disease in patients with type 2 myocardial infarction: a prospective cohort study. Circulation. 2022;145:1188–1200. doi: 10.1161/CIRCULATIONAHA.121.058542 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Shah ASV, Anand A, Strachan FE, Ferry AV, Lee KK, Chapman AR, Sandeman D, Stables CL, Adamson PD, Andrews JPM, et al. High‐sensitivity troponin in the evaluation of patients with suspected acute coronary syndrome: a stepped‐wedge, cluster‐randomised controlled trial. Lancet. 2018;392:919–928. doi: 10.1016/S0140-6736(18)31923-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Kimenai DM, Lindahl B, Chapman AR, Baron T, Gard A, Wereski R, Meex SJR, Jernberg T, Mills NL, Eggers KM. Sex differences in investigations and outcomes among patients with type 2 myocardial infarction. Heart. 2021;107:1480–1486. doi: 10.1136/heartjnl-2021-319118 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Raphael CE, Roger VL, Sandoval Y, Johnson M, Jaffe A, Lerman A, Rihal CS, Bell MR, Singh M, Gulati R. Causes of death after type 2 myocardial infarction and myocardial injury. J Am Coll Cardiol. 2021;78:415–416. doi: 10.1016/j.jacc.2021.05.026 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Wereski R, Kimenai DM, Bularga A, Taggart C, Lowe DJ, Mills NL, Chapman AR. Risk factors for type 1 and type 2 myocardial infarction. Eur Heart J. 2022;43:127–135. doi: 10.1093/eurheartj/ehab581 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35. Reynolds HR, Maehara A, Kwong RY, Sedlak T, Saw J, Smilowitz NR, Mahmud E, Wei J, Marzo K, Matsumura M, et al. Coronary optical coherence tomography and cardiac magnetic resonance imaging to determine underlying causes of myocardial infarction with nonobstructive coronary arteries in women. Circulation. 2021;143:624–640. doi: 10.1161/CIRCULATIONAHA.120.052008 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. Gerbaud E, Arabucki F, Nivet H, Barbey C, Cetran L, Chassaing S, Seguy B, Lesimple A, Cochet H, Montaudon M, et al. OCT and CMR for the diagnosis of patients presenting with MINOCA and suspected epicardial causes. JACC Cardiovasc Imaging. 2020;13:2619–2631. doi: 10.1016/j.jcmg.2020.05.045 [DOI] [PubMed] [Google Scholar]
- 37. Fedele D, Cavallo D, Bodega F, Suma N, Canton L, Ciarlantini M, Ryabenko K, Amicone S, Marinelli V, Asta C, et al. Pathological findings at invasive assessment in MINOCA: a systematic review and meta‐analysis. Heart. 2024;111:291–299. doi: 10.1136/heartjnl-2024-324565 [DOI] [PubMed] [Google Scholar]
- 38. Gudenkauf B, Hays AG, Tamis‐Holland J, Trost J, Ambinder DI, Wu KC, Arbab‐Zadeh A, Blumenthal RS, Sharma G. Role of multimodality imaging in the assessment of myocardial infarction with nonobstructive coronary arteries: beyond conventional coronary angiography. J Am Heart Assoc. 2022;11:e022787. doi: 10.1161/JAHA.121.022787 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39. Gaggin HK, Liu Y, Lyass A, van Kimmenade RR, Motiwala SR, Kelly NP, Mallick A, Gandhi PU, Ibrahim NE, Simon ML, et al. Incident type 2 myocardial infarction in a cohort of patients undergoing coronary or peripheral arterial angiography. Circulation. 2017;135:116–127. doi: 10.1161/CIRCULATIONAHA.116.023052 [DOI] [PubMed] [Google Scholar]
- 40. Taggart C, Monterrubio‐Gomez K, Roos A, Boeddinghaus J, Kimenai DM, Kadesjo E, Bularga A, Wereski R, Ferry A, Lowry M, et al. Improving risk stratification for patients with type 2 myocardial infarction. J Am Coll Cardiol. 2023;81:156–168. doi: 10.1016/j.jacc.2022.10.025 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41. Eggers KM, Baron T, Chapman AR, Gard A, Lindahl B. Management and outcome trends in type 2 myocardial infarction: an investigation from the SWEDEHEART registry. Sci Rep. 2023;13:7194. doi: 10.1038/s41598-023-34312-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42. Taggart C, Ferry A, Barker S, Williams K, Souter G, Bularga A, Wereski R, McDermott MJ, Williams MC, Boeddinghaus J, et al. Targeting investigation and treatment in type 2 myocardial infarction: a pilot randomized controlled trial. JACC: Advances. 2025;4:101738. doi: 10.1016/j.jacadv.2025.101738 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43. Lambrakis K, French JK, Scott IA, Briffa T, Brieger D, Farkouh ME, White H, Chuang AM, Tiver K, Quinn S, et al. The appropriateness of coronary investigation in myocardial injury and type 2 myocardial infarction (ACT‐2): a randomized trial design. Am Heart J. 2019;208:11–20. doi: 10.1016/j.ahj.2018.09.016 [DOI] [PubMed] [Google Scholar]
- 44. McCarthy C, Murphy S, Cohen JA, Rehman S, Jones‐O'Connor M, Olshan DS, Singh A, Vaduganathan M, Januzzi JL Jr, Wasfy JH. Misclassification of myocardial injury as myocardial infarction: implications for assessing outcomes in value‐based programs. JAMA Cardiol. 2019;4:460–464. doi: 10.1001/jamacardio.2019.0716 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45. Armillotta M, Bergamaschi L, Paolisso P, Belmonte M, Angeli F, Sansonetti A, Stefanizzi A, Bertolini D, Bodega F, Amicone S, et al. Prognostic relevance of type 4a myocardial infarction and periprocedural myocardial injury in patients with non‐ST‐segment‐elevation myocardial infarction. Circulation. 2025;151:760–772. doi: 10.1161/CIRCULATIONAHA.124.070729 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46. Mehta SR, Wood DA, Storey RF, Mehran R, Bainey KR, Nguyen H, Meeks B, Di Pasquale G, Lopez‐Sendon J, Faxon DP, et al. Complete revascularization with multivessel PCI for myocardial infarction. N Engl J Med. 2019;381:1411–1421. doi: 10.1056/NEJMoa1907775 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data S1. Supplemental Methods
Tables S1–S7
Figures S1–S4
References 45,46