Fifth universal definition of MI: what changes

In a fictional case, a 78-year-old woman admitted for pneumonia develops rapid ventricular response atrial fibrillation on the second day, with a heart rate around 150 beats per minute and low blood pressure. She reports a vague chest discomfort. High-sensitivity troponin rises between two draws and the electrocardiogram shows diffuse ST-segment depression. Until August 2026, many people would write "type 2 infarction" in the chart and move on.

The fifth universal definition of myocardial infarction, published on August 28, 2026 by the European and American societies and by the World Heart Federation, changes that wording (Mills et al., 2026). The numbered types 1 to 5 are gone. In their place came three clinical contexts: primary infarction, secondary infarction and procedure-related infarction. For the patient in the opening, the infarction label now depends on objective criteria, and the final word may be different: acute myocardial injury.

This text explains the new classification, the criteria for each context, the role of troponin with a sex-specific threshold and the alignment with ICD-11. Management of acute coronary syndrome is left for the text on the 2023 European guideline in the same series; here the subject is the name given to the event and what that name requires investigating.

Key points

• The fifth universal definition replaces types 1 to 5 with three contexts: primary, secondary and procedure-related (Mills et al., 2026).
• Primary infarction includes all acute coronary diseases: atherothrombosis, spontaneous dissection, embolism, vasospasm and late failure of a stent or graft.
• Secondary infarction requires more than troponin and symptoms: confirmation comes from obstructive coronary disease without acute lesion or from a new segmental abnormality on imaging.
• Procedure-related infarction no longer depends on troponin multiples and uses the same criteria for angioplasty and surgery.
• Myocardial injury uses the sex-specific 99th percentile, to reduce underdiagnosis in women.
• MINOCA now means myocardial injury with non-obstructive coronary arteries, a working diagnosis.
• ICD-11 codes accompany each context and each mechanism of primary infarction.

Summary

• Why the definition changed again
• Myocardial injury: acute, chronic and the sex-specific threshold
• Primary infarction: acute coronary disease in all its forms
• Secondary infarction: objective criteria for the imbalance
• Procedure-related infarction: no biomarker threshold
• Sudden death, unrecognized infarction, unstable angina and MINOCA
• ICD-11: the classification reaches the record
• How NEXO helps train this reasoning
• Frequently asked questions
• Take-home messages

Why the definition changed again

The first international definition of infarction came from the World Health Organization in 1979 and relied on symptoms, electrocardiogram and cardiac enzymes. Then came the 2000, 2007, 2012 and 2018 versions, each adjusted to the arrival of troponin and, later, of high-sensitivity assays (Mills et al., 2026). The fourth definition, from 2018, had seven numbered subtypes and was a conceptual advance, but the working group itself acknowledges that it was hard to apply consistently.

The weak point lay in three places. Type 2 grouped, under the same label, spontaneous coronary dissection, vasospasm and coronary embolism with the supply-demand imbalance caused by another acute disease. These patients follow very different diagnostic paths and treatments. Types 4a and 5, after angioplasty and surgery, used arbitrary thresholds of 5 and 10 times the 99th percentile of troponin, when myocardial injury is almost universal in these procedures. And the alphanumeric labels said little to the patient about the mechanism of the event.

An international prospective study cited by the document showed that, after cardiac surgery, almost all patients exceeded 10 times the reference limit, and that only values above about 218 times that limit separated those at higher risk of dying (Mills et al., 2026). A threshold that almost everyone crosses says little about who had an infarction.

The discomfort of those who already used type 2

Before publication, a Delphi study with experts from 19 countries already pointed to the need for revision. There was consensus among 78 out of every 100 participants that the type 2 criteria should be revised, and 70 out of every 100 supported moving spontaneous coronary dissection to the type 1 group (Taggart et al., 2025). The same experts rejected fixed criteria for supply and demand imbalance, because susceptibility to ischemia varies from person to person. The fifth definition followed this line: instead of thresholds for heart rate, blood pressure or hemoglobin, it requires proof of the ischemic consequence.

What stays the same

The pathological basis remains: infarction is death of myocardial cells from prolonged ischemia. The clinical diagnosis still requires acute myocardial injury plus symptoms or other evidence of ischemia, new ischemic changes or pathological Q waves on the electrocardiogram, imaging of acute coronary disease or new loss of viable myocardium with an ischemic pattern (Mills et al., 2026). Classification by electrocardiogram, with and without ST elevation, also continues, because it identifies who needs immediate reperfusion.

Myocardial injury: acute, chronic and the sex-specific threshold

The whole classification starts with troponin. Acute myocardial injury is a rise or fall in troponin I or T with at least one value above the sex-specific 99th percentile (Mills et al., 2026). Chronic injury is considered when two or more values are above this limit in a stable clinical setting, and is confirmed when a cardiac or non-cardiac disease with remodeling is found.

Raised troponin, by itself, says little about the mechanism. The definition's table of mechanisms lists inflammation (myocarditis, sepsis), hemodynamic stress (tachyarrhythmia, acute heart failure, pulmonary embolism), catecholamine stress (Takotsubo syndrome, subarachnoid hemorrhage, seizures), chemotherapy toxicity and trauma, including cardioversion and procedures (Mills et al., 2026). Pulmonary embolism and sepsis are common examples in which troponin rises without infarction.

The sex-specific threshold

Troponin concentration varies greatly between men and women. With high-sensitivity assays, the 99th percentile for women falls below the single threshold, and that for men falls above it. For the most recent generation of troponin T, the female limit is half the male limit in a global healthy reference population (Mills et al., 2026). Using a single threshold tends to miss injury and infarction in women, a topic explored further in the text on troponin with a sex-specific threshold.

The document separates two uses of troponin. In the accelerated 0 h and 1 h or 0 h and 2 h algorithms, the cutoffs serve to stratify risk and are usually the same for both sexes. The definition of injury, and therefore the final diagnosis of infarction, uses the sex-specific 99th percentile. Age matters as much as sex: one in three people over 75 with suspected infarction has raised troponin, compared with one in twenty under 50 (Mills et al., 2026).

What the large troponin studies taught

In the cohort that validated a very low single cutoff for troponin I, values below 5 ng/l ruled out infarction in about two thirds of patients without infarction on arrival, with a negative predictive value of 99.6 percent (Shah et al., 2015). The stepped trial that followed, with more than 48 thousand patients, showed the other side: the high-sensitivity assay reclassified 1771 people as having injury or infarction, without a drop in events at one year (Shah et al., 2018). The authors questioned whether the 99th percentile of a healthy population is the best cutoff point. The fifth definition responds by insisting on imaging to confirm the mechanism, instead of relying on troponin alone. The use of serial troponin together with clinical scores also appears in the text on the HEART Pathway.

When troponin misleads

A high troponin can reflect assay interference, without injury. Antibodies can form macrotroponin complexes, which reduce clearance and raise the measured value. In a population cohort of 19,499 people, biochemical analysis attributed 92.5 percent of discordant troponin I elevations to macrotroponin, and those with suspected macrotroponin had cardiovascular risk similar to that of people with normal values (de Bakker et al., 2026). The definition recommends suspecting interference when the value does not match the clinical picture and varies without a pattern across serial samples (Mills et al., 2026).

Primary infarction: acute coronary disease in all its forms

Primary infarction is the old type 1, broadened. It is considered in people with spontaneous onset of symptoms or signs of acute ischemia with acute myocardial injury, with no other acute condition or cardiac procedure explaining the picture (Mills et al., 2026). Atherothrombosis, from plaque rupture or erosion, remains the most common cause. The difference is that spontaneous coronary dissection, epicardial or microvascular vasospasm, coronary embolism and restenosis, stent thrombosis or graft failure more than 30 days after the procedure also fall here.

The criterion prioritizes sensitivity, so as not to miss an acute coronary disease. The diagnosis is probable when there is acute injury and at least one of these elements: symptoms compatible with ischemia, new ischemic changes on the electrocardiogram or new pathological Q waves. It is confirmed when imaging shows the acute coronary lesion or a new segmental wall motion abnormality, or absence of viable myocardium, with an ischemic pattern (Mills et al., 2026).

Angiography as part of the diagnosis

Under the new definition, coronary angiography, with intravascular imaging when needed, is essential to identify the mechanism and complete the classification. Optical coherence tomography and intravascular ultrasound help when angiography does not show the cause. Spontaneous dissection should be kept in mind especially in women under 50, during pregnancy and postpartum (Mills et al., 2026), a topic covered in the text on MINOCA, dissection and Takotsubo in women.

Where coronary angiography and cardiac imaging are out of reach, or are not appropriate, the final diagnosis of primary infarction can be made with troponin, electrocardiogram and clinical assessment. In that case, the code records unknown etiology. The definition includes a specific framework for low-resource settings, without biomarkers or imaging, acknowledging that much of the world diagnoses infarction this way (Mills et al., 2026).

A simpler name for the patient

The document notes that "primary infarction" is a synonym for spontaneous infarction, and that in practice the clinician may call it simply infarction (Mills et al., 2026). The term exists to separate this picture from the other two contexts, not to change the conversation at the bedside.

Secondary infarction: objective criteria for the imbalance

Secondary infarction is the part of the old type 2 that remained after acute coronary causes moved to primary infarction. It is considered when the acute injury results from another acute condition that unbalances myocardial oxygen supply and demand and produces symptoms or signs of ischemia (Mills et al., 2026). Demand rises in tachycardia and severe hypertension; supply falls in hypotension, hypoxia and anemia. Tachyarrhythmia is the most common trigger.

Here the criterion shifts and prioritizes specificity. The diagnosis is probable when, in addition to the clinical picture, there is known coronary disease or strong suspicion from the extent of ischemia on the electrocardiogram or from the troponin elevation. Confirmation, when imaging is feasible and appropriate, requires one of two things: obstructive coronary disease without an acute lesion, or a new segmental wall motion abnormality or loss of viable myocardium with an ischemic pattern (Mills et al., 2026).

What counts as obstruction

The document defines obstructive coronary disease as stenosis of at least 70 percent in an epicardial artery on angiography, or at least 50 percent with flow limitation on physiological assessment (Mills et al., 2026). If imaging shows an acute lesion, the diagnosis becomes primary infarction. Sepsis and bleeding can trigger atherothrombosis, and so primary infarction remains in the differential of the acutely ill patient.

Why require imaging

Troponin in current assays does not distinguish the ischemic cause from the non-ischemic one, and the electrocardiogram in imbalance often shows diffuse changes, without a defined territory. Most patients in this scenario have neither coronary obstruction nor new myocardial loss on imaging, and in these cases the diagnosis of secondary infarction is ruled out (Mills et al., 2026). The correct name becomes acute myocardial injury, with the cause investigated.

When ischemia persists and primary infarction is a possibility, invasive coronary angiography is indicated. Outside that, imaging usually waits for the acute condition to resolve, and coronary computed tomography angiography is an option. In very frail patients or those with limited life expectancy, imaging may not be appropriate, and clinical judgment decides (Mills et al., 2026).

The weight of hidden coronary disease

The requirement for imaging has a prognostic justification. In a cohort of 858 patients with raised troponin, those who had the old type 2 with atherosclerotic coronary disease had more major cardiovascular events than those who had type 1, with an adjusted hazard ratio of 1.95 (Knott et al., 2026). Coronary disease was present in about one third of people with type 2. Finding the obstruction changes treatment: secondary prevention, control of risk factors and, sometimes, revascularization. A recent narrative review recalls that this group is older, has more comorbidities and includes many women, in whom symptoms and biomarker kinetics make the diagnosis even harder (Kościanek et al., 2026).

Table: the three contexts side by side

Context	When to think of it	What confirms it
Primary infarction	Spontaneous onset, without another acute disease or procedure	Acute coronary lesion or new ischemic segmental change (Mills et al., 2026)
Secondary infarction	Another acute disease causes an imbalance between supply and demand	Stenosis of at least 70 percent without an acute lesion or new ischemic segmental change (Mills et al., 2026)
Procedure-related infarction	Up to 30 days after a cardiac procedure (Mills et al., 2026)	Coronary complication or new segmental change in the territory; both if the complication arose during the procedure
Acute myocardial injury	Troponin rises or falls without infarction criteria	Identified non-ischemic cause, such as myocarditis or Takotsubo

Procedure-related infarction: no biomarker threshold

The third category covers infarction that arises as a complication of a cardiac procedure within 30 days (Mills et al., 2026). Cardiac procedure here includes coronary angiography with or without intervention, structural intervention, catheter ablation and any open or minimally invasive cardiac surgery. The 30-day window is arbitrary, but it follows most complication studies.

The central change is that the diagnosis no longer depends on troponin multiples, and the criteria are the same for the percutaneous and the surgical route. Suspicion arises with acute injury and some clinical sign: symptoms, new ischemic changes, new Q waves or unexplained sudden worsening in a sedated patient. Confirmation comes from angiography, with a coronary complication such as side branch occlusion, dissection, slow flow, stent thrombosis or graft failure, or from cardiac imaging, with a new segmental change in the territory of the artery involved (Mills et al., 2026).

When both criteria are mandatory

If the complication happens during the procedure itself, or if the procedure is performed in the treatment of an acute infarction, the definition requires both things: the angiographic complication and the new segmental change. Some complications during the procedure are transient and resolved on the spot, and the dual requirement avoids labeling as infarction what left no damage (Mills et al., 2026).

The role left for troponin

Troponin remains essential for deciding who needs imaging. Soon after the procedure, the diagnosis becomes more likely when values rise relative to baseline and keep rising at 6 hours, with more than 5 times the limit after intervention, and at 24 hours, with more than 35 times after surgery (Mills et al., 2026). The authors themselves call these cutoffs arbitrary and say their sensitivity for ruling out imaging is unknown.

Prognosis studies show why a single cutoff does not work. After surgical revascularization, the median peak troponin T at 72 hours was 52 times the reference limit, and 30-day mortality rose markedly only when CK-MB was in the upper tertile together with troponin in the middle or upper tertile (Winter-Pölzl et al., 2026). Prognosis and diagnosis are different questions, and the fifth definition left diagnosis to imaging.

Outside the heart

Noncardiac surgery follows the same logic as the other contexts, with no category of its own. In a high-risk cohort, troponin rose after surgery in about one in five patients, and the cause was attributed more often to extracardiac mechanisms, especially kidney failure, than to cardiac causes (Schockaert et al., 2026). Inter-rater agreement was lower precisely in supply-demand imbalance, the same point the new definition tries to make objective.

Sudden death, unrecognized infarction, unstable angina and MINOCA

The old type 3, for cardiac death before testing, was removed from the classification. When infarction is the probable cause of a sudden death, the primary, secondary or procedure-related context applies, depending on the setting or the autopsy (Mills et al., 2026). Without an electrocardiogram, troponin or post-mortem examination, the code is unspecified infarction. Autopsy and post-mortem imaging alternatives confirm or rule out the diagnosis; the comparison between methods of investigating the cause of death appears in the text on verbal autopsy and complete autopsy.

Unrecognized infarction

Pathological Q waves on a routine electrocardiogram have limited specificity. The definition requires imaging confirmation, preferably late-enhancement magnetic resonance, before recording an unrecognized infarction (Mills et al., 2026). The document avoids the term "silent" when possible, because many patients, when asked, recall symptoms they did not value at the time.

Unstable angina

Unstable angina remains within the spectrum of acute coronary syndrome. It is considered when there are new ischemic symptoms at rest or with minimal effort, or sudden worsening of chronic angina, and serial troponins rule out infarction (Mills et al., 2026). With high-sensitivity assays, the diagnosis became rarer, because many cases previously labeled unstable angina were small infarctions.

MINOCA with another meaning

The acronym MINOCA came to mean myocardial injury with nonobstructed coronary arteries, and no longer names an infarction. The term is a working diagnosis for patients with a possible infarction picture and coronary arteries without stenosis of 50 percent or more on angiography (Mills et al., 2026). Magnetic resonance, ideally within two weeks, confirms infarction in a minority of cases and more often finds myocarditis or Takotsubo. A survey of British clinicians showed wide variation in the investigation of these patients and concern about the terminology itself, which helps explain the change (Sykes et al., 2026).

Chronic kidney disease

In chronic kidney disease, troponin is often raised and stable. In a study of nearly 47 thousand patients, the proportion with troponin I above the 99th percentile rose from 10 percent with a filtration rate of 90 or more to 66 percent below 30 ml/min/1.73 m² (Gallacher et al., 2021). The definition keeps the same diagnostic pathways in this group and reinforces the importance of serial troponin and imaging, rather than stage-specific thresholds (Mills et al., 2026). High troponin in these patients deserves investigation, and attributing it only to reduced clearance is a common mistake.

ICD-11: the classification reaches the record

One of the reasons the fourth definition had little effect on statistics was the lack of alignment with ICD-10. In two European countries, fewer than one fifth of patients meeting criteria for type 2 infarction received an infarction code (Mills et al., 2026). The fifth definition was written together with the World Health Organization and proposes ICD-11 codes for each context.

The code BA41.0 identifies infarction with ST elevation, and BA41.1, infarction without ST elevation. A sixth digit adds the mechanism of primary infarction (atherothrombosis, dissection, embolism, vasospasm, restenosis, stent thrombosis, graft failure, undetermined or unknown etiology), the letter A marks secondary infarction and the letters B and C mark that related to percutaneous or surgical procedure (Mills et al., 2026). Secondary infarction also receives the code of the acute condition that caused the imbalance. Unstable angina is in BA40, unspecified infarction in BA41.Z and unrecognized infarction in BA50.

At the time of publication, these codes were still under review by the WHO committee. Implementing ICD-11 will take time in each country, and the text about ICD-11 already in force explains where the transition stands. The document calls for training of clinicians and coders so that the classification appears in statistics.

Back to the case

In the patient from the opening, rapid atrial fibrillation with hypotension is a clear trigger of imbalance, and troponin rose. By the fifth definition, the diagnosis of secondary infarction is considered, but it is only confirmed with obstructive coronary disease without acute lesion or with a new segmental change on imaging (Mills et al., 2026). An echocardiogram after rate control and, if the course allows, a coronary CT angiography after discharge resolve the doubt. If both are normal, the correct record is acute myocardial injury due to tachyarrhythmia. If the pain persists, primary infarction returns to the top of the differential diagnosis, with serious causes on the list early. The same discipline of separating cardiac from non-cardiac appears in the evaluation of syncope of cardiac cause.

How NEXO helps train this reasoning

In NEXO, the case of the day is fictional, with a real clinical presentation and up to six progressive clues, and you propose the diagnosis by searching for the ICD-10 code.

The limits of use of the game are described on the NEXO economy page. If you research teaching of clinical reasoning, the page for researchers explains how to get in touch. Professionals who write or review cases find information in curation and in creators, and institutions can learn about the proposal on the partnerships page. Reading the electrocardiogram in other contexts, such as in hyperkalemia, completes this training.

Frequently asked questions

Has type 2 infarction ceased to exist?

The term left the classification. The coronary part of the old type 2 (dissection, embolism and vasospasm) moved to primary infarction, and the imbalance between supply and demand became secondary infarction, with confirmation criteria by imaging (Mills et al., 2026).

Are high troponin and tachycardia enough for secondary infarction?

That only makes the diagnosis possible. Confirmation requires obstructive coronary disease without acute lesion or new ischemic segmental change; without these findings, the correct name is acute myocardial injury (Mills et al., 2026).

Is there still a troponin threshold for infarction after angioplasty or surgery?

As a diagnostic criterion, it is gone. Values above 5 times the limit after intervention or 35 times after surgery, still rising, only indicate who needs imaging (Mills et al., 2026).

What does MINOCA mean now?

Myocardial injury with non-obstructive coronaries, a working diagnosis. After investigation, which usually includes magnetic resonance, the condition is reclassified as primary or secondary infarction or receives another diagnosis, such as myocarditis (Mills et al., 2026).

Why use troponin with sex-specific thresholds?

Because the 99th percentile in women is lower, and a single threshold fails to recognize injury and infarction in them. For the most recent troponin T, the female limit is half the male one (Mills et al., 2026).

Is the new definition a guideline?

It is a consensus statement from the European and American societies and the World Heart Federation, to be used together with national and regional guidelines (Mills et al., 2026).

Take-home messages

• Types 1 to 5 are gone; infarction is now primary, secondary or procedure-related.
• Primary infarction includes all acute coronary disease, and coronary angiography defines the mechanism.
• Secondary infarction requires imaging proof: obstruction without acute lesion or new segmental change.
• After a procedure, the diagnosis comes from angiography and imaging, with troponin only pointing to who should be investigated.
• Myocardial injury uses the 99th percentile by sex, and a high troponin without infarction criteria is injury.
• MINOCA is injury with non-obstructed coronary arteries, a working diagnosis.
• ICD-11 codes accompany each context and each mechanism.

Play today's case on NEXO.

References

1. Mills NL, Newby LK, Zaman S, et al. Fifth Universal Definition of Myocardial Infarction (2026): On behalf of the Joint European Society of Cardiology (ESC)/American College of Cardiology (ACC)/American Heart Association (AHA)/World Heart Federation (WHF) Task Force for the Universal Definition of Myocardial Infarction. Glob Heart. 2026. PMID 42666939. DOI 10.5334/gh.1578. Published simultaneously in Eur Heart J (DOI 10.1093/eurheartj/ehag101, PMID 42661425), J Am Coll Cardiol (DOI 10.1016/j.jacc.2026.07.025, PMID 42663357) and Circulation (PMID 42661421). Read in full in the Global Heart version.
2. Shah AS, Anand A, Sandoval Y, et al. High-sensitivity cardiac troponin I at presentation in patients with suspected acute coronary syndrome: a cohort study. Lancet. 2015;386(10012):2481-2488. PMID 26454362.
3. Shah ASV, Anand A, Strachan FE, 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(10151):919-928. PMID 30170853.
4. Gallacher PJ, Miller-Hodges E, Shah ASV, et al. Use of High-Sensitivity Cardiac Troponin in Patients With Kidney Impairment: A Randomized Clinical Trial. JAMA Intern Med. 2021;181(8):1237-1239. PMID 34096969.
5. Taggart C, Ferry AV, Chapman AR, et al. The assessment and management of patients with type 2 myocardial infarction: an international Delphi study. Eur Heart J Qual Care Clin Outcomes. 2025. PMID 40905366.
6. Knott JD, De Michieli L, Ola O, et al. Prevalence and Prognostic Implications of Atherosclerotic Coronary Artery Disease in Type 2 Myocardial Infarction. JACC Adv. 2026. PMID 41863201.
7. Kościanek J, Zabawa N, Stanaszek A, et al. Contemporary Challenges and Strategies for Diagnosing and Managing in Type 2 Myocardial Infarction: A Narrative Review. J Pers Med. 2026;16(8):438. PMID 42645932.
8. Winter-Pölzl L, Kletzer J, Lohmann R, et al. Using CK-MB and hs-cTnT to Predict 30-Day Mortality After CABG: A Combined Biomarker Risk-Stratification Approach. JACC Adv. 2026. PMID 42790282.
9. Schockaert BRBK, van Bruchem R, Stolker RJ, et al. Aetiology and associated outcomes of myocardial injury as defined by postoperative troponin elevation in high-risk patients after noncardiac surgery: a single-centre retrospective cohort study. BJA Open. 2026. PMID 42633194.
10. Sykes R, Berry C, Brown S, et al. Heterogeneity in the diagnostic management of MINOCA: a UK clinician survey. Open Heart. 2026. PMID 42901730.
11. de Bakker M, Thurston AJF, Hammarsten O, et al. Macrotroponin Complexes and Risk of Cardiovascular Events. Circulation. 2026. PMID 42708217.