ESC · ACC · AHA · WHF | 28 August 2026
The Numbers Are Gone
What the Fifth Universal Definition of MI Actually Changes
Bishnu Subedi, MD, FACC
Types 1 through 5 are retired. Three named categories replace them, troponin thresholds go sex-specific, and "Type 2 MI" finally has a bar to clear.
For nineteen years we have been writing numbers in charts. Type 1. Type 2. Type 4a. On 28 August 2026, the European Society of Cardiology, the American College of Cardiology, the American Heart Association and the World Heart Federation retired them.
The Fifth Universal Definition of Myocardial Infarction replaces the numerical classification with three named categories based on mechanism: primary, secondary and procedure-related myocardial infarction. It also mandates sex-specific troponin thresholds, strips the biomarker multiples out of the periprocedural criteria, and demotes MINOCA from a diagnosis to a working diagnosis.
Presented at ESC Congress 2026 and published simultaneously in the European Heart Journal, JACC, Circulation and Global Heart. The task force was co-chaired by Professor Nicholas Mills for the ESC and Professor Kristin Newby for the ACC/AHA. Mills was blunt about the motivation: the previous definition's numerical categories "were not always easy to apply in clinical practice." Newby framed the payoff at the bedside — clinicians can now discuss the cause of an MI with a patient in terms the patient can actually follow.
This is not a relabelling exercise. Four things genuinely change.
1. Three names, and three things that crossed categories
The translation is mostly one-to-one — Type 1 becomes primary MI, Type 2 becomes secondary MI, Types 4a and 5 merge into procedure-related MI. The interesting part is what moved.
Late stent thrombosis, in-stent restenosis and late graft failure are now primary MI. Under the numbered system these were permanently procedural — Type 4b and 4c — however many years after the index PCI they occurred. The 2026 framework treats a coronary event beyond the 30-day procedural window as de novo disease, which is what it pathophysiologically is. Thrombus on a stented segment at fourteen months behaves like thrombus on a plaque.
Type 3 is gone as a category. It described a circumstance — cardiac death before biomarkers could be drawn — rather than a mechanism, which is why it always sat awkwardly beside the others. It is replaced by explicit criteria for diagnosing MI following sudden death, after which the case is named for whatever mechanism is established.
The translation, at a glance
| You used to write | You now write |
|---|---|
| Type 1 | Primary MI |
| Type 2 | Secondary MI — but see below |
| Type 3 | No longer a category |
| Type 4a | Procedure-related MI |
| Type 4b, within 30 days | Procedure-related MI |
| Type 4b, beyond 30 days | Primary MI |
| Type 4c | Primary MI |
| Type 5 | Procedure-related MI |
| MINOCA | A working diagnosis, then reclassify |
2. Secondary MI finally has a bar to clear
This is the change most likely to alter your hospital's numbers, and the one most worth reading carefully.
Under the 2026 framework, secondary MI requires more than an imbalance and a rising troponin. It requires either obstructive coronary disease without acute coronary pathology, or a new regional wall motion abnormality.
Consider what that excludes. A septic patient in fast AF with a troponin that rises and falls, no prior coronary imaging and no echocardiogram, does not have secondary MI. That patient has acute myocardial injury — which is not a lesser diagnosis, carries a serious prognosis of its own, and is simply the accurate term.
Watch for this
"Type 2 MI" had drifted into being a way of noting a raised troponin in a sick patient rather than making a diagnosis. The new criteria stop that. Expect your secondary MI counts to fall and your acute myocardial injury counts to rise — and expect somebody to misread that as a clinical improvement.
3. Sex-specific troponin thresholds become the standard
Healthy women have lower circulating troponin than men, principally because left ventricular mass is lower. A single overall 99th percentile therefore sits above the true female limit — and women with genuine myocardial injury get reported as normal.
The magnitude is not subtle. On a widely used hs-cTnI platform the published sex-specific limits are roughly 16 ng/L for women against 34 ng/L for men, with an overall limit of 26 ng/L. Every woman falling between 16 and 26 ng/L has been sitting in a diagnostic blind spot. The High-STEACS programme showed a decade ago that closing it roughly doubles the proportion of women diagnosed with MI, and that those newly identified women carry real risk while being less likely than men to reach angiography or secondary prevention.
Two caveats worth stating plainly. First, correcting the threshold identifies injury; the ischaemia gate is unchanged, and these women still need the same diagnostic reasoning as anyone else. Second, this is a laboratory information system project, not a memo — the reference limit, the result flag, the chest pain pathway, the decision support and the audit queries all have to ship in the same release, or you will produce results flagged abnormal that your pathway still treats as normal.
4. Procedure-related MI: one criterion set, no biomarker multiple
The 5× and 10× thresholds are gone. PCI and cardiac surgery are now assessed by identical criteria within a 30-day window, and the diagnosis turns on a demonstrated coronary complication — dissection, side-branch occlusion, no-reflow, acute closure, a graft problem — rather than on how high the troponin climbed.
Anyone who has adjudicated a revascularisation trial will recognise why this matters. Holding the surgical arm to a numerically higher bar than the percutaneous arm made every composite endpoint containing periprocedural MI internally inconsistent. That structural problem is now fixed.
The corollary for clinical work: troponin release after uncomplicated cardiac surgery is procedural myocardial injury. Expected, prognostically meaningful, worth recording — and not an infarction. When you are asked to see a post-procedure patient with a rising troponin, the first question is no longer "how high?" but "what did the operator or surgeon actually see?"
What to do on Monday
- Call your laboratory. Ask whether the troponin report applies a sex-specific 99th percentile and whether it says so on the result. If not, every "normal" troponin in a woman in that band is being mishandled by your pathway right now. Until it changes, know your assay's female limit and interpret manually.
- Stop writing bare "NSTEMI". Write the category and the mechanism as a clause: "Non-ST-elevation primary myocardial infarction due to plaque rupture, mid-LAD." Coders code the diagnosis line, not your narrative.
- Use "acute myocardial injury" out loud. In handover, on the ward round, in the note. If it is harder to say than "MI", people will say "MI".
- Get the echo. Under the new criteria, a new regional wall motion abnormality is often the deciding evidence between secondary MI and acute myocardial injury.
- Do not let a patient leave with "MINOCA" as the final diagnosis. Either you can name the mechanism, or the discharge summary says which investigation is outstanding and who is arranging it.
- Warn your audit team. Every time series that crosses this transition has a break in it. Map your historical codes to the new categories before anyone draws a trend line.
The part that is easy to miss
Two smaller changes deserve more attention than they will get.
The document provides explicit guidance for settings without access to coronary and cardiac imaging — the first universal definition to acknowledge that most of the world's infarctions occur where the diagnostic pathway it describes cannot be performed. And the proposed ICD-11 alignment means the three categories can finally be recorded distinctly and compared across health systems, with better surveillance of mechanisms that have been chronically under-captured. The societies name spontaneous coronary artery dissection as the example — a predominantly female condition that a mechanism-blind coding system was never going to count properly.
The honest caveat
Everything above reflects the societies' release material and the first wave of reporting. The primary document is the authority on its own wording, and the exact criteria — particularly the supporting evidence required for secondary MI and the specifics of the procedure-related definition — should be read in full before anyone rewrites a protocol. Assay-specific numbers quoted here are illustrations of magnitude, not cut-points to adopt.
But the direction is unambiguous, and it is the right one. A number could be written by a clinician who had noticed a raised troponin and gone no further. A name is a claim about what happened — and the 2026 criteria now ask you to have evidence for it.
Sources
Fifth Universal Definition of Myocardial Infarction (2026), J Am Coll Cardiol, doi:10.1016/j.jacc.2026.07.025 · ESC guideline page · ESC Congress 2026 press materials · Thygesen K, et al. Fourth Universal Definition of Myocardial Infarction (2018).
Educational commentary for a professional audience. Not a clinical guideline and not a substitute for the primary document.
