Fourth universal definition of myocardial infarction

The new Fourth Universal Definition of Myocardial Infarction, published in August, introduced the term “myocardial injury,” defined as an elevated troponin value above the 99th percentile upper reference limit.


The new Fourth Universal Definition of Myocardial Infarction (MI), published in August, introduced the term “myocardial injury,” defined as an elevated troponin value above the 99th percentile upper reference limit (URL). In chronic myocardial injury, troponin remains elevated at consistent levels. Myocardial injury is considered “acute” when an elevated level increases further or there is a decrease from the elevated level, indicating that the earlier level had been acutely elevated.

Image by Thinkstock
Image by Thinkstock

MI is defined as acute myocardial injury with clinical evidence of myocardial ischemia. The diagnosis of myocardial ischemia involves application of the physician's clinical judgment to the patient's clinical circumstances. The concept of “myocardial injury” replaces the term “necrosis,” allowing a clearer distinction between ischemic and nonischemic causes.

The recommended biomarkers for evaluation of myocardial injury and MI are troponins. Under the new universal definition, troponin is the only recommended diagnostic biomarker for MI. Creatine kinase-MB and myoglobin are no longer recommended. Creatine kinase-MB is not as accurate or reliable as troponin.

Troponin reference ranges depend on the methodology of the specific test used. It is important to know the 99th percentile reference range for your hospital's lab. Troponin I is more accurate than troponin T, and high-sensitivity troponin (hs-cTn) has been newly endorsed to supplement troponin I or T.

Causes of myocardial injury may be ischemic (such as MI) or nonischemic. Ischemic causes include coronary artery disease (CAD), supply-demand mismatch, and postprocedural infarction. Nonischemic causes include heart failure, myocarditis, cardiomyopathy, catheter ablation, defibrillation, cardiac contusion, and systemic causes such as sepsis, cerebrovascular accident/hemorrhage, pulmonary embolism, hypertension, chemotherapy, and infiltrative diseases (e.g., amyloidosis, sarcoidosis).

When troponin levels are persistently elevated and do not fall, it indicates a chronic nonischemic myocardial condition causing myocardial injury. Troponin levels are commonly elevated without any myocardial injury in chronic kidney disease.

The types of MI according to the Fourth Universal Definition are shown in Table 1. The new definition includes some changes from those outlined in the November 2017 Coding Corner addressing the Third Universal Definition. The diagnostic criteria for myocardial ischemia have been enlarged and described more specifically for types 1 and 2, with nonischemic causes described in greater detail. Type 4c has been affirmed where it previously was not.

Type 1 MI includes ST-elevation MI (STEMI), Q-wave infarction, and non-ST-elevation MI (NSTEMI) with coronary thrombosis due to CAD. STEMI or Q-wave infarctions are usually self-evident with characteristic electrocardiogram (EKG) findings and typically require immediate reperfusion therapy or percutaneous coronary intervention (PCI). NSTEMI is identified by elevated troponin levels with only minor nonspecific ST/T-wave changes or even a normal EKG.

The diagnosis of type 1 and 2 MI requires at least one of the following: symptoms of acute myocardial ischemia, new ischemic echocardiogram changes, imaging evidence of new loss of viable myocardium or of new regional wall motion abnormality consistent with an ischemic etiology, or development of pathological Q waves (in type 2, usually only when due to coronary embolism or dissection). Another criterion for type 1 is the identification of a coronary thrombus by angiography or by autopsy.

Type 2 MI is characterized by an elevated troponin level primarily due to supply/demand imbalance without coronary thrombosis (not primarily due to CAD). Common causes include coronary artery spasm, embolism, or dissection, sustained tachy- or brady-arrhythmia, severe anemia, hypotension, shock, or severe hypertension.

ICD-10-CM codes for the types of MI are shown in Table 2. Prior to October 1, 2017, type 2 MI was assigned the same code as NSTEMI, but now has its own separate code (I21.A1) distinguishing it from NSTEMI. Type 2 MI should not be identified as NSTEMI since the latter is due to CAD with thrombosis and improperly describes the patient's true condition and prognosis. Type 2 and NSTEMI are two completely different conditions having different causes, management, prognosis, and outcome. Types 4a, b, c, and 5 have complex definitions best left to the judgment of cardiologists and cardiac surgeons involved in the care of the patient.

STEMI codes have specificity to identify the involved artery or myocardial wall. Codes for all the other types of MI do not have any greater specificity. If the type of MI is not identified, the code assigned is I21.9.

Ask Dr. Pinson

Q: I know the Sepsis-2 definition of severe sepsis and the Sepsis-3 criteria do not mention non-ST-segment elevation myocardial infarction (NSTEMI) as acute organ dysfunction that would qualify for severe sepsis (code R65.20). However, if the clinician specifically stated that NSTEMI (code I21.4) was due to sepsis, could the R65.20 code be assigned?

A: This is a very interesting question. Yes, under Sepsis-2 criteria, but not Sepsis-3, a clinician may specify that an MI is a consequence/manifestation of severe sepsis. However, there would need to be a statement confirming it as “acute organ dysfunction” due to sepsis or as a manifestation of severe sepsis. Just documenting “MI due to sepsis” would not be enough to allow assignment of code R65.20.

That said, I should mention that a type 2 MI (primarily due to myocardial oxygen supply/demand mismatch), coded I21.A1, is the most likely diagnosis for MI related to sepsis. NSTEMI (type 1 primarily due to coronary artery disease) is not common in these circumstances.

Q: A brain MRI on a patient with head injury from a fall showed a subdural hematoma with midline shift and mass effect suggesting brain compression. I believe that brain compression (code G93.5) is not separately coded with traumatic brain injuries, so a query for documentation of brain compression would be pointless. Is that correct? Is there anything else that should be documented or queried?

A: This is a really insightful and important question. Brain compression is included in and coded as diffuse traumatic brain injury, so a query on that issue would be pointless.

Midline shift and mass effect can be associated with brain herniation (also coded G93.5), which is not indexed as “traumatic” but does have an “excludes 1” note in the ICD-10-CM tabular list for diffuse and focal traumatic brain injury (S06.2- and S06.3). It is not separately coded in those situations.

However, there is no prohibition of separately coding brain herniation (G93.5) for traumatic cerebral edema (subcategory S06.1); other specified or unspecified traumatic brain injury (subcategories S06.8 and S06.9, respectively); or traumatic epidural, subdural, and subarachnoid hemorrhage (subcategories S06.4, S06.5, and S06.6 respectively). Brain herniation associated with these conditions should be queried and coded since it can impact the diagnosis-related group.