Skip to content

Acute Myocardial Infarction (AMI) Revisited: Commonly Misunderstood Aspects

On February 2020 one of our colleagues wrote a blog reviewing acute myocardial infraction (MI) and the 4th Universal Definition. There are still many documentation issues involving this condition that call its validity into question.  This blog will address these using a Q & A format, with a focus on types 1 and 2 MI.


For this discussion, the following terms will be used: 

  • CTn: cardiac troponin 
  • URL (upper reference limit of normal range)
  • ACS: “acute coronary syndrome” is a concept that includes patients with STEMI, NSTEMI, or unstable angina 
  • AMI: acute myocardial infarction 
  • STEMI/NSTEMI: ST segment elevation myocardial infarction and non-ST segment elevation myocardial infarction– descriptions of EKG findings that are used to assist in triggering more rapid and specific treatment of patients who present with possible acute MI.  
  • “Significant change” in cTn: one of 2 criteria will establish presence of acute myocardial injury in types 1 and 2 AMI 
    • Detection of a rise and/or fall of cTn values with at least 1 value above the 99th percentile URL 
    • For chronically elevated cTn values, a change of greater than 20% between values* 

*The use of an acute >20% elevation for patients with chronic cTn readings above the 99th percentile URL of normal is a criterion used by Enjoin to provide defensible advice to clients. The issue is not clearly addressed by the 4th Universal Definition beyond procedurally related acute MIs with findings of all cTns remaining above the 99th percentile of URL. Types 4b and 4c MIs require the >20% elevation above baseline; for clarity we have extrapolated from that guidance to apply it to types 1 and 2 MIs. 

Question: if a patient comes to the ED with acute ST segment elevation on an EKG and has chest pressure and diaphoresis, does this satisfy the definition of acute MI? 

Answer: No. The 4th Universal Definition classifies acute MI by clinical classification– types 1-5. For all types (except for type 3), these require two components: 

  • Evidence of myocardial injury (most commonly by a rising and/or falling pattern of cTn values) as described above 
  • Symptoms or signs of acute myocardial ischemia that would suggest the myocardial injury is due to an ischemic process (instead of a non-ischemic process such as myocarditis, cardiac trauma or surgery, rhabdomyolysis or cardiotoxic medications) 

While ST elevations in the context of chest pressure and diaphoresis would be evidence to urgently take the patient to the heart catheterization lab, to definitively validate the acute MI diagnosis there still needs to be a significant troponin change.  

Question: if there is only a single normal cTn in the above scenario, and a culprit lesion is identified on coronary artery catheterization and a thrombectomy is performed, are you suggesting that single cTn is insufficient to support an acute MI? 


Answer: Acute MI is not adequately validated. If a single cTn is negative and no follow-up is drawn, we are unable to distinguish between an acute MI and an aborted MI. There is no ICD-10-CM code for aborted MI, making the distinction critical.  

Coding Clinic, 2nd Quarter 2001, Pages 7-8 instructs unstable angina would be reported as the principal diagnosis (Note:  if CAD is determined to be present and documented, then coding rules would require selection of CAD, and not unstable angina, as the PDx with unstable angina as a SDx). 

Unstable angina would exclude the patient from the acute myocardial infarction re-admission/mortality cohorts for hospital-centric quality data. 


Question: What if that same patient has an initially elevated troponin-I of 0.8 ng/ml? Is that sufficient to establish an acute MI? 

Answer: The key phrase is “a significant troponin change.” A single troponin cannot be evidence of a change. This elevated troponin could represent: 

  • A chronic elevation, which if trended, may not show significant change and thus support only an aborted MI 
  • It may represent a significant change which would validate an acute MI. Such a significant change could be established by finding a troponin in recent prior records that was either  
    • Below the 99th percentile of the upper reference limit (URL) or 
    • Greater than 20% lower than current troponin. 

Alternatively, query to clarify if the single troponin is likely to represent 

  1. A clinically significant change supportive of acute myocardial injury 
  2. A clinically insignificant (“flat”) troponin change with chronic troponin elevation 
  3. Other 
  4. Unable to determine 
  • Only the first response would validate the acute MI. 

Question: a patient has EKG evidence of acute ischemia and serial cTns as follows: 0.41, 0.51, 0.42. How do I determine if these are flat or clinically significant? 

Answer: we feel that to defend against a denial, in a patient with all cTns above the URL of normal, the patient ought to have >20% change. Here is the math: 

  • Look for the lowest value (here it is 0.41) 
  • Multiply by 1.2 (0.41 x 1.2 = 0.504) 
  • Determine if the patient’s highest troponin value exceeds the calculated cTn (patient peak cTn is 0.51, which is higher than the calculated cTn of 0.504) 
  • In this case the change is significant, and an acute MI is supported by the presence of acute myocardial injury (cTn changes) that appears to be due to acute myocardial ischemia (as evidenced by the EKG changes). 

Question: I learned that a “silent MI” does not need to have any symptoms suggestive of acute myocardial ischemia. Does that mean only a significant troponin change is necessary to validate acute silent MIs? 

Answer: in patients who survive, isolated troponin changes do not validate any type of MI, including silent MIs. In such cases, there must still be (along with the significant troponin change) clinical evidence to support acute myocardial ischemia. This can be evidenced by any one of the following: 

  • Ischemic EKG changes that are either definitively new or assumed to be new 
  • Development of pathological Q waves
  • Imaging evidence of new loss of viable myocardium or new regional wall motion abnormality in a pattern consistent with an ischemic etiology; 
  • Identification of a coronary thrombus by angiography including intracoronary imaging 

Question: a provider notes “patient does not have ACS.” Does this exclude the possibility of a type 2 MI being present? 

Answer:  Some sources advise that ACS only applies to unstable angina and type 1 MIs (STEMI and NSTEMI). The 4th Universal Definition paper, however, does not limit ACS to apply just only type 1 AMI. Therefore, “no ACS” would exclude unstable angina and any type of MI, including a type 2 MI. 


Question: when a STEMI is diagnosed, is it appropriate to assume it is a type 1 MI? 

Answer: “STEMI” and “NSTEMI” are EKG descriptions and not formal clinical classifications of MI like types 1-5.  

The Official ICD-10-CM codebook improperly assumes a link between either unspecified STEMI or NSTEMI and type 1 MI. Any clinical type of MI can be either a STEMI or NSTEMI, including a type 2 MI. While rare, type 2 STEMIs do occur. A type 2 MI has a directive to code first the underlying cause, whereas a type 1 can be principal diagnosis; clarifying this distinction is imperative for sequencing. Additionally, when a patient is coded as a STEMI or NSTEMI that ought to be a type 2 MI, the patient is inappropriately included in potential readmissions and mortality quality cohorts for patients with an index admission for acute myocardial infarction. A type 2 AMI description would exclude such patients. 

It is useful to review the coronary angiography procedural description for further insights into clinical type of MI. If a “culprit vessel” with plaque disruption is described (particularly if an intervention on that culprit vessel is undertaken), this strongly suggests a type 1 MI. Absence of such a description suggests a type 2 MI, particularly when a clear underlying etiology for a coronary arterial oxygen supply/demand mismatch is evident (e.g., sepsis, profound anemia, coronary artery vasospasm, etc.). 


Question: a 50-year-old lady who had no risk factors was admitted with an acute NSTEMI. She had recently found out her daughter was severely injured in a bicycle vs automobile accident. The mother’s heart catheterization demonstrated a spontaneous circumflex coronary artery dissection which was stented.  Minimal plaque was noted. No link was made between the NSTEMI and the coronary dissection. My CDS colleague and I were debating whether to query for a type 1 or type 2 acute NSTEMI. Which is correct? 

Answer: This is unequivocally a type 2 AMI. A type 1 AMI is due to a localized acute coronary artery plaque disruption (rupture or erosion). A spontaneous dissection is neither of these. The dissection led to a circumflex coronary artery oxygen supply deficit, thus leading to a type 2 AMI. Query choices could be:  

  • Unspecified acute NSTEMI 
  • Type 2 NSTEMI from coronary artery dissection 
  • Other (please specify) 
  • Unable to determine 

Remember that if the type 2 NSTEMI is selected, sequencing rules would place the type 2 MI (I21.A1) as ODx and Coronary artery dissection (I25.42) as PDx. 


In summary, remember the following when attempting to validate type 1 or type 2 acute MI: 

  • Acute myocardial injury must be present. It is evidenced by a significant change in cTns (see above for specifics) 
  • Acute myocardial ischemia must be evidenced by either signs (eg, acute EKG ischemic changes, imaging evidence of acute wall motion abnormality, etc.) or symptoms (eg, chest pressure, diaphoresis, etc.) 
  • When only an EKG description of STEMI or NSTEMI is documented, look for evidence to clarify type (eg, type 1 vs type 2 AMI).  


Thygusen K, et al. Fourth Universal Definition of Myocardial Infarction (2018), Circulation. 2018;138:e618–e651 

Coding Clinic for ICD-9-CM 

2021 ICD-10-CM Codebook 


Back to All Blog Posts

Sign Up For Updates From Enjoin