Monday, May 27, 2024

Chest pain: Are these really "Nonspecific ST-T wave abnormalities", as the cardiologist interpretation states?

Written by Jesse McLaren, with a very few edits by Smith


A 60-year-old presented with chest pain. The ECG did not meet STEMI criteria, and the final cardiology interpretation was “ST and T wave abnormality, consider anterior ischemia”. But are there any other signs of Occlusion MI?

There’s only minimal ST elevation in III, which does not meet STEMI criteria of 1mm in two contiguous leads. But STEMI criteria is only 43% sensitive for OMI.[1] But there are multiple other abnormalities that make this ECG diagnostic of Occlusion MI, localized likely to the right coronary artery:

1. AV block, which can be a transient complication of RCA occlusion since it perfuses the AV node [2]

2. Inferior hyperacute T waves, which have been added to the 2022 ACC consensus on chest pain as a “STEMI equivalent”[3]

3. Reciprocal ST depression and T wave inversion in aVL, which is 99% sensitive for inferior OMI.[4]

4. Ischemic ST depression V2-4, which is 97% specific for posterior OMI[5]


So this patient with chest pain has an ECG that is diagnostic of inferior and posterior OMI complicated by 3rd degree AV block, likely representing an RCA occlusion. (See Ken Grauer discussion below for more on AV blocks in today's case).

These findings are diagnostic even without comparison with an old ECG.  

But they are even more striking when you do compare with the old one: 

Here’s a side-by-side comparison of the inferior leads from the baseline ECG (left) and new ECG (right) showing hyperacute T waves with reciprocal change in aVL:

The hyperacute T waves are not tall in absolute terms (only 4mm) but they are tall relative to the QRS, broad based, bulky and look inflated. The T wave inversion in aVL is not necessarily abnormal because it is concordant with the negative QRS, but it is huge relative to the QRS, reciprocal to the hyperacute T wave in III

The ECG also shows how reciprocal change can be the more obvious finding of OMI. Here are leads aVL and V2 of the new ECG, flipped upside down:


With the leads flipped, aVL has an obvious hyperacute T wave larger than the entire QRS complex, while V2 has ST elevation.

If simply flipping leads upside down can show signs of occlusion, then there’s a problem with a paradigm that dichotomizes ST segment deviation between elevation and depression as seen on the ECG paper, because they can reflect the same current of injury in the patient from a different view.[6] When there’s a current of injury towards a territory it produces reciprocal change in its wake, which is sometimes more obvious than the territory of injury (for inferior OMI) and is sometimes the only sign on the 12 lead (for posterior OMI). AVL and V2-V3 are oriented to reflect the reciprocal change to inferior and posterior OMI respectively – but this helpful information is disregarded by STEMI criteria because these leads manifest ST depression instead of ST elevation. Hence the first ECG was labeled 'anterior ischemia' based on ST depression, rather than identifying this as reciprocal from posterior OMI.


With a “STEMI negative” ECG and ongoing chest pain, the ECG was repeated in 10 minutes:

Now the AV block has improved from 3rd to 1st degree. But there are ongoing inferior hyperacute T waves, reciprocal TWI in aVL, and ischemic STD in V2-3. The final interpretation is now “nonspecific ST abnormality” based on STEMI criteria, but it is still very specific for infero-posterior OMI. There’s also more greater reciprocal change - here’s aVL flipped, with a T wave that now towers over the tiny QRS:

But now the patient has now had serial ECGs which are “STEMI negative”. So a troponin level was sent: 90 minutes later it returned at 250ng/L (normal <26 in males and <16 in females). Then the ECG was repeated again:

Now the ECG just barely meets STEMI criteria, along with more obvious hyperacute T waves.

A 15 lead was done 7 minutes later:

Now greater inferior ST elevation, and some convex ST segment in V5-6. Posterior leads V8-9 are also elevated, but this only confirms the posterior OMI that was seen on the first ECG.

Here’s a view of the last ECG showing V2-3 flipped (on the left) compared with V8-9 (right) :

The flipped V2-3 show clear ST elevation and hyperacute T waves, while V8-9 have lower voltage and proportionally less obvious ST/T changes which add no additional value.

Since the ECG now meets STEMI criteria, a code STEMI was activated. There was a 99% mid RCA occlusion, and peak troponin was greater than 25,000ng/L (upper limit of assay)

Discharge ECG showed resolution of changes, and subtle inferior reperfusion T wave inversion:

This patient received the standard of care under the current STEMI paradigm: serial ECGs for ongoing chest pain, and waiting for troponin level for "STEMI negative" ECGs. But this led to a 90 minute delay to reperfusion for an occlusion that could have been identified on the first ECG. This is typical of the current paradigm, as OMI findings can identify occlusions a median of 1.3 hours earlier.[7] 

Earlier intervention could have saved myocardium possibly a huge amount of myocardium, and could easily have been done if the mind set and approach was OMI-NOMI rather than STEMI/NSTEMI and if the Queen of Hearts app was used.


There is now AI trained to identify OMI.[8] If this been available in this case, the diagnosis could have been made on the first ECG, and saved 90 minutes of reperfusion delay. 

Here’s the Queen of Heart’s interpretation on the first ECG, along with explainability:

Even without comparison to the prior, and without any reference to the minimal ST elevation in III, the Queen identifies OMI with high confidence based on inferior hyperacute T waves, reciprocal T wave inversion in aVL, and anterior ST depression. She is not bothered by lead orientation, and is equally comfortable identifying OMI by the reciprocal hyperacute T waves or in the reciprocal ST depression.

Click here to sign up for Queen of Hearts Access


Take away

1. STEMI criteria misses the majority of OMI, and leads to delayed reperfusion for STEMI(-)OMI

2. Findings of RCA occlusion may include AV block, subtle inferior ST elevation, hyperacute T waves, reciprocal STD/TWI in aVL, and anterior STD from associated posterior OMI

3. Reciprocal change can be more obvious than the primary injury, and posterior leads rarely add any additional value to the ischemic STDmaxV1-4 that can be seen on 12 lead

4. OMI ECG findings can lead to rapid diagnosis, and can be widely disseminated through AI



1. De Alencar Neto. Systematic review and meta-analysis of diagnostic test accuracy of ST-segment elevation for acute coronary occlusion. Int J Cardiol 2024

2. Nikus et al. Conduction disorders in the setting of acute STEMI. Curr Cardiol Red 2021

3. Kontos et al. 2022 ACC expert consensus decision pathway on the evaluation and disposition of acute chest pain in the emergency department: a report of the American College of Cardiology solution set oversight committee. J Am Coll Cardiol 2022

4. Bischof et al. ST depression in lead AVL differentiates inferior ST-elevation myocardial infarction from pericarditis. Am J Emerg Med 2016

5. Meyers et al. Ischemic ST-segment depression maximal in V1-4 (versus V5-6) of any amplitude is specific for occlusion myocardial infarction (versus nonocclusive ischemia). J Am Heart Assoc 2021

6. Phibbs and Nelson. Differential classification of acute myocardial infarction into ST- and non-ST segment elevation is not valid or rational. Ann Noninvasive Electrocardiol 2010

7. Meyers et al. Accuracy of OMI ECG findings versus STEMI criteria for diagnosis of acute coronary occlusion myocardial infarction. IJC Heart and Vasc 2021

8. Herman et al. International evaluation of an artificial intelligence-powered electrocardiogram model detecting acute coronary occlusion myocardial infarction. Eur Herat J Digital Health 2024


MY Comment, by KEN GRAUER, MD (5/27/2024):


Superb presentation by Dr. McLaren on a case that clearly could have been (should have been) diagnosed as acute infero-postero OMI based on the initial ECG.
  • The fact that this ECG was initially diagnosed as showing, “ST-T wave abnormality — consider anterior ischemia” — means that the provider failed to recognize the AV block. The reason this happens is simply because of a failure to spend the 2-3 seconds it takes for an ‘educated eye” to routinely scan the long lead II rhythm strip, looking to see if there are upright P waves that precede each QRS complex with a fixed PR interval. Try this. Isn’t it easy to see in no more than 2-3 seconds that regular P waves do not precede each QRS with a fixed PR interval in today’s initial ECG?

  • PEARL: If you do not want to overlook 2nd- and 3rd-degree AV Blocks — Make it a habit as soon as you are given any ECG to always spend your first 2-3 seconds (which is all it should take) ensuring that P waves are present and are (or are not) related to neighboring QRS complexes
  • As per Dr. McLaren — the rhythm in today’s initial ECG is 3rd-degree ( = complete) AV Block — here at the AV Nodal level, given the narrow QRS complex of ventricular beats at an escape rate ~50/minute (well within the 40-60/minute usual range for AV nodal escape).
  • PEARL: In my experience — the BEST clue to distinguish 2nd-degree from 3rd-degree AV block — is that most of the time with 3rd-degree, the ventricular response will be regular. This is because most of the time escape rhythms are regular (or at least fairly regular). IF instead of being regular — you see a ventricular beat that clearly occurs earlier-than-expected — the chances are that the degree of AV block is not complete.
  • The initial ECG in today’s case is complete AV Block beause: i) The atrial rhythm is quite regular; ii) The ventricular rhythm is quite regular; andiii) P waves are not related to neighboring QRS complexes (ie, P waves “march through the QRS complexes” — which is easily established by noting the constantly changing PR interval) — andiv) No P waves are conducted despite having adequate chance to do so (ie, While in today’s initial ECG, we would not expect those P waves with a short PR interval to conduct — there is more than adequate opportunity for those P waves occurring in the middle of the R-R interval to conduct in this tracing, yet they fail to do so).
  • As per Dr. McLaren — it is common to see AV Block in patients with acute inferior, posterior and/or infero-postero OMI. The “culprit” artery will almost always be either the RCA or LCx, as either vessel may supply the AV node (the “culprit” being the RCA in today’s case)PEARL: Even if providers failed to appreciate the diagnostic features of acute infero-postero OMI in today’s initial ECG —The failure to spend those 2-3 seconds surveying the long lead II rhythm strip should have prompted recognition that the chest lead ST-T wave depression was unlikely to be the result of “anterior ischemia” — and much more likely to reflect acute infero-postero OMI — since 2nd- and/or 3rd-degree AV block with a narrow QRS is common with infero-postero OMI — but it is rare with anterior ischemia/OMI.

  • PEARL: When AV block with a narrow QRS occurs in association with acute inferior and/orposterior OMI — there is often a step-wise progression and/or regression (ie, from 1st-degree — then to 2nd-degree, Mobitz Type I — then to 3rd-degree AV block at the AV nodal level — or — from 3rd-degree — to Mobitz I 2nd-degree — to 1st-degree, IF the patient began with complete AV block). As per Dr. McLarren, the “good news” — is that most of time, the associated AV block will resolve, especially if there is prompt reperfusion of the “culprit” artery.

Playing “Devil’s Advocate”: In view of this last Pearl — I suspect that the 2nd ECG in today’s case (done just 10 minutes after the 1st ECG) represents 2nd-degree 2:1 AV Block of the Mobitz I Type — and not just sinus rhythm with 1st-degree AV Block.
  • We often see both 1st-degree AV block (ie, of conducted beats) — and 2nd-degree Mobitz I AV block together!
  • The atrial rate in today’s 1st ECG is just under 100/minute — and the junctional escape rate is ~50/minute.
  • The ventricular rate remains the same ~50/minute in the 2nd ECG — but for this to be sinus rhythm with 1st-degree, the atrial rate would have to have decreased from just under 100/minute to 50/minute over a 10-minute period.
  • The very subtle “extra” widening of the T wave peak in the inferior leads, and of the inverted T wave in lead aVL occurs at an appropriate moment in the R-R cycle to potentially “hide” a 2nd P wave (especially if there is some ventriculophasic sinus arrhythmia — as there so commonly is with 2nd-degree AV block). And, if a 2nd P wave is “hiding” within each T wave — this would represent an atrial rate that is still just under 100/minute — instead of having to postulate that the atrial rate so rapidly decreased to ~50/minute.
  • The atrial rate in the 3rd ECG is clearly much faster than 50/minute.
  • And, the reason I so carefully considered the possibility of 2:1 AV block in today’s 2nd ECG — is that as per the above Pearl, it is common for 3rd-degree AV block at the AV Nodal level to resolve with stepwise regression (ie, passing from 3rd-degree — to 2nd-degree, Mobitz I — and then to 1st-degree, before finally returning to normal sinus conduction).

  • To EMPHASIZE: I can not prove whether or not my suspicion that the 2nd ECG in today’s case represents 2:1 Mobitz I instead of sinus rhythm with 1st-degree.  To prove this — I’d need to see serial ECG monitoring revealing what happens with respect to the atrial and ventricular rates as the case evolves.
  • That said — Clinically, it does not matter if the 2nd ECG shows sinus rhythm at ~50/minute with 1st-degree vs Mobitz I 2nd-degree with 2:1 AV block and 1st-degree — because the ventricular rate is the same in either case and chances are that the AV block will resolve with treatment.

  • PEARL: The above said — there are times when it will matter clinically whether your patient with acute infero-postero OMI is in sinus rhythm vs 2nd-degree, Mobitz I with 2:1 AV block (in which the 2nd P wave is hidden within T waves). Awareness of when and how to look for 2:1 block in subtle cases (as I describe above for today’s 2nd ECG) can greatly facilitate identification.


ADDENDUM (5/28/2024):
I wrote the above comment yesterday — while traveling, without large screen computer and without calipers.
  • To EMPHASIZE: Calipers were not needed to optimally treat the patient in today's case — as Dr. McLaren superbly demonstrates in his discussion. But I find it helpful after the emergency has passed to reflect on events for my own edification — so that I can learn and do even better in the future.

  • The above said — After returning home and able to view these 2 ECGs better — it literally took me no more than seconds using calipers to verify what took so much longer without them.

I've labeled the first 2 ECGs in today's case in Figure-1.
  • RED arrows make the ventriculophasic sinus arrhythmia obvious in ECG #1. Of note — most of the time when ventriculophasic sinus arrhythmia is seen in AV block rhythms — it is the P-P interval that "sandwiches" the P wave that is shorter (thought to be due to transient increased perfusion from the QRS) — but we see the opposite in Figure-1.
  • Complete AV block (as per Dr. McLaren) is clearly present in ECG #1 (quite regular junctional escape — but absolutely no relation between the sinus arrhythmia and neighboring QRS complexes despite adequate opportunity to conduct).

  • Impossible to prove anything in ECG #2, because there is no long lead rhythm strip — but the PINK arrows suggest ever-so-subtle deflections that I suspect represent "hidden" P waves from regression of 3rd-degree — to 2nd-degree AV block, Mobitz Type I with 2:1 AV conduction in this repeat ECG done 10 minutes after ECG #1.

Figure-1: I've labeled the 1st 2 ECGs in today's case.

No comments:

Post a Comment

DEAR READER: I have loved receiving your comments, but I am no longer able to moderate them. Since the vast majority are SPAM, I need to moderate them all. Therefore, comments will rarely be published any more. So Sorry.

Recommended Resources