Wednesday, April 13, 2022

A man in his 50s with acute chest pain and history of prior MI

Written by Pendell Meyers

A man in his 50s with prior history of anterior MI with LAD stent presented with acute chest pain similar but more intense than his last MI. He presented around midnight with pain that had started around 9pm the night before. He had taken NTG at home with no improvement, and immediately received morphine on arrival at the ED for severe chest pain (a very bad idea if your accuracy for finding OMI on ECG is low, since ongoing pain will be your last chance to identify those with ongoing untreated OMI).

Here is his triage ECG at 0012:

What do you think? What is the differential of this ECG?

There is sinus rhythm. The QRS is narrow but has poor R wave progression and QS waves in V1-V3, followed by STE and upright potentially large T waves. There is some movement in the limb leads portion, but there is the question of subtle STE in aVL, and subtle reciprocal STD in some inferior leads.

The differential could include:

#1) Old anterior MI with no active ischemia (old LV aneurysm morphology)

#2) Old anterior MI with superimposed acute occlusion

#3) Subacute MI (no prior MI at that location, and now at least 6-12 hours into full thickness infarction)

T waves are tall and hyperacute (use the ratios below) in both #2 and for at least a few hours in #3, then fall with time, loss of tissue, and rising troponin.

Troponin will be positive already in #3, will quickly become positive in #2, and will be negative in #1.


We have derived and validated a rule to differentiate "LV aneurysm" ST elevation from STEMI.  The rule depends on the principle that acute STEMI has a tall T-wave and LV aneurysm does not.  There are two versions:

In the first rule, if there is any single T/QRS ratio in V1-V4 that is greater than 0.36, it is likely STEMI:  for the ECG from 3 years prior, that would be lead V2.

T/QRS ratio to differentiate anterior STEMI from anterior LV aneurysm:

These studies showed that acute anterior MI (symptoms less that 6 hours) almost always had a T/QRS ration greater than 0.36 in at least one of leads V1-V4.  A subacute MI (symptoms of at least 6 hours) and Old MI both had ratios less than 0.36.  This just demonstrates that in acute MI, the T-wave is large.


Here are my calculated T/QRS ratios in the ECG above:

8/26.5 in V3 = 0.30

4.5/14 in V4 = 0.32

1.5/4 in V5 (first QRS complex) = 0.375

2.5/4.5 in V5 (second QRS complex) = 0.55

So there are multiple leads with concerning ratios, and lead V5 is probably diagnostic according to the literature above.

But more importantly for visual learning, let us simply compare to baseline!

Baseline ECG from several months ago: 

What do you think now? Did you need ratios?

Side by side comparison with baseline:
The difference is obvious if you have been taught the OMI progression. This is what it looks like when you have acute OMI superimposed on old LVA morphology.

In other words, in the context of this patient, with acute chest pain, the ECG on this presentation (the one at the top), is diagnostic of acute LAD occlusion, LAD OMI

Yet, it was read as no change from prior.

The initial hs troponin I returned elevated at 24 ng/L (upper reference limit for men in this assay is 20ng/L).

For such severe chest pain without recognized "STEMI" on ECG, the patient was given more morphine and sent to CT for CT angio to rule out aortic dissection, which was negative.

Repeat ECG at around 2am:
Ongoing OMI.

The ECG was read again as no acute ischemia.
The second troponin returned at 1,751 ng/L.

Due to rising troponin, another ECG was ordered around 3 am:

T waves are still hyperacute, but have deflated just slightly since last ECG, and there is the tiny beginning of TWI in V4, V5. These findings are typical of the evolution of completed infarction, plus or minus some reperfusion.

There is no mention of whether the pain was active at this time, but it had clearly been constant for the 3 hours from arrival to this ECG.

Cardiology was contacted for admission, but reportedly refused saying there were no signs of "STEMI" and requested medicine admission. Neither party seemed to consider the idea of emergent catheterization for acute coronary occlusion, or ongoing chest pain with rising troponins, such as the ACC/AHA class 1A indication for emergent cath within 2 hours for ACS with ongoing ischemia despite medical management regardless of ECG findings. 

So the patient was admitted to medicine floor, with cardiology consulting.

The next troponin was 2,955 ng/L.

The next was 12,216 ng/L, and none further were ordered.

An ECG was recorded around 8 hours after presentation:

Further evolution of terminal T wave inversions seen in both reperfusion and completed OMI.


Formal echo was done before cath, which showed EF 20-25%, with global hypokinesis and focal anterior and apical WMA.

Angiography was performed around 12 hours after arrival and showed:

Basically all vessels had CAD with at least 70% stenosis or more. Among these, there was a 95% mid LAD stenosis that has no TIMI flow listed, and no indication of trying to figure out whether it was the culprit (no indication of whether there was thrombus, haziness, etc.). The cath report states there was no acute culprit found. No intervention was done.

His troponin elevation was attributed to NSTEMI, and the notes state there is uncertainty whether this was type 1 MI or type 2 MI due to "systolic heart failure with acute volume overload." 

Cardiac MRI was then performed several days later which showed:

"27% EF with akinesis of the mid anterior, anteroseptal, and entire apical LV segments, indicating large transmural infarct along the mid anterior/anterior septal and entire apical LV segments (LAD territory), which is nonviable except for the first diagonal distribution which is entirely viable, RCA/circumflex/OM territories are entirely viable."

Medical management was chosen. He survived the index visit was discharged.

Learning Points:

After a patient already has LVA morphology in a particular area of the myocardium, recurrent OMI in that area is more difficult to appreciate, but can be done by understanding hyperacute T waves and using the formulas above. As always, the most viable and at risk tissue during OMI registers in the T wave as hyperacute T waves. This is still true even in the context of LVA.

Ongoing ischemic pain, rising troponins, despite maximal medical management is a class 1A indication for emergent catheterization in the current ACC/AHA guidelines. This patient is one of the roughly 90+% whose management violates these guidelines.

Learning the OMI progression helps free your mind from the prison of the STEMI criteria and allows you to start understanding OMIs.

Remember, STEMI(-) OMIs with delayed management have double the mortality and long term morbidity than their RCT counterparts who did not have OMI.

Get more practice with LVA and OMI here:

A woman in her 40s with acute chest pain

Subtle Anterior STEMI Superimposed on Anterior LV Aneurysm Morphology


MY Comment, by KEN GRAUER, MD (4/13/2022):


As I often acknowledge — "Hindsight is 100% in the retrospectoscope". It's EASY for me to sit back and "pick" at the details of these cases that I was not present on the "firing line" for. That's not our purpose. Instead — our purpose is to take a hard, soul-searching and objective look at what transpired — with our sole goal being to learn from these cases in the hope of improving care the next time. We are all learning every day from these cases.

I focus my comments on a few different aspects of this case than those covered in the thorough and excellent discussion by Dr. Meyers.

PITFALLS in Today's Case:
Rather than concern about the morphine administered for ongoing severe chest pain on arrival in the ED — I was much more concerned about the features of this case that were apparently overlooked.
  • This man in his 50s has documented CAD (ie, prior anterior MI with LAD stenting). He presents to the ED at midnight with a 3-hour history of new-onset chest pain that was worse than the chest pain that occurred with his previous MI. NTG provided no relief.

COMMENT: This history suggests as much of a "high-prevalence likelihood" situation for an acute event as one could possibly imagine. Especially in view on ongoing (for 3 hourssevere chest pain — the "onus of proof" should be on demonstrating why prompt cath is not needed, rather than the other way around.
  • Even without seeing the initial ECG — this clinical scenario (especially with any elevation at all in troponin) — should be viewed as clear indication for prompt cath.

Dr. Meyers has reviewed serial ECGs in today's case in detail. I wanted to look at the initial ECG from a different perspective — so I've reproduced and labeled this tracing in Figure-1:

Figure-1: I've labeled the initial tracing in today's case (See text).

MY Thoughts on Figure-1:
When interpreting serial ECGs — I find it helpful to start by systematic asssessment of the entire tracing before correlating my findings with the clinical scenario. Contrary to popular belief — this approach does not slow me down at all.
  • There is significant baseline artifact in multiple leads on this tracing. Not being at the scene — I have no idea why there is so much artifact — but in this patient with such high likelihood of needing prompt cath — I would have immediately repeated this ECG (or repeated it within no more than 10-15 minutesin the hope of obtaining a more interpretable tracing.
  • That said — there is a regular sinus rhythm at ~80/minute in ECG #1. All intervals (PR, QRS, QTc) are normal.
  • There is marked left axis deviation (ie, at least -60 degrees) — which is consistent with LAHB (Left Anterior HemiBlock).
  • Voltage criteria for LVH are satisfied in ECG #1, primarily because of the very deep (~25 mm) S wave in lead V3 (both Cornell and Peguero Criteria are satisfied — as discussed in My Comment of the June 20, 2020 posof Dr. Smith's Blog).

NOTE: Before going further — I think it worthwhile to comment on potential implications of the above findings.
  • Because the left anterior hemifascicle lies anatomically in front of (ie, "anterior") to the posterior hemifascicle — ventricular activation with LAHB is initially posteriorly-directed (toward the intact left posterior hemifascicle). As a result — LAHB may cause (or exacerbatepoor R wave progression.
  • As we have shown on many ECG Blog posts in Dr. Smith's Blog — the presence of LVH can complicate assessment of anterior ST elevation — because the "reciprocal" of lateral lead LV "strain" in patients who manifest LVH with deep anterior S waves, is ST elevation in these anterior leads (See My Comments in the December 27, 2018 post — the February 6, 2020 post — and the June 20, 2020 postto name a few).
  • Today's case is further complicated — because the known history of prior anterior MI may itself result in deeper anterior S waves not due to LVH (ie, loss of anterior forces from the anterior infarction may result in unopposed posterior forces). 
  • That said — ECG #1 lacks the tall R waves so common with LVH in left-sided leads (ie, in leads I, aVL and V6)  — and it lacks deep S waves in anterior leads V1,V2 — so despite satisfying textbook "voltage criteria" for LVH — I'm not convinced we can diagnose LVH on the basis of this tracing (ie, formal Echo would be needed to do so).

  • BOTTOM LINE: Even though definitive conclusions can not be reached — It's important to be aware of potential interaction that LAHBLVH and prior anterior infarction may all have when attempting to assess the initial ECG in today's case!

Completing My Systematic Assessment of ECG #1:
Regarding Q-R-S-T Changes in ECG #1:
  • Q Waves: QS complexes are seen in leads V1 and V2. A small-but-present initial positive deflection (r waveis present in lead V3. This initial r wave gets a little bit larger in leads V3 and V4 — until transition to a predominant R wave finally occurs in lead V6.
  • There is PRWP (Poor R Wave Progression) — given how small the initial r wave is until lead V6. This is consistent with the patient's prior anterior infarction.
It is assessment of ST-T wave changes in ECG #1 that is so problematic:
  • As I emphasized earlier — significant baseline artifact complicates assessment. Two of the most concerning leads to me in ECG #1 were leads aVL and aVFBut WHICH of the 3 beats in these 2 simultaenously-recorded leads are the "real" ones? In this patient with new-onset and persistent chest pain — there clearly is worrisome ST elevation in complex "B" in lead aVL (with corresponding reciprocal ST depression in lead aVF). There is still ST elevation in complex "A" in aVL, albeit to a lesser degree. But I don't think we can say there is any definite ST elevation in the artifact-laden ST segment of complex "C" in this aVL lead.
  • Similar pitfalls in interpretation of ST-T wave changes are encountered in simultaneously-recorded leads V4 and V5. If complex "C" in leads V4 and V5 represents the "real" ST segment — then there is no doubt that this is a hyperacute T wave. But although the ST-T waves for "A" and "B" in leads V4, V5 are still abnormal — the abnormality is not nearly as decisive.
  • Lead V3 at least shows consistency — in that ST-T wave morphology for each of the 3 complexes recorded in this lead looks the same. Let me emphasize that I completely agree with Dr. Meyers' assessment — namely, that despite the deep S wave in lead V3, the amount of J-point ST elevation and height of T wave peaking is highly suspicious. But given potential complicating factors I described above — I was not at this point 100% certain that ECG #1 represented an acute process.

BOTTOM LINE about Today's Case: Nearly 2 hours passed after obtaining the initial ECG — until a repeat ECG was finally obtained at 2:00 am.
  • Given technical faults with the quality of ECG #1 — plus persistence of severe chest pain — plus a slightly elevated initial troponin — this delay between tracings was far too long.
  • prior tracing on this patient was found soon after. Although this prior tracing was apparently looked at by the health care team — lead-by-lead comparison could not have been done — otherwise there would be no way not to notice the difference between the prior tracing and the initial ECG (ECG #1) in today's case (See side-by-side comparison figure by Dr. Meyers above). The most accurate (and time-efficient) way to compare serial tracings, is that after you have interpreted one of the ECGs — to compare lead-by-lead the 2 tracings in front of you (I illustrate this approach in My Comment in the August 8, 2020 post — and in the August 21, 2020 post of Dr. Smith's Blog).

  • Closing Thought: As detailed above by Dr. Meyers — Oversights in today's case continued for ~12 hours after ED arrival, until cardiac cath was finally performed. My belief has always been that the best way to learn from mistakes made — is to retrace events, and soul-search the decisions made. The decision to perform cardiac cath should have been made long before the 2nd ECG ( = the 2:00am tracing) was obtained.

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