Friday, October 23, 2020

A 40-something male complains of worrisome chest pain and possible "fever"

This was sent by an EM colleague at Highland Hospital in Oakland.  His name is "Deep" 

A 40-something male complained of chest pain and SOB that began 2 hours prior at work and was becoming progressively worse.  He had additional nausea and vomiting and complained of fever.  The pain was constant, pressure-like, substernal, without radiation, and was 10/10 in intensity.  

He stated that his wife had been diagnosed with Covid 3 months prior and that he, too, had been tested but never received the results. 

BP was 213/128.  Apparently no temperature was recorded as the patient looked very uncomfortable.  Temp was not recorded until much later when it was 36.5.

Here was his triage ECG:

There is ST Elevation that meets STEMI criteria in V2 and V3
What do you think?

The initial 4th generation troponin I was less than 0.010 ng/mL (undetectable).

The ECG looks a lot like early repolarization because: 

1. There are asymmetric T-waves (slower upstroke, faster downstroke) 2. There is upward concavity in all of leads V2-V6 (in this case, even in V1) 3. There is no reciprocal ST depression 4. There is no terminal QRS distortion (there are S-waves in both V2 and V3) 5. There are no Q-waves 

However, there is lots of ST Elevation, and Upright T-wave, in V1:  (In our study, Upright T-wave in V1 was found in 46% of Normals and 73% of LAD occlusion.  T-wave in V1 larger than T-wave in V6 was found in 15% of Normals and 39% of LAD occlusion.  Both results were highly significant but did not add value to the multivariate logistic regression formula.

Because of V1, and also due to the patient's clinical presentation, Deep was pretty sure this was a STEMI, but because it also has a lot of features of benign ST Elevation (also frequently called "Early Repolarization") which also has STE in V2-V4, he used the relatively new 4-variable formula, which is used to differentiate subtle LAD occlusion (in that it looks normal) from normal variant ST Elevation.

See use of the formula here: 12 Example Cases of Use of 3- and 4-variable formulas to differentiate normal STE from subtle LAD occlusion

WARNING: the formula is not perfect.  Beware of using it to reverse your prior opinion that the ECG represents LAD occlusion.  I recommend using it when you are worried that an ECG with apparent normal ST Elevation might be LAD occlusion.  Sensitivity is not perfect.

Using these values, he calculated the formula using these 4 variables: QTc-Bazett = 418, QRSV2 = 33, R-wave V4 = 12.5, STE (60 ms after J point) = 4.  Value = 17.65.  18.2 is the most accurate cutpoint.

In this external validation, "the published cut-point of 18.2 had a sensitivity, specificity, and diagnostic accuracy of 83.3%, 87.7%, and 85.9%, respectively."  A tale of two formulas: differentiation of subtle anterior MI from benign ST segment elevation.

Remember that it was only subtle LAD occlusion that was studied, so the sensitivity for all LAD occlusion is substantially higher.

In our Hennepin data, the sensitivity at a cutpoint of 17.0 was 97% (only 3% of LAD occlusions had a value less than 17.0).

Clinical Course

Deep realized that this was a false negative formula value, and he activated the cath lab.  A 100% LAD occlusion was found and opened and stented.

Here is the post PCI ECG:

Classic Wellens' Pattern A T-waves diagnostic of reperfusion.
Notably, there is not much R-wave amplitude, suggesting that there was substantial myocardial loss.

(Wellens' pattern A T-waves biphasic, with terminal T-wave inversion -- up-down).  
These will always evolve into Wellens' Pattern B over hours to day.  

Classic Evolution of Wellens' T-waves over 26 hours

Peak troponin was 19.9 ng/mL, which is typical of STEMI, but is at the low end of peak troponins in anterior STEMI.


MY Comment by KEN GRAUER, MD (10/24/2020):


TOUGH case today! — and one that I fully acknowledge that I was not at all certain about my answer after seeing the initial ECG. Reasons this case is so challenging include:

  • Reason #1: The initial ECG looks a lot like a repolarization variant (because of the 5 findings noted above by Dr. Smith).
  • Reason #2: There is at-the-least voltage for LVH (by Peguero Criteria) — and given how markedly increased this patient’s BP was in triage (ie, 213/128 mm Hg!) — there is most likely true chamber enlargement.


NOTE: Among the criteria for ECG diagnosis of LVH that I favor — are Peguero Criteria, which state: LVH is present IF sum of deepest S in any chest lead + S in V4 ≥23 mm (female) or ≥28 mm (male). If the deepest S wave is in lead V4 — then double this value.

  • Applying Peguero Criteria to ECG #1 in today’s case (Figure-1 below) — the deepest S wave is ~21 mm in lead V2 + an S wave ~ 11 mm in lead V4 = 32 mm, which satisfies voltage criteria for LVH.
  • For those wanting review of “My Take” on a user-friendly approach to the ECG diagnosis of LVH  Please SEE My Comment at the bottom of the page in the June 20, 2020 post of Dr. Smith’s ECG Blog.


PEARL #1: Although the ED physician in this case knew to immediately activate the cath lab — You do not have to definitively decide on whether or not there is acute OMI on the basis of this single ECG!

  • We have shown numerous cases on Dr. Smith’s ECG Blog of acutely evolving OMIs in which a 2nd ECG done no more than a few minutes later showed obvious hyperacute changes or frank ST elevation. Therefore — Repeat the ECG ~5-15 minutes later (and if needed, frequently thereafter) — and, the chances are good that “the Answer” will soon become apparent.
  • Other modalities (ie, high-sensitivity troponin, stat Echo during chest pain, finding a prior ECG for comparison) may all help clarify if ongoing OMI is in progress.
  • That said, even without a definitive ECG diagnosis of OMI — persistence of chest pain + the ECG findings that we do see in ECG #1 are enough to justify prompt cath.

TAKE another LOOK at the initial ECG in today’s case (Figure-1).

  • HOW MANY LEADS show suspicious ECG findings? WHAT are the abnormalities?
  • HINT: Look at the leads in which there are horizontal RED or PINK dotted lines.

Figure-1: The initial ECG in this case (See text).

MY Thoughts on ECG #1:

When I first saw this case (before I read what happened) — I was aware of 3 types of “input” information that went into my decision-making process:

  • Input Type #1: Consideration of the History = which is of a 40-something man who presented to the ED with new-onset chest pain that began just 2 hours earlier. Chest pain was pressure-like and constant, with a 10/10 severity rating. Clearly, this presentation places this patient in a high-prevalence likelihood for OMI even before you look at his ECG!
  • Input Type #2: Consideration of objective ECG findings in the initial tracing that further increase concern. These include: i) ST elevation in lead V1 that looks disproportionate given modest depth of the S wave in this lead; ii) Somewhat more-than-expected ST elevation in the next 2 anterior leads (leads V2 and V3); iii)Unexpected ST flattening in leads I, V5 and V6 (short RED horizontal lines in these leads); iv) flattening of the ST segment before the inverted T wave in lead aVL (ie, Although the T wave in aVL may normally be inverted when the QRS complex in this lead is predominantly negative — the preceding ST segment in lead aVL is usually not as flat as seen here) — andv) The upright T wave in lead V1 is larger than the T wave in lead V6.
  • Input Type #3: Consideration of “stuff-that-I-feel” but am not able to put into objective terms. In view of the fact that this patient presented with new-onset worrisome chest pain + an initial ECG showing more-than-expected ST elevation in anterior lead V1 — if anything, it looked to me as if despite less deep S waves, the relative amount of J-point ST elevation in lead V3 is more than what is seen in lead V2. My “intuitive” sense of concern was heightened by this suspicion of abnormal anterior ST elevation (in leads V1, V2 and V3) — in the context of unusual and abnormal ST segment flattening in the 4 lateral leads (leads I, aVL; V5 and V6).

PEARL #2: An under-appreciated important clue to potential acute changes is the above noted finding of new upright T wave in lead V1 that is taller than the T wave in lead V6. When found in a patient with new chest pain who does not have LBBB (that normally produces tall, upright anterior T waves) — one should be suspicious of acute ischemia, if not impending OMI.

  • NOTE: This Pearl #2 is not to say that tall, upright T waves in lead V1 might not sometimes be the result of a repolarization variant or a mirror-image reflection of LV “strain”. Instead, it is simply to say that on occasion — I have found recognition of a tall, upright T wave in lead V1 that is clearly much taller than the T wave in lead V6 to be an insightful clue (as it was for me in today’s case) of impending acute anterior OMI.

BOTTOM Line: It’s not common that anterior OMI is seen on an ECG that satisfies criteria for LVH. Today’s case provides one such example.

  • ECG findings suggesting OMI on the initial ECG are extremely subtle. Nevertheless — suspicious ECG findings are seen in 7/12 leads. I’d bet that a repeat ECG done no more than a short while later would have been more definitive.


  1. I also think it's worth noting that the St elevation is significantly more in lead 3 than lead 2. In combination with the features of V1 V2 and V3 I think this is further evidence that we need to be concerned about the right side of the heart.

    I haven't seen any research to this effect but my intuition is that because leads II and III are not that different from an axis standpoint when you have one that is clearly more St elevation than the other it's dangerous to discount this finding especially when it correlates with typical chest pain and suggestive precordial lead findings.

    1. @ DrMusicMan — THANKS for your comments. Re your 1st comment — I made a big point of this in My Comment above (Please see my “Input Type #3”). The reason I described this as “stuff-that-I-feel” but am not able to put into objective terms — is that there are 3 complexes in simultaneously-recorded leads V1,V2,V3 — and while the overall “shape” of the ST segment looks “more elevated” in lead V3 compared to V2 — there ARE slight differences in ST-T wave morphology from 1 beat-to-the-next (and there are only 3 beats …) — and I actually measure a slightly higher J-point in lead V2 for the 1st complex — about equal for the 2nd complex — and slightly higher in V3 than V2 for the 3rd complex. Given this slight-but-real variation in ST-T wave morphology among the 3 beats that we see — I think it difficult to know for certain which ST-T wave complex is “the accurate one” — and for that reason, I subjectively integrated what I saw as a suggestive “stuff-that-I-feel” finding (that lacks obective verification), but which DOES support the clearly disproportionate amount of ST elevation in V1 + the subtle-but-real ST straightening in the 4 lateral leads.

      Otherwise — I unfortunately do not understand what you are saying and asking in your 2nd comment about ST elevation in leads II and III. I do not see any ST elevation in leads II and III in ECG #1. Please clarify this for me. Thank you again — :)

  2. Hello Dr. Smith,
    Once again great post ( Fri Oct 23 2020 ), absolutely sparkling ! Excellent academic treat. Once the angio reveals 100% LAD occlusion, the suspense is over. However, for completion sake, we may need to know the exact site of occlusion and whether the LAD is of wrap-around type. The trivial but real 0.5mm ST depression in I & aVL speaks in favour of LAD occlusion and against Early Repolarization. By the same token, the 0.5 mm STD in I aVL can be taken as almost isoelectric and conclude that the LAD occlusion must be distal in a non wrap-around type.
    With regards,

    1. @ dr. R.Balaubramanian — THANK YOU for your comment! It would indeed be interesting to know more details about cardiac cath findings. The history given says only that chest pain began 2 hours prior to obtaining ECG #1 in a 40yo man — and, we only have 1 ECG to go on (given prompt recognition by the ED physician of the need for cath). My guess would be a more proximal LAD occlusion — because ST elevation begins (and is marked) already in lead V1 — although I’d usually expect some ST elevation in aVL and reciprocal inferior ST depression with proximal LAD occlusion. I’d expect more obvious inferior ST elevation if there was “wraparound”. Another possibility might be multi-vessel disease — although that would seem less common in a 40-year old. So I didn’t know how to “localize” the likely location of the culprit LAD without more info, serial tracings and/or cath results …

  3. Subtle ST depressions in I, aVL, and ?V6 should negate use of subtle STEMI algorithm, at least from my understanding of it. This is technically an obvious OMI. Would this ECG not have been excluded from your validation study for the algorithm?

    1. @ Ryan — I completely agree with you. The emergency physician in this case (Deep) recognized a number of clearly abnormal ECG findings in association with a very worrisome history of new-onset chest pain — which immediately prompted him to activate the cath lab. I emphasize in My Comment above, that no less than 7/12 leads in ECG #1 are abnormal. Therefore, Dr. Smith’s formula isn’t “needed” to make the diagnosis of suspected OMI. That said — Dr. Smith often likes to still refer to his formula, as it provides additional insight into the relative likelihood of an acute event. As described above by Dr. Smith in his discussion — today’s case provides one example in which there was a false negative formula value. This can happen — with the optimal response being exactly the way in which Deep proceded, namely to recognize that despite the negative formula value — ECG #1 was nevertheless very suggestive of acute evolving OMI.

      My understanding is that this type of tracing would be excluded from Dr. Smith’s data gathering. But I am passing your comment directly to him for his additional clarification of how he validates his data. Thank you again for your comment!

    2. Good comment! For exclusion in our study, we required a sum of 1 mm of STD in inferior leads, which this does not meet. This is very subtle here and may be significant. Emre Aslanger thought it was. I am not so certain.

  4. This ECG #1 is really challenging. As per the great physician William Osler taught us "
    Listen to Your Patients. They're Telling You the Diagnosis!". I am in doubts, because I’m completely confused with some used terms.It's Normal Variant the same thing that Early Repolarization and benign ST Elevation(Normal Variant=Early Repolarization=Benign ST Elevation)? What is diference between them?
    Thanks a lot my teachers. I LOVE ECG.
    Anderson Santos from Brazil.

    1. Muito obrigado (THANK YOU) Andereson for your comment! You are correct that the terminology (even among different cardiologists) is confusing! I will therefore give you “My Take” on the issue (realizing that others may favor a different approach to terminology). The problem with the term, BER ( = “Benign” Early Repolarization) — is that this is “a benign disease that can kill you … “. As described in this article by Ali et al (World J Cardiol 7(8): 466-475, 2015 — GO TO = ) — certain patients with this type of ST-T wave appearance are at somewhat increased risk of sudden death. The overall increased risk is low — and extremely low in asymptomatic patients for whom a routine ECG shows incidental “early repolarization”. But risk may be increased IF a similar repolarization change is seen in a SYMPTOMATIC patient (ie, with syncope/presyncope or a sustained ventricular arrhythmia) or in a patient with a positive family history for malignant arrhythmias.

      As a result — I long ago stopped using the term “BER” — and I no longer say this type of ECG appearance is “benign” or “benign” early repolarization. I also generally do not use the term, “Normal Variant” — because of the potential (albeit very small) of increased risk (ie, in which case this ECG finding would retrospectively turn out to not have been a “normal” variant). Instead — I favor use of the term, “Repolarization Variant” — because this is an accurate description of what we see on ECG — and it is a “variant” appearance of the textbook picture of “normal” ST-T waves. We KNOW that in asymptomatic patients — this “Repolarization Variant” pattern will almost always be benign — but I prefer NOT to add the word “benign” to my “official” interpretation, because of a possible rare adverse outcome. I hope this approach makes sense. I welcome other views from our readers — :)

  5. Thank you for sharing this case. I think some of us may have repeated the ECG after treating the patient's hypertension. Just wondering if this occured in parallel to activating the Cath lab?

    1. Thanks for your comment. I agree that it would have been nice to see a follow-up ECG to ECG #1 prior to PCI. My understanding is that this was not done … because Deep recognized the need for cath from the findings in ECG #1 in association with the VERY worrisome history ( = new-onset chest pain that began just 2 hours earlier, and which worsened to attain a 10/10 intensity scale). This should BE ENOUGH to justify prompt cath ( = persistent, severe and worsening cardiac chest pain + an ECG that DOES show some definite worrisome signs, regardless of whether or not you accept that this ECG suggests acute OMI). This patient was fortunate that Deep was working that day with a cardiologist who agreed with his astute assessment.

  6. excellent. thank you Deep! actually, i am now in Alameda, about 15 minutes from your shop!
    and my close friend has worked with you (unless there are two "Deep"'s at Highland ER! i just showed her your case, and she referred me to your youtube video : " Sepsis the Highland Way"...! hope you don't mind me sharing that.

    thank you , Deep, for sharing this case. (and of course thank you to Steve and Ken)..
    very enlightening.. pulling the trigger, ie, activating the cath lab cannot be done willy-nilly, and it's knowledge of the finer points discussed here that helps us in that critical decision.
    thank you.


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.