Monday, May 11, 2020

A man in his 50s with fever and shortness of breath

Written by Pendell Meyers

A man in his 50s with HTN, HLD, obesity, and restrictive lung disease presented with shortness of breath worsening over the past 3 days. He also had a cough and subjective fevers. He denied chest pain. His vitals were within normal limits.

Here is his triage ECG (no baseline available):

What do you think?

 - sinus rhythm at about 100 bpm
 - STE in I and aVL (meets STEMI criteria)
 - hint of STD in III and aVF
 - STD in V1 and V2
 - hyperacute T-waves in I and aVL (with reciprocal negative hyperacute T in III)

This is definitive evidence of acute transmural injury and active infarction of the lateral and posterior walls. In any adult population with symptoms potentially compatible with ACS, the most common etiology of this focal transmural ischemia is Occlusion MI (type 1 ACS). But there are other etiologies such as coronary spasm, takotsubo/stress cardiomyopathy (which rarely has reciprocal findings because it is usually widespread), focal myocarditis (rare, but usually does have reciprocal findings), etc. The cells of the posterior and lateral walls do not know the etiology of the transmural injury, and the ECG findings are the same regardless of the etiology.

Back to the Case:

With his infectious and respiratory symptoms and complete lack of chest pain, we did not activate the cath lab despite the ECG meeting STEMI criteria. Instead, we called cardiology to assist us in the ED to make sure that they agree he does not need emergent catheterization. We performed a bedside (non-contrasted) cardiac ultrasound together which showed globally decreased EF, and we were unable to appreciate a discrete wall motion abnormality in the posterior or lateral walls (very limited to due point of care and lack of contrast). If there is no wall motion abnormality on very high quality contrast-enhanced echo, it is unlikely that the ECG findings represent acute OMI.

Ultimately cardiology decided not to proceed with cath lab activation due to a combination of lowered clinical suspicion of ACS given lack of chest pain and presence of more infectious symptoms, as well as preliminary ultrasound findings. However, I think it would also have been perfectly reasonable to cath him regardless to definitively rule out ACS as the cause, because the harms of a diagnostic cath without intervention are minimal compared to the harm of untreated OMI.

The first troponin T resulted at 1.34 ng/mL (quite elevated), and subsequent measurements were 1.29 and 1.27 ng/mL.

Formal echo confirmed diffuse hypokinesis with severely depressed systolic function, EF 25%. No contrast was used, but there was no focal wall motion abnormality seen.

Covid PCR resulted positive.

CXR also revealed bilateral opacities consistent with covid pneumonia.

No cath was performed.

A repeat ECG was done several days later:

Poor quality. Focal lateral findings are much less pronounced, but still slightly present.

The patient had an otherwise uneventful course and was discharged home with follow up to evaluate his long term cardiac function.

Learning Points:

COVID may cause myocarditis. However, myocarditis is not definitively proven in this case. If a high quality echo with contrast had shown no focal wall motion abnormality during ECG findings, then we could be almost assured that this case was definitively myocarditis. Short of that, myocarditis is only certain with biopsy or MRI (both of which were not performed in this case). It is still possible that this patient had true Occlusion MI (and STEMI), and he also happened to have COVID.

Myocarditis can be focal, and can be completely indistinguishable from OMI on ECG. Although the clinical picture is vitally important, clinical features alone are not enough to definitively distinguish focal myocarditis from OMI.

The decision to avoid emergent cath lab activation in a patient with shortness of breath, highly elevated troponin, and OMI findings on ECG (even STEMI[+] in this case) should not be made lightly. It may arguably be the right thing to do on a population level to perform the cath even knowing there is a high chance of a false positive, because the harms of one missed OMI are likely far greater than the harms of many false positive diagnostic caths without intervention.

MY Comment by KEN GRAUER, MD (5/11/2020):
Given the prevalence of Covid-19 infection — this case by Dr. Meyers is extremely relevant. It provides a superb illustration of clinical decision-making, in which despite ECG findings consistent with acute STEMI — the decision was made (wisely) not to proceed to cardiac catheterization.
  • Presumably (although not definitively proven) — the patient in this case had acute myocarditis.
  • To re-emphasize the key LEARNING Point highlighted by Dr. Meyers — Acute myocarditis can be focal, and can at times be completely indistinguishable from acute OMI on ECG. Consideration of clinical features may help in distinguishing between these 2 entities — but even then, a definite diagnosis cannot always be made. When doubt persists — prompt cardiac cath may be needed to exclude the possibility of acute OMI.

I have little to add to the excellent clinical discussion by Dr. Meyers. I limit my comments to a few additional “picky” points I’d add regarding ECG interpretation of the 2 tracings shown in this case (Figure-1):
  • As per Dr. Meyers — ECG #1 shows sinus rhythm at a rate just under 100/minute. All intervals are normal. I would describe the frontal plane AXIS as indeterminate — as all 6 limb leads are close to being isoelectric.
  • PEARL #1 — Determination of frontal plane axis is usually easy — and with experience, can usually be accomplished within seconds. That said — IF ever you find it taking you more than 2-3 seconds to figure out the frontal plane axis — chances are that the axis is indeterminateThat’s the case here. Looking at lead I in ECG #1 — one might think the axis is rightward. But looking at the 3 inferior leads — one might think there is LAHB. The reason for this discrepancy is that the axis lies in “No-Person’s Land” (ie, in the upper right quadrant). (NOTE: For those wanting a quick primer on Axis determination & the Hemiblocks — CLICK HEREIf you’d like a Video Review on Axis & the Hemiblocks — CLICK HERE).
  • PEARL #2 — It’s helpful to keep in mind the 3 most common Causes of an Indeterminate Axis. These are: iRVH; ii) Chronic pulmonary disease; and/or, iii) Large body habitus. Clinically, in this case — we are told that this patient is obese and has restrictive lung disease.

Figure-1: The 2 ECGs in this case (See text).

Continuing with our Descriptive Analysis of ECG #1:
  • Chamber Enlargement — None. Although the frontal plane axis is indeterminate — the rest of the ECG is not particularly suggestive of RVH. There is no sign of atrial abnormality or LVH.
  • Q-R-S-T Changes — As per Dr. Meyers, there is ST elevation with hyperacute T waves in high-lateral leads I and aVL — with reciprocal (mirror-image) ST depression in leads III and aVF. There is also J-point ST depression in leads V1 and V2 (as per Dr. Meyers) — with an ST-T wave shape in lead V2 consistent with acute posterior involvement.
  • WHY do you think the QRS complex in lead V6 of ECG #1 is so tiny? I’d surmise this may be due to either the patient’s obesity and/or his chronic pulmonary disease. The T wave flattening we see in leads V5V6 is not a normal finding (Normally the T wave should be positive and of reasonable amplitude in leads V5,V6). I was not initially sure whether the flat ST-T waves in leads V5, V6 were a reflection of acute ischemia, the patient’s large body habitus, his chronic pulmonary disease — or some combination of these features.
  • Clinical IMPRESSION of ECG #1 — As per Dr. Meyers, the above ECG findings could certainly be consistent with acute postero-lateral OMI. Alternatively (given the lack of chest pain, worsening dyspnea with positive Covid-19 assay) — the findings more likely represented focal acute myocarditis.

My THOUGHTS on ECG #2: This 2nd ECG tracing was obtained several days later. The patient had clinically improved — and was soon to be discharged from the hospital. As per Dr. Meyers — “focal lateral findings were much less pronounced, but still slightly present”. That said — WHAT makes it difficult to compare the chest leads in these 2 tracings?
  • PEARL #3 — In my experience, the most time-efficient way to compare serial ECGs is to: i) First choose 1 of the tracings — and interpret this first tracing in its entirety (as we did above for ECG #1); THEN; ii) Compare the 2nd ECG with the 1st one — by looking lead-by-lead to see if there are any changes between the 2 tracings in lead I — in lead II — in lead III — and in each of the other 9 leads. The reason performing this lead-by-lead comparison is important — is that otherwise, you might MISS the fact that QRS morphology of lead V2 is so different in ECG #1 compared to ECG #2! Given this patient’s large body habitus — the most likely reason for this marked difference in QRS morphology in lead V2 between the 2 tracings is lead misplacement of this chest lead on one of the tracings. On occasion, this technical mishap can be very important to recognize — because trying to compare the ST-T wave changes we saw in ECG #1, with what we now see for the ST-T wave in ECG #2 — is like trying to compare “apples with oranges” (given how very different the QRS complex in lead V2 is in these 2 tracings ...).
  • Clinical IMPRESSION of ECG #2 Compared to ECG #1 — Both tracings show sinus rhythm at a rate ~100/minute. There’s more artifact in ECG #2. I’d once again interpret the frontal plane axis of ECG #2 as indeterminate. I agree entirely with Dr. Meyers — that the hyperacute high lateral and reciprocal inferior lead changes are far less pronounced than they were in ECG #1. And although there almost certainly is some difference in chest lead electrode placement between the 2 tracings — the overall ST-T wave appearance in the chest leads of ECG #2 looks benign.

Our THANKS to Dr. Meyers for presenting some important clinical considerations for ECG interpretation during the current Covid-19 pandemic.


  1. An ECG that is very a propos in these times!

    Ken... You have me confused. You describe the mean QRS axis in the frontal plane as "indeterminate" and yet you immediately place the axis in the right upper quadrant. In fact, the axis is not indeterminate. For it to be indeterminate all the limb leads would have to be both biphasic and equiphasic. Had there been no STE in Leads I and aVL, I would quickly agree that those two leads were equiphasic with zero net voltage. But with Leads II, aVR (mustn't forget about aVR!) and aVF not being equiphasic, there should be no problem determining the mean QRS axis. Granted, it probably won't be so simple for introductory level readers.

    Otherwise, a great discussion on a very complicated and seldom-discussed topic (myocarditis). There are very few articles that discuss myocarditis without the discussion being mostly about pericarditis. Perhaps you guys could do a few more.

    1. Hi Jerry. It’s ALWAYS a Learning Experience for me when you comment! I believe you and I use some different terminology and different conceptualizations with regard to axis description and determination (I noted that a number of years ago on one of your comments on the ECG Guru that approached the subject quite differently than I have). So while I completely agree that an “easy” diagnosis of “indeterminate” axis is when all limb leads are biphasic and equiphasic — the calculation of axis is not so easy for the tracings here … I’d put the axis for ECGs #1 and #2 in the upper right quadrant — which is not “indeterminate” in the sense of being unable to come up with some number of degrees — but it IS “indeterminate” in the sense of not being able to determine with confidence if there is marked RAD or marked LAD. In my experience, determining a precise number of degrees for the axis when you are located in the right upper quadrant is highly challenging, as well as clinically irrelevant — since clinical implications for an axis anywhere between +181 to +270 degrees are for practical purposes quite similar. And from a teaching point of view — it should take even the novice interpreter no more than seconds to determine that the axis is (by my definition) “indeterminate” (ie, predominantly negative in both leads I and aVF). And for those occasional circumstances like this tracing (looking at my ECG #1) — since I’m hard pressed to calculate a specific number of degrees from looking at the 6 limb leads (and using “area-under-the-curve” rather than just number of positive and negative little boxes) — we know (by my Pearl #1) that the axis is in that right upper quadrant. Jerry — I realize and fully accept that you may disagree with my above explanation. I’m happy to agree-to-disagree with you on this one. And as always — it’s GREAT to debate these concepts with you! — :)

      As to cases on Myocarditis — Steve always publishes these as they come up. Here are a few links on cases I’ve participated in over the past 2 years.

    2. "the calculation of axis is not so easy for the tracings here …"


      Actually, Ken, I don't disagree with anything in your comment. The ECG looks "almost" like an indeterminate axis (though it really isn't) and the axis is definitely in the right upper quadrant. Also, it is a rare occasion when we have to be any more specific than noting the quadrant in which the axis is located.

      I enjoy our discussions because I certainly learn from you. It's proof that two physicians can argue two different viewpoints and still be respectful and cordial to each other. I see too many posts where someone tries to be aggressive and demeaning in their attitude toward anyone who disagrees with them.

      And thanks for the links to the other myocarditis cases.

    3. THANKS so much Jerry. YOUR Comments are always a most welcome addition to this ECG Blog! — :)

  2. Maarten Van HemelenMay 12, 2020 at 3:47 AM

    In addition to myocarditis, I'd be worried about COVID-related thrombotic events. While probably the patient did indeed have myocarditis, I'd be hesitant to forgo cath in the absence of MRI or high quality echo?

  3. interesting.
    pendell, i wonder. we know that COVID is a hyper-coagulable state. yes, COVID is associated with a number of associated entities, including cardiomyopathy, myocarditis, much more. But these patients clot, in their lung, brain,
    ? kidney micro-vasculature, and i suspect their coronaries.
    here we had a patient with dyspnea. an ECG with a ST elevation with reciprocal changes, in the ? high LAD or circumflex(?) territory.
    you already asked the question: why not take the patient to cath?
    what was the d dimer? just curious. can this be, Pendell, a pulmonary embolism, with positive trop? (no CT-A i suspect).
    also, isn't there usually some pericarditis with myocarditis, but there was no ecg or echo sign of pericarditis. just a i off-base?

    interesting case. thank you so much for sharing


    1. I think his d-dimer was between 1-2000, I remember it being elevated but not the exact number.

      Really nothing can be excluded for sure in this case. I do not think that PE would account for the ECG findings, but PE was not excluded in this patient.

      Nor was OMI.

      I do think that myocarditis can be focal and not involve pericarditis, but that is just my hunch thinking back on several posts in the blog. No solid evidence that I have there.

      Very annoying that the same process can cause both myocarditis and acute thrombotic events.

  4. it's funny... Maarten had the same thought, but his comment had not yet posted when i wrote mine thirty minutes later..

    1. Tom — I suspect MANY providers these days are having the "same thought" — as we continue to learn more about this "novel" virus (that initially was thought to primarily affect the lungs ...). As has become apparent — it affects MANY more systems in many patients ...

  5. Very challenging initial management.
    Personally I (if I would have been consulted as Cardiologist) would have activated cath lab despite the absence of chest pain and the clincal scenario highly suggestive for an infectious disease based on the following reasons: as mentiond above, despite a very likely infectious disease (essentially acute myocarditis here), acute myocarditis is a diagnosis of exclusion and therefore define coronary anatomy is of paramount importance. Another important feature suggestive of an acute coronary event is the textbook localization of the ST-T wave abnormalities, confined to a restricted coronary territory (=posterolateral); moreover there are reiprocal changes (STD in III-aVF).
    Finally, Ken, don't you think that there are some ECG findings favoring LVH, such as the indeterminate axis and persistence of S in chest leads (the transition is in V6!)? It would have been very useful to have data on RV function and morphology on echo results.
    Very interesting and timely (with regard to COVID-19) case!

    Mario Parrinello

    1. Thanks for your comment Mario! As you can see from Pendell’s discussion and his answer to the concern voiced by Maarten Van Hemelen (above) — He agrees on the need for strong consideration of cardiac cath for this patient (Sounds like it wasn’t his decision to make … ). As to the possibility of RVH (You said “LVH” in your comment — but I know you meant “RVH”) — you’ll note in my Pearl #2 above that RVH is one of the 3 common causes I list for an indeterminate axis. Persistence of lateral chest lead S waves is supportive of this — but really not “enough” (in my opinion) to make a definitive ECG diagnosis of RVH. I usually would write down something like, “Findings consistent with pulmonary disease; Suggest clinical correlation”. As you know — a definitive ECG diagnosis of RVH is often extremely difficut to make in adults — because the LV typically is 3X as thick, and has up to 10X the mass of a normal RV — so in order for RV forces to predominate in adults — the degree of RVH must usually be marked before you see a clear ECG picture of true RVH. This is different for infants and children, in whom there is not so much discrepancy between LV and RV mass — so the ECG diagnosis of RVH in young children is much easier to make. THANKS as always for your comments! — :)


Recommended Resources