Monday, August 26, 2019

An ECG sent to me with concern for hyperacute T-waves

Written by Pendell Meyers

A woman in her 70s with diabetes, hypertension, and hyperlipidemia suddenly developed nausea, diaphoresis, and brief syncope while eating at a restaurant. She did not report any chest pain or pressure.

She was brought to the Emergency Department and this ECG was recorded while she was still feeling nauseous but denied chest pain, shortness of breath, or other symptoms:

What do you think? No baseline was available for comparison.

Sinus rhythm
Grossly normal QRS complex
Less than 1mm STE in II, III, and aVF, as well as V4-6, all with extremely upward concavity
aVL has the smallest possible amount of STD in aVL with a shallow negative T-wave
The T-waves in the inferior leads seem to have potentially large amplitude compared to their QRS complexes
STEMI criteria are not met

Although inferior OMI commonly presents with submillimeter STE in the inferior leads with very subtle reciprocal STD and/or T-wave inversion, the morphology of this ECG is not convincing to me. The T-waves have significant amplitude but they are not "fat" enough to be hyperacute. The morphology of STE is not diagnostic of being due to acute transmural ischemia. It is difficult to describe exactly why, but is something that simply comes with time, after following up hundreds of cases to see the ECG progression and patient/angiographic outcomes.

This ECG was texted to me with no clinical information, with the sender being concerned for possible hyperacute T-waves and STE in the inferior leads.

I replied: "I see why you would be worried about hyperacute T-waves, but they just don't look fat enough to be hyperacute. There is definite STE inferior and lateral but it just doesn't look like true positive STE to me. I wouldn't activate the lab for this EKG alone, but if the patient is clinically compatible with ACS you could call a heart alert. I would do echo for WMA, q15 min EKGs - if it's real it will be dynamic and become clear soon. Let me know what happens." If I had known there was no definite chest pain or pressure this would have been slightly more confident.

I later found out that the cath lab was activated soon after arrival.

The patient went emergently to the cath lab, where all coronary arteries were found to be normal.

Note: Normal coronary arteries at the time of cath of course does not rule out the possibility of OMI. There can be OMI at the time of the ECG followed by spontaneous (or medication induced) lysis, such that the artery is open at the time of cath. This can be so brief that even troponins do not significantly rise. Of course that is very rare. But it demonstrates the point that the angiogram alone cannot be used to adjudicate the presence or absence of OMI. The ECG progression, clinical picture, and biomarkers are also involved in the full diagnosis of OMI. The ECG progression in particular is important: True positive OMI findings on ECG must always resolve or evolve. Conversely, an ECG that remains static throughout the entire clinical course means that the ECG findings were not due to OMI (even if OMI was truly present!).

Serial troponins were negative.

Echo showed no wall motion abnormality.

Most importantly, several repeat ECGs were basically unchanged:
This is the ECG the next day. This one likely does meet STEMI criteria in II, III, and aVF. But it is the same morphology as the first ECG. Absolute STE is irrelevant - the fact that the prior ECG did not meet criteria and this one does, is meaningless in my opinion.

Learning Points:

This was an example of a false positive cath lab activation due to ECG findings that were thought to represent OMI.

Learning how to differentiate hyperacute T-waves from normal variants like this is difficult and requires comparing cases like this with our database of true positive hyperacute T-waves. True hyperacute T-waves are tall, fat, wide, symmetric, and by these properties have large area under the curve compared to the size and morphology of the QRS complex. 

Take this example: first I'll show you a close up of leads II, III, aVF, and aVL from a true positive inferior hyperacute case. Below you'll see the same leads from this case.

These are fat, symmetric, true positive hyperacute T-waves in III and aVF, with their reciprocal counterpart in aVL. This patient had RCA OMI. From: 

A female in her 60s who was lucky to get expert ECG interpretation

This case. False positive, proven by no evolution and no evidence of OMI, negative trops, and normal angiogram.

Use these cases to see more true positive inferior hyperacute T-waves:

You have two hours to save this patient's life

MY Comment by KEN GRAUER, MD (8/26/2019):
I find this an important case for helping to clarify the semantics surrounding use of the term hyperacute” when referring to T wave appearance. I fully AGREE with Dr. Meyers astute ECG assessment and management plan for the initial ECG in this case ( = ECG #in Figure-1).
  • That said — I offer a different perspective on the semantics that I favor to describe the ECG findings in this case.
Figure-1: The 2 ECGs shown in this case (See text).

MTHOUGHTS on ECG #1: The most remarkable findings in ECG #1 are seen in the inferior leads. That said, systematic assessment of this tracing reveals a number of additional findings of interest:
  • There is baseline artifact in several leads. In ECG #1 — this is most marked in leads II, III and aVF. Since electrical derivation of these 3 leads is shared by involvement of the left leg electrode — this is the probable source of this artifact. I’ll emphasize that despite the artifact — this tracing is clearly interpretable. The reason I note the artifact, is that on occasions when artifact does impair interpretation, it may be helpful to recognize from which extremity the artifact arises (Details on this subject are beautifully described in this article by Rowlands and Moore.)
  • Beyond-the-Core: In contrast, the limb lead artifact in ECG #2 is maximal in leads I, II and aVR, with electrical derivation of these 3 leads shared by involvement of the right arm electrode — which is the probable source of artifact in ECG #2.
Additional findings of interest in ECG #1 include the following:
  • There are multiple small and narrow Q waves. These are seen in leads I, II, aVL, aVF; and V3-thru-V6. There appears to be a small-but-present initial r wave in lead III. The clinical significance of these Q waves is uncertain — but their presence should be noted because this finding might be relevant if there was ongoing infarction.
  • There is early transition — because the R wave becomes predominant (ie, taller than the S wave is deep) as early as lead V2. This finding could also be relevant — because posterior infarction is among potential causes of early transition.
  • Finally — I was struck by how similar the appearance is for QRS complexes and ST-T waves in leads V3-thru-V6 of ECG #1. This made me wonder if there might not be a potential technical mishap in this recording.

AGREE completely with Dr. Meyers that:
  • There is <1 mm ST elevation in the inferior and lateral chest leads.
  • There is ST segment flattening with no more than minimal T wave inversion in lead aVL. Lead aVL does not have the appearance of a true “reciprocal change”.
  • The T waves in the inferior leads of ECG #1 are each taller-than-expected considering the appearance of the QRS complex in each of these leads.
  • Criteria for a STEMI are definitely not met in ECG #1.
  • The T waves in each of the inferior leads in ECG #1 are not as “fat” as the T waves in the true positive example of inferior hyperacute T waves that Dr. Meyers shows above.
  • Although this 70yo woman with cardiac risk factors did have acute symptoms (ie, nausea, diaphoresis and syncope) — she did not report any chest discomfort! On the whole — her history is not convincing for acute OMI (although we need to remember that not all acute MIs are associated with chest pain).
  • I would not activate the cath lab on the basis of ECG #1 alone.

Where MPerspective Differs: While the history for this patient is not at all convincing for acute OMI — and, reciprocal changes are lacking — and, chest leads fail to show acute changes — we still have the unescapable finding of waves that are taller-than-they-should-bin each of the 3 inferior leads. That said:
  • ECG #1 is not diagnostic of acute OMI.
  • I would not have activated the cath lab on the basis of ECG #1 alone.
  • BUT — this patient did present to the ED with new non-chest pain symptoms (including syncope) that could reflect an acute cardiac event.
  • AND — in my opinion, in the absence of a prior ECG for comparison — we can not rule out the possibility that these might be hyperacute T wave changes in the inferior leads of ECG #1.

Why this is 
all Symantics: Regardless of how one classifies the T waves in the 3 inferior leads of ECG #1 — optimal management will be similar:
  • A definitive diagnosis can not be made from ECG #1 alone.
  • Continued close observation of this patient frequent serial ECGs stat Echo in the ED (looking for wall motion abnormalityserial troponins — will almost certainly yield the correct diagnosis in very short order.

MTHOUGHTS on ECG #2: There is essentially NO change in the appearance of ECG #2 compared to ECG #1. We are told that several other ECGs done after the initial tracing all looked the same.
  • I find it of interest that both early transition and similar appearance of the QRS complex and ST-T waves in lateral chest leads persist in ECG #2 done the next day. This makes a technical mishap much less likely — and makes me wonder if an unusual body habitus might account for these findings ...

BOTTOM LINE: Dr. Meyers and I are essentially saying the same thing. Considering this patient presented to the ED with new symptoms (including syncope — albeit without chest pain) — ECG #1 is not a normal tracing.
  • Regardless of whether or not you classify the inferior T waves in ECG #1 as possibly hyperacute or not — additional testing, including repeat serial ECGs, troponin + stat Echo are all needed to arrive at a more certain diagnosis.
  • “Ya gotta be there” to make the decision of whether or not to activate the cath lab on the basis of the history and ECG #1. While fully acknowledging that “hindsight is 100% in the retrospectoscope” — for learning purposes, I’ll put forth the thought that repeat ECG, stat Echo and initial troponin could probably have been enough to dissuade the decision to activate the cath lab in this case.

Our THANKS to Dr. Meyers for presenting this case!


  1. This is an excellent “false positive ECG” case to learn from.

    I agree entirely with Dr. Meyers and prof. Grauer: I wouldn’t activate cath lab on the basis of this ECG alone, rather I would have made the usual management of any suspicious ACS (close observation, serial ECGs and troponins, bedside echo).
    As to the ECG findings, I would add another ECG sign and I refer to the difference between the ischemic versus non-ischemic hyperacute T waves (applying also to ischemic or non-ischemic STE).
    In the case at hand, we can see the abrubt passage from the isoelectric ST-segment to the hyperacute T wave in true OMI (abrupt angle), whereas we can see a more gentle angle (if any) in non-ischemic ECG. The above ECG sign is not of course standardized or proved by clinical trials, it's a tiny detail but is often (=not always), at least in my experience, very useful.
    Thanks for sharing this instructive case.
    Mario Parrinello

  2. Dear dr Smith,
    I'm a cardiac surgeon working in Milan, Italy. I'm deeply indebted to you for this fantastic blog and your book about ecg in mi and reperfusion. They are really improving my clinical practice.
    In order to achieve objective criteria for hyperacute T waves, could it be useful to measure the angle between ascending and descending T limbs?
    Thank you.
    Best regards,
    Aldo Cannata, MD

    1. Thanks for the feedback, happy to hear it helps your practice!
      I have not considered the concept of that angle before, but it does seem that a wider angle may correspond with a "fatter" T-wave. We are in the beginning stages of studying hyperacute T-wave more definitively. Our first analysis will be aimed at the area under the curve of the ST-T complex compared against the size/area/etc of the QRS complex. Perhaps your angle idea would also be helpful. Until we have better data, it remains to be human pattern recognition. Thanks!

    2. Thank you Aldo for your excellent question. As per Dr. Meyers — qualitative assessment is the current standard — though in addition to the point you mention — I’ll ADD a few descriptive points regarding the example of TRUE POSITIVE Hyperacute T waves that Dr. Meyers added (showing leads II, III, aVF & aVL) above. These are: i) As per Dr. Meyers, this true positive example looks “fat” not only at the T wave peak, but ALSO at the wide T wave base; ii) The short ST segment prior to the upslope of this hyperacute T wave is FLAT (instead of gently upsloping); iii) The junction of this flat ST segment with the initial upslope of the T wave is ANGULATED (ie, abrupt, instead of gently upsloping); iv) The changes in i,ii,iii are seen in ALL 3 inferior leads; and v) The negative T wave in lead aVL is the MIRROR-IMAGE opposite picture of what we see for the T wave in lead III, in that the negative T in aVL is WIDE and FAT at its nadir. Not all hyperacute T waves have all of these changes — but the example shown by Dr. Meyers DOES — so “qualitatively”, we CAN state with certainty (until proven otherwise) that this is a TRUE positive. That said — in order to achieve an acceptable (ie, >95%) level of certainty to RULE OUT the possibility that the taller-than-they-should-be inferior T waves in the case presented here (in this patient with new syncope) were NOT due to an acute event — additional testing (serial ECGs, Echo, troponin) was needed. Grazie ancora per il tuo ottimo commento! (Thanks again for your excellent comment!) — :)

  3. Dr. Meyers...

    Thanks for a very informative presentation. Whether the T waves were hyperacute or not is not an easy call in this case for anyone, so those who are relatively new to ECG interpretation should not feel too bad if they called it wrong. If one is going to err, it's much better to err on the side of caution. As both Dr. Meyers and Dr. Grauer indicated, you need to base your decisions on more than just the ECG. I think activation of the cath lab here was probably unavoidable based on the history and presentation. I would have felt more comfortable holding the patient and repeating the ECG every 15 - 20 minutes if there had been less of an issue with the ECG (though lack of WMA on echo or a positive troponin could also have been significant factors in deciding whether or not to wait).

    Regarding the contribution of history of present illnes in this case, one should always be very cautious of the combination of older age group, female gender and diabetes in a patient. Any of the three can result in atypical symptoms experienced by the patient so one should never require chest pain as a symptom of OMI when any one of these background factors are present and certainly not when more than one are present.

    Thanks to Drs. Meyer and Grauer for their comments and for a very educational presentation.

    If I may make a practical suggestion to those who are just developing their skills at ECG interpretation - even to this day I keep copies of three or four NORMAL ECGs next to my computer for the sake of comparison. Even though I am not comparing ECGs from the same patient, it often helps in getting a "feel" for what is normal, what is abnormal and what is "suspicious."

    1. Jerry, agreed. We should always err on the side of more false positives than false negatives. If no CAD is present, then no intervention is needed; if no intervention, the risk of the PCI to the patient is quite low. Thus a false negative is not so untenable (unless you ask the cath team).

      So for those who are not at the stage of learning to differentiate things like this case from true positive hyperacute T-waves, I totally agree that this case should be met with immediate concern which may justify cath lab activation.

    2. Thanks (as always!) for your comment Jerry. To follow-up on your important point that “chest pain” (CP) is NOT needed for there to be an acute MI — I always refer back to the Framingham studies which showed AT LEAST 1/4 (if not 1/3) of all MI’s in their MANY-year follow-up studies were “silent” — in that they were not associated with chest pain. But OF THOSE 1/4-to-1/3 of all MIs in the Framingham studies — about HALF of that group had NO chest pain at all — and the other HALF had “something else” in association with their MI. The most common “something else” ( = non-CP) symptom was shortness of breath — but other non-CP symptoms incuded GI upset; “flu”-like symptoms; weakness; mental status changes; and syncope. We always look for "chest pain" when taking our history — but especially in a population of a certain age, any NEW symptom that temporally correlates with what you are seeing on ECG could be a non-CP "equivalent" symptom. HISTORY is key! Thanks again for your input! — :)

  4. very cool discussion, very interesting case and ecg. thank you Pendell, et al.
    one ecg begets another.



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