Saturday, November 24, 2018

Large T-waves and a Computer Interpretation of ***Acute MI***

This ECG was texted to me with no information:

I answered: "Show me the whole 12-lead."

Here it is:
Computer Interpretation:
***ACUTE MI***

What do you think?

This was my answer, in which I suspected that he was worried about possible hyperacute T-waves:

"I suspect these do not represent hyperacute T waves.  Although the T-waves tower over the R-waves, they have extreme upward concavity.  Although MI can easily have upward concavity, it is not usually this pronounced.  If it is a chest pain patient, I would get a formal echo and serial ECGs.  And look for an old EKG."

They found this old one scanned into the chart from a stress test.  There were no others:
Notice the similarities to the ECG above, confirming that they are baseline

Then he sent the clinical history, which was of a malfunctioning ICD, but without any chest pain or SOB.

And it turns out that this was the baseline ECG. There was no ACS here.


On this blog, I show a lot of ECGs that were sent to me and show subtle findings of OMI (acute coronary occlusion).

However, the vast majority of ECGs that are sent to me are NOT occlusion, but mimics.  Thus, the vast majority of my responses are "No, you don't have to worry about that one."

I need to show more such cases.

Not all large T-waves are hyperacute!
Some are baseline normal, especially in Early Repolarization
Some are hyperkalemia, but they are peaked and sharp.
Some are large but also with a high voltage R-wave, S-wave, or QRS, or by a wide QRS (e.g., LBBB, paced rhythm, LVH, early repol) and so not proportionally large

What makes a hyperacute T-wave?

1. Most important: Size of the T-wave, or total area under the curve, is greater in ischemic hyperacute T-waves.  It is NOT height (voltage) by itself that matters: the T-waves of early repolarization may be very tall in V2-V4, but the QRS voltage is also high.  I always say that hyperacute T-waves are "Fat," not just tall.
2.  T-wave size, or fatness, is larger when there is less concavity.  A straightening ST segment (less upward concavity) increases the area under the curve (or size, fatness) of the T-wave.*
3.  It is not absolute size, but size in proportion to the QRS.
4. Symmetry is important.  Hyperacute T-waves are more symmetric.  The T-waves of early repolarization have a slower upstroke than downstroke, and thus more upward concavity.

*Interestingly, in our study in which the LAD-Early repol formula was derived and validated, we attempted to evaluate the difference in upward concavity between subtle LAD occlusion and early repol, but could not find a difference.  Nevertheless, we were not evaluating LAD occlusion in the hyperacute phase when T-waves are most hyperacute, and so I am still convinced that there is a difference.

10 Cases of Inferior Hyperacute T-waves

Here are some other examples of V1-V3 hyperacute T-waves:

Notice on the right a typical normal T-wave.  
It has a slower upstroke than downstroke -- it is asymmetric
There is a lot of upward concavity on the upstroke.
There are well-formed R-waves

Comment by KEN GRAUER, MD (11/24/2018):
Excellent case — with the challenge being to interpret the initial ECG without any history (TOP tracing in Figure-1). The obvious concern is the presence of very tall and peaked T waves in multiple leads of ECG #1.
  • Are these hyperacute T waves in ECG #1, indicative of acute OMI (Occlusion-related Myocardial Infarction)?
Figure-1: TOP ( = ECG #1) — Initial ECG in this case. BOTTOM ( = ECG #2) — A prior tracing on this patient (See text). 
ECG #1: ( = the Initial Tracing):
As per Dr. Smith, despite dramatic peaking of these high-amplitude T waves in multiple leads — my suspicion was also that this probably did not represent an acute finding. My reasoning:
  • The abnormal T waves in ECG #1 have a very similar appearance with nearly identical-looking peaked T waves in no less than 8 leads (ie, leads I, II, III, aVF; V3-thru-V6 — and possibly in V2). Acute OMI is much more likely to be localized, rather than generalized to so many lead areas.
  • There are no reciprocal changes. While not an invariable finding — most acute OMIs that manifest hyperacute ST-T waves as markedly abnormal as seen in Figure-1 will show at least some mirror-image reciprocal change in at least one opposing lead area.
ECG #2: ( = a Prior Tracing):
I found the prior tracing to be quite interesting (BOTTOM ECG in Figure-1). As per Dr. Smith — there are striking similarities in T wave appearance in this prior tracing compared to the T waves seen in ECG #1. This suggests that these tall, peaked T waves in ECG #1 are a feature of this patient’s baseline ECG. That said — there clearly are some differences in ST-T wave appearance between ECG #1 and ECG #2. These include:
  • Despite no more than minimal change in frontal plane axis — the ST segment in lead aVL was previously (in ECG #2) coved and associated with fairly deep T wave inversion. This is not seen in ECG #1.
  • Similar ST coving with shallow T wave inversion was seen in the prior tracing in lead V1 — but is no longer present.
  • T waves seem to be disproportionately more prominent in the newer tracing (ie, in ECG #1) — to a much greater extent than I would expect from the changes I see in chest lead QRS morphology.
It is interesting to speculate on WHY these differences might exist between ECG #1 and ECG #2:
  • It could represent ischemia.
  • Since the reason for obtaining ECG #1 was “ICD malfunction” — perhaps there was a sustained arrhythmia — and T wave changes reflect a “memory effect”?
  • Since the prior ECG ( = ECG #2) was taken “from a stress test” — perhaps the patient was not supine at the time, or was exercising (or had just exercised) — ALL of which might alter ST-T wave appearance without necessarily indicating ischemic change.
BOTTOM LINE  Despite these differences in T wave appearance between ECG #1 and ECG #2 — the “theme” of the prior tracing is that disproportionately tall and peaked T waves were previously seen in multiple leads. This suggests that this finding in ECG #1 is less likely to represent a meaningful acute change (although admitedly, it does not completely rule out this possibility).
Additional Learning Points from This Case:
  • When worrisome-looking T waves are generalized and, look nearly-identical throughout the tracing — this finding is less likely to represent acute OMI, especially if reciprocal change is also absent. That said, IF you still have any doubt about whether the ECG is acute — then serial tracings; search for prior ECGs on the patient (for comparison); cardiac markers; bedside Echo during chest pain (looking for wall motion abnormality); and ongoing clinical evaluation may be needed until you can be more certain!
  • A history is essential for intelligent interpretation. Learning that the tall, peaked T waves in ECG #1 are not associated with new-onset chest pain immediately reduces the likelihood that these T waves represent a “hyperacute” change.
  • Availability of a prior ECG may provide invaluable assistance in support of our impression either for or against “hyperacute” changes. Lead-by-lead assessment is needed to determine IF meaningful change has (or has not) occurred since the earlier tracing was done.
Our THANKS to Dr. Smith for this superb discussion!


  1. Dr. Smith,

    Very much appreciate your blog.

    Quick question regarding QTc and paced rhythms
    1. How much do you read into the QTc?
    2. Should I avoid agents that prolong the QTc in paced rhythm?
    3. If you have a pt that has a QTc of 650 and has been there on all previous EKG's, is there anything to do?

    Thank you!!!

    1. I would use the Tpeak to Tend measurement. That's what we described for LBBB.

  2. @SteveSmith - is would seem that there is an extra wave at the qrs ending on v2. On the second full 12 lead picture? Could it be an epsilon?

    1. I suppose it's possible, but does not look a lot like it.


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