Thursday, February 15, 2018

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

Submitted and written by Alex Bracey, with edits by Pendell Meyers and Steve Smith:

I was walking through the critical care section of the ED when I overheard a discussion about the following ECG. I had no history on the case and no prior ECG for comparison.

What do you think?

Here are inferior leads, and aVL, magnified:
A closer inspection of the inferior leads and aVL
Sinus bradycardia. 
The T-wave in lead III is slightly tall and broad (increased area under the curve) compared to its QRS complex. In isolation, this probably could not be called a hyperacute T-wave, but you may suspect it.   
There is T-wave inversion (TWI) in aVL.  

T-wave inversion in aVL: when is it abnormal?

There is no LVH or LBBB on which to blame the TWI (i.e., the QRS is normal). While T-wave inversion in aVL may be normal in the presence of a normal QRS, this is only true when the T-QRS angle is small. That is to say, when the T-axis and QRS axis are similar. In other words, if the QRS is negative, the T-wave may be negative. However, here QRS axis is about 35 degrees and the T-axis is about 85 degrees. Thus the T-QRS angle is 25 - 85 = (-60) degrees, which is abnormal. Any absolote value greater than 45 degrees is suspicious for T-wave inversion (however, this is very complex; see table posted at the bottom of this post.)

Now that we know the T-wave inversion in lead aVL following a normal QRS complex is abnormal, it helps to confirm that the T-wave in lead III is indeed hyperacute. The flattened T-wave in V2 suggests likely posterior involvement.

The cath lab was not activated based on this initial ECG. The patient was a female in her 60s with history of HTN and smoking who presented for chest pressure x1 hour. The initial ECG was taken at 0839. Based on the initial ECG and presenting complaint, the attending involved in the case opted to keep the patient in our critical care unit for close monitoring and serial ECGs.

She went on to describe her chest pain as a "buffalo sitting on my chest" and a "weird" sensation in her jaw for 1 hour prior to arrival, associated with lightheadedness and diaphoresis.

The patient was given fentanyl initially for chest pain with minimal effect and then vomited which was followed by zofran and famotidine. Initial troponin T from 0840 was less than 0.010 ng/mL (undetectable).

The following ECG was obtained at 0910:

Repeat ECG recorded 30 minutes after initial ECG. The patient still had chest pain.
Again, a close up of the interior leads and aVL

Sinus rhythm with borderline 1st degree AV block. There has been interval marked increase in the area under the ST-segment and T-wave in leads II, III, and aVF, with concomitant increase in the area above the inverted T-wave in aVL, all confirming that that these truly represent inferior hyperacute T-waves. The STE in the inferior leads is larger. There is now TWI and small STD in lead V2, highly suspicious for posterior MI. 

This ECG is diagnostic of an acute coronary occlusion of an artery supplying the inferior and posterior walls.

The cath lab was activated following this ECG and the cardiology fellow came to the bedside. The patient was given aspirin and ticagrelor and was scheduled for urgent cath.

In the cardiac cath holding area, a repeat troponin T was 0.01 ng/mL (positive). The following ECG as recorded at 1206:

Pre-cath right sided ECG. V1 and V2 are unchanged from the normal 12-lead sytem; V3-V6 are actually V3R-V6R.
The T-waves are even more hyperacute. Right sided leads (V3-V6 on this ECG correspond to V3R-V6R) have STE and hyperacute T-waves indicative of RV infarction.

Progression of Inferior leads and aVL

In this magnified arrangement you can see progression of subtle changes including the progressively increasing area underneath the hyperacute T-waves. Even at ~4 hours into her acute coronary occlusion there is barely any ST Elevation. 

Also notice that the ST segments in this example are concave, which is often erroneously mythologized as a non-ischemic pattern.  

Progression of V2 showing posterior involvement.

The patient was then taken to the cath lab an found to have a proximal RCA 100% thrombotic occlusion which was successfully stented.

100% occluded RCA with TIMI 0 flow

Post drug-eluting stent placement with TIMI 3 flow

While in the cath lab, she transiently developed complete heart block and became hypotensive requiring transvenous pacemaker placement and transient pressors. A right heart cath revealed increased right heart pressures and a similarly timed echo revealed mild right heart failure.

Peak troponin T was 3.00 ng/mL (highly elevated).

Post-cath ECG with resolution of acute changes.

The transvenous pacemaker was removed the following day and pressors were not required again. She was discharged to home 2 days later without further complications.

Learning Points:

1) As we have previously demonstrated, aVL was once again the key initial clue to diagnosing subtle RCA occlusion.

2) Contemporary troponins only start to rise 4-6 hours after the onset of acute coronary occlusion. Relying on troponin elevation to diagnose acute coronary occlusion after at least 4 hours of infarction when the ECG can identify it immediately is poor choice.

3) STEMI criteria failed to identify this acute coronary occlusion, like many others. Only expert ECG interpretation combined with strong clinical suspicion were able to identify this case. Remember that some acute coronary occlusions will present with totally normal serial ECGs, and some patients with extremely concerning symptoms warrant emergent cath lab activation even without ECG findings.

Normal QRS-T angle

From this article: Ziegler R and Bloomfield DK.  A study of the normal QRS-T angle in the frontal plane.  Journal of Electrocardiology 3(2):161-167; 1970.  

Yes, there are valuable articles from 50 years ago! 

For each QRS angle on the left column, Ziegler and Bloomfield have found the normal range of T angles. You can see that, for a QRS angle of 25, as in our patient, the normal range of T angles would be 2-69. So our patient really does have an abnormal T axis at 85 degrees, and the "inverted," or negative, T-wave really is abnormal inversion. A negative T-wave in aVL must have an angle of greater than 60 degrees.


  1. Dr. Smith and Contributors,

    The posts on your blog are truly incredible. I always leave this site feeling totally stupid and thanking God that I did not manage the patient discussed.

    Thank you for all of your efforts to educate.



  2. - Thank you so much for this case

    - Ziegler-Table 1970 !

    merci Al

  3. Thank Dr Smith Sir
    For sharing
    One more point v3,v4 swapped ??

    1. Ilyas, good point! Looks like they are!

  4. How was the QRS axis and T wave axis
    calculated with such accuracy? (i. e.QRS axis is about 25 degrees and the T-axis is about 85 degrees)
    Was the computer interpretation used?

    1. The computer does do this. I did not see the computer's numbers as the case was not mine. when editing the entire presentation, I simply looked at the ECG and calcluated the approximate axes.

    2. Correct my approach for rough estimation of QRS AXIS. I see QRS in Lead one and aVF and then lead 2,if all positive then axis is normal.then I look at these 3 leads and determine which lead has greatest R wave height. In this case its lead2, so roughly I estimated the axis little less than 60. But here its 25 which is way less, but cannot devote much time to calculate the exact axis. Any tips? What do you feel about my approach?

    3. This comment has been removed by the author.

    4. I look for the lead that is most perpendicular. In this case, lead III (120 degrees; it is as much positive as negative). Therefore, the axis is approx perpendicular to Lead III (30 degrees), but a little less (25). Use the computer measurements; they are accurate.

    5. I will offer a simple approach. Practically speaking — ALL you care about is determining if a negative T wave in lead aVL might be normal vs being indicative of a reciprocal change for a subtle acute inferior MI. That is, we don't need to worry about T wave axis if the T in aVL is positive. And, if there is frank ST elevation or frank ST depression — then we know this is abnormal. So, keeping in mind that T wave axis usually trails the QRS axis by a relatively modest amount — you ought to see an rS-type complex (with predominant negativity) in lead aVL IF the negative T wave in this lead is the result of a "normal variant". If instead you do not see a predominantly negative QRS in lead aVL — then it becomes unlikely that a negative T in aVL will be a "normal variant" — :)

  5. NICE case! As is well explained — in the context of new-onset worrisome chest discomfort that this patient had, the subtle-but-real finding of a taller-and-fatter-than-it-should-be T wave in lead III (given the QRS in this lead) + isolated T inversion-that-shouldn’t-be in lead aVL (and which looks like a “mirror-image” opposite of the ST-T wave in lead III) — should be assumed to be early acute inferior infarction until proven otherwise. And in the context of leads III and aVL — one has to acknowledge that the J-point in the other 2 inferior leads ( = leads II and aVF) is elevated, with a T wave shape identical to that seen in lead III. While ALL of these findings are extremely subtle — this makes for 4 suspicious limb leads in a patient with new chest pain, so the onus is on us to prove this isn’t early acute rather than the other way around … P.S. Lead V2 on this initial ECG most probably is the result of incorrect lead placement (One should not get an r’ in V2 that is not seen in V1 — and then transition so very abruptly occurs by the time V3 is recorded …). THANKS for presenting this insightful case!

  6. Very interesting. Any idea of calculating T axis other than computerized method?

  7. This blog is fantastic for the ecg lovers.There are plenty of explanations how to calculate QRS axis, in fact an online calculator is also available at present whereas there are no enough documentation regarding the calculation of T axis. Is anyone there to help?

    1. Any QRS axis calculator will work for the T axis just as well.


Recommended Resources