Wednesday, December 27, 2017

Ed Burns (from Life in the Fast Lane) sent me this one....

Ed Burns (who is the creator of the incredible, fantastic, Life in the Fast Lane ECG library) sent me this one....

This is a from a patient with chest pain:

What do you think?
There is very little ST elevation, only 1 mm at the J-point in V2 and V3

Ed's message was this:

Hi Steve,

A colleague sent me this ECG today.
I responded saying that it was a STEMI...
I thought the STs were a little too straight in V2-3 and the T waves a little tall, plus the reciprocal changes in lead III.

Can you improve on my assessment a little?
What is it that definitively makes the diagnosis of STEMI in this case?



Here was my response:

Good call!

There are several features:
First, as you said, there is a nearly straight ST segment.  It is very rare to have non-concavity (convex or straight) in any one of leads V2-V6 in normal variant ST elevation
Second, the QT appears slightly long for early repol.  Remember early repol is called early repol because repolarization comes early (relatively short QT)
Third, there is not enough R-wave amplitude in V4 for the T-wave size
Fourth, there is not enough QRS amplitude in V2 for that T-wave size.

The 3- and 4-variable formulas take the second, third, and fourth issues into account (see below):

Leads III and aVL are indeed somewhat suspicious; however, there really is no ST depression in III and a negative T-wave is normal, especially in the presence of a negative QRS (QRST angle is very small, less than 30 degrees)

Thanks for sending!


One more comment: I would not call it a STEMI, as this diagnosis is associated with ST Elevation "criteria" which this ECG does not meet.  One might call it:

1. Subtle STEMI

2. "Semi-STEMI"
3."Subtle occlusion" (in this case, subtle LAD occlusion)

Formulas use the following measurements
QTc =  (manually measured)                                                      = 400 ms  
ST Elevation at 60 ms after the J-point in lead V3 (STE60V3) = 2.5 mm
R-wave amplitude in V4 (RAV4)                                               = 9.5 mm
QRS amplitude in V2 (QRSV2)                                                 = 17 mm.

3-variable (STE60V3, RAV4 and QTc) = 23.48 (greater than 23.4 predicts LAD occlusion)
4-variable (adds QRSV2)                       =18.34 (greater than 18.2 predicts LAD occlusion)

These 2 cutoffs are the most accurate, not the most sensitive, nor the most specific.  

At 23.4, 3-variable formula had sens, spec, and acc of 86%, 91% and 88%

At 18.2, 4-variable formula had sens, spec, and acc of 89%, 95%, and 92%

The 4-variable is better for both but only in a derivation sample (needs validation!  Anyone??).

The patient did indeed have an LAD occlusion.

Final note on T-wave size: how do the formulas take T-wave size into account??   In our study, T-wave amplitude was not significantly greater for LAD occlusion vs. normal variant ST elevation (early repol also has large T-waves).  But large T-waves are only normal when there is high QRS voltage, as in normal variant.  ST elevation at 60 ms after the J-point is a measure of the slope of the ST segment; the higher the STE60V3, the steeper the slope. A steep slope correlates with a large T-wave and a flat slope with a smaller T-wave. 

Links to articles

3-variable formula:

4-variable formula:


  1. Thanks for sharing
    I hv issue about calculations
    R in v2 10(u hv 17)
    St elevation in v3 60ms after j point 2.5
    R in v4 9.5
    Qtc 338 ms (400 ms according to u, pz clarify which lead is selected)
    Four variable formula show 16.18 which is away from final outcome
    Pz elaborate in details (correct me where I made mistake)

    1. I measured QT in V2 and it is 400, with a rate of 60, the QTc is also 400. 338 is the reason your value is so low. As for V2: it is the total QRS, not the R-wave; total QRS is 17 mm

    2. Thanks sir for yr valuable reply
      If computerised Qtc is not available than which is the best lead to select for manual calculations

    3. Choose the longest QT interval of the 12 leads. This is usually II or V2 or V3

  2. Highly insightful tracings submitted by Ed Burns (from LITFL). Dr. Smith nicely documents the abnormalities in both his 3- and 4-variable formula. I’ll add a few qualitative thoughts. One already begins with a high-prevalence situation, given that the patient apparently presented to an ED with chest pain. The shape of the ST-T wave in lead V3 is eye-catching (not only because of straightening of the ST segment takeoff, but because of the uncharacteristically wide T wave base). Neighboring leads (ie, V2 and at least V4, if not also V5) manifest similar, albeit less marked shape abnormality. There is subtle-but-real ST elevation in lead aVL (which in the context of suspected acute anterior stemi IS significant) — and not only lead III, but also other inferior leads II and aVF show subtle-but-real ST-T wave abnormalities (there is ST segment straightening in both of these leads for the initial part of the ST segment — and the T wave in lead aVF, in the context of the tiny amplitude QRS in this lead is disproportionately tall). So while “criteria” for a stemi are not actually met (since there isn’t ST “elevation”) — one has to be highly suspicious from the overall ( = “Gestalt”) picture of subtle-but-real qualitative ST-T wave abnormalities in at least 8/12 leads, that this patient with new chest pain is about to evolve a stemi. THANKS for presenting this highly illustrative case.


Recommended Resources