An elderly male presented with 48 hours of off and on chest pain.  This ECG was recorded with pain:

The QRS is 130 ms.  It is essentially LBBB, though some might quibble with the tiny Q-wave in aVL or the fact that the intrinsicoid deflection (onset to peak of R-wave in V5, V6) is not quite 60 ms (it is at least 55 ms). When I first glanced at this, I did not think it looked like STEMI (that is, I did not think it met the modified Sgarbossa criteria)

But then I measure it.  Here is V3 magnified:

Here I point out where the J-point and the PQ junction are:

You can see that, no matter which complex you measure, the ST elevation at the J-point is at least 3.5 mm.  The S-wave is no more than 13 mm. Thus, the ST/S ratio in V3 is greater than 0.25. There is thus proportionally excessively discordant ST elevation. Only one lead is required!

The cath lab was activated.

Another ECG was recorded, with reportedly ongoing chest pain:

Now the ST elevation is less.  The ratio is about 3/15 = 0.20. 0.20 is still very good for diagnosing coronary occlusion. However, look at lead V4: there is 1.5/5.0 = 0.30 ratio (the QRS is so different from the first ECG that I suspect some lead placement changes) This improving ECG makes it appear as if there is some reperfusion of the artery.

At angiogram, a 90% thrombotic lesion was found and stented.  I suspect that the artery had spontaneously reperfused prior to the angiogram.  That is why the ST segments were falling.

Here is the post-PCI ECG:

It is now much more atypical for LBBB and should be called a nonspecific intraventricular conduction delay.  QRS is 126 ms. The disproportionally excessive discordant ST elevation is gone.

This is 2 days later:

LBBB is gone.  There is normal conduction and anterior reperfusion T-waves.

Learning Point:

In LBBB, measure the ST elevation at the J-point and the PQ junction.  In a patient with the right clinical scenario, if there is one lead with a ratio greater than 0.25, then there is occlusion until proven otherwise.