Saturday, May 21, 2011

LBBB: is there STEMI?

A 45 year old male with no history of cardiac disease presented with new onset pulmonary edema.  He was intubated prehospital.  BP before and after intubation was 110 systolic, with HR of 120.

Click on this ECG and try to interpret it yourself before reading the answer below:

There is sinus tach with LBBB.  There is no concordant ST elevation.  V4 has 2 mm of discordant ST elevation (at the J-point, relative to the PR segment) following a 5 mm S-wave.  The ST/S ratio is 0.40 in this lead.  Lead II has proportionally excessively discordant ST depression, with 1.25 mm STD and only 4.0 mm R-wave, for a ratio of 0.31.  This is also a sign if ischemia (reciprocal inferior ST depression).              Also, look at V3: complexes vary slightly: 2nd complex has approx 2.5-3.0 mm STE following a 14 mm S-wave; complex 4 has 2-2.5 mm STE following a 10.5 mm S-wave.   So these approach an ST/S ratio of 0.20, but it is not definite.   
Definitions in LBBB:
Concordance: ST segment is in the same direction as the majority of the QRS
Discordance: ST segment is in the opposite direction to the majority of the QRS
Rule of appropriate discordance: ST segments in all leads should be discordant to the majority of the QRS.
Smith rule: Discordance should be proportional to the QRS, with an ST/S or ST/R ratio no greater than 0.2

In a study of 20 patients with LAD occlusion, vs. 129 controls with ischemic symptoms and LBBB, at least one complex in V1-V4 with at least 2mm of STE and an ST/S ratio > 0.20 was highly specific for LAD occlusion (1).   Here is the reference for the abstract on proportionally excessively discordant ST depression (2).

Cases with excessive discordance of at least 5mm [Sgarbossa criteria 3] that did not have proportional discordance, did not have LAD occlusion.  The mean highest ST/S ratio for those without occlusion was 0.10 (95% CI: 0.09-0.11); the mean highest ST/S ratio for those with occlusion was 0.44 (95% CI: 0.19-1.05)

Because of this study, I believe the following rule is as good for diagnosis of STEMI in the setting of LBBB as standard interpretation of STEMI in the absence of BBB (and that it is more sensitive and specific than the Sgarbossa rule):

Smith modified Sgarbossa rule:

1) at least one lead with concordant STE (Sgarbossa criterion 1) or
2) at least one lead of V1-V3 with concordant ST depression (Sgarbossa criterion 2) or
3) proportionally excessively discordant ST elevation in V1-V4, as defined by an ST/S ratio of equal to or more than 0.20 and at least 2 mm of STE. (this replaces Sgarbossa criterion 3 which uses an absolute of 5mm)

It is important to remember that this is not sensitive for "MI" which is diagnosed by biomarkers. The lack of sensitivity of the Sgarbossa rule in previous studies is because the ECG is always (even without BBB) insensitive for MI.  It is, however, much more sensitive for occlusion.

Followup (Answer):
Because of proportionally excessive discordance in lead V4, (and, of course, clinical instability), the patient was taken for immediate angiography, which confirmed a 100% mid-LAD occlusion.

For a case with more than 5 mm of ST elevation in V1-V4, but without excessive proportional discordance, see this post:

Tom Bouthillet has done a great job of describing my ratio rule here:

To learn more about the meaning of New LBBB, look here:

Caution: these data have not been published in a peer review journal, and the ACC/AHA still (though I believe wrongly, and this recommendation is rarely followed) recommends reperfusion for patients with ischemic symptoms and new LBBB, even without any specific findings of STEMI.

1.      Dodd KW. Aramburo L. Broberg E.  Smith SW.  For Diagnosis of Acute Anterior Myocardial Infarction Due to Left Anterior Descending Artery Occlusion in Left Bundle Branch Block, High ST/S Ratio Is More Accurate than Convex ST Segment Morphology (Abstract 583).  Academic Emergency Medicine 17(s1):S196; May 2010.

2.     Dodd KW.  Aramburo L.  Henry TD.  Smith SW. Ratio of Discordant ST Segment Elevation or Depression to QRS Complex Amplitude is an Accurate Diagnostic Criterion of Acute Myocardial Infarction in the Presence of Left Bundle Branch Block (Abstract 551).  Circulation October 2008;118 (18 Supplement):S578.

If you want a .doc or .pdf of these abstracts, email me at:

If you request these, and I do not send them, it will be because I have received too many requests; in this case I will find a way to post them.  So look back here for more info.



  1. that looks like cabrera's sign in v3

  2. It is indeed Cabrera's sign. Notching of the upstroke of the S-wave in any of leads V3-V5 in LBBB. It is associated with MI in LBBB, but not necessarily with acute MI. Good pickup.

  3. Dr. Smith... thanks again for the great case.. you reference proportionally excessive discordant depression in lead II... i have noticed such before, but have not ever seen that referenced until your post. how much significance do you put in excessively discordant depression?

  4. Actually quite a lot. See our research in the second abstract. But this data is preliminary and would need confirmation in other studies. It's just hard to get anyone to study it.

  5. In LBBB, I thought one of the criteria is the lack of q waves in lateral leads. How does that fit in when considering STEMI in LBBB, though I understand the main concept with sgarbossa is discordance and concordance. -Josh

  6. By most definitions of LBBB, you're correct, and this should be called an nonspecific intraventricular conduction delay. However, that LBBB cannot have Q-waves is not universal. In a great article that has been lost to history, Chapman (of Chapman's sign) found that LBBB often develops Q-waves when the patient undergoes MI. full text:
    In any case, whether IVCD or LBBB, the issue as you say is concordance/discordance, a principle which applies to IVCD as well as LBBB. And in this case the excessive discordance in V4 was the clue to diagnosis of LAD occlusion.