Paucis Verbis: Sgarbossa’s Criteria with LBBB

Paucis Verbis: Sgarbossa’s Criteria with LBBB



It is difficult to determine if a patient with a left bundle branch block (LBBB) has an acute myocardial infarction (AMI) because ST segments are “appropriately discordant” with the terminal portion of the QRS. That means if the QRS complex is negative (or downgoing), the ST segment normally will be positive (or elevated). Similarly if the QRS complex is positive (or upgoing), the ST segment will be negative (or depressed).

PV Card: Sgarbossa’s Criteria

In 1996, Sgarbossa et al looked through the GUSTO-1 trial patients with LBBB and AMI. They derived 3 criteria which may help diagnose the “hidden” AMI. The criteria are:

1. ST elevation ≥ 1 mm concordant with QRS complex (most predictive of AMI of the 3 criteria)
2. ST depression ≥ 1 mm in lead V1, V2, or V3
3. ST elevation ≥ 5 mm where discordant with QRS complex

Use these criteria with caution though. None of these criteria are perfect. They are to help you risk-stratify. For instance, criteria #3 (ST elevation ≥ 5 mm) can exist in asymptomatic patients with LBBB because of concurrent left ventricular hypertrophy and high voltages.

Sgarbossa's Criteria


Thanks to Tom Bouthillet at for the useful illustration above.


Feel free to download this card and print on a 4” x 6” index card.
Go to the ALiEM Cards site for more resources.

Michelle Lin, MD
ALiEM Editor-in-Chief
Academy Endowed Chair of EM Education
Professor of Clinical Emergency Medicine
University of California, San Francisco
Michelle Lin, MD
Michelle Lin, MD

Latest posts by Michelle Lin, MD (see all)

  • Anonymous

    Great Overview!

  • Thanks! I figure people are looking at this brief review, since when you google “Sgarbossa Criteria”, my link is 3rd on the front page! Happy to help.

  • Our hospital still adheres to AHA criteria of activating the cath lab with new LBBB. There’s some great spiel of Sgarbossa and how Dr Steve Smith modified it (the 3rd criterion).

    Dr Smith:



    Your last statement is absolutely right, they’re never sensitive, just like generally the ECG isn’t.

  • Hi JD:
    Yes, Tom Bouthillet at told me about Dr. Smith’s modified Sgarbossa criteria using a ST/S ratio of >0.2 in discordant segments to be than Sgarbossa’s criteria #3. I would love to advocate for it, but I’m still waiting for some sort of publication to come out on this.

  • anuraj ms

    awesome explanation. thanks

  • Happy to help!

  • 1) Does the concordance refer to the st segment in relation to the QRS vector or the terminal portion of the QRS?

    2) In real life sometimes its difficult to tell where ST segment starts and ends?

    3) Do we measure from the highest portion of the ST segment? If it is rising do we start measuring at the J point or at the beginning of (t wave ) which is sometimes hard to see! This is important as sometimes the ST segment slopes up!

    4) In real life it is rarely LBBB often a mix of things! such as a wandering baseline from moving pts ( as they often do when having MI chest pain). Also what about fasicular blocks!? Is there a rule for this as this is what I often see.

    • Excellent questions. Sounds like the LBBB gives you a lot of angst, as it does for me as well!

      1. ST segment concordance is measured in relation to the QRS vector, although this is also the terminal QRS portion for LBBB. See great review:

      2. Yes, it is sometimes challenging to see where the ST segment starts.

      3. I typically measure at the start of the J point, although I often factor in the patient’s risk profile and history. I have a lower threshold to call an abnormal ST segment in the setting of a high pretest probability case.

      4. Wandering baselines and mixed blocks can indeed be challenging. You can have the patient sit on their hands for a few seconds or place layers of warm blankets over their chest to get temporarily stillness.

      Best of luck!

  • Well most of the pts have a cardiac abnl…that is why they warrant a ekg! If they are not a risk there pretest prob is often to low to warrant a un necessary test. however given they have baseline disease they often have abnl ekg with Strain patterns and LVH and LBBB. The problem i have especially with the faster rhythms of a standard pt in the ER having chest pain. The problem i have is find the point where the ST segement ends and the T wave begins! and i dont know if i measure where the elevation at the beginning of the ST segment or the End of the segment? if it is at the end of the segment, then its hard to tell if that portion is elevated vs part of the T wave! Maybe im splitting hairs here but it is very important! as i can see the future of this will be that pts likely benefit from PCI if they are having a STEMI and if it is NSTEMI troponin leak or UA then likely the benefit is none. So it this really is our call…

    however if you are measuring from the start of the ST segement that makes it much easier and i will try that however i guess i would rather call a false positive than a false negative as type II errors can be costly!

    I tried to have the patient sit on his hands! He got angry with me…” you want me to do WHAT”! “im hurtin doc!”

    1) good point about the terminal segement is also the direction of the vector as by definition this is the LBBB. As i was confused what i should do with a Rs pattern. However i guess i wont see that with LBBB right?