Thursday, September 17, 2020

A 60-something Woman with Chest Pain and a Wide QRS


MY Comment by KEN GRAUER, MD (9/17/2020):


Today’s patient is a previously healthy, 60-something year-old woman who presented with chest pain that began at a reception. Her initial 12-lead ECG that was obtained by paramedics in the field is shown in Figure-1.

  • How would YOU interpret this 12-lead ECG?
  • Should you activate the cath lab?

Figure-1: The initial ECG in today’s case, obtained by paramedics in the field. (See text).

(Smith's quick comments are at the bottom)


MY Thoughts on ECG #1: There is a regular sinus rhythm at ~80/minute in ECG #1. The PR interval is at the upper limit of normal (ie, not more than 0.21 second in duration) — but the QRS complex is clearly widened.

  • The reason for QRS widening in ECG #1 is that there is complete LBBB (Left Bundle Branch Block). LBBB is diagnosed by the presence of QRS widening with a monophasic R wave in lateral leads (as seen here in leads I and aVL+ a predominantly negative QRS in anterior leads (as seen here in lead V1 and the chest leads that follow). NOTE: Usually with LBBB — a monophasic R wave is also seen in lead V6, which is not the case here (I’ll discuss the reason why in detail momentarily).
  • Beyond-the-Core: It is of interest to note that the QRS does not appear to be widened in leads V5, V6 — and perhaps also in lead aVF. However, a look at QRS width in leads aVR and aVL (obtained simultaneously to lead aVF) — and, at QRS width in lead V4 (obtained simultaneously to leads V5 and V6) — confirms that the QRS complex is indeed wide throughout the entire recording of ECG #1. A portion of the QRS complex in those leads that look narrow simply occurs on the baseline. This highlights the importance of always measuring intervals in that lead that looks widest!

Returning to Our Systematic Approach to ECG #1:

Since LBBB is present in ECG #1 — We need to adjust our criteria for interpretation of the QTc interval, axis, chamber enlargement, and detection of QRST abnormalities accordingly. By this I mean that the usual criteria for ECG diagnosis are different when there is a conduction defect. This is especially true with complete LBBB — in which the initial vector of ventricular depolarization is altered, because the left side of the interventricular septum can no longer be activated (since the path of the common left bundle branch courses down the left side of the septum).

  • The QTc interval in ECG #1 appears slightly prolonged. I measure a QT = 400 msec. in lead V3, which corrects to a QTc ~450-460 msec., given the heart rate of ~80/minute. That said, given the conduction defect — this slight QTc prolongation is not unexpected (and it does not significantly impact management decisions). The QTc is virtually always prolonged when there is LBBB.
  • Since LBBB changes the sequence of LV depolarization — the usual voltage criterial for LVH (which have been derived from patients with a normal QRS duration) are no longer valid. The criterion I favor for diagnosis LVH on ECG in association with LBBB — is the presence of extremely deep anterior S waves (ie, ≥25-30 mm in lead V1 or V2 or V3). This is clearly not present in ECG #1.
  • The concept of “QRS axis” means little in the presence of LBBB.
  • The presence of Q waves (or QS complexes) in inferior and anterior leads is much more difficult to interpret when there is LBBB. This is because (as I note above) — the initial vector of ventricular depolarization is altered with LBBB. As a result — the Q wave we see in lead III of ECG #1 — and the QS complexes in anterior leads V1V2 and V3 are not indicative of infarction — since Q waves are commonly seen in all of these leads with uncomplicated LBBB.
  • PEARL #1: There should never normally be a Q wave in lead I or lead V6 with simple LBBB. This is because block of the left bundle branch (which runs down the left side of the septum) causes a reversal of the normal left-to-right direction of septal activation. As a result, with LBBB — septal activation occurs from right-to-left — which means that initial electrical activation of the LV moves toward the left ventricle, which should write an upright R wave on the ECG. IF ever you see an initial Q wave in either lead I or lead V6 with LBBB — this strongly suggests there has been septal infarction at some point in time!
  • In the context of PEARL #1 — I interpreted the tiny rsr’ complex in lead V6 of ECG #1 as a Q wave “equivalent” in this patient with LBBB. As we will see in a moment in ECG #2 — a definite infarction Q wave develops just a few minutes later in lead V6.
  • Beyond-the-Core: With simple LBBB (in which there has not been infarction) — the predominantly negative QS or rS complexes seen in anterior leads undergo transition (ie, become more-positive-than-negative) at some point between leads V4-to-V5-to-V6. It is uncommon (though not impossible) for there to be no positive R wave at all with LBBB as far lateral as lead V5. That said — the negative QRS complex in lead V5 of ECG #1 clearly distinguishes itself because in addition to being all negative, the QS in lead V5 manifests a terminal notch — and THAT is clearly abnormal for simple LBBB. In the context of the Q wave “equivalent” pattern we see in lead V6 — I interpreted the entirely negative QRS with terminal notch in lead V5 as also being an infarction Q wave “equivalent”.

Interpreting ST-T Wave Changes with LBBB:

In addition to changing the sequence of LV depolarization — complete LBBB also changes the sequence of ventricular repolarization. As a result, we need to alter our approach for interpreting the meaning of ST-T wave changes when there is LBBB — because anterior leads will often show ST elevation with tall, peaked T waves — and lateral leads commonly show ST-T wave depression, simply because of the LBBB.

  • NOTE: Because LBBB changes the sequence of LV repolarization — it may be more difficult to recognize acute coronary occlusion on ECG.
  • KEY Point: There are still many cardiologists who believe that you can not recognize acute MI when there is LBBB. Those cardiologists are wrong. As we’ve shown many times on Dr. Smith’s ECG Blog — it most definitely is possible in a surprising percentage of cases to recognize acute ECG findings despite the presence of LBBB.

We are indebted to Dr. Smith for developing Modified Smith-Sgarbossa Criteria for assessing ST-T wave changes in chest pain patients with LBBB. I describe and illustrate in detail my approach to incorporating these criteria in My Comment at the bottom of the December 16, 2019 post in Dr. Smith’s ECG Blog. Dr. Smith illustrates how to measure these parameters with magnified views in his December 21, 2015 post.

  • While I definitely find Modified Smith-Sgarbossa Criteria helpful in assessing LBBB tracings in patients with chest pain — I favor in addition, a qualitative approach that is based on ST-T wave appearance rather than measurements. Since measurements aren’t used in my qualitative approach — I fully acknowledge that this approach is experiential. It has worked well for me over many years ...
  • PEARL #2: As helpful as Smith-Sgarbossa Criteria can be — I like to look for ST-T wave changes that I know are not normal in a patient with LBBB. Thus, although the QS complex in lead III of ECG #1 is not necessarily abnormal when there is LBBB — there should not be ST elevation in leads III and aVF of the shape we see here. In lead III — the J-point is clearly 2 mm elevated — the ST segment takeoff is straight — and the T wave is disproportionately tall considering the amplitude of the QS complex in this lead. In lead aVF — it is even easier to appreciate the hyperacute T wave, which is disproportionately tall with respect to the very small QRS complex in this lead. There is also 1 mm of ST elevation in lead aVF.
  • Continuing with my qualitative approach in ECG #1 — there is also a hyperacute T wave in lead V5 (ie, the T in V1 is disproportionately tall with respect to the small QRS complex in this lead.).
  • In view of the Q wave “equivalent” we noted earlier in lead V6 — the T wave we see in this lead is also hyperacute (ie, Normally the ST-T wave will be negative in lateral leads with LBBB).

Final Touches re ECG #1: Had I not recognized the above-noted definitely abnormal findings in ECG #1 — I probably would not have identified abnormal findings in the remaining leads. But since I recognized the Q wave “equivalents” in leads V5 and V6 — and since I know the ST-T waves in leads IIIaVF and V5V6 are all abnormal in this patient with LBBB — I looked again at the remaining leads:

  • In addition to the negative ST-T wave in lateral leads I and aVL (that is a normal accompaniment of simple LBBB) — there is also ~2 mm of J-point depression in both of these leads. In the context of the abnormal ST elevation we see in leads III and aVF — I interpreted this mirror-image J-point depression as a reciprocal change in this LBBB patient whose ECG is diagnostic of an acute inferior STEMIAnd, in the context of our diagnosis of acute inferior STEMI — I strongly suspected that the ST-T wave in the remaining inferior lead (ie, in lead II) — was probably also abnormal (ie, showing early signs of broadening the T wave peak before becoming frankly hyperacute).
  • Although the ST elevation with tall positive T waves in leads V1, V2 and V3 of ECG #1 is not unexpected with simple LBBB — in the context of hyperacute T waves leads V5 and V6, I thought the T wave in neighboring lead V4 was most probably also hyperacute (ie, disproportionately taller, fatter-at-its-peak and wider-at-its-base than it should be).
  • BOTTOM Line: To address the Questions I posed at the beginning of this case — In a 60-something year-old woman with new-onset chest pain — despite the presence of LBBB (and even without the benefit of a prior tracing for comparison) — ECG #1 is diagnostic of acute infero-lateral OMI. As a result — the cath lab could have been activated on the basis of the History given and ECG #1.

The Case Continues: The initial ECG was transmitted to the ED physician at the hospital. In route — ECG #2 was obtained approximately 6 minutes after ECG #1 (Figure-2). On seeing both of the tracings in Figure-2 — the paramedic team and the ED physician activated the after-hours cath lab. However, the cath lab activation was cancelled by the on-call interventionalist — who felt the diagnosis of acute STEMI could not be made because of the LBBB.

  • Is ECG #2 helpful in YOUR assessment of this case?
  • Do YOU agree with the decision of the paramedic team and the ED physician to activate the cath lab? — OR — Do YOU agree with the interventionalist to cancel cath lab activation?

Figure-2: The 2 ECGs in today’s case. Both tracings were obtained by paramedics in the field. (See text).

MY Thoughts on the Serial ECGs in Figure-2: If there was doubt about whether or not to activate the cath lab on the basis of ECG #1 — the decision to activate the cath lab should be indisputable after seeing ECG #2.

  • The amount of ST elevation in leads III and aVF has clearly increased in ECG #2 compared to ECG #1 (which was done just 6 minutes earlier).
  • What I previously suspected was the earliest signs of a hyperacute T wave in lead II — has now evolved into ST elevation with a much-fatter-than-it-should-be peak to its T wave.
  • The amount of J-point depression in high lateral leads I and aVL has increased since ECG #1. In addition, the shape of the depressed ST segment in lead aVL has straightened — so that this now clearly reflects a perfect mirror-image opposite picture of the ST elevation in lead III — which confirms that these are indeed reciprocal ST-T wave changes of an acute inferior STEMI.
  • Subtle: Did you notice in ECG #2 that there is loss in the amount of J-point ST elevation, with a change in shape of the ST-T wave in leads V1 and V2? Leads V1 and V2 most definitely do not look like these leads should look like with simple LBBB. In the context of everything else that is occurring — this changed appearance in leads V1 and V2 of ECG #2 indicates acute posterior involvement!
  • Finally, in the lateral chest leads — there is now a definite Q wave in lead V6, and the hyperacute ST-T wave in this lead is now clearly elevated. The QS complexes and hyperacute ST-T waves in leads V4 and V5 have stayed about the same, and remain distinctly abnormal.
  • BOTTOM Line: ECGs #1 and #2 are diagnostic of LBBB with an extensive acute infero-postero-lateral STEMI that shows ongoing evolution (ie, dynamic changes) occurring within the short span of 6 minutes between the timing of these 2 tracings. This emphasizes the diagnostic utility of obtaining frequent repeat ECGs.

CONCLUSION to the Case: The interventionalist finally took the patient to the cath lab. There was 100% occlusion of the RCA, which was stented.


NOTE: My sincere THANKS to Emmanuel Reisman (New York) for sharing the tracings and this case with us!


Smith Quick Comments:

Great case and great discussion!
The modified Sgarbossa criteria are only 84% sensitive (if you use 20%) in our studies (Meyers Validation study), and if used on a consecutive group of chest pain patients with LBBB, it would probably be lower.  So indeed we need to look beyond these criteria in order to NOT miss OMI in LBBB.

That said, this initial ECG does meet the modified Sgarbossa criteria.

Here, aVF has discordant STE of 25%.  The ST segment is 1 mm and the S-wave is 4 mm.  Thus, there is one lead with at least 1 mm of STE and a ratio of 25%, and so the criteria are met.

Even if you measure the ST segment as 0.75 mm, 0.75 divided by 4 = 0.19 (19%) which is also quite specific for OMI.  However, all our measurements in our studies were to the nearest 0.5 mm. So cases like this would have been classified as "1 mm concordant STE.

In our studies, the mean maximal ST/S ratio for LBBB is 0.11.  The mean overall is ratio 0.85, so a value of 0.1875 is very high indeed.

Also, lead II has a concordant ST segment.  To be positive by the modified Sgarbossa criteria, 1 mm of concordant STE is required in at least 1 lead. This does not meet that criteria.  However, we did find that the specificity of 0.5 mm of concordant STE in any lead was 92% (see reference).

So even if you don't have the experience of morphology recognition, ECG 1 can be identified as representing OMI by using measurments and applying the Modified Sgargossa Criteria.


  1. Am I wrong or the tracings would even meet the original Sgarbossa
    criteria? Not to say that we should use them as such, obviously.

    1. @ Marco Garrone — The original 3 criteria from the 1996 Sgarbossa scale that are used to diagnose infarction in patients with LBBB were:
      i) Concordant ST elevation > 1mm in leads with a positive QRS complex (score 5);
      ii) Concordant ST depression > 1 mm in V1-V3 (score 3);
      iii) Excessively discordant ST elevation > 5 mm in leads with a -ve QRS complex (score 2).

      A total score of ≥ 3 is reported to have a specificity of 90% for diagnosing myocardial infarction.

      In ECG #1 — I do not see concordant ST elevation — nor do I see concordant ST depression. Because the original Sgarbossa criteria do NOT account for relative proportionality — even criterion iii) is not satisfied, because the amount of ST elevation in leads III and aVF is clearly LESS than 5 mm — so NO, original Sgarbossa criteria would not have been met.

      The benefit of the Smith-modified Criteria — is that it DOES account for proportionality, as he illustrates in his comment above in which he explains how his Modified Criteria ARE met for this tracing. But particularly for ECG-1 — there are so many leads that are clearly abnormal in shape, in this patient with new-onset chest pain — that the diagnosis should NOT be missed. It DOES take some practice looking at these BBB tracings to appreciate these ST-T wave abnormalities (as well as to appreciate the abnormal QRS morphology in leads V5,V6) — but with careful attention to detail + increasing your comfort to recognizing what is “normal” with LBBB (and RBBB) patterns — and at-the-least, remembering to get FREQUENT serial tracings when doubt exists that you compare using the Lead-BY-Lead technique — acute OMI despite LBBB can often be recognized much of the time. Remember that in today’s case — in just 6 MINUTES — indisputable serial changes became evident! THANKS again for your question!

  2. Thought I read comments on here that you cannot tell the location of the occluded coronary when interpreting wide complex.

    1. rediction of the occluded coronary artery is not 100% accurate when there is sinus rhythm and a narrow QRS. Things such as anatomic variants — variations in collateral flow — multi-vessel diseaase — prior infarctions — etc. can all affect the accuracy of predictions. When there is sinus rhythm with bundle branch block, prediction of the “culprit” artery may be even more challenging. That said — sometimes you CAN still tell the location of the acutely occluded artery, despite the presence of BBB. We saw this in today’s case — in which despite the LBBB, the 2 serial ECGs showed acute evolving infero-postero-lateral infarction that allowed us to accurately predict acute RCA occlusion.

  3. Great case! I wonder can this be an atrial sense ventricular pace rhythm? since there are notch before the nadir of S wave in aVR, v1, v2, referred as josephson sign, which are not the typical morphology for LBBB(also mentioned in brugada algorithm and vereckei algorithm). Or maybe this sign isn't specific enough?

    1. There was no pacemaker. As per the history (provided in the 1st sentence above) — this 60-something year old woman had previously been healthy.


DEAR READER: We welcome your Comments! Unfortunately — due to a recent marked increase in SPAM — we have had to restrict commenting to Users with a GOOGLE Account. If you do not yet have a Google account — it should not take long to register. Comments give US feedback on how well Dr. Smith’s ECG Blog is addressing your needs — and they help to clarify concepts of interest to all readers. THANK YOU for your continued support!

Recommended Resources