Monday, December 16, 2019

A 50-something with 5 hours of typical chest pain and Left Bundle Branch Block

This case was texted to me by one of our residency graduates, and with the outcome, so I don't know how I would have interpreted it blindly.

A 50-something male who is healthy and active with no previous medical history presented with 5 hours of continuous worrisome chest pain.

Here is his ECG (there was no previous ECG available):
What do you think?

There is sinus rhythm with Left Bundle Branch Block (LBBB).  This is presumably and new LBBB, but that is uncertain.

There is less than 1 mm of concordant ST Elevation in lead V5.  This is pretty specific for OMI but not very sensitive (see table below).

The highest ST/S ratio is in V3, and is 2.5/13 = 0.19.   0.19 is a high ratio, but not the most accurate.  The average highest ST/S ratio in V1-V4 for a normal LBBB is about 0.11.  0.20 is about 84% sensitive and 94% specific in our validation study.   So 0.19 is worrisome!

Chest pain with New LBBB: It helps to actually measure the ST/S ratio

A Fascinating Demonstration of ST/S Ratio in LBBB and Resolving LAD Ischemia

The cath lab was activated. Shortly thereafter, the first troponin I returned at 0.689 ng/mL (URL = 0.034), all but diagnostic of acute MI.

The LAD was 100% occluded.  It was opened and stented.

Subsequent Peak cTnI was 46.84 ng/mL (consistent with LAD occlusion).

There was an ECG recorded the next AM:
It does not appear much different!
But there are subtle differences:

1. The T-wave is not as tall.
2. There is the beginning of terminal T-wave inversion in V1-V3.

Here they are, side by side for comparison:
On the left is pre-PCI, right is post-PCI
T-waves are taller in V1-V3 on the right
There is very subtle terminal T-wave inversion on the left.
This is consistent with reperfusion
I do not have later ECGs, but unless there is a no-reflow phenomenon, the ECGs should evolved still more evidence of reperfusion.  Often this would include resolution of the LBBB.

The provider, and former HCMC EM resident, was exceedingly perceptive and able to accurately diagnose LAD occlusion from this very subtle ECG.


The doctor saw this ECG and had a sense, in combination with the worrisome clinical presentation, that this was an ischemic LBBB.  But it is by no means obvious, and does not strictly meet the Modified Sgarbossa Criteria, which require (in at least one lead):
1.   1 mm of concordant STE or
2.   1 mm of concordant STD in V1, V2, or V3
3.   Excessively discordant STE, as defined by at least 1 mm and with an ST/S ratio of 25% or more.

How about "New (or presumably new) Left Bundle Branch Block"?  This was formerly an indication for cath lab activation, but was abandoned in the 2013 guidelines because of poor specificity.  The best study was an ED study that showed that about 7% of patients with chest pain and new LBBB in the ED were diagnosed with acute MI.  The number with Occlusion MI (OMI) would be far fewer, in the range of 2-3%.  This is why the Modified Sgarbossa Criteria, and use of troponin and the following algorithm are so important.

Chang et al.  Lack of association between LBBB and Acute MI in symptomatic ED patients.

Interestingly, the rate of OMI among ALL patients with ischemic symptoms in the ED is only 1-5%, no matter whether there is LBBB or not! 

Here is an algorithm that is very useful in such cases, published by Cai and Sgarbossa in 2013.
Algorithm for LBBB (also good for managing OMI in paced rhythm)

So in this case, if the provider had not been certain of the ECG, he could have waited for the troponin.  Since the pain had been constant for 5 hours, it would be unlikely to be negative in the presence of acute OMI.  Then, after a short wait for the troponin which was indeed elevated, cath lab activation is indicated regardless of the ECG.

This is also true in normal conduction, not just LBBB!

If a patient has symptoms of ACS, and they are persistent, and a diagnostic troponin, then cath lab activation is indicated.

Here is a paper we published on other measurements and their diagnostic utility for occlusion (see Table):

--> Dodd KW.  Elm K.  Smith SW.   Comparison of the QRS Complex, ST-Segment, and T-Wave Among Patients with Left Bundle Branch Block with and without Acute Myocardial Infarction.  Journal of Emergency Medicine 2016. 

MY Comment by KEN GRAUER, MD (12/17/2019):
Great case about some subtleties in association with LBBB. I’ll focus on the 2 ECGs done in this case — which I have labeled and put together for clarity in Figure-1.

Figure-1: The 2 ECGs shown in this case (See text).

The patient in this case is a previously healthy 50-something male — who presented with 5 hours of continuous worrisome chest pain at the time ECG #1 was done. There was no prior tracing available for comparison.
  • As per Dr. Smith — ECG #1 shows sinus rhythm with complete LBBB (Left Bundle Branch Block).
I would make the following points regarding ECG #1. Some of these points were already emphasized above by Dr. Smith. Others reflect a slightly different perspective ...
  • As per Dr. Smith — the fact that this patient had a “worrisome” history for new chest pain that has persisted + the diagnostic initial troponin are indications for prompt cardiac cath, even if there were no ECG signs of acute OMI.
  • This patient presented to the ED with complete LBBB. As per Dr. Smith — without availability of a prior tracing, we simply do not know if this LBBB is new or not. The important point is that new (or presumably new) LBBB is no longer by itself an indication for cath lab activation. However, it would be an indication for cath lab activation — IF there were acute findings on this ECG with LBBB.
ECG #1 is a difficult tracing to interpret. Assessment of ST-T wave changes in the presence of LBBB is challenging, even for the expert interpreter.
  • We are indebted to Dr. Smith for developing Modified Smith-Sgarbossa Criteria for assessing chest pain patients with LBBB. Measuring the amount ST elevation as the difference between the PQ junction baseline and the height of the J-point — in a patient with the right clinical scenario — IF there is ≥1 lead with an ST/S wave ratio ≥0.25, then assume acute OMI until proven otherwise (Excellent illustration of these measurements in the Dec. 21, 2015 post by Dr. Smith).
  • 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 — this approach is experiential. It has worked well for me over many years ...
CONFESSION: When I first looked at ECG #1 — I was not convinced from this ECG that the changes were diagnostic of acute OMI. It may be helpful to expand on my initial assessment of this ECG.
  • For clarity in illustrating measurement of Dr. Smith’s ST/S ratio — I’ve labeled the spot I am calling the J-point in leads V1, V2 and V3 of ECG #1 with GREEN arrows.
  • In lead V2 — I measure the J-point at 3mm, and the S wave = 25 mm. This comes to a ratio of 3/25 = 0.12, which is clearly not diagnostic. This illustrates the purpose of proportionality that Dr. Smith introduced with his “modified” Smith-Sgarbossa criteria. Considering how deep the S wave in lead V2 is — the amount of ST elevation in V2 is not significant in the presence of LBBB. Similarly, the amount of ST elevation in lead V1 of ECG #1 is not significant, given the depth of the S wave in lead V1.
  • However, in lead V3 — the S wave is not nearly as deep (I measure 14 mm). I measured 2.5 mm of J-point ST elevation in lead V3 — which comes to a ratio of 2.5/14 mm = 0.18 — which is a relatively high ratio, albeit falling short of the 0.25 diagnostic ratio.
  • Lead V5 in ECG #1 caught my eye. Normally, the ST segment and the T wave should both be oppositely directed to the positive QRS deflection seen in left-sided leads (as it is in leads I, aVL and V6 in ECG #1). Instead of the T wave in lead V5 being negative (as it should be) — it is upright! And instead of slight, downsloping ST depression — this is a tiny-but-real amount of ST elevation in lead V5. This is not normal. That said — the shape of the ST segment in lead V5 is flat (horizontal) — and therefore to my eye, not necessarily acute (with NO prior tracing available for comparison)!
  • Given these subtle (but not diagnostic) abnormalities in leads V3 and V5 — I focused on the in-between lead V4. While it clearly is difficult to assess disproportionate T wave peaking in a lead like V4 with overall reduced QRST amplitude + LBBB — I thought the T wave in V4 was disproportionately taller-and-more-peaked-than-it-should-be.
  • I didn’t think any limb leads showed any ST-T wave changes of note.
  • BOTTOM LINE on ECG #1  This 50-something man presented with new-onset worrisome and persistent chest pain. He has complete LBBB + a number of subtle abnormal ST-T wave findings on his ECG. That should be enough to merit prompt cath — even though it is not diagnostic of acute OMI. As soon as the initial troponin came back positive — there definitely was enough to merit prompt cath. CREDIT to the treating clinicians who ensured optimal management of this patient!
MThoughts on ECG #2: As noted above — ECG #2 was obtained the next day after PCI for total LAD occlusion was performed. Dr. Smith has drawn attention to some of the subtle changes between ECGs #1 and #2. For the purpose of HONING your ECG diagnostic skills — Let’s IMAGINE that ECG #2 was the initial tracing you saw, on this patient who presented with new chest pain.

QUESTION: If the initial (and only) ECG that you saw on this patient with new-onset chest pain was ECG #2 How would you interpret this tracing?

MANSWER: In this context — ECG #2 would be diagnostic of acute OMI!
  • The rhythm is sinus. There is again complete LBBB.
  • There can be NO doubt that lead V5 is abnormal and looks acute! Although the QRS complex is tiny — there is significant (relative to QRS size) ST elevation with a hyperacute appearance of the ST-T wave in this lead. (By comparisonI thought the ST-T wave in lead V5 of ECG #1 did not necessarily look acute.)
  • I believe Lead V6 also shows a tiny amount of ST elevation. This is significant — because as mentioned earlier, the ST segment and the T wave should be oppositely directed to the positive QRS deflection seen in left-sided leads (as it is in lead aVL). Instead of the ST segment being slightly depressed and the T wave in lead V6 being negative — both are upright in ECG #2!
  • By the principle of neighboring leads” — lead V4 in ECG #2 also looks qualitatively abnormal. That is, the J-point in V4 is angled — the ST segment itself is straightened (BLUE slanted line in V4) — and, the T wave in V4 looks a bit fatter-and-more-prominent-at-its-peak than expected. I will emphasize that these changes are subtle, and by itself — I would not be at all certain that the ST-T wave appearance in lead V4 is abnormal in a patient with LBBB. But given definite acute abnormalities in neighboring leads V5 and V6 — these subtle changes we see in lead V4 are virtually certain to be real.
  • Extending the principle of “neighboring leads” one lead further — I think the ST segment straightening (slanted BLUE line) that we see in lead V3 is probably also real.

Our THANKS to Dr. Smith for presenting this case!


  1. Replies
    1. Thanks for your comment Nicolas. By the Qualitative Approach that I mention above (in My Comment, at the bottom of the page) — I have not found the inferior leads reliable in manifesting appropriate discordance with LBBB. If we apply this to ECG #1 — lead II is problematic because of baseline artifact that produces 3 different appearances for the 3 QRS complexes in this lead. That said, while the 1st and 3rd complex in lead II of ECG #1 clearly show unusual ST segment flattening — I do not believe the upright T wave in lead II is necessarily abnormal. Similarly, lead aVF in ECG #1 shows suspicious ST segment flattening — but in my opinion, the upright T wave in this lead is NOT necessarily abnormal. On the other hand — by the qualitative approach I favor — left sided leads with LBBB in which there is a monophasic R wave should show an oppositely directly ST segment and T wave — and I AGREE completely with you, that the upright T wave in lead V5 of ECG #1 is NOT normal (Please see My Comment above — that I believe I wrote AFTER you had already submitted your comment). THANK YOU again for your interest! — :)


DEAR READER: I have loved receiving your comments, but I am no longer able to moderate them. Since the vast majority are SPAM, I need to moderate them all. Therefore, comments will rarely be published any more. So Sorry.

Recommended Resources