Friday, February 18, 2022

A comatose patient with a carbon monoxide level over 50%

A young man had an accidental exposure to carbon monoxide (CO).  He was comatose and intubated and his initial Carboxyhemoglobin level was over 50%.

An ECG is always recorded for CO toxicity.  This was his ECG.  

It was shown to me with worry for ischemic ST elevation, which is certainly possible from severe CO toxicity, or concomitant ACS.  

In fact, there is laboratory evidence that CO toxicity increases Platelet–neutrophil aggregates and plasma myeloperoxidase (MPO) concentration and thus may precipitate ACS (though this is by no means clinically proven).  Intravascular Neutrophil Activation Due to Carbon Monoxide Poisoning

What do you think of this ECG?

My interpretation was that this was all normal variant ST Elevation.

Inferior STE: there are prominent J-waves in all leads.  There is no reciprocal ST depression in aVL.

Anterior STE: Prominent J waves and high QRS amplitudes.

If we use the LAD-normal variant formula, we get:

QTc = 436 ms 

R-wave V4 = 13 mm, QRS V2 = 21 mm, STE at 60 ms after J pt in V3 = 2.5

Formula value = 18.7.  This is suggestive of LAD occlusion.  However, the variable which most contributes to this is the QTc of 436 ms, which is quite long and skews the result of the formula to LAD occlusion.  For comparison, if the QT were an average of that of LAD occlusion (420 ms) vs. Normal variant (390 ms) from our study, the QTc would be 405 ms and the formula value would be 17.06, which is rarely seen in LAD occlusion

What we have evidence for, but is not widely known, is that CO Toxicity likely lengthens the QT interval.  So this relatively long QT interval is NOT due to ischemia but may be a result of CO Toxicity.  Yelken B et al. The assessment of QT intervals in acute carbon monoxide poisoning

I was not worried about this ECG.

The coma completely recovered with 3 hyperbaric oxygen treatments in our world class state-of-the-art-facility.  

Read more about CO poisoning and cardiac ischemia here (ECG is pasted below): What is the treatment for this subendocardial ischemia?

Troponin peaked at 43 ng/L (barely elevated; most patients with CO toxicity this severe have significantly elevated troponin simply from the oxygen deprivation AND from directly toxicity of CO on myocardial cells)

Here is the next day ECG:

Essentially the same, with no evolution, proving that the previous ECG is his baseline normal variant ST Elevation

Sometimes Occlusion MI and CO poisoning are simultaneous!

The patient was discharged neurologically intact.


See this case of an awake patient without chest pain but with a CO level of 34%:

CO poisoning. Routine ECG recorded before hyperbaric therapy....Are they related?

With this ECG:


See this typical case of cO poisoning: 

What is the treatment for this subendocardial ischemia?


MY Comment, by KEN GRAUER, MD (2/17/2022):


Insightful case presented by Dr. Smith of this young adult who presented with accidental CO poisoning. I focus my comments on an often-overlooked technical point — and — on an important clinical point that is unfortunately not well appreciated by clinicians outside of those who regularly follow Dr. Smith’s ECG Blog.

The Often-Overlooked Technical Point:
As per Dr. Smith — in support of our impression that the initial ECG in today’s case represented a benign repolarization variant, was the lack of any significant change in overall ST-T wave appearance in the repeat ECG (done the next day) — compared to the initial ECG that was done in the ED ( = ECG #2 compared to ECG #1 in Figure-1). That said — there IS a difference between the 2 tracings!
  • The KEY to accurate comparison of serial tracings — is to be sure you are “comparing apples with apples — instead of comparing apples with oranges”
  • It is common to see changes in frontal plane axis and/or in chest lead R wave progression on serial tracings. This may be due to slight changes in the incline of the patient’s bed (ie, if on the initial tracing the patient was too dyspneic to lie flat) — or, to changes in chest lead electrode placement — or, simply due to the difficult-to-account-for slight variation in QRST morphology that sometimes occurs from one tracing to the next. These changes may produce differences in QRS and ST-T wave morphology from one tracing to the next not due to any change in the patient’s clinical condition.

CHECK OUT the 2 serial tracings in Figure-1:
  • Is the frontal plane Axis the same in both tracings?
  • Is R Wave Progression (and the area of Transition) the same?
  • IF either the Axis or R Wave Progression changed — How did this change affect QRS and ST-T wave appearance?


The frontal plane Axis is identical in ECG #1 and ECG #2 in Figure-1 (ie, approximately +70 degrees). Comparing these 2 tracings lead-by-lead — QRS morphology is virtually identical in each of the 6 limb leads. This tells us that lead-by-lead comparison of ST-T wave changes in the 6 limb leads is "comparing apples with apples", so that any differences in ST-T wave morphology in any of these 6 limb leads will be real! 

  • Since there has not been any change (other than minimal increase in T wave amplitude in lead aVL of ECG #2) — we can confidently say "No change".

In contrast — R Wave Progression is different in ECG #2, compared to what it was in ECG #1.

  • The area of Transition (where height of the R wave becomes taller than the S wave is deep) — had been between leads V3-to-V4 in ECG #1. Note that R wave amplitude equaled S wave depth already in lead V3 in ECG #1.
  • Transition occurs later in ECG #2 (ie, R wave height does not exceed S wave depth until leads V4-to-V5 — and the QRS was still predominantly negative in lead V4).
  • This means that serial comparison looking for ST-T wave changes in Figure-1 between these 2 tracings is not like "comparing apples with apples" — because transition is delayed in ECG #2, and QRS morphology is not the same.

  • Isn't the T wave in lead V1 of ECG #2 now negative — whereas it was positive in ECG #1?
  • Isn't the T wave in lead V2 of ECG #1 significantly taller than the T wave in lead V2 of ECG #2?
  • Isn't there more J-point ST elevation in leads V2 and V3 in ECG #1?

BOTTOM Line: When frontal plane Axis or R Wave Progression changes in serial tracings (as it so often does!) — We have to account for such changes with a "value judgment" as to whether we think such differences in ST-T wave morphology represent a "real" (ie, clinically significant) change — or whether such differences are unlikely to be clinically significant.

  • I completely agree with Dr. Smith's assessment — that in today's case, the differences we see in ST-T wave morphology between the 2 tracings is unlikely to be clinically significant (and is most likely the result of slight change in chest lead electrode placement).
  • Take-Home POINT: Remember to take into account frontal plane Axis and R Wave Progression when comparing serial tracings!

Figure-1: Comparison of the initial ECG in the ED ( = ECG #1) — with the follow-up ECG done the next day ( = ECG #2).

The Important Clinical Point:

Prior to working with Dr. Stephen Smith — the concept of Terminal QRS Distortion (T-QRS-D) was unknown to me. It’s a beautiful concept that on occasion may provide invaluable assistance for distinguishing between early repolarization vs acute OMI. The 3 ECGs shown above by Dr. Smith in today's case add insight to what is, and what isn’t T-QRS-D.
  • T-QRS-— is defined as the absence of both J-wave and an S-wave in either lead V2 or lead V3 (and sometimes lead V4). Although simple to define — it’s taken me a bit of practice to become comfortable and confident in its recognition.

There are an increasing number of other examples of what T-QRS-D is, and what it is not on Dr. Smith’s Blog. For easy reference — Consider these 2 links:

Figure-2: Comparison between ST elevation in lead V3 due to a repolarization variant (TOP — from the 4/27/2019 post) — vs acute OMI (BOTTOM — from the 9/20/2015 post), which manifests T-QRS-D (See text).

What About the ECGs shown in Today's Case?
To illustrate the difference between what T-QRS-D is and is not — I've put the 1st and 3rd tracings shown in today's case together in Figure-3.
  • ECG #1 ( = TOP tracing in Figure-3): As per Dr. Smith — the ST-T wave changes in this tracing strongly suggest a repolarization variant. In addition to prominent J-point notching in multiple leads (RED arrows in Figure-3) — there is normal R wave progression with ample QRS amplitude — no reciprocal ST depression — similar ST-T wave morphology in multiple leads (ie, lack of the localizing changes that are typically seen with acute infarction) — and, terminal QRS "slurring" (PINK arrows), that convey similar clinical implications as the prominent J-point notching. Lack of evolution on the repeat tracing done the next day (ECG #2) supported the conclusion that ECG #1 represents a repolarization variant, and is not indicative of an acute cardiac event.

  • ECG #3 ( = BOTTOM tracing in Figure-3) — was added to today's case by Dr. Smith to provide an example in which acute CO poisoning was accompanied by an obvious STEMI (The case discussed in detail in the December 9, 2019 post in Dr. Smith's ECG Blog). Note the BLUE arrows — which highlight obvious T-QRS-D, seen here in the 3 leads with the most ST elevation.

Take-Home POINT: Clearly — the diagnosis of acute LAD occlusion is readily evident in ECG #3 without need to identify T-QRS-D. But there will be occasional cases in which you may be uncertain if there is (or is not) an acute OMI — in which case, recognizing T-QRS-D may instantly confirm your suspicion of the need for prompt cath.

Figure-3: Comparison of ECG #1 and ECG #3 from today's case to illustrate what is — and what is not T-QRS-D (See text).

No comments:

Post a Comment

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