Friday, September 27, 2019

Chest pain with NonDiagnostic ECG but Diagnostic CT Scan

An elderly woman presented with chest pain that radiated to the back for several hours.

Here is here initial ECG:
There is only a nonspecific flat T-wave in aVL.  It is essentially normal.
The first troponin returned at 0.099 ng/mL (elevated, consistent with Non-Occlusion MI)

Providers were concerned with aortic dissection, so they order a chest aorta CT.

This showed no dissection but did show the following:
Notice the area of the lateral wall (lower right) that has no contrast enhancement 
(It is dark, where areas of enhancement are light-colored). 
This transmural ischemia, but not necessarily completed infarction (yet).
 A slice at a slightly different level:
Again, an area with absence of contrast enhancement (dark-, not light-colored).
These were read by our fantastic chief of radiology, Gopal Punjabi, who has his own blog on Spectral CT: https://www.ctspectral.com/    https://twitter.com/CtSpectral

Here is the image using Spectral CT:
It is much more obvious with this technique!

See more images of this case at Gopal's Spectral CT Blog:

It's all about confidence


With continued symptoms, an elevated troponin, and no other explanation, this is acute MI with ongoing ischemia until proven otherwise.

The CT scan confirms no other explanation and also confirms that this is acute Transmural ischemia diagnostic of OMI (Occlusion MI).

If the patient were no longer symptomatic, one could conclude that the infarct is completed, and emergent angiogram +/- PCI would not be necessary.

The cardiologists did not want to go to the cath lab.

She had subsequent ECGs:

First at 1.5 hours:
This possibly shows some inferior STD, but probably it is baseline wander.


This one at 5.5 hours:
No significant change


She was admitted on a Nitro drip.

Her troponins went like this:
0.099 ng/mL
1.250
3.712
6.073
8.092
12.170
15.680
21.051
23.159 (this is not a small MI)

Next day echo:


The estimated left ventricular ejection fraction is 60%.
Regional wall motion abnormality-inferolateral

Angiogram:

Culprit for NSTEMI is thrombotic occlusion of small-medium caliber OM1.  It was stented.




===================================
MY Comment by KEN GRAUER, MD (9/27/2019):
===================================
As suggested by the title of this Blog post — confirmation of the diagnosis in this case was made not by ECG — but instead by chest aorta CT! That said — there are some Observations that should be made about the first 2 ECGs in this case.
  • For clarity — I’ve put these first 2 tracings together in Figure-1.

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



OBSERVATION #1  This case describes a high-prevalence situation — because the patient is an elderly woman with new-onset chest pain of several hours’ duration. Once dissecting aneurysm was ruled out by chest aorta CT — the possibility of acute ischemic heart disease becomes paramount.
  • Given this History — our scrutiny for detecting ECG abnormalities is heightened. The onus falls on us to prove questionable ECG findings are not acute — rather than the other way around.


OBSERVATION #2  While not diagnostic — the initial ECG in the ED ( = ECG #1) is not normal (Figure-1). Descriptive Analysis of ECG #1 reveals that the rhythm is sinus @ 80-85/minute — all intervals are normal — small septal q waves are seen in leads V4-6 — and transition (where the R wave becomes taller than the S wave is deep) occurs normally between leads V2-to-V3. Findings to note on ECG #1 include the following:
  • There is some baseline artifact. Given that artifactual baseline undulations are largest in leads II, III and aVF — the left leg is most likely the source of this artifact. This artifact was no longer seen in the next 2 tracings. NOTE While the amount of artifact in ECG #1 is small (ie, not enough to prevent accurate interpretation of this tracing) — knowing how to quickly determine which extremity is the cause of the artifact (due to tremor, a faulty lead connection/inadequate skin contact, etc) is helpful, because this may expedite correction, which may be important when artifact does impede interpretation. (BONUS  The 3 page article by Rowlands and Moore that I’ve posted in Figures-23 and 4 below is the BEST description I’ve seen of how to quickly determine which extremity is the source of the Artifact).
  • Returning to ECG #1 — There is probably some type of misplacement of lead V2 (perhaps due to body habitus of this elderly woman — perhaps due to placement of the V2 electrode too high on the chest). It just doesn’t make good physiologic sense for there to be an rSr’ only in lead V2, and not in V1. That said, given nearly identical QRS morphology in all leads for the 3 tracings shown in this case — this technical point does not affect assessment or management in this case.
  • The T waves in leads V2 and V3 in ECG #1 look disproportionately tall and peaked compared to R wave amplitude in these respective leads. One reason this abnormal T wave morphology stood out to me — is that the ST segments prior to onset of T wave ascent are relatively flat. Whether this T wave appearance in leads V2 and V3 of ECG #1 is a benign longstanding finding — or — whether it might represent either hyperacute T waves and/or reperfusion changes in this elderly woman with new chest pain of several hours’ duration is (in my opinionnot known from assessment of this single ECG.
  • There clearly is voltage for LVH in ECG #1 (ie, R wave amplitude in lead V5 ~25mm — CLICK HERE for Review of Criteria for ECG Diagnosis of LVH). The next bullet tells why recognition of voltage for LVH is relevant in this case.
  • What does this patient’s baseline ECG show? Was there LVH with an ST-T wave repolarization abnormality (ie, with LV “strain” ?) on the baseline ECG? IF so — then perhaps the ST segment flattening in V5,V6 with relatively small T wave amplitude in lead V6 represents the combination of baseline LV “strain” (ie, ST-T wave depression+ hyperacute T waves that result in an “intermediate pattern” that might not look acute in the lateral chest leads of ECG #1.
BOTTOM LINE: In my opinion, at the point that we saw ECG #1 in this case about this elderly woman with new chest pain of several hours’ duration — it would be impossible to rule out acute changes on the basis of this single initial ECG.
  • Finding a prior ECG on this patient could be invaluable! Otherwise — more information will be needed (ie, serial ECGs, troponin, Echo, etc.) before we'll be able to tell if the findings I describe above might be acute ...


OBSERVATION #3  Compared to ECG #1 — the 2nd ECG (obtained 1.5 hours later) in this case ( = ECG #2) — does show a change that should be noted and which may be relevant (Figure-1).
  • Isn’t T wave amplitude in both leads V2 and V3 of ECG #2 proportionately taller than it was in ECG #1, compared to R wave amplitude in these respective leads? That is — the T wave in lead V2 of ECG #2 measures 5 mm tall, whereas it was only 3 mm tall in ECG #1. In lead V3 — the T wave in ECG #2 is clearly taller than the R wave in this lead — whereas in ECG #1 — the T wave in lead V3 was clearly not as tall as the R wave in this lead. BOTTOM LINE: From comparison of ECG #1 with ECG #2 — I do not know how to rule out the possibility that T waves in leads V2 and V3 have become more hyperacute in ECG #2.


CONCLUSION of this Case: The 3rd ECG (done at 5.5 hours) did not show significant change compared to ECG #2 — so at this point, it had become apparent that this patient’s serial ECGs were not acutely evolving. And, as noted in detail by Dr. Smith — serial troponins had become markedly positive — the chest aorta CT showed regional contrast enhancement abnormality — and, Echo the next day confirmed regional wall abnormality. Therefore — the ECG findings I point out above were not needed in this case for diagnosis. That said, in the interest of optimizing ECG interpretation ability:
  • Please take another look at the tracings in Figure-1 if you did not initially note the findings I described above. While it turned out in this case that serial ECGs did not evolve — this could not have been known from assessment of only those first 2 ECGs!
  • Keep in mind that finding a prior ECG on your patient can sometimes prove invaluable! I wish a prior ECG on this patient was available! IF a prior ECG had been available — it is possible that recognition of an acute cardiac event could have been accomplished much earlier (ie, as early as the moment ECG #1 was recorded)!
  • As should be apparent from the unique clinical correlations routinely provided on Dr. Smith’s ECG Blog — the BEST way to continue honing your ECG interpretation skills — is to GO BACK and review tracings in each case after the definitive diagnosis is revealed. In this particular case, the most important teaching point to me — is to appreciate that we can not be certain the ECG findings I describe above for ECG #1 are not acute on the basis of this single initial ECG. It was not until time had passed and the 3rd ECG recorded at 5.5 hours was obtained, that lack of serial ECG evolution became evident ...

Our THANKS to Dr. Smith for this illustrative case!



BONUS  The 3 Figures below reproduce the article by Rowlands and Moore for determining the source of artifact when a single limb lead is responsible. 
  • Knowing which extremity is at fault is helpful for expediting correction (ie, from a faulty lead connection/inadequate skin contact, etc.).



Figure-2: Page 475 from Rowlands — Determining the extremity responsible for Artifact (See text).




Figure-3: Page 476 from Rowlands — Determining the extremity responsible for Artifact (See text).




Figure-4: Page 477 from Rowlands — Determining the extremity responsible for Artifact (See text).


4 comments:

  1. Are the T waves not considered hyper acute in V2-4?

    ReplyDelete
    Replies
    1. @ Chad B — You must have submitted your excellent Question BEFORE I wrote My Comment for this post (above). In my opinion (as I detail above) — YES, the T waves especially in V2 and V3 of ECG #1 (possibly also in V4, though I intentionally chose not to mention that) ARE hyperacute. PLEASE take a look at My Comment above — and let me know after reading it if you still have this question. (NOTE: Since I never know when Steve or Pendell will be publishing a case — there OFTEN is a slight lag time until I am able to write My Comment ... — :)

      Delete
  2. I always just look at I, II, and III and whichever one looks good, the limb (RA, LA or LL) not included in that lead is the problem electrode.

    ReplyDelete
  3. @ Jody — YES, your system works. Lead I in ECG #1 (of Figure-1) looks good — and the left Foot electrode is not included in derivation of lead I. Then the NEXT thing I do, is to look at the “a” ( = augmented) leads — and you’ll note that the greatest amplitude of artifact in the 3 “a” leads in ECG #1 is aVF — which CONFIRMS the left Foot as the “culprit” extremity. As in the short article by Rowlands that I’ve reproduced in Figs 2,3,4 — formulas are given, which further confirm the left Foot — because ~1/2 the amplitude of the artifact is seen in the other 2 “a” leads. THANKS again for your comment — :)

    ReplyDelete

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