Sunday, October 15, 2023

This ECG was handed over at triage. See how the Queen of Hearts AI does with these explainability images.

Submitted by anonymous, written by Pendell Meyers


I was at triage when this ECG was presented to me.

What do you think?





Here's what the Queen of Hearts thinks:

Note that the Queen has "HIGH" confidence that this is OMI. 



As noticed by the AI, if the patient presented with chest pain, then even this subtle ECG is diagnostic of OMI.  Specifically, this ECG is diagnostic of acute LAD OMI, with hyperacute T waves likely in V3 and V4, very small but abnormal STE in V2-V4 (because of its there is a lot of STE relative to the very small QRS), and there is also some terminal T wave inversion in V4 which may suggest some element of reperfusion.

Here is her baseline below for comparison (which is totally normal, and you can see that Queen of Hearts says its totally normal also).

Normal baseline ECG on file.





Here are the precordial leads side by side (left old, right new). If you can see this diagnostic difference, then use the Queen of Hearts side by side explainability images to help you see the differences! (just to be clear: the AI has no ability to compare current ECGs to prior in her current version, her diagnoses are only based on the current ECG you input)






I read her triage ECG as very concerning for OMI without any clinical context and without the prior ECG, then went to find the patient and confirmed a clear ACS history:

This woman in her 60s presented with acute chest pain off and on for the past week. Prior episodes had simply resolved after brief symptoms, but the current episode had lasted for 2 hours without improvement, so she presented to the ED. Her symptoms were described as "pressure" with left arm radiation. No diaphoresis or vomiting. 

At triage she stated her pain is still persistent, but it is mildly improved compared to when she decided to come to the ED. 

Here is the triage ECG:


I got a repeat ECG during my history:





I activated the cath lab.

The cardiologists decided to overrule the STEMI criteria and the cath lab activation was NOT cancelled, and she was taken immediately to the cath lab and found to have an acute thrombotic distal LAD occlusion (TIMI 0). There was some faint collateral noted from the distal RCA, which likely explains the terminal T wave inversion (reperfusion finding).

Pre cath. Distal LAD occlusion
(Just to the right of middle in the image, closest to the bottom, there is a cutoff).

Post PCI to distal LAD.




Her initial hs trop I (collected before PCI) returned at 2,507 ng/L.

The troponin rose quickly to 12,828 ng/L, but peak troponin was not measured.

Formal echo showed apical hypokinesis with preserved EF.

Here are her ECGs after PCI, showing anterolateral reperfusion findings.






Learning Points:

1. The Queen of Hearts PM Cardio AI app is extremely sensitive to subtle ECG findings of OMI

2. All ST Elevation and T-wave size must be assessed relative to the QRS size.

3. In many of your institutions, ST elevations are the only way the cath lab may be activated. Here is a terrific example of how that finding could be missed due to its subtle changes on her ECG.  The AI Bot, however, didn’t miss it, and demonstrates how critical this technology is for high quality care.


The Queen of Hearts PM Cardio App is now available in the European Union (CE approved) the App Store and on Google Play.  For Americans, you need to wait for the FDA.  But in the meantime:

YOU HAVE THE OPPORTUNITY TO GET EARLY ACCESS TO THE PM Cardio AI BOT!!  (THE PM CARDIO OMI AI APP)

If you want this bot to help you make the early diagnosis of OMI and save your patient and his/her myocardium, you can sign up to get an early beta version of the bot here.  It is not yet available, but this is your way to get on the list.




===================================
MY Comment, by KEN GRAUER, MD (10/16/2023):
===================================
Today’s case by Dr. Meyers is subtle. It is very subtle. I think we sometimes learn the most from subtle cases. The process we use is that we find out what “the answer” is (ie, in this case — cath proven distal LAD occlusion) — and then we go back to the tracing to reexamine the subtle clues that may have been missed.
  • I thought the best way to demonstrate this was through review of the initial ECG — and then to compare this first tracing with the repeat ECG done a little while later (Figure-1).


Why was Today’s Case so Subtle?
There are several reasons why I found the initial tracing in today’s case so very subtle. These include:
  • The very low voltage on this ECG — that made it more difficult to appreciate the subtle acute ECG findings.  
  • In addition to the small size of the ECG complexes on this tracing — I did not think the ECG features of an acute event were at all obvious. This was because i) There are no Q waves in ECG #1ii) There is no reciprocal ST depression; iii) Potentially acute ECG changes are limited to only a few leads; andiv) ST-T waves are relatively “flat” in multiple leads — making it difficult to appreciate meaningful ST segment deviations.


Taking a Closer LOOK:
  • We have periodically reviewed cases with low voltage in Dr. Smith’s ECG Blog (Please see My Comment at the bottom of the page in the November 12, 2020 post, among others). Among the causes of low voltage that I note in this Nov. 12, 2020 post — are 2 causes potentially relevant to my initial considerations in today’s patient, who was found to have acute LAD OMI. These 2 causes are: i) Low voltage from myocardial “stunning” — resulting from low cardiac output that may follow a large acute infarction; andii) Takotsubo cardiomyopathy — which can at times present with similar ECG findings of acute MI.
  • As today’s case evolved — Dr. Meyers was able to rule out myocardial stunning and Takotsubo cardiomyopathy by results from Echo, cardiac cath — and from finding a prior tracing on this patient that showed similar diffuse low voltage (making it more likely that something related to body habitus may provide the answer for this patient’s persistent low voltage).

  • NOTE: The KEY clue for me that today’s initial tracing might indicate LAD OMI — is the subtle-but-real terminal T wave negativity in lead V4 (RED arrows in lead V4 of Figure-1). In view of the history (Dr. Meyers tells us that the patient’s CP had decreased compared to when she decided to come to the ED) — I thought the shallow-but-real terminal T wave negativity in lead V4 might represent the subtle reperfusion wave of Wellens’ Syndrome. After all, the ST segment in this tiny V4 lead is slightly elevated (probably disproportionately — considering how tiny the R wave in V4 is in ECG #1)There normally should not be any terminal T wave negativity in this lead.
  • Even more subtle is the hint of terminal T wave negativity that we see in leads V2 and V3 of ECG #1 (BLUE arrows in Figure-1).

  • Definition: Technically, for the tracing in ECG #1 to truly “qualify” as Wellens’ Syndrome — there should be no CP at the time the ECG is recorded. CP was still present in today’s case — although this patient’s CP was decreasing — so criteria for “true Wellens” would be satisfied if CP was to completely resolve as reperfusion T waves deepened in a follow-up tracing (See ECG #2 below).
  • NOTE: For more on “My Take” regarding a historical perspective, including current clinical relevance of recognizing Wellens' Syndrome — See My Comment at the bottom of the page of the August 12, 2022 post.


How to Check Your Findings ...
To Emphasize — I would not be at all sure from the subtle findings in ECG #1 that there was ongoing LAD OMI. This is where follow-up tracings can be so helpful. Take another LOOK at Figure-1 — Compare the repeat ECG with the initial tracing.
  • Other than loss of the tiny R wave and some T wave flattening in lead aVF — there is essentially no difference in the limb leads between the 2 tracings in Figure-1.

  • However, in the chest leads — there has definitely been evolution, in that ST segments are more coved and slightly more elevated in virtually all chest leads.

  • KEY Point: Did you notice that the terminal T wave inversion is now deeperand much more readily apparent in leads V2-thru-V5? Doesn't seeing the evolved deeper terminal T wave negativity in ECG #2 help when going back to ECG #1 — in terms of facilitating recognition of the subtle beginning stage of these Wellens' T waves highlighted in Figure-1 by the RED and BLUE arrows?


Closing Thought: Lots to learn from going back over subtle findings in the initial ECG — and then realizing that they were there the whole time!  




Figure-1: Comparison between the initial ECG in today's case — with the repeat ECG done a little while later.





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.