Friday, February 4, 2022

This ECG was interpreted as completely NORMAL by the computer: What about it is THE critical finding??

A 40 something otherwise healthy man presented with substernal chest pain.  It had occurred once 3 days prior and resolved without any medical visit.

He had a triage ECG at time zero:

Here is the computer interpretation (Normal)
This was the Veritas algorithm.

What do you think?

This ECG is DIAGNOSTIC of acute LAD Occlusion.  The T-waves are hyperacute, but most important, the minimal ST Elevation is accompanied by Terminal QRS Distortion, or at least nearly so.

The ST Elevation in V2 and V3 is NEVER normal if it is accompanied by terminal QRS distortion, as defined by absence of S-wave and J-wave in EITHER V2 or V3.  There is absence of both in V3 here (although one of the complexes in V3 has a tiny S-wave).

See 12 cases of Terminal QRS distortion here.

See: Best Explanation of Terminal QRS Distortion in Diagnosis of Electro-cardiographically Subtle LAD Occlusion

This ECG was seen by one of my partners who was very astute and recognized that something was wrong.  He came to me with the ECG and I said, "Yup, this is an LAD Occlusion."  

I sent it to the Queen of Hearts later.  She gets it again.

So we immediately activated the cath lab.

37 minutes later, the patient's pain spontaneously resolved and a 2nd ECG was recorded: 

Now the artery has opened up and the S-waves are re-constituted!!
But even this ECG by itself is quite suspicious.
The T-waves remain hyperacute, but as long as the artery stays open, they will be "on the way down" to normal.

This was also called Normal by the Veritas algorithm

Let's see what the Queen has to say:

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:


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.

Unbeknownst to us at the time, there was an old ECG for comparison from 3.5 years prior which I only found a day later:
This is a truly normal ECG, with normal sized T-waves and normal S-waves in V2 and V3.

The patient went to angiogram and was found to have two 99% LAD thromboses with TIMI-3 flow.  They were stented.

The peak troponin was 1863 ng/L.  Why not very very high?  Because it reperfused on its own and because we intervened before it could re-occlude.

Here is the post intervention ECG:

Reperfusion T-waves identical to Wellens' syndrome, because it is the same pathophysiology.

Formal Echocardiogram:

The estimated left ventricular ejection fraction is 58 %.

Left ventricular hypertrophy concentric.

Regional wall motion abnormality-distal septum anterior and apex

Today, shortly after this case presented, I received in my inbasket from Journal Feed this article which promotes the idea that "normal" ECGs by computer do not need to be overread by a physician.  

Emergent cardiac outcomes in patients with normal electrocardiograms in the emergency department. Am J Emerg Med. 2022 Jan;51:384-387. doi: 10.1016/j.ajem.2021.11.023. Epub 2021 Nov 17.

This study looked at less than 1000 cases, which is not nearly enough (see below for analysis) and they used cardiologists as the gold standard (a very poor gold standard), NOT presence or absence of Occlusion MI (which we have done in all of our ECG studies, and must be ascertained by 1) TIMI 0/2 flow on angiogram or 2) culprit + TIMI 3 flow and very high troponin. 

So this study is worthless and must be ignored.  And the Clay Smith at Journal Feed comment was right on.

In our case, the emergency physician was well trained in identifying subtle LAD Occlusion MI and ignored the computer.

I have here 30 cases of "Computer Normal" ECGs which were critically abnormal and the vast majority are missed acute coronary occlusions (Missed Acute OMI) and most were recognized by the physician.

We wrote this Editorial in the Journal of Electrocardiology in 2019.  Litell JM, Meyers HP, Smith SW. Emergency physicians should be shown all triage ECGs, even those with a computer interpretation of “Normal.” J Electrocardiol [Internet] 2019;54:79–81. Available from:


"To illustrate the limitations imposed by sample size, recent data from our institution reveal that we identify approximately 225 type I myocardial infarctions (MI) in a typical year. These include about 60 occlusion MI (OMI) with clear ST segment elevation (none of which would be called “Normal” by the computer) and about 165 Non-STEMI. Of the Non-STEMI in our cohort, about 25% will actually have acute coronary occlusion.  While most of these roughly 40 NSTEMI occlusions would be read by the computer algorithm as abnormal in some way (typically nonspecific ST segment or T wave abnormality), they would not be labelled STEMI.  We might conservatively estimate that 5 of these 40 acute OMIs without diagnostic ST segment elevation would be erroneously read by the computer as “Normal.” That is five OMIs per year misread by the computer algorithm as normal. In that same year we collect approximately 24,000 ECGs in our ED, of which 20% are called “Normal” by the computer. Taken together, these data suggest that out of 5000 “Normal” ECGs in a given year, about 5 (0.1%) will actually be acute OMIs that have been misinterpreted by a computer algorithm as completely normal. A sample size of 855 has no chance of generating a meaningful conclusion about the reliability of computer “Normal”.   In fairness, it is not certain that an average emergency physician will catch these few false normals, but they will absolutely go missed if the physician never sees them.We would prefer to be interrupted."

In this case, an emergency physician who is not particularly focused on ECG interpretation as I am saw that this "normal ECG" was an acute OMI.


MY Comment by KEN GRAUER, MD (2/4/2022):


I wanted to comment on the initial ECG in today's case — which for clarity, I have reproduced in Figure-1. This tracing was interpreted as "normal" by the computerized report. As noted by Dr. Smith — the Am J Emerg Med article he cites above floats the notion that ECGs interpreted as "normal" by the computer "do not need to be overread by a physician".

completely agree with Dr. Smith. The methodology of this Am Emerg Med study is flawed — and results of this study must be ignored.

  • It is hard to believe that the computer interpreted the ECG in Figure-1 as "normal". To amplify remarks by Dr. Smith above — considering the clinical context (ie, new-onset substernal chest pain) — no less than 9/12 leads in ECG #1 are clearly abnormal.
  • There is frank ST coving with elevation and T wave inversion in lead V1, which is never a normal finding.
  • As per Dr. Smith — the ST-T wave changes with T-QRS-D (Terminal QRS Distortion) in leads V2, V3 is diagnostic.
  • Hyperacute T waves are seen in leads I, II, aVF; and V2-thru-V6. These T waves are "hypervoluminous" with respect to QRS amplitude of the complexes in these respective leads. Although these changes might seem subtle — they are not subtle to anyone experienced in emergency care. Each of these T waves is taller and fatter-than-expected, as well as being wider-at-their-base in the context of QRS amplitude in each respective lead. 
  • As a specific example — one needs to keep in mind the tiny amplitude of the QRS in lead aVF — for which the T wave in this lead literally towers over the QRS.

Figure-1: The initial ECG in today's case (See text).

I like the computerized ECG report. This is because I know how best to use the computerized interpretation (CLICK HERE — for my approach for Using the Computer). I realize that many clinicians do not like computerized interpretations. It's a matter of personal preference and clinical experience with computerized interpretations. In my opinion — the following are the BASICS for optimal Computer Use:

  • Never look at the computerized interpretation until after you have made your own independent assessment! 
  • In general — the computer is excellent for calculating rates, axis, and intervals. 
  • In my experience — the computer is terrible for assessing any rhythm other than sinus!
  • Overall — the computer does OK for assessing normal tracings. That said — the CAVEAT is that the computer may overlook certain findings (such as ST elevation and hyperacute T waves — as was highlighted by today's case in which the computer called the ECG in Figure-1 a "normal" tracing).

BOTTOM LINE: The computerized report may help by suggesting certain findings that you may not have thought of — BUT — the clinician must always overread what the computer says. IF you disagree with what the computer says — then cross out the computer report!

  • In an emergency setting — clinicians must promptly interpret all ECGs from any acute patient themselves. Only after doing so — should they dare to look at what the computer said — but regardless of what the computer says, if YOU think the ECG suggests an acute process — YOUR opinion is the one that needs to be followed.

ADDENDUM (2/8/2022): 

Dr. Mario Parrinello, an esteemed cardiology colleague of ours from Cremona, Italy — wrote the following comment regarding today’s post on the EKG Club. I (KG) thought his comment worthy of adding as an Addendum. 

Dr. Parrinello wrote: Great case! I entirely agree with the interpretation, and as a result with the excellent management that has saved myocardium. ECG interpretation is a critical point in our daily practice and we wisely and always should not trust the computer interpretation, as Drs. Stephen Smith and Ken Grauer have excellently emphasized.

  • That said — Let me “stick up for“ the computer interpretation. While I am not an expert on computer software programmed for ECG interpretation — as Ken Grauer has pointed out, computer interpretation is very useful for calculation of parameters (rate, intervals, axis, etc.).
  • Computer interpretation is not useful for other important and vital ECG signs, such as terminal QRS distortion (T-QRS-D). But why? I suspect the reason is a result of what the “programmer” has put into the software. Interval calculations are relatively simple to implement — but T-QRS-D is not (this requires expert knowledge on clinical trials and literature...). 
  • What is surprising (from a negative standpoint) — is that computer interpretation in this case failed to recognize hyperacute T waves, that in turn would have concerned a less experienced and astute interpreter.
Therefore — The moral is, as always — Never trust the computer interpretation

  • To paraphrase a very well known statement (ie, “Time is muscle”) — I would suggest that "Correct ECG interpretation is muscle!" 

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