Normal ECG by many measures. Is it normal?

This was submitted by Hans Helseth, who is applying to med school and who is working as an EKG tech right now.  

Smith: This patient presented years ago, but came in again while Hans was working.  He looked back in time in the patient's chart and saw these ECGs and immediately recognized that they manifested subtle OMI.

Also: See Ken Grauer's excellent comments at the bottom.

Case

A 68 year old man with a medical history of hypertension, hyperlipidemia, and CAD with stent deployment in the RCA presented to the emergency department with chest pain. He developed it only 20 minutes prior to presentation while cutting branches outside. He had an EKG recorded right away.

EKG 1, 1646:

What do you think?

Smith: this is suspicious for a very subtle inferior OMI.  There is minimal STE in inferior leads, but also with terminal QRS distortion and some reciprocal ST depression in aVL.  Terminal QRS distortion is always abnormal in leads V2 and V3, but we have also found it to be useful in other leads, though not quite as specific.  Although normal variant STE can have reciprocal STD in aVL

I want to mention that Hans saw this immediately.  Pretty impressive for someone who has not yet attended med school, or even been a nurse or paramedic yet.

Back to the case

The conventional computer algorithm (Marquette 12 SL) diagnosed “Normal Sinus Rhythm, Normal ECG”. This interpretation was confirmed by the overreading physician. The EKG is not normal, however. There is ST segment straightening in the inferior leads.  There is baseline wander in aVL which makes it difficult to assess for reciprocal changes, but behind the artifact, the first QRS complex in aVL shows probable ST depression. In a patient with new chest pain and multiple cardiac risk factors this EKG is very suspicious for inferior OMI, although it is extremely subtle and should be repeated to eliminate the artifact in aVL.

Queen of Hearts for EKG 1: Not OMI with high confidence

The first 4th generation troponin T (URL 0.05 ng/mL) was drawn at 1650 and resulted below the limit of detection. The patient was given nitroglycerin which partially alleviated the pain. He was worked up non-emergently in the ED with pain recurring and resolving multiple times during his stay. Another EKG was eventually taken. It is unclear if the patient had pain or not at this time.

EKG 2, 2122:

The computer algorithm called this EKG normal, and this interpretation was again confirmed by the overreading physician who noted no changes from EKG 1. Since EKG 1, however, the T waves in the inferior leads have become concave. If the patient’s artery was indeed occluded at the time of EKG 1, then his artery was most likely open at this time. This change further supports the case for an OMI diagnosis.

Another 12-lead was taken 44 minutes after EKG 2.

EKG 3, 2206:

While extremely subtle, the ST depression in aVL has worsened slightly since EKG 2. This EKG was again called normal by both the computer and the overreading physician, although in the context of the previous two tracings, this finding suggests re-occlusion of the culprit artery. 

Queen of Hearts for EKG 3: Not OMI with high confidence

See this post, in which the first sign of inferior OMI is ST depression isolated to leads I and aVL: (https://hqmeded-ecg.blogspot.com/2019/04/a-40-yo-with-chest-pain-only-ecg.html)

And this post, in which T wave inversion isolated to leads V2 and aVL are the only signs of inferior OMI: (https://hqmeded-ecg.blogspot.com/2020/08/computer-and-transferring-physician-say.html)

Despite an undetectable troponin and three “normal” EKGs, the nature of the patient’s symptoms and his positive cardiac history warranted concern for ACS. As such, the patient was placed on a heparin drip and transferred by ambulance to a cardiac cath-capable facility. 

En route to the next hospital, the paramedics recorded another 12-lead tracing.

EKG 4, 2327:

In V2-V5, the T waves have begun to flatten and ST depression has begun to develop. The T waves in the inferior leads are becoming convex again. The computer algorithm (Glasgow) called this normal, but it is highly suggestive of inferior and posterior OMI. This suggests continued occlusion of the culprit artery.

On arrival to the cath-capable facility, another EKG was recorded.

EKG 5, 0028

This EKG in a patient with new chest pain and multiple risk factors, especially after examination of the previous 4 EKGs, is diagnostic of inferior and posterior OMI. The T waves in the inferior leads have once again lost their concavity. The ST depression in aVL is horizontal and the T wave is subtly terminally upright. There is slight ST depression in V3. The Marquette 12 SL algorithm called this EKG normal, and as with all previous EKGs, this interpretation was confirmed by the overreading physician. 

Queen of Hearts for EKG 5: Not OMI with mid confidence; this is to say that she does NOT think this EKG is normal.

At 0120, a repeat troponin T was drawn: 0.13 ng/mL (this is elevated, and thus in this context is now diagnostic of acute MI)

At 0330, about 11 hours after presentation, the patient was taken to the cath lab:

There is a 99% occlusion of the proximal segment of the RCA with TIMI 2 flow. In this screenshot, a small amount of contrast can be seen opacifying the vessel distal to the lesion.

After stent placement:

The vessel is now open with TIMI 3 flow, although it is diffusely diseased and the middle segment is ectatic.

Left ventriculography demonstrated inferobasal hypokinesis and an EF of 45%. The troponin peaked at 0.4 ng/mL, which is relatively low for OMI. The patient was lucky to have periodic spontaneous reperfusion. 

A post-cath EKG was recorded at 0719:

The computer interpretation read “Sinus bradycardia, otherwise normal ECG”. 

The overreading physician confirmed this interpretation, adding that in comparison to EKG 5, “rate has decreased”. 

While ST depression and T wave inversion persists in aVL, the T waves in inferior leads have become concave and slightly smaller in size. The T waves in the anterior leads are once again upright. These findings support inferior and posterior reperfusion.

Every 12-lead recorded on this patient during his presentation was called completely normal by all conventional computer algorithms and attending physicians involved in this man’s care. Only the Queen of Hearts alerted any concern about an EKG!

Three years later, the patient presented again to the ED with chest pain. Here is his EKG for this presentation:

It is now truly normal! He was found not to have OMI during this presentation.

The progression of leads III and aVL can be appreciated below:

Smith comment: this case is well managed.  There was intermittent pain and eventually a diagnostic ECG.  As soon as the troponin was diagnostic, with persistent pain, cath lab is indicated and the patient was taken.

===================================

MY Comment, by KEN GRAUER, MD (1/15/2024): 

===================================

I love this post by Hans Helseth and Dr. Smith — because it raises the important issue regarding how we define an ECG as "normal"? As my answer to this challenging question — I will paraphrase the well known saying, "Beauty is in the eye of the beholder". My answer is as follows:

• The definition of a "normal" ECG — is dependent on the clinical circumstances under which the tracing is recorded and dependent on the person who is interpreting the tracing.

As emphasized by Hans Helseth in his stimulating discussion — despite the first 6 ECGs in today's case being passed off as "normal" by both conventional computer interpretation and by each of the overreading physicians — none of these ECG were truly "normal" by our interpretation!

• My Experience: As one who has studied and taught the "art" of clinical ECG interpretation over decades — I have long followed and contemplated with interest and curiosity the designation of "normal tracing" bestowed on the ECGs of our patients by excellent, experienced cardiologists. To paraphrase another well known saying, "The devil is in the details". The reality (as it pertains to the interpretation of ECGs) is all in the details — and those crucial details are all-too-often ignored, even by clinicians who should know better.
• To Emphasize: The designation, "normal" ECG — is not the same as saying there are "nonspecific ST-T wave abnormalities that do not appear to be acute". And, what then needs to accompany this recognition of nonspecific ST-T wave changes — is clinical correlation to the circumstances of the case at hand.
• For example — an identical ECG that is truly "normal" in a healthy, asymptomatic young adult (who needed an ECG as part of a routine employment physical) — might in an older adult with new-onset cardiac-sounding chest pain, appropriately raise suspicion of a possible early OMI because of T waves that are seemingly larger-than-expected considering amplitude of the QRS complex in those leads being looked at (whereas identical-looking T waves in that healthy young adult would clearly by default represent a repolarization variant).

Applying these principles to today's case — The failure by multiple overreading physicians to appreciate the subtle ECG findings highlighted above by Hans Helseth indicates 3 important deficits in the expertise of these overreading physicians: 

• i) Failure to correlate ECG findings to the clinical situation (ie, When was each ECG recorded with respect to the presence, severity and duration of symptoms?). 
• ii) Lack of awareness and appreciation for the concept of "pseudo-normalization" (in which "on the way" from the ST elevation of acute coronary occlusion — to the evolutionary development of reperfusion ST-T wave changes — the ECG may look almost "normal" for a short period of time) — and, 
• iii) An obvious failure to place serial ECGs side-by-side when comparing serial tracings for potential evolution of ST-T wave findings (a suboptimal practice that we have shown in numerous posts will result in failure to recognize subtle-but-real "dynamic" ST-T wave changes).

=======================================

For clarity in Figure-1 — I've reproduced and placed together the first 2 ECGs in today's case. 

• Take another LOOK at these 2 tracings in Figure-1. Why are we saying that neither of these tracings is totally "normal"?
• And why are we saying that some evolution has occurred?

Figure-1: I've labeled the initial ECG and the 1st repeat ECG in today's case.

Why is Today's Initial ECG not "Normal"?

I interpreted today's initial ECG as showing sinus rhythm — normal intervals (PR/QRS/QTc) — normal axis — and no chamber enlargement. 

Regarding Q-R-S-T Changes in Figure-1:

• The tiny Q wave in lead III is narrow and not clinically significant.
• R wave progression is appropriate, albeit a little abrupt (rapid transition to a predominant R wave occurs between leads V2-to-V3).

Regarding ST-T wave Changes:

• My "eye" was immediately drawn to the depressed ST segment in lead aVL (RED arrow within this lead in Figure-1). But why does the ST segment in the 2nd QRS complex in lead aVL look so different? (RED question mark in this lead).
• 
  
• PEARL #1: It is often difficult to distinguish between artifact and baseline wander from true (potentially ischemic) ST depression. Here, the fact that we have a simultaneously-recorded long lead rhythm strip is helpful. Note that among the 10 beats recorded in this long lead V1 rhythm strip — it is the T waves of beats #5 and #7 that appear to be the most altered by artifact. This suggests that the ST-T waves of these 2 beats in other leads are the most likely ST-T waves to be distorted by artifact.
• As a result — I strongly suspected that the flattened and subtle-but-definite ST depression seen for the 1st complex in lead aVL (corresponding to the 4th beat recorded in this tracing) is "real".
• For the same reason — I also suspected that the coved and slightly elevated appearance of the ST segment for the 2nd complex in lead V1 (also highlighted by a RED question mark) has been distorted by artifact.
• PEARL #2: Because neighboring leads often display similar abnormalities — knowing there is subtle-but-definite abnormal ST depression in lead aVL led me to look especially close at the ST segment appearance in the other high-lateral lead = lead I. Doing so — I suspected the subtle ST segment flattening in lead I is probably of similar significance as the ST-T wave abnormality that we see in lead aVL.

To Emphasize: This subtle ST-T wave flattening in leads I and aVL of ECG #1 was the only deviation I perceived from what otherwise would fit my definition for a "normal" ECG. I perceived this ST-T wave flattening in high-lateral leads I and aVL as a nonspecific finding — that in isolation — would not appear to be acute.

• BUT — We need to correlate this subtle change in these 2 neighboring leads with the clinical situation — and, this older man with known coronary disease presents to the ED for new CP (Chest Pain) that began just 20 minutes earlier. In this clinical context — We have no idea what (if anything) this nonspecific ST-T wave finding in these 2 high-lateral leads might mean.
• Without ready availability of a prior ECG — we also have no idea of whether or not there may be additional "new" findings compared to this patient's usual baseline tracing.
• BOTTOM Line: In the clinical context of today's case — this initial ECG is not a "normal" tracing. It may or may not represent early findings in a new acute event. 
• The clinical reality is that it will not be possible for us to know the significance (if any) of the findings in ECG #1 until we have more information (ie, repeating the ECG within no more than 30 minutes — obtaining more than a single Troponin, since even with OMI the initial hs-Troponin may sometimes be normal — and — maintaining close follow-up on this patient's CP symptoms).

  

Comparing ECG #1 and ECG #2:

Waiting 4+ hours until repeating this patient's ECG was clearly a mistake (ie, given that this patient's CP had begun just 20 minutes before his arrival in the ED — repeating the ECG within 10-to-20 minutes with notation of the presence and relative severity of CP at the time this 2nd ECG is done may have been all that was needed to confirm an acute evolving OMI).

• In addition — today's patient had CP that was "recurring and resolving multiple times" during the 4+ hours since the initial ECG was done until the ECG was repeated. And so was missed "golden opportunity" to make a definitive diagnosis long before the 11 hours that it ultimately took to get this patient to the cath lab.

Providers in today's case interpreted both ECGs in Figure-1 as "normal". By inference — this means that providers must have thought that there was no change between ECG #1 and ECG #2.

• To me — this means that providers in this case did not compare the 2 tracings in Figure-1 by looking lead-by-lead to see if there were any changes.

Whereas the ST-T wave changes that have occurred between the time that these 2 tracings were done are subtle — in this older man with known coronary diseae and new CP — Aren't there real differences in leads aVL,V2,V3?

• Note that there is virtually no artifact in ECG #2.
• There is also no change in either frontal plane axis or in R wave progression or QRS morphology between the 2 tracings. This means that lead-by-lead comparison for any subtle differences is valid.
• Lead aVL — now shows coving of the ST segment and T wave inversion that was not present in ECG #1.
• Leads V2,V3 — now show flattening of their ST segment, and reduced T wave amplitude. Thus, there has been loss of the slight, gentle upsloping ST elevation in leads V2,V3 that we so often highlight as the 1st sign of ongoing posterior OMI.
• In the context of these subtle-but-real changes in leads aVL,V2,V3 — I thought there is suggestion of beginning ST elevation, with more acute-looking T wave peaking in each of the inferior leads.
• BOTTOM Line: ECG #2 is not a "normal" tracing. Much more than this — in this older man with known coronary disease and new CP — we see "dynamic" ST-T wave changes in 6 leads — which should suggest acute infero-postero OMI until proven otherwise.

LESSONS to-be-Learned:

• Clinical electrocardiography dictates that ECGs can not be interpreted in the absence of correlation to clinical circumstances. New CP severe enough to prompt evaluation in the ED, occurring in an older man with known coronary disease (as occurred in today's case) — means very high likelihood that the patient is undergoing an acute event — which means that even seemingly slight ECG abnormalities must be accepted as potentially meaningful until we can prove otherwise.
• Quantification of relative CP severity at the time each ECG is recorded is essential for optimal understanding of the ECG changes we are looking for (A number scale from 1-to-10 probably works best for documentation of CP severity).
• When CP recurs (as it apparently did on multiple occasions in today's case) — a repeat ECG needs to be done.
• Attention to detail when comparing serial ECGs lead-by-lead is the BEST way to avoid the errors in today's case of calling ECG #1 and ECG #2 both "normal" — with "no change" between the 2 tracings.