ST segment concavity is just one small piece of the puzzle that is pattern recognition

Written by Pendell Meyers

Here are two striking examples from a single shift highlighting the fallibility of the standard "smiley face" or upwardly concave ST segment morphology "rule".

Case 1.

Obvious STEMI and OMI with massive STE in the inferior and lateral leads, even extending back into V3. 
All STE has concave upward (smiley face) morphology. V2 shows STD indicating posterior involvement. 
V1 may be in a tug-of-war between STD from posterior involvement and STE from possible RV involvement. 
Reciprocal STD in I and aVL. 
Sinus bradycardia with first degree heart block present, clearly high risk for worsening bradycardia and/or heart blocks.

I received this ECG via telemetry from an ambulance far from our hospital. There was another cath center closer than our institution, so we agreed they should go there and advised they activate their cath lab immediately. Before our conversation was finished, the patient went into VF arrest and they hung up. I was unable to get back in touch with them for the outcome.


Case 2.

Hyperacute T-waves in V2-V6, I, and aVL. STEMI criteria met by the STE in V6, I, and aVL. Reciprocal STD in III and aVF. V5 has very diminished voltage and poor quality, suggesting some combination of lead misplacement and/or physical object between it and the patient (hair, defibrillation pad, medical cables, clothing, etc.).

I received this also by telemetry, and again we agreed with their plan of the nearest cath center and activated their cath lab protocol. He was found to have proximal LAD occlusion, but other outcome details are not available.

We are taught in medical school that concave upward ST segment morphology is a feature of benign STE, pericarditis, etc.

In actuality, 40-50% of acute LAD occlusion have upwardly concave ST segment morphology in all of V2-V5.

Smith SW. Upwardly concave ST segment morphology is common in acute left anterior descending coronary artery occlusion. Journal of Emergency Medicine 2006; 31(1):67-77.

In anther study of 355 LAD occlusions, only 36 of 355 LAD occlusions were excluded as "obvious" due to non-concave morphology:

Smith SW et al. Electrocardiographic Differentiation of Early Repolarization from Subtle Anterior ST-Segment Elevation Myocardial Infarction. 
https://www.annemergmed.com/article/S0196-0644(12)00160-6/pdf

Brady et al. looked at all sites of MI and found that non-concave morphology had sensitivity and specificity of 77% and 97%, with PPV and NPV of 94% and 88%.

Brady, W. J., et al. 2001. Electrocardiographic ST Segment Elevation: The Diagnosis of Acute Myocardial Infarcton by Morphologic Analysis of the ST Segment. Academic Emergency Medicine 8 (10): 961–67.

Learning Points:

Concavity (aka "upward concavity") is just one small piece of pattern recognition of OMI, and is not reliable to assuage concern by itself.

Up to half of LAD occlusions have concave ST segments, and overall non-concavity (upwardly straight or convex) is optimistically 77% sensitive for OMI of all MI sites (including inferior), but is far less sensitive for anterior OMI.

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

Comment by KEN GRAUER, MD (6/22/2019):

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

I suspect I am the person MORE than any other who has popularized the concept of “smiley” vs “frowny”-shape configuration as a visual AID for clarifying description of ST segment elevation morphology. 

• My first ECG publication (in 1983) featured this concept — and I’ve published illustrations similar to that shown in Figure-1 in all editions of each of my 4 ECGs books over the past 36 years.

Figure-1: Illustration of the concept of “smiley” vs “frowny” ST elevation (adpted from My Comment at the bottom of the page in Dr. Smith’s March 11, 2019 blog post).

-------------------

KEY POINT: The purpose for using the terms, “smiley” vs "frowny” is merely descriptive. Textbooks describe upward and downward concavity or convexity — but students (as well as experienced clinicians) are all-too-often not sure if one is supposed to be looking from above or below the “concavity”. But the instant one says, “the ST segment has a ‘smiley’ or ‘frowny’ shape” — ALL doubt is removed (Figure-1).

• In general — in a setting in which ECGs are done for some purpose other than assessment of acute cardiac-sounding chest pain — ST elevation with a “smiley”-shape ( = upward concavity), especially if seen in association with J-point notching and in the absence of other findings suggestive of acute ischemia (ie, Q waves, reciprocal ST depression) — will most often indicate a benign repolarization pattern (Panel B in Figure-1).
• Clinical Correlation is essential! ALL BETS are OFF — IF the patient with upward concavity (ie,“smiley”-configuration) ST elevation presents with new-onset worrisome chest pain! In this case — clinical decision making depends on a constellation of evaluative factors (ie, details of the history, exam, serial ECGs, troponins, stat Echo at the bedside, etc.). True "smiley"-shaped ST elevation in selected leads that is not out-of-proportion to QRS amplitude in the same lead — and, which occurs in the absence of large Q waves and ST depression are factors that may favor a non-ischemic cause — but which can not be depended upon as definitive. The whole clinical picture needs to be considered ...
• On the other hand — a coved shape ( = “frowny”-configuration) to ST elevation is more likely to suggest ischemia (Panel A in Figure-1). This is especially true when seen as a new ECG finding in a patient with acute symptoms.
• Panel C — shows subtle ST segment coving ( = “frowny”-configuration) in the inferior leads, with perhaps slight ST elevation and early T wave inversion in lead III. In a patient with new symptoms — this should suggest ischemia until proven otherwise.
• In contrast is Panel D — which shows J-point ST elevation with an upward concavity ( = “smiley” configuration) in leads V5 and V6. In addition, there is J-point notching (RED arrow in lead V5 of Panel D). In a patient without acute cardiac symptoms — this pattern of ST elevation + J-point notching almost certainly reflects early repolarization.
• Exceptions to the above generalities abound! — as for certain types of benign repolarization variants in which the identifying feature is a special form of “frowny”-shaped ST elevation (See My Comment at the bottom of the page in Dr. Smith’s June 20, 2019 blog post — from just 2 days ago! ).

BOTTOM LINE: There is no “rule” about any ST segment “smiley” shape. Instead, please use the shape descriptions in Figure-1 as they were intended by me, ever since 1983 — as a simple visual aid for clarifying ST segment elevation shape (and for effectively communicating this shape to your colleagues) — which in context with the clinical scenario may assist in clinical assessment.

----------------------------------

About the 2 ECGs in this blog post:

• For clarity — I have adjusted resolution and reproduced both ECGs in this blog post in Figure-2. Both patients had symptoms and were transported for immediate care.

Figure-2: The 2 ECGs shown in this blog post (See text).

My THOUGHTS on ECG #1 — The rhythm appears to be regular, at a rate just under 60/minute. The QRS is narrow. Although there is baseline artifact — there appears to be an upright (sinus) P wave in lead II, with a long PR interval. Thus, the rhythm is sinus bradycardia with 1st-degree AV block.

• The QTc appears to be normal.
• The frontal plane axis is normal (about +70 degrees).
• There is no chamber enlargement.

Regarding Q-R-S-T Changes:

• There are no Q waves.
• Transition occurs slightly early (the QRS complex is equiphasic by lead V2, and predominantly positive by lead V3).
• The most remarkable finding is dramatic ST elevation in each of the inferior leads, and in leads V3-thru-V6. In addition — there is almost equally dramatic reciprocal (mirror-image) ST depression in lead aVL, and to a lesser extent in lead I. This is an obvious extensive acute infero-lateral STEMI.
• There is probably also acute posterior MI — because there is subtle-but-real ST depression in lead V2 in association with a large acute inferior STEMI.
• There is probably also acute Right Ventricular MI — because the T waves in leads V1 and V2 are so prominent (much fatter-than-they-should-be at their peak) in association with a large acute inferior STEMI. This is almost certainly the reason there is no ST depression in lead V1, and no more than minimal ST depression in V2 (ie, because of opposing RV forces trying to elevate ST segments in right-sided leads).
• NOTE #1: The concept of ST segment “shape” (ie, smiley-configuration) does NOT enter into play in this case — because this is an obvious acute STEMI (ie, it is irrelevant whether ST segments are upward or downward sloping).
• NOTE #2: The “culprit” artery is almost certain to be the proximal RCA because: i) ST elevation in lead III > II (the opposite tends to be true with LCx occlusion); ii) There is marked reciprocal (mirror-image to lead III) ST depression in lead aVL; iii) ST elevation in lead III > than in lead V6 (the opposite tends to be true with LCx occlusion); and, iv) The finding of probable acute RV MI almost always localizes the “culprit” artery to the proximal RCA.

QUESTION: Did you see the Terminal QRS Distortion ( = T-QRS-D) ?

• T-QRS-D — is defined as the absence of both a J-wave and an S-wave in either lead V2 or lead V3. When present in either lead V2 or V3 — T-QRS-D tells you that there is acute OMI! In ECG #1 — there is neither a J-wave nor an S wave in lead V3 (the negative deflection that is elevated by the ST segment in lead V3 never comes back down to the baseline).
• While recognition of T-QRS-D seen here in lead V3 is obviously not needed to diagnose this large acute STEMI — T-QRS-D is a wonderful concept (conceived by Dr. Smith) that is well worth knowing about, because it may sometimes be the deciding clue that tells you there is acute OMI (CLICK HERE — for more on T-QRS-D; Please check out My Comment at the bottom of this page).

---------------------------

My THOUGHTS on ECG #2 — The rhythm looks to be fairly regular (although difficult to assess given poor quality of the tracing). This again looks like sinus bradycardia.

• The QRS is narrow.
• The QTc appears to be normal.
• The frontal plane axis is normal (about +15 degrees).
• There is no chamber enlargement.
• Something is wrong with lead V5. We get no useful information from lead V5.

Regarding Q-R-S-T Changes:

• There are small q waves in lead I and V6. (It is hard to be certain whether there are or are not small q waves in leads III, aVF and/or aVL).
• Transition (where the R wave becomes taller than the S wave is deep) occurs normally between leads V3-to-V4 in ECG #2. That said — R wave amplitude looks to be reduced in leads V1, V2 and V3 — which given other findings probably is significant!
• The most remarkable finding is the dramatic increase in T wave size! This hyperacute T wave pattern seen in leads V2, V3 and V4 resembles the picture of DeWinter T Waves, albeit without any J-point ST depression. In association with the ST elevation we see in leads I, aVL and V6 + inferior reciprocal ST depression — these findings in ECG #2 strongly suggest acute proximal LAD Occlusion. So, the loss of anterior R wave we noted a moment ago, may indeed reflect ongoing loss of anterior myocardium.
• NOTE #3: A proximal location of this LAD occlusion is strongly suggested because: i) The finding of hyperacute T waves begins as early as lead V2; ii) There is also significant ST elevation in lead aVL; and, iii) There is reciprocal inferior ST depression (mid- and distal LAD occlusions tend to show acute ST-T wave changes that aren't as marked by lead V2 — and they often lack reciprocal inferior ST depression and ST elevation in lead aVL).
• NOTE #4: Once again, the concept of ST segment “shape” (ie, smiley-configuration) does NOT enter into play in this case — because once again, there is an obvious acute STEMI (ie, it is irrelevant whether ST segments are upward or downward sloping when it is otherwise obvious that an acute STEMI is occurring).