ST-Segment Morphology

In a previous post we discussed the problem of ST segment elevation.

Because acute myocardial infarction (STEMI) is not the most common cause of ST-segment elevation amongst chest pain patients, we need to consider other factors like reciprocal changes to shore up the diagnosis.

It’s also a good idea to be well versed in the typical appearance of the STEMI mimics (e.g., left ventricular hypertrophy, early repolarization, left bundle branch block, paced rhythm, pericarditis, ventricular aneurysm, hyperkalemia, and so on).

Another factor that can assist you is an analysis of the morphology of the ST-segment.

The normal ST segment should not be flat. It should have an upward concavity sometimes referred to as a “take-off”. When an ST segment loses its upward concavity and becomes straight or upwardly convex, it’s suggestive of acute myocardial infarction.

Consider this image from Brady W, Syverud S, Beagle C et al. Electrocardiographic ST-segment Elevation The Diagnosis of Acute Myocardial Infarction by Morphologic Analysis of the ST Segment. Acad Emergency Med. 2001;8(10):961-967.


You can draw an imaginary line between the J point and the apex of the T wave. If the ST-segment is below that line, then it’s upwardly concave. If it’s even with or above that line, then it’s “non-concave” (straight or upwardly convex) which is suspicious for acute myocardial infarction.

This phenomenon was described earlier by Pardee in An Electrocardiographic Sign of Coronary Artery Obstruction. Arch Intern Med. 1920;26:244-257 where upward convexity was referred to as “coving” of the ST-segment, and you still hear the ST-segment elevation of Brugada syndrome referred to in this way.

concave_nonconcave_smile_frownIf it helps you to remember, an upwardly concave ST segment makes a “smiley face” (good) and an upwardly convex ST segment makes a “frowny face” (bad).

Does that mean that acute myocardial infarction always presents with non-concave ST-segments when ST-segment elevation is present?

Not at all! This finding is not particularly sensitive. For example, the ST-segments retain their upward concavity 40% of the time with LAD occlusion. It is, however, fairly specific.

The STEMI mimics are more likely to present with upwardly concave ST-segments and an absence of reciprocal changes (although left ventricular hypertrophy, left bundle branch block, and paced rhythm can show pseudo-reciprocal changes due to a pattern of T-wave discordance).

Stephen Smith, M.D. from Dr. Smith’s ECG Blog has shown me a couple of cases of left ventricular hypertrophy with upwardly convex ST-segments in the right precordial leads that were not experiencing STEMI so with a strain pattern it’s more important to consider the depth of the S-wave.

You might have noticed that I used the phrase “upwardly concave” as opposed to simply “concave”. That’s because “concave” is “convex” depending on your perspective. That’s why I always mention the direction of the concavity or convexity.

This can get confusing! Consider this image from the AHA’s STEMI Provider Manual.

AHA_STEMI_44_markup 1

The caption says “concave down” even though it’s referring to the first window showing upwardly concave ST-segments. It also indicates that the ST-segment elevation is “possibly due to early repolarization” but that is unlikely because the R-waves are not well developed and there is a tiny Q-wave in lead V3.

The second window shows the same patient 10 minutes later with non-concave (straight) ST-segments and hyperacute T waves consistent with LAD occlusion.

The third window shows the same patient 1 hour later following PCI. Note the Q-waves and terminal T-wave inversion. If you didn’t know the history this would be a difficult ECG. It can be hard to distinguish between late presents and the so-called left ventricular aneurysm pattern (persistent ST-segment elevation from previous myocardial infarction).

Do you need comprehensive training in 12-lead ECG interpretation?

Check out our video-based 12-lead ECG course at ECG Medical Training!

Updated 12/21/2015


  • Shaggy says:

    Thanks for presenting the concave vs. non-concave in a format easier to remember. I admit I always had difficulty in this when discussing pericarditis.

  • Tom B says:

    You're welcome, Shaggy! One of the most important clues for pericarditis is ST elevation in leads I and II (two leads that are normally reciprocal).Add in PR segment depression and an absence of reciprocal changes (with the exception of lead aVR) and the history, and the correct dx emerges! At a recent AHA ML meeting, a paramedic said that his medical director didn't expect paramedics to be able to determine whether or not ST elevation was due to acute STEMI or pericaditis (for example).I thought this was unfortunate! Paramedics can identify the mimics with a high degree of accuracy with the right education! We should help spare our patients from unnecessary and potentially risky procedures.Tom

  • Shaggy says:

    "Paramedics can identify the mimics with a high degree of accuracy with the right education!"The key is, do medics currently have the "right education". If not, that MD is not incorrect in his opinion. I am not sure if all medic programs are currently on the same page when it comes to 12 lead interpretation.

  • Shaggy says:

    Besides, have you ever considered doing more advanced 12 lead programs for pre-conference workshops on the continent? Something to think about. If Mike Smith, and other big names can be in hot demand just to make people laugh or feel good about themselves, I am sure there would be a demand for you to actually teach people critical knowledge that may be lacking in some regions.

  • Tom,I think it would be helpful to use a systematic approach to rule out mimics. Just as we look at the 12-lead to interpret the rhythm and axis every time, we should have a system to rule out early repol, hyperkalemia, pericarditis, hypertrophy, brugada etc…How would you go about this? From most common to least common[mimics], or do you just know right away?

  • to clarify, I mean a step by step approach, using the rules we know/you've taught us.

  • Shaggy says:

    That is a good idea, but I guess with time, you should be able to look at it like you look at a 3 lead. That is of course if you are very proficient like Tom. People like me could use any additional help we can.

  • Tom B says:

    Shaggy – What I thought was disappointing was the fact that it was the paramedic's Medical Control Physician who didn't expect paramedics to be able to tell the mimics apart.I could sympathize with an ED physician who was only recognizing that paramedics could not routinely differentiate pericarditis from STEMI. If you're building a STEMI system, and you're a Medical Control Physician, then I think you have a responsibility to make sure that your paramedics can identify the most common STE-mimics.You're certainly correct that paramedics are not being adequately taught in school. I'm not entirely satisfied with the aftermarket education, either.Tom

  • Tom B says:

    Shaggy – I agree with you that a pre-conference workshop would be a good place for some advanced 12 lead ECG interpretation.Tom

  • Tom B says:

    Adam – You're correct in that screening for STE-mimics should be a systematic part of every 12 lead ECG interpretation.I can spot them fairly easily now, but in the beginning I would follow a 6-step method.1.) Rate and rhythm2.) Axis3.) QRS duration (intervals)4.) Morphology5.) STE-mimics6.) Ischemia, Injury, InfarctI do look for the most common STE-mimics first, and I combine some of them (LBBB/paced, BER/pericarditis).It's also a good idea to familiarize yourself with the so-called "strain patterns" of LVH.Left ventricular aneurysm is a tough one! I'll definitely be revisiting the topic of STE-mimics in the months ahead! Tom

  • Shaggy says:

    "I'll definitely be revisiting the topic of STE-mimics in the months ahead! "Can't Wait! Can we hold you to that? 🙂

  • Tom B says:

    You sure can! That's the whole point of the PH12ECG blog.Tom

  • London Cardiologist says:

    "After PCI, you can see the development of Q waves and terminal T wave inversion (which usually indicates a STEMI that's been around for a while)".It's worth pointing out that these changes only occur after PCI if the patient develops a large infarct (for example, if there is a delay from inital symptoms to the procedure, or if the procedure is not successful). If the patient is rapidly revascularized, the resulting infarct is smaller and these changes are not typically seen.MRI studies have shown that the presence of Q-waves is associated with larger infarcts, not with a full-thickness infarct (which is what's commonly taught).

  • Tom B says:

    Excellent point. Thanks for the comment! Tom

7 Trackbacks

Leave a Reply

Your email address will not be published. Required fields are marked *