Friday, March 8, 2019

Chest pain resolved, and a "Normal" ED ECG

A previously healthy middle-aged male presented shortly after the acute onset of chest pain very shortly before calling 911.

On arrival, he was pain free:

What do you think?

There is a very subtle inferior MI.  Some may quibble that this ECG is not diagnostic of MI (although I would say it is "all but diagnostic.")

Notice the computer called it normal -- but it is clearly not normal.

The computer should at least call this "Nonspecific ST-T abnormalities."

Jerry Jones commented:
"Any ST depression on the ECG of a patient with chest pain credible for ACS represents a reciprocal change until proved otherwise." 

And also: 
"I have, for many years now, taken exception to the term "non-specific ST-T changes."" There is a finality to that term that suggests "inconsequential, unimportant, pay-no-attention or let's-move-on." I prefer the term "unexplained ST-T changes" --Great comments, thanks Jerry!

I like to say: 
"It is not the ECG that is nonspecific; it is the interpreter who is nonspeciific."  Many "nonspecific" findings are only non-specific because the physician does not recognize their meaning.

In this case, the diagnosis was much easier because there was a prehospital ECG:
less than 1 mm ST Elevation in II, III, aVF, with minimal ST depression in aVL
This is diagnostic of inferior OMI

No clear evidence of ischemia

Computer interpretation
Impressive!  This algorithm called it a STEMI.
It measured all the ST segments and recorded them (see bottom right)
It only measure one lead (III) as greater than 1 mm, so there are not "2 leads with at least 1 mm".  Yet it gave a diagnosis of STEMI.
I have not seen this before, and perhaps they are altering their algorithms?

There was prehospital cath lab activation (as always) and the patient was taken to the cath lab with a door to opening time of about 20 minutes (very fast).

Left Circumflex (LCx):
Circumflex is a large caliber co-dominant vessel with mild, diffuse plaque
There is a single large caliber first obtuse marginal (OM-1) branch that supplies the lateral wall and inferolateral wall via a significant lateral branch, which is occluded in the proximal segment, suggestive of plaque rupture.  Distal/AV groove Cx supplies a cascade of small caliber LPL branches.

This branch is 100% occluded in the proximal subsection, with Pre procedure TIMI 0 flow and Post Procedure TIMI III flow. 

Here is the post-PCI ECG:
Reperfusion T-waves (T-wave inversion) is present in inferior leads.

Peak cTnI = 9.9 ng/mL

Next day echo:

Regional wall motion abnormality-inferior.
Regional wall motion abnormality-inferolateral.
Regional wall motion abnormality-lateral.
Normal estimated left ventricular ejection fraction lower limits of normal.
My Conclusion: a large myocardial territory

Even though the ECG findings were very subtle, there was a large myocardial territory at risk.  Rapid reperfusion resulted in a relatively low peak troponin.

Learning points

1. This is, in effect, a transient STEMI
2. Although transient, and the pain resolved, the artery remained occluded.
3. Computer interpretations cannot be trusted.  (They may be helpful.)
4. Subtle findings on the ECG may be associated with a large myocardial territory at risk

This is another case in which the chest pain was resolved but the artery remained occluded, as with this post 2 days ago:

What will you do for this patient transferred to you who is now asymptomatic?

See the learning points in that case

Comment by KEN GRAUER, MD (3/8/2019):
Superb teaching case presented by Dr. Smith regarding this middle-aged man who was transported to the ED after new-onset chest pain. I focus my comments on a number of additional points regarding ECG interpretation.
  • For clarity — I show in Figure-1 the initial ECG obtained in the ED ( = ECG #1— together with the prehospital ECG that I have pieced together = ECG #2).
  • For educational purposes — I think it helpful to initially interpret ECG #1 as if it was the only ECG available in this case, keeping in mind that this patient’s chest pain had resolved at the time ECG #1 was recorded.
Figure-1: The first 2 ECGs shown in this case (See text).
COMMENT: As per Dr. Smith — in a patient with new-onset chest pain, ECG #1 should be interpreted as acute inferior MI until proven otherwise.
  • The rhythm in ECG #1 is sinus arrhythmia. PR, QRS and QT intervals are normal. The frontal plane axis is about +30 degrees. There is no chamber enlargement. Small q waves are seen in multiple leads — including 2 of the 3 inferior leads (ie, leads II and aVF). The diagnosis of acute inferior MI is suggested by subtle-but-real ST elevation in leads III and aVF, with reciprocal change in lead aVL in this patient who shortly before had complained of new-onset chest pain.

QUESTION: If you focus your attention on simultaneously-recorded leads aVR, aVL and aVF — you’ll see 2 complete QRST complexes. Don’t the ST-T wave complexes in leads aVL and aVF for these 2 beats look quite different?
  • Which complex (or B) in these 2 leads is the “correct” complex?
  • Why is this question important?
  • Would your interpretation of ECG #1 change if both of the complexes in leads aVL and aVF looked like A, instead of like B?

ANSWER: Unfortunately, artifact has been introduced into the recording of complexes A and B in leads aVL and aVF.
  • IF the “correct” complex was — then I would have absolutely NO doubt about the occurrence of acute inferior MI in this patient. That’s because straightening of the ST segment takeoff and, ST elevation in lead aVF for complex is much more evident than it is for complex B. In addition — the scooped ST depression for complex in lead aVL looks to be the exact mirror-image opposite of the subtle ST elevation seen in lead III. One of the best clues to acute inferior MI is this finding of a near-magical mirror-image picture between ST elevation in lead III with reciprocal ST depression in lead aVL.
  • On the other hand — I would be less certain about acute inferior MI if the “correct” complex in leads aVL and aVF was complex B. Considering how reduced QRS amplitude in lead aVF is — I’d still interpret the T wave in lead aVF as taller-than-expected, and probably hyperacute — but, complex B is clearly less convincing than complex A in lead aVF. Similarly — the ST segment in lead aVL of complex B is inappropriately FLAT, which is definitely abnormal. However, complex B in aVL does not provide the mirror-image reflection of what we see for the ST segment in lead III.
  • The SOLUTION: Especially when ECG changes are subtle (as they are in ECG #1) — Have a low threshold for immediately repeating the ECG when critical leads (ie, like leads aVL and aVF in ECG #1) are technically inadequate. We need to know if complex A or complex B is the “correct” one!

Regardless of whether complex A or B is the “correct” one — there are other ECG findings in ECG #1 that are strongly suggestive of acute inferior MI. What are they?
  • HINT: What abnormal findings do you see in the anterior chest leads?

ANSWER: intentionally omitted assessment for R wave progression in my initial interpretation of ECG #1 above. This parameter is often forgotten when a systematic approach to ECG interpretation is not used ... — but assessment for R wave progression is important in this case.
  • Height of the R wave in lead V2 of ECG #1 = depth of the S wave in this lead. Therefore, transition occurs early (ie, between V1-to-V2). In addition — the R wave in lead V3 looks unexpectedly tall.
  • The ST segments in leads V2 and V3 of ECG #1 are straighter-than-they-should be, if not slightly depressed (There is usually slight concave-up ST elevation in these leads).
  • Anterior leads V1, V2 and V3 provide a mirror-image of electrical activity in the posterior wall of the left ventricle. Note the positive mirror-test” shown in the insert in lead V2. This suggests posterior wall involvement.
  • BOTTOM LINE regarding ECG #1: Even if it is complex B in leads aVL and aVF that is correct — in a patient with new-onset chest pain — but whose chest pain has resolved by the time ECG #1 was recorded — the combination of residual abnormal findings in leads III, aVL, aVF + leads V2 and V3 strongly suggest acute MI has occurred (likely infero-postero MI ).

As per Dr. Smith — diagnosis of acute MI was made much easier in this case — because a prehospital ECG ( = ECG #2was found. I’ll add a few thoughts regarding interpretation of this prehospital ECG:
  • Although once again, small q waves are seen in multiple leads of ECG #2 — a large and wide Q wave was seen in lead III that was not seen in lead III of ECG #1. What is strange — is that QRS morphology of the other 5 limb leads is not appreciably different in ECGs #1 and #2. We therefore cannot explain disappearance of this large Q wave in lead III from the prehospital tracing on the basis of an axis shift. Change in patient positioning might be a factor (ie, acutely ill patients may have difficulty lying completely flat for the recording of an ECG) — but unless specific note was made of a change in patient position, this would be impossible to verify.
  • In any event — the ST segment in lead III of ECG #2 is coved and more clearly elevated — this time, with undisputable mirror-image reciprocal ST depression in lead aVL. Together with inferior Q waves, and frank ST elevation present also in lead aVF — ECG #2 is diagnostic of acute inferior MI.
  • Although subtle — note again the hint of ST segment flattening (if not slight depression) in leads V2 and V3 of ECG #2 — once again accompanied by unexpectedly early transition.
  • For more on the magical mirror-image” relationship between leads III and aVL with acute MI — CLICK HERE (Please See My Comment at the bottom of the page.)
  • For “My Take” on use of the Mirror Test for Diagnosis of Acute Posterior M CLICK HERE (Please See My Comment at the bottom of the page.)
  • For “My Take” on use of the Systematic Approach to ECG Interpretation  CLICK HERE.


  1. Steve...

    Thanks for posting these last two OMI cases because they certainly give guidance to those physicians who may have only an intermediate level of ECG interpretation expertise and those with more advanced ECG skills (i.e., those who would have recognized the occlusion) but may have had questions about what approach to take next.

    Without the benefit of the prehospital tracing, I was immediately concerned about an acute inferior OMI because of the ST depression in aVL. According to my own rule (Jones's Rule), "any ST depression on the ECG of a patient with chest pain credible for ACS represents a reciprocal change until proved otherwise." Although this patient was pain-free at the time the second ECG was performed, chest pain was his reason for transport. My rule does not mean you will ALWAYS diagnose an OMI because sometimes you WILL rule it out (e.g., previous ECGs with exactly the same pattern).

    Also, I have, for many years now, taken exception to the term "non-specific ST-T changes." There is a finality to that term that suggests "inconsequential, unimportant, pay-no-attention or let's-move-on." I prefer the term "unexplained ST-T changes" which implies more that perhaps one should take a closer look. I am, however, well-aware that the term "non-specific" was never intended to have those connotations I mentioned but unfortunately, for too many people reading ECGs, it does.

    And, as always, thanks for an incredible website and forum for learning!

    1. Thanks, Jerry. I'm going to put a bit of this with reply in the main part of the post.

    2. Hi Jerry. Your point about use of the term, “non-specific ST-T wave changes” is well taken. That said, I submit that the ST-T wave changes in ECG #1 were VERY specfic (especially if complex A in my Figure-1 represents the correct QRST complex). And, taken together — the combination of abnormalities I cite above for ECG #1 in this patient with new chest pain are diagnostic of a “specific” entity = acute MI. Someone once counted the number of entities that might affect ST-T waves — and came up with over 50 possible causes (many of them non-cardiac). But coronary disease (including acute ischemia/infarction) ARE among those causes. So my preference in teaching ECG interpretation has always been to use the term “non-specific” for ST-T wave flattening for which no definite cause is apparent — at which point clinical correlation is essential to determine if recent ischemia/infarction might be relevant to the case at hand. Jerry — I realize I am talking “semantics” — and I know that you and I agree that ST-T wave changes for the ECGs in Figure-1 are definitely not “non-specific”. THANKS as always for your insightful comments!

    3. Ken...

      Thanks for addressing my comment, but I think you may have missed my point. It's not a question of whether the ST-T changes are specific or non-specific; it's that the term "non-specific" is itself too often interpreted as inconsequential, as in "you can't know any more about this." As you mentioned, the "non-specific" changes are, at times, actually quite specific (and we have a good example with this case)! But we must be curious enough to inspect these "non-specific changes" a bit further rather than accept the term as a "write-off" for this particular finding. Neither you nor I, nor Steve nor Pendell would do that, but there are many people reading ECGs who would accept a machine's interpretation of "nonspecific changes" and not pursue it any further. That would have been disastrous in this case without the first ECG! I am aware that there exists a long list of factors that affect the ST-T segment; but I've never been especially impressed with (nor intimidated by) long lists following an incident during rounds when I was an IM resident. Our attending was listening to one of my fellow residents recite a differential diagnosis for his patient. This dragged on and on and on. When he finally finished he gave everyone a very proud and smug look. The attending looked at him for a moment, then said, "Young man, I'm never impressed by a physician who can recite a lengthy differential diagnosis; I'm much more impressed with the physician who has done such a thorough history and physical exam on the patient that such a long list is completely unnecessary." Though there may be many reasons for ST-T changes, the specific clinical circumstances are usually sufficient to narrow the possibilities to a much more manageable number (as it did in this case).

      Personally, I have always found the term "non-specific" insufficient. Nothing appears on an ECG without reason, so I feel the term "unexplained" is more fitting. Certainly, much of what we are discussing may appear to be nothing more than a matter of semantics, but words can affect attitudes and attitudes can have a direct impact on the quality of care.

      By the way, I really enjoyed your comparison of the two QRS complexes (A and B) and your analysis of their disparity. This is also a problem that anyone who reads ECGs on a regular (or even semi-regular) basis will encounter rather frequently. I really admire your stepwise, methodical approach to these issues and find it very refreshing. Best Wishes,

      Jerry W. Jones, MD FACEP FAAEM

    4. This comment has been removed by the author.

    5. Thanks for your detailed comment Jerry! I actually did not misunderstand your point — it’s just that I have a different experience and a different viewpoint. For me the term, “unexplained” implies that the ECG finding may be more serious than it actually is. Realizing that you and I are discussing FINE details of sophisticated interpretation — I’ll continue by rhetorically asking, “What is a ‘nonspecific’ [or ‘unexplained’] ECG finding?” I see MANY cardiologists calling an ECG “normal”, despite the presence of ST segments tending toward flatter-than-they-should-be-if-everything-was-truly-normal, and despite the presence of T waves that are smaller-than-they-should-be. To call these findings, “unexplained” to me implies that there is a definite pathology. To instead call them “nonspecific” ( = my preference) — instead to me means that the ECG is not completely normal, though it is not necessarily reflective of worrisome pathology. I always complete my written interpretation of such tracings with “Suggest clinical correlation” — and that’s the way I teach. In fact, the reason I approach ECG interpretation in this way gets at PRECISELY the goal that I believe you have = getting clinicians to recognize that ST flattening in a number of leads is NOT a “normal” finding — and that the next step is to CORRELATE clinically with what’s going on with the patient. BOTTOM LINE — I really believe you and I finish at a VERY SIMILAR end point in ~99+% of cases — though we have slightly different ways of getting there (and sometimes different ways of teaching the same important concepts). I’ve always felt it GOOD to have different approaches to similar problems — and that’s how we all learn from each other! THANKS (as always!) for your thoughtful commentary. P.S. Glad you liked the part about Complexes A & B in this case — which to me, made a tremendous difference in my certainty about what was going on with this patient. This point is all-too-often ignored by less experienced interpreters — :)

    6. Ken...

      It's always very enlightening to disagree with you - though our disagreements always seem to involve infinitesimal nuances of topics upon which we basically agree. I learn from them and I hope others following along also learn that disagreement is not always an "all-or-nothing" thing nor should it result in any acrimony. What we've been debating here is our respective interpretations of a term that, in electrocardiography, has no actual agreed-upon definition.

      Thanks for the discussion.

  2. Hi, Dr.Smith, I am a medical student from Taiwan, in the figure 1, I could only see mild ST depression in lead aVL, mild ST depression in lead aVF, did I neglect any other important hints? If without prehospital ecg, I might just let this patient leave the emergency department.

    1. See what I added to the post in response to Jerry Jones

    2. — There's more in My Comment (which I added after you wrote your question — See above — :)


Recommended Resources