Wednesday, November 1, 2017

A Middle-Aged Man with crescendo angina

A middle-aged male with several comorbidities including DM presented with chest pain.  It had been on and off all day, then constant for 2.5 hours.  It was central without radiation, sharp but also tight, and was 9/10.    There was some cough, but no SOB; he reports COPD and has been using his inhalers without improvement. He also reports diaphoresis and tingling of his hands since the worsening of pain at 1700.

His BP was 160/90.

He was given aspirin and nitro via EMS with good improvement in his chest pain.

First ED ECG at 2:40 after chest pain became severe.
Sinus rhythm.
Slight left axis deviation (mostly negative QRS in II, R-wave aVL; axis about -40)
There is a bit of suspicious flattening of the ST segment in inferior leads and a tiny bit of STE in aVL, but it is really non-diagnostic and nearly normal.
Precordial leads all have minimal, if any, ST elevation

Here was a previous ECG while dehydrated a few months prior:
Sinus tachycardia. 
In this ECG, there is a hint of ST elevation in III, with reciprocal ST depression in aVL.
None of this was ischemic, but it does suggest the patient's baseline is different from the presentation ECG above.
Here I put limb leads one after the other for more close inspection:

II,III,aVF,aVL only: Presentation ECG

II,III,aVF,aVL only: Previous ECG

At approximately 30 minutes after arrival, the first troponin returned at 0.046 ng/mL (99% URL = 0.030 ng/mL)

The patient was put on heparin and Nitroglycerin, but his pain could not be controlled.

Another ECG was recorded:
There is lead reversal of limb leads.
As for the precordial leads, there is not much change.

A bedside echo was performed:

The segment on the upper left is not contracting. This is the anteroseptal wall.

Here is a 4-chamber view:

The apex is not contracting.

From all this data, one should concluded that there is a coronary occlusion that is not showing on the ECG, until proven otherwise.

25% of acute coronary occlusion does not have diagnostic ST elevation.  Many of those cases have ECG that are highly suggestive of ischemia.  An unknown number of acute occlusions have normal, or near normal ECGs, but it is NOT uncommon, and one should be on the lookout for them.

A positive troponin in this clinical situation [no reason for an elevated troponin: no end stage renal disease, no heart failure, no reason for a type II MI (BP of 160/90 should not be thought to be a cause)], ALONG with the inability to control the pain with medical therapy is an indication for immediate angiography, according to the guidelines of both the American Heart/American College of Cardiology and the European Society of Cardiology.

This patient, based on just the 1) troponin and 2) refractory pain, should go emergently to the cath lab even with a normal ECG.

Actually, although he was put on IV NTG, it was not maximal. This should be titrated to a significant drop in BP, down to perhaps 110-120 systolic.  This was not done.

The echo is just extra data to support that plan.

However, the vast majority of patients with NonSTEMI that really need the cath lab now do not go in a timely way.  In fact, in this large meta-analysis by Khan et al. of 40,000 NonSTEMIs published in the European Heart Journal in August 2017, 25% had occlusion and it took 24 hours to get to angiography.

Unfortunately, this patient proved the rule rather than the exception.  He was admitted with a diagnosis of NonSTEMI without immediate angiography.

But Fortunately, he was admitted to excellent internal medicine residents, who immediately recorded another ECG when he arrived upstairs.

Here it is, one hour later:
Now there is diagnostic ST Elevation

The cath lab was activated.

This was recorded just before angiography:

Left ventricular descending coronary artery: 100% thrombotic occlusion in the midportion of the vessel distal to the origin of the first diagonal.

Next Day
Deep QS-wave in V3

Troponin profile

This was a big myocardial infarction!

Learning points:

1. Some coronary occlusion do not show on the ECG, or show only minimally.
2. Even serial ECGs may not show the MI
3. If the troponin is elevated in a patient with a chief complaint of chest pain, it is a type I MI until proven otherwise.
4.  In a type I MI, if the symptoms can't be controlled with antiplatelet, antithrombotic, and anti-ischemic therapy (NTG), then immediate cath lab activation is indicated.
5. Read your cardiac ultrasounds carefully. Wall motion abnormalities are hard to see. You can only say there is no wall motion abnormality if: 1) The images are very high quality, with bubble contrast, and 2) you are expert.


  1. Please compare the 5th and 6th recordings! 4
    "Not much change." ???
    Laszlo Farkas

    1. the only place that I state "not much change" is between the 1st and 2nd recordings.

  2. Yet another outstanding case!

    ps -minor point, does the 30min/3rd EKG also show R/L limb lead reversal?

    Infinite thanks!

  3. I would like to ask if the change of axis in the vertical plane in the 5th ECG is just an artifact? Thank you

    1. I have assumed it is lead placement, but thanks for pointing that out.

  4. Between the first and second ECGs, I believe there is a subtle but notable change in morphology, by way of straightening of the ST segment. Would this not be suspicious for evolving anterior STEMI?

    Dave B

    1. Dave,
      If you see it, then you're better than I am!

  5. Superb case by Dr. Smith that emphasizes the essential points about how acute coronary occlusion will not always be recognizable from the first few ECGs. It is in cases like these in which initial tracings were not diagnostic of the large MI that this turned out to be — that I find it helpful to look back at those tracings we felt were non-diagnostic to see if “knowing the answer” allows us to go back and now better discern subtle changes that had been present. The 1st point to emphasize — is that the limb leads on ECG #1 were not adequate technically for ST-T wave assessment. For example, for each of the 4 beats in each of the 3 inferior leads of ECG #1 — there is a marked difference in ST-T wave appearance due to artifact. Some of the beats (ie, the ST-T wave for the 3rd QRS complex in leads II and III) clearly look worrisome — but not so for other complexes. In a patient with new-onset severe chest pain — this 1st ECG should have been repeated! Even when one looks at the 15 beats in the long lead II — we see significant variation in ST-T wave morphology, which should leave us aware of simply not being able to render appropriate judgment about subtle initial ST-T wave changes. As per Dr. Smith — comparison of ECG #1 with the Prior ECG (done a few months earlier) does yield some differences in ST-T wave morphology. While possible this could be due to a difference in heart rate and clinical condition — one should also acknowledge the possibility of a “dynamic” ST-T wave change in a patient with new chest pain. Then, there is ECG #2 — which showed limb lead reversal (global negativity in lead I; global positivity in lead aVR). Although we are not told if the treating clinicians recognized this limb lead reversal — what was NOT done, was to repeat this 2nd ECG. I strongly suspect that if this 2nd ECG had been repeated with limb leads in correct position, that we would have seen some recognizable ST-T wave abnormalities — because the ST segment straightening and slight deviation that we see in the inferior leads of ECG #2 suggests abnormality superimposed on the limb lead reversal. ECG #3 (done 1 hour later) is obviously abnormal. Looking at this ECG #3 in “the retrospectoscope” makes me wonder if those T waves in ECG #1 in leads V2-V5 that I thought were non diagnostic, but which did look a bit fatter-than-usual at their peak (and which clearly looked different than the corresponding T waves in the ECG baseline done a few months earlier) — were in fact a harbinger of the chest lead ST elevation to follow. Even in “the retrospectoscope” — I would not have called a stemi from ECG #1 — but I probably should have been more suspicious than I was. BOTTOM LINE: Dr. Smith’s KEY points about this case hold true: i) coronary occlusion will not always be readily apparent on the first 1-2 ECGs …; ii) emergency providers should do their utmost to hone their echocardiographic skills! and iii) regardless of whether diagnostic changes are or are not apparent on initial ECGs — the patient with new worrisome chest pain and either some troponin and/or wall motion abnormality on echo deserves prompt cardiac catheterization to define the anatomy. THANKS to Dr. Smith for presenting this case.

  6. It is amazing and especially concerning that 25% of NSTEMI shows coronary occlusion on angiography. Rather frequently seen during our clinical practice, but very scaring when put in number as the that meta-analysis!
    Thanks for presenting this very interesting and instructive case!

    1. But Mario — this case WAS a STEMI (See the ECG @ 1 hour) — and as per my detailed comment just above yours — my strong hunch is that the inferior leads on the 2nd ECG (done 30 minutes later) would probably have shown worrisome changes IF providers would have immediately repeated that ECG as soon as they recognized (IF they recognized) that the limb leads were reversed. So beginning acute changes WERE in fact seen as early as 30 minutes after arrival in the ED — and, if you compare the initial ECG with the baseline tracing from a few months earlier — there WAS already a "difference" in ST-T waves from the 1st ECG done in this case (on ED arrival) compared to that earlier baseline — suggesting in a patient with new chest pain "dynamic change" and potential ongoing event. And, that 1st ECG should ALSO have been repeated — because artifactual distortion in each of the 3 inferior leads really prevented accurate assessment ... (though that was apparently not recognized ... ). So, yes TRUE that a number of NSTEMI's show coronary occlusion on angiography — but this case should not have been called a "NSTEMI" ... As per Dr. Smith, there are LOTS of important lessons to be learned from this insightful case!

    2. I slightly disagree. NSTEMI is the term applied to MI that does not have ST elevation that meets the so-called "criteria." This one definitely does not. What is not appreciated is that the criteria are NOT met for 25% of acute coronary occlusion, and that is why they are so often missed. I agree that there are clues to the presence of occlusion, and that in many occlusions that are not STEMI, the presence of occlusion is obvious to anyone with experience, and especially to you and me. I don't think this is one of them. The changes could easily be either nonspecific or suggestive of ischemia that is not due to total occlusion. The indication for the cath lab here is not the ECG, but the troponin and/or echo in the presence of uncontrolled pain.

  7. Yes, Ken, the present case was really a STEMI, clearly manifesting itself on ECG one hour later (I do have already read your insightful comments).
    I was referring instead to the prevalence of corarary occlusion in NSTEMI, a very concerning percentage (1 patient out 4!), as the mentioned recent meta-analysis shows.
    The ECG one hour later shows diagnostic ST elevation, thus fulfilling “formal” criteria for STEMI. As we all know, millimetrical STEMI criteria are unsensitive as this case too proves: the present myocardial injury does start with NSTEMI criteria (according to formal STEMI criteria) and then does develop into clear STEMI (always according to millimetrical criteria). Probably, if the diagnostic ECG (1 hour later) would have not been recorded and the patient's symptoms would have relieved, perhaps this case would have fallen into NSTEMI category and coronary angiography would have been performed the next morning, then with the “unexpected” surprise to find an occluded coronary artery.
    As already magistrally underlined by both of you, this case does show the importance (if any) of comparing prior ECGs and the need to get serial tracings, in order to see dynamic changes which, in an appropriate clinical scenario, should prompt cath lab activation, regardless of 1 or 2 or n- mm STE.

    Thanks for the very interesting discussion!

    1. Grazie Mario — Siamo completamente d'accordo! (We comletely agree)


Recommended Resources