Monday, June 15, 2020

A man in his 60s with shockable arrest, then ROSC, and no STEMI - what will you do?

Case submitted by Allie Oberst and Ashley Mogul, edits by Pendell Meyers and Steve Smith

A male in his late 60s with PMH of poorly controlled Type 2 DM, HTN, HLD, and “silent heart attack” per family presented to ED with ROSC following witnessed cardiac arrest with initial bystander CPR performed. Per friends, he had been out on a walk in the park. EMS reported that patient was in pulseless VT initially, for which defibrillation and 1 round of CPR and epinephrine were given with ROSC achieved. The patient was unresponsive upon arrival.  
Here is the presenting EKG:

What do you think?

Relevant findings include:
 - Slight STE in anterior leads (V1-V5, approximately 0.5mm V1, 1.0 mm V3, and less in the other leads)
 - reciprocal ST depression in V6, I, and II
 - Extremely poor R-wave progression
 - Q-wave in V3
 - Slightly large, possibly hyperacute T waves in V3-V5 (by themselves not large, but quite large in comparison to the diminished R waves of the poor R wave progression)

I sent this to Dr. Smith with no other info (including no history of cardiac arrest), and this is what he wrote back:
"very suspicious for LAD occlusion, possibly subacute or superimposed on previous anterior MI."

These findings combined are extremely concerning for OMI of the LAD, and fits a pattern we have described multiple times on this blog:

A man in his 30s with chest pain

The anterior OMI vs. normal variant STE formula would not technically be applicable in this case due to STD in II and V6, and also due to Q-wave in V3, however here are the results:

STE 60 msec after J point in V3 = 1.5 mm
QTc = 367 ms
R-wave amplitude V4 = 1.0 mm
QRS amplitude V2 = 6.5 mm

Formula result = 19.43 (supporting LAD OMI at when greater than 18.2; The most accurate cutoff is 18.2 with about 85% sensitivity and specificity; a cutoff of 19.0 is about 97% specific, and 17.0 is about 97% sensitive)

One might think that this is an old MI (old MI with persistent STE, otherwise known as LV aneurysm), and that the patient has old scar as a nidus for ventricular dysrhythmias. But the ECG tells otherwise!! Old MI has a small T-wave; acute MI has a large T-wave. In two studies, a derivation and validation, Smith et al. showed that if the differential dx is between acute and old MI, if any one of leads V1-V4 has a T/QRS ratio greater than 0.36, then it is acute. If less than 0.36, it is old or subacute (symptoms over 6 hours). In this case, V3 has a large T/QRS ratio: 4.5/6.5 = 0.69, so this is definitely acute (or acute superimposed on old MI).

Note: ECG "LV aneurysm morphology" usually has an associated echocardiographic anatomic aneurysm, also known as "diastolic distortion." At the very least, there is akinesis.

Back to the case:

Given the history of collapse with initial EMS rhythm of VT reported, there was a high suspicion of OMI as etiology of arrest. The patient was treated with an amiodarone bolus and drip. 

Initial high sensitivity troponin I was within normal limits at 9 ng/L (upper limit of normal for males is 20 ng/L). 

Cardiology agreed to take patient for urgent catheterization.

A 100% occlusion was identified at the mid LAD and stents were deployed to the proximal and mid LAD achieving TIMI 3 flow. A 100% distal RCA calcified chronic occlusion was identified as well and was stented but was unable to achieve persistent reflow. Distal collateral retrograde flow was noted. An EF of 30-40% with probable anterolateral and inferobasal hypokinesis were noted during the cath.

Repeat troponin I post cath increased to 440 ng/L and peaked at 777 ng/L.

Echo following intervention noted concentric LVH, LVEF 40-50%, apical akinesis, and diastolic dysfunction.

Prior records from about 7 months ago were obtained including a prior ECG report (no image available) and prior stress test results. The ECG was read as having possible inferior and anteroseptal infarct of undetermined age. The stress test reported "infarction in apical and anterior walls with minimal reversibility, hypokinesis of the apical wall, and EF 42%." At the time of that evaluation the patient was offered catheterization given intermittent exertional chest pain and dyspnea at that time, but they decided on outpatient noninvasive management.

Repeat EKG on Day 2 of hospitalization:

This shows resolution of findings in the first ECG, and Q-waves in V1-V4, showing significant loss and/or stunning of myocardium, with V2 and V3 being classic for LV aneurysm morphology.

Learning Points:

Not all OMIs present as STEMIs. Complete LAD occlusion can be subtle. We have described this subtle pattern many times, including minimal STE in the right precordial leads with reciprocal STD in the lateral and inferior leads. This is only one of many recognizable patterns that come with experience.

Make sure to see the many other examples to lock in this pattern in your mind:

A man in his 30s with chest pain

In real life we treat individual patients and must make individual estimations of the probability of ACS as the cause of cardiac arrest. This case shows why it is inappropriate to conclude from the COACT trial that ALL post-arrest patients without STEMI criteria on ECG do not need emergent catheterization. The key to the benefit of reperfusion is selecting patients with OMI (those who can benefit form emergent reperfusion). The STEMI criteria have been shown to be inaccurate for detecting OMI, and thus cannot possibly be sufficient in isolation to rule out the need for emergent cath. In this case, this particular ECG pattern in addition to initial rhythm of VT led to high suspicion for acute MI as etiology of arrest and helped build case for urgent catheterization despite lack of STEMI criteria.

For more information on the COACT trial, see this post where we have already discussed its flaws in detail:

ROSC: does the ECG rule out OMI? And why does a heart just stop beating? And what rhythm is this?

If there is STE with Q-waves, especially QS waves, it cannot be normal variant STE. Use the LV aneurysm rule.

If there is STE but no Q-waves, no inferior or precordial STD, no convex ST segment, no terminal QRS distortion, no 5mm STE, no T-wave inversion, then it might be normal variant and the LAD formula can be applied. Practice with these cases:

12 Example Cases of Use of 3- and 4-variable formulas to differentiate normal STE from subtle LAD occlusion

MY Comment by KEN GRAUER, MD (6/12/2020):
Important case presented by Drs. Oberst, Mogul and Meyers — regarding the benefit of timely cardiac catheterization post-ROSC, even when frank ST segment elevation is not present.
  • Initial ECG findings are not reliable for detecting all post-arrest patients who may potentially benefit from prompt reperfusion. This is because ECG evidence of acute ischemia will not always be seen in this group of patients (Staer-Jensen et al Circ: Cardiovasc InterventVol. 8; No. 10, 2015). As a result — other parameters (ie, as in this case — sudden collapse from VT) may assist in facilitating selection of patients most likely to benefit from prompt cardiac catheterization.
  • I focus my comments on the 2 ECGs shown in this case. For clarity and ease of comparison — I’ve put these 2 tracings together in Figure-1.

Figure-1: The 2 ECGs in today’s case (See text).

My THOUGHTS on ECG #1: Drs. Oberst, Mogul and Meyers have highlighted key findings from the initial ECG that was obtained in the ED, following resuscitation of this patient by the EMS team (ECG #1 in Figure-1). I’ll “zoom in” on some of the details:
  • Looking Systematically at ECG #1: There is artifact with baseline wander in leads I and III — and for the last complex recorded in leads V4,V5,V6 — but other than this, the tracing is technically sound.
  • The rhythm is sinus at ~80/minute. The PR interval is slightly prolonged at 0.22 second. The QRS complex is narrow — and the QTc is normal. The frontal plane axis is normal at +50 degrees. There is no chamber enlargement.

Regarding Q-R-S-T Changes:
  • Re Q Waves and R Wave Progression —There are narrow Q waves in leads III and aVF. There may be slight misplacement of some precordial leads — as the QS complex in lead V1 becomes an rS complex by lead V2 — only to “lose” the initial r wave again in lead V3. A small initial r wave returns in lead V4. Lead V5 looks a bit unusual, in that the positive deflection (r wave) is much wider than what one usually sees with such a small amplitude r wave. Transition (where the R wave becomes taller than the S wave is deep) finally occurs between leads V5-to-V6. Conclusion: I suspect one or more of the anterior leads was placed too high on the chest (especially given the deep negative P wave in lead V1) — but regardless, the poor R wave progression we see in ECG #1 is consistent with prior anterior infarction (and this patient’s past medical history is remarkable for a prior “silent” heart attack).
  • Re ST-T Wave Changes — There are some nonspecific ST-T wave changes in some limb leads (ie, leads I, II, aVL) — but these do not look acute. Of much more concern (as per Drs. Oberst, Mogul and Meyers) — there is 0.5-1.0mm of J-point ST elevation in leads V1-through V5, with a similar amount of J-point ST depression in lead V6. KEY Point: The T waves in leads V3, V4, V5 and V6 are all disproportionately tall, peaked, and wider-than-they-should-be at their base, especially given the modest R wave amplitude in these leads. In the context of the acute clinical situation of this case — this fits my definition of hyperacute T waves. Prompt cardiac cath is clearly indicated based on this ECG.

COMPARING the 2 Tracings: I like to put serial ECGs together when comparing them — as this greatly facilitates seeing subtle differences (Figure-1).
  • There is much more baseline artifact in ECG #2. That said — the rhythm is still sinus, and QRS morphology in the limb leads is virtually the same. I don’t think there has been any significant change in ST-T wave morphology in the limb leads between ECG #1 and ECG #2.

In contrast — there have been some changes in the chest leads on the follow-up tracing done the day after PCI (ECG #2):
  • Note some difference in QRS amplitudes in ECG #2 — in which transition now occurs a bit sooner (between leads V4-to-V5) than it did in ECG #1. Note that instead of the negative P wave in lead V1 (that we saw in ECG #1) — the P wave in V1 is now predominantly positive. Although these details do not change my impression regarding the acute ST-T wave changes in ECG #1 — I believe they do support my suspicion that there was some lead misplacement in the initial ECG, that has now been corrected in ECG #2.
  • R wave progression remains poor in ECG #2. More than this — there is “loss of r wave” from lead V3 (which did feature a tiny-but-real initial positive deflection) to lead V4 (which appears to feature a QS complex without any r wave, albeit difficult-to-discern given all the baseline artifact).
  • KEY Point: The hyperacute T wave changes I described above in leads V3, V4, V5 and V6 for ECG #1 — have for the most part resolved in ECG #2. The T wave in lead V4 of ECG #2 remains disproportionately tall with an overly wide base — but even in this lead, T wave amplitude is less than it was in ECG #1. This confirms that T waves in these 4 chest leads of ECG #1 were indeed hyperacute.
  • Finally — there remains ~1mm of residual ST elevation in leads V1,V2 of ECG #2 — which in the context of the very poor R wave progression we see in this tracing, is consistent with possible LV aneurysm.

Two More Learning Points:
  • Point #1: Memorize in your mind the “picture” of the T waves we see in leads V3, V4 and V5 of ECG #1 (and by the principle of “neighboring leads” — the “picture” that we also see for the T wave in lead V6). In the context of an acute cardiac event (be this new-onset worrisome chest pain, or the episode of VT causing cardiac arrest that occurred in this case) — this T wave “picture” indicates hyperacute T waves. These T waves tell us this patient needs to be considered for prompt cath.
  • Point #2: The BEST way to get really good at recognizing hyperacute ST-T wave changes — is to regularly review serial tracings on your patients, that you then correlate with what happened. In this case — the fact that the T wave changes in ECG #1 almost completely resolve following PCI proves that the T waves in those 4 chest leads of ECG #1 were indeed hyperacute.

Our THANKS to Drs. Oberst, Mogul and Meyers for presenting this case.


  1. Hello sir ,
    how can we safely apply the 4 variable formula in the above mentioned case , while ST depression and Q wave are already prsent ?
    best regards.

    1. It should not be applied. One should just assume that it cannot be normal variant ST Elevation if those either of those are present. We only do so to make a point.


DEAR READER: I have loved receiving your comments, but I am no longer able to moderate them. Since the vast majority are SPAM, I need to moderate them all. Therefore, comments will rarely be published any more. So Sorry.

Recommended Resources