Friday, April 20, 2018

OMI Can be Diagnosed by "Pseudonormalization of ST Segments"

This post was written by Tarissa Lai, one of our outstanding EM residents at Hennepin County Medical 
Center, with comments by Steve Smith and Dan Lee.


A 30 something y.o. female with HTN, HLD, diabetes, ESRD on dialysis is brought in by EMS with 
sudden onset, left -sided chest pain for the past four hours.

This is her pre-hospital ECG:

 This is her first ECG in the ED:
What do you think?

I interpreted this as normal sinus rhythm with LVH, but no significantly peaked T waves concerning 
for hyperkalemia. I did not appreciate any significant ST elevation.

However, the prehospital ECG is more worrisome: the T-wave inversion in V5 and V6 is preceded
by ST Elevation. In LVH, T-wave inversion in leads V5 and V6 should be preceded by ST depression, or
at least by an isoelectric ST segment. These look like ischemic ST segments and T-waves.

More Clinical history:
I first met her on an overnight shift while she was sleeping in a hallway bed. She was a difficult 
historian as she kept answering in one word replies before falling back asleep. The patient had gone 
to dialysis that day without any significant change in her regimen. While she was in her bed at home, 
she had sudden onset of left sided chest pain that radiated to her shoulder.  The pain
was pleuritic, without nausea or diaphoresis. The nitro she took in the ambulance did not help. 
Her physical exam was remarkable for a young woman sleeping comfortably in bed, whose chest pain
was easily reproducible with palpation.

Although her comfortable appearance with pleuritic and reproducible chest pain were reassuring, some 
components of her story were concerning, including the sudden left sided chest pain and radiation to 
her shoulder. Given her risk factors (HTN, HLD, ESRD from diabetes) I decided to obtain a broad 
cardiac workup for the patient: serial ECGs, labs, serial troponins, CXR and bedside cardiac ultrasound. 
She also received an additional nitro in the ED after receiving aspirin and nitro via EMS.

This is the patient's old ECG from ~2 months prior:
There is ST elevation that is most notable in V1, V2 and aVR with ST depression that is most obvious 
in the inferior leads and V6. I interpreted these as findings consistent with "secondary" ST-T abnormalities due to
her known LVH.
When comparing the old and new ECGs, there is:
New relative ST depression in V1-V3 and
New relative ST Elevation in: II, III, aVF and in V5 and V6.
When I say "relative," I mean relative to the previous ECG, which is the baseline, chronic, non-ischemic ECG.

Here is the Point of Care Cardiac Ultrasound (POCUS), short axis:

This shows the anterior wall (top) and septum contracting perfectly, but the postero-lateral walls not contracting.

Parasternal Long Axis View

There is a posterolateral wall motion abnormality. This appears to be new, as her last formal 
echocardiogram 2 years ago was relatively normal.

All of this is consistent with an acute postero-lateral MI. [New relative ST elevation in V5 and V6 (lateral
wall), new relative ST depression in V1-V3 (posterior wall) and new relative ST Elevation in II, III, aVF
(inferior wall).

At this time, her initial troponin came back at 103.2 ng/mL (103,000 ng/L, extremely high!!),
and her K was 4.6 mEq/L.

While the patient was known to have mild troponin elevations in the past, this was clearly diagnostic of acute MI.

Given the WMA on her echo, a repeat ECG was obtained with posterior lead placement:
There is now STE in V7, V8 and V9, which is consistent with a posterior MI.
The decreased voltage noted in the first couple ECGs is even more obvious now.
There is now also ST depression in V2 and V3 (previously only relative)
On closer evaluation of her old ECG compared to the ones from today, you can see what might
be called "ST Segment Pseudonormalization." Her previous ST elevation in V1-V3 is now gone,
and the voltage from her LVH is decreasing with each successive ECG. This would make sense in
the setting of an acute posterior MI, which can present with precordial ST depression.

V1-V3 side by side:


The patient was emergently taken to the cath lab and found to have severe multi-vessel disease with 
high thrombus burden, and the culprit lesion was a 100% thrombotic occlusion in the proximal
left circumflex.

Diagnostic angiogram:

Post PCI of the circumflex artery. She had an uncrossable lesion in the LAD with an estimated 99% 
occlusion. Arrowheads represent some of the remaining thrombotic lesions.

Troponin peaked around 325 ng/mL (massive). Pertinent results of the formal Transthoracic Echo:
    -Moderately severe concentric left ventricular hypertrophy.
    -Severely decreased left ventricular systolic function with an estimated EF of 22%.
    -Akinesis of the apical septal, apical inferior, and apical segments
    -Hypokinesis of the basal to mid anterolateral and basal to mid inferolateral segments. (Basal
    Inferolateral = Posterior)

    Learning Points:
    1. Pseudonormalization of STE and/or STD, as in this patient with LVH, but also in LBBB and other
    etiologies of chronic ST shift, should raise concern for OMI (Occlusion Myocardial Infarction).
    2. Likewise, ST shift from a previous ECG (relative ST elevation or depression) is equivalent to STE or STD.
    3. Ultrasounds can be very helpful in guiding your diagnostic pathway: location of WMA on US led 
    to obtaining posterior leads.
    4. Clinical presentation is important, but so is history. Have a higher suspicion for true pathology in 
    anyone with significant comorbidities such as ESRD from diabetes, even if they are young and 
    appear “not sick” on presentation.
    5. Refractory Chest pain with a clearly positive troponin, without an alternative explanation, is an indication
    for emergent cath lab activation regardless of ECG or bedside echo findings.

    Comments by Dan

    This case was particularly interesting for several reasons. First, the patient is such a young woman to have an occlusion MI (OMI), the likely difference being her significant risk factors. If she had no risk factors, it is doubtful that she would have developed such extensive coronary artery disease as we see on the angiogram. 

    Because the troponins were so high, the clinicians did not need to rely on other diagnostics, including posterior leads and ultrasound, to confirm OMI and salvage what myocardium could be saved after hours of constant chest pain before the patient presented.  However, this additional information was supportive.

    Note that by current guidelines, this patient had a NSTEMI, but it would hard to argue that her condition would not have worsened if she was left to 'next day cath'.  I took part in her ICU care and she was extubated and stable to transfer to a stepdown unit after a few days. Her repeat ECHO showed an improving EF of 37%. Patients like her are the reason we are advocating for a change in the ACS paradigm from STEMI to OMI

    Her first EKG in isolation has no hard findings that are diagnostic for an acute coronary occlusion. The importance of having a baseline EKG for comparison cannot be understated. The old EKG shows LVH by Sokolow-Lyon and Cornell criteria but much of that voltage is resolved in the new EKG, which is commonly seen in acute occlusion superimposed onto LVH. For some reason (with debatable physiology), coronary occlusion often causes decrease in the high voltage of LVH on the EKG. Thus it is difficult to study occlusions, particularly subtle occlusions, in the context of LVH. Armstrong et al attempted to study it but may have included too many 'obvious' cases - the criteria from that paper would certainly have missed this case.

    There is an interesting finding in this EKG which is the positive notching of the latter portion of the QRS complex, upright in I, II, aVL and downward notching in III, aVF.  

    I believe these represent the equivalent of Q-waves, or infarcted myocardium, of the posterior (inferobasal, lateral) wall. They appear in the latter part of the QRS complex because the posterior/basal wall is the last part of the ventricle to depolarize. The notching is not seen in the precordial leads because as it is traditionally taught, a tall R wave in V1 represents the 'upside down' Q wave in posterior infarction.  Thus, if you have LVH in which baseline precordial voltages are already predominantly negative, a posterior infarction should 'cancel out' some of that voltage and possibly diminish the voltage evidence of LVH on the EKG. Note that this is completely different from 'terminal QRS distortion'

    This paper by Nui et al describes this finding (the "Delayed Activation Wave") in left circumflex occlusions but (I believe mistakenly) compares it to terminal QRS distortion. This finding is not new, and is analogous to Cabrera's and Chapman's sign.  As a Q-wave equivalent, this notching would be expected to persist in subsequent EKGs, as seen in the patient's follow up EKG 6 days later:


    1. Excellent case & discussion by Drs. Lai, Lee & Smith! I would add to the Learning Points that there was also “pseudonormalization” of QRS amplitude in serial tracings here, due to loss and/or cancellation of QRS forces from evolving infarction. Two additional comments I would make: i) The ST segment coving with symmetric T wave inversion seen in leads V5,V6 for the first 2 tracings is not normal. In a patient with symptoms — this finding should suggest ischemia and/or possible acute injury — and ii) The mean frontal plane axis for the first 2 tracings compared to the prior ECG from 2 months earlier is dramatically different (ie, the isoelectric QRS in lead II from those first 2 tracings was all positive in the earlier ECG). As a result — CAUTION is needed in comparing ST-T wave changes from the current tracings compared to the earlier ECG! That said, I agree there are still enough differences between the 3 tracings to support the conclusions in this case. THANKS for presenting!


    Recommended Resources