Friday, August 7, 2015

Is this Saddleback a STEMI??

I saw this Middle-aged patient with no previous cardiac disease who presented with worrisome substernal chest pain:
Sinus rhythm with saddleback ST elevation in V2.
There is significant ST elevation here.  QTc is 416 ms.
There is an upwardly convex ST segment in V3 and V4, which is worrisome.
But it really doesn't look very much like a STEMI, does it?
It does not look like early repolarization.
There is high voltage (large S-waves and R-waves) and it could all be LVH.
Notice there are also prominent U-waves.

In all my years, I have only reviewed one ECG with a Saddleback ST elevation that was due to STEMI.

Click on this link to see many examples of false positive ST elevation from Saddleback.

A recent previous ECG was available:
Looks like early repol except for the Q-wave in V2.  QTc is 388 ms.  All ST segments are concave upward.

Here are the precordial leads side by side:
It still didn't look like STEMI to me but it was too different to ignore.  

U-waves could be the source of some of the difference, so I checked a K.  It was 4.5 mEq/L.

So I activated the cath lab.

While waiting for the cath lab, we did a bedside echo (not shown).  We did not see an anterior wall motion abnormality.

I started to get more doubtful.

I ordered a second ECG:
This looks much more like the old one.  There is no saddleback.  There is less ST elevation, and less upward concavity

Knowing that a saddleback can be caused by lead placement too high on the chest (see this paper), I asked the tech if she had changed the lead placement.

She had not.

However, she said she had recorded the first one fully supine.  ECGs should be recorded with the patient at a 30-45 degree angle, if possible.  (see this reference:  The second one had been recorded in the 30 degree upright position.

I asked her to record another one with the patient supine:
This again looks a lot like the first one. 

Was this pseudoSTEMI due to artifact due to positioning?   Possible, but we do record supine ECGs frequently in patients who are too ill to sit up and they don't look like this.

Now I was quite certain that this would be a false positive cath lab activation.  But it was too late to turn that ship around.  The cath team was ready and he underwent an angiogram.

The interventionalist, who was not at all purturbed, said that the coronaries were "huge pipes" and totally clean.

Here is his formal ultrasound result:

Hyperdynamic systolic performance; EF 75-80%.
No regional wall motion abnormality.
Left ventricular hypertrophy-concentric, moderate.

Learning Points:

1. Saddleback ST Elevation is rarely due to STEMI.
2. Make sure the ECG is recorded with good patient and lead position.
3. LVH has a variety of PseudoSTEMI patterns.
4. If there is high precordial QRS voltage, either large S-waves or R-waves, anterior STEMI is unlikely.


  1. As a former interventional cardiologist, I think if we're not sending some patients with normal coronaries to the cath lab for scenarios like this one, we're not sending enough patients to the cath lab. Thank you for all of your good work-I learn something every week!

  2. Thanks so much for your comment! I totally agree.
    Steve Smith

  3. Love your work! Can you provide link/rationale for 30 degree angle 12 lead ECG acquisition please

    1. It is just the convention. Actually, this paper says 45 degrees. So 30-45 is ok.

      Steve Smith

    2. Thanks, have always been taught ECG should be acquired in supine position!

    3. But semi-recumbent (30-45 degrees) is also supine. Those who recommend supine are not being specific enough.

  4. I would have been very suspicious of V3-4 and certainly dicussed it with the cardiologist, but in my hospital the echo result would likely have stopped the cath lab and waited for serial ECGs and Troponin. However if the symptoms were convincing I can't be sure. What was the blood pressure? It all depends on local protocol, lab capacity etc.

    1. Charles,
      BP was normal. If I had been more circumspect, I would have done some more investigation first (e.g. bedside echo and serial ECGs before activation).

  5. NICE post! I'd add a #5 to your Learning Points (as you and I have corresponded about) — namely, that patient positioning ( = angle of the bed) should ideally be recorded. I know from stress testing baselines (that we did for years in supine, then sitting for brief hyperventilation — then standing prior to starting the treadmill) — that in some patients (hard to predict in whom) there could be significant change in QRST morphology — but not in others. Sounds like from our brief informal surveying, that there is lack of convention at what angle the bed should be set at — so given that numerous different providers in any given institution might be called on to obtain a 12-lead ECG at any given time — it would seem optimal procedure would be simple notation on the tracing at the angle of the bed at which the tracing was done ... THANKS again Steve for this insightful case — :)

  6. What could have accounted for the 'worrisome' CP ? Was it due exclusively to the LVH ?

    Thanks for the link to the PDF document concerning patient positioning during ECG examination. I pleasantly spotted David Richley's name among the authors, he's quite a quality contributor to the FOAMed movement.


    1. Olivier,
      We evaluate 1000 chest pain patients in the U.S. for every 100 who get diagnosed with ACS. Worrisome chest pain is all around us, but much of it never gets diagnosed. GI reflux?? Who knows?

  7. brilliant case, brilliant use of serial ecgs and meticulous hx,

    was gonna ask if she had abnormal cardiac anatomy but u got that too - Left ventricular hypertrophy-concentric, moderate

    did you find the cause of her chest pain? not standard IHD and was the pain positional as well?

    1. Thanks! Pain was not positional and we never found the etiology


Recommended Resources