Wednesday, April 26, 2017

A 60-something with Syncope, LVH, and convex ST Elevation

Note 2 other similar cases at the bottom that come from my book, The ECG in Acute MI.


While I was busy seeing patients, a resident brought me this ECG of a 60-something with a history of syncope only.  There was no chest pain or SOB at the tim of the ECG:

Computerized QTc is 464 ms
A previous ECG from 8 years prior was normal.
What do you think?

There is sinus rhythm at a rate of nearly 100 (nearly tachycardic)
There is 2.5 mm STE in lead V1 and 3 mm in lead V2, with convexity, and 1.5 mm in V3.
This meets "STEMI criteria"
However, there is very high voltage, with a very deep S-wave in V2 and tall R-wave in V4.
This is massive LVH and the ST elevation is not atypical for LVH with superimposed stress.
The morphology is not right for STEMI.

My interpretation: LVH with secondary ST-T abnormalities, exaggerated by stress, not a STEMI.

Is there a formula to help with this? Maybe.  See discussion at the bottom.

I did not have more information at the time.  To the ED providers, the patient denied CP, SOB, or drug use.

My opinion was that it was not a cath lab case, but I did suggest they do a bedside ultrasound to look for an anterior wall motion abnormality.

Absence of chest pain or SOB at the time of the ECG is important; had the patient had active chest pain, I would have recommended at least an emergency formal echo, if not cath lab activation.

This is the ED bedside echo, recorded during ST elevation:

Parasternal short axis shows huge concentric LVH.  The anterior wall is contracting normally.  This is very good evidence that the ST elevation is not due to STEMI.

Notice also the very small LV cavity size at both diastole and systole (poor LV filling with good contractility) and apparent low stroke volume.

So there is poor LV filling.  This can be due to one or both of very poor compliance (which is no doubt present here, due to LVH) and to low LV filling pressures.  When there is poor compliance, as with massive LVH, the heart is particularly vulnerable to low LV filling pressures.

Low LV filling pressures are due to several etiologies, most commonly due to volume depletion (dehydration or hemorrhage), but also due to other etiologies including, but not limited to: mitral stenosis, pulmonary hypertension (chronic, or due to pulmonary embolism), or poor RV performance.

A view of the inferior vena cava showed it to be completely collapsed, confirming volume depletion.

This is just the sort of stress that can lead to ST elevation when there is this degree of LVH.

Here is the long axis view:

Massive LVH with good septal wall motion, and poor LV filling.

The RV was small and IVC empty, making pulmonary embolism extremely unlikely.  (Acute PE can also result in right precordial ST elevation due to RV stress and ischemia)

90 minutes later, I was shown this followup ECG without any knowledge of the ultrasound results:

There is significantly less ST elevation, and the ST segment is now concave.
Does this make you think that the previous ECG did indeed represent LAD occlusion, and the followup represents reperfusion?


I had not seen the cardiac ultrasounds at this time.

This repeat ECG did not convince me.  When a heart with LVH is under stress, the ST segments may rise and become convex.  When the stress is relieved, it resolves.  There was some stress associated with the syncope, whatever the cause of the syncope.

I remained unconvinced that this was due to ACS.

The history given was still that there was no chest pain and no other significant findings.

There was a positive initial troponin I at 0.115 ng/mL.

The patient was given aspirin and heparin and admitted to the hospital.

This ECG was recorded at 3 hours:
Consistent with LVH alone

This was recorded the next AM:

Further clinical data

Later physicians obtained history that  the patient had been on 3-day binge of cocaine, heroin, and marijuana.  He reportedly did complain of atypical chest pain and intermittently, including some measurements in the ED, had a very elevated blood pressure (up to 210/110) because he had not been taking his antihypertensives.

With the added history, and the entirety of the presentation, it was determined by cardiology that the clinical presentation was not due to ACS.

See similar cases below from my book, The ECG in Acute MI

New 4-Variable Formula

I have published a new formula for Early Repolarization vs. Subtle LAD Occlusion that solves the problem of false positives due to LVH by adding a 4th variable, QRS in V2.  The higher the QRS voltage in V2, the less the likelihood of LAD occlusion.  The article is published online (manuscript only at this point, corrected proofs available in a few days) in the Journal of ElectrocardiologyA New 4-Variable Formula to Differentiate Normal Variant ST Segment Elevation in V2-V4 (Early Repolarization) from Subtle Left Anterior Descending Coronary Occlusion - Adding QRS Amplitude of V2 Improves the Model.

The new formula is:  0.052*QTc-B - 0.151*QRSV2–0.268*RV4 + 1.062*STE60V3.   The cutpoint with the highest accuracy (92.0%) was at a cutoff value  of 18.2 (higher indicative of LAD occlusion).  

QTc-B is the computerized Bazett-corrected QT interval
QRSV2 is the entire QRS in millimeters in lead V2
RV4 is the R-wave amplitude in millimeters in lead V4
STE60V3 is ST elevation, relative to the PQ junction, at 60 milliseconds after the J-point, in millimeters.

This formula was derived in populations of LAD occlusion and early repolarization. Does it apply to LVH also?  I'm not sure, but I suspect it is.  That is worth study.

Let's apply the formula to the first ECG aboveWith values for QTc of 464 ms, STE60V3 of 2 mm, QRSV2 of 44 mm, and RAV4 of 35 mm:
The value = 10.3.   This is very low. 

The sheer amplitude of the R-wave in V4 and the S-wave in V2 make LAD occlusion extremely unlikely.

Here are a couple cases from my book, The ECG in Acute MI:

Case 1, from chapter on Cocaine Associated Chest Pain (CACP)

Case 2, from chapter on LVH


  1. Hello doctor, looks like a complicated formula to apply in a busy ED.. Any app/website/link to jus enter the values n get the value? ��
    Also.. If we send such a patient to cath and they are found to have occlusions(occlusions probably silent meaning not the reason fr the current complaints of the patient) how to sum this all up?

    1. I will eventually get around to making a new iPhone app and putting the formula on the web site. Always so much to do!

  2. Are you doing a validation study? Will there be a new iPhone app with the new formula?

    1. Peter,
      I do not have the data to do a validation. Maybe someone else in this wide world could do it?
      I will eventually get around to making a new iPhone app and putting the formula on the web site. Always so much to do!

  3. Brilliant as ever! Thanks for the post! I have seen this several times in the ED in cocaine OD...and AMI as a result of too.

  4. I have two important questions doctor..
    1. Can we diagnose LBBB in the presence of LVH
    2.if the answer is yes then can we use modified sgarbosa criteria in a pt with LVH with LBBB?

    1. If there is LBBB, then the modified Sgarbossa works well. Remember that lower ratios were also found in coronary occlusion, but not quite as specific. so a ratio of 0.20 is still VERY worrisome. For LVH, Armstrong found a ratio of 0.25, but I find his study not to be valid, and for LVH a much smaller ratio is worrisome.

  5. Will you be changing/updating the app with this new formula? I've been using the app more and more (I'm an ECG tech and help screen inpatient ECGs for STEMI).

  6. Am I making a mistake, not getting a qrs of 44 mm in V2 ??

    1. Hi AKS. WHICH tracing are you referring to? (There are 4 ECGs shown in this case). Are you talking about R wave, S wave or R wave + S wave QRS amplitude in a particular tracing being equal to 44mm? As soon as you can specify what it is that you are measuring — I'll be glad to reply! — :)

  7. I was referring to ecg 1. I quote the calculation made below.

    Let's apply the formula to the first ECG aboveWith values for QTc of 464 ms, STE60V3 of 2 mm, QRSV2 of 44 mm, and RAV4 of 35 mm:
    The value = 10.3. This is very low.

    1. @ AKS — Thanks for the extra information. I am NOT an expert regarding Dr. Smith's 4-variable formula — so I completely defer to HIM on this issue. Based on your question — I looked at the 1st ECG in this case and applied the following values to his 4-variable formula (QTc = 490msec; QRSV2 = 33; RV4 = 34; STE60V3 = 2.5) — which using his formula calculates a value = 14.04. This is still a LOW value, but it is higher than the value cited above. I thought the entire QRS in V2 = 33. I am sending your question to Dr. Smith for his further comment.

    2. Agree with Ken. Perhaps you did not measure QRS V2 correctly because you thought the S-wave in V3 was part of V2? It is common in LVH to not be able to find the V2 S-wave because it overlaps with V3 S-wave.

  8. Ken I say it frankly.
    I have lot to learn.

  9. Tks Ken, I also measured around that but after Dr Smith clarification, it's OK.

  10. I also face problems with 3 ad 4 variables. Any blog with labelled illustrations of these, especially in situations like this where hard to find end of qrs.

    1. @ AKS — Please CHECK OUT the November 3, 2017 post by Dr. Smith — in which he shows 12 examples using his formulas — — :)


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