Saturday, March 26, 2016

"Shark fin" ECG in I, aVL, V4 and V5. Which artery? Hint: patient is in shock and was put on ECMO

This was contributed by Rohin Francis (Twitter: @MedCrisis), a cardiologist from England and FOAM enthusiast.


A 55 year old lady initially presented to hospital with an acutely ischaemic arm. An embolic occlusion of her brachial artery was diagnosed by CT and treated with anticoagulation. The following day she developed sudden severe chest pain. This ECG was obtained:
Sinus rhythm.
The rather alarming appearance of the QRST may be mistaken for a broad complex QRS but,  in fact, her QRS complex can be clearly seen in V2, V3 and II and is narrow. What has manifested as triangular complexes is actually huge ST segment elevation seen in V2-5 and laterally in I and aVL. There is also ST segment depression inferiorly.  A proximal LAD occlusion can produce anterolateral ST elevation, but if the circumflex is also occluded, it is possible the anterior ST elevation might be attenuated and indeed here the ST elevation is less pronounced in V2 and V3.

Smith comment: Note there is ST depression in aVR, a true sign of Left Main occlusion.  Many authors state that ST elevation in aVR is a good ECG sign of left main occlusion.  This is erroneous.  (See this post for more explanation).

ST elevation in aVR is a sign of left main or 3 vessel insufficiency (not occlusion), due to diffuse subendocardial ischemia, which results in leftward and anterior ST depression (towards lead II and V5) and reciprocal ST elevation in aVR.
Left main occlusion is identical to simultaneous occlusion of the Proximal LAD (anterolateral STE: STE in V2-V6, I, aVL) and circumflex (posterolateral STE: STE in aVL, STD in V1-V4).  The opposing ST vectors of anterior and posterior can (at least partly) cancel each other out in V1-V4, so that the lateral has the most obvious STE, as in this case.  

Lead aVR is opposite an imaginary lead between leads I and II, often called (-aVR).  You can see that left main occlusion causes high lateral STE (aVL) with reciprocal STD (leads II and III); since aVR is reciprocal to this ST elevation, there is ST depression in lead aVR.

Case continued

She was immediately transferred to the primary PCI centre. On arrival she was in cardiogenic shock and haemodynamically unstable. Her BP was 80 systolic, GCS 13 and she had a grey appearance.

An intra-aortic balloon pump was inserted first and the cardiac arrest and ECMO teams were alerted. This was the initial shot of her angiogram:
The guide catheter was deliberately not engaged with the left main artery, as an occlusion was suspected. Here one can see contrast being injected towards the left main and a large thrombus sitting at the ostium. A small amount of flow is passing but not reaching further than proximal circumflex nor LAD. There is sluggish clearance of contrast from the aortic root indicative of poor cardiac output. An intra-aortic balloon pump can be seen inflating and deflating in the descending aorta.Smith comment: There is near total obstruction of the Left Main.  I am no angiographer, but I believe that this is TIMI-1 flow, which results in ST elevation.  It is not TIMI-0 flow (total obstruction/occlusion), as some minimal contrast gets through.   All thrombi are dynamic, however, and the only certainty one can have about the condition of the artery 45 minutes earlier, during recording of the ECG, comes from ECG analysis.

Passing angioplasty wires into her coronary arteries restored flow to the circumflex but not LAD. Her left main and LAD were opened using very careful aspiration and she received a drug eluting stent to her proximal left main.

Here is the aspirated thrombus:
Large red thrombus aspirated

Despite revascularisation and inotropic support, she remained in cardiogenic shock. An on-table echocardiogram revealed severe global impairment of her left ventricle. She was shocked out of VF on several occasions but continued to arrest so was placed onto femoral VA ECMO.

Peripheral ECMO increases afterload by introducing blood into the aorta and if there is no cardiac ejection, it can cause dilatation of the LV. Unfortunately despite maximal therapy in ICU, she continued to deteriorate. 

A trans-oesophageal echocardiogram revealed extensive thrombus in the aortic root and left ventricle. She died later that day. A post mortem was not performed but disparate embolic events in her arm and coronary circulation might point to an intra-cardiac source of thrombus. Thrombi tend to form in impaired ventricles, but it is unclear if the severe left failure seen on echo was pre-existing or simply a result of a catastrophic acute coronary syndrome.


  1. Very informative thanks for posting this.

  2. To my knowledge If the ST depression vector is inferior and leftward,there should be ST elevation in aVR ? any explanation

    1. Just look at it. It is the ST elevation vector that is leftward, though superior. But certainly not rightward. Thus, there is ST Depression in aVR.

  3. Fascinating about aVR, I did not know that.

    Do you know the rough sensitivity of aVR ST-depression in left main occlusion?


  4. The patient was a young woman. After the arm thrombosis, she was sent to home?? She was admitted in the hospital for study? I think un accutely ischemic arm has to be treated into the hospital...

  5. Sad case in which the patient unfortunately could not be salvaged despite prompt maximal therapy. But it does provide insight into ECG recognition of the uncommon situation in which a patient with acute LMain occlusion is still alive to receive treatment in a care facility.

    Excellent explanation by Dr. Smith as to why when LMain occlusion IS seen, the relative amount of ST elevation in anterior leads V2,V3 may be less than in V4,V5. Remarkable on this tracing is the huge amount of ST elevation in the high lateral leads (I,aVL) — a finding which suggests left circumflex blockage, as well as explaining why lead aVR shows ST depression rather than elevation (aVR being the reciprocal lead to events viewed at location between leads I and II).

    QUESTION: Is ST depression in lead aVR a “requirement” for diagnosis of acute LMain occlusion? It would seem that depending on the relative amount of high lateral ST elevation vs the amount of lead II ST depression — that almost any picture may be seen for the ST-T wave in lead aVR with acute LMain occlusion (ie, flat ST; ST depression as seen here; or slight ST elevation) — depending on whether the ST depression in lead II or the ST elevation in lead I predominates.

    THANKS again to Dr. Smith for key insights from this important case!

    1. Ken,
      Exactly. Very variable. That is how I answered the above questioner.
      aVR does not really help in diagnosis of LM occlusion.

  6. Dr Smith,

    for some years many authors have confused LMCA-Occlusion versus LMCA-Stenosis: ok (and thank you for tis "battle").

    - STE in aVR looked "a common point" for LMCA-Occlusion (Septum-Ischemia, etc.) and LMCA-Stenosis (reciprocal to

    widespread STD ... apical-lateral)

    - now for you LMCA-Occlusion "corresponds" to STD in aVR + "ECG-Combination" between LAD and LCX-Occlusion

    - it is really so for you or only 1 example and 1 (new) hypothesis ?

    - STE in aVR is also typical for prox.(1.septal) LAD-Occlusion: concomitant LCX-Occlusion is able to

    counterbalance this STE ?

    - this time, after a lots of opinions, it is also "academic" (for everyone) important !

    merci beaucoup comme toujours, Al

    1. Al,
      I did not mean to imply that this is a universal, or even common, finding. See above two answers. aVR is variable depending on too many factors.


DEAR READER: We welcome your Comments! Unfortunately — due to a recent marked increase in SPAM — we have had to restrict commenting to Users with a GOOGLE Account. If you do not yet have a Google account — it should not take long to register. Comments give US feedback on how well Dr. Smith’s ECG Blog is addressing your needs — and they help to clarify concepts of interest to all readers. THANK YOU for your continued support!

Recommended Resources