Thursday, August 16, 2012

Spontaneous Reperfusion and Re-occlusion - My Bad Thinking Contributes to a Death

This is a case I had about a decade ago.

This is a 51 year old who was playing cards with his friends when he started to have left hand numbness.  They were worried he was having a stroke and so called 911.  The medics had just learned to do ECGs, and so recorded one.  Here it is, at 1915:

Diagnostic of Anterior STEMI.  It cannot be anything else.

I activated the cath lab at 1929 based on this ECG.

Then, I questioned the patient at length and his only symptom was subjective left hand paresthesias.  He had no pain, discomfort, or tightness of any kind, no weakness, and no SOB.  So I had a hard time believing the ECG.  I thought perhaps it was recorded with lead misplacement.

So I did the first ED ECG at 1931:
STE is resolved (but there are de Winter's T-waves in V2 and V3 (hyperacute T-wave with depressed ST takeoff)        
Previously published in:  Harrigan (Ed.). The ECG in Emergency Medicine.  
Smith SW and Whitwam W.  The ECG in Acute Coronary Syndromes.   
EM Clinics of N Am 24(1):53-89; Feb 2006]

With no more overt STEMI, and (through bad thinking and "Nah, couldn't be.....") I thought that there must have been some mistake in recording the first ECG.  At worst, if it was a STEMI, I thought that it is reperfused.  I cancelled the cath lab activation for the team that would have to come in from home.   

(Today I would have unequivocally interpreted leads V2 and V3 as LAD occlusion).

At 1942, the patient started becoming hypotensive, so I recorded another ECG at 1946:
Need I say more?  Obvious anterior STEMI.
Previously published in:  Harrigan (Ed.). 
The ECG in Emergency Medicine.  
Smith SW and Whitwam W. The ECG in Acute Coronary Syndromes.   
EM Clinics of N Am 24(1):53-89; Feb 2006

I activated the cath lab again at 1946, so that I had caused a 17 minute delay by cancelling.

He went to the cath lab, had an LAD occlusion, then died just before it could be opened.

I learned 2 major lessons from this:

1. STEMI, even if it spontaneously resolves, is very high risk and must go to the cath lab.
2. A clearly diagnostic ECG is diagnostic even if it does not match the symptoms.
  (One cannot make the same conclusion about ECGs that are only highly suspicious - these are more likely to be false positives in the context of atypical symptoms.)


  1. That's a rough case, and it sounds like it certainly affected the culture at your shop.

    Given a similar case recently, I was doing some reading. It looks like a certain amount of the literature suggests that spontaneous-resolution of a STEMI prior to cath is a very good prognostic sign. Bainey 2008 (, found a big drop in mortality if the ST segments resolved by ≥ 70%.
    Similar results have come out of Israel ( and (

    Bottom line, though, I agree: I'm waking up the interventionalist early next time!

  2. Yes, it is a good prognostic sign compared to STEMI that do not reperfuse, but any ischemic ST elevation is a bad prognostic sign (compared to none)!

  3. Steve- Gosh im sorry, it really hurts and all ER doctors go through this, of blaming themselves thinking they could have done more. But as you know hindsight is 20/20. The important questions to ask are:

    1) are your cardiologists onboard with the the STEMI equivalents? and have you discussed with them taking "stable" ( as your patient appeared) pts with a stemi equivalent and only ekg findings of a stemi equivalent to the cath lab? Your patient was pain free and stable and presented with a stemi equivalent and im not sure many cardiologist would be too pleased to rush to the cath lab.

    2) what is the Likelihood ratio for De winter's waves and your pretest prob in this patient? This can help you determine your post test probability and then you can see if you over thought this. Likely your prestest prob was low...given your patients vague complaints and stable presentation.

    1. Thanks, I appreciate the note. Fortunately, I don't dwell on these things very long. I just admit the mistake and learn from it. Blaming oneself and feeling bad, though unavoidable to some degree, doesn't change anything.

      Our interventionalists come in when we activate the cath lab and generally don't question it if it's reasonable.

      If there had only been the de Winter's T-waves, without the prehospital ECG, and without any symptoms, I might not activate the cath lab but would do serial ECGs, stat echo, etc. But in this case there WAS the prehospital ECG, and was diagnostic in itself, then especially so with the next one having de Winter's T-waves.

  4. Tell me about EKG #2. #1 is so terrible, I feel like looking at #2 in isolation is difficult. The hyperacute T's anteriorly: are you adding the flat/flipped T in AVL or the inferior Q's to your "I would have called it STEMI" math, or is it the J point on those leads being depressed that makes you so sure?

    1. There is slight ST depression in leads V2 and V3, with almost no R-wave and large fat T-waves. These are typical of de Winter's T-waves. See several cases by searching for winter:

  5. thanks for sharing, even if it is what you describe as your "bad thinking." that is admirable.

    one of the STEMIs that stuck out for me was from a few years ago. it's like you wrote - not your typical symptoms, no CP/SOB, nothing else. the patient's only complaint: L triceps pain, very specific and very localized. the only reason we ran a 12-lead was that he was sweaty and did not "look right."

    his 12-lead was clear as day - inferior STE, lateral STD.

  6. at the time, what specifically was the mistake you thought occurred with the first 12-lead? is there misplacement that causes ST-T morphology such as those seen in that tracing?

    (i'm asking to learn, not to judge.)

    1. Tom,

      There is a lot of literature on lead misplacement causing differences in RS amplitude, but not a lot on ST segments, though it stands to reason that if it changes one kind of ammplitude, it should change others. However, I would expect such changes to be proportional. Thus, in this case, the fact that ST amplitudes were very large without correspondingly large RS amplitudes, argues against any lead misplacement. I caution that this is just my reasoning and not supported by any other literature or data.

      I think I just let myself look for an excuse for the ST Elevation in someone who had no symptoms, and knowing that the medics had only recently learned to record ECGs.


    2. Here is a good reference:

    My question about Lytics in case of unavailability of Cath-Lab. Would u give lytics in such case where the STE already resolved?

    1. In this case, yes, because 1. there are no symptoms to guide you in this case ("hand numbness????"). AND

      2. the ECG actually shows de Winter's waves.

      In most cases, I would not give lytics to someone with resolution of the STEMI. However, I would add eptifibatide to the medical treatment, in addition to aspirin, clopidogrel (or ticagrelor), heparin.

  8. Thank you for sharing this case. very educational.

  9. I'm curious how you go about handling these cases when patient symptoms are so very non-typical. I've seen enough Takotsubo cardiomyopathy masquerading as an "obvious" acute anterior STEMI in an individual with a very poor story, that I think caution is warranted prior to sounding the STEMI alarm bells in these scenarios. We actually had 2 M&Ms in our program in the last 3 months of individuals with ICH who had stories, symptoms, and physical exam that all pointed to ICH, but an ECG that demonstrated a "diagnostic" anterior STEMI who received antiplatelet therapy and anticoagulation prior to cath, only to find the "octopus trap" on angiography, and a big, bright bleed on their subsequent CT head assuredly made worse by our overzealousness to get these patients to cath lab.

    In my opinion, the story still matters. While isolated hand numbness is probably not enough to prevent me from immediately activating cath lab, severe headaches, severe altered mental status, GI bleeds, will definitely make me reconsider my priorities in management.

    1. There is a very simple solution to that: get a head CT if you're worried. Takes minutes only. But not for this case: head bleeds do not present with hand numbness only. As you say, in your cases the story and exam fit for ICH. This patient's story fits for either ischemic (not hemorrhagic) stroke or acute MI. The ECG made it clear which it was. Patients with Takotsubo do not sit there quietly with hand numbness. They are always in some sort of distress.

  10. Dr. Smith, what I learned from this case is that De Winter is not only an early sign of acute occlusion... It is also an early sign of reperfusion, isn't it? And then it will turn into Wellen, right?

    Anyway, thank you for sharing this valuable experience.

    1. It is an intermediate stage between subendocardial ischemia and STEMI, in which there is enough flow that there is no ST Elevation (and instead there is ST depression of subendocardial ischemia), but so little flow that there are hyperacute T-waves. Wellens' is a stage after complete or near complete occlusion when there has been full reperfusion. So de Winter is only a sign of minimal reperfusion, and only if there is previous evidence of full occlusion, as in this case. Wellens' is a sign of reperfusion which could have been preceded by either ST elevation, de Winter's, or both (but in true Wellens' none of these ever get recorded - you only see the aftermath)


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