Saturday, May 19, 2018

A middle aged man with ST depression and a narrow window of opportunity

Written by Pendell Meyers

I received a text at 18:13 of an ECG taken several minutes prior, with no clinical information and only the question "De Winters?"

Here is the ECG:

What would you tell the treating team???

I responded at 18:14 PM:

"I think it's posterior STEMI (OMI) instead of de Winter. Cath lab immediately is indicated."

I clarified further:

"De Winter would need hyperacute T waves (not present here), and would indicate acute occlusion of the territory in the affected leads; so if there was de Winter in anterior leads, that would mean the anterior wall is the one involved. Here we have isolated posterior STD, with no hyperacute T wave, so that's OMI of the posterior wall. Posterior wall may have hyperacute T's if posterior ECG is recorded."

Let's go back in time and see the full case play out.

A middle aged man with HTN, DM, and CAD (with two prior stents) presented for chest pain, shortness of breath, and palpitations that started several hours ago (2-3 hours) while walking his dog. He was triaged at 17:34, had normal vital signs except tachycardia, was not in cardiogenic shock, and had this ECG obtained:

There is atrial fibrillation with rapid ventricular response at about 150bpm. There is massive ST depression in leads V2-V3, with smaller amounts of STD in V4-5, I, II, III, and aVF, with obligatory reciprocal STE in aVR. The J point in V6 is isoelectric (or perhaps even a little elevated), which would be unusual in the case of widespread supply/demand mismatch ischemia because there would normally also be STD in V6; the fact that V6 is isoelectric implies that there is relative STE in this lead.

When there is rapid AF and diffuse STD with elevation in aVR, the differential does include rate-related demand ischemia (supply/demand mismatch), as well as non-occlusive ACS in the setting of three vessel disease or left main disease, as well as OMI.

However, the fact that the STD is so much greater in V2-V3 than the other leads with STD suggests that it is in fact primary STD (posterior elevation) with superimposed widespread STD from supply/demand mismatch in the setting of rapid AF. Additionally, STD in V1 is almost never present in the case of diffuse STD from global supply/demand mismatch ischemia, whereas STD in V1 is commonly present in posterior OMI.

Also, remember that the rule of thumb "STEMI (or OMI) does not produce tachycardia unless the patient is in cardiogenic shock" does not apply to patients who have an arrhythmia which bypasses the normal physiologic determiners of heart rate, such as atrial fibrillation or flutter. Any patient with underlying AF who has an acute severe illness of almost any etiology may have rapid ventricular response due to catecholamine surge or other responses to illness. So this rule of thumb does not apply to our patient in this case.

The treating team was concerned for OMI vs. rapid AF with rate related ischemia, so they very appropriately administered aspirin and IV rate controlling medications over approximately 20 minutes and collected repeat ECGs.

At 18:08, the rhythm changed and the rate decreased to about 110bpm, but the patient complained of ongoing chest pain and dyspnea. Here was the ECG at that time:

The rhythm is not entirely clear because there are not definite P-waves, but it is certainly regular and therefore not atrial fibrillation. Now the diffuse STD is resolved, leaving very focal STD from V1-V5, maximal in V2-V3 consistent with classic, obvious posterior STEMI (a very obvious case of OMI).

The treating physicians saw the focal STD in the anterior leads and were considering posterior STEMI vs. possible de Winter's pattern. They texted me at 18:13 and my opinion was posterior STEMI.

At 18:15 a Code STEMI was called, and the cardiology team responded immediately. Unfortunately they believed that the STD was more likely to be caused by rate-related ischemia from rapid AF. They advised the ED team to give nitrates and cardizem drip.

Because they were not convinced, the ED team performed a posterior EKG while the cardiologists were at beside, approximately 15 minutes after the last ECG:

It appears that leads V2 through V6 have all been moved to a posterior location, though it is unclear exactly where they were placed on the chest, or which leads are supposed to represent V7-V9. Regardless, it is irrelevant because ALL leads show diagnostic STE, confirming posterior STEMI. 

Somehow the cardiologists were still not impressed by this posterior ECG. They cancelled the Code STEMI and asked the ED team to administer nitro drip and let them know what the troponin shows.

At 18:41, the first troponin T (drawn at 17:40) returned significantly elevated at 0.44 ng/mL. The patient had ongoing pain.

Code STEMI was called a second time.

The cardiology team responded and this time agreed to take the patient to the lab. He had a delay of 87 minutes from the first, appropriate Code STEMI (18:15) to arrival in the cath lab at 19:42.

Here is what they found:

Normal RCA.
The left main coronary artery branches into a very small LAD (the vessel going vertically down the image) which has a 50% ostial stenosis, and a LCX which is 100% occluded at the ostium.

Arrows at the site of LCX occlusion.

Another view, showing the relatively small LAD in the upper half of the image, and the empty territory of the occluded LCX in  the lower half.

Arrows show the site of LCX occlusion.

A wire has crossed the ostial LCX lesion and you can now see the large vessels distal to the occlusion.

Arrows show the site of the (prior) LCX occlusion.

The epicardial vessels are now open, revealing an enormous territory supplied by the occluded LCX. As you can see, an Impella (cardiac output assist device) has also been placed, as the patient has gone into cardiogenic shock on the table.

Arrows highlight the territory that had been occluded.

 The patient became progressively more dyspneic, hypoxic, and hypotensive during the procedure, despite opening the artery as shown above. An Impella was placed for assisting cardiac output, and the patient was intubated.

Remember, the angiographic result does not ensure that the actual downstream myocardium is receiving blood supply. Only the clinical symptoms and ECG can show whether there is true reperfusion on a cellular level.

So what do you expect to see on his repeat ECG? You are looking to see if he shows signs of reperfusion vs. "No Reflow Phenomenon" (in which the ECG changes progress as if there was no reperfusion at all, because there is either no reperfusion at the level of the cells, or reperfusion was too late and the infarct is already irreversible). See the diagram below for the patterns of reperfusion vs. continued occlusion.

 Here is the patient's post-intervention ECG:

What do you make of this? Why are the anterior T-waves so big now?

This ECG shows posterolateral reperfusion. The large T-waves in V1-V3 are reciprocal to massive negative reperfusion T-waves in the posterior leads (remember: the diagram above assumes you are looking at leads directly over the site of the infarct). The inverted T-wave in V6 and I is indicative of lateral reperfusion. So this ECG is evidence that the infarct was not yet complete at the time of cath, and that there was truly successful reperfusion on a cellular level as well as the angiographic level.

Despite reperfusion, the patients troponin T peaked at over 32 ng/mL at just under 24 hours from presentation (extremely high troponin, indicative of enormously large territory of infarction). It is impossible to convert this directly to troponin I, however our experience suggests a roughly 10:1 conversion between troponin I:T, so for those of you using contemporary troponin I assays, this patient would be predicted to have a troponin I of over 300 ng/mL.

Unfortunately the patient's course was complicated by acute renal failure requiring dialysis, and the patient ultimately passed away 7 days later of a combination of complications.

It is plausible that he may have had a better outcome if his duration of acute coronary occlusion had been reduced, but we can't know for sure. But we can make sure to learn from his case and deliver reperfusion therapy as rapidly as possible to those with diagnostic ECGs.

Learning Points:

You must advocated for your patients with OMI, because the STEMI guidelines and some current practice patterns do not. Even though this particular case does have STD diagnostic of "posterior STEMI," this is not actually recognized formally as an entity in our current ACC/AHA 2013 STEMI guidelines, despite the fact that it is recognized in other ACC/AHA documents). There are no formal recommendations for posterior STEMI in the 2013 STEMI guidelines, not even millimeter thresholds for STE in V7-V9 as are given in other documents.

Posterior OMI may manifest on the classic 12-lead ECG as STD proportionally maximal in leads V2-V4.

Diffuse supply/demand mismatch ischemia, such as during atrial fibrillation with rapid ventricular response, may manifest widespread STD, but this will usually be proportionally maximal in V4-V6. Additionally, I have never seen a case of widespread STD from supply/demand mismatch with STD in lead V1 (whereas V1 is involved in posterior STEMI). Changes due to diffuse supply/demand mismatch without ACS should resolve within 10-30 minutes of resolution of the condition causing abnormally increased demand.

A delay of even just 1 hour may have been the difference between life and death in some cases such as this. Whether the patient meets STEMI criteria is irrelevant; what matters is whether the patient has an acutely occluded coronary artery that could be opened emergently in order to improve the outcome of acute MI.

The diagnosis of OMI (or STEMI) does not rely on troponin, and should be made based on clinical findings and the ECG if possible. Furthermore, troponin T level of 0.44 ng/mL does not differentiate between early-mid OMI and supply/demand mismatch from rapid atrial fibrillation with underlying structural heart disease. Our troponin assay, for example, usually does not even start to elevate from zero until at least 2-4 hours after onset of acute coronary occlusion. This period of time is in fact the most valuable for the patient, as they have the most at risk but salvageable myocardium. The whole idea of "STEMI" or "OMI" is to prevent the cells affected by acute coronary occlusion from becoming measurable troponin if possible.

The ECG predicts reperfusion on a cellular level better than the angiogram, possibly even better than the patient's symptoms. Without understanding the progression of ECG findings in continued acute occlusion vs. reperfusion, you may not understand whether your patient has had successful intervention, and more importantly you may not know when the patient has re-occluded.

Posterior leads may help convince others of diagnostic STD maximal in V2-V3, but are not mandatory for diagnosis.


  1. excellent case, Pendell, with very good cath photos and descriptions.

    my partner and i had a similar case a few nights ago,equally frustrating, an obvious major vessel acute occlusion, , but not by "classic" STEMI criteria, so they didn't cath her immediately, and she bumped trop's. she was 39, severe chest pain.
    its somewhat surprising that in 2018 there isn't a generally greater awareness of STEMI equivalents.
    thank you for this case. i am disheartened by the outcome.


  2. I'm still surprised how hard it is sometimes to convince certain cardiologist of STEMI. I agree 100% that the guidelines are misguiding. I feel that i have more luck convincing them when i can argue that i think a certain vessel is occluded. Say for instance subtle high lateral infarcts. Therefore i have a request. Would you be bothered to make a post on how to distinguish infarcts of Cx, obtuse marginal and diagonal branch.

    1. We have had many such cases on the blog previously with occlusions of LCX, OM1, diagonals, etc. Look back at these cases for examples. If you mean distinguishing LCX from OM or from diagonal occlusions, this can be very difficult as the arterial supply is variable in this area. There are some patterns that can be recognized, such as the "mid-anterolateral" occlusion which is usually a diagonal, but this is just not definite enough to say that I could distinguish these particular vessels reliably. Only cath will tell!

  3. Great post and great explanation! I am kinda confused because of the not so steep initial positive deflection of the R wave.

    1. Not sure what you mean, please explain your question

  4. So, I suppose this begs the following question: At what point could/should the ED physician simply decide to give lytics in this situation? Would doing so have potentially changed this patient's outcome? Would love to hear thoughts on this from my fellow ED colleagues.

  5. Really well presented and informative case. The use of OMI as opposed to STEMI and NonSTEMI really proves a point on this case I feel.

  6. Excellent case by Pendell Meyers! I’ll add 2 comments: i) On the initial ECG (that Dr. Meyers received at 18:13) — the SHAPE of the ST segment in lead V6 is coved! I have always believed that “shape” is more important than amount, and this shape for the ST segment in lead V6 (in conjunction with what we see in the anterior leads), is perfectly consistent with acute LCx occlusion; and ii) I suggest use of my “Mirror Test” to facilitate recognition of the shape of ST-T wave deviation indicating acute posterior infarction. To do a “Mirror Test” — simply flip the ECG over, and hold it up to the light. Doing so in this case, would produce coved ST elevation with beginning symmetric T wave inversion, especially in lead V2 (that looks like an acute ST elevation MI). Given how localized the ST-T wave changes in this 1st ECG are (maximal in V2,V3) — this picture should suggest acute posterior MI until proven otherwise (clearly, a “STEMI-equivalent” = OMI). THANKS again to Dr. Meyers for presenting this case!

  7. Great post.Thanks

  8. Great case!

    Did the patient get an echo before cath?

    1. I am not sure whether there was a bedside echo done before cath. I cannot find one.


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