Thursday, February 22, 2018

Shoulder pain after lifting a heavy box

Written by Pendell Meyers, edits by Steve Smith

This will be too easy for most long-time readers, but if you are at that level, sit back and enjoy noticing how few milliseconds it takes to recognize this thanks to so many prior examples on this blog!

I was sent this ECG from EMS with only the information that it belonged to a middle aged male with left shoulder pain.
What do you think?

There are hyperacute T-waves in leads V1-V6, as well as in leads II, III, aVF. The J-points are all at baseline with the exception of leads V2-V3 which show a small amount of STD (which makes de Winter morphology in the presence of hyperacute T-waves).

How can you explain that the most obvious findings are in the anterior leads, yet the inferior leads are also hyperacute??

The occluded vessel must supply the anterior wall and also the apex and/or inferior wall. The most common variant that satisfies this is a type III "wraparound" LAD. This is a large and long LAD that wraps around the apex of the heart, supplying the apex and sometimes even parts of the inferior wall.

By ECG, this acute coronary occlusion is predicted to be of very short time duration, with very high acuity and very high viability. As shown in our reference diagram below, hyperacute T-waves generally exist only within a few hours of persistent acute coronary occlusion, or immediately after reperfusion ("on the way up, and on the way down," as Dr. Smith says).

We activated the cath lab based on this EMS ECG, because it is obviously diagnostic of acute coronary occlusion involving the anterior, lateral and inferior walls. When I make this decision prospectively on this particular highly diagnostic ECG, I estimate that the likelihood of acute coronary occlusion as the etiology of these ECG findings is approximately 99%, with the remaining 1% being the occasional takotsubo cardiomyopathy with indistinguishable ECG findings, and which can only be differentiated by angiogram.

The patient arrived in the resuscitation bay at the same time as the cardiologist.

He was a middle aged man with history only of HTN who called EMS for "soreness" of the left shoulder while working in his garage. He stated he lifted a box weighing approximately 75 lbs, then set it back down, then noticed severe pain in his left shoulder described as "soreness" and "pressure." He stopped working, but the pain persisted. He waited 2-3 hours at home before calling EMS thinking the pain might simply go away.

Here is his initial ED ECG:
Essentially the same findings, hyperacute T-waves without dramatic ST segment changes.

The cardiologist was somehow not impressed by these findings. He also thought that the pain was musculoskeletal because it started around the time of lifting a heavy box. Yet on exam the patient had full range of motion without any change in his constant severe shoulder pain.

I advised the cardiologist that this patient must be taken immediately for cath and intervention. He stated that this ECG does not meet STEMI criteria. I said that the patient has an acute coronary occlusion based on the hyperacute T-waves, the same pathology as an obvious STEMI. The only difference being that there is even more viable myocardium to save than a classic obvious STEMI because there is not yet STE.

Note: The reason there is even more myocardium to save than classic STEMI is because acutely ischemic myocytes first "register" in the T-wave and create increased area under the T-wave, then as they start undergoing the process of death they register in the ST segment, and finally when they are stunned or dead they cannot conduct the action potential and register in the Q-wave. As far as I know this is not proven on a cellular level but is well supported by my experience and hundreds of cases on this blog.

He asked me where I thought the lesion was based on the ECG, and I said "mid LAD or higher, and the LAD will be a type III wraparound."

I stood by the monitor, getting repeat ECGs every 5 minutes for the next 20 minutes while trying to convince the cardiologist, expecting the repeat ECGs to show evolution to frankly obvious STE. But the ECGs did not change - hyperacute T-waves were present non-stop for approximately 45 minutes (from EMS ECG to my last ED ECG). The patient stated that his pain had been exactly the same for 3-4 hours, with no episodes of decreasing and then returning pain.

In my experience (and Dr. Smith agrees), it is unusual for hyperacute T-waves to last this long without progression or evolution. Most cases we have on this blog show evolution to obvious ST elevation, or you see the predictable progression of reperfusion and reocclusion with hyperacute T-waves in both directions. It is possible that there was reperfusion and then reocclusion between the EMS ECG and the ED ECG, although this is less likely because the patient denied temporary improvement in symptoms. Interestingly, Dr. Smith notes that de Winter himself stated that his characteristic morphology was stable for several hours, although Dr. Smith's opinion is that de Winter's data did not actually support that assertion.

It is possible that he had some very small source of collateral flow which was just barely enough to prevent progression, keeping him on the upper end of the de Winter pathology spectrum. It is also possible that the patient had recurrent brief episodes of reperfusion and reocclusion which did not have enough time to show the progression of ECG findings before reversing.

My fellow resident performed a bedside US showing a very dense anterior and apical wall motion abnormality, further confirming the diagnosis.

Thankfully, the cardiologist took the patient to the cath lab at this point.

Here's what they found!

100% mid-LAD occlusion.

With red arrows at the site of occlusion.

Mid-intervention, you can see the occlusion has been opened and there is a very large territory of myocardium supplied by the previously occluded LAD.

After intervention with good flow. This LAD is large and long such that it cannot be captured in one frame, and is seen extending down further in other images (not shown), wrapping around the apex of the heart. This confirms wraparound LAD.

Here is his post-cath ECG approximately 2 hours after the above ED ECG (no ECGs available between my ED ECGs and this one):

What do you think?

This ECG shows progression of acute myocardial infarction to almost complete transmural anterior wall loss. There are new deep wide pathologic Q-waves in V1-V5, with persistent STE and some loss of T-wave hyperacuity. Hyperacute T-waves in the inferior leads are now much less hyperacute than seen in the prior ECG.

But we're done, right? The angiogram confirms successful reperfusion, right?


The ECG is much better at confirming or denying reperfusion than the angiogram. The reason is that epicardial large vessel flow does not necessarily ensure that the actual myocytes downstream are receiving blood flow. The epicardial vessel obviously branches into innumerable smaller branches to provide capillary circulation to the cells. The ECG measures the viability of the cells rather than the flow through the epicardial coronary vessels. Expert ECG interpretation is better than angiogram, better than troponin, even better than patient reported symptoms for determining the state of occlusion and reperfusion!

"No reflow" phenomenon is described when a patient seemingly has successful reperfusion on angiogram but persistently progressing myocardial infarction. This is usually evidenced by ECG changes that progress along the occlusion progression below, and looks similar to the progression seen in patients who receive no intervention at all. This is thought to be due to downstream showering thrombi and platelet aggregates into the microcirculation or other unknown pathophysiology.

So knowing that this patient had relatively good door to balloon time, good angiographic result, and hyperacute T-waves just prior to emergent cath, what do you expect to see for his peak troponin over the next 24 hours?

Answer: Very very high, because despite the angiographic result the ECG shows complete infarction.

Indeed, his peak troponin T was 8.89 ng/mL (very highly elevated). Echo showed EF 35% with dense anterior and apical wall motion abnormality.

Here is the next day ECG:
T-waves further deflating as expected in the course of subacute total LAD occlusion infarction. Remember, when this QRS morphology is present the T-wave becomes almost the only reliable indicator of progression/reocclusion/reperfusion, and is in fact the main consideration used in decision aids to differentiate acute LAD occlusion versus persistent STE (LV aneurysm morphology).

These ECGs are unfortuantely an excellent example of the progression to "LV aneurysm morphology" (see below for reference). The ECG will likely remain similar to the above indefinitely. It is critical to recognize this morphology as this patient is at highest possible risk for the classic complications of transmural infarction including anatomic LV aneurysm, mural thrombus with subsequent stroke, free wall rupture, VSD, Dressler's syndrome, etc. He is also at risk of another complication that gets less discussion: misdiagnosis of his new baseline ECG! He will likely have persistent STE which will be alarming to his future providers, yet the T-waves will likely be the most reliable electrocardiographic feature to tell us whether he is experiencing further acute coronary occlusion affecting his anterior wall. He is at risk of mismanagement in both directions. Should he unfortunately suffer a pulmonary embolism or pericarditis or simply GERD or chest wall pain, for example, he may present with chest pain and this ECG, prompting premature diagnostic closure and immediate catheterization rather than further workup for other causes. Should he suffer another acute coronary occlusion, diagnosis may be missed or delayed because his ST segments are not appreciated as different from his baseline ECG which will now have STE forever.

The patient recovered and did well. Long term outcome unknown.

Learning points: 

1) You MUST be able to recognize hyperacute T-waves such as these. They are common early in the course of acute coronary occlusion.

2) You must learn and advocate for your patients, as these ECG findings are not widely taught or known since our current guideline-promoted strategy focuses only on the misguided STEMI criteria. 

3) Hyperacute T-waves and/or de Winter morphology may be present for hours without obvious evolution, but in most cases you can find evidence of progression with serial ECGs. They may even represent complete occlusion with ongoing necrosis (infarction)!

4) Type III Wraparound LAD is a common anatomic variant which produces anterior and inferior findings on ECG.

5) The ECG is the most accurate measure of occlusion and reperfusion, even better than angiogram, laboratory values, or patient symptoms.

6) You must become familiar with the ECG findings of complete full thickness infarction, as well as the clinical syndromes and complications that ensue.


  1. Thanks once agian Dr. Smith! For about a year now i've been following youre website. Im a paramedic now for 1 year in the Netherlands (small country in Europe) and youre website has drasticly improved my knowlegde. Youre making a difference whilst not even aware of it. Thank you! Kind regards from the Netherlands!

  2. GREAT case by Drs. Pendell Meyers & Steve Smith! I’d add that there is loss of r wave (from lead V1-to-lead V2) plus fragmented Q waves evident in leads V2-thru-V4 (as well as q waves in all lateral leads) on the very 1st ECG. So despite the acute appearance of this event from ST-T wave morphology — would these anterior fragmented Q waves (which later evolve to frank QS complexes) signal significant myocardial injury has already been sustained? Thanks again for this wonderfully written insightful case!

  3. Don't forget to read the EKGs from right to left in any condition. That's what I call "Arabic or Hebrew style".
    Why? Because depolarization (QRS) depends of repolarization (ST-segment, T-wave), and once you'll have an obvious T-Wave abnormality (like this particular case), that will progress further to the ST-segment, then to the QRS.
    Don't wait till the QRS is reached by the trouble! Act Now!

  4. Without arrogance, that cardiologist should follow your blog...
    Another excellent, and as usual, instructive case. Thank you!

  5. Great case.... Thank you. . Any links to MI in a LV aneurysm patient?

  6. Being that I have been reading the blog for a while, I immediately was able to realize that this EKG is diagnostic. In fact its not even a subtle finding. Which is why i forwarded this post to a few people. They were amazed at how much they still need to learn.

    So thank you Dr Smith.

  7. Nice case, thank you!! Another learning point I would like to add (even if it's an ECG blog): you must use echocardiography at bedside ad soon as possible!! In order to support our ECG diagnosis and because STEMI or STEMI-equivalent not always is due to classical coronary occlusion (what if it were an aortic dissection with coronary involvement and we didn't perform echo?). I follow you with great attention, thank you for your job! Giulia Geremia, Verona Italy

    1. But no echo finding would counter this ECG finding. You don't need it in this case! Anyway, it was done: see the post!

  8. Excellent learning here! Any recommendations for free online coronary anatomy and/or angiography atlases or references to review both normal and unique anatomy?

    1. I don't know of any offhand, but just search and I'm sure you will find.

  9. steve and pendell, thank you kindly.
    even i have come to a place where i can say " that ekg looks funky!!".
    then i say, " is pendell trying to trip us up?" is there any way on god's earth that this something other than very bad?"

    in my shop, we don't even have stat cardiac cath. so if i say "bad, bad", the nurses know to trust me , and everyone says "dr fiero are you calling a STEMI?" , and i pause for a sec, and say , one of three things (for an EKG such as this: 1. "yes". a lie, but is easiest. 2. "its a semi-STEMI" , which confuses everyone enough to get the right stuff activated. or 3. "its a STEMI- equivalent". all have the same effect, initially. it gets the patient out of the triage EKG room into a real room, two IV's , and the STEMI box comes out, all within a few seconds. (the team in our shop is expert).
    but... i better be damn sure of what i am diagnosing, because the next immediate step is stat transfer to another hospital by ground, or by chopper. (both of which , as you know, are no little things. choppers can crash, are 50,000., and stat ground has its own risk/expense).
    so i need to be as sure as i can be about what i am seeing, interpreting, when there's an EKG placed under my nose from triage.
    and, that gentlemen, is why i, and from the comments, a great many others of us, from that "small" country in europe( the Netherlands), to verona , italy, to merced, california, and elsewhere, say " thank you , pendell, and stephen, and kenny g.


  10. I am really so glad to be here for this educative blog. Really so informative


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