Wednesday, April 20, 2016

A 37 year old woman with Chest Pain

I have put up one post with 10 similar cases here.

Don't miss the comments at the bottom!

I was sent this ECG with the following information:

"A 37 year old female with no comorbidities, a non-smoker, with no known hyperlipidemia and no family h/o of CAD presented to ED with central chest pain since 5 hours with no radiation, increased by moving her arms and associated with SOB.  This occurred after a stressful interpersonal conflict. She had had similar episodes before when angry or stressed."

"Her vital signs are BP 110/70, P 70/m, RR 18/m, O2 sat 98% RA.  Physical exam was normal." 

Here is the ECG:
What do you think?

Here was my response:

"I looked at the ECG and immediately thought "This is an acute LAD occlusion."  It is diagnostic of LAD occlusion, but really only to someone who has expertise.  You will virtually never see an EKG like this that is a patient's baseline.  The T-waves are huge in proportion to the QRS, the QRS amplitude is very small, the T-waves are symmetric and fat, there are sagging ST segments in I, II, V3 and V4.  V3 should have some ST elevation, but it actaully has a bit of ST depression." 

These should be thought of as subtle de Winter's T-waves.


"A repeat ECG showed no change."

"Labs shows normal cardiac enzymes and normal troponin. 
She was discharged with diclofenac 75 mg IM injection.  The pain was relieved. 
The patient was discharged home with reassurance and analgesia."

"12 hours later the patient was found collapsed at home with no signs of life.  The family refused a post-mortem.   As she visited our ED within 24 hours from her death, her ECG was reviewed and 50% of the consultants (the most senior physicians at the institution) said there is no abnormality of in it.  The other 50% said there are hyperacute T waves suggestive of early presentation of MI."

My response:

"Sorry to hear about the outcome.  This is a very hard ECG for a non-expert to recognize.  My hope is that this blog, with cases like this, will educate others about these kinds of subtle findings and prevent future cases like this."

Other comment:

"It would have been very helpful to record an ECG after the pain was relieved, to see if there is resolution of the hyperacute T-waves.  I would not use absence of change to be reassured that this is NOT ischemia, as it is too abnormal to be anything else.  But resolution (change) would be confirmatory evidence."

Learning Points:

1. We all must learn these high risk findings of coronary occlusion.
2. Hyperacute T-waves have a unique morphology.  The are like a face that you must recognize.
3. Negative troponins must not be trusted in the setting of an ischemic ECG.
4. Most "misses" like this will never be considered a "miss."  The patient will be admitted for "rule out MI", will "rule-in", will get a delayed angiogram, and will have a completed anterior MI.  Or the patient will have ventricular fibrillation while on the monitor in the hospital and then go to angiogram.  So most of these will be missed opportunities to save myocardium, and will not result in death at home.  This is the situation with these cases: 

Ten (10) Examples of Hyperacute T-waves in Lead V2 (a few in V3), due to acute LAD occlusion

I received a good question:

So as we move to "rapid rule outs" do you think a second troponin would likely have shown a change? I am trying to understand if this is an acute problem how similar previous episodes are connected vs red herring.

My answer:

Not necessarily. 

Even with high sensitivity troponins, there will be false negatives. 

Not all unstable angina will be detected in the future by hs trop. You will still need to be able to read the ECG. 

For this patient:

1) the artery could spontaneously open up before there is any cell death (i.e., before any elevation of troponin). Then, later, the lesion could close off and kill. 
2) if the artery does not reperfuse, then by the time you have a second troponin back, most of the damage is done.

You must be able to recognized these patterns or you will miss an opportunity to make a big difference.

Steve Smith

Some more comments and answers that I want to feature:

Comment: Unfortunately without the post-mortem, or more information (which may not have been included in the article for brevity's sake), I would argue it is not possible to say for certain that an MI was missed here. Perhaps the cause of death was overdose, drugs, or some other factor. The fact that the family refused a post-mortem in this situation makes me think something else may have been at play.

AnswerBut the EKG is diagnostic of LAD occlusion. That is by far the most likely cause of death.

CommentT waves are not hyperacute by definition, and it's a difficult scenario. Looking retrospectively it's easy to say an echo would have been helpful but I would have probably done the same except repeating an ecg and two sets of troponins

Answer:  They are indeed hyperacute. There is no definition. You just have to learn to recognize it. Denying it will just put you in trouble some day, and that's why I'm trying to teach you. I did not look retrospectively. I saw the ECG before I knew any outcome and knew immediately that it was LAD occlusion. Many of my readers tell me this was obvious, not even subtle. Rather than fighting it, try to learn this morphology. It may save a patient of yours. Learn. Don't be closed-minded.

Steve Smith

Finally, Ken Grauer's excellent comment:

THANKS to Dr. Stephen Smith for posting this highly illustrative case. There recently has been a good number of similar tracings like this posted on various ECG forums (each with slight variation from the other) — but ALL with virtually the SAME finding — namely that the T waves in lead V2 (and to a lesser extent in lead V3) are disproportionately tall compared to the amplitude of the QRS complex in these leads. And, the responses I’ve seen from even experienced clinicians show the same remarkable range as that reported by Dr. Smith in this Blog — namely that some clinicians correctly recognize the tracing for a DeWinter variant with acute occlusion (or about to be acute occlusion) of the proximal LAD — and others (unfortunately all-too-many other clinicians) call such findings “normal”.

THIS is a tracing that probably should be shown to ALL clinicians (MD and non-MD) who are called upon to analyze acute 12-lead ECGs — both for teaching and perhaps assessment of ECG interpretation ability … In my opinion, this is a case that just should NOT be missed because: i) the patient presented to an ED with 5 hours of chest pain — which should of itself dramatically lower your threshold for what is “normal” vs “abnormal”; ii) the T wave in lead V2 is almost twice QRS amplitude in this lead — and the width of this T wave is comparable to QRS amplitude in this V2 lead. There is no way this is normal in a patient wth new chest pain.

The findings in multiple other leads that Dr. Smith describes are more subtle, but given the picture described by i) and ii), they all support the diagnosis of DeWinter-like T waves.

Dependence on troponins (even the highest of sensitivity troponins) has no place in a case like this. I would be happy initial troponin was negative — since it means that we have recognized this acute LAD occlusion in time to dramatically improve prognosis.

Most cases of anterior hyperacute T waves will not be as glaring as this one is. But attention to the learning points of this case can markedly help to reduce the chance of oversight: i) When a patient presents to the ED with new chest pain — one has to look that much more carefully at their ECG; ii) Engrain the picture of the ST-T wave that we see in lead V3 in your mind. The ST-T wave in V3 is not as obvious as that in V2, but the T wave is still clearly disproportionately taller, fatter at its peak and wider at its base than it should be given QRS amplitude in this lead (Even without V2, this ECG should be of great concern; with V2 this ECG should be alarming); iii) Train your eye to look extra carefully at the “other leads” on the tracing for “patterns”. Virtually every lead on this tracing (except perhaps aVL) has at least a subtle abnormality that taken together in context with the history of new chest pain + obvious abnormality in V2,V3 adds further support of acuity until proven otherwise.

THANKS for posting this case.

Recommended Resources