Sunday, March 30, 2014

The ECG told the whole story, but no one listened: ECG interpretation skills are critical to patient outcomes.


This was sent by a medical student somewhere in the world who will remain anonymous.  Details are scant so that it cannot be recognized.

The minute this medical student saw the first ECG, he knew the diagnosis without any further information.  Reading ECGs is hard, but can be done with commitment to learning, which comes from an awareness of its importance.  My most talented blog readers are paramedics because they have to put themselves on the line every time they activate the cath lab.  And they teach me a lot.  One of my most talented readers is a health care assistant (a nursing assistant) who has taken a keen interest in ECGs.  He can beat nearly anyone.  So you don't have to be highly trained.  You don't have to be a genius. You need to be interested and understand the value of the ECG.   It is not the ECG which is nonspecific, it is the interpreter.


A female in her 60s presented with 3 days of nausea, vomiting, and diarrhea, and not feeling well.  She complained of jaw pain and shortness of breath.  There was a 30 second episode of syncope, and another one the previous evening, without any tonic clonic movements, tongue-biting, or incontinence. She was ill appearing.  She had a remote history of lung cancer many years ago, cured.  She had no other medical  history.

She had this ECG recorded:
ECG 1: QTc is 484.  What do you think, especially compared to the previous, pasted below?

Previous ECG. completely normal

The diagnostic features of the first ECG were entirely missed by a whole team of emergency physicians, cardiologist, surgeons and others, even as the patient became increasingly hypotensive to a systolic BP of 60.  The lactate was elevated and pH low.  They were worried that the syncope was seizure and that she had brain mets.  Then they were worried about sepsis as an etiology of hypotension.  Then the notes mention "cardiogenic shock" but without any reference to a cardiac echo or to a chest x-ray.  Was there pulmonary edema?  Not mentioned in physicians'  notes.  Then she had some RUQ tenderness which led to a long process of evaluation for an abdominal source of sepsis, which turned up negative.  Initially fluids were given, then stopped in favor of pressors, as a CVP line showed CVP = elevated.

Another ECG was recorded:
ECG 2: Findings are still present and have evolved to a small degree

At time 6 hours, Troponin T returns at between 0.50 and 1.00 ng/mL (not giving exact number, but significantly high)

Now having frequent PVCs.

Now chest pain free.  Cardiologist note says: "Elevated troponin explained by type II MI due to her shock."  "Repeat ECG showing no STEMI, only non-specific ST-segment and T-wave abnormalities, unchanged from prior"

Transferred to surgery for exploration but diagnostic studies were too indeterminate to be certain of intra-abdominal pathology.  Besides, the surgeons say she is "too unstable for surgery."

Day 2:

A cardiac echo was done which showed normal LV function but moderate RV systolic dysfunction and diastolic septal flattening consistent with right ventricular volume overloadRight ventricle: moderately enlarged, hypocontractile free wall, moderate global decrease function.

Trop T now very high, well into the range one sees with a STEMI; very unusual in type II MI.

Still the pieces were not put together.

Another ECG is recorded:

ECG 3: More evolution, but not recognized.

I have only shown you 3 of the many ECGs recorded over the 3 days, all of which were diagnostic but not diagnosed.

After midnight (now day 3), she complains of shoulder pain and dyspnea.  Troponin was repeated and returned higher still.

She was taken to the cath lab.

What was the diagnosis?  See below.

Outcome and Analysis:

ECG 1 is diagnostic of inferior and right ventricular MI.  There is a junctional bradycardia.  There is less than 1 mm of STE in inferior leads with diagnostic reciprocal ST depression in aVL.  Additionally, there is ST elevation in V1 which, in the presence of inferior MI is diagnostic of Right ventricular MI.  This is the etiology of the syncope and hypotension and shock and elevated CVP. and right ventricular echo findings..  Furthermore, there are T-wave changes in V2 and V3 which are highly suggestive of ischemia, but difficult to localize: anterior? posterior? right ventricular?  In any case, these further support the diagnosis of coronary occlusion or near occlusion.  ECG 2 is similar, also with a junctional rhythm that is resulting in inverted P-waves in inferior leads (retrograde).  ECG 3 is yet more evolved.

When they finally did it, the cath showed a 100% RCA thrombotic occlusion at the ostium (proximal, before the RV marginal branch), confirming all clinical and ECG findings (except for diarrhea).  There was also disease of the LAD and Left circ.

She turned out to not have any abdominal pathology.
All cultures were negative.
There were multiple complications.

This was an inferior and right ventricular MI and once the first ECG was recorded, the diagnosis could have been made: everything about the presentation, after the ECG, screamed this diagnosis.


I cannot read the minds of the physicians involved, but I suspect there were two components to the delayed diagnosis:

1. Atypical presentation (nausea, vomting, and diarrhea)
2. Failure to regognize subtle but diagnostic features of the ECG.

Many say generally that the ECG is often a "non-diagnostic" test, or a nonspecific test.  Most often, it is the interpreter who is nondiagnostic and nonspecific.  The ECG here told the whole story, very specifically.  It was the failure of accurate interpretation which led to a 2 day delay in diagnosis and significicant myocardial loss and much unnecessary diagnostic and therapeutic intervention.

The ECG told the whole story, but no one listened.


  1. Interesting case, but quite alarming, and of cause terrible for the patient... First ECG should have been picked up by the cardiologists."time is muscle".

  2. And ST elevation in lead III is more than in lead II.

  3. ...astonishing that something that blatantly obvious was missed by the "professionals".

    1. I hesitate to be critical because it is very easy to get tunnel vision and miss things. I hope that by publishing such cases others are less likely to repeat the mistakes. We have come to rely so heavily on newer technologies that we forget to look for the information that is readily at hand, and many do not know how much information the ECG contains.

  4. Thanks for the kind words!

    I think folks tend to take their experience with other testing modalities and apply it to the EKG, thus conflating "subtle" with "non-specific" without realizing the exact points you explain so well.

  5. inattentional blindness? kinda like the way we say, "if you don't look for hyper-K, you won't be thinking hyper-K."

    ECG is very obvious.

  6. Interesting case. Ecg 1 clearly shows inferior wall MI. I think pt with inferior wall MI should have undergone right sided ecg for right ventricular infarction. Its upto the interpretor.

  7. This blog is excellent way to learn Advanced ECG interpretation.

  8. Thank you for this, from the perspective of a female patient much like the one described in this case: I was misdiagnosed in mid-MI despite textbook cardiac symptoms yet "normal" ECGs, sent home from Emerg with a confident: "You are in the right demographic for GERD!"

    As reported in Circulation (Masoudi et al, 2006), "12% of high-risk EKG abnormalities are not detected by physicians working in Emergency Medicine."

    Q: Even with the misinterpreted ECGs, why wouldn't the high Trop T here have raised alarms earlier?

    1. Kennarina, Sorry! I just saw this now. much too late. Thanks for your great comment. They must have attributed the elevated trop to type II MI. Very inappropriate!
      Steve Smith

  9. While I think it's an important case to post, stating that ECGs are not non-specific is an overstatement. ECGs are only the gold standards for most arrhythmias, QT prolongation (obviously), and intraventricular conduction delay (i.e. bundle branch blocks). Should an acute MI be the highest item on this patient differential diagnosis based on the brief history and this ECG? Absolutely. But it's not the only thing. I've seen many patients with mild ST segment elevation get taken straight to the cath lab only to find their coronaries unremarkable. The problem here is not that people refer to ECGs as "non-specific", but rather a failure to recognize that abnormalities from "non-specific" tests still have differential diagnoses which often warrant further (and sometimes urgent) evaluation.

    1. This ECG is very specific for inferior and RV ischemic injury. With the inferior ST elevation and reciprocal ST depression in aVL and the ST elevation in V1,there is no other pathology other than localized myocarditis (which is very rare) that causes this specific pattern. It is possible that the ischemic injury is due to another etiology than ruptured plaque: stress cardiomyopathy, coronary embolism, even fixed stenosis in the presence of hypotension. But it is very specific for ischemic injury. And by far the most common cause will be ACS of the RCA.

    2. While this particular ECG is specific (given the inferior location of STe and reciprocal changes), I would still maintain that ECGs in general are non-specific tests for most non-arrhythmia diagnoses, ST-T changes included (for example, the more common pattern of mild ST elevation in the septal/anterior leads without reciprocal changes).

    3. I agree with that. However, much of the time that an ECG is called "Nonspecific", it is NOT nonspecific. Many infarcts that have "nonspecific, nondiagnostic ECGs" actually had quite diagnostic ones. It was the interpreter who did not know what to look for.

  10. thank you dr Smith.........very helpful n i will keep in my dd always!

  11. This is a great blog, and you are clearly a very advanced ECG reader. BUT, it seems that you cherry pick ambiguous cases that turn out to have real disease and then criticize the treating physicians retrospectively. I'm sure we could present just as many cases with similar ECGs that turn out NOT to have ischemic injury.

    1. I read almost all of these ECGs prospectively. Notice that, in this case, the ECG was immediately recognized as inferior and RV MI by a medical student. This one is not difficult at all if read by someone with a prepared mind.
      Steve Smith


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