Monday, December 13, 2010

Computer misses again -- two cases: one very obvious, one very subtle

Case 1.

A 63 year old male had chest pain and syncope after playing hockey (this is Minnesota). A prehospital ECG showed inferior ST elevation and the cath lab was activated by paramedics from the field. Initial BP was 96/60 with a pulse of 65. While waiting for the cath team, an ED ECG was recorded.

This is an obvious inferior STEMI. However, the computer algorithm read "ST elevation with normally inflected T-waves, probably early repolarization. Marked ST depression, probable subendocardial ischemia."

Just remember: if it can be so wrong in this case, then it can also be wrong in the cases you may miss.

There was no right sided ECG available, but a bedside ED echo showed good LV fct, a very large and poorly functioning RV, and a full IVC, all indicative of an RV MI. Volume loading improved blood pressure.

Immediate cath confirmed a very proximal RCA occlusion.

Case 2.

A 67 yo male with h/o mechanical aortic valve and h/o MI stopped taking his coumadin months ago due to depression. In the last couple weeks he has had some CP with exertion. On the day of presentation he had the onset of CP 5 hours previous.

3:01 PM. Read by computer as "nonspecific ST and T wave abnormality"

There is ST elevation of 1 mm in at least 2 inferior leads, and reciprocal ST depression and T-wave inversion in aVL. There is also some ST depression in V2 and V3; any ST depression in these leads is abnormal. This is diagnostic of inferior posterior STEMI.

There was some disagreement about intervention, partly because of the possibility of an emoblism from the valve, and after aspirin and heparin, the patient had some resolution of symptoms, so he underwent a second ECG 1.5 hours later.

4:30 PM

There is now resolution of ST elevation.
The next day, cath revealed an embolism in OM-1, 100% occluded. Peak troponin I was 22 ng/ml. The ECG showed increased T-wave amplitude in V2 and V3, which I have noticed frequently in reperfused posterior STEMI and call "posterior reperfusion T-waves."


  1. very interesting indeed. details are often overlooked and undervalued.

  2. Very interesting cases. In particular, I find it remarkable that a S1Q3T3 pattern is absent in case 2. However, no diagnostic tool is perfect I suppose. Any insight on the accuracy/sensitivity of that pattern in PE detection, Dr. Smith?

  3. In both cases i can see STD in the anterior leads even in the first case i was pretty sure of posterior MI .

    This provokes me to ask how we can differentiate between STD in the anterior leads due to RVI and right venticualr strain OR due to posterior STEMI .

    In the 2nd case you mention the STD is due to posterior MI based on what ? without performing right sided leads to exclude STD SECONDARY to RV strain !!

    Don't you think that the classic 12 leads ECG is becoming insuff. for precisse diagnosis of inferior MI ??

  4. Sorry, this was NOT a PE. It is a right ventricular infarct. Some of the pathophysiology is similar, especially the large RV on echo.

    S1Q3T3 is only about 8% sensitive for PE.

    Here is a good article: Marchick MR. 12-Lead ECG Findings of Pulmonary Hypertension Occur More Frequently in Emergency Department Patients With Pulmonary Embolism Than in Patients Without Pulmonary Embolism. Annals of EM 2010; 55(4):331.

  5. Any time there is inferior STEMI with ST depression in right precordial leads, that ST depression is due to posterior STEMI (in additon to inferior). RV MI usually has some ST elevation in V1. In this case, it did NOT because that ST elevation was overwhelmed by the ST depression of the posterior STEMI.

    Right side ECG is not for diagnosis of posterior MI, but to confirm or detect RV MI. This was done but the ECG was not on the computer system.

  6. thank u 4 this intersting case.. my Q is what the change will be in ECG when apply it in Right side for posterior MI?

  7. Right side ECG in general does not help in Diagnosis of posterior MI. Posterior leads help.

  8. What is the rhythm in the first ECG? There are negative P waves in II with pauses, so it's not sinus. Looks more like an ectopic focus. Could there be sinus node dysfunction because of the proximal RCA occlusion?

    1. Good question! It looks like a bradycardic low atrial rhythm with 1st degree AVB, PVC's, and at least one early junctional escape (it escapes after the p-wave but before the p-wave conducts!)

      What do you think?

      Steve Smith

    2. Indeed. With the third and the last QRS as junctional escapes. So, can the lack of sinus activity be interpreted as a sign of ischemia? (not that one would need subtle signs of ischemia in this ECG, I was just curious because I never thought of that, first thing that comes to mind with inferior infarcts is AV Block)

    3. Inferior MI often has high vagal tone, not just AV node ischemia. See this very old paper:

    4. Of course, that had slipped my mind. Thank you so much for the answers and the article. And for this amazing blog!

    5. Thanks for the feedback!

      Steve Smith