Was the intern correct?

Written by Pendell Meyers, case submitted by Max Macbarb, edits by Steve Smith

A 71 year old gentleman with history of CAD and PCI presented with acute chest pain and normal vitals signs.  He was triaged to the general area of the emergency department after an initial review of this ECG by a senior resident or attending physician correctly interpreted "No STEMI."

An intern who has attended my lectures and has begun reading this blog picked up the chart and flipped to the ECG and saw this:

Presentation ECG at 6:57 AM. What do you think?

I texted this to Dr. Smith with no clinical information and he replied immediately after viewing it on his phone: "Wow, this could be a very subtle LAD occlusion."

The intern was also concerned about LAD occlusion.  So he then looked in the chart and found the patients last ECG on record for comparison:

Prior ECG on file. The most important point of this baseline is that you see how the presentation ECG above is diagnostic of LAD occlusion by comparison to this baseline. Going from this baseline to the presentation ECG above, the R wave progression is blunted and the T-waves have massively grown in relative size, height, and area. The increase in the ratio of area under the ST-T waves compared to the size and progression of the QRS is diagnostic.

However, this "baseline" ECG turns out not to truly be a "baseline." It was the last of a series of ECGs recorded years ago during another presentation with chest pain. The next ECG below is the first of that series of ECGs. What do you think happened during this prior presentation based on the ECG below?

This shows very subtle Occlusion MI of the high lateral wall, and possibly implies posterior wall occlusion as well (though I do not believe this can be diagnosed on this ECG). There is the smallest possible amount of STE in aVL, with much more noticeable STD in II, III, and AVF which is absolutely contradictory to the preceding normal QRS complexes and indicative that they are reciprocal to primary STE in aVL. Lead I does not have any evidence of occlusion to my eyes. This does not meet STEMI crtieria, but to me is highly concerning for Occlusion MI of the high lateral wall. Because I recognized this as occlusion and therefore cannot call this a "baseline" ECG, I went back in the chart and found that the patient indeed had a large troponin rise and was found (on delayed cath) to have a fully occluded LCX which was stented.

Let's get back to the present case by reviewing the presentation ECG again:

Presentation ECG at 6:57 AM.  Sinus rhythm with one PAC.
Why does it look like an LAD occlusion? 
There is less than a mm of STE in V2-V4, which is normal by itself.
But R-wave amplitude is too low for normal STE.
Also, normal STE is NOT associated with reverse R-wave progression: the R-wave in V3 is smaller than the R-wave in V2.
Moreover, there is a Q-wave in lead V4, which should not be seen with normal STE.
The T-waves are too symmetric for normal STE.
There are also Q-waves in inferior leads: old inferior MI.  When an ECG shows old MI (h/o coronary disease - CAD), it makes CAD in other coronary arteries more likely, increasing the probability that the patient has an MI.
Put it all together, and acute LAD occlusion is very likely.

In any case, the differential diagnosis for this ECG should be normal variant vs. subtle LAD occlusion. Therefore, it is appropriate to apply Dr. Smith's formulas for this difficult situation.

Use the blog itself, MD Calc, or your smartphone app called "Subtle STEMI" (hopefully we will change the name sometime), and see what you get. The computerized QTc was 456 ms (this result is correct: raw QT = 397 ms; using Bazett's formula, we divide by square root of RR interval (0.76) = 0.87 and get 456). Look below to see my results:

4-variable formula, cutoff 18.2, now validated. 22.74 is VERY high.

3-variable formula, not as good!  Cutoff 23.4.  27.83 is VERY high.

Back to the case:

The clinician was worried about hyperacute T-waves in the anterior leads, documenting "hyperacute T-waves in V2-V5." He immediately evaluated the patient and found a well-appearing gentleman with normal vitals who was complaining of persistent chest pain of about 2 hour duration which started at rest.

Having confirmed his suspicion of clinical ACS, he alerted his attending immediately and they performed a bedside echo which showed an apical wall motion abnormality. They considered takotsubo cardiomyopathy given the well-contracting base, however they also recognized that the LAD supplies the apex, so an apical WMA would be consistent with Occlusion MI as well. With hyperacute T-waves in the LAD distribution and an apical wall motion abnormality, the diagnosis is Occlusion MI until proven otherwise by emergent angiogram.

They activated the cath lab (against the advice of the ACC/AHA guidelines).

The cardiologists agreed with the ED interpretation, in turn documenting "hyperacute T-waves and anterior-apical wall motion abnormality."

Here is another ECG recorded just prior to transport to cath lab:

Repeat ECG at 7:24. Mostly unchanged.

The patient is currently in the hyperacute T-wave stage of the Occlusion MI progression above.

The patient went quickly to cath lab and was found to have total thrombotic occlusion of the mid-distal LAD. It appears to me that they found a 90% lesion of the proximal-mid LAD which they believe was where the initial plaque rupture occurred and then the thrombus embolized downstream to the mid-distal LAD where it became fully occluded.

Here are the relevant images:

Diseased but overall acutely patent RCA.

Left system showing mid LAD occlusion.

Intra-intervention.

Post-intervention.

Here's a different view, before intervention. Compare this image with the next to see the very large LAD which is occluded in the middle of its course.

Post-intervention.

Let's say you now receive this patient in the CCU. What do you hope to see on this patient's ECGs over the next hours-days to reassure you that he was truly reperfused and stays reperfused on a cellular level?

Answer: you want his ECGs to display the reperfusion sequence shown in our Occlusion MI diagram below. You want his hyperacute T-waves to deflate, then slowly start to invert starting with the terminal portions of the T-wave and progressing to the entire T-wave over the course of days. If you acted quickly enough, and downstream reperfusion is complete enough, you may even witness avoidance of Q-waves and reformation of normal R wave progression.

Now that you know exactly what you hope to see, here are three serial ECGs on the patient during the next 24-36 hours after his intervention. Pretend you are taking care of him in your mind and see whether you are confident that he is reperfused, or whether he reoccludes at some point.

Post intervention.  Notice hyperacute T-waves are resolved. R-waves have been obliterated.

Post intervention 2.  Beginning of T-wave inversion.

Next day.  More T-wave inversion, and some increased ST Elevation

The ECGs show the expected progression of reperfusion, without any signs of reocclusion. Unfortunately there are Q waves and persistent STE in the affected distribution. However the T-waves tell us that the territory is in fact reperfused despite significant loss prior to reperfusion.

The EF on day 2 was 35%, with a persistent wall motion abnormality of the inferior wall, anterior septum, and the entire apex. Peak troponin T was 3.19 ng/mL (this is quite high and indicates a significant myocardial loss).

This is an outstanding display of ED and cardiology experts working together and making the right call, even when our guidelines suggested otherwise. I believe that this level of skill is accomplished in only a tiny percentage of cases around the world in 2018, and these ED and cardiology teams should be commended for this case. I am so glad not to have to tell the story of this case from the perspective of a missed occlusion, as about 25% of all Occlusion MIs are in 2018.

Learning Points:

This was an NSTEMI that needed the cath lab immediately, otherwise called Occlusion MI (OMI).

The usual story is this: the patient does NOT get emergent reperfusion.  Serial troponins are positive and NSTEMI is diagnosed and the patient gets next day angiogram (unless he develops cardiogenic shock, which is likely) and intervention.

There are some who object that there is no proof that emergent reperfusion therapy for NSTEMI has better outcomes than next day angiogram.

However, this is false.  First, most studies of actual emergent angiography DO show benefit. Second, none enroll patients with KNOWN acute occlusion, and, third, they all exclude patients with ongoing symptoms.

If you know that a patient has an acute, symptomatic, total thrombotic occlusion of the mid-LAD (or any artery) without sufficient collateral circulation (as he did in this case), then it would be unethical to randomize him/her into a trial in which "NSTEMIs" undergo either immediately or delayed cath.

Hyperacute T-waves are difficult to recognize, and you must train specifically for this task.

Use the formulas when they are applicable and understand their limitations.

Teach other people with these blog posts. It is rewarding to watch them improve and help patients.

-----------------------------------------------------------

Comment by KEN GRAUER, MD (11/18/2018):

-----------------------------------------------------------

Another superbly illustrated and recounted case by Drs. Meyers & Smith (appreciation to Dr. Max Macbarb for submitting this case!) — showing serial ECG evolution of acute OMI (Occlusion-related Myocardial Infarction) with indisputable evidence justifying immediate catheterization despite not fulfilling the usual “STEMI criteria”.

• I focus my comments on some additional teaching points relating to the first 2 ECGs in this case (Figure-1).
• NOTE — I’ll emphasize in advance that my comments below relate to advanced concepts in ECG interpretation that go beyond-the-core!

Figure-1: Comparison between the initial presentation tracing ( = ECG #1) — and the last ECG on file for this patient ( = ECG #2). 

==========================

ECG #1: ( = the Initial ED Tracing @ 6:57am):

As emphasized by Dr. Meyers, in view of the history (ie, a 71-year old man with a history of CAD, who presents with new-onset chest pain) — the initial presentation ECG (ie, ECG #1) provides definitive evidence supporting the need for acute cath until proven otherwise!

• Although sufficient ST elevation to qualify as a “STEMI” is not present in ECG #1 — the obvious concern regards possible acute LAD occlusion. The point to emphasize is that it does not matter whether or not “criteria for a STEMI” are, or are not present — since regardless, the indisputable hyperacute changes seen here indicate acute OMI until there is evidence to the contrary.

In a patient with a documented history of coronary disease — there are numerous subtle abnormalities in ECG #1. These include:

• The rhythm is sinus with a PAC ( = the 7th beat).
• There is Baseline Artifact. All-too-often I find no mention of this finding in the interpretation of even skilled clinicians. It is important to be aware of baseline artifact — because it explains why the ST-T waves for successive beats in the long lead V1 (at the bottom of ECG #1) continually change. For example — there actually is some ST elevation for the 1st QRS complex in lead II — but really not for the other 2 QRS complexes in this lead. Similar subtle discrepancies are seen for the 3 complexes in lead aVF. When faced with the presence of baseline artifact that alters ST-T wave appearance — I favor use of an overall “Gestalt”, looking for the “theme” for ST-T waves for all beats in all leads in a given lead area. Using this approach — I believe the “theme” for assessment of the ST-T waves in the 3 inferior leads (II,III,aVF) in ECG #1 — is that of being taller-and-fatter-than-they-should-be given the height of the R waves in these leads. Thus, in addition to chest lead hyperacute changes — there are hyperacute Twaves in each of the inferior leads in ECG #1.
• LAHB (= Left Anterior HemiBlock) is present. Criteria for LAHB are met in ECG #1 — because the QRS complex in each of the inferior leads is clearly more negative than positive (ie, the frontal plane axis is clearly more negative than -30 degrees). NOTE — Criteria for LAHB were not met in the last ECG on file for this patient (ie, = ECG #2) — because the QRS complex in lead II of ECG #2 is not more negative than positive (which means that the frontal plane axis was less than -30 degrees!). Therefore — development of LAHB is a new finding in this patient!
• There are inferior Q waves in ECG #1 — which as per Dr. Meyers, suggests previous inferior infarction. This finding deserves further comment — because the appearance of the Q waves in each of the inferior leads in ECG #1 is even more suggestive of previous inferior infarction than would be the case with “simple” inferior Q waves (that often are not definitive for previous MI). By this I mean, that the QRS complex in lead III is not only all negative (ie, a QS complex) — but the downslope of the S wave in lead III is clearly delayed (ie, it is not straight, as is typically the case with a “simple” Q wave). Even more abnormal — is the qrS complex that we see for 2 of the 3 complexes in lead aVF. The finding of an initial small q, followed by a small r, that then evolves into a predominant s wave in one or more of the inferior leads suggests 2 things: i) that there almost certainly has been prior inferior infarction; and, ii) that in addition to previous inferior infarction — there is also LAHB.
• REMEMBER — Because of opposing initial vectors — the presence of inferior MI may “mask” recognition of LAHB — and vice versa. The finding that I’ve found most helpful for recognizing the presence of both prior inferior MI and LAHB is fragmentation of the QRS complex in one or more inferior lead, especially when there is a qrS pattern (as we see here in lead aVF). This is relevant in this case — because the presence of new LAHB in ECG #1 (compared to the last previous tracing on file = ECG #2) — provides one more finding of concern in this patient suspected of new acute LAD occlusion.
• In addition to the high scores on the 3- and 4-variable formulas emphasized by Dr. Meyers — I favor qualitative assessment of ST-T wave SHAPE when assessing for potential hyperacute changes. Chest lead T waves are taller-and-fatter-than-they-should-be-at-their-peak in no less than 4 of the chest leads (ie, in leads V2-thru-V5). I like to “extract” two of the parameters from the formulas for use in my qualitative assessment = i) R wave progression; and, ii) Relative QTc lengthening. In ECG #1 — there is loss of R wave (as per Dr. Meyers, height of the r in V3 is less than it was in lead V2, which is the opposite of what is expected). In fact, transition (where the R becomes taller than the S wave) does not occur until between V5-to-V6 in ECG #1, whereas it normally should occur before lead V5.  And, as per Dr. Meyers, the computerized QTc  = 456 msec in ECG #1 — which clearly is prolonged (whereas early repolarization variants with ST elevation tend to have relatively short QTc values).
• In support of the finding that there is meaningful loss of r wave (ie, not simply due to lead placement error), and meaningful poor R wave progression — is the finding of very small-but-real q waves in leads V4, V5, and V6 of ECG #1. While “normal septal q waves” may often be seen in one or more lateral leads — they should not normally be seen as far over as lead V4 when the r wave in V4 is as small as it is here relative to the depth of the S wave in this lead. This finding of new chest lead Q waves is subsequently borne out by post-intervention tracings in this case.

==========================

Comment on ECG #2 — Dr. Meyers emphasizes the importance of assessing the clinical context in which comparison tracings were obtained. Thus, in this case — ECG #2, which was the last ECG on file for this patient — was not truly a “baseline” ECG for this patient, since it was part of a series of tracings obtained during a prior acute event. Nevertheless — assessment of ECG #2 reinforces our belief that acute OMI is occurring in ECG #1. Let’s again compare the changes seen between ECG #1 and ECG #2 in Figure-1:

• There is new LAHB and new fragmented q waves in the inferior leads of ECG #1 that were not seen previously.
• There are new hyperacute T waves in both the inferior leads, as well as in most chest leads.
• The QTc has lengthened in ECG #1 since the last ECG on file for this patient.
• There is now loss of r wave from V2-to-V3, poor r wave progression with delayed transition — and new-but-real q waves in leads V4, V5 and V6.
• Finally — this patient with known CAD is now presenting with new-onset chest pain. Therefore — ECG #1 should be interpreted as acute OMI until proven otherwise!

Our THANKS again to Drs. Macbarb, Meyers & Smith for this superb case!