Sunday, September 25, 2022

A 50 year old with chest pain? What is going on? By Emre Aslanger.

Posted by Emre Aslanger, our newest blog Editor. Emre is a distinguished cardiologist in Turkey, and has published widely on the ECG in OMI and other areas.

Emre Aslanger Google Scholar Profile

A 50-year-old male with a 20 years’ history of diabetes mellitus treated with metformin only presents with chest pain that started 20 minutes ago. The pain radiates to left inner arm and is now about to resolve. His admission ECG is given below.

What do you think?

You can click on it to make it larger, but let's make it a bit easier to see here:

What do you think? (you can still click on it to make it larger)

Although the wandering baseline makes it a bit hard to interpret, ST-segment depression in inferior leads are quite evident. This should immediately urge the interpreter to look at high lateral (lead I and aVF) and anterior leads for subtle ST-segment elevation. There seems to be a subtle ST-segment elevation in aVL, which can be easily overlooked if interpreted in isolation. The ST-segment elevation in V1-3 is also highly suspicious. Moreover there is ST-segment depression in V4-6. All indicate a proximal left anterior descending artery (LAD) occlusion myocardial infarction (OMI), which sadly does not meet ST-elevation myocardial infarction (STEMI) criteria.

Despite this, the automated interpretation suggests acute STEMI!  

If you were use the LAD-normal variant STE formula (which is not advised when there is ST depression, which make normal variant all but impossible), you would get 18.89, which indicates LAD occlusion.


Smith comment: This is strange because the millimeter criteria for STEMI are not met.  How did this conventional algorithm (not a neural network) diagnose STEMI?


Smith and Meyers comment: We have posted several prior cases of STE in V1 with STD in V5, 6 due to LAD occlusion.  These seem to indicate septal MI, proximal to septal perforator (see below for other cases).  We have coined the term "Procordial Swirl" pattern based on the upward ST-segment shift usually seen in V1-V2/V3 and downward shift in V4/V5-V6, as shown in another example below. Please note: the most important mimic of the precordial swirl sign is seen in cases of LVH, which routinely have STD in lateral leads and STE in V1-V2. 

Here is another example of "Precordial Swirl":


The clinician in charge was not sure and ordered ECG and troponin follow-up.

A second ECG was taken after 15 minutes. The patient said the pain was almost resolved. 

What do you think?

The subtle ST-segment elevation in lead I, aVL and V2 continues. The ST-segments in inferolateral leads are resolving, but there is still significant ST depression especially in lead III (this should always be considered to be reciprocal to high lateral STE, whether that STE is evident or not -- it is NOT "inferior (subendocardial) ischemia"!!  . Also note the tiny Q wave in V2. Even this ECG is highly suspicious, if not diagnostic, for anterior OMI !


Smith comment

Since I started writing this blog, I have been making the point that "ST depression does not localize," meaning that ST depression in inferior, anterior, or lateral leads does NOT tell you that there is subendocardial ischemia in those leads.  This has been proven by many studies of stress ECGs.  ST depression in II, III, aVF is generally reciprocal to (often unseen) STE in aVL.

ST depression limited to Inferior leads is reciprocal to high lateral wall and represents STEMI

ST depression does not localize: 2 cases of "inferior" ST depression diagnostic of high lateral STEMI (references can be found here)


The first troponin (hsTnT) turned out to be 9 pg/dL (normal <14 pg/dL). [This is the same as the ng/L, the units used in most of the US and Europe].  As pain was then completely resolved, they decided to wait for a second troponin. 


Smith comment: If the ECG is diagnostic, as it is here, waiting for the troponin is only acceptable if BOTH ECG and symptoms have resolved


The second troponin came 69 pg/dL. A diagnosis of non-STEMI was made and a search for a PCI-capable center was commenced. 

Another ECG was taken at 4th hour.

Nearly all ST segment changes returned to normal. Note spontaneous reperfusion caused T-wave inversion in aVL and relatively bulky T-waves in inferior leads ("reciprocal reperfusion T-waves".) 

The patient was transferred to PCI-capable center. A third troponin at three hours after admission turned out to be 320 pg/dL.

Angiogram showed exactly what is expected: a thrombosed stenosis just before the first septal and diagonal artery. 

LAD lesion was successfully stented. The next day echocardiogram showed mid and apical anterior and septal dyskinesia with an ejection fraction of 40%. 


Take home messages:

  • A subtle ECG does not mean a mild, unimportant lesion.
  • Troponin lags and does not reflect the seriousness of the situation. 
  • Although this lesion is spontaneously reperfused, the risk of reocclusion is quite high and if it occurred, a significant amount of myocardium would be lost before reaching to catheter laboratory. These patients should urgently be taken to catheterization laboratory.
Research on the ECG in OMI:
If this case were in one of our studies, it would be considered a false positive EKG!!  Our methods in determining retrospectively if the artery was occluded at the time of the ECG recording require EITHER 1) TIMI 0/1 or sometimes 0/1/2 flow OR a culprit PLUS a very high troponin, which for high sensitivity Troponin T is above 1000 ng/L.  In this case, reperfusion was so fast that the peak was at 320 ng/L, and this case would be considered a false positive in one of our studies!!

See these other cases with precordial swirl sign:

A man in his 50s with "gas pain"

Comment by KEN GRAUER, MD (9/25/2022):
We welcome Dr. Emre Aslanger as our newest Associate Editor to Dr. Smith's ECG Blog! The problem he addresses in today's case relates to a "theme" well known to our readers: Recognizing subtle signs of an acute OMI that does not meet "STEMI criteria" — but which nevertheless needs prompt intervention.

For clarity and ease of comparison — I have reproduced the 3 serial tracings in today's case in Figure-1.
  • I find it easiest to appreciate the serial ECG findings highlighted by Dr. Aslanger — by lead-to-lead comparison — in association with close correlation to the clinical history.

Figure-1: Comparison of the 3 serial ECGs in today's case.

MY Thoughts on the Serial ECGs in Today's Case:
As is so often true — the History in today's case is KEY: The patient is a 50-year old man with the longstanding risk factor of diabetes. He presented to the ED with new-onset chest pain that began just 20 minutes prior to ECG #1
  • Not only does the presentation of today's patient immediately place him in a "high-prevalence" group for a likely coronary event — but the fact that he presented to the ED so soon (ie, within 20 minutes!) after the onset of symptoms: i) Makes it quite possible that any ECG findings found on his initial tracing will be subtle and not yet fully evolved; andii) Makes this patient an optimal candidate for maximum benefit from prompt intervention IF it turns out that he is in the process of evolving an acute cardiac event.

ECG #1:
The rhythm in ECG #1 is sinus arrhythmia at ~90/minute. The axis is normal at +15 degrees. The PR, QRS and QTc intervals are normal. No chamber enlargement.
  • Q waves — A tiny, narrow q wave is seen in lead aVL.
  • R Wave Progression — Transition (ie, where the R wave becomes taller than the S wave is deep) — is slightly delayed, occurring between leads V5-to-V6. Of NOTE (and potentially relevant given this patient's clinical course!) — is the fact that the R wave increases from lead V1-to-V2 — but then decreases from V2-to-V3 — increasing again in lead V4 — but inexplicably decreasing one more time from V4-to-V5. I thought this repetitive increasing-then-decreasing R wave progression strongly suggested one or more errors with chest lead electrode placement.

Regarding ST-T Wave Changes in ECG #1:
I counted ST-T wave abnormalities in no less than 11/12 leads!
  • I thought the most "eye-catching" ST-T wave abnormality was in lead V2. Although slight ST elevation is commonly seen as a normal finding in anterior leads V2 and V3 — this usually does not attain 2 mm, as seen in lead V2 of ECG #1. Even more remarkable is the disproportionately enlarged T wave in lead V2 (with respect to the QRS complex in this lead) — that in a patient with new-onset chest pain strongly suggests a hyperacute change.

  • Neighboring leads V1 and V3 also manifest taller-than-they-should-be T waves — that I interpreted as a continuation of the hyperacute change seen in lead V2.
  • Leads V4 and V5 manifest the unusual appearance of an initially downward-sloping ST segment — that evolves into disproportionate taller-and-fatter-than-they-should-be T waves.
  • This appearance of the ST-T waves in leads V4 and V5 made more sense to me when I considered anterior neighboring leads V1,V2,V3 (all of which show hyperacute T waves) — and the neighboring lateral chest lead V6, which shows primarily reciprocal ST depression (ie, leads V4 and V5 being obviously acute — manifest an intermediate picture to its anterior and lateral neighboring leads).

  • As per Dr. Aslanger — the overall picture of ECG #1 in this patient with new-onset chest pain strongly suggests acute proximal LAD OMI. This impression of proximal LAD occlusion is further supported by the subtle-but-real ST elevation in high-lateral leads I and aVL — and the reciprocal ST depression in all 3 of the inferior leads

  • QUESTION: How might the appearance of the acute ST-T wave changes seen in all 6 of the chest leads been altered — IF chest lead electrodes had been correctly placed?

  • BOTTOM LINE: Even before troponin results come back positive — this high-risk clinical presentation in today's patient with new-onset chest pain + abnormal ST-T waves in 11/12 leads, including multiple leads with hyperacute T waves — clearly justifies prompt cath after this 1st ECG.

ECG #2:
It's important to emphasize how optimal it is that the initial ECG was repeated in just 15 minutes! The need for quickly repeating the initial ECG became especially important in today's case when the patient's clinical condition changed! (ie, His chest pain had almost resolved within 15 minutes after ECG #1 was done!).
  • Interpretation of ECG #2 is best accomplished by direct lead-by-lead comparison (which is facilitated by side-by-side viewing of the first 2 ECGs in today's case, as shown in Figure-1).
  • Although subtle — there is definitely less ST elevation in lead V2 of ECG #2 — and there is no longer any ST elevation at all in lead V3. The T waves in these 2 leads clearly look less hyperacute.
  • In lead V1 — Not only has the ST elevation completely resolved in ECG #2 — but instead of the straight ST segment takeoff seen in the 1st ECG, the ST segment is now distinctly flat.
  • In the lateral chest leads (V4,V5,V6) — although T wave size has not appreciably changed — there is no longer ST segment flattening or depression in any of these leads.
  • In the inferior leads — the reciprocal ST depression previously seen in ECG #1 has now almost resolved.
  • In lead aVL — the T wave has become smaller in ECG #2.
  • The only lead not to show any ST-T wave change is lead I.

Putting It All Together:
Although only 15 minutes passed between the time that ECG #1 and ECG #2 were recorded — the patient's chest pain virtually resolved during this period and at the same time, there was improvement in ST-T wave morphology in 10/11 leads that showed abnormalities in ECG #1. This qualifies as clear evidence of "dynamic" ST-T wave changes in a patient who presented to the ED for new chest pain less than 1 hour earlier.
  • As per Dr. Aslanger — Acute proximal LAD OMI should be assumed until proven otherwise. Although relief of chest pain and resolution of acute ST-T wave changes suggests spontaneous reperfusion of the occluded vessel — the risk of reocclusion remains high.
  • Troponin may not become significantly elevated if the period of vessel occlusion is very brief (ie, when there is rapid spontaneous reperfusion). The decision to perform cardiac catheterization should not have to wait for troponin to become elevated.
  • Clinically in Today's Case Conditions could not have been better for potential optimal benefit from prompt cath with PCI of the "culprit" vessel to prevent reocclusion.

ECG #3:
To Emphasize — Prompt cath was indicated in today's case, if not immediately (after ECG #1) — then 15 minutes later after ECG #2.
  • Although ECG #3 was not needed for diagnostic purposes — interpretation of this 3rd tracing adds further support to the conclusions drawn from review of the first 2 ECGs. As per Dr. Aslanger — Chest lead ST-T wave changes have essentially normalized in ECG #3 — with reperfusion changes evident in the limb leads (ie, T wave inversion in lead aVL — and flat ST segments with bulky inferior T waves).

Thursday, September 22, 2022

A man in his 50s with acute chest pain without STEMI criteria. Trop negative. Cath lab cancelled. But how about the ECG and echo?

Case submitted by Matt Tanzi MD, written by Pendell Meyers

A man in his early 50s presented with substernal chest pain and that started 1 hour prior to arrival. There was some radiation to the left jaw and diaphoresis. He had ongoing pain on arrival.

Initial triage ECG:

What do you think?

I sent this to Dr. Smith who immediately replied that it is diagnostic of OMI, but difficult to tell whether it is 1) anterolateral with de Winter morphology, or instead 2) A combination of Aslanger's pattern (inferior OMI with single lead STE in III and reciprocal STD in I, aVL, plus widespread STD of subendocardial ischemia) with posterior (V2 STD) and RV extension (V1 STE). In other words, does V2 indicate ST depression of posterior OMI? or is the ST depression and hyperacute T wave indicative of de Winter hyperacute OMI morphology of the anterior wall? It is difficult to tell, but either way: OMI needing reperfusion.

The ECG shows sinus tachycardia, grossly normal QRS complex, ST depression from V2-V6, and in I and aVL. There are possibly hyperacute T waves in V2 (maybe also V3 and V4), constituting de Winter morphology (hyperacute T wave taking off from depressed ST segment). V1 has a tiny amount of STE and a hyperacute T wave also. The ECG overall is diagnostic of OMI.

The cath lab was activated. He received aspirin and heparin bolus. The cardiologist immediately evaluated the patient and stated that the ECG did not meet STEMI criteria, so the cath lab was cancelled. The cardiologist stated that they should try nitroglycerin and see if that controls the pain. 

He was given sublingual nitro and then a continuous drip.

ED2 Post nitro:

Mostly similar, but ST segments in V3 and V4 have risen. Still de Winter morphology in V2, I, and aVL.

At this time, POCUS was performed by the ED team and is shown below:

I believe the POCUS clips above show a clear anterior wall motion abnormality.

The patient had ongoing chest pain. The ultrasound was reviewed with the cardiologist and together they reactivated the cath lab, about 25 minutes after the initial cancellation.

If you had waited for the initial (conventional prior generation) Troponin T, it was undetectable (less than 0.01 ng/mL).

Angiogram results:

Prox LAD 99%, thrombectomy (no TIMI flow listed)

"Distal nitroprusside administration and distal LAD thrombectomy performed with improvement in no reflow."

"Integrellin double bolus given in lab. Continue Integrelin for 18 hours."

Post cath ECG:

What do you think? What has changed?

Now there is RBBB (the right bundle is supplied by the LAD in all cases I am aware of), with persistent concordant STE in precordial leads, as well as Q-waves in V1 and V2. In the situation of LAD OMI, this ECG does not yet show significant signs of reperfusion, and may instead be concerning for No Reflow phenomenon, meaning lack of downstream perfusion despite epicardial intervention. 

Formal echo next day:
Severely reduced global LF systolic function, EF 25%
WMA: mid anterior, mid anterolateral, mid inferior, mid anterior septum, inferior septum, entire apex, all severely hypokinetic 
No evidence of clot with usage of echo enhancing agent. Swirling of contrast in LV apex consistent with low flow state
Normal RV systolic function

Peak Troponin T = 4.94 ng/mL (large MI, similar to the average STEMI+OMI in our data of 265 OMIs, using the same assay, which had mean and median peak troponin T of 5.50 and 3.78 ng/mL)  

Discharge ECG:

Persistent RBBB. The Q wave in V1-2 has widened, concerning for LV aneurysm morphology in RBBB. There is conspicuous lack of reperfusion T wave inversions in anterolateral areas, which would usually be present if there had been meaningful reperfusion. This ECG further supports No Reflow and long term poor outcome.

Learning Points:

When the OMI is most hyperacute, when there is maximal tissue to be salvaged by emergent reperfusion therapy, the troponin is lowest and there is very often insufficient STE to meet STEMI criteria. We have shown that, even in OMIs that DO meet STEMI criteria,  expert ECG interpretation can identify OMI hours sooner than STEMI criteria. Would you like your EM physician and cardiologist to be ECG experts? or just use STEMI criteria?

When the ECG is not diagnostic to a provider, immediate bedside echo can help to diagnose OMI.

The LAD supplies the RBBB and LAFB. New RBBB/LAFB in OMI is a sign of extremely dangerous LAD OMI.

No Reflow phenomenon indicates inadequate downstream reperfusion and is diagnosable by ECG.


MY Comment, by KEN GRAUER, MD (9/22/2022):


I liked this case — because it presents yet one more variation in the de Winter T Wave Pattern that we all need to recognize.

  • For clarity — I have reproduced and show together in Figure-1 — the first 2 tracings in today's case.

Dr. Meyers showed Dr. Smith and myself the 1st ECG in today's case, initially without the benefit of any history. As noted above — Dr. Smith was immediately suspicious of, "de Winter, LAD OMI".

I focus my comment on my slightly different response. I replied as follows for my interpretation of ECG #1:

  • "Sinus tachycardia — some potentially significant Q waves (in leads aVL, V1,V2) — and LOTS of ST depression, which interestingly is most marked in lead V2 (!), as well as in high-lateral leads I and aVL. In addition — there is ST elevation in “opposing” leads (ie, especially in lead aVR — but also in leads III and V1)."
  • My Impression of ECG #1 was: "Diffuse subendocardial ischemia in a patient with multi-vessel disease (and probably some prior events). IF associated with a history of new chest pain — then this patient should clearly undergo prompt cath to define the anatomy."

The reason why I did not promptly recognize de Winter T Waves (as Dr. Smith immediately did) — was because I didn't see very tall T waves arising from the marked J-point ST depression.
  • The "Take Home" point for me regarding today's case — was that the simple reason I didn't see the very tall anterior T waves usually associated with de Winter T Waves — is that we were seeing this extensive anterior infarction at a very early stage in the process before the T waves had a chance to enlarge.

  • I believe the process in today's patient did begin with diffuse subendocardial ischemia (supported by the diffuse severe hypokinesis on formal Echo done the following day). 
  • Then followed initial marked J-point ST depression (in leads I, aVL and V2 — as seen in ECG #1 of Figure-1) — with development of terminal T wave positivity in these 3 leads.
  • And as we look back on ECG #1 (after seeing the subsequent serial ECGs that evolved) — those Q waves I pointed out in ECG #1 were in fact early indicators of the infarction to come.

  • The follow-up ECG (ECG #2 — done a short while later) — then showed progression of extensive evolving infarction. Specifically — the amount of J-point ST depression in leads I, aVL and V2 is much less in ECG #2 (with reduction in the amount of ST depression in other chest leads) — and T waves in multiple leads (ie, leads I, II, aVL; V2,V3,V4) are becoming larger and "fatter"-at-their-peak (ie, more hyperacute).

Figure-1: The first 2 ECGs in today's case.

Additional Learning Points from Today's Case:
We continue to frequently review the gamut of de Winter T Wave "variants" in Dr. Smith's ECG Blog (See the May 2, 2019 post — among many others).
  • I discussed in detail in My Comment at the bottom of the page in the July 11, 2022 post — how common it is to see "variations on the theme" of de Winter T Waves. In the interest of brevity — I'll simply reproduce in Figure-2, my adaptation from the original manuscript by de Winter et al, published in this 2008 NEJM citation — which shows you 8 selected ECG presentations from his series of patients.
  • I've seen scores of other variations of de Winter T Waves. I have added to "My Mental List" — the additional ECG variation seen in ECG #1 from today's case, in which we catch a very early glimpse of de Winter T waves evolving from initial diffuse subendocardial ischemia.

  • P.S.: Clinically — the fact that I did not recognize de Winter T waves as soon as Dr. Smith did in today's case does not matter — because regardless of what you call the picture of ECG #1 in Figure-1 — in a patient with new-onset chest pain — prompt cardiac cath is clearly indicated. Unfortunately that was not done in today's case

Figure-2: The de Winter T Wave Pattern, as first described by Robbert J. de Winter et al in N Engl J Med 359:2071-2073, 2008. ECGs for the 8 patients shown here were obtained between 26 and 141 minutes after the onset of symptoms. (See text).

Wednesday, September 21, 2022

Posterior leads can give false reassurance

A 40-something woman with diabetes and peripheral vascular disease who frequently needs the ED for chronic pain called 911 for sudden severe chest pain.  The patient was very agitated and could not hold still.

I greeted medics at the door to view the prehospital ECG.

Here it is:

It is not normal, but there is no specific evidence of Occlusion MI (OMI)
There are some Q-waves in inferior and lateral leads (of uncertain age) and there is some "terminal QRS distortion" in inferior leads and V4-V6, but this finding may be present normally in these leads (it is NOT normal in lead V2/V3 in the presence of normal variant ST elevation in V2-V4)

The patient was quite agitated and needed sedation.

Because of this, it was very difficult to record an ED ECG, but eventually it became possible:

What do you think?

There is NEW very subtle ST depression in V3 and V4, and new subtle ST Elevation in V5 and V6.  

There was an old ECG available from 15 months prior:
This shows no STE and no STD.  In fact, as in most patients, there is a touch of normal STE in V3 and V4.  The presenting ECG has a touch of STD in V3 and V4.  
This is diagnostic of Occlusion MI (OMI).

I diagnosed Occlusion MI (OMI) and activated the cath lab.

Before going to cath, we recorded a posterior ECG:
Lead V4-V6 have been moved to the posterior thorax.  
Thus, they have much less voltage.  They have ZERO ST Elevation.  However, you can see that V3 still has ST Depression.  
Does this mean that the ST depression in V3 represents "anterior" subendocardial ischemia, and not posterior OMI?  
That is what many would have you believe.

This did NOT dissuade me in the least!

Does ST depression in V3 diagnose OMI of the posterior wall better than the posterior leads?  It may indeed.

She went to the cath lab and had a 100% Obtuse Marginal Occlusion (supplying the lateral and posterior wall) with TIMI-0 flow.

High Sensitivity Troponin I (Abbott Architect: URL = 16 ng/L for women, 34 for men)
Initial: 26 ng/L
Peak: greater than 50,000 ng/L.  A massive acute OMI.


The estimated left ventricular ejection fraction is 34%

Regional wall motion abnormality-lateral, akinetic.

Regional wall motion abnormality-inferior base (this is the posterior wall).

This is most consistent with ischemia/infarction in the distribution of

the left circumflex coronary artery.

Next Day ECG:
T waves in V2, V3 are larger than they were at baseline. This may be due to "posterior reperfusion T-waves", maybe not.  If it was, I would usually expect reperfusion T-wave inversion in V5 and V6 also.

Learning Points.

1. We have shown that ANY ST depression maximal in V1-V4 is 96% specific for posterior OMI.

Meyers, Bracey, Smith, et al.  Journal of the American Heart Association.  Ischemic ST depression maximal in V1-V4 (vs. V5-V6) of any amplitude, is specific for Occlusion Myocardial Infarction (vs. non-occlusive ischemia)

2. Posterior leads are unnecessary in this situation and although they could be helpful, there is a risk that they will dissuade you from making the accurate diagnosis of posterior OMI.

3.  OMI that are not STEMI can be very subtle and difficult to diagnose even though the findings are very specific.  

4.  In many such cases, it is not the ECG which is nonspecific; it is the interpreter who is nonspecific.


MY Comment, by KEN GRAUER, MD (9/21/2022):


I love cases that support my beliefs — which is why I couldn’t wait to write up my comment on this most recent post by Dr. Smith. The message conveyed by Dr. Smith is simple: — Posterior leads sometimes provide false reassurance. As occurred in today’s case — there are times when despite definitive recognition of posterior OMI by other means — no ST elevation at all will be seen in posterior leads.
  • Failure of posterior leads to consistently demonstrate ST elevation in association with subtle posterior OMI — should not be surprising. This is because posterior placement of leads V7, V8 and V9 situates these leads in a position from which electrical activity must pass through the thick musculature of the back before being recorded on the ECG. As a result — even under optimal circumstances, QRST amplitudes (and therefore the amount of ST-T wave elevation) in posterior leads is often modest.

To facilitate instant recognition of posterior OMI — I favor the MirrorTest, which I first popularized nearly 4 decades ago (circa 1983, in my 1st ECG publication).
  • In the years since I first proposed this visual aid — I have used it to identify many hundreds of posterior infarctions within seconds — and without need to delay management (or transport of the patient) in order to record additional (ie, posterior) leads. The standard 12-lead ECG is all that is needed.

What is the “Mirror” Test?
The "Mirror" Test is nothing more than a visual aid to facilitate recognition of the abnormal shape of ST depression seen in one or more of the anterior leads in association with acute posterior infarction.
  • Figure-1 — shows the rational for the Mirror Test (ie, the anterior leads provide a mirror image of electrical activity in the posterior wall). By simply inverting a standard 12-lead ECG, and then holding it up to the light — you can easily visualize the "mirror-image" of leads V1-thru-V4.
  • With a little bit of practice — it becomes EASY to recognize the "shelf-like" (flat) shape of ST depression — that looks like a stemi when this image is inverted.
  • Posterior infarctions often (though not always) result in an increase in R wave amplitude in anterior leads. The taller anterior R waves become — the more this looks like a deepening Q wave when the image is inverted.

  • KEY Point: If instead of seeing a comparable diffuse amount of ST depression in chest (and limb) leads — you see maximal ST depression in leads V2, V3 and/or V4THINK posterior infarction! Inverting the image (ie, applying the Mirror Test) then facilitates recognizing the characteristic ST-T wave depression shape indicative of posterior infarction.

  • NOTE: Because you are not having to traverse the thick back musculature to record a standard ECG (as you have to do when recording posterior leads) — the relative amplitude of ST-T wave segment deviations tends to be significantly larger than the ST-T wave amplitude seen with posterior leads. This is why I believe the Mirror Test is superior to use of posterior leads.

  • EDITORIAL Note: I do not believe I have ever seen a case in which a posterior infarction diagnosed by posterior leads was not evident by use of the Mirror Test on the standard 12-lead ECG. But as Dr. Smith illustrates in today's case — the opposite is not true (ie, Today's patient was correctly diagnosed on ECG as having posterior OMI despite the complete lack of ST elevation in posterior leads).

Figure-1: Illustration of the rational for the Mirror Test (Figure excerpted from Grauer K: ECG-2014 Pocket Brain ePub).

Applying the Mirror Test to Today's Case:
To illustrate my use of the Mirror Test in today's case — I've reproduced in Figure-2 the initial ECG obtained in the ED (which is the 2nd ECG shown above by Dr. Smith— and a prior baseline tracing on this patient (which is the 3rd ECG shown above).
  • Of NOTE  The fairly large Q waves seen in the inferolateral leads in ECG #2 are not new! Instead — comparable Q waves were seen in the prior baseline ECG, suggesting this patient had a prior infarction.

  • What is new  is lateral ST elevation, which is marked in lateral chest leads V5,V6 — and which is subtle-but-present in high-lateral leads I and aVL. This lateral ST elevation was not present on the earlier tracing.

  • What is also new — is subtle-but-real "shelf-like" ST depression that is maximal in leads V3 and V4 (and hinted at in lead V2)
  • QUESTION: Isn't it EASIER in the mirror-image (inverted) views of leads V2,V3,V4 to appreciate that this shape of anterior ST depression in this patient with new-onset chest pain is diagnostic of acute posterior OMI (ie, This is a positive Mirror Test).

  • To Emphasize — Comparison of ECG #2 with the prior baseline tracing confirms that the ST depression in ECG #2 is clearly new! But before I looked at the prior tracing — I already knew from the positive Mirror Test that the shelf-like (flattened) shape of the depressed ST segments in leads V3 and V4 was clearly abnormal — and indicative of acute posterior OMI until proven otherwise.
  • Although ST-T wave changes in lead V2 are much more subtle than what we see in leads V3 and V4 — given that the T wave is typically upright in lead V2 and the ST segment is usually sightly elevated in this lead as a normal finding — I knew that in a patient with new chest pain, the ST-T wave appearance in lead V2 was not normal.

  • Putting It All Together — As stated, the inferior Q waves (and inferior infarction) in ECG #2 is old. But the combination of lateral lead ST elevation + the positive Mirror Test in leads V2,V3,V4 — indicated acute posterolateral OMI, which strongly suggested the "culprit" artery to be a branch of the LCx (Left Circumflex). This was confirmed on cath.

Figure-2: Comparison in today's case of the initial ECG in the ED ( = ECG #2) — with a prior baseline tracing ( = ECG #3) on this patient obtained 15 months earlier. The inverted (mirror-image) views of leads V2,V3,V4 in ECG #2 constitute a positive Mirror Test — that in this patient with severe, new-onset chest pain — is diagnostic of acute posterior OMI (See text).


Selected LINKS for More on the Mirror Test:

  • ECG Blog #246 — Reviews the concept of the "Mirror Test" with a clinical example.

  • The February 10, 2022 post in Dr. Smith's ECG Blog — My Comment (at the bottom of the page) illustrates the Mirror Test in a case with posterior reperfusion waves (ie, tall anterior T waves).

The following posts in Dr. Smith's ECG Blog provide additional examples in My Comment (at the bottom of the page) that illustrate application of the Mirror Test for diagnosis of acute Posterior MI.
There are many more examples of posterior OMIs with positive Mirror Tests sprinkled throughout Dr. Smith's ECG Blog ...

Recommended Resources