Tuesday, January 22, 2019

Anterior ST Elevation and a High Initial Troponin.

I was shown this ECG with the history that the patient had chest pain:

What do you think?

This is what I said: "This is not anterior STEMI; it is normal variant, though it is unusual in appearance."

If you wanted to use the formula, the computerized QTc on all ECGs was 360 ms - 370 ms

4-variable formula value = 12.14 (this is VERY low).

Then I was told that the first troponin I was 14.0 ng/mL (very high!)

Here is the history:

A 40-something male presented with pleuritic left-sided chest pain, radiating to the neck, on and off for 3 days.  He has also noted some dull generalized chest pain which he locates more so on the right side of his chest with some associated fatigue and shortness of breath.  He does feel generally unwell and notes chills at home.

As soon as the ECG was seen, the patient was moved to the critical care area and cardiology was consulted.  They stated that the ECG was normal variant, and the patient was moved out of the critical care area.  Then that troponin returned at 14 ng/mL, and he was moved back to the critical care area.

Comment: When I heard all this, I thought it must be myocarditis. But before assuming myocarditis with chest pain and a positive troponin, one must ascertain absence of wall motion abnormality (WMA) or, if there is a WMA, then one must prove by angiography that it is not due to occlusion MI (OMI), acute occlusion of a coronary artery.

If it is OMI, then it is subacute OMI, as the troponin is already very high.

There were several further ECGs, and all looked the same. Here is one:

An emergent formal echo was obtained:

Normal estimated left ventricular ejection fraction .
Regional wall motion abnormality-inferolateral.


There was no anterior wall motion abnormality.

At this point, in a patient with persistent pain, wall motion abnormality, and high troponin, the cath lab was activated.

Because of the arrival of another more critical patient, and also the intervening resolution of pain, he was treated medically and taken for angiogram the next day and found to have a 100% obtuse marginal occlusion (OMI).  This was opened and stented.

His ECG never evolved in any way, over several days.  When STE is due to ischemia, the ECG always evolves.  Absence of evolution is proof that the first ECG showed no evidence of ischemia.

Troponins did not rise any further, which is a bit atypical even for subacute OMI.  Opening the artery usually releases troponin into the circulation.

What do we make of this?

The outcome tells us that this ECG was the patient's baseline ECG.  This ECG, though there was ST Elevation, was this patient's NORMAL ECG!   He has normal variant ST Elevation AND an occlusive MI that is invisible on the ECG.

Remember that an ECG can be completely normal in the presence of OMI, especially in the circumflex distribution.

Remember that one of the manifestations of completely normal is normal variant.

Other similar situations:

A patient with ST elevation due to LVH can have OMI and present with his baseline ECG that shows pseudoSTEMI due to LVH.

A patient with ST elevation due to LV Aneurysm can have OMI and present with his baseline ECG that shows pseudoSTEMI due to LV Aneurysm.

Learning Point

The patient needed cath lab activation, but not because of the ECG.  He needed it because of refractory chest pain and elevated troponin, and also with a new inferior wall motion abnormality!

Sunday, January 20, 2019

Pre-existing Left Bundle Branch Block and Atrial Fib: what is alarming on this routine pre-procedure ECG?

An elderly patient had a pre-procedure ECG (ECG-1).  She was asymptomatic.  The patient had known Left Bundle Branch Block (LBBB) and atrial fibrillation (see ECG-2 below), and was rate controlled on metoprolol. 

What is the problem?

ECG-2 (previous for comparison):

There are enormous U-waves, best seen in V1-V3, but also in V4 and V5.  This was seen by the overreading cardiologist.  The cardiologist called the The K was checked and it was 1.9 mEq/L.

The patient was sent to the ED for hypokalemia.

While being assessed and having potassium orally and IV, her heart rate dropped to 30 beats per minute (2 second pauses).  She was dizzy whenever not supine (symptomatic).

This 12-lead ECG (ECG-3) captures her at a heart rate in the 40s, but the monitor at times had a much slower rate.

Still in atrial fibrilation with a slow ventricular response and LBBB, and with large U-waves.  This implies the AV node is not working well.

What do you want to do?

This is what we did:

We gave atropine and the heart rate responded perfectly, with rate rising to 70-80.

After replenishment of KCl, orally and IV, the heart rate stabilized and the etiologies of the hypokalemia were also corrected.

The patient was discharged the next day.

Severe hypokalemia causes life threatening dysrhythmias.  At exactly what level one is at risk of such dysrhythmia is uncertain, but certainly any K below 2.0 is life threatening.  Probably under 2.5 is life threatening.  I believe that any hypokalemia that results in ECG abnormalities such as large U-waves, long QT, or ST-T abnormalities is potentially life threatening.

There are few reports of AV block due to hypokalemia, but it has been described.  It is also rare to see U-waves in the setting of LBBB.

Two case reports of hypokalemia associated with AV block

Hypokalemia Associated With Infra-His Mobitz Type Second Degree A-V Block

Atrioventricular conduction disturbances with hypokalemia in renal tubular acidosis

See this post for an in-depth discussion of several cases of severe hypokalemia and what to do about it:

STEMI with Life-Threatening Hypokalemia and Incessant Torsades de Pointes

Saturday, January 19, 2019

Anterior STEMI Evolves to de Winter's T-waves

A middle-aged man called EMS for chest pain.

This prehospital ECG was recorded:
Obvious Anterior STEMI due to proximal LAD occlusion (with STE in aVL and reciprocal STD in inferior leads).

On arrival to the ED, this ECG was recorded 10 minutes later:
Almost all STE is gone, but the hyperacute T-waves remain

While waiting for the cath team, this was recorded 30 minutes after 2nd ECG:
Now there are classic de Winter's T-waves.
This shows the dynamic nature of coronary thrombus.  Presumably, the thrombus had autolysis to a very small degree, allowing a trickle of blood flow through the LAD, enough to eliminate the ST elevation.

The patient was found to have a 100% proximal LAD occlusion.

By the time of the angiogram, which is never at the same the time as the ECG, there was not even a trickle of blood.

See this related post, with discussion of de Winter's T-waves:

Is the LAD really completely occluded when there are de Winter's waves?

Wednesday, January 16, 2019

Inferolateral STEMI: is there right ventricular MI also?

A middle-aged patient called 911 for 1 hour of chest pain.

He was hemodynamically stable.

Here is the prehospital ECG:
Obvious inferior MI, but also with STE in V3-V6

Here is the first ED ECG:
Again, inferior and lateral STEMI.
Is there any right ventricular (RV) MI?

85% of inferior MI are due to RCA occlusion.  Even the majority of inferolateral MI are due to RCA occlusion. The RCA may have a large lateral branch.

RV MI is caused by RCA occlusion proximal to the RV marginal branch, especially when there are no collaterals from the LAD to the RV.

We showed that, if there is zero ST depression in lead V2, that lead V1 is 85% sensitive for RVMI.  When there is STD in V2, only 35% have any STE in V1.

In Inferior Myocardial Infarction, neither ST elevation in lead V1 nor ST depression in lead I are reliable findings for the diagnosis of right ventricular infarction.

We should have titled it: In Inferior Myocardial Infarction, for diagnosis of RV MI, ST elevation in lead V1 is only sensitive in the absence of ST depression in V2; lead I is not reliable.

We care about RV MI because it can cause RV failure, with shock and hypotension.

But only a minority of patients with RV MI have these adverse hemodynamics, and an even smaller number with proximal RCA occlusion have these hemodynamics because they may have blood supply from LAD collaterals.

Is it important to diagnose RV MI here?

Probably not so much.  He is hemodynamically stable.

Nevertheless, we took a look with a right sided ECG.  Here it is:
V1-V6 are really V1R (=V2) to V6R
There is STE in V3R to V6R.
This is diagnostic of RV MI.  This is one of the 15% of RVMI that have zero STD in V2 but no STE in V1.
Perhaps the absence of STE in V1 (=V2R) is a good prognostic sign?
Perhaps it means that the critical anterior wall of the RV is not ischemic, even if more right lateral portions are ischemic?  This is pure speculation.

The patient went to the cath lab and had a proximal RCA occlusion opened and stented.  He was never hypotensive.

Here is the post PCI ECG:

Acute Inferior ST elevation MI.
Culprit is 100% occlusion of the Proximal RCA .

LMCA: The LMCA has mild plaque.
LAD: No angiographic significant obstructive disease.
LCx: No angiographic significant obstructive disease.
RCA: RCA has Normal take off.  The Proximal segment of the RCA has 100% disease.
Lesion on Prox RCA:


The estimated left ventricular ejection fraction is 41 %.
Left ventricular hypertrophy concentric .
Regional wall motion abnormality-distal septum and apex.
Regional wall motion abnormality-distal inferior wall.
Regional wall motion abnormality-anterolateral.


Findings c/w ischemia/MI in vascular territory of LAD, proximal to a diagonal branch.
(However, concomitant MI/ischemia in distal PDA/watershed territory cannot be excluded)

Learning points:

1. Inferior STEMI is usually caused by RCA occlusion
2. RCA occlusion may be proximal to the RV marginal branch
3. Proximal occlusions may or may not result in ECG or hemodynamic evidence of RV MI
4. When there is ST depression in lead V2 AND a proximal RCA occlusion, there is STE in V1 approximately 85% of the time.
5. STD in Lead I is not useful to diagnose RV MI except to differentiate RCA from circumflex occlusion.  (RVMI can only occur with RCA occlusion)
6. Diagnosis of RV MI is important in the presence of hemodynamic instability

Sunday, January 13, 2019

What will you do for this altered and bradycardic patient?

Written by Pendell Meyers

A female in her 60s with COPD, DM, hypothyroidism, CAD, and severe bladder cancer presented from a nursing home with altered mental status, hypotension, hypoxia, and bradycardia.

Here is her initial ECG (no prior for comparison):
What do you think?

Here is another ECG minutes later:
There is a regular wide complex bradycardia.
There are P-waves at a rate of approximately 100bpm with no clear relationship to the QRS complexes, diagnostic of complete heart block (see correction below!).
The QRS morphology is wide (computer QRS duration 179 msec) but it does not fit any clear bundle branch block pattern (it is similar to LBBB but not truly LBBB).
The T-waves in the precordial leads are peaked. Overall this is highly concerning, if not diagnostic, for hyperkalemia.

Important correction by Dr. K Wang:
"By the way, it's not complete AV block right away because the QRSs do not occur regularly.  In complete AV block, the QRSs occur regularly whether its junctional or ventricular escape rhythm. In this case, actually QRSs have a P wave in front of them with an identical PR interval, revealing 3:1 ,4:1 or higher degree AV conduction ratio. Regardless, the most important thing clinically in this patient is to recognize  that the T wave of the V3 in the first tracing is tented, pointed, highly suggestive of hyperkalemia, a life threatening condition, which it turned out to be. Yes, the pacemaker induced QRSs are very wide and the T waves are very tall, again making one to suspect hyperkalemia."

Heart block (of some sort) was recognized, but hyperkalemia was not initially. She was given push dose epinephrine and atropine with no change. External pacing was attempted but failed.

A transvenous pacemaker was then placed, with capture obtained:
Ventricular paced rhythm. 
Computerized QRS duration 192ms, but it appears to be ~200 msec in several leads. 
It is unusual for any patient with any conduction pattern to reach 190-200 msec in the absence of hyperkalemia or other sodium channel blockade. 
The J-point is not easy to find, but there is excessively discordant ST deviation in many leads (this is why we excluded hyperkalemic patients in our ventricular paced rhythm study of acute coronary occlusion using the modified Sgarbossa criteria!). 

The first potassium level was reported to be hemolyzed by the laboratory.

She was placed on an epinephrine drip. Her mentation improved after pacing and epinephrine drip.

The second potassium then returned at 8.1 mEq/L. Labs confirmed new renal failure.

Around this time the patient's family arrived and explained that the patient's goals of care were palliative. Comfort care was instituted. The patient expired.

Learning Points:

Hyperkalemia is an important cause of bradycardia and heart blocks.

Never place a transvenous pacemaker without considering hyperkalemia and a trial of IV calcium.

Comment by KEN GRAUER, MD (1/13/2019):
Superb case with essential teaching points by Dr. Pendell Meyers that have been emphasized many times on Dr. Smith’s blog, but which never cease to be insightful and instructive. As stressed by Dr. Meyers — HyperKalemia is an important cause of bradycardia and heart blocks! It needs to be given high priority whenever clinical circumstances are potentially consistent with this diagnosis.
  • The goal of My Comment is to illustrate some advanced pointers in clinical arrhythmia interpretation. Although none of these pointers were “needed” for the essentials of diagnosis and management of this case — awareness of these ECG & Arrhythmia PEARLS may prove invaluable in evaluation and management of other cases.
  • For clarity — I’ve labeled the first 2 tracings in this case (Figure-1):
Figure-1: The first 2 ECGs in this case (See text).

My thoughts ( = my opinions) are the following:
  • As per Dr. Meyers — the rhythm in the initial ECG (ECG #1) is slow and irregular at a rate between 30-40/minute. The QRS is very wide. Baseline artifact prevents reliable recognition of atrial activity. I think it important to recognize when artifact impedes identification of the rhythm — and to state this in your written interpretation.
  • Under normal circumstances (ie, when the patient is not markedly hyperkalemic) — one of the most helpful hints for recognizing that a rhythm is not complete AV block, is if the ventricular response is not regular. Very often — beats that occur earlier-than-expected in an otherwise regular rhythm are conducted. That said, I’ve found that this general rule does not hold true well when there is marked metabolic abnormality, as is the case here. Therefore, given the absence of reliable atrial activity in ECG #1 — I don’t believe any conclusion can be reached regarding the rhythm beyond stating there is a marked, somewhat irregular slow rhythm with a wide QRS.
  • To highlight the point Dr. Meyers made regarding T wave morphology — I’ve enclosed within RED dotted rectangles the QRS complexes in leads V3 and V4 of ECG #1. Note that the T waves of these 2 QRS complexes are exceeding thin, pointed at their peak, and display a narrow base. It is these 2 T waves that should immediately heighten your suspicion of hyperkalemia. T waves elsewhere on both of these tracings show more-than-you-might-expect T wave peaking — but none are as dramatically suggestive of hyperkalemia as the T waves within the RED dotted rectangles.
  • QRS morphology in ECG #1 is potentially consistent with LBBB — because there is an all upright QRS complex in lead I — there is predominant negativity with fairly steep S wave downslope in anterior leads — and there is a positive (albeit small) QRS in lead V6. Some patients with LBBB do not transition to an all-positive R wave until leads V7 or V8 — so the pattern in ECG #1 is consistent with LBBB.
  • Atrial activity is clearly present in ECG #2! Although there is still significant baseline artifact — RED arrows across the long lead V1 rhythm strip show a fairly regular atrial rhythm at a rate just over 100/minute. I’ve placed a GREEN arrow at the point in the long lead V1 strip where a P wave is not seen in lead V1 — but — we know that a regular P wave does occur at this point, because we can clearly see it in simultaneously-recorded leads aVL and aVF (vertical BLUE lines). PEARL — If you are not already doing so, LEARN to use simultaneously-recorded leads in assessment of challenging arrhythmias!
  • There are only 3 QRS complexes in the long lead V1 rhythm strip of ECG #2. That said, the PR interval (horizontal thick PURPLE line) is the SAME for each of these 3 beats. Whether this is by chance … — or whether it indicates conduction of these 3 beats can not be determined from this single tracing. But IF a longer period of monitoring were to confirm that the PR interval in front of each QRS complex that appears is identical to that seen in ECG #2 — then the rhythm would be Mobitz II, 2nd-Degree AV Block (and not complete AV Block). And if the rhythm in ECG #2 is Mobitz II — then we are seeing a manifestation of why Mobitz II is a potentially lethal rhythm, because there is no QRS complex at all for the last 3 seconds of this ECG!
  • NOTE: As I said at the outset — determination of the precise rhythm in these 2 tracings is not clinically essential, because regardless of what the rhythm may be, this patient has life-threatening bradycardia with some form of AV block as a result of potentially treatable hyperkalemia. My point is simply to illustrate how we can not distinguish between an irregular idioventricular rhythm vs some form of AV block in ECG #1 — and, how we can not distinguish between high-grade Mobitz II vs something else in ECG #2.
  • QRS morphology in ECG #2 remains potentially consistent with LBBB. I’ve drawn in vertical BLACK lines to illustrate the point of lead change. Note that beat #1 in ECG #2 occurs right at the point of the first lead change — and that the QRS is all positive in lead I (PURPLE arrow). The QRS manifests a qR pattern in lead aVL in ECG #2, which is also consistent with LBBB — there is again predominant negativity in anterior leads, with steep S wave descent (which is consistent with LBBB) — and no QRS complex at all occurs in leads V4,V5,V6. Whether or not this patient has underlying LBBB vs a pattern of QRS widening solely from hyperkalemia can only be determined if a repeat ECG is done after K+ has normalized.

Friday, January 11, 2019

The Interventionalist Refuses Angiography, and even to speak to the Emergency Physician

A recent residency graduate, let's call her "The Graduate" or "TG," texted me these ECGs from somewhere far away across the country, in real time, in the hopes of being able to persuade the interventionalist to take the patient to the cath lab.


An otherwise healthy middle-aged patient presented with chest pain of uncertain duration.

Here is the initial ED ECG (I apologize for the poor quality of these images -- they were mobile phone photos of computer screens, texted to me -- but they are good enough!):
What do you see?
There was a second ECG after the patient had symptom resolution:
What do you see?

This is what TG and I saw: ST elevation in lead V1 during pain (with a tiny amount of terminal T-wave inversion), then resolution of STE and more prominent T-wave inversion after resolution of pain (ECG 2).  Looks like Wellens' in V1 only!  This ECG is typical of the rare isolated right ventricular STEMI.

There is also some very minimal inferior STE, with reciprocal STD in aVL, which is gone after reperfusion (ECG 2).  The very narrow-based, peaked T-waves would be an unusual manifestation of MI.

There is quite a bit of diffuse T-wave peaking.  Hyperkalemia can result in ST elevation in V1 and V2, so one might wonder about hyperkalemia, although such ECGs really look very different from this, especially because the QRS should be prolonged.  In our case here, the K was normal.

Here are examples of ST elevation in V1 due to hyperkalemia.

Hyperkalemia and ST Segment Elevation, Post 1

Case Continued

At this point, the first troponin I returned at 1.1 ng/mL.  The potassium was normal.

The pain returned and a 3rd ECG was recorded:
STE in V1 recurs.  Again, there is some minimal inferior STE, with reciprocal STD in aVL.  Very suspicious for inferior and RV OMI (Occlusion MI)
The very narrow-based, peaked T-waves are an unusual manifestation of MI.

This is the point at which TG texted me.

The interventionalist did not want to take the patient to the cath lab.  Antiplatelet and antithrombotic therapy had already been initiated.

TG wrote: "By being persistent, I'm not making friends with cardiology.  They are refusing cath.  They say to treat the MI medically only."

I wrote: "Tell him that by the recommendations of the American College of Cardiology and the American Heart Association, a patient with persistent symptoms in spite of medical treatment who is having acute coronary syndrome should go emergently to the Cath Lab. This is by the recommendations of their own societies.  ACC/AHA Guidelines for the management of Non ST Segment Elevation  Myocardial Infarction. And this is regardless of the ECG findings."

"Moreover, in this case you have ST Elevation in V1.  This is an RV infarct."

I continued:

"The European society of Cardiology recommends emergent and angiography for patients even with normal biomarkers if you believe they are having acute coronary syndrome with refractory symptoms."

"All studies randomizing patients to early versus delayed catheterization exclude patients with refractory symptoms, and this patient has refractory symptoms."

"This is because people can have a major coronary occlusion and no EKG findings."

"And this one has EKG findings!"

TG responded:  "So far he is refusing to talk with me and only wants my name and refusing to give recommendations."  

I replied: "Is there any code of conduct in your hospital? It's hard to believe that he can refuse to talk to you."

She replied: "I don't know - I'm new here.  Thanks for the backup!"

Smith: "Any time. Sorry can't fix personality problems!"

Would an emergent echocardiogram have helped?  Not necessarily.  The cardiologist recognized it was an MI but only wanted medical therapy; he/she probably would have expected a wall motion abnormality.

She: "Update: new interventional is just came on and will take to cath lab. Still in pain.  Here is a 4th EKG:"
Now there is an obvious inferior STEMI, with persistent right ventricular MI.  The ST segment in V2 shows posterior involvement as well.
The patient was taken to the cath lab and found to have an acute proximal RCA occlusion.

Learning Points:

1. See all of the above.
2. Hospitals should consider adopting a Code of Conduct.

Code 1: One of the elements would be that a consultant must at least speak to any physician asking for help in caring for a patient.

Code 2:  When there is disagreement on critical next steps in emergency patients, there needs to be some mechanism to rapidly resolve the disagreement.

Comment by KEN GRAUER, MD (1/11/2019):
I don’t have much to add to the excellent discussion by Dr. Smith of this case:
  • Perhaps right-sided leads could have helped to convince the 1st interventionalist that this case merited prompt cardiac catheterization with reperfusion therapy? Given the dramatic amount of ST elevation in lead V1 — it would seem that there almost certainly would have been extensive ST elevation in sequential right-sided leads, that would have painted an even more impressive picture. Then again, given resistance by this initial interventionalist to even talk to the treating physician — nothing might have been “enough” to convince him of the need for prompt cardiac cath ... (P.S.: It should be emphasized that in this case — right-sided leads are not needed to make the diagnosis of acute RV MI, because V1 alone is sufficient. I'm simply suggesting that a dramatic right-sided lead picture might help to convince a skeptical cardiologist ... ).
  • As per Dr. Smith — the finding of tall, peaked T waves with narrow base in multiple leads (8/12 leads on the 1st ECG) is an unusual manifestation of acute MI. Serum K+ was normal — confirming that these tall, narrow T waves were indeed a “hyperacute” finding. I’ve not before seen T waves so prominent and pointed with such a narrow base in so many leads, in association with a normal serum K+ in a patient with acute OMI. Though not common — this is a picture worth remembering!
  • Final Thought: It is not easy when you are the clinician “on the front line” — and your consultant cardiologist, surgeon, or other specialty physician that you have consulted refuses to engage in active dialog with you about your patient. Fortunately, a 2nd interventional cardiologist took over this case. Otherwise: iCalling someone else, regardless of the hour (either medical chief of staff or another consultant) may be your only option; andiiDevelopment of a protocol for rapid resolution of such clinical impasses needs to take 1st priority at the very next medical staff meeting.
  • Important lessons to be learned in this case! Our thanks to Dr. Smith for presenting it!

Recommended Resources