Monday, October 31, 2016

An apparent SVT that does not persistently correct with adenosine

A middle-aged woman presented for abdominal pain.  She was found to have a heart rate of 150.  A 12-lead was recorded:
There is a regular, narrow complex tachycardia.
What is it?

Regular narrow complex tachycardias can be sinus tach, atrial tach, atrial flutter, AV nodal reentrant tachycardia, orthodromic AV reciprocal tachycardia [using a bypass tract (accessory pathway), that is WPW], or junctional tachycardia.

Here is a one hour lecture on Supraventricular Tachycardias; everything you ever need to know.

MAT and atrial fibrillation, other causes of supraventricular tachycardia, are not possible because they are always irregular, and this rhythm is regular.

There are no definite P-waves before the QRS and there are definite retrograde P-waves (see V2 below).
Here the black arrows point out the retrograde P-waves that are clearly present in V2
Once you find this, you go up to V1 and you can see that the retrograde P-wave is all negative.

--Thus, this cannot be sinus tachycardia or atrial tachycardia, which would have upright P-waves in V2.
               Suppose the P-wave was upright in V2 and in the same location in the middle of the QRS?  Then you would have to consider that there is sinus or atrial tach with a very long PR interval.

--When there is a narrow complex tachycardia, especially at a rate of 150, you should also look for flutter waves, which are not present.

--Junctional tachycardia is possible, as it also may have retrograde P-waves.  This is much less common, so let's discard this.

--Thus, we are left with re-entrant tachycardias that use the AV node (AVNRT and orthodromic AVRT).

This can be treated with vagal maneuvers, the best of which is Modified Valsalva, which has a success rate of about 40%.  More reliable is adenosine.

Case continued

A bedside echo showed good LV function.
The patient was given adenosine 6 mg without effect.  Then she was given adenosine 12 mg and had an asystolic pause (without underlying flutter waves), followed by one sinus beat, then one early narrow beat, followed by SVT again.  


Aside: if this were atrial flutter, you would see underlying flutter waves revealed when adenosine blocks the AV node and thus removes the QRS which obscures the flutter waves.  See this image which comes from a great article (the third of the 12 rhythms of Christmas) by Vince DiGiulio on EMS 12-lead and which Vince graciously permitted me to use:
Adenosine blocks AV conduction of flutter to ventricle, resulting in a ventricular escape rhythm and uncovering the flutter waves.

This did NOT happen in our case here.

So what happened?  Did adenosine work?

Yes!  Adenosine's only job is to stop the AV nodal re-entrant circuit.  Its effect is of extremely short duration, so it will be gone immediately and be unable to terminate a subsequently initiated episode of Paroxysmal SVT.  So one may have to prevent initiation of PSVT.

How does PSVT start?  PSVT requires 2 pathways.  In AVRT, one of them is the accessory pathway.   In AVNRT, there are 2 pathways in the AV node, a fast pathway with a long refractory period, and slow pathway with a short refractory period.  PSVT starts with a premature atrial beat (PAB) that reaches the AV node when the fast pathway is still refractory, and the slow one is not.  Then it travels down the slow pathway to the inferior part of the node where it then finds the fast pathway is no longer refractory and ascends that fast limb.  When it gets to the top it goes to BOTH the atrium and back down the slow pathway. Then around and around. For a more detailed explanation with an image, watch this lecture from minute 3:18 to 4:54.

PSVT is stopped by stopping the circuit.  If it recurs, further recurrence can be prevented by either 1) having a continuous AV blockade, as with a longer acting medication such as a calcium channel blocker or 2) preventing PABs.  PABs can be prevented by beta blockers or amiodarone.

Beta blockers may be detrimental in patients with poor LV function because of their negative inotropic effect.  In such cases, I have tried esmolol, as it has a short half-life and can be turned off.  I did not have success.   See this case:

Paroxysmal SVT (PSVT) that repeatedly recurs in spite of successful conversion with adenosine.

In that case in which esmolol did not work, I then gave amiodarone bolus of 150 mg, followed by an infusion, after which intervention the patient stayed in sinus rhythm.

In this case today, the adenosine worked, but within one beat, there was another PAB that initiated the PSVT again.

What to do?  Amiodarone?  Calcium channel blocker?  Something else?  As this patient has good LV function, a calcium channel blocker is an excellent and safe choice.  Before adenosine, verapamil was the standard way to terminate PSVT.

Verapamil IV dosage: the recommend IV dose of verapamil is 5 mg over 2 minutes, followed in 5 to 10 minutes (if no success) by a second 5 to 7.5 mg dose.
Diltiazem IV dosage: The recommended IV dose of diltiazem is 20 mg followed, if necessary, by a second dose of 25 to 35 mg.
An oral dose may be given for ongoing maintenance.

Case continued:

In this case, 2 g of Mg (8 mmoles, 16 mEq) was given by one of my excellent partners, Laura Shrag, and the patient converted!

Here is the post conversion ECG:
What is this? Sinus rhythm?

This was read as sinus rhythm, but it looks more like an ectopic atrial rhythm.  The P-wave in V1 is upright, whereas it should be biphasic.  This means that lead V1 is to the left of the source of the P-wave.  Either the heart is far to the right, or V1 was placed too far left, or there is a left-sided atrial source.  In any case, it is a stable rhythm at a good rate and we shouldn't worry about it any more than that.

--There are no delta waves, so this is less likely to be WPW, though absence of delta waves does not rule out WPW.
See explanation of this at these two posts:

--A large R-wave in lead V1. And why is the PR interval not short?

Also of concern: this patient had tachycardia and was completely unaware of it. She presented only for abdominal pain.  This implies that she is not aware of her PSVT and may live with it for long periods of time.  Long duration tachycardia may lead to cardiomyopathy, so it is particularly important for her to be on long term medications, or have an ablation, so that this does not recur.

Magnesium in PSVT:

Magnesium may be loosely thought of as a nonspecific Calcium channel blocker.  I was unaware of its effect on PSVT until this case.  There is data to show that Magnesium, in a dose of approximately 0.15-0.30 mEq/kg (10.5- 21 mEq per 70 kg, or 1.3-2.6 grams per 70 kg) is often effective for termination of PSVT with a low incidence of mild side effects (transient sensation of warmth, flushing, and diaphoresis).
10 mmoles = 2.5 g = 20 mEq

In the electrophysiologic (EP) literature below, it had variable effects on termination, but consistently prolonged antegrade AV node conduction, which is a necessary component to termination. 
1. In an EP study, after a bolus of 0.3 mEq/kg, followed by a 0.2 mEq (0.025 gm)/ kg/hr maintenance infusion, Mg was shown to increase the tachycardia cycle length by about 10% by slowing the antegrade conduction in the AV node of the re-entrant circuit.  However, it did not appear to convert any of the 4 AVNRT studied or any of the AVRT.  It did not prevent inducibility of SVT or affect the refractory period:

2.  In this EP study, 8 of 10 episodes of SVT that involved the AV node (4 AVRT and 6 AVNRT) were terminated with 2 grams of MgSO4 by rapid bolus, flushed with saline.

3.  In this dose-finding EP study, prolongation of the atrial-His interval was no more pronounced at doses higher than 5 mmols (=1.25 g) than at 5 mmoles:

4.  In this EP study, Mg was effective in termination of 5 of 15 with dual AV node physiology and 0 of 8 cases without dual AV nodal pathways: 

See this post for an explanation of dual AV nodal pathways: 

Wide Complex Tachycardia in a 20 something.

5.  Here is a case report of termination of SVT in the ED, from Annals of EM in the 1990s:


The patient was started on Diltiazem CD 120 mg daily.

Learning Points:

1. SVT that is converted by adenosine to sinus may revert back to SVT.
2. SVT is initiated by PABs.  Prevention of PABs helps to prevent recurrence.
3. Calcium channel blockers are very effective in conversion to, and maintenance of, sinus rhythm.
4. Especially in patients who cannot tolerate calcium channel blockers, it is reasonable to attempt an IV bolus of Mg at 1.25 to 2.5 g.
5. In patients who do not respond to Mg and cannot tolerate calcium channel blockers, amiodarone 150 mg followed by an infusion, may be effective in preventing further episodes of PSVT.  One may have to additionally convert the patient to sinus with adenosine.
6. Chronic tachycardia may lead to cardiomyopathy

Friday, October 28, 2016

Is this STEMI? LVH? Early Repolarization?

This 60-something with diabetes presented with abdominal pain, nausea, and vomiting.

We all know that diabetics with acute MI may present like this (see this case: "I have food poisoning").

An ECG was recorded:
This was the automated interpretation:
***ACUTE MI***
Is this Acute MI?

--There is much "anterior" (right precordial) ST elevation.
--There is also high voltage, which suggests LVH.  We know that LVH causes false positive ST elevation.
--Leads I, II, aVL, aVF and V5 and V6 have the typical LVH "hockey stick" pattern of discordant ST depression and T-wave inversion.
--However, right precordial ST elevation due to LVH should be discordant to a deep S-wave. There is no deep S-wave in V2 and V3.   So this is somewhat atypical.  
--There is also a saddleback in V2 and these are rarely due to STEMI.
--Sometimes LVH may be combined with early repolarization: high voltage and typical lateral leads on the one hand, but also right precordial leads that look like early repol with well-formed J-waves on the other hand.

Here is an extensive discussion of the LVH pseudoSTEMI phenomenon:

LVH with anterior ST Elevation. When is it anterior STEMI?

Can we use the early repolarization vs. LAD occlusion formula to help?
In the setting of LVH, this often leads to false positives, but should not lead to false negatives.  
Here it is:
Formula: (STE 60 ms after J-point in lead V3, computerized QTc-Bazett, R amplitude in V4)
Computerized QTc = 452 ms
Formula value (3, 452, 45) = 15.58
Formula value (4, 452, 45) = 16.78
These are very low values, far below 23.4, and thus very unlikely to be due to STEMI

Better yet, look for an old ECG.  Here is one from 1 week prior:
No significant difference.  This all but confirms that the first ECG is not due to STEMI.

However, you can also be fooled by previous ECGs:

Here is one from 3 months prior:
Much less STE.  This would lead you to believe the STE on the present ECG is due to STEMI
If this one had been found, one might think the presenting ECG is in fact due to STEMI.

And here is one from 1 year prior:
This is somewhat intermediate, but the computer also read STEMI.

The patient ruled out for MI.  He was, in fact, septic.  Going to the cath lab would have been detrimental to his health.

Learning Point:

Be familiar with this pattern of combined LVH/Early repolarization.  It will fool you.

Here is another example of it:

Wednesday, October 26, 2016

RSR' with ST elevation: is this Right Bundle Branch Block with STEMI? Type 2 Brugada?

A very healthy 40 year-old man went to his primary care physician for a check-up. He denied any chest pain, dyspnea, or exertional intolerance, but an ECG was obtained:
The ECG is fairly normal with the exception of the rSr’ pattern and ST segment elevation in V1 and V2. The QRS duration is 108 milliseconds, consistent with incomplete right bundle branch block (RBBB).  The ST elevation in V1 is less than 2 mm, and so does not meet criteria for type 1 Brugada.
The "saddleback"
 in V2 suggests RBBB with STEMI, although very unlikely (see 2 cases below).
Type 2 Brugada might also be considered, but
the beta-angle is far too acute

Notice the P-wave in V1 is all negative.
As you can see, the computer warned of possible STEMI

See these 2 posts:

ST elevation (Saddleback), is it STEMI?

Is this Saddleback a STEMI??

Of course, he was sent by ambulance to the ED for this possible STEMI.

The ECG was immediately repeated in the ED: 
The rSr’ pattern is not seen in V1, but it is seen in V2, with ST elevation. Such ST elevation in the presence of RBBB, or incomplete RBBB, might strongly suggest RBBB with STEMI. 
However, a saddleback STE in V2 is rarely due to STEMI. 
The P-wave in V1 is mostly negative.
What is going on?

After review by a physician, the cath lab was not activated.

The patient’s history was not suggestive of either ischemia or occult Brugada, and the physical exam demonstrated only an athletic physique.

The ED physician suspected upwards misplacement of V1 and V2 as the cause of the ECG findings. Such misplacement is very common - Wenger and Kligfield found that > 40% of V1 and V2 placements were too high.

Upwards misplacement of V1 and V2 can generate false Q waves, poor R wave progression, and ischemic patterns. Such misplacements can also easily produce psuedo-Brugada patterns, especially in athletic males.

There are certain clues to this upwards misplacement of V1 and V2. Garcia-Niebla analyzed 101 normal subjects, and found that the P waves provided most of the evidence.

Reexamine the first ECG:
The P is wholly negative in V1, and V1 also has a rSr’ morphology. This only happened when V1 and V2 were placed in the second intercostal space, almost at the clavicles!

The ED physician asked for the ECG to be repeated, and placed the precordial leads himself.
(Written consent obtained from patient)
With proper lead placement, the ECG was repeated:
This last ECG shows a normally biphasic P in V1, and an upright P in V2, confirming proper position. The rSr’/IRBBB pattern has disappeared, along with the saddle-back complex. Some STE remains, but is normal variant STE (early repolarization).

Learning Point:

1.  Lead misplacement can result in abnormalities that simulate pathology.
2.  RSR' in lead V1 is particularly common and due to high lead placement of leads V1 and V2
3.  Saddleback in V2 is also common
4. Analysis of the P-wave morphology greatly aids in assessing good lead placement

Monday, October 24, 2016

Anaphylaxis and ST Elevation

This was written by one of our excellent 2nd year residents, Nathan Ansbaugh, with some editing and commentary by Smith.

An otherwise healthy middle-aged male age presented as an outpatient for a routine MRI. Immediately after receiving the IV contrast load, he became anxious, nauseous, flushed, and proceeded quickly to become apneic and pulseless.

The emergency response team arrived approximately 2 minutes later and found the patient pulseless and in presumed cardiac arrest. They began CPR. They gave 1mg epinephrine with immediate ROSC and the patient was transported to the Emergency Department for further stabilization and further treatment of presumed anaphylaxis.

An initial ECG was recorded shortly after arrival:
There is sinus tachycardia. There is right bundle branch block (RBBB) with additional left posterior fascicular block (LPFB). Aside: As for LPFB, note the small inferior Q-waves and the rightward axis to approximately +140 degrees, even when disregarding the last 40 ms of the QRS, which is due to the late depolarization of the right ventricle due to RBBB.) Most concerning is the ST elevation in inferior leads, concordant to the majority of the QRS in leads III and aVF, with concordant reciprocal ST depression in I and aVL. This is diagnostic of inferior STEMI.

The prior ECG from 3 months prior was obtained for comparison:
The RBBB and LPFB are old, but this ECG shows no sigificicant ST elevation.
Therefore, the ST elevation on the presentingn ECG is new.

RBBB should not have any significant ST elevation. And here we confirm that it is new relative to the old ECG. So this is new (ischemic) ST elevation.

Should the cath lab be activated?

Approrpiately, the cath lab was not activated.

Instead, the ECG was repeated 10 minutes later:
There is still mild ST elevation the inferior leads but this appears to be somewhat improved.

ECG repeated at 20 minutes:
Inferior ST elevation has essentially fully resolved at this time.

Is this transient thrombosis? Or is it type 2 MI that is resolving as the supply and demand issues improve? It could be either, but type 2 MI is more likely. Both are reported in the literature (see references below) in the context of epinephrine to treat anaphylaxis.

Cardiac arrest can also, of course, lead to type 2 MI, or type 2 STEMI, due to poor perfusion. [Aside: pulselessness is not good evidence of cardiac arrest. Rather, in this case, it is likely that there was severe shock and hypotension such that pulses were hard to palpate. There was no bedside ultrasound to assess cardiac function, and, as far as I know, no monitor showing asystole or wide complex. Probably, there was a narrow complex tachycardia with extremely weak pulses.]

In any case, such poor perfusion can lead to ischemic ST elevation, mimicking ACS STEMI.

Due to the possibility that it was thrombosis, cardiology was contacted for possible angiography +/- PCI.

The patient was admitted to the MICU for anaphylaxis and evaluation for possible ACS. The peak troponin was 1.710 ng/mL. The patient underwent coronary angiography prior to hospital discharge and this showed only mild plaque without angiographic evidence of significant obstructive CAD or cultprit lesion or thrombosis. Echocardiography did not demonstrate wall motion abnormality, and showed normal systolic function and normal EF.

Learning Points:
1. Cardiac Arrest or hypotension can result in ischemic ST elevation without coronary thrombosis
2. Anaphylaxis treated with epinephrine is associated with ischemic ST elevation without coronary thrombosis.
3. Not all electrocardiographic STEMI is due to ruptured plaque and thrombosis. Only the clinical scenario +/- angiography can determine the etiology of the ischemic ST elevation.

Selected Literature

STEMI due to spasm after epinephrine for anaphylaxis:

STEMI with thrombus after epinephrine for anaphylaxis:

Supply/Demand Type 2 Myocardial Infarction: should we be paying more attention?
Yader Sandoval, Stephen W. Smith, Fred S. Apple
JACC 63(020); May 2014.

Myocardial Infarction and Type 2 Myocardial Injury
Yader Sandoval and Kristian Thygesen
Clinical Chemistry (online now)

Here is a classification of Acute Myocardial Injury:

Saturday, October 22, 2016

Sepsis with Pulmonary Edema and Elevated Right Sided Pressures

I was reading from our list of unconfirmed ED ultrasounds and saw a cardiac ultrasound with good LV function, but with B-lines of pulmonary edema, pleural effusions, and a very dilated inferior vena cava. 

So I went to look at the chart.

It was a chronically ill patient with sepsis, a heart rate of 120, presumed sinus tachycardia, and did not have chronic renal insufficiency.

But there was no ECG recorded.

It did not make sense to me that someone who was septic would have ultrasound evidence of fluid overload unless they had renal failure, which this patient did not.

Then it turned out that an ECG was recorded upstairs, but only after admission:
What is it?

This is atrial flutter at a rate of 120.
This was not only recorded late, but misread as sinus tachycardia.
The giveaway is the upright P-wave in lead V1.
Sinus P-waves always have a negative component.  The right atrial component comes first and is upright, the left atrial component comes next and is negative, such that the P-wave in V1 should be biphasic.  Upright P-waves in V1, as in this ECG, are classic for atrial flutter.

See these posts for slow atrial flutter:

--Tachycardia with Pericardial Effusion

The clinicians probably did not consider that tachycardia at a rate of 120 in a septic patient might be something other than sinus tachycardia.  

Just recording a 12-lead on an ill patient may reveal the unexpected.  Atrial flutter commonly comes in atrial rates as slow at 240 (and ventricular rates as slow as 120), and even slower in the presence of sodium channel blocking drugs.

Also, pulmonary edema should raise high suspicion of a cardiogenic cause, and this is due to either pump function (systolic vs. diastolic), valvular dysfunction, or dysrhythmia.  The patient had good pump function, and sepsis usually leads to volume depletion, or relative volume depletion.  Thus, the pulmonary edema had to be either noncardiogenic (or pneumonia), or due to either dysrhythmia or valvular disease.

Most dysrhythmias are easily diagnosed by ECG.

Learning Points:

1. Always record a 12-lead on a sick patient
2. Flutter can be at rates much lower than you expect.
3. Flutter often mimics sinus tachycardia
4. Fluid overload must be explained, and a cardiac cause is very likely.  This is especially true in clinical situations in which volume depletion is expected.

Friday, October 21, 2016

Atrial Flutter. What else??

A 50-something presented with bradycardia.  No other history is available.  Here is the EKG:
There is obvious atrial flutter.  The ventricular rate is 29.
The QRS has the morphology of Right Bundle Branch Block (RBBB) and Left Anterior Fascicular Block (LAFB)

There is also a very long QT, primarily due to a long ST segment
What is really going on?

This is NOT really RBBB and LAFB.  The flutter waves are not conducting through the AV node.   This is not Flutter with 7 or 8 to 1 conduction. Instead, there is third degree (complete) AV block.

How do I know?  When atrial flutter conducts, the QRS should occur at the same part of the flutter wave for every QRS.  Every time the flutter circuit goes around the atrium it arrives at the AV node at the same part of the wave and, if the AV node is ready to conduct, that is when it conducts.

If you look closely, this does not happen:
On beat 1, the QRS starts at the peak.
On beat 2, it starts earlier
On beat 3, it starts earlier still
On beat 4, it starts just after the peak
Beat 5 looks like beat 2.
Instead, there is a very regular escape rhythm

Thus, there is complete AV dissociation.

[There are times when there is atrial flutter and Wenckebach, and the lengthening PR interval changes this otherwise fixed relationship between flutter waves and QRS, but it is far more rare than complete block AND it does not occur with 6:1 block.]

So this is atrial flutter with complete AV block and a regular escape rhythm.

Is it a junctional escape?  A junctional escape should be around a rate of 40, and it should be narrow, unless there is a junctional escape + RBBB + LAFB.  This is possible.

Much more likely is that this is a ventricular escape.  Then why the RBBB and LAFB morphology? Because the escape is originating in the posterior fascicle.  When there is a supraventricular rhythm with RBBB and LAFB, the impulse gets to the posterior fascicle only, then spreads to the ventricle from the posterior fascicle.  Thus, RBBB + LAFB also, in a sense, originates from the posterior fascicle.  In this case, there is no supraventricular stimulus.  There is only the automatic escape, and it originates from the posterior fascicle.

Diagnosis: Atrial Flutter with complete, third degree, AV block.  The long ST segment is probably due to hypocalcemia.  Consider hyperkalemia.

Consider reversible causes (especially hyperkalemia or drug effect from beta blockers or calcium channel blockers before pacing).

Tuesday, October 18, 2016

LAD occlusion or Early Repolarization?

This male in his 20's presented with chest pain:
There is sinus rhythm with a PVC.
There is 2.5 mm ST elevation at the J-point, relative to the PQ junction, in both of leads V2, V3.
This meets the ACC/AHA criteria (for age less than 40) for anterior STEMI.
The ST segments in V2 and V3 are not upwardly concave (they are straight).

Is it STEMI?  Is it Early Repolarization (Normal Variant ST Elevation)?

We have derived a formula to help with this:

Go here for the formula:

Here is the study, a derivation and validation:

It is critical to use it only when the differential is subtle LAD occlusion vs. early repol. Thus, there must be ST Elevation of at least 1 mm. If there is LVH, it may not apply. If there are features that make LAD occlusion obvious (inferior or anterior ST depression, convexity, terminal QRS distortion, Q-waves), then the equation MAY NOT apply. These kinds of cases were excluded from the LAD occlusion group as obvious anterior STEMI.


--Bazett-corrected QTc is the computer measurement.
--RAV4 = R-wave amplitude, in mm, in lead V4.
--ST elevation (STE) is measured at 60 milliseconds after the J-point, relative to the PR segment, in millimeters.

Formula that was derived and validated:
(1.196 x STE at 60 ms after the J-point in V3 in mm) + (0.059 x computerized QTc) - (0.326 x R-wave Amplitude in V4 in mm).
A value greater than 23.4 is quite sensitive and specific for LAD occlusion.

A value less than 23.4 might still be LAD occlusion, but it is unlikely.  A value less than 22.0 is extremely unlikely to be LAD occlusion.

I and at least hundreds of other with whom I have been in contact have used this with a high degree of accuracy.

How to use it?  It does not rule in or rule out LAD occlusion.  Rather, it serves as a warning to evaluate intensively with use of clinical skills, serial ECGs, stat echo and, if needed, angiography.

For this ECG: Strictly speaking, the formula may not apply, as there is a straight ST segment in V2 and V3.  Thus, if we had seen such an ECG in the LAD occlusion group, it would have been excluded as "obvious" STEMI, as early repolarization should have upwardly concave ST segments.  

Unfortunately, there are exceptions to this rule.

Therefore, let's try to apply the rule.  But be very careful with the result.

If we do apply the formula, the numbers are:
--QTc = 410
--STE60V3 = 4.0
--RAV4 = 18
Score = 23.1, which is less than 23.4 and therefore most compatible with early repolarization.

It would be appropriate to do serial ECGs, troponin, and emergent echocardiogram to look for anterior, septal, and apical WMA.


The patient ruled out for myocardial infarction.  This ECG manifests an unusual early repolarization variant.

Recommended Resources