Regular Wide Complex Tachycardia with poor LV function and hypotension. Duration unknown. How to manage?

An older patient with no previous medical history arrived at triage complaining of SOB.  Her heart rate was very fast, so we obtained an ECG immediately:

ECG:

What do you think?

There is a regular wide complex tachycardia.  The QRS morphology is RBBB with LAFB (tiny r-waves in inferior leads followed by deep S-waves; tiny q-wave in aVL followed by large R-wave; Left axis deviation).  There appears to be a sawtooth pattern in lead II.  There is a fully upright "P" wave about midway between each QRS in V1.  

A fully upright P-wave is typical atrial activity of atrial flutter as seen in V1.  

See these example cases of upright P-waves:

Case Continued

Thus, I was all but certain that this was atrial flutter.

I brought the patient to the critical care area and told the providers I thought it was atrial flutter with 2:1 AV conduction, but there is an outside chance that it is VT.  If they want to confirm, they could give adenosine and "uncover" the flutter waves.

Remember: Adenosine is safe in Regular Wide Complex Tachycardia.  If it is VT, there will be no effect.  If it is antidromic AVNRT or AVRT (WPW, accessory pathway), it will convert.  If it is flutter, it will reveal the underlying flutter waves.

[Adenosine is only unsafe in atrial fibrillation with WPW, which presents with a very rapid rate, polymorphic QRS, and some very short RR intervals]

Bedside echo showed poor LV fct.  BP was 100 systolic. 

They chose to give adenosine, and this is what happened: 

Adenosine to verify flutter: flutter waves made obvious

Not actually from this case.  Rather, from this one: 

Very Fast Very Wide Complex Tachycardia

Ideally, one would cardiovert.  

But the duration of symptoms was prolonged and so to avoid the risk of stroke, AV nodal blockade (rate control) was preferred.  But with poor LV function, it could be hazardous to give a calcium channel blocker or beta blocker. 

What AV nodal blocker should we give? 

The patient was relatively stable and did not need emergent Ventricular slowing.  Therefore, digoxin is a good choice.  It takes hours to work, but does not decrease LV contractility.  

How about amiodarone?  

That would likely slow the ventricular rate as well, but would have a higher likilihood of converting to sinus.  But we just decided that we DON'T want a conversion to sinus until we can ascertain absence of thrombus in the atrial appendage!  We want to avoid a stroke. 

See this study, in which digoxin performed better than amiodarone.

Acute treatment of recent-onset atrial fibrillation and flutter with a tailored dosing regimen of intravenous amiodarone: A randomized, digoxin-controlled study

I recommended Digoxin, and so a Dig load was given.  

Several hours later, this was the effect:

NT pro-BNP elevated to 7000

Furosemide was also given.

A formal ultrasound later showed reasonably good LV function, and so he later received carvedilol and diltiazem,  Unfortunately, those led to hypotension at 80/40 with a HR 40.

Eventually the patient stabilized on Carvedilol 25 mg bid and Diltiazem 60 mg po q 6 hours

The patient did not want a TEE (transesophageal echo to rule out atrial thrombus), so anticoagulation was started with a plan for outpatient cardioversion in 4 weeks

What happens when you give adenosine to a patient with this rhythm?

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

MY Comment, by KEN GRAUER, MD (7/5/2024):

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

Assessment of WCT (Wide-Complex Tachycardia) rhythms without clear sign of atrial activity is always challenging. I'd add the following thoughts to Dr. Smith's discussion.

• As we often emphasize (See My Comment in the May 5, 2020 post of Dr. Smith's ECG Blog) — Statistically (ie, even before we look at the ECG itself) — at least 80% of all regular WCT rhythms without clear sign of P waves will turn out to be VT (Ventricular Tachycardia). This is especially true in an older adult, as is the patient in today's case. As a result — we need to always assume VT until proven otherwise.
• That said — As per Dr. Smith, there are times when morphologic clues suggest that a WCT rhythm is supraventricular, either with preexisting BBB (Bundle Branch Block) — or, with aberrant conduction.

 

WHY Did We Think Today's ECG was Supraventricular?

For clarity in Figure-1 — I've reproduced and highlighted atrial activity from today's initial ECG, as well as the long lead II rhythm strip from the repeat ECG (obtained after IV Digoxin loading).

• As per Dr. Smith — Today's initial ECG shows a regular WCT rhythm at just under 150/minute, without sinus P waves (ie, there is no clearly upright P wave in lead II with a reasonable chance to conduct).
  
• QRS morphology is consistent with RBBB/LAHB conduction (ie, upright QRS in V1 with wide terminal S wave in V6 — and predominant negativity in the inferior leads). That said — the lack of a terminal S wave in lateral limb leads I and aVL is atypical for true RBBB conduction. As a result — We can not rule out VT on the basis of this single ECG.

PEARL #1: The most commonly overlooked arrhythmia is AFlutter (Atrial Flutter). Because the atrial rate range for untreated AFlutter in an adult is ~300/minute (usual atrial range between ~250-350/minute) — the usual ventricular response to untreated AFlutter is ~150/minute. This is a result of the automatic "protective" mechanism of the AV Node that seeks to avoid 1:1 AV conduction at these fast atrial rates. Instead — the most common conduction ratio with untreated AFlutter is 2:1 (ie, 300 ÷ 2 ~150/minute).

• The BEST way not to miss the diagnosis of AFlutter — is to always consider AFlutter whenever you have a regular SVT (SupraVentricular Tachycardia) at a rate close to 150/minute.
• 
  
• PEARL #2: Using calipers facilitates recognition. What makes the rhythm in ECG #1 so challenging to diagnose — is that atrial activity is partially hidden within the QRS and ST-T wave. This is where calipers help. Simply set your calipers to precisely half the R-R interval. Then see if you can "walk out" 2:1 atrial activity in any of the 12 leads.
• KEY Point: My "Go-To" leads when looking for subtle atrial activity are the inferior leads (II,III,aVF) — lead aVR — and lead V1, as these are the leads most like to reveal "hidden clues" when atrial activity is not readily apparent.

Take another LOOK at today's initial ECG in Figure-1. The RED arrows I've added in the inferior leads highlight an "extra something" in the form of a notch that is consistently seen just after the QRS. 

• Normally, this beginning portion of the ST segment should be smooth. The magnified view of lead II in Figure-1 makes it easier to appreciate the perfect 2:1 regularity of these small-amplitude positive deflections that occur at a rate of ~300/minute. The only thing that does this is AFlutter!
• 
  
• To Emphasize: This 2:1 atrial activity that I highlight in Figure-1 is subtle. It will not be seen unless you specifically look for it. And, the reasons I immediately looked in my 5 "Go-To" leads for subtle signs of extra atrial activity are: i) That QRS morphology in ECG #1 is consistent with RBBB/LAHB conduction, therefore potentially consistent with a supraventricular etiology; and, ii) That the rate of today's regular tachycardia is close to 150/minute (ie, Therefore by PEARL #1, the need to consider AFlutter as the potential etiology).

Figure-1: I've labeled atrial activity in today's ECGs.

What is the AV Conduction Ratio after Digoxin?

I show the long lead II rhythm strip from the 2nd ECG in today's case at the bottom of Figure-1. Regular flutter activity is now readily apparent (BLUE arrows in Figure-1). There is a tendency to miscount the ratio of atrial activity as representing 3:1 conduction — but BLUE numbers show there is 4:1 AV conduction.

• PEARL #3: Odd conduction ratios with AFlutter are uncommon (if not rare!). As emphasized above — by far, the most common conduction ratio of untreated AFlutter is 2:1, which typically results in a ventricular rate of ~150/minute (usual range ~140-160/minute).
• The next most common conduction ratio with AFlutter is 4:1 — which is what we see in Figure-1.
• 1:1 and 3:1 conduction ratios are seen far less often. The clinical relevance of appreciating these conduction ratios — is that a regular tachycardia of ~110/minute is far less likely to represent untreated AFlutter — because 110 X 2 = 220/minute, which is considerably slower than the usual ventricular rate response for untreated AFlutter (ie, The atrial rate of flutter may be less if the patient is being treated with an antiarrhythmic agent).
• 1:1 AV conduction with AFlutter is rare. About the only times I have seen this is with WPW (ie, accelerated conduction over an accessory pathway) — or, in a patient with AFlutter who is treated with an antiarrhythmic such as Flecainide, but without an AV nodal blocking agent to control the ventricular response.

As explained by Dr. Smith — the choice of IV Digoxin was appropriate and effective in today's case.

• Having trained in the 1970s — I had lots of experience using oral and IV Digoxin. At the time, it seemed that virtually all cardiac patients with chronic AFib or heart failure were on this medication. The drug works — but caution is needed in use because of the risk of Dig toxicity.
• In recent decades — Digoxin has been used much less often. As a result — current generations of physicians have far less experience using this medication. I believe ( = My opinion) this reduced use of Digoxin is appropriate for 2 reasons: i) In most circumstances — other more effective agents are available; and, ii) It is because recent generations of physicians have less experience using Digoxin — that the risk of developing Dig toxicity in their hands is increased. The drug is effective — but there is an "art" to optimal use.
• The above said — Dr. Smith's skillful use of IV Digoxin loading in today's case illustrates successful use in this special situation. 
• 
  
• PEARL #4: While onset of action following oral ingestion of Digoxin is between 30 minutes to 2 hours — the onset of action following IV administration may begin within 15-to-30 minutes (Ferrari et al — Current Cardiol Rev, 16(2): 141-146, 2020 — and — London Health Sciences Centre, 2018). IV Digoxin may begin to slow the ventricular response of AFib or AFlutter sooner than many clinicians realize.

Concerning ECG from the patient in trauma bay 2. Activate the Cath Lab?

Written by Willy Frick

You are sitting at your desk, catching up on your charting, and a nurse hands you the following ECG for interpretation.

What do you think?

Here is Queen of Hearts interpretation with explainability:

She says OMI with high confidence. She is particularly worried by the inferior HATW with apparent posterior extension suggested by STD maximal in V2.

Version 2 of the Queen of Hearts, which is not yet available publicly, stated that this is a false positive (that is without knowing any history!)

Click here to sign up for Queen of Hearts Access

I sent this to Dr. Meyers with no context, and he responded "Something a little weird to me, but I'll just go with OMI if the patient has symptoms with a reasonable pre-test probability of ACS, especially chest pain."

Smith: it is atypical for OMI, but as Pendell says, with high pretest probability you would need to assume OMI.  But to my eye it is not diagnostic of OMI.  Queen version 1 may be wrong here.  Maybe we can get an interpretation from Version 2. 

Ten minutes later, repeat ECG was performed.

Again, Queen of Hearts interpretation with explainability:

She again says OMI with high confidence. This time, she sees HATW in V6. It is common for a large, dominant RCA to supply the lateral wall. Shortly after this ECG was obtained, cath lab was activated and the patient underwent emergent angiography. What did they find?

The patient had normal coronary arteries.

In addition, her echocardiogram was normal. What explains this? The answer is that it was predictable based on the history. The patient is a middle aged woman with epilepsy and no other medical history. She had a completely normal day with no complaints and was playing cards with family when she had a generalized tonic clonic seizure lasting 30 minutes, finally terminating with midazolam 10 mg IM per EMS. Upon arrival to the ED, her temperature was 103.2 °F, and her other vital signs were within normal limits. Her ECG changes may be baseline (no prior available), or they may be related to prolonged seizure or anti-epileptic drugs.

Now, forget all of the ECGs in this post along with the Queen's interpretation. Imagine all you knew about the patient was the preceding paragraph. Based on this, what would you say the likelihood of OMI is? You would have to say it is extremely low. If you saw 100 similar patients (i.e., in otherwise normal health presenting with seizure and no other symptoms), it would be surprising if even 1 of them had OMI based on history.

Stated differently, her pre-test probability for OMI is at most 1%. So, what does it mean that the Queen of Hearts sees OMI with high confidence? You might be asking yourself, "Aren't the Queen's sensitivity and specificity exceptionally good?" Yes! They're phenomenal. In the largest published study to date, her sensitivity is 80.6% and her specificity is 93.7%. Her area under the receiver operating characteristic curve  was 0.938 which is superb.

And yet, all of this can be overcome by misapplication of an exceptionally good test in the wrong patient population. Remember the following framework.

Imagine we have 100,000 patients, and 1% of them have OMI. Before reading further, see if you can fill in the numbers in the above grid using these values -- 100,000 patients, 1% incidence of OMI, 80.6% sensitivity, and 93.7% specificity.

If the incidence is 1%, this means 1,000 patients will have OMI, and 99,000 will not have OMI. If the sensitivity of the test (QOH) is 80.6%, then 80.6% of the 1,000 patients with OMI will have a positive test (true positive), and therefore 19.4% of the 1,000 patients with OMI will have a negative test (false negative).

Similarly if the specificity is 93.7%, then 93.7% of the patients without OMI will have a negative test (true negative), and 6.3% will have a positive test (false positive). This is shown below:

Note that the middle column (true positives and false negatives) sums to 1000, and the right column (true negatives and false positives) sums to 99,000. So, back to the question, what does it mean to have a QOH positive ECG in this context? 

What you want to know is the positive predictive value of the test. This tells you the likelihood that a positive test represents a true positive (and not a false positive). Test yourself by seeing if you can calculate this number before reading further.

To calculate this number, divide the number of true positives by the number of total positives. In this case, that is:

806/(806 + 6,237) ≈ 11%

In other words, a false positive is much more likely than a true positive. This is why Dr. Meyers said "I'll just go with OMI if the patient has symptoms with a reasonable pretest probability of ACS, especially chest pain."

Learning points:

• ECG is a diagnostic test which must be interpreted in clinical context.

• Estimation of pre-test probability is an essential part of taking good care of patients.

• Diagnostic tests with phenomenal test characteristics perform poorly when they are used in the wrong context.

• The Queen of Hearts was trained on patients presenting to the ED with new onset chest pain, and that is the context in which she performs the best.

Smith comment: the converse is also true: if the pretest probability is very high, a negative test has a low negative predictive value.  

Here are more cases related to pretest probability: https://hqmeded-ecg.blogspot.com/search?q=pretest+probability

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

MY Comment, by KEN GRAUER, MD (7/2/2024):

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

Appreciation of the concept conveyed in Dr. Frick's post today — is essential for optimal clinical ECG interpretation.

• In a word — Dr. Frick synthesizes the essence of Bayes' Theorem — which amazingly was first described in 1763 by the English statistician, philosopher and minister, Thomas Bayes.
• In my role as a primary care educator from 1980-2010 — I regularly cited Bayes' Theorem when teaching ETT (Exercise Treadmill Testing) to primary care residents and clinicians.
• Simply stated, as applied to ETT in the office — the likelihood of a patient having coronary disease is proportional to the prevalence (ie, pre-test likelihood) of coronary disease in the population being tested. Viewed from another perspective, "You can't catch any goldfish — if there are no goldfish in the pond."
• Applied to  Stress Testing in the office — When the patient in front of me was a middle-aged or older male smoker with classic anginal symptoms on minimal exertion, relieved by rest — I knew before that patient stepped on the treadmill that the likelihood that the ETT would be positive was at least 90%! Realistically, my expectation was less to "confirm" what I already knew (ie, that this patient had coronary disease) — and more to estimate relative severity of his disease (ie, downsloping ST depression within minutes of beginning exercise with reproduction of his angina meant the need for asap cath).
• In contrast — atypical CP (Chest Pain) in a younger woman without significant risk factors meant that even if the ETT resulted in downsloping ST depression — this was more-likely-than-not going to be a false positive test.

What does this have to do with ECGs in the ED?

Practically speaking — the fact that a patient calls EMS or presents to the ED because of new CP automatically places him or her in a higher-risk group for having an acute cardiac event. 

• This is not to say that all such patients are having an acute OMI (because we know that is not the case) — but it is to say that the pre-test likelihood for a patient who calls EMS or presents to the ED for new CP is much higher than the pre-test likelihood of acute OMI for a patient who is seen the next day in the office.
• Working in primary care for 30 years — it was rare that we would see a patient with CP come to the office as they were having an acute OMI. I believe there is an intuitive process of self-selection — in that patients who came to the office were much more likely to have some non-cardiac cause of their symptoms — or perhaps angina pectoris in need of evaluation — but they were relatively unlikely to be in the process of having an acute OMI.

BOTTOM Line: As emergency providers — the great majority of patients we encounter who either contact EMS or come to the ED themselves for new CP —automatically fall in a higher-prevalence group with greater likelihood that they are having an acute event.

• Specifics in the history may further increase this risk of an acute cardiac event — but the onus is clearly on us to rule out rather than rule in an acute event — because this patient came to us for new CP.
• That said — the woman in today's case is different from patients who seek our care because of new CP — in that the reason this woman came to the ED was for evaluation of a seizure associated with fever — in the absence of chest pain.
• Conclusion: As per Drs. Frick and Meyers' comments in today's case — the History is of critical importance for optimal clinical ECG interpretation.

P.S.: Without realizing we are using Bayes' Theorem — there are many other situations in emergency medicine, in which we use this principle all the time! 

• For example — the previously healthy 20-year old with CP, modest troponin elevation, and ECG changes that don't quite fit an anatomic distribution — is far more likely to have acute myocarditis than an acute OMI (far fewer "goldfish" for OMI in that pond).
• In contrast — the older patient with known severe heart disease who presents in a regular WCT (Wide-Complex Tachycardia) rhythm, without clear sign of sinus P waves — has a ~90% likelihood of having VT even before you look at their ECG (Lots of VT "goldfish" in that pond — but very few with aberrant conduction).