Friday, May 22, 2020

A 50ish Man whose Wide Tachycardia was Treated

===================================
MY Comment by KEN GRAUER, MD (5/22/2020):
===================================
The ECG in Figure-1 was obtained from a 50-ish year-old man, who was found by the EMS team to be in a wide tachycardia. The patient was treated on the scene with medication. ECG #1 shows the results of such treatment.
  • Is this patient now in sinus tachycardia (RED arrow)?

Figure-1: The 12-lead ECG and long lead II rhythm strip obtained after medication converted a wide tachycardia (See text).



My THOUGHTS on ECG #1: It is tempting to accept the RED arrow in Figure-1 as highlighting an upright sinus P wave in lead II.
  • The “good news” — is that this patient had been in a wide tachycardia prior to administration of medical treatment. The QRS complex in ECG #1 is now narrow — so the patient is in a regular SVT ( = SupraVentricular Tachycardia) rhythm at a rate of ~115/minute.


QUESTION: Is there extra atrial activity in Figure-1?
  • HINT: Using calipers makes it much easier to answer this question!




ANSWER: Although tempting to accept the RED arrow in Figure-1 as pointing to a sinus P wave — several factors should make you skeptical of this conclusion:
  • IF the rhythm in Figure-1 was sinus tachycardia — the ventricular rate of ~115/minute is fairly fast ...
  • Several leads seem to manifest an “extra notching” in the early part of the T wave (slanted RED lines in Figure-2).
  • Stepping back a bit from the tracing — the image of “sawtooth” activity (reminiscent of AFlutter) can be imagined (BLUE sawtooth in the long lead II of Figure-2).

Figure-2: Slanted RED lines — as well as suggestion of a BLUE “sawtooth” pattern have been added to Figure-1 (See text).



NOTE: We have previously worked through the diagnostic approach to the regular SVT rhythm (See the November 12, 2019 post in Dr. Smith’s ECG Blog). Today’s case is insightful — because of how easy it is to be fooled into thinking that the RED arrow we saw in Figure-1 is pointing to an upright sinus P wave in lead II.
  • The BEST way to prove what the rhythm diagnosis is in Figure-2 — is to carefully set your calipers to precisely HALF the R-R interval of the regular SVT rhythm.
  • Doing so allows you to perfectly walk out 2:1 atrial activity in numerous leads (slanted RED and PINK lines in Figure-3). Since these slanted lines suggest there are 2 atrial deflections for each QRS complex — and since we calculated the ventricular rate in Figure-1 to be 115/minute — this means that the atrial rate must be 115 X 2 = 230/minute.
  • As we emphasized in the November 12, 2019 post — the atrial rate for untreated AFlutter in adults is almost always close to ~300/minute (ie, 250-350/minute range) — which, with 2:1 AV conduction means that the ventricular rate of untreated AFlutter will most often be close to 300/2 ~150/minute. BUT we were told at the beginning of today’s case that the patient has been treated with antiarrhythmic medication. As a result — an atrial rate of ~230/minute would not be too slow for 2:1 AFlutter (since antiarrhythmic medication may slow the rate of flutter).
  • Finally — I fully acknowledge that it is impossible to rule out ATach (Atrial Tachycardia) with 2:1 block as the rhythm diagnosis in Figure-3. That said — I feel the suggestion of “sawtooth” activity (BLUE lines in Figure-2) make AFlutter a much more likely diagnosis than ATach with 2:1 block!

Figure-3: Slanted RED and PINK lines highlight 2:1 atrial activity in numerous leads (See text).



=====================
— My sincere appreciation to Edward Brunacci (of Sydney, Australia) for contributing this case.
=====================



6 comments:

  1. Great case, Ken, and my thanks also to Dr. Brunacci!

    I've seen so many of these dysrhythmias that the overall pattern or Gestalt immediately suggests atrial flutter to me. I think many people are confused by the interference of the classic saw-tooth pattern by the T waves. This is just another complication of two ECGs (atrial and ventricular) being superimposed on each other. The slower flutter rate also contributes significantly to the confusion. I've taught myself to draw the hidden flutter waves in my mind but calipers help, too. Atrial flutter is one of the most overlooked dysrhythmias and it is NOT benign!

    ReplyDelete
    Replies
    1. Thanks so much for your comment Jerry! In my experience — AFlutter is (by far!) the most commonly overlooked arrhythmia — so, it is important to always consider it whenever you have an SVT rhythm of uncertain etiology — especially if the ventricular rate is close to ~140-160/minute — and especially if atrial activity is unclear. This case is potentially confusing because the ventricular rate is slower than usual for flutter (ie, it wasn't "untreated" AFlutter) — and because to the "quick observer", it could be easy to think there were sinus P waves in lead II ... THANKS again for your active participation on our ECG Blog posts! — :)

      Delete
  2. Agree that there is reasonable suspicion of AFl on the ecg.exposing the flutter waves by carotid massage or iv adenosine would be a simple method of confirmation.

    ReplyDelete
    Replies
    1. @ Subhasish — Thank you for your comment! I agree that one could "confirm" AFlutter by use of either carotid massage or "chemical valsalva" ( = use of Adenosine). But the MAIN point of My Comment is that you could CONFIRM AFlutter within less than 5 seconds, simply by using CALIPERS to confirm there is 2:1 AV conduction — which in association with the "sawtooth" appearance (BLUE lines in my Figure-2) should be sufficient to confirm AFlutter as the rhythm — :)

      Delete
  3. Thank you for a great case as usual!

    I would do Lewis Lead (S5), to spot the atrial activity, although I'm sure its a. Flutter 2:1 at ~115 bpm .
    Thanks!

    ReplyDelete
    Replies
    1. Always nice to see clear atrial activity with a Lewis Lead — Thanks — :)

      Delete

Recommended Resources