Monday, October 3, 2011

Fusion Beat During Supraventricular Tachycardia: No criterion is absolutely accurate in differentiating wide complex tachycardia

Yesterday, I posted this case of wide complex tachycardia, and some steps to help in differentiating VT from SVT with aberrancy.

One step was this:  
"Do a quick look for obvious fusion beats and AV dissociation.  If found, then VT."  It is generally thought that fusion beats are diagnostic of VT.

This prompted our excellent electrophysiologist, Dr. Rehan Karim, to make the point that no criterion or algorithm can make a certain diagnosis of wide complex tachycardia.  And so he provided me with the interesting tracing below (sorry, no 12-lead because this just happened to occur in the EP lab while they were recording):

There is a wide complex tachycardia; rate was 171.   It was proven at EP study to be SVT.  You can see the RBBB morphology, with rSR' in V1 (green arrow) and a wide S-wave in lateral leads (see lead I, red arrow).   There is a fusion beat (black arrow) which was proven in the EP lab to be due to a PVC occurring in the midst of the SVT!!

Learning point: whatever system or rule you are applying, it is not perfect.  Dr. Karim also wanted to make the point that Brugada's algorithm had very good sensitivity and specificity in the derivation group, but attempts to validate it were not nearly as successful.  This may be just as true for Sasaki's rule. 

Nevertheless, while appreciating the limitations of all algorithms, I like Sasaki's rule because it depends on many of same principles as the other algorithms (principles which, though fallible, are pretty reliable), but is quite a bit simpler to apply.


  1. That's very interesting Dr Smith, I remember being taught that the presence of fusion beats in wide complex tachycardias is virtually diagnostic of VT.

    Stupid question though, how would you differentiate this from a capture beat? Would you just expect a capture beat to come earlier, or is the morphology different?

    1. Well, it's now 2015 so I feel like I'm speaking from the future; however, my answer would have been the same in 2011. Capture beats and fusion beats do not diagnose a wide tachycardia as ventricular tachycardia. They simply diagnose AV dissociation. From that, one may (or, as this case so aptly demonstrates, may not) infer a diagnosis of VT.

      A capture beat should be early with respect to the R-R interval of the underlying rhythm. As long as there is no 3rd degree AV block, the ratio of atrial rate to ventricular rate should eventually result in an opportunity for the atrial impulse to make it through the AV node and activate the ventricles. "Capture beat" is really a poor choice of words because so many people take that to mean ANY supraventricular impulse that is successfully conducted through the AV node thus activating the ventricles. Those would be more correctly designated "captured beats." In order for a true capture beat to indicate AV dissociation, its QRS must terminate an R-R interval that is shorter than the R-R interval of the dominant rhythm. Assuming that the QRS is narrower than the wide-complex rhythm, by being early it proves that 1) the AV node is quite capable of transmitting impulses even faster than the dominant pacemaker and that the wide complexes are not due to a pre-existing bundle branch block or a rate-related bundle branch block. If the "captured beat" terminates an R-R interval longer than the R-R interval of the dominant rhythm, who's to say that the wide-complex tachycardia simply didn't self-terminate (they frequently do) thus allowing the supraventricular impulse to conduct? We still wouldn't know if a rate-related bundle branch block had occurred.

      A fusion beat must occur when one would reasonably expect an impulse from the dominant rhythm, otherwise there would be no fusion between the supraventricular and ectopic ventricular impulses. So you aren't going to see fusion beats that are especially early. The fusion beat usually has a morphology intermediate between the supraventricular impulse and the ectopic ventricular discharge. However, it may not resemble the supraventricular impulse much at all. The initial vector will resemble the supraventricular impulse only if the PR interval is the same as the regular supraventricular beats. This would mean that the supraventricular impulse made it into the ventricles before the ectopic impulse could spread very far. The terminal vector is always provided by the ectopic impulse, so the end of the QRS never resembles the supraventricular impulse.

  2. I don't think you can differentiate it. The only way we know that this is SVT with a PVC is that there were internal cardiac recordings in the EP lab when it happened.