Saw this post on Lifehacker and was reminded of AF and most AF studies:

First dinner!

First dinner at the flat! Which was spaghetti bolognese, with cookies and strawberries with cream for sweet.

Home, finally furnished

Took a while to get everything sorted out, but got there eventually. 

Just outside Sydney Street entrance of the the RBH. Beautiful day.

Atrial extrastimuli, atrial overdrive pacing for SVT: When is it useful?

Today, I thought I’d write a bit about atrial extra-stimuli and atrial overdrive pacing during SVT since these manoeuvres aren’t used so often. Personally, I often take a little time to think through what they mean in real-life cases. Also, there are some interesting responses and tips I learned about recently.

First, when are they useful? Atrial extra-stimuli  (or PACs) and atrial overdrive pacing (AOD) are basically used to differentiate a junctional tachycardia from a slow-fast AVN reentry tachycardia. In clinical practice, AVNRT is much more common than JT, so one would think this isn’t much of an issue. Mostly this is true, but occasionally, it can be a problem post ablation of a AVNRT slow pathway, when a slower narrow complex rhythm with a short or zero VA time is seen. This is particularly common during isuprel infusion towards the end of the case:

18 year old female with recurrent SVT. AVNRT was diagnosed at initial EP study and the slow pathway was ablated. Following ablation, this tachycardia (which was a little slower than the original tachy) remained inducible.
Is this still AVNRT?

So how can one resolve this problem?

• Just wait! Often, the VA relationship becomes variable. This doesn’t occur with AVNRT. If the A moves in and out of the V electrograms, then this can only be JT.
• Place a His-refractory PAC. PACs timed to His-refractoriness, or early PACs are both useful. In the case of a PAC timed to His-refractoriness during a JT, the JT will not be affected (because the PAC cannot have travelled down to affect the tachycardia). So no reset of the tachycardia occurs. However, if the rhythm is a AVNRT and the PAC falls during His refractoriness, the PAC can still engage the slow pathway early and advance the subsequent His. As an example of this, look back to Figure 1. Here, we have placed a PAC slightly earlier than the His. The His electrogram arrives on time and is of similar morphology as previously, indicating the PAC cannot have travelled down to reset the tachycardia. However, the subsequent His is advanced (HH 322ms decreased to 295ms) - this must have been through early engagement of the slow pathway. A schematic of the response to a His-refractory PAC is shown in this diagram (taken from the original paper describing this manoeuvre by Padaniliam et al. JACC 2008):

• Place a early (pre-His) PAC. As mentioned earlier, placing an early PAC (before the His signal) can also be useful. Look at this example:

Is this tachycardia AVNRT or JT?

Consider what would occur if the rhythm were a JT. The early PAC would be able to advance the immediate His, but would not terminate the tachycardia. With AVNRT, an early PAC should often be able to engage the fast (and possibly the slow) pathway, leading to refractoriness and thereby terminating the tachycardia. Hence a early PAC which consistently terminates the tachycardia indicates AVNRT and not JT. Schematically, one can show this as below.

(Note that if the tachycardia is not terminated, then one needs to look at the position of the His to determine whether the rhythm is JT or AVNRT - absence of termination of the tachycardia does not mean JT. I think this is quite a subtle point - I didn't appreciate this immediately anyway).

• Atrial overdrive (AOD) pacing. This is really just a special case of a PAC but it is worth describing separately. Look at this example:

Atrial overdrive pacing from proximal CS. Is this AVNRT or JT?

In many ways, AOD here is analogous to using a VOD to distinguish between AT and AVNRT/AVRT. With AOD, one would look for either a A-H-A response (consistent with AVNRT) or A-H-H-A (consistent with JT):

To correctly interpret AOD, (just as in VOD), it is is absolutely essential to measure the intervals to ensure one is not caught out by "pseudo" responses. First, check that the tachycardia is accelerated to the pacing cycle length and not terminated by the AOD. Next, check which is the last entrained beat:

The reason why it is crucial to check which is the last entrained beat is apparent after measuring out the intervals. As the red arrows show, it is the fifth ventricular electrogram which is the last entrained beat and not the fourth. Therefore, this is a A-H-A response and not a A-H-H-A response. Hence, the rhythm is AVNRT and not JT.

(In fact, all of these manoeuvres were done on the same patient, so it is good that they all agree!)

I'd like to close with some EGMs from George Klein's excellent book, "Electrophysiological Manoeuvres for Arrhythmia Analysis":

• It is quite frequent to get one or two extra beats like this following atrial extra-stimulus testing after slow pathway ablation. Again, the usual question is, echo beat (i.e. persistent AVNRT) versus junctional beat. The timing of the His can be useful, but often, putting in a S3 will also sort things out immediately. A S3 will be able to advance the next QRS over a wide range of coupling intervals for JT, whereas it would only rarely be able to do so with AVNRT:

In this case, comparing panels A and B, S3 advances the subsequent QRS (587ms versus 677ms), hence this is likely to be a junctional beat.

•  What about this interesting EGM? Here, a early PAC is able to advance two subsequent beats! In this case, the most likely explanation is a 2 for 1 response, where the early PAC is able to engage both the fast and slow pathways and give rise to two QRS complexes. This implies the slow pathway is still intact.

Retrograde Right Bundle Branch Block

Recently, I've been trying out a SVT manoeuvre I had learned about during the Mayo EP course, but never really used. It was a quiet afternoon, and I was reviewing some old SVT procedures. I came across something like this a few times:

Figure 1. Is there (1) a slow pathway or (2) a septal accessory pathway?

Here, the measured interval shows that the VA time "jumps" from 148 to 202ms with a 40ms (360 to 320ms) shortening of the S1S2 coupling interval.

What does this indicate? In the antegrade direction, an increase in the AV conduction time with a small decrease in coupling interval is usually taken to indicate a switch from fast to slow pathway conduction - this is confirmed by showing that the increase in conduction time is related to an increase in the AH interval ("AH jump").

What about in the retrograde direction? Here, the His signal, which in Panel A must have been buried in the ventricular electrogram, has been released, and is now apparent (blue arrow in Panel B). The increase in VA time must therefore have been related to an increase in VH time. Therefore, this cannot imply the presence of a slow pathway, but instead suggests retrograde right bundle branch block. Assuming the His was buried in the ventricular electrogram but is now released, it looks to have shifted about 40 or 50ms to the right, which is roughly the amount of time one would expect for conduction to cross the ventricular septum and reach the left bundle to go up to the His.

Why is this observation of use? Well, the first point is that this occurs fairly frequently (about 80% of the time). It is not hard to induce retrograde right bundle branch block using RV extrastimulus testing. The second point is that this manoeuvre is more or less analogous to a para-Hisian pacing!

The idea of para-Hisian pacing is basically to compare VA time on loss of His capture. If the VA time remains similar whether the His is captured or not, it suggests there must be an accessory pathway used to conduct from the ventricle to atrium. In the absence of such a pathway, one would expect the VA time to prolong because on loss of His capture, locally captured ventricular myocardium would need to conduct down to the terminus of the right bundle branch before conducting retrogradely up to the atrium - this clearly will take a significantly longer time than if a (septal) accessory pathway is available for retrograde conduction.

In the case of retrograde right bundle branch block, one would observe the same phenomenon. If VA time is significantly prolonged on induction of retrograde right bundle branch block (marked by release of the His i.e. a VH "jump"), then this suggests that there isn't a (septal) accessory pathway available for retrograde conduction. If however VA time is increased minimally, then there is a strong suggestion of a septal accessory pathway (able to conduct retrogradely).

This maneuver was, I believe, first described by Sam Asirvatham's group in the Mayo clinic. In a retrospective study of 105 patients undergoing EPS and ablation for AVNRT or AVRT, the average V-H interval increase with induction of RBBB was 53.7 ms for patients with AVRT and 54.4 ms for patients with AVNRT (P = NS). However, the average V-A interval increase with induction of RBBB was 13.6 ms with AVRT and 70.1 ms with AVNRT (P < 0.001). All patients with a greater V-H than V-A interval change had AVRT, and those with a smaller had AVNRT.

Turning back to the example tracing in Figure 1, the VA interval increased significantly, suggesting absence of septal accessory pathway conduction. (No comment can be made about presence or absence of a slow pathway).

      

The Exercise Stress Test: An Electrophysiologist's Perspective (Part 1)

I received an email from Singapore today, which I found quite interesting:

"Hi Eric, I think some of the CAs are not entirely sure what to look out for when running a treadmill with a modified protocol, such as those for WPW, Brugada, etc. I don't blame them, even I am not sure exactly what to look out for in such instances. Do you guys have any materials/resources that could help? The SRs would probably find it quite useful too. Thanks!"

Although treadmill stress tests are, of course, most often used to look for myocardial ischaemia, they also have utility in a number of EP contexts, most notably:

1. To assess robustness (or otherwise) of AV nodal conduction
2. To assess robustness (or otherwise) of antegrade accessory pathway conduction (i.e. WPW) 
3. For risk stratification in Brugada syndrome 
4. To aid in the diagnosis and risk stratification of long QT syndrome
5. Very occasionally, to assess implanted device function

Lets take a look at these in turn.

1. To assess the robustness of AV nodal conduction

EP is often referred patients who are bradycardic and appear to have some problem with AV conduction, and the issue is: (1) is impaired AV conduction pathological? and (2) if so, does it merit anti-bradycardia device implantation? In such cases, it is very useful to be able to assess whether AV conduction "gives up" as the atrial rate increases. From a teleological perspective, AV conduction which is unreliable with increasing atrial rates must clearly be pathological. Put another way, it makes no sense for AV conduction to worsen as the sinus rate increases e.g. with exercise, since this implies that with exertion, the ventricular rate falls. Remembering this simple rule is frequently enough to deduce whether a pacemaker implant is likely to be helpful.

In fact, exercise stress testing can often be taken further and used to localise the level of AV block to either the AV nodal level, or the infra-nodal level. With sympathetic stimulation, AV nodal conduction is improved. However, this is not the case for infra-nodal conduction, and indeed, as atrial impulses reach the infra-nodal conduction system faster and faster, successive impulses encroach more and more on the refractory period. The most usual pattern of infra-nodal block then is abrupt loss of AV conduction, otherwise commonly referred to as Mobitz type II block. In severe cases, it is even possible to observe higher degrees of AV block and complete heart block. Such cases will always require a pacemaker implant, unless some reversible cause for the impaired AV conduction can be found.

2. To assess the robustness (or otherwise) of antegrade accessory pathway conduction

Some folks are born with so-called accessory pathways. These are basically connections between the atrium and ventricle at a point other than the AV node, and are generally composed of working myocardial cells. In some cases, accessory pathways conduct only in the antegrade direction (i.e. A→V), in others they conduct in the retrograde direction (V→A), and in still others, they are bidirectional. When they conduct in the antegrade direction, accessory pathways may activate the ventricular myocardium earlier than conduction through the AV node would - this is then known as pre-excitation and on the surface ECG, this is manifested as a delta wave. We would call this a Wolff-Parkinson-White pattern ECG, and if the patient were symptomatic, we would say the patient had Wolff-Parkinson-White syndrome.

Why would an exercise test be helpful in this context? It is helpful to think through this in a few steps:

(i) The feared complication of WPW is that of ventricular fibrillation (and therefore, sudden death). The reason for this is that in a normal individual, AV conduction is regulated by the AV node - the AVN has a property called decremental conduction, where the faster atrial impulses reach the node, the slower conduction through the node. This is why in a normal individual, atrial fibrillation (where atrial impulses may reach the AVN at rates of 600/min) does not translate into ventricular fibrillation and sudden death. In contrast, for patients with WPW, accessory pathway conduction is usually not decremental (since the pathways are composed of normal, working myocardial cells). Therefore, in such individuals, AF can result in VF. This is especially unfortunate as individuals with accessory pathways are also more prone to AF.

(ii) Therefore, risk stratification for SCD in WPW patients basically boils down to evaluating whether  the accessory pathway allows robust antegrade conduction. If the accessory pathway allows very rapid antegrade conduction, then the risk of AF degenerating into VF (and therefore sudden death) is higher.

Accurate assessment of the electrophysiological properties of accessory pathways can only be made in the EP lab. However, the exercise stress test can sometimes be useful as a non-invasive alternative. The idea is to evaluate the point at which accessory pathway conduction is lost when the atrial rate increases (as a result of exercise) - if this occurs easily, then the risk of ventricular fibrillation and sudden death will be low, and vice-versa.

This theory has been tested.  Daubert and colleagues demonstrated that (only) abrupt and complete loss of preexcitation during exercise confirms poor antegrade accessory pathway conduction a long anterograde APERP. In EP-speak, what they showed, in a predominantly adult prospective study, was that persistence of preexcitation during exercise stress showed a sensitivity of 96% (but a specificity of only 17%) in predicting either a SPERRI in AF of  less than 250ms or an APERP of less than 250ms. What this means is that while the exercise test has a reasonable negative predictive value (about 88%), it has quite poor positive predictive value (about 17%).

So, to summarise the above:

a) Report whether there is abrupt and complete loss of pre excitation during the exercise test. I would also state the maximum heart rate reached.

b) If there is abrupt and complete loss of pre excitation during the exercise test, and the patient is clinically asymptomatic (no faints / loss of consciousness), then the patient is most likely to be low risk for ventricular fibrillation.

c) However, the converse is not true - just because pre-excitation persists throughout the exercise test does not mean that the patient is high risk.

I think that's probably enough for a single post. I'll post some examples of (1) and (2) in the future, and also discuss more about (3) to (5).