Cover Story

Components of the EP Study, Part 4: Atrial Extrastimulus Testing (AET)

Kriegh Moulton, MD1 and Linda Moulton, RN, MS2
1Director, Cardiac Electrophysiology, Prairie Heart Institute, Springfield, Illinois; Faculty, Order and Disorder Electrophysiology Training Program, New Berlin, Illinois 
2Owner, Critical Care ED and C.C.E. Consulting; Faculty, Order and Disorder Electrophysiology Training Program, New Berlin, Illinois

Kriegh Moulton, MD1 and Linda Moulton, RN, MS2
1Director, Cardiac Electrophysiology, Prairie Heart Institute, Springfield, Illinois; Faculty, Order and Disorder Electrophysiology Training Program, New Berlin, Illinois 
2Owner, Critical Care ED and C.C.E. Consulting; Faculty, Order and Disorder Electrophysiology Training Program, New Berlin, Illinois


This is the fourth and final installment of a series of articles in EP Lab Digest discussing the components of a basic electrophysiology study. It is preceded by articles on baseline intervals, decremental ventricular pacing, and decremental atrial pacing. The material has been adapted from the Order and Disorder Training Program.

This part is commonly referred to as atrial extrastimulus testing (AET). It will provide information regarding the behavior of impulse propagation from the “antegrade” perspective, but different from that described with decremental atrial pacing. Rather than assessing AV nodal function by continuous pacing at progressively increasing rates to determine the limits of impulse input to and transit time through the AV node, AET will use a single premature atrial complex (PAC) as the impulse delivery technique. In a manner analogous to DAP’s decremental pacing, the single PAC is being delivered at progressively shorter coupling intervals. A coupling interval is the interval between the last basic paced atrial complex and the premature atrial complex, the A1A2 interval. 

The AET process presumes that the premature impulse has the ability to propagate from its “high right atrial” source, reach the region of interest (the AV junction) without delay, and penetrate the compact AV node with the expectation of emerging from it. The input limit is reached when a particularly short coupling interval results in failure to reach the other side of the compact AV node: a non-conducted PAC. The coupling interval associated with block is known as a refractory period, a measurement unique to AET. More on cardiac refractoriness is discussed below. The transit time of a premature impulse conducting through the compact AV node, the A2H2 interval, tells the story of which route (the fast or slow AV nodal pathway) was traversed: <150 msec is unequivocally the fast pathway, and >200 msec is unequivocally the slow pathway conduction. 

AET was probably designed to do one thing: determine the AV nodal refractory periods. Historically, this would aid in the identification of the patient having a slow AV nodal pathway. However, as we will see, the defining feature of a slow AV nodal pathway as traditionally defined by AET is not transit time. It was based on the magnitude of change in the AH interval between two consecutive PACs. This approach would often fail to meet the criteria despite having AH intervals well over 200 msec. 

The rules regarding the options for routes of impulse conduction during AET are the same as those applying to decremental atrial pacing. It might be a good idea to first review the discussion on the relevant anatomy and physiology of antegrade impulse conduction as outlined in the introduction section of the DAP article from the March 2017 issue of EP Lab Digest (


Refractoriness refers to a state of reduced or absent cell excitability, and is usually discussed in the context of cell responsiveness during the period of repolarization. Excitability is entirely dependent on the ability to activate the cell’s inward sodium current, responsible for depolarization in “fast fibers” (atrial, Purkinje, and ventricular cells) or the inward calcium current responsible for depolarization in “slow fibers” (sinus and AV nodes). Repolarization in the cardiac action potential may be more meaningfully viewed as a sustained period of the depolarized state. It exists to provide a period of sustained cell contraction, in turn, necessary to produce a force-enabling ejection of a hefty mass of blood from the ventricle. That is what the heart does. Imagine if cardiac cells behaved like skeletal muscle cells, whose action potentials last only a few milliseconds and produce a “twitch” contraction. It would be as ineffective as trying to put a bowling ball into motion with the slap of your hand.  Thus, cardiac cells have a long action potential duration (prolonged repolarization or recovery of excitability) when compared to other bioelectric cells of the body, and the duration is measurable vis-à-vis the response to a premature stimulation. 

In order for a single cardiac cell to respond to a stimulus, its level of membrane potential, or difference in intracellular charge from that in the extracellular space, must be large enough to prepare the sodium or calcium channels for responsiveness. Responsiveness is most optimal in the cell membrane’s resting state (e.g., -85 mV) and least optimal in the midst of an action potential when the level of membrane potential is low (less than -65 mV). If a premature stimulus is delivered at the very end of an action potential when membrane potential is high (-85 mV), a full-fledged action potential will result. However, if the coupling interval of the premature impulse is shortened, it may be timed such that the level of membrane potential is too low to allow the sodium (or calcium) channels to open sufficiently to produce a proper action potential. The result may be a poor action potential (timing within the cell’s “relative refractory period”) or no action potential (earlier timing within the cell’s “absolute refractory period”). By decrementing the coupling interval in small steps, one can “discover” these two refractory periods sequentially, based on the cell’s response. The message is that it’s not so much about the duration of the action potential as it is about the level of membrane potential a cell is at when stimulated. That’s the ticket.

In a manner analogous to the case of a single cell, refractoriness can be measured in cardiac tissue: atrial, AV nodal, or ventricular. The difference is that instead of directly stimulating a single cell, a natural wavefront of activation produced from a pacing catheter can be delivered prematurely to spread through tissue. The success of the premature impulse propagating into tissue is determined by the refractoriness of the tissue ahead of the wavefront. More precisely, it is the level of membrane potential of all the cells about to be invaded by the leading edge of the wavefront which determines the success of propagation. Prematurely releasing an impulse will allow disclosure of the tissue’s functional refractory period (FRP) if the response of the premature impulse is conduction slowing. This is analogous to the single cell’s relative refractory period. Discovery of the tissue’s effective refractory period (ERP) is noted if the premature impulse experiences conduction failure (block). The tissue’s ERP is analogous to the cell’s absolute refractory period.

Since refractoriness is “cycle length dependent”, refractory period determination requires that the tissue of interest (atria, AV node, or ventricle) be paced at a constant, fixed rate for at least 8-10 beats to allow the tissue to accommodate to a steady-state condition. Generally, the shorter the basic pacing cycle length (BCL), the shorter the tissue refractory periods, and vice versa. After a suitable number of paced atrial complexes at the BCL (A1A1 interval), a single premature stimulus is delivered at a particular “coupling interval” (A1A2 interval). This process is repeated at progressively shorter coupling intervals until delay or block in impulse conduction occurs. The very first evidence of delay (beginning of A2H2 interval lengthening in the case of AV junctional tissue) means you have reached the FRP, while the occurrence of impulse block means you have reached the ERP of the tissue being evaluated.

During AET, we will discover the AV nodal and atrial refractory periods. Since most patients have both a fast and slow AV nodal pathway, the ERP of both pathways can be determined. Because of its minimal clinical significance, the FRP of any of these tissues will not be discussed in any detail; however, be aware that the FRP of any tissue is always reached first, followed by its ERP. In the case of AV junctional tissue, the longest ERP is that of the fast pathway and will always be reached first as the coupling interval is shortened. Next is that of the slow AV nodal pathway, and the shortest is the atrial ERP. Occasionally, the slow pathway ERP is similar to that of the fast pathway, in which case, the discovery of the slow pathway may be missed. As the process of AET unfolds and the coupling intervals are progressively decreased, appreciate the fact that we are delivering the premature beats from the “high right atrial” location — miles away from the region of interest where the refractoriness measurements are actually being made (the AV junction). Only because the atrial refractory periods are the shortest of the three, one will nearly always be guaranteed the timely arrival of an atrial premature impulse to the region of the AV junction to enable the discovery of its component refractory periods.

AET Process

The following is a description of how to perform AET and what to expect to see on the tracings. While a given EP study may have as few as three intracardiac recordings or as many as 12, the only channel that actually requires your attention is the His bundle channel. The A, H, and V electrograms must all be readily visible in that channel.

Recall that the atrial electrogram (A) seen in the His tracing signals the arrival of the impulse to the “entrance” region of the compact AV node. The node’s “exit” indicator is the His spike (H). Therefore, the AH interval describes the transit time “through” the compact node. In order to provide for a constant state of refractoriness, AET involves 8 atrial paced complexes at a fixed pacing cycle length (the BCL). The BCL can be at 600 msec, 500 msec, or 400 msec. The 9th beat is the PAC and is introduced with a coupling interval (A1A2 interval) shorter than the A1A1 interval (Figure 1). The A1A2 interval is then decreased in decrements of either 10 or 20 msec until the PAC reaches the fast pathway’s input limit and blocks there (i.e., “jumps” to the slow pathway). With further decrementing of the A1A2 interval, the A2H2 interval continues to lengthen until the PAC reaches the input limit of the slow pathway, creating a “non-conducted” PAC, at which point AET ends.

Near the beginning of AET when the coupling intervals are long, the PAC’s A2H2 interval will be identical to the BCL’s A1H1 interval, because the PAC timing is well outside any refractoriness. However, at a shorter A1A2 interval, the A2H2 interval will begin to lengthen, identifying the FRP of the fast pathway. Continued 10- to 20-msec A1A2 decrements result in small increases in the A2H2 interval until a critical A1A2 interval is reached, wherein there is an abrupt and large change in the A2H2 interval. If the magnitude of that A2H2 interval increase is at least 50 msec, a jump to the slow pathway is declared (Figure 2). This A1A2 interval identifies the ERP of the fast pathway. From this point onward, the A2H2 intervals will be >200 msec. Further small decrements in the coupling interval will result in small increases in the A2H2 interval, as the impulse is now maintaining conduction limited to only the slow pathway, until reaching an A1A2 that results in AV nodal block. This A1A2 interval is the ERP of the slow pathway. This process is summarized in Figures 3, 4, and 5.

As mentioned above, the traditional defining feature of the presence of a slow pathway during AET is a “>50 msec jump in the A2H2 interval”. You may be surprised to learn that there is no journal declaring that to be a scientific certainty — it was accepted convention among the EP world in the early 1970s. Practically speaking, since the transition from fast to slow may be subtle (<50 msec change in the A2H2 interval), one may be led to believe a slow pathway isn’t present. This situation can occur when the fast and slow pathways have similar ERP values. Thus, the best marker for the slow pathway’s presence is an A2H2 interval >200 msec (transit time) — not how much it changed from one coupling interval to the next. 

Assessment of AV Nodal Function

As previously seen in DVP and DAP, the fundamental aspects of assessing AV nodal function emphasizes two things: the limits of impulse input to the AV node (and its component pathways), and the limits of transit time through the AV node (and its component pathways). Consistency of this construct is also maintained during AET in any given patient: how short of a coupling interval can be employed and still maintain conduction over the fast pathway? How short of a coupling interval can be employed and still maintain conduction over the slow pathway? What is the longest fast pathway A2H2 interval? What is the shortest slow pathway A2H2 interval? What is the longest obtainable A2H2 interval? These latter questions regarding the A2H2 interval are probably more clinically relevant than the refractory period determinations, because they more directly reflect information about what pathways are being utilized for impulse travel. 

Input limits are defined by the ERP (an A1A2 interval). For the fast pathway, this is the longest A1A2 (coupling) interval resulting in block over the fast AV nodal pathway. More simply stated, it is the first A1A2 to conduct over the slow AV nodal pathway. For the slow pathway, this is the longest A1A2 interval resulting in block over the slow AV nodal pathway. Under normal conditions, when there are no other routes over which to conduct, the A1A2 interval of the first non-conducted PAC will represent the slow pathway’s ERP.

The limits of transit times (an A2H2 interval) through the AV node define the path taken by the impulse: fast vs slow. By tabulating the A2H2 interval values among a large number of patients, one can appreciate that the longest A2H2 interval preceding a jump is infrequently >150 msec and virtually never >200 msec. The shortest A2H2 interval following a jump is infrequently <200 msec, and the longest A2H2 is always >200 msec. Thus, any AH interval <150 msec represents fast pathway conduction, any AH interval >200 msec represents slow pathway conduction, and AH intervals between 150 and 200 msec could be either a slowly conducting fast pathway or a rapidly conducting slow pathway. The “overlap” in this range of AH interval values is the consequence of converging fast and slow pathways. An example of sorting out the meaning of AH intervals between 150-200 msec is discussed in the Figure 4 legend. 

If, because you have all the time in the world, you decide to continue decreasing the coupling interval after reaching the slow pathway ERP, you are headed for discovery of the atrial tissue FRP and ERP. At very short A1A2 intervals, one will begin to see the local atrial electrogram in the RA tracing start to separate from the accompanying stimulus artifact. Remember, the distal electrodes (1,2) of the high right atrial catheter are dedicated to pacing function, while the proximal pair (3,4) provides the recording seen on the screen labeled RA or RAA. Thus, the recording pair is only 2-3 mm away from the actual origin of the wavefront and provides a recording in which the stimulus artifact imperceptibly blends into the local electrogram. As the A1A2 interval reaches the atrial FRP, the local electrogram will begin to separate from the stimulus artifact, illustrating a phenomenon known as latency. You are actually seeing the delay or slowing of a premature wavefront as it sets sail from its origin only 2-3 mm away. It signals that the ERP is imminent. Upon reaching the atrial ERP, the “block” is represented by atrial non-capture. This type of non-capture is called “physiologic” since the tissue’s refractoriness caused the capture failure; it is not to be confused with non-capture seen during pacemaker lead testing, wherein capture failure is due to inadequate stimulus strength.

Disclosures: The authors have no conflicts of interest to report regarding the content herein.