First of all, note that during ventricular pacing there is ventriculoatrial (VA) block. The fourth pacing stimulus does not capture the ventricle because it occurs during ventricular refractoriness that is present shortly after activation from the sinus complex. The third and fifth QRS complexes conduct from the sinus origin to the ventricle. The first sinus complex occurs very soon after the ventricular paced beat and does not conduct through the atrioventricular (AV) node as noted by absence of a His bundle electrogram. Thus, there is concealed conduction into the AV node.

The fifth QRS complex is a result of conduction from the atrium over an AV accessory pathway. It is able to conduct because the time from the previous ventricular paced beat to the sinus complex, 510 milliseconds, is sufficient to allow conduction over the accessory pathway. Since this pathway is not noted on the first sinus complex, either there is also concealed conduction into the accessory pathway preventing AV conduction over the accessory pathway or the ventricle is simply refractory since it comes too early after the initial paced beat.

The third QRS complex is not normal but it does not show conduction over the AV accessory pathway. The paced V to the subsequent atrial interval is 300 milliseconds, and there is substantial prolongation of the AH interval suggesting concealed conduction into the AV node but not to the degree that it causes anterograde block. The HV interval is very short. No matter what the AH interval was in this patient, whenever the AV accessory pathway was blocked there would be a similar short HV interval. This represents conduction over a fasciculoventricular (Mahaim) pathway. Clearly there is concealed conduction into the AV accessory pathway here because the sinus complex can conduct to the ventricle over the normal system as well as the fasciculoventricular pathway. With the longer paced V to A interval of 510 milliseconds, there is now manifest conduction over the AV accessory pathway with a shorter AH interval, but the His is within the QRS complex, characteristic of this type of accessory pathway.

In summary, the electrophysiologic observations in this tracing include concealed conduction into both the accessory AV pathway and the AV node, manifest preexcitation over an AV accessory pathway, and manifest preexcitation over a fasciculoventricular pathway.

The baseline QRS complex noted on the right side of the figure is normal. The wide QRS on the left side of the figure can only realistically be preexcited or left bundle branch block (LBBB). During incremental atrial pacing a rate-dependent LBBB occurred and the patient also demonstrated a markedly long AH interval. Thus, the paced ventricular responses are essentially “one complex off,” that is, they are conducting to the subsequent LBBB QRS complex (arrow). The fourth paced atrial complex blocks in the AV node and the next atrial paced complex conducts with a narrow QRS complex and subsequent 2:1 block in the AV node. The LBBB is no longer present because the 2:1 AV block results in an H–H cycle length, 714 and 741 milliseconds, which is outside of the refractory period of the LBB.

During the LBBB complexes, the His potential occurs at nearly the same time as the pacing stimulus artifact, totally obscuring the His on the first beat, but a close look at the subsequent complexes reveals a His potential well in front of the LBB QRS (H) ruling out preexcitation. Of note, the patient actually had tachycardia secondary to an epicardial ventricular tachycardia site in the posterior area of the heart but otherwise had normal cardiac function.

The first QRS complex has the classic preexcited pattern with a short PR interval followed by a delta wave noted most prominently here in ECG lead II. This has sometimes been referred to as an “Eiffel Tower” appearance. The location of this AV accessory pathway was in the atrial septum very close to the normal AV conduction system. The second QRS complex was secondary to a catheter-induced junctional beat. The His electrogram is readily seen in this cycle, suggesting that it is “buried” in the preceding preexcited QRS complex even if not seen.

The most common type of accessory pathway is an AV muscle connection, seen here, that originates at the base of the atrium and conducts into the base of the ventricle. This is confirmed when a His extrasystole conducts to the ventricle without preexcitation, clearly demonstrating that the accessory pathway originates above the level of the His bundle. “True” Mahaim fibers as described by Mahaim originate in either the AV node or the His–Purkinje system (HPS), namely, nodoventricular/fascicular or fasciculoventricular pathways, respectively. (The descriptor “Mahaim” has been applied to many atypical pathways that exhibit “rate-dependent” or “decremental” conduction for historical reasons, namely, that all such pathways were thought to involve the AV node as per the pathway described by Mahaim. We believe it is more precise to name the pathways by their origin and insertion sites to be clear since decremental pathways may really be anywhere on the AV ring and the great majority of those (often atriofascicular) seen clinically have nothing to do with the AV junctional connections described by Mahaim.) While one cannot totally rule out a nodoventricular accessory pathway in this instance, they are rare and would not be expected to have such prominent preexcitation with such a short PR interval.

During ventricular pacing, retrograde atrial activation sequence is slightly eccentric with the mid-coronary sinus (MCS) electrogram occurring somewhat earlier than the His bundle atrial electrogram (HBE). On the last beat of the drive train (S₁), the HA interval measured on the HBE tracing is 82 milliseconds. This observation alone cannot differentiate conduction over an accessory pathway or both an accessory pathway and the AV node. However, the premature ventricular stimulus (S₂) causes a sudden prolongation of the VH interval with a concomitant shortening of the HA interval to 24 milliseconds. The sudden lengthening of the VH interval likely results from retrograde block in the RBB with subsequent transseptal ventricular conduction and then conduction to the His bundle over the LBB. Note that the general retrograde atrial activation sequence remains much the same.

A sudden shortening of the HA interval does not occur with retrograde conduction over the normal AV conduction system. There would be a similar or slightly longer HA interval. It is also noted that atrial activation after S₂ as measured at the MCS electrograms now precedes the retrograde His, clearly untenable with retrograde AV node conduction. This observation clearly shows that the HBE atrial electrogram is activated over the accessory pathway and not over the AV node. Of note, this does not mean that conduction over the AV node could not occur at another time during the study, only that it could not be demonstrated in this tracing.

This maneuver is known as para-Hisian pacing. The distal His bundle electrode (HBED) was positioned near the His bundle recording and the pacing energy was incrementally increased to get initial capture only of the ventricle (first two QRS complexes) and subsequently capture of the ventricle and the His bundle (second two, more narrow QRS complexes). Note that the stimulus to atrial electrogram on the ablation catheter (ABL) positioned in the septum is 120 milliseconds without His capture and 94 milliseconds with His capture. This response is characteristic of VA conduction over the AV node and not over a septal accessory pathway. A septal accessory pathway connecting the A and V near this site would typically show a similar VA interval with or without capture of the His bundle during this maneuver. Importantly, while this observation suggests retrograde conduction over the AV node, it does not rule out conduction over an accessory pathway with a long conduction time in which the VA interval might be longer than that noted over the AV node or a nonseptal accessory pathway located farther from the pacing site. This patient did have AV node reentry that was successfully ablated.

The atrial extrastimulus results in an echo cycle but the resulting atrial activation is complex with “double” atrial potentials after the QRS. In the distal coronary sinus, there is a large “near-field” atrial electrogram followed by a barely visible “far-field” electrogram. In the adjacent electrode CS 3–4, the first atrial electrogram is far-field (single asterisk) and the second is near-field (double asterisk), as it is for the subsequent two electrode pairs. At the proximal CS (CS 9–10), there is a single electrogram that is relatively early.

It must be remembered that the coronary sinus is often known as the “third” atrial chamber and generally has a muscular coat that is capable of conducting cardiac impulses. This coat is most prominent proximally, although it can extend quite far into the left lateral region. The CS catheter is closer to the CS muscle than the adjacent left atrium so that its potential would generally be expected to be near-field, whereas the LA potential is generally farther and more far-field. This varies somewhat in individuals.

With this background, one can postulate that the atrial extrastimulus conducts to the ventricle and echoes back to the atrium over a left lateral accessory pathway. Conduction then propagates from this left lateral region via the left atrium (not the CS musculature) proximally (first faint line) and then proceeds distally again via the CS musculature (second faint line). There is no apparent communication between the CS muscle and the LA in this individual until the proximal CS region where the two atrial components merge.

The “dual” conduction is hinted at but not readily apparent in sinus rhythm, presumably since conduction is proceeding over pathways providing a fused complex representing both atrial and CS muscle.

It is conceivable that the first of the two atrial components represents an accessory pathway potential but this would then necessitate an extremely long accessory pathway, a little far-fetched. Further, AP potentials generally have a more near-field or “rapid activation” appearance.

The coronary sinus muscle conduction was considered a bystander in this example and ablation was targeted to the LA endocardium adjacent to CS 1–2, which abolished accessory pathway conduction.

The “checklist” for assessing para-Hisian pacing includes assessing the following:

1. 
   

   Measure the stimulus to A interval, best at a septal site or orifice of coronary sinus region. A short S–A interval at this site (certainly less than 50 milliseconds) should arouse suspicion of inadvertent atrial capture.

   
   
2. 
   

   Measure the stimulus to V interval at the His bundle electrode (HB) site of pacing. Evidence of capture and effective pacing of the para-Hisian muscle here is indicated by very short S to V interval, in range of 20 milliseconds or less.

   
   
3. 
   

   Measure the stimulus to V interval at the RV apical electrogram, which is generally relatively close to the RBB terminus. This should be relatively shorter with His capture than with only RV para-Hisian muscle capture since the latter has to proceed from the base of the heart to the apex via muscle.

Individual cycles have been labeled 1–4 for clarity of explanation. Cycle 1 by the above criteria shows His and local muscle capture and no direct atrial capture. His capture is also signaled by the relatively narrow QRS. Cycle 2 shows a lengthening of the S–V at the His pacing catheter indicating loss of para-Hisian muscle capture. Pure His capture is indicated by normal QRS morphology and no change in the S–V at the RV apical electrogram. The S–V is about the same as the HV interval in sinus rhythm. Further, the similar S–A interval with pure His capture and His plus direct local muscle capture would be expected with retrograde conduction over the normal AV conduction system. The only change from cycle 2 to 3 is shortening of the S–A interval at the proximal coronary sinus. This signals direct local atrial capture in addition to His capture seen in cycle 2. In cycle 4, the S–A at the proximal coronary sinus again indicates atrial capture but here the S to V becomes later at all sites and this indicates atrial capture only.

The answer to the original question posed is that this was not helpful in assessing retrograde conduction since His pacing was never lost in the presence of para-Hisian muscle capture.

The tachycardia has a normal QRS with normal HV interval and an eccentric atrial activation sequence with earliest atrial activation at the distal coronary sinus. This can only be atrial tachycardia or AVRT over a left lateral accessory pathway. The mode of onset, initiated by a ventricular extrastimulus with the VA interval following the first cycle being the same as subsequent cycles, leaves little doubt that it is AV reentry. This is so even though there is some variability in the initial AH interval. Further, the initiation is with a VAV sequence, characteristic of either AVNRT or AVRT.

The mode of onset is nonetheless a little unusual. The atrial activation sequence (highlighted by the faint lines) is different for each of the ventricular drive cycle, ventricular extrastimulus, and the tachycardia. During tachycardia, it is clearly eccentric with the distal coronary sinus activating first. The ventricular extrastimulus is clearly “central” with activation first at the HB electrogram. The ventricular drive is very similar but shows activation relatively earlier at the distal coronary sinus. This is most compatible with retrograde fusion over the AV node and the accessory pathway. The ventricular extrastimulus blocks in the accessory pathway and proceeds retrogradely over the AV node with a longer VA interval. A rapid deflection appearing in the middle of the QRS at the His site (arrow) may well be a retrograde His deflection, although it is difficult to be sure. Conduction proceeds to “echo” back to the ventricle via a slow AV node pathway (AH = 270 milliseconds) to start AVRT. AVNRT was observed as a separate arrhythmia at other times in the study.

This is an exercise in extracting data from whatever information is available, that is, a “minimalist” approach. The ECG leads suggest that the pacing site is the RV outflow region since lead 1 is negative and V₁ has LBBB pattern.

One can consider the two cycles as a nonsustained rhythm. The first cycle may be an echo with LBBB following the A after the ventricular extrastimulus or possibly a repetitive ventricular response (RVR) due to bundle branch reentry. Regardless, the QRS is LBBB type and the very early V in the RV apical electrogram supports this.

The VA interval for this LBBB cycle is 230 milliseconds. The next cycle has right bundle branch block (RBBB) morphology and a VA of 155 milliseconds. This is most compatible with a left lateral accessory pathway since the return cycle to the atrium is much longer (230 milliseconds vs. 155 milliseconds) with the LBBB than with the RBBB cycle.

A prolongation of VA by >30 milliseconds with change to LBBB from normal QRS or RBBB QRS during tachycardia is most compatible with retrograde conduction over a left lateral accessory pathway.

One might also note that termination with an atrial electrogram would not be expected with atrial tachycardia and suggests that anterograde conduction to the ventricle after the atrial cycle is necessary to complete a tachycardia circuit in this individual.

The first cycle has LBBB morphology and the second has RBBB morphology and terminates after a retrograde A. This can be thought of as AVRT that terminates spontaneously. The last event is an anterograde His, very similar to the anterograde His in the sinus cycle at the end of the tracing.

Termination with a His suggests that the His and bundle branch are critical to this short “tachycardia” since block below the His stops it.

The point of interest is indicated by the “local” VA intervals noted at CS 1–2. This is the site of the accessory pathway and corresponds to the site of successful ablation. The “local” VA interval is the interval between the rapid ventricular activation and the rapid atrial activation at the site of the accessory pathway. At the AP site, this represents the conduction time over the AP. The local VA time in this case lengthens with LBBB, that is, 25–50 milliseconds. This is best explained by “slanting” of the accessory pathway (i.e., the atrial and ventricular ends are not exactly concordant at the AV ring). If the accessory pathway went directly from ventricle to atrium at the AP site, one would not expect the “local” VA interval to change with a change in ventricular activation sequence (i.e., a change in the direction from which ventricular depolarization reaches the accessory pathway).

The “local” VA interval is to be distinguished from the (global) VA interval that is from the “onset of ventricular depolarization to the rapid atrial deflection at a designated site.”

The electrophysiologic phenomenon exhibited is concealed conduction. Concealed conduction occurs because the PVC conducts retrogradely into the AV node but not to the atrium, and conduction into the AV node is not visible (concealed). The result is prolongation of AV nodal refractoriness. The sinus complex after the PVC has an increase of the AH interval from 100 to 105 milliseconds (upper panel). This increase in AH interval would likely be missed during analysis of the surface electrocardiogram performed at 25 mm/s paper speed. It is, however, demonstrated during intracardiac measurements performed at faster paper speeds. Conduction of the PVC into the AV node is confirmed because of the subsequent prolongation of AV node conduction.

The effects of concealed conduction are more prominent in the lower panel. Note that the AH interval after the PVC is 140 milliseconds compared with 105 milliseconds in the upper panel. The longer AH interval occurs because the PVC is introduced later in diastole and consequently closer to the next sinus complex. There is less recovery of AV nodal excitability as a consequence. Note that the V₂A′ in the upper panel is 620 milliseconds but only 390 milliseconds in the lower panel. This demonstrates the well-known inverse relationship of the effect of PVCs on PR and RP intervals as identified in the electrocardiogram. In other words, as a PVC occurs closer to the next sinus P wave (short RP), there is a more marked effect on subsequent PR prolongation or even block. This relationship is identified in Fig. 3–11C. As V₂A′₁ shortens, A′₁ H′₁ progressively lengthens until block of the subsequent sinus complex occurs.