Abstract
Atrioventricular nodal reentrant tachycardia (AVNRT) is the most common form of paroxysmal supraventricular tachycardia. The exact electroanatomic circuit responsible for AVNRT remains elusive. Current evidence suggests that dual atrioventricular nodal (AVN) pathway physiology constitutes the substrate for AVN reentry. The different atrial inputs to the AVN, rather than functional longitudinal dissociation within the compact AVN, represent the fast and slow pathways involved in the reentrant circuit.
AVNRT can manifest in different forms depending on the anatomic substrate forming the anterograde and retrograde pathways incorporated in the reentry circuit. Traditionally, AVNRT has been classified into “typical” or “atypical” forms. Typical AVNRT (anterograde slow-retrograde fast) accounts for 90% of AVNRTs. Atypical AVNRT variants are traditionally subclassified as either fast-slow or slow-slow types.
Maintenance of AVNRT is dependent on AVN conduction; hence, maneuvers or drugs that slow AVN conduction and prolong AVN refractoriness are used to terminate the tachycardia. For chronic management of AVNRT, pharmacological therapy and catheter ablation can be considered. Long-term pharmacological therapy (including beta blockers and calcium channel blockers) can be effective in 30% to 60% of patients. On the other hand, catheter ablation is associated with higher efficacy (>95%) and low incidence of complications and therefore has become the preferred initial therapeutic approach. The slow pathway is the target of ablation for all variants of AVNRT. Initially, the rightward inferior AVN extension is targeted. Then, the leftward inferior AVN extension is targeted if needed.
Keywords
atrioventricular nodal reentrant tachycardia, dual atrioventricular nodal physiology, fast pathway, slow pathway, junctional tachycardia
Outline
Anatomy and Physiology of the Atrioventricular Node, 560
Pathophysiology, 562
Tachycardia Circuit, 562
Types of Atrioventricular Nodal Reentry, 563
Epidemiology, 566
Clinical Presentation, 567
Initial Evaluation, 567
Principles of Management, 567
Electrocardiographic Features, 569
Electrocardiographic Manifestations of Dual Atrioventricular Nodal Physiology, 569
Electrocardiographic Manifestations of Atrioventricular Nodal Reentrant Tachycardia, 569
Electrophysiological Testing, 570
Baseline Observations During Sinus Rhythm, 570
Induction of Tachycardia, 575
Tachycardia Features, 578
Diagnostic Maneuvers During Tachycardia, 580
Diagnostic Maneuvers During Sinus Rhythm After Tachycardia Termination, 586
Exclusion of Other Arrhythmia Mechanisms, 587
Ablation, 591
Target of Ablation, 591
Ablation Technique, 591
Endpoints of Ablation, 595
Outcome, 595
Cryoablation of the Slow Pathway, 596
Anatomy and Physiology of the Atrioventricular Node
The atrioventricular node (AVN) is the only normal electrical connection between the atria and the ventricles; the fibrous skeleton acts as an insulator to prevent electrical impulses from entering the ventricles by any other route. The main function of the AVN is modulation of atrial impulse transmission to the ventricles; it introduces a delay between atrial and ventricular systole, thereby allowing atrial systole and ventricular filling to complete prior to the initiation of ventricular systole. Another primary function of the AVN is to limit the number of impulses conducted from the atria to the ventricles. This function is particularly important during fast atrial rates (e.g., during atrial fibrillation [AF] or atrial flutter [AFL]), in which only a few impulses are conducted to the ventricles, and the remaining impulses are blocked in the AVN (facilitated by the relatively long refractory period of the AVN). In addition, fibers in the lower part of the AVN can exhibit automatic impulse formation, serving as a subsidiary pacemaker.
Triangle of Koch
The triangle of Koch constitutes the endocardial surface of the region of the lower right atrium (RA) septum. It is bordered anteriorly by the insertion of the septal leaflet of the tricuspid valve and posteriorly by the fibrous tendon of Todaro. The apex of the triangle is formed by the junction of these two boundaries. The base of the triangle is formed by the anteromedial edge of the coronary sinus ostium (CS os) and is continuous with the sub-eustachian pouch ( Fig. 17.1 ; see Fig. 9.2 , eFig. 12.1 ).
The tendon of Todaro is a fibrous band that connects to the central fibrous body as a fibrous extension of the membranous septum. It courses obliquely between the fossa ovalis and the CS os, crossing the eustachian ridge in the floor of the RA, and connecting to the valve of the inferior vena cava (IVC) (eustachian valve).
Of note, the interatrial sulcus is displaced to the far left of the interventricular sulcus; the atrioventricular (AV) valves are not isoplanar; and the attachment of the septal leaflet of the tricuspid valve into the most anterior part of the central fibrous body is displaced a few millimeters apically relative to the attachment of the septal leaflet of the mitral valve. Thus the true septal part of the AV junction (the RA–left ventricle [LV] sulcus) actually separates the inferomedial RA from the posterior superior process of the LV (the right side above the tricuspid valve while the left side is below the mitral valve). Hence, the triangle of Koch can be considered the RA side of the AV muscular septum.
The mean distances from the His bundle (HB) electrogram recording site (at the apex of the triangle of Koch) to the upper and lower lips of the CS os are 10 mm (range, 0 to 23 mm), and 20 to 25 mm (range, 9 to 46 mm), respectively. However, it is important to note that electroanatomic mapping studies revealed individual variations in the locations of the HB and slow pathway recording sites as well as anatomic variations in the sizes of the Koch triangle and the CS os. A downward deviation of the HB to the midseptum is not uncommon, especially in older patients. Such a downward deviation of the HB may account for unexpected AV block during slow pathway ablation.
Atrioventricular Node
The AVN is an intraatrial structure, measuring approximately 5 mm long, 5 mm wide, and 0.8 mm thick in adults. The compact node is adjacent to the central fibrous body on one side but is uninsulated by fibrous tissue on its other sides, thus allowing contiguity with the atrial myocardium. The AVN is located beneath the RA endocardium at the apex of the triangle of Koch, anterior to the CS os and directly above the insertion of the septal leaflet of the tricuspid valve, where the tendon of Todaro merges with the central fibrous body. Slightly more anteriorly and superiorly is where the HB penetrates the AV junction through the central fibrous body and the posterior aspect of the membranous AV septum.
When traced inferiorly, toward the base of the triangle of Koch, the compact AVN area separates into two extensions, usually with the artery supplying the AVN running between them. The prongs bifurcate toward the CS os along the tricuspid annulus (right posterior [inferior] extension) and toward the mitral annulus (left posterior [inferior] extension). The right posterior nodal extension and its corresponding atrionodal (AN) approaches have been implicated as the anatomic substrate for the so-called “slow pathway” in the atrioventricular nodal reentrant tachycardia (AVNRT) circuit. The tachycardia circuit also can involve the left posterior nodal extension (see later). The “fast pathway” is less well defined from an anatomic and structural standpoint. The probable anatomic substrate of this pathway consists of the transitional cell layers located around the compact AVN (in the anterior portion of the triangle of Koch) at the interface between the compact node and the transitional cells.
Functionally, based on activation times during anterograde and retrograde propagation, and on the action potential characteristics from microelectrode recordings in the rabbit AV junction, the cells of the AVN and perinodal region are frequently described as AN, N (nodal), and NH (nodal-His), with the AN and NH regions more sodium dependent for depolarization, and the N region being calcium dependent, and the likely region where calcium channel blockers act. The transition from one cell area to the other is gradual, with intermediate cells exhibiting intermediate action potentials with great changes related to the autonomic tone.
Atrionodal Region
The AN region corresponds to the cells in the transitional region that are activated shortly after the adjacent atrial cells. Transitional cells are histologically distinct from both the cells of the compact AVN and the working atrial myocytes, and they are not insulated from the surrounding myocardium, but tend to be separated from one another by thin fibrous strands. Transitional cells do not represent conducting tracts but a bridge funneling atrial depolarization into the compact AVN via discrete AN inputs (approaches). AN approaches connect the working atrial myocardium from the left and right sides of the atrial septum to the left and right margins of the compact node, with wider extensions inferiorly and posteriorly between the compact node and the CS os and into the eustachian ridge. In humans and animals, two such AN inputs are commonly recognized in the right septal region: the anterior (superior) approaches, which travel from the anterior limbus of the fossa ovalis and merge with the AVN closer to the apex of the triangle of Koch; and the posterior (inferior) approaches, which are located in the inferoseptal RA and serve as a bridge with the atrial myocardium at the CS os. Although both inputs have traditionally been assumed to be RA structures, growing evidence supports the AV conduction apparatus as a transseptal structure that reaches both atria. A third, middle group of transitional cells has also been identified to account for the nodal connections with the septum and left atrium (LA).
Nodal Region
The N region corresponds to the region where the transitional cells merge with midnodal cells. The N cells represent the most typical of the nodal cells, which are smaller than atrial myocytes, are closely grouped, and frequently are arranged in an interweaving fashion. Sodium channel density is lower in the midnodal zone of the AVN than in the AN and NH cell zones. The inward L-type calcium current is the basis of the upstroke of the N cell action potential. The N cells are characterized by a less negative resting membrane potential and low action potential amplitude, slow rates of depolarization and repolarization, few intercellular connections (e.g., gap junctions), and reduced excitability compared with surrounding cells. Therefore conduction is slower through the compact AVN than the AN and NH cell zones. In fact, the N cells in the compact AVN appear to be responsible for the major part of AV conduction delay and exhibit decremental properties in response to premature stimulation because of their slow rising and longer action potentials. Fast pathway conduction through the AVN apparently bypasses many of the N cells by transitional cells, whereas slow pathway conduction traverses the entire compact AVN. Importantly, the recovery of excitability after conduction of an impulse is faster for the slow pathway than for the fast pathway, for reasons that are unclear.
Nodal-His Region
The NH region corresponds to the lower nodal cells, typically distal to the site of Wenckebach block, connecting to the insulated penetrating portion of the HB. The action potentials of the NH cells are closer in appearance to the fast-rising and long-action potentials of the HB.
Pathophysiology
Tachycardia Circuit
The exact electroanatomic circuit responsible for AVNRT remains elusive. Current evidence suggests that dual AVN pathway physiology constitutes the substrate for AVN reentry. The different atrial inputs to the AVN, rather than functional longitudinal dissociation within the compact AVN, represent the fast and slow pathways involved in the reentrant circuit. The right and left inferior extensions of the AVN and their corresponding AN inputs have been implicated as the anatomic substrate for the slow pathway(s) in the AVNRT circuit. However, the fast pathway is less well defined from an anatomic and structural standpoint. The probable anatomic substrate of this pathway consists of the transitional cell layers located around the compact AVN (in the anterior portion of the triangle of Koch) at the interface between the compact node and transitional cells.
It is important to recognize that dual AVN physiology characterizes the normal AVN electrophysiology, and the presence of dual (or multiple) AVN pathways can be demonstrated in most individuals with or without AVNRT, and is not necessarily indicative of the existence of functional reentry. However, although the potential substrate for AVNRT (dual pathways) is normally available, only a minority of normal individuals develop AVN reentry. This is likely related to the fact that reentry requires several other conditions to be met at the same time, including appropriately timed conduction times and refractoriness over the two pathways as well as the prevalence of perfectly timed triggers (premature atrial complex [PACs] or premature ventricular complex [PVCs])
The understanding of the AVN as having superior (anterior) and inferior (posterior) right and left inputs that form the fast and slow pathways, respectively, is a simple conceptual framework that seems to enable the clinician to confront most cases. Reentry occurring along these pathways is the basic mechanism for the various subtypes of AVNRT. The proximal atrial insertions of the fast and slow pathways are anatomically distinct during retrograde conduction, and several important functional differences exist between the two pathways.
Differences Between the Fast and Slow Pathways
The fast and slow pathways exhibit different electrophysiological (EP) properties. In general, the fast pathway demonstrates faster conduction velocity but longer refractory periods than the slow pathway. However, many exceptions exist.
The fast pathway forms the normal physiological conduction axis, and the atrial–His bundle (AH) interval during conduction over the fast pathway is usually no longer than 220 milliseconds. Longer AH intervals can be caused by conduction over the slow pathway.
Furthermore, the AVN dual pathways are greatly influenced by changes in the sympathovagal balance. Sympathetic stimulation shortens conduction time and refractoriness, whereas vagal stimulation provides the opposite effect. However, the relative extent of these effects on the two pathways can be different, which can potentially cause conduction to shift from one pathway to the other. An increase in vagal tone preferentially prolongs the effective refractory period (ERP) of the fast pathway compared to the slow pathway. Adrenergic stimulation tends to shorten the anterograde and retrograde ERP of the fast pathway to a greater extent than that of the slow pathway. Conversely, beta-blockers tend to prolong ERP of the fast pathway more than that of the slow pathway. Notably, adenosine produces differential effect on anterograde and retrograde conduction over the different AVN pathways. The effect of adenosine seems to be far less potent on retrograde fast pathway conduction than on the retrograde slow pathway and anterograde fast and slow pathway conduction. The mechanism of adenosine resistance of retrograde fast pathway conduction (both in patients with slow-fast AVNRT and in normal subjects) remains unclear.
Investigators used several approaches to identify the location of the atrial input of AVN pathways involved in the AVNRT circuit. The pathway serving as the retrograde limb of an AVNRT circuit can be identified by mapping the site of earliest retrograde atrial activation during AVNRT. The anterograde limb of the AVNRT circuit can be identified by using the resetting response to late-coupled atrial extrastimulation (AES); the site where the AES (with the longest coupling interval) resets the tachycardia identifies the anterograde limb. Atrial entrainment of AVNRT can also identify the anterograde limb of the reentry circuit; the site of the shortest stimulus-to-His interval with a postpacing interval (PPI) equal to the tachycardia cycle length (TCL) identifies the location of the atrial input to the anterograde pathway. Several studies demonstrated that slow pathways with longer conduction times frequently have a more inferior location in the triangle of Koch when compared with locations producing shorter AH intervals. However, atypical locations of those pathways are not infrequent.
Fast Pathway
Typically, a single fast pathway is identified in normal subjects and in patients with AVNRT. During retrograde fast pathway conduction, the earliest retrograde atrial activation is typically recorded at the apex of the triangle of Koch (superior to the compact AVN, at the same site recording the proximal His potential). However, detailed mapping localized the earliest site of atrial activation to the anterior interatrial septum posterior to the tendon of Todaro and eustachian ridge (outside the triangle of Koch, at a level approximately 10 mm inferior to the level recording the proximal HB potential).
Slow Pathways
Data suggest the presence of several slow pathways that can potentially participate in various forms of AVNRT, either as the anterograde or retrograde limbs of the reentry circuit. The slow pathway most commonly incorporated in the AVNRT circuit is formed by the rightward inferior AVN extension (and its corresponding AN approaches), which travels in the triangle of Koch between the tricuspid annulus and the CS os and connects selectively to the floor of the CS os. The earliest atrial activation during retrograde conduction over this pathway is typically located at the inferior aspect of the triangle of Koch, close to the CS os. The second most commonly used slow pathway is formed by the leftward inferior AN extension, which travels within the myocardial coat of the proximal CS leftward (transseptally) toward the left inferoseptal region and mitral annulus. An eccentric activation sequence in the CS is observed during retrograde conduction over this pathway, with the earliest activation site at the roof of the proximal CS (1 to 3 cm from the CS os). Two other slow pathways (inferolateral left atrial pathway and anteroseptal pathway) have been described but are less frequently observed. The earliest retrograde atrial activation is located close to the inferolateral mitral annulus in the LA (for the inferolateral left atrial slow pathway) or the anterior atrial septum, close to the proximal HB (for the anteroseptal slow pathway).
Multiple slow pathways (as demonstrated by multiple discontinuities in the AVN function curves; see later) are present in up to 14% of patients with AVNRT, although not all these pathways are involved in the initiation and maintenance of AVNRT. For example, an eccentric CS activation pattern has been reported frequently (14% to 80%) among patients with atypical forms of AVNRT. Whether the retrograde left-sided AN connection constitutes the critical component of the reentrant circuit or is only an innocent bystander in atypical AVNRT with the eccentric CS activation pattern remains controversial. It may be possible for both the leftward and rightward extensions, either together or separately, to participate in AVN reentry. Right-sided ablation is probably sufficient for most of these patients. However, in some patients, the slow pathway participating in the reentrant circuit cannot be ablated from the posteroseptal RA or the CS os, and requires ablation along the roof of the CS, as much as 5 to 6 cm from the CS os, or mitral annulus.
Upper Turnaround Point
Based on rare cases of dissociation of atrial activation from the tachycardia (e.g., persistence of AVNRT during different patterns of ventriculoatrial [VA] block or during AF) and on similarities between fast-slow AV conduction and longitudinal-transverse conduction in nonuniform anisotropy, early studies proposed that AVNRT results from reentry within the compact AVN (i.e., subatrial) secondary to functional longitudinal dissociation within the AVN into fast and slow pathways. Those studies suggested the presence of an upper common pathway, at least in a subset of patients.
However, current evidence, derived from histological studies, computer modeling, multielectrode recordings, and optical mapping, supports the role of perinodal atrial myocardium and suggests that the fast and slow pathways involved in the reentrant circuit of AVNRT represent conduction over different AN connections, thus making at least a small amount of atrial tissue a necessary part of the reentrant circuit (see Video 17.1 
). The common presence of multiple atrial breakthroughs, and the frequent changes in timing and location of retrograde activation without significant alteration in AVNRT cycle, argues against the existence of an “upper common pathway” or a focal atrial exit site from the AVNRT circuit in the majority of patients.
Lower Turnaround Point
The AVNRT circuit does not involve the ventricles; however, the location of the lower turnaround point of AVN reentry relative to the HB has been controversial. There is good evidence that the distal junction of the slow and fast pathways is located in the AVN, with the existence of a region of AVN tissue extending between the distal junction of the two pathways and the HB (called the “lower common pathway”), at least in a subset of patients.
The existence of a lower common pathway was proposed to explain several observations, including (1) AV block occurring without interruption of AVNRT and without recording of a His electrogram; (2) ventricular extrastimulation (VES) during AVNRT prematurely depolarizing the HB without affecting the tachycardia; (3) AES during AVNRT resulting in changes in the relative activation of HB and atrium (i.e., varying His bundle-atrial [HA] intervals); and (4) the HA interval during ventricular pacing at the TCL is longer than that during AVNRT.
The HA interval (measured from the end of the His potential to the earliest atrial activation in the HB recording, assuming stable activation sequence) during ventricular pacing represents a true conduction time between the HB and atrium. In the presence of a lower common pathway, the HA interval during AVNRT represents the relative activation times between the HB and junctional atrium, since the reentrant wavefront travels retrogradely up an AVN pathway to activate the atrium, while at the same time propagating down the lower common pathway to activate the HB (i.e., the HA is a “pseudo-interval”). Consequently, in the presence of a lower common pathway between the AVNRT circuit and HB recording site, the HA during AVNRT is expected to be shorter than that during ventricular pacing. The difference between the two HA intervals (ΔHA interval) is proportional to the length of the lower common pathway.
However, these phenomena can be interpreted in ways that do not involve the presence of a common pathway. For example, the first two phenomena assume that the recorded “proximal HB” potential corresponds to the actual “proximal end of the HB,” which is probably inaccurate in many cases. Hence, those phenomena also can be explained by intra-Hisian block occurring beyond the site of HB recording rather than in a “lower common pathway.” Furthermore, the last two phenomena assume that retrograde conduction follows the same pathway during pacing and tachycardia, and the difference between the HA interval during tachycardia and that during ventricular pacing at the TCL was assumed to reflect the conduction time over the lower common pathway. On the contrary, mapping studies have demonstrated that the breakthrough of atrial activation during AVNRT is frequently slightly discordant from that observed during ventricular pacing. Hence, relying on the measuring the HA interval from the same recoding sites, rather than detailed mapping for the true site of “earliest” atrial activation, may not be accurate.
Types of Atrioventricular Nodal Reentry
AVNRT can manifest in different forms depending on the anatomic substrate forming the anterograde and retrograde pathways incorporated in the reentry circuit. Traditionally, AVNRT has been classified into “typical” or “atypical” forms. Typical AVNRT (anterograde slow-retrograde fast) accounts for 90% of AVNRTs. Atypical AVNRT variants are traditionally subclassified as either fast-slow or slow-slow types. The distinction between these various forms of AVNRT has been based on: (1) the absolute values of the AH and HA intervals; (2) the AH/HA ratios; (3) the pattern of earliest retrograde atrial activation; and (4) the identification of a lower common pathway ( Table 17.1 ).
| Typical (Slow-Fast) | Atypical (Fast-Slow) | Atypical (Slow-Slow) | |
|---|---|---|---|
| AH interval | >200 ms | <200 ms | >200 ms |
| HA interval | <70 ms | ≥70 ms | ≥70 ms |
| AH/HA ratio | >1 | <1 | ≥1 |
| Site of earliest retrograde atrial activation | Apex of the triangle of Koch | CS os or within 1–2 cm of the proximal CS | CS os or within 1–2 cm of the proximal CS |
| Lower common pathway | Short or absent | Long | Long |
However, this classification approach has several limitations. As noted previously, the methodology for determining the presence or absence of a lower common pathway is not reliable (as discussed above). In addition, retrograde atrial activation over both fast and slow conduction patterns exhibit significant heterogeneity in all forms of AVNRT. Both typical and atypical AVNRT are compatible with varying retrograde atrial activation patterns. Furthermore, the conduction properties of the fast and slow AVN pathways (and thus the absolute and relative values of the AH and HA intervals) depend on the autonomic status, and can change in the same patient during the same EP study in response to changes in the sympathovagal balance.
A more simplified and clinically practical scheme was recently proposed, which classifies the different types of AVNRT into only two categories, typical versus atypical, based only on the AH/HA ratio and absolute HA (or VA) intervals, while disregarding the retrograde atrial activation sequence and the demonstration of a lower common pathway ( Table 17.2 ). Also, both fast-slow and slow-slow atypical AVNRTs were combined together since the distinction between the two types is often arbitrary in view of the lack of a unanimously accepted definition.
| Typical (Slow-Fast) | Atypical (Fast-Slow or Slow-Slow) | |
|---|---|---|
| HA interval | ≤70 msec | >70 msec |
| VA interval | ≤60 msec | >60 msec |
| AH/HA ratio | >1 | Variable |
The HA interval is measured from earliest deflection of the HB activation to the earliest rapid deflection of the atrial activation in the HB electrogram. The VA interval (measured from the onset of ventricular activation on surface electrocardiogram [ECG] to the earliest rapid deflection of the atrial activation on the HB electrogram) is also a practical and easily obtainable criterion, when the His potential cannot be reproducibly and reliably recorded during tachycardia.
Typical (Slow-Fast) Atrioventricular Nodal Reentrant Tachycardia
Anterograde conduction.
The anterograde limb of the reentry circuit is formed by a slow pathway. Mapping studies localized the input to the anterograde slow pathway to the midseptum (in 50%) or inferior septum (33%), and infrequently to the superior septum (13%) or CS (3%). Some investigators subdivided typical slow-fast AVNRT according to the putative slow pathway used anterogradely, based on the site of successful ablation. At least three subtypes were described: (1) rightward inferior extension slow-fast AVNRT (most common), which is typically eliminated by ablation at the inferior aspect of the triangle of Koch; (2) leftward inferior extension slow-fast AVNRT (uncommon, 5%), which requires ablation within the triangle of Koch superior to the level of CS os and closer to the compact AVN or at the roof of the proximal CS (1 to 3 cm from the CS os); and (3) inferolateral left atrial slow-fast AVNRT (rare), which requires ablation close to the inferolateral mitral annulus in the LA.
Retrograde conduction.
Typical AVNRT uses the fast pathway for retrograde conduction. During slow-fast AVNRT, the earliest retrograde atrial activation is usually recorded at the anterior interatrial septum posterior to the tendon of Todaro, close to the apex of the triangle of Koch ( Fig. 17.2 ).
Importantly, retrograde atrial activation has been mapped to the inferior triangle of Koch, the roof of the CS (1 to 3 cm from the CS os), or the left side of the septum in up to 9% of patients ( Fig. 17.3 ). These forms of AVNRT were considered variants of slow-fast AVNRT with an inferior exit for the retrograde fast pathway. More recently, however, those AVNRTs with earliest retrograde atrial activation outside the traditional location of the fast pathway (formerly described as “posterior” or “leftward inferior extension” or “left variant” slow-fast AVNRT) have been considered as a variant “slow-slow AVNRT.” In the latter setting, two slow pathways (the right and left inferior AVN extensions) are used in the AVNRT circuit. The very short (and occasionally negative) HA interval observed during tachycardia, despite retrograde conduction over a slow pathway, could be explained by the presence of a relatively long lower common pathway. Retrograde conduction over the slow pathway with simultaneous bystander conduction over the lower common pathway (from the distal junction of the two slow pathways to the HB) results in simultaneous atrial and ventricular activation and shortening of the recorded HA interval, which mimics slow-fast AVNRT. Older patients (>60 years) often have longer conduction times over the retrograde fast pathway, resulting in HA intervals similar to those in slow-slow AVNRT, but still with the earliest atrial activation at the tendon of Todaro.
Lower common pathway.
The presence of a lower common pathway in typical AVNRT remains controversial. Nevertheless, it is recognized that the lower common pathway in typical AVNRT, if present, is very short (as assessed by the degree of HB prematurity required for a VES to reset the tachycardia, and by comparing the HA interval during AVNRT with that during ventricular pacing at the TCL).
AH/HA ratio.
The AH interval is long (>200 milliseconds), due to anterograde conduction over the slow pathway. The HA interval is relatively short (<70 milliseconds) given the fast retrograde conduction. This produces an AH/HA ratio of greater than 1 and simultaneous atrial and ventricular activations (the onset of atrial activation appears before, at the onset, or just after the QRS complex).
Atypical (Fast-Slow) Atrioventricular Nodal Reentrant Tachycardia
Anterograde conduction.
The nature of anterograde pathway conduction during fast-slow AVNRT remains poorly understood. Although this variant was initially thought of as using the same circuit as typical slow-fast AVNRT but in the reverse direction, recent data suggest that anterograde “fast pathway” during atypical AVNRT is distinct from the retrograde “fast pathway” during typical AVNRT. Some investigators have suggested that anterograde conduction during fast-slow AVNRT is mediated by a slow pathway, and that the fast pathway is a bystander that mediates conduction to the HB (and results in a short AH interval) but without contributing to the reentry circuit (analogous to AVNRT with ventricular preexcitation over a bystander bypass tract [BT]). Of note, patients with fast-slow AVNRT often exhibit multiple AH interval jumps during AES testing, which is consistent with the presence of multiple slow pathways, and supporting reentry between two “slow pathways.”
Retrograde conduction.
A slow pathway forms the retrograde limb of the reentry circuit. The earliest retrograde atrial activation (over the slow pathway) during fast-slow AVNRT is traditionally reported at the base of the triangle of Koch, near the CS os (see Fig. 17.2 ). However, other locations are also frequent, including mid or superior septum, distal CS, or left side of the septum (see Fig. 17.2 ). Thus fast-slow AVNRT can be of posterior, anterior, or middle type according to the mapped location of the retrograde slow pathway.
Lower common pathway.
A relatively long lower common pathway has been observed in fast-slow AVNRT, in contrast to slow-fast AVNRT.
AH/HA ratio.
The AH interval is shorter than the HA interval (30 to 185 milliseconds vs. 135 to 435 milliseconds), resulting in long RP tachycardia and an AH/HA ratio less than 1. The long HA interval is a result of slow retrograde conduction over the slow pathway. The short AH interval represents anterograde conduction over the fast pathway; however, whether this pathway serves as the anterograde limb of the reentry circuit, or as just a bystander, is being debated.
Atypical (Slow-Slow) Atrioventricular Nodal Reentrant Tachycardia
As the name implies, the reentrant circuit in slow-slow AVNRT uses two slow pathways (the right and left inferior AVN extensions). These patients often exhibit multiple AH interval jumps during AES testing, which is consistent with multiple slow pathways.
Anterograde conduction.
A slow pathway forms the anterograde limb of the reentry circuit. Data suggest the slow pathway used for anterograde conduction in slow-slow AVNRT is similar to the one used for retrograde conduction during fast-slow AVNRT.
Retrograde conduction.
A second slow (or “intermediate”) pathway serves as the retrograde limb of the reentry circuit. The earliest retrograde atrial activation occurs along the roof of the proximal CS (1 to 3 cm from the CS os) or, less commonly, at the inferoposterior aspect of the triangle of Koch (see Fig. 17.2 ). Of note, patients with this form of AVNRT can also have retrograde conduction over the fast pathway, which can be demonstrated during ventricular pacing, which is not part of the reentrant circuit.
Lower common pathway.
Slow-slow AVNRT exhibits a relatively long lower common pathway, significantly longer than that in typical AVNRT.
AH/HA ratio.
The AH interval is long (>200 milliseconds) due to the slow retrograde conduction over the slow pathway. The HA interval is often shorter than the AH interval, but is usually greater than 70 milliseconds, and the AH/HA ratio remains greater than 1. The short HA interval observed during slow-slow AVNRT (despite retrograde conduction over a slow pathway) could be explained by the presence of a relatively long lower common pathway. Retrograde conduction over the slow pathway with simultaneous bystander conduction over the lower common pathway (from the distal junction of the two slow pathways to the HB) results in simultaneous atrial and ventricular activation, and shortening of the recorded HA interval, which mimics slow-fast AVNRT.
Although the HA interval during slow-slow AVNRT is usually longer than that during slow-fast AVNRT, a significant overlap exists. Slow-slow AVNRT can exhibit very short AH intervals, mimicking slow-fast AVNRT. In fact, this form of AVNRT was considered a variant of slow-fast AVNRT, formerly described as “posterior” or “type B” AVNRT (which accounted for approximately 2% of patients with slow-fast AVNRT). Nonetheless, several features of slow-slow AVNRT can help distinguish it from slow-fast AVNRT, including: (1) the earliest retrograde atrial activation is recorded at the roof of the CS or at the inferior triangle of Koch, rather than the fast pathway area; (2) there is a much wider range of HA intervals; (3) there is a much more common cycle length (CL) changes and especially changes in HA interval during tachycardia; and (4) there is a relatively long lower common pathway (HA interval during ventricular pacing at the TCL exceeding the HA interval during tachycardia [ΔHA] by ≥15 milliseconds).
Epidemiology
AVNRT is the most common form of paroxysmal supraventricular tachycardia (SVT). The absolute number of patients with AVNRT and its proportion of paroxysmal SVT increase with age. The reason may be related to the normal evolution of AVN physiology over the first two decades of life, as well as to age-related changes in atrial and AVN physiology observed in later decades. AVNRT is unusual in children younger than 5 years, and it typically initially manifests in early adult life (e.g., in the teens). Conversely, atrioventricular reentrant tachycardia (AVRT) manifests earlier, with an average of more than 10 years separating the time of clinical presentation of AVRT and that of AVNRT. AVNRT onset has been reported after the age of 50 years in 16% and before the age of 20 years in 18%. There is also a striking 2 : 1 predominance of AVNRT in women, in whom symptoms start at a significantly younger age. As such, female sex and older age (i.e., teens vs. newborns or young children) favor the diagnosis of AVNRT over AVRT. Gender differences in the anterograde and retrograde AVN EP properties have been observed and may contribute to the pathogenesis of AVNRT. There is no significant association of AVNRT with other types of structural heart disease; patients with ventricular tachycardia (VT) in the absence of structural heart disease have a higher prevalence of AVNRT.
Clinical Presentation
Patients with AVNRT typically present with the clinical syndrome of paroxysmal SVT. This is characterized as regular rapid tachycardia of abrupt onset and termination. Patients commonly describe palpitations and dizziness. Rapid ventricular rates can be associated with complaints of dyspnea, weakness, chest pain, or presyncope, and can at times be disabling. True syncope is uncommon but can occur, especially in elderly patients. Episodes can last from seconds to several hours. AVNRT can occur spontaneously or on provocation with exertion, caffeine, or alcohol. Patients often learn to use certain maneuvers, such as the carotid sinus massage or the Valsalva maneuver, to terminate the arrhythmia, although many require pharmacological treatment.
About half of patients with typical AVNRT report experiencing a pounding sensation in the neck during tachycardia, which likely is related to pulsatile reversed flow when the RA contracts against a closed tricuspid valve due to simultaneous contraction of atria and ventricles. The physical examination correlate of this phenomenon is continuous pulsing cannon A waves in the jugular venous waveform (described as the “frog” sign). This clinical feature has been reported to distinguish paroxysmal SVT resulting from AVNRT from that caused by orthodromic AVRT. Although atrial contraction during AVRT occurs against closed AV valves, the longer VA interval results in separate ventricular and then atrial contraction and a relatively lower RA and venous pressure. Therefore the presence of palpitations in the neck is experienced less commonly (about 17%) in patients with AVRT. Polyuria is particularly common with AVNRT and is related to higher right atrial pressures and elevated levels of atrial natriuretic protein in patients with AVNRT compared with patients who have AVRT or AFL.
Initial Evaluation
History, physical examination, and 12-lead ECG constitute an appropriate initial evaluation. In patients with brief, self-terminating episodes, an event recorder is the most effective way to obtain ECG documentation. An echocardiographic examination should be considered in patients with documented sustained SVT to exclude the possibility of structural heart disease. Further diagnostic studies (e.g., cardiac stress testing) are indicated only if there are signs or symptoms that suggest structural heart disease.
The diagnosis of AVNRT as the mechanism of SVT can be strongly suspected based on the surface ECG but can be difficult to confirm, especially when only single-lead rhythm strips are available during the SVT. However, EP testing is not indicated unless a decision to proceed with catheter ablation is undertaken.
Principles of Management
Acute Management
Because maintenance of AVNRT is dependent on AVN conduction, maneuvers or drugs that slow AVN conduction and prolong AVN refractoriness can be used to terminate the tachycardia. Initially, maneuvers that increase vagal tone (e.g., Valsalva maneuvers, gagging, carotid sinus massage) are used. When vagal maneuvers are unsuccessful, tachycardia termination can be achieved with antiarrhythmic drugs whose primary effects increase refractoriness or decrease conduction (negative dromotropic effect) over the AVN ( Fig. 17.4 ). Adenosine is the drug of choice and is successful in more than 95% of cases. In addition, intravenous (IV) diltiazem and verapamil are particularly effective in terminating AVNRT and can be used in hemodynamically stable patients. Beta-blockers are also a reasonable option. Digoxin, which has a slower onset of action than the other AVN blockers, is not favored for the acute termination of AVNRT, except if there are relative contraindications to the other agents.
Class IA and IC sodium channel blockers can also be used in treating an acute event of AVNRT when other regimens have failed, a strategy that is rarely needed. Electrical cardioversion is recommended for hemodynamically unstable patients and those with persistent arrhythmia refractory to pharmacological therapy. Energies in the range of 10 to 50 J are usually adequate.
Chronic Management
Because AVNRT is generally a benign arrhythmia that does not influence survival, the primary indication for its treatment relates to its impact on a patient’s quality of life ( Fig. 17.5 ). Factors that contribute to the therapeutic decision include the frequency and duration of tachycardia, tolerance of symptoms, the effectiveness and tolerance of antiarrhythmic drugs, the need for lifelong drug therapy, and the presence of concomitant structural heart disease. Patients who develop a highly symptomatic episode of paroxysmal SVT, particularly if it requires an emergency room visit for termination, may elect to initiate therapy after a single episode. In contrast, a patient who presents with minimally symptomatic episodes of paroxysmal SVT that terminate spontaneously or in response to Valsalva maneuvers may elect to be followed clinically without specific therapy. These patients should be taught how to correctly perform vagal maneuvers and be educated about when to seek medical attention.
Catheter Ablation
Once it is decided to initiate treatment for AVNRT, the question arises whether to initiate pharmacological therapy or to use catheter ablation. Because of its high efficacy (>95%) and low incidence of complications, catheter ablation has become the preferred therapy over long-term pharmacological therapy and can be offered as an initial therapeutic option. It is reasonable to discuss catheter ablation with all patients suspected of having AVNRT. However, patients considering radiofrequency (RF) ablation must be willing to accept the risk, albeit low, of AV block and pacemaker implantation.
Pharmacological Therapy
For patients with AVNRT who are not candidates for, or prefer not to undergo, catheter ablation, long-term pharmacological therapy can be effective in 30% to 60% of patients. Most pharmacological agents that depress AVN conduction (including beta-blockers and calcium channel blockers) can reduce the frequency of recurrences of AVNRT. If those agents are ineffective, class IC (flecainide or propafenone in patients without structural or ischemic heart disease) or class III antiarrhythmic agents (sotalol or dofetilide) may be considered. Given the potential adverse effects of digoxin and amiodarone, these agents are generally reserved as last-resort therapy.
Outpatients may use a single dose of verapamil, diltiazem, or propranolol to acutely terminate an episode of AVNRT. This so-called pill-in-the-pocket approach (i.e., administration of a drug only during an episode of tachycardia for the purpose of termination of the arrhythmia when vagal maneuvers alone are not effective) is appropriate to consider for patients with infrequent episodes of AVNRT that are prolonged but well tolerated, and it obviates exposure of patients to long-term and unnecessary therapy between rare arrhythmic events. This approach necessitates the use of a drug that has a short onset of action (i.e., immediate-release preparations). Candidate patients should be free of significant LV dysfunction, sinus bradycardia, and preexcitation. Single-dose oral therapy with diltiazem (120 mg) plus propranolol (80 mg) has been shown to be superior to both placebo and flecainide in terminating AVNRT.
Electrocardiographic Features
Electrocardiographic Manifestations of Dual Atrioventricular Nodal Physiology
As noted, dual pathway physiology characterizes the normal AVN electrophysiology, which is prevalent in most normal individuals. However, the ECG normally reveals only the conduction of the fast pathway, while slow pathway conduction generally remains concealed. The most common ECG manifestation of dual AVN physiology is AVNRT. In addition, several ECG manifestations can be explained by dual pathways physiology.
Two Families of PR Intervals
A shift of anterograde conduction from the fast pathway to the slow pathway can manifest as sudden prolongation of the PR interval, and vice versa ( Fig. 17.6 ). This shift can occur spontaneously and result in alternans of the PR interval (i.e., long and short PR intervals alternate with each other) or manifest as two families of PR intervals (short and long), with sudden and sustained prolongation or shortening of the PR interval that can be observed on different occasions. The shift also can be precipitated by a change in autonomic tone or by a PAC or a PVC that causes conduction block or concealment in one pathway and allows conduction over the other ( Fig. 17.7 ).
Dual Ventricular Response to a Single Supraventricular Beat
Rarely, a single atrial impulse (e.g., PAC) can conduct simultaneously along the slow and fast pathways, producing two QRS complexes (“double fire”). The reverse can also occur, with two atrial impulses from one ventricular complex.
Rapid Ventricular Rates During Atrial Fibrillation
Ventricular rates during AF can exhibit a bimodal distribution of R-R intervals on Holter recordings. The shorter R-R intervals (the faster ventricular rates) are believed to be a result of conduction over the slow pathway (because of its shorter refractory periods), while the fast pathway mediates relatively slower ventricular rates. Ablation of the slow AVN pathways in these patients can potentially eliminate the fast ventricular rates and produce a unimodal R-R interval distribution.
Electrocardiographic Manifestations of Atrioventricular Nodal Reentrant Tachycardia
P Wave Morphology
In typical (slow-fast) AVNRT, the P wave is usually not visible because of the simultaneous atrial and ventricular activation. The P wave can distort the initial portion of the QRS (mimicking a q wave in the inferior leads), lie just within the QRS (inapparent), or distort the terminal portion of the QRS (mimicking an s wave in the inferior leads or a terminal r wave in lead V1) ( Fig. 17.8 ). When apparent, the P wave is significantly narrower than the sinus P wave, and is negative in the inferior leads, findings that are consistent with concentric retrograde atrial activation over the fast AVN pathway. In atypical AVNRT, the P wave is relatively narrow, negative in the inferior leads, and positive in lead V1 (see Fig. 17.8 ). Compared to orthodromic AVRT using a posteroseptal BT, the retrograde P waves in slow-slow AVNRT are more negative in lead aVF (>0.16 mV negative amplitude).
QRS Morphology
The QRS morphology during AVNRT is usually the same as in normal sinus rhythm (NSR). The development of prolonged functional aberration during AVNRT is uncommon, and it usually occurs following the induction of AVNRT by ventricular stimulation more frequently than by atrial stimulation, or following the resumption of 1 : 1 conduction to the ventricles after a period of block below the tachycardia circuit. At times, alternans of QRS amplitude can occur when the tachycardia rates are rapid. Occasionally, AVNRT can coexist with ventricular preexcitation over an AV BT, whereby the BT is an innocent bystander.
P-QRS Relationship
In typical (slow-fast) AVNRT, the RP interval is very short (−40 to 75 milliseconds). Variation of the P-QRS relationship with or without block can occur during AVNRT, especially in atypical variants of the tachycardia. This phenomenon usually occurs when the conduction system and the reentry circuit are unstable during initiation or termination of the tachycardia, which is likely secondary to decremental conduction in the lower common pathway. The ECG manifestation of P-QRS variations with or without AV block during tachycardia, especially at the initiation of tachycardias or in cases of nonsustained tachycardias, can be misdiagnosed as atrial tachycardias (ATs). Moreover, the variations can be of such magnitude that long RP tachycardia can masquerade for brief periods of time as short RP tachycardia.
Usually, the A/V ratio during AVNRT is equal to 1; however, 2 : 1 AV block can be present because of a block below the reentry circuit (usually below the HB and, infrequently, in the lower common pathway). In such cases, narrow, inverted P wave morphology in the inferior leads inscribed exactly between QRS complexes strongly suggests AVNRT ( Fig. 17.9 ). The incidence of reproducible sustained 2 : 1 AV block during induced episodes of AVNRT is approximately 10%. Rarely, VA block can occur because of a block in an upper common pathway; however, some of these cases may represent reentry using a retrogradely conducting nodofascicular or nodoventricular pathway with intranodal block above the level of the circuit ( see Chapter 19 ).
In atypical (fast-slow) AVNRT, the RP interval is longer than the PR interval. In slow-slow AVNRT, the RP interval is usually shorter than, and sometimes equal to, the PR interval. Occasionally, the P wave is inscribed in the middle of the cardiac cycle, thus mimicking AT with 2 : 1 AV conduction (see Fig. 17.2 ). Slow-slow AVNRT can be associated with RP intervals and P wave morphology similar to that during orthodromic AVRT using a posteroseptal AV BT. However, although both SVTs have the earliest atrial activation in the posteroseptal region, conduction time from that site to the HB region is significantly longer in AVNRT than in orthodromic AVRT. The results are a significantly longer RP interval in lead V1 and a significantly larger difference in the RP interval between lead V1 and inferior leads during AVNRT. Therefore ΔRP interval (V1 − III) of more than 20 milliseconds suggests slow-slow AVNRT (sensitivity, 71%; specificity, 87%).
Electrophysiological Testing
EP testing is used to study the inducibility and mechanism of the SVT and to guide catheter ablation. Typically, three quadripolar catheters are positioned in the high RA, the right ventricular (RV) apex, and the HB region, and a decapolar catheter is positioned in the CS ( see Fig. 4.4 ). A typical programmed electrical stimulation protocol used for EP testing in patients with AVNRT is outlined in Box 17.1 .
Atrial burst pacing from the RA and CS (down to AV Wenckebach CL)
Single and double AESs at multiple CLs (600–400 ms) from the high RA and CS (down to atrial ERP)
Ventricular burst pacing from the RV apex (down to VA Wenckebach CL)
Single and double VESs at multiple CLs (600–400 ms) from the RV apex (down to ventricular ERP)
Administration of isoproterenol infusion as needed to facilitate tachycardia induction (0.5–4 µg/min)
AES , Atrial extrastimulus; AV , atrioventricular; CL , cycle length; CS , coronary sinus; ERP , effective refractory period; RA , right atrium; RV, right ventricle; VA , ventriculoatrial; VES , ventricular extrastimulus.
Baseline Observations During Sinus Rhythm
Programmed Atrial Stimulation During Sinus Rhythm
Anterograde dual AVN physiology.
Demonstration of anterograde dual AVN pathway conduction curves requires a longer ERP of the fast pathway than the slow pathway ERP and the atrial functional refractory period (FRP), as well as a sufficient difference in conduction times between the two pathways. Dual AVN physiology can be diagnosed by demonstrating one of the following: (1) a “jump” in the AH interval in response to progressively more premature AES; (2) two ventricular responses to a single atrial impulse; (3) a PR interval exceeding the R-R interval during rapid atrial pacing; or (4) different PR or AH intervals during NSR or fixed-rate atrial pacing ( Box 17.2 ).
Manifestations of Anterograde Dual Atrioventricular Node Physiology
- •
A “jump” in the AH interval of ≥50 ms in response to a 10-ms decrement of the AES coupling interval or atrial PCL
- •
Dual ventricular response to a single atrial beat (“double fire”)
- •
PR interval exceeding the R-R interval during rapid atrial pacing
- •
Two distinct PR or AH intervals during NSR or fixed-rate atrial pacing
Manifestations of Retrograde Dual Atrioventricular Node Physiology
- •
A “jump” in HA interval of ≥50 ms in response to a 10-ms decrement of the VES coupling interval or ventricular PCL
- •
Two atrial responses to a single ventricular impulse
AES , Atrial extrastimulation; AH , atrial-His; CL , cycle length; NSR , normal sinus rhythm; PCL , pacing cycle length; VES , ventricular extrastimulation.
AH interval jump.
In contrast to the normal pattern of AVN conduction, in which the AH interval gradually lengthens in response to progressively shorter AES (i.e., shorter coupling intervals), patients with dual AVN physiology usually demonstrate a sudden increase (“jump”) in the AH interval at a critical AES (A1-A2) coupling interval ( eFig. 17.1 ). Conduction with a short PR or AH interval reflects fast pathway conduction, whereas conduction with a long PR or AH interval reflects slow pathway conduction. The AH interval jump signals block of anterograde conduction of the progressively premature AES over the fast pathway (once the AES coupling interval becomes shorter than the fast pathway ERP) and anterograde conduction over the slow pathway (which has an ERP shorter than the AES coupling interval), with a longer conduction time (i.e., longer A2-H2 interval). A jump in the A2-H2 (or H1-H2) interval of ≥50 milliseconds in response to a 10 millisecond shortening of either the A1-A2 interval (i.e., AES coupling interval) or the A1-A1 interval (i.e., pacing cycle length [PCL]) is defined as a discontinuous AVN function curve and is considered evidence of dual anterograde AVN pathways ( see Fig. 4.23 ).
Two ventricular responses to a single atrial impulse.
Rapid atrial pacing or AES can result in two ventricular complexes to a single paced atrial impulse (referred to as “1 : 2 response”). The first ventricular complex is caused by conduction of the atrial impulse over the fast AVN pathway, and the second complex is caused by conduction over the slow AVN pathway ( Fig. 17.10 ). This response requires a unidirectional retrograde block in the slow AVN pathway. Typically, in the presence of dual AVN pathways, conduction propagates simultaneously over both fast and slow AVN pathways. However, the wavefront conducting down the fast pathway reaches the distal junction of the two pathways before the impulse conducting down the slow pathway; hence, it conducts retrogradely up the slow pathway to collide with the impulse conducting anterogradely down that pathway. Thus the anterograde impulse conducting down the slow pathway does not have the opportunity to reach the HB and the ventricle. Rarely, however, the slow pathway conducts only anterogradely or has a very long retrograde ERP. In this setting, the wavefront traveling anterogradely down the fast pathway blocks (but does not conceal) in the slow pathway retrogradely and fails to retard the impulse traveling anterogradely down that pathway. Consequently, the wavefront traveling down the slow pathway can reach the HB and ventricle to produce a second His potential and QRS in response to a single atrial impulse. Because retrograde block in the slow pathway is a prerequisite to a 1 : 2 response, when such a phenomenon is present, it indicates that the slow pathway cannot support reentrant tachycardia using the slow pathway as the retrograde limb. The 1 : 2 response should be differentiated from pseudo-simultaneous fast and slow pathway conduction, which is a much more common phenomenon during rapid atrial pacing. In the latter case, all paced atrial impulses block anterogradely in the fast pathway and conduct exclusively down the slow pathway with prolonged AH intervals (with PR intervals longer than atrial PCL), so that the last paced atrial impulse falls before the His potential caused by conduction of the preceding paced atrial impulse. Thus the last paced atrial impulse is followed by two His potentials and two ventricular complexes. The last response may then be followed by induction of AVN echo beats or AVNRT, mimicking simultaneous fast and slow pathway conduction ( Fig. 17.11 ).
PR interval longer than RR interval.
The PR interval gradually prolongs as the atrial pacing rate increases. When a critical pacing rate is reached, the PR interval typically exceeds the R-R interval, with all AVN conduction over the slow AVN pathway (see Fig. 17.11 ). This manifests as crossing over of the pacing stimulus artifacts and QRSs; that is, the paced atrial complex is conducting not to the QRS immediately following it, but rather to the next QRS, because of a very long PR interval. There should be consistent 1 : 1 AV conduction that remains stable over the span of several cycles for this observation to be interpreted (i.e., without Wenckebach block). Such slow AVN conduction, sometimes called “skipped P waves,” is seen only when conduction propagates over a slow AVN pathway, and it is not seen in the absence of dual AVN physiology. This phenomenon is diagnostic of the presence of dual AVN physiology, even in the absence of an AH interval jump and therefore is very helpful in patients with smooth AVN function curves. In fact, 96% of patients with AVNRT and smooth AVN function curves have a PR/RR interval ratio greater than 1 (i.e., PR interval longer than PCL) during atrial pacing at the maximal rate with consistent 1 : 1 AV conduction (vs. 11% in controls).
Two distinct AH intervals during NSR or at identical atrial PCLs.
This phenomenon can occur when the fast pathway anterograde ERP is long relative to the sinus or paced CL (see Fig. 17.6 ). Such a phenomenon also requires a long retrograde ERP of the fast pathway. Otherwise, AVN echo beats or AVNRT would result, because once the impulse blocks anterogradely in the fast pathway and is conducted down the slow pathway, it would subsequently conduct retrogradely up the fast pathway if the ERP of the fast pathway were shorter than the conduction time (i.e., shorter than the AH interval) over the slow pathway.
Determinants for the occurrence of sustained slow pathway conduction include markedly abnormal, anterograde and retrograde conduction properties of the fast pathway and, possibly, differential sensitivity to vagal activity of the fast pathway, compared with the slow pathway.
Multiple AVN pathways.
Multiple AH interval “jumps” in response to AES, a finding suggesting the presence of multiple AVN pathways, can be observed in up to 14% of patients with AVNRT, especially in those with atypical variants of AVNRT. These phenomena are characterized by multiple AH interval jumps of 50 milliseconds or more in response to an increasingly premature AES. In these patients, a single AES can initiate multiple jumps in only 68%, whereas double AESs or atrial pacing is required in 32%. Such patients can have AVNRT with longer TCLs and longer ERP and FRP of the AVN. It is uncommon for multiple AVNRTs with different TCLs and P-QRS relationships to be present in the same patient.
Prevalence of dual AVN physiology.
The presence of dual AVN pathways can usually be demonstrated by using a single AES or atrial pacing in 85% of patients with clinical AVNRT. In 95% of patients, the presence of dual AVN pathways can be revealed by using multiple AESs, multiple-drive CLs (typically 600 and 400 milliseconds), and multiple pacing sites (typically high RA and CS).
Occasionally, the AH interval continuously prolongs (without a discrete “jump”) in response to more premature AES, with “smooth transition” of anterograde conduction from the fast pathway to the slow pathway, until retrograde conduction starts over the “fast pathway” and AVNRT is induced. Failure to demonstrate dual AVN physiology in patients with AVNRT can be caused by minimal differences in the anterograde refractory periods of the fast and slow AVN pathways. In this setting, dissociation of refractoriness of the fast and slow AVN pathways is required and can be achieved by any of the following: (1) introduction of an AES at a shorter pacing drive CL; (2) introduction of multiple AESs; (3) burst atrial pacing; or (4) pharmacological modulation of AVN dual pathway conduction and refractoriness.
In general, if fast pathway conduction is suppressed at baseline (as evidenced by a long AH interval at all atrial pacing rates or VA block during ventricular pacing), isoproterenol infusion (and occasionally atropine) usually facilitates fast pathway conduction. In contrast, if the baseline ERP of the fast pathway is very short, conduction over the slow pathway can be difficult to document. Increasing the degree of sedation or infusion of esmolol can prolong the fast pathway ERP and allow recognition of slow pathway conduction.
Another potential reason for the inability to demonstrate dual AVN physiology is a block in the fast AVN pathway at the pacing drive CL (i.e., fast pathway ERP is longer than pacing drive CL). In addition, atrial FRP can limit the prematurity of the AES. Consequently, AVN activation cannot be adequately advanced to produce block in the fast pathway because a more premature AES would result in more intraatrial conduction delay and less premature stimulation of the AVN. This obstacle can be overcome by the introduction of an AES following a shorter pacing drive CL, introduction of multiple AESs, burst atrial pacing, or stimulation from multiple atrial sites.
Programmed Ventricular Stimulation During Sinus Rhythm
Retrograde dual AVN physiology.
Demonstration of retrograde dual AVN pathway conduction curves requires a longer retrograde ERP of the fast pathway than slow pathway ERP and ventricular and His-Purkinje system (HPS) FRP, as well as a sufficient difference in conduction times between the two pathways. In a pattern analogous to that of anterograde dual AVN physiology, ventricular stimulation can result in discontinuous retrograde AVN function curves, manifesting as a jump in the H2-A2 (or A1-A2) interval of 50 milliseconds or more in response to a 10 millisecond decrement of the VES coupling interval (V1-V2) or ventricular PCL (V1-V1). This finding must be distinguished from sudden VA prolongation caused by VH interval (but not HA interval) prolongation related to retrograde functional block in the right bundle branch (RB) and transseptal activation of HB through the left bundle branch (LB) ( see eFig. 4.9 ). A 1 : 2 response (i.e., two atrial responses to a single ventricular stimulus) can also be observed (see Box 17.2 ).
Failure to demonstrate retrograde dual AVN physiology in patients with atypical AVNRT can be caused by similar fast and slow AVN pathway retrograde refractory periods. Dissociation of refractoriness of the fast and slow AVN pathways may be required and usually can be achieved by any of the following: (1) introduction of VESs at a shorter pacing drive CL; (2) introduction of multiple VESs; (3) burst ventricular pacing; or (4) administration of drugs such as beta-blockers, verapamil, or digoxin. In addition, retrograde block in the fast AVN pathway at the pacing drive CL (i.e., the PCL is shorter than the fast pathway ERP) and ventricular or HPS FRP interval limiting the prematurity of the VES can also account for such failure.
Differential-site RV pacing.
Differential-site RV pacing can help exclude the presence of a retrogradely conducting septal AV BT. The response to differential RV pacing can be evaluated by comparing two variables between RV basal and RV apical (or midseptal) pacing: the VA interval (i.e., the stimulus-to-atrial [SA] interval) and atrial activation sequence ( see Fig. 18.28 ). This maneuver is discussed in detail in Chapter 20 .
The RV apical septum, although anatomically more distant from the atrium than the RV base, is nonetheless electrically closer because of the proximity of the distal RB to the pacing site. As a result, in the absence of a retrogradely conducting septal AV BT, pacing at the RV apex allows entry into the rapidly conducting HPS and results in a shorter VA interval during pacing from the apex than from the base. In addition, retrograde atrial activation sequence remains constant during pacing both at the RV apex and at the RV base because the atrium is activated over a single route (the AVN) in both settings.
A shorter VA interval during RV basal pacing than during RV apical pacing or a change in retrograde atrial activation sequence in response to differential RV pacing (RV base vs. RV apex) indicates the presence of an AV BT. However, differential-site RV pacing does not exclude the presence of a distant right or left free-wall BT or slowly conducting BT, whereby retrograde conduction occurs preferentially over the AVN. Also, the occurrence of right bundle branch block (RBBB) (but not left bundle branch block [LBBB]) also can alter the significance of the VA interval criterion ( see Fig. 20.6 ).
Para-Hisian Pacing During Sinus Rhythm
Para-Hisian pacing helps exclude the presence of a septal AV BT, which can mediate orthodromic AVRT with a retrograde atrial activation sequence similar to that during AVNRT. In the absence of a BT, para-Hisian pacing results in a shorter SA (or VA) interval when the HB-RB is captured (S − H = 0 and SA = HA) than the SA interval when only the ventricle is captured (SA = S − H + HA) with no change in the atrial activation sequence or HA interval. This response to para-Hisian pacing is termed pattern 1 or AVN/AVN pattern .
A change in the retrograde atrial activation sequence with loss of HB-RB capture indicates the presence of a retrogradely conducting BT. Similarly, an SA (VA) interval that is constant regardless of whether the HB RB is being captured indicates the presence of a BT, whereas prolongation of the SA (or VA) interval on loss of HB capture, compared with that during HB capture, excludes the presence of a retrogradely conducting BT, except for slowly conducting and far free-wall BTs. Please refer to Chapter 20 for a more detailed discussion of para-Hisian pacing.
Induction of Tachycardia
Initiation by Programmed Atrial Stimulation
Typical (slow-fast) AVNRT.
Clinical AVNRT almost always can be initiated with an AES that blocks anterogradely in the fast pathway, conducts down the slow pathway, and then conducts retrogradely up the fast pathway. Only when anterograde conduction down the slow pathway is slow enough (“critical AH interval”) to allow for recovery of the fast pathway to conduct retrogradely does reentry occur (see eFig. 17.1 ). This critical AH interval is not a fixed interval. It can change with changes in pacing drive CL, changes in autonomic tone, or after drug administration, thus reflecting changes in the fast pathway retrograde ERP.
There is a zone of AES coupling intervals (A1-A2) associated with AVNRT induction called the tachycardia zone. This zone usually begins at coupling intervals associated with marked prolongation of the AH intervals. This AVN conduction delay (AH interval prolongation), and not the AES coupling interval, is of prime importance for the genesis of AVNRT.
Atrial pacing can initiate AVNRT at PCLs associated with sufficient AVN conduction delay (see Fig. 17.11 ), especially during atypical Wenckebach periodicity, when anterograde block occurs in the fast pathway and conduction shifts to the slow pathway.
Rarely, AES or atrial pacing can produce a “1 : 2 response” with anterograde conduction over both the fast and slow pathways, as explained earlier (see Fig. 17.11 ). Such a response predicts easy induction of slow-fast AVNRT by ventricular stimulation because poor slow pathway retrograde conduction would increase the opportunity for the ventricular stimulus to block in the slow pathway and conduct up the fast pathway to return down the slow pathway and initiate AVNRT.
The site of atrial stimulation can affect the ease of inducibility of AVNRT, probably because of different atrial inputs to the AVN or different atrial FRPs. Therefore it is important to perform atrial stimulation from both the RA and CS.
AVN echo beats and AVNRT usually occur at the same time that dual pathways are revealed ( see Fig. 4.23 ). In 20% of patients, the dual AVN pathway AH interval jump occurs without concurrent occurrence of echo beats or AVNRT because of failure of retrograde conduction up the fast pathway. This failure can be caused by the absence of a distal connection between the two AVN pathways, a long retrograde ERP of the fast AVN pathway, or concealment of the AES anterogradely into the fast AVN pathway (i.e., the AES propagates some distance into the fast pathway before being blocked). The last event results in anterograde postdepolarization refractoriness, which would consequently make the fast pathway refractory to the wavefront invading it in the retrograde direction. The latter phenomenon can be diagnosed by demonstrating that the AH interval following the AES that fails to produce an echo beat is longer than the shortest ventricular PCL with 1 : 1 retrograde conduction. Such a PCL is a marker of the fast pathway retrograde ERP. This finding implies that an AES blocking in the fast pathway and conducting over the slow pathway, with an AH interval exceeding fast pathway ERP and still not conducting retrogradely over the fast pathway, is caused by anterograde concealment (and not just block) into the fast pathway.
Markers of poor retrograde conduction over the fast AVN pathway predict difficulty inducing AVNRT. These markers include the absence of VA conduction, poor VA conduction (manifest as retrograde AVN Wenckebach CL longer than 500 milliseconds), and retrograde dual pathways (indicative of long retrograde ERP of the fast pathway, which must exceed the refractoriness of the slow pathway for retrograde dual pathways to be demonstrable). In fact, retrograde fast AVN pathway characteristics (i.e., ERP) are the major determinant of whether reentry (AVN echoes or AVNRT) occurs, whereas conduction delay anterogradely over the slow pathway (i.e., “critical AH interval”) determines when reentry is to occur.
Although isolated AVN echoes can occur as long as VA conduction is present, the ability to initiate sustained AVNRT also requires the capability of the slow pathway to sustain repetitive anterograde conduction. In other words, sustenance of AVNRT requires that the TCL be longer than the ERP of all components of the circuit. Typically, for AVN reentry to occur, the fast pathway should be able to support 1 : 1 VA conduction at a ventricular PCL shorter than 400 milliseconds (i.e., retrograde Wenckebach CL shorter than 400 milliseconds), and the slow pathway should be able to support 1 : 1 AV conduction at an atrial PCL shorter than 350 milliseconds (i.e., anterograde Wenckebach CL shorter than 350 milliseconds). The shorter the AH interval during anterograde conduction over the fast pathway, the better the retrograde conduction over the same pathway (i.e., the shorter the HA interval), and the better the inducibility of AVNRT. Nevertheless, it is important to recognize that during EP testing, these criteria are dependent on the cardiac autonomic tone at that moment, and they can change dramatically by changing the level of patient sedation or the use of isoproterenol or by prolonged periods of rapid pacing (particularly ventricular) that cause hypotension and a reflex increase in adrenergic tone, which then affect inducibility of AVNRT.
Atypical AVNRT.
Anterograde dual AVN physiology is usually not demonstrable in patients with atypical AVNRT. In addition, as noted, the presence of a “1 : 2 response” to AES predicts noninducibility of atypical AVNRT because it indicates failure of the slow pathway to support retrograde conduction, a prerequisite for the atypical AVNRT circuit.
When atypical AVNRT is initiated with atrial stimulation, it is usually with modest prolongation of the AH interval over the fast pathway and anterograde block in the slow pathway, followed by retrograde slow conduction over the slow pathway ( Fig. 17.12 ). Therefore a critical AH interval delay is not obvious.
