Curative Catheter Ablation for Supraventricular Tachycardia: Techniques and Indications

Published on 21/06/2015 by admin

Filed under Cardiovascular

Last modified 22/04/2025

Print this page

rate 1 star rate 2 star rate 3 star rate 4 star rate 5 star
Your rating: none, Average: 2 (1 votes)

This article have been viewed 1565 times

Chapter 93 Curative Catheter Ablation for Supraventricular Tachycardia

Techniques and Indications

The current popularity of radiofrequency (RF) catheter ablation is, in large part, attributed to its contribution to the management of supraventricular tachycardias (SVTs). The electrophysiologist is able to ablate as well as analyze mechanisms, evaluate the results of ablation, and re-ablate, if necessary, all with minimal morbidity.

This chapter briefly reviews the basic principles of performing curative catheter ablation for SVTs. The arrhythmias considered below include arrhythmias involving accessory atrioventricular (AV) connections, AV nodal re-entry tachycardia (AVNRT), atrial tachycardia (AT), including typical flutter and other macro–re-entrant right and left AT (atypical flutters), and non–re-entrant AT.

Accessory Atrioventricular Connections

The anatomic substrate of accessory AV connections is the myocardium bridging the AV annuli, which, in normal individuals, are fibrous and electrically insulating (Box 93-1).1 The sequence of normal initial ventricular septal depolarization is altered by conduction through these connections inserting into the ordinary myocardium and bypassing the normal insulated and septally conducting His-Purkinje system. The relatively slow spread of activation through the ordinary myocardium contrasts with the coordinated septal endocardial breakthrough of Purkinje ramifications and results in the δ-wave in the surface electrocardiogram (ECG). In addition to providing an additional route for impulse conduction between the atria and the ventricles, nearly all accessory connections exhibit conduction properties different from the AV node. Decremental conduction is not ordinarily seen; that is, with increasing frequency or shortening coupling intervals, the conduction time across the pathway does not significantly increase.

Box 93-1 Checklist for Catheter Ablation of Accessory Atrioventricular Connections

Evaluation

ECG, Electrocardiogram; IVC, inferior vena cava; AV, atrioventricular; VA, ventriculoatrial; SVT, supraventricular tachycardia; AVNRT, atrioventricular nodal re-entrant tachycardia; AP, accessory pathway; RF, radiofrequency.

Ventricular Pre-excitation and Its Mechanisms

Pre-excitation is defined as ventricular myocardial activation by a pathway other than the His-Purkinje system during sinus rhythm or atrial pacing. The normal H-V interval includes the time required for activation to proceed from the bundle of His recording site down the bundle branches to the distal ramifications of the Purkinje fibers before exiting to depolarize the working myocardium. Therefore pre-excitation is inferred if the H-V interval is abnormally short during sinus rhythm or atrial pacing. The H-V interval may be normal if too little myocardium is pre-excited (consequently generating feeble voltage) to be evident on the surface ECG. With an increased frequency of supraventricular impulses, more of the ventricular myocardium is pre-excited through the accessory connection (with a progressive widening of the QRS and shortening of the H-V interval) because of decremental conduction through the AV node and nondecremental conduction through accessory AV connections. Incremental atrial pacing is an integral part of the evaluation of accessory AV connections and increases pre-excitation, thus allowing the optimal surface ECG localization of pathway insertion. Pre-excitation may be difficult to discern on the surface ECG in sinus rhythm in case of accessory connections with long anterograde conduction times or short conduction times through the AV node. Pre-excitation should, however, be detectable by rapid pacing or slowing conduction through the AV node.

An accessory pathway with a long antegrade conduction time can manifest with an isoelectric interval separating the end of the P wave from the onset of ventricular activation, which may persist even during atrial pacing or AT. An electrical connection between the AV node and the ventricular myocardium bypassing the His-Purkinje system has also been postulated to be responsible for such an ECG but has not been conclusively demonstrated. The so-called fasciculoventricular connections can also produce similar ECG manifestations with a high septal breakthrough from the normally insulated bundle of His or bundle branch being the anatomic correlate. Interestingly, no clinical arrhythmia correlate has been described for what may be no more than an anatomic variant.

Electrophysiological Characteristics of Accessory Pathways

Normal retrograde (ventriculoatrial [VA]) activation over the AV conduction system depolarizes the atria from the septal region and decrements by at least 20 ms with faster stimulation. Nondecremental free wall activation (the so-called eccentric activation) suggests VA conduction over an accessory AV connection. Dynamic maneuvers are required to distinguish between septally situated accessory pathways and normal routes of VA conduction.

During sinus rhythm, ventricular extrastimuli resulting in atrial activation preceding retrograde bundle of His activation indicate an accessory connection. If moving the ventricular pacing site from the apex toward the septum decreases the stimulus to atrial activation time instead of increasing it, an accessory VA connection should be considered. Moving away from the apex increases the conduction time to the normal AV conduction system through the distal Purkinje myocardial interface, whereas it decreases the conduction time to the annular insertion of an accessory pathway.2 Similarly, high output–dependent capture of the insulated right bundle or the bundle of His contrasted with lower output ventricular myocardial capture at the same site can show changes in atrial activation sequence, retrograde His to atrial activation time, and stimulus to atrial activation timing, which suggest the presence of more than one retrograde pathway of VA conduction.3 Unchanged atrial activation sequence coupled with a constant H-A interval and prolongation of the stimulus-A interval resulting from loss of His–right bundle capture indicate the presence of the normal VA conduction alone. Conversely, the absence of change in any of the intervals and sequences indicates the sole presence of accessory pathway retrograde conduction. If the accessory pathway is remote from the pacing site or is captured only with a long conduction time or if conduction through the AV node is very rapid, conduction through the accessory pathway may be completely masked. In practice, left free wall pathways remote from a right ventricular pacing site may fulfill these conditions and are therefore likely to be masked.

During a tachycardia, evidence of conduction through an accessory AV connection can be obtained by delivering late ventricular extrastimuli coincident with or 10 ms before activation of the bundle of His, thus ensuring the encountering of complete refractoriness within the bundle of His. If the extrastimulus is earlier than the His electrogram, the lack of anticipation of the ventricular electrogram, the bundle of His electrogram, or both confirms His-Purkinje refractoriness. The presence of conduction through an accessory connection is indicated if the ventricular extrastimulus advances or delays atrial activation or terminates the tachycardia without conduction to the atria.4 Tachycardia termination by a His-synchronous ventricular extrastimulus without conduction to the atria or with anticipation of the succeeding ventricular or bundle of His electrogram indicates participation of the accessory pathway in the tachycardia.

In addition to establishing the presence of an accessory connection, the electrophysiological study (EPS) allows assessment of the arrhythmogenic potential of the accessory connection. The indications for curative ablation of accessory pathways chiefly depends on their proven threat—pre-excited AF degenerating to VF—or their potential threat, indicated by R-R intervals shorter than 200 to 250 ms during AF or the presence of clinical tachycardias using the accessory pathway.5

Successful ablation of an accessory AV connection requires precise localization, and the surface ECG is a vital starting point. Although many algorithms have been described, those using the δ-wave vector are more difficult to use compared with the mean QRS vector during full or maximal pre-excitation. ECG pattern recognition allows the planning of a strategy specific to the presumed location.

Retrograde Transaortic Approach

In our laboratory, a retrograde arterial approach is preferred, whereas the trans-septal approach is used secondarily. Trans-septal access is the first-line approach in case of aortic or arterial abnormalities, such as the presence of prosthetic valves; aortic stenosis; or severe aortic, femoral, or iliac atherosclerosis. In pediatric patients, the trans-septal approach may be preferred to avoid injury to the aortic valve.

Entering the left ventricle in a retrograde fashion across the aortic valve is an important part of the retrograde arterial approach. When the catheter is brought down to the root of the aortic valve, it meets the resistance of the aortic valve, and catheter flexion combined with continued gentle pressure facilitates the formation of a loop. The loop generally crosses the aortic valve into the left ventricular cavity before a 180-degree flexion. This may be facilitated by gentle torquing. It is imperative to avoid the catheter tip entering a coronary artery, and the catheter should be promptly withdrawn in case of any doubt. Rare instances of complications resulting from an unrecognized position within the left coronary system have been reported. The catheter can also easily enter the right coronary artery ostium, particularly if it has a downward takeoff. Entry into the left ventricle occasionally produces mechanical trauma and block within the normal AV conduction axis, which does not become apparent until the accessory pathway conducting in an anterograde fashion has been ablated. Fortunately, spontaneous recovery of normal conduction is the most common outcome. After crossing the aortic valve, the catheter should be straightened before it is gently advanced toward the posterolateral left ventricular free wall. Progressive flexion of the catheter tip as it touches the free wall brings the catheter tip near or at the level of the mitral annulus and under the mitral valve leaflet, as indicated by the recording of a significant atrial electrogram. The posterior and lateral mitral annulus should be mapped at this level.

Although catheter stability is the strong suit of the retrograde left-sided approach, this same characteristic renders mapping the mitral annulus difficult. Moving from one position to another requires catheter withdrawal from under the leaflet and repositioning it anew. Clockwise rotation positions the catheter tip more laterally and anteriorly, whereas counterclockwise rotation brings the tip around more medially. The size of the catheter curve is important; a large curve does not allow the catheter tip to reach the annulus level, and a small curve means that the catheter tip “floats” or bounces without stable contact. A more atrial position (where the catheter tip makes contact with the atrial side of the mitral leaflet) can be achieved by torquing the catheter counterclockwise so that it slips medially onto the atrial side through the posterior commissure. The further anterior the accessory pathway, the more difficult it is to reach the atrioventricular annulus with the catheter tip from the retrograde approach. This situation may call for a larger curve or a trans-septal approach. The catheter tip can be much more freely moved to map the annulus on the atrial side of the mitral leaflet but, typically, is less stable than when positioned under the leaflet. Ectopy, not uncommon during RF delivery in this position, can easily dislodge the catheter.

Electrophysiological Localization

A multi-catheter approach can cover both AV annuli and provide corroboration of localization rapidly. Successful and equally rapid ablation can, however, be achieved with fewer catheters—typically two or three. In the case of evident pre-excitation, a single ablation catheter may be successfully used, which may be followed by an adenosine test; but the assessment of retrograde VA conduction usually requires an additional intracardiac catheter.

When even the best unipolar and bipolar endocardial electrograms are not good enough, an epicardial or intramyocardial pathway insertion may need to be evaluated or considered. Ventricular electrograms close to or at the site of insertion can be late, not only because the insertion may be far from the endocardium but also because of the endocardial insertion of an oblique pathway. Changing the pacing site (e.g., from the right ventricular apex to the lateral left ventricular or the right ventricular infundibulum) can help distinguish apparently early atrial electrograms (during ventricular pacing) because of an oblique pathway course. Simultaneous comparison of endocardial and epicardial recordings obtained from within the coronary sinus is useful; bracketing, as well as electrogram timing and dv/dt (rate of ventricular electrogram depolarization) comparison, can provide valuable clues. Ablation within the coronary sinus may be necessary (Figure 93-1), although conventional RF delivery achieves only low power and is frequently ineffective. Ablation in the coronary sinus and veins with a catheter with an open irrigated tip can achieve good results; however, stepwise increments in RF power (a cautious maximum of 25 W) are prudent. Pops in the thin-walled coronary venous structure can be devastating; damage to adjacent coronary arteries has also been reported.

Localizing a pathway conducting in an antegrade fashion involves sampling the annulus of interest for the shortest local AV intervals and the earliest V (local ventricular electrogram)-δ intervals. Some posteroseptal pathways exhibit long AV times at successful sites, which suggests slow conduction through the accessory pathway. The correct assessment of the timing of ablation catheter electrograms requires comparison with the surface ECG lead showing the maximum pre-excitation.

Local Electrogram Characteristics

Bipolar and unipolar electrograms should both be used for mapping (Figure 93-2)—the former because of their higher signal/noise ratio and the latter because of their simple morphologic pattern recognition–based analysis.6 Localization based on bipolar electrograms requires distinction of atrial electrograms from ventricular electrograms by using late-coupled ventricular and atrial extrastimuli. However, these maneuvers can be difficult to perform or analyze and may even induce arrhythmias. The contribution of the proximal ring electrode to bipolar electrograms from the distal bipole can be misleading. Atrial electrograms can be distinguished from ventricular electrograms by using unipolar electrograms from the distal electrode.

Unipolar electrograms with a steep QS morphology are particularly useful for localizing the site of ventricular insertion on the basis of a steep QS morphology (the absence of an initial R wave) during sinus rhythm, pacing, or even ongoing AF and also in patients with Ebstein’s anomaly who exhibit low-amplitude, fractionated bipolar electrograms on the tricuspid annulus. Unipolar electrograms should be recorded with wide band filters—0.05 to 500 Hz—because the low-frequency content makes important contributions to the generation of RS or QS patterns. Instead of Wilson’s central terminal, a remote cutaneous or inferior vena cava (IVC) electrode may be useful as a ground, allowing common mode rejection of contaminating 50- or 60-Hz line noise. Notch filters should also be used with caution, if at all. Once the atrial and ventricular electrograms have been recognized, the intervening deflections represent presumptive accessory pathway potentials (see Figure 93-1).7 Certainly, the best validation is the prompt abolition of accessory pathway conduction by RF ablation at this site (assuming appropriate power delivery and contact). In practice, accessory pathway potential validation often is a retrospective exercise.

For patients without pre-excitation, the target of choice is the shortest VA interval during orthodromic AVNRT because this effectively rules out the fusion of activation through the normal AV axis with activation through the accessory connection. Ablation during ongoing AVNRT can, however, lead to dislodgment of the ablation catheter at tachycardia termination. Prompt initiation of ventricular pacing after tachycardia termination has been advocated to minimize or prevent instability. If no tachycardia is inducible and retrograde conduction through the normal AV axis can be excluded or distinguished, earliest atrial activation during ventricular pacing is a reasonable target. In the presence of an obliquely coursing accessory pathway, changing the ventricular pacing site is useful in evaluating electrogram timing as pointed out above.

Individual Pathway Locations

Right Free Wall Atrioventricular Accessory Connections

Right free wall pathways are defined by a location within the arc of the tricuspid annulus extending from approximately the 12 o’clock position to the 6 o’clock position, as viewed from the left anterior oblique 45-degree view. The tricuspid annulus has a much more vertical orientation compared with the mitral annulus, the right ventricle is much thinner than the left ventricle, and there is no counterpart of the coronary sinus. Access to the right AV annulus is much more direct than on the left side, but a position on the annulus is more difficult to achieve, particularly from the femoral route. Right atrial free wall contraction tends to dislodge the catheter tip and long sheaths or large curve catheters facilitate ablation by improving stability. The radiolucent shadow of annular fat is a useful clue to the level of the tricuspid annulus, particularly when annular electrograms are fractionated and of low amplitude, as in Ebstein’s anomaly. Unipolar electrograms can be of help, and recordings from within the right coronary artery have been used, although the latter run the risk of significant complications. Unlike the usual QS morphology at successful ablation sites of other pathway locations, the ventricular electrogram has a two-stepped deflection; the first steeper deflection indicates local right ventricular activation, and the second represents far-field activation probably originating from the septum and the left ventricle.

The typical successful ablation site for right free wall pathways shows ventricular electrograms with timings approximately 20 to 30 ms earlier than for left free wall pathways. The atrial insertion of the pathway is activated before the end of atrial activation so that right ventricular pre-excitation actually begins within the P wave, and pre-excitation of the thin-walled right ventricle does not become evident on the surface ECG as early as for pathways inserting into the thicker left ventricle. Local ventricular activation of –10 to –20 ms (preceding the QRS) is therefore usually not “early enough” for pathways in this location. Ablation at this location is characterized by the lower electrode temperatures during RF delivery—related to catheter contact, electrode orientation, and stability and high-power delivery during temperature-controlled RF applications. Mechanical block of accessory pathway conduction may be frequent, and an eventual recurrence is more likely for pathways at this location despite the “security” RF applications.

Septal Atrioventricular Accessory Connections

Septally situated pathways have been divided into anteroseptal, mid-septal, and posteroseptal pathways. The anteroseptal pathways are located on the tricuspid valve (TV) annulus (because of the aortomitral fibrous continuity on the left side) between the 12 o’clock position (in the left anterior oblique view) and the bundle of His region—with the provisio that any His potential recorded at the site be less than 0.1 mV. A larger His deflection defines the pathway as para-Hisian. Posteroseptal pathways are defined by a location between the coronary sinus ostium to approximately the 6 o’clock position on either AV annulus. The mid-septal pathways are more difficult to define in terms of location and are broadly considered to be between the His and coronary sinus ostial locations, excluding the para-Hisian pathways. Except for the anteroseptal pathways, the septal pathways have in common the possibility of being accessible from either annulus and proximity to the AV node or the bundle of His. An epicardial location is more frequent for the posteroseptal pathways; therefore mapping of the coronary sinus and the middle cardiac vein is often necessary.

The anteroseptal pathways can be ablated with advantage from the superior vena cava (SVC) approach, with active catheter flexion bringing the catheter tip in contact with the annulus; with the femoral approach, relaxing the catheter tip flexion is required to achieve contact, and unless the catheter is bi-steerable, this is a passive movement providing much less stable contact. It may also be easier to achieve a position under the tricuspid valve by making a loop in the right ventricle using an approach from above.

The main concern with regard to the mid-septal pathways is to avoid damage to the AV node and the normal conduction axis. As for the para-Hisian pathways, proximity to the bundle of His allows an estimation of this risk. Proximity to the compact AV node is, however, difficult to estimate in the absence of an electrogram marker. The appearance of junctional rhythm is a clear warning that should prompt cessation of RF delivery, and a narrow QRS complex without a preceding P wave should not be mistaken for loss of pre-excitation. Using conventional RF, the strategy for pathways estimated to be close to the AV node or the bundle of His should center around careful mapping for the best electrograms and delivering low RF power at sites thought to be farthest from the conduction axis, usually on the ventricular side of the AV ring. At prospective ablation sites, it is useful to verify the presence and the amplitude of a bundle of His deflection concealed by pre-excitation by using programmed stimulation to induce antegrade pathway block or sustained orthodromic AVNRT. RF power may be increased cautiously in steps of 5 W, but energy delivery should be terminated immediately in case of junctional rhythm or if loss of pre-excitation does not occur promptly. Cryoablation offers the theoretical advantage of reversible cryomapping. In practice, although a greater margin of reversible lesion creation with cryoablation and therefore a lower risk of AV block may exist, this energy source has a clearly higher risk of recovery of pathway conduction.8

The posteroseptal pathways have a higher likelihood of an epicardial course or insertion. Moreover, the anatomic boundaries of the posterior pyramidal space frequently require a choice to be made between the right or left endocardial sites and the sites within the proximal coronary sinus or the middle cardiac vein. A steep QS complex in lead II or an rS complex in leads V5-V6 during pre-excitation may be a clue to an insertion into the middle cardiac vein.9 If endocardial mapping is not good enough or the ablation is unsuccessful, mapping within the coronary sinus and the middle cardiac vein is performed under the guidance of a coronary sinus angiogram. Occlusion balloon angiography provides the best opacification of the great cardiac vein and related branches, but adequate visualization of the proximal coronary sinus and the middle cardiac vein can be achieved from the femoral approach by using an Amplatz catheter. Successful ablation sites are frequently clustered in proximity to venous anomalies such as aneurysms or diverticula. A superior approach from the internal jugular vein provides a relatively straight and vertical catheter course to the middle cardiac vein and should be considered in case of difficulty with the femoral approach. Ablation within the coronary sinus or cardiac veins with a conventional nonirrigated ablation catheter is frequently ineffective because of low delivered powers and high electrode temperatures as a consequence of limited blood flow around the electrode. Ablation in the middle cardiac vein can damage the posterior descending and posterior left ventricular branches of the distal right coronary artery. Ineffective low power delivery can be overcome by using an irrigated tip catheter that allows power to be titrated up to a limit of 25 to 30 W to avoid pops or damage to nearby coronary arteries (within 2 to 3 mm of the site of ablation).

Specific Situations

During an ablation procedure, sustained AF is not infrequent, which renders mapping difficult. No alternative to electrical cardioversion may be available because type I or III antiarrhythmic drugs may alter the accessory pathway properties and may even eliminate pre-excitation. Mapping the earliest ventricular activation during the widest QRS (indicating maximum pre-excitation) is feasible, as is ablation, particularly when guided by unipolar electrograms. Verification of bi-directional conduction block (assessment of VA conduction) is not possible during AF.

Multiple pathways are not common but may be encountered, particularly in association with Ebstein’s anomaly. Changing patterns of pre-excitation and VA intervals and sequences are important clues. However, the same principles of mapping and ablation described above are usually effective.

The substrates of the “Mahaim” pathways (decremental atriofascicular or atrioventricular pathways) and permanent junctional reciprocating tachycardias are both thought to be accessory pathways with long conduction times—antegrade in case of the atriofascicular or AV Mahaim pathways and retrograde in case of persistent junctional reciprocating tachycardia (PJRT). In addition, these two accessory pathway variants share another characteristic—that of one-way conduction only. The few available histologic studies suggest that the anatomic substrate of PJRT is a long and tortuous muscular fascicle, whereas in the case of the Mahaim fiber, an accessory node–like structure is thought to exist at its atrial origin.10 Atriofascicular and AV Mahaim fibers are most effectively ablated by targeting the pathway potentials at the level of the annulus; they resemble the bundle of His potentials but continue to precede ventricular activation even during pre-excitation. PJRT is ablated by targeting the earliest atrial activation during the tachycardia, and, as with every ablation within that posteroseptal area, care must be taken to ensure a reasonable distance from the normal AV conduction axis.

An accessory pathway with a large insertion or an insertion with multiple branches may occasionally be encountered.11 Multiple coalescent lesions, each of which modifies local electrogram parameters, have been used.

Another uncommon variant is an appendage to the ventricular connection characterized by an insertion bridging the appendage tip to the ventricle away from the annulus.12 Careful mapping, aided by three-dimensional mapping as needed, can clarify the exact location of the insertion. Similarly, the unusual variant of surgically acquired pre-excitation is encountered rarely after right atrial appendage anastomosis to the right ventricular outflow tract (RVOT) (historically performed as a palliative procedure for tricuspid atresia). In the appropriate surgical context and with the pre-excited QRS resembling an RVOT tachycardia, careful mapping has allowed successful ablation.

An additional arrhythmia substrate such as AVNRT or AT may coexist. The electrophysiological maneuvers described above can assist in deciding whether the accessory pathway participates in the tachycardia.13 However, in practice, elimination of the accessory pathway substrate typically unmasks the AVNRT or AT, which can then be ablated in the standard fashion.

Indications for catheter ablation include the following:

Atrioventricular Nodal Re-entrant Tachycardias

AVNRT is the result of a re-entry circuit in the AV junctional region (Box 93-2), although debate about anatomic delimitations continues. The functional heterogeneity of AV junctional tissues, primarily with respect to conduction velocity and refractory periods, permits the sustenance of an excitable gap re-entry circuit. Because of anatomic factors and the lack of distinct electrophysiological markers of activation, it is difficult to delineate the anatomic extent of the circuit. Nevertheless, available evidence suggests that the peri-nodal atrium, the compact AV node, and possibly a part of the proximal bundle of His are involved. Although no significant anatomic abnormalities have been found in patients with AVNRT, multiple posteriorly situated pathways or approaches to the AV node have been described.14

Box 93-2 Checklist for Catheter Ablation of AV Nodal Re-entrant Tachycardia

Evaluation

AP, Accessory pathway; AV, atrioventricular; VA, ventriculoatrial; CS, coronary sinus; RF, radiofrequency; AVNRT, atrioventricular nodal re-entrant tachycardia.

At least three different types of AVNRT have been described: (1) slow antegrade/fast retrograde, (2) fast antegrade/slow retrograde, and (3) slow antegrade/slow retrograde. In addition to differences in conduction velocity and refractory periods, the fast and slow pathways manifest relatively disparate anterior and posterior retrograde exit sites, which helps delineate the different types of AVNRT. Most laboratories today establish baseline evidence of antegrade “slow” pathway conduction in the form of a long A-H interval (>200 ms) with or without a discontinuity (50-ms increments in A-H for a 10-ms decrement in coupling interval). In addition, an H-A interval ranging from 25 to 80 ms during tachycardia is indicative of a typical slow-fast AVNRT. Variants of AVNRT (fast-slow or slow-slow) can be distinguished from AVNRT and AT by the response to ventricular extrastimuli introduced during the AV nodal refractory period, by rapid ventricular stimulation, and by the atrial activation sequence and tachycardia behavior during AV block.13

Earliest retrograde activation at the anterosuperior tricuspid annulus during typical AVNRT localizes the fast pathway exit site, and retrograde activation near the posteromedial tricuspid annulus during fast-slow AVNRT localizes the slow pathway exit site. Techniques of AV nodal modification have therefore targeted these sites to produce selective fast or slow pathway ablation.

The fast pathway exit can be approached by slow withdrawal of the catheter a few millimeters from the bundle of His position while concurrently applying clockwise torque to maintain good contact. A monitoring catheter kept in the bundle of His–recording position is a convenient reference. Because no accepted electrogram markers of fast pathway activation exist, indirect parameters such as an A/V electrogram amplitude ratio greater than 1 and a His deflection less than 0.05 mV are used to ensure relative separation from the bundle of His. RF energy applied for a short period at such sites results in P-R interval prolongation and elimination or marked attenuation of retrograde VA conduction. A junctional tachycardia is frequently noted; this may require atrial pacing to allow monitoring of AV conduction. P-R interval prolongation by more than 50% or AV block make prompt discontinuation of RF energy delivery (which should be titrated in steps of 5 W) mandatory.

Evaluation after fast pathway ablation typically reveals abolition or marked attenuation of VA conduction accompanied by an increase in the A-H interval and elimination of dual AV nodal physiology. VA conduction is eliminated in more than one third of patients, whereas the VA block cycle length is increased in the remainder. Similarly, the A-H interval is prolonged markedly (<50%) but without significant change in the H-V interval, AV nodal effective refractory period (ERP), or anterograde Wenckebach cycle length.

Although fast pathway modification is successful in more than 90% of patients, complete heart block occurs in up to 21%.15 In case of transient conduction block, in-patient telemetric monitoring may be advisable for 1 to 2 days after ablation to watch for delayed complete heart block. The high incidence of AV block and the efficacy of slow pathway ablation have led to this technique being abandoned.

The slow pathway has been targeted by selective ablation, and because of the greater distance from the compact AV node, the incidence of complete heart block is consistently lower. Slow pathway ablation or modification has become the therapeutic procedure of choice.

Two approaches have been used: anatomic and electrophysiological. The anatomic approach uses fluoroscopic landmarks to guide the positioning of the ablation catheter, with the target area being the junction of the middle and posterior thirds of the medial inter-atrial septum at the level of the tricuspid annulus or in the vicinity of the ostium of the coronary sinus. Arrhythmia re-induction is attempted after each RF delivery (for 30 to 60 seconds) followed by slow pathway assessment. If AVNRT remains inducible, subsequent RF energy is delivered nearer the AV node; however, the most posterior (inferior) successful ablation site is the safest.

One electrophysiological approach targets the earliest atrial activation during retrograde slow pathway conduction during fast-slow AVNRT or ventricular pacing but is limited by the difficulty in inducing consistent retrograde slow pathway conduction. This is possible in only approximately 10% of patients. More commonly, characteristic electrograms representative of slow pathway conduction have been used to guide RF energy application. Two distinct types of slow pathway potentials have been described. One is a sharp, spikelike potential (Asp) preceded by a lower frequency, lower amplitude potential (A) during sinus rhythm.16 Asp usually follows A by 10 to 40 ms; such double potentials are recorded in the vicinity of the coronary sinus ostium near the tricuspid annulus. During retrograde conduction over the slow pathway, the sequence of these double potentials is reversed, that is, Asp now precedes A. RF energy applied to the latest Asp potential close to the tricuspid annulus successfully eliminated AVNRT in 99% of patients (with only one AV block). Experimental data have shown that similar double potentials are produced by asynchronous activation of muscle bands flanking the mouth of the coronary sinus.17

Other types of the slow pathway potentials are characteristically low-amplitude, low-frequency signals that are concealed within or follow the atrial electrogram and occupy some or all of the A-V interval in sinus rhythm.18 They are easily found by withdrawing the catheter posteriorly from the bundle of His position. In the posterior septum, they are usually hump shaped, whereas more anteriorly they are rapid, narrower, often biphasic, and with a superimposed bundle of His deflection. They are typically recorded at the junction of the anterior two thirds and the posterior one third of the area between the bundle of His and the coronary sinus ostium. During incremental atrial pacing, these slow potentials characteristically separate from the atrial electrograms, are prolonged in duration, and decline in amplitude. They fractionate and disappear at rapid pacing rates so they are not discernible during tachycardia. Animal studies indicate that the low-frequency deflections coincide with the activation of the cells around the tricuspid annulus possessing AV node–like properties. During reverse ventricular echoes, these cells are activated before the earliest atrial activation during retrograde slow pathway conduction but fail to be activated during antegrade conduction over the fast pathway. In view of their wide recording area, they may represent dead-end pathway activation overlying the actively participating cells. Activation of the posterior part of the slow atrio-nodal approaches may give rise to the Asp potential, and the (transitional) tissue anteriorly (beyond the coronary sinus ostium) gives rise to low-frequency slow potentials.

After successful ablation of the slow pathway, an increase in the antegrade AV block cycle length and the AV nodal ERP is usually noted without a change in baseline A-H intervals or retrograde conduction. The maximum A-H interval during incremental atrial pacing is characteristically curtailed. However, in approximately 50% of patients, residual slow pathway conduction, in the form of persistent antegrade AV nodal duality, single AV nodal echoes, or both, is evident, although AVNRT typically remains noninducible even with intravenous infusion of isoproterenol.

RF ablation guided by either approach offers essentially equivalent results, although fewer applications may be required when ablation is guided by low-frequency potentials. Complete AV block during slow pathway ablation is definitely uncommon but may be related to an abnormally posteriorly situated fast pathway.

To avoid AV block, a careful search should be made for the most posterior site with typical slow potentials, without His deflections, of course, but also avoiding sites that exhibit slow potentials that persist and coincide with the end of the A-H interval during rapid pacing. Junctional ectopy is elicited at 70% to 90% of effective sites, and it is important to monitor VA conduction during this rhythm. VA block—even intermittent—and faster rates of junctional rhythm are useful markers of impending AV block. In case of doubt, it may be wise to stop RF delivery to check the P-R interval during sinus rhythm. Atrial pacing may be helpful, although the flip side is that it may mask junctional rhythm and prevent monitoring of VA conduction, which is an early sign of encroachment on antegrade conduction. In the event of AV block, early recovery (within 2 to 3 minutes) indicates a good prognosis.

Recurrence after successful ablation is uncommon (approximately 2%) and may be lower if complete elimination of the slow pathway is achieved. However, tachycardia noninducibility in spite of isoproterenol infusion is an adequate endpoint.

The lower incidence of AV block (approximately 1%) has made slow pathway ablation the technique of choice, although fast pathway ablation may be considered if the slow pathway approach is ineffective. Cautious ablation of the slow pathway is usually effective even in patients with prolonged P-R intervals at baseline because of the persisting so-called “intermediate” pathways that permit AV conduction. Cryoablation is theoretically attractive but has not been shown to be superior and, in fact, has a high incidence of recurrence.19

Symptomatic patients who do not wish to have drug therapy or cannot tolerate standard drug therapy may be offered this intervention. The ablation should probably not be performed for initial or infrequent episodes of AVNRT because of the small but definite risk of AV block. This risk is thought to be higher in small children.

Indications for catheter ablation include the following:

Atrial Flutter

Negative sawtooth flutter waves in leads II, III, and aVF between 200 and 350 beats/min characterize typical atrial flutter (Box 93-3). The absence of a diastolic isoelectric baseline distinguishes it from other SVTs. A typical flutter includes the counterclockwise as well as the clockwise form of the cavotricuspid isthmus–dependent flutter. Beyond this characterization, macro–re-entrant ATs are often considered to be forms of atrial flutter, although they are termed atypical to distinguish them from cavotricuspid isthmus–dependent (typical) flutter.

Typical Atrial Flutter

The key to the widespread use of catheter ablation for the treatment of typical atrial flutter has been the realization that the anatomically limited and easily accessible cavotricuspid isthmus is critical for maintenance of the arrhythmia. The understanding that stable isthmus conduction block is a clearly demonstrable endpoint in sinus rhythm and the absence of significant adverse effects have contributed to the popularity of this treatment.

Modern mapping techniques have merely confirmed the data derived more than 20 years ago that the macro–re-entrant circuit of typical atrial flutter is confined to the right atrium, resulting in counterclockwise or clockwise activation when viewed in the left anterior oblique perspective, with the tricuspid valve en face. In analogy with Mines’ ring models, the tricuspid valve is the outer boundary, and the posterior intercaval right atrium–crista terminalis complex is the inner boundary of a ring of re-entrant activation.20

The surface ECG morphology of counterclockwise typical flutter is remarkably consistent, allowing effective ablation even without proof of the participation of the cavotricuspid isthmus. In contrast, the surface ECG morphology of clockwise flutter is variable and difficult to distinguish from non-isthmus-dependent flutters. Intracardiac activation and entrainment mapping often are necessary for confirming the diagnosis. Because the cavotricuspid isthmus is a critical segment of the circuit of both clockwise and counterclockwise forms, the strategy of cavotricuspid isthmus ablation is effective for both forms.

Ablation of Typical Atrial Flutter

The aim of catheter ablation for typical atrial flutter is to create complete and stable bi-directional cavotricuspid isthmus block because terminating or interrupting the flutter is not enough as an endpoint and because recurrences are frequent if isthmus conduction persists. The ablation procedure involves (1) creation of linear lesion, (2) elimination of residual conducting gaps, and (3) verification of isthmus conduction block.

Creation of Linear Lesion

We use a 4-mm ablation catheter with an open irrigated tip for temperature-controlled, sequential, point-by-point RF applications at the isthmus between the IVC and the tricuspid annulus, although ablation catheters with an 8-mm tip have been shown to provide good results as well. A series of linear and contiguous point lesions need to be created to expeditiously achieve complete conduction block. The lesions may be delivered under fluoroscopic guidance, alone or supplemented by nonfluoroscopic navigation (e.g., the Biosense system), with or without the use of long sheaths for superior stability. During counterclockwise flutter, RF may be delivered point by point from the TV annulus on electrograms within the isthmus region coinciding with the center of the surface ECG flutter wave plateau, all the way from the TV annulus to the IVC edge. This ensures a lesion perpendicular to the advancing wavefront and allows catheter displacement to either side to be recognized by the altered timing of the recorded electrogram; for example, in case of lateral displacement, the electrogram coincides with the beginning of the surface ECG plateau, and in case of medial displacement, it coincides with the end of the plateau. During low lateral right atrial pacing in sinus rhythm, sequential RF may be similarly delivered at the 6 o’clock position in the left anterior oblique view in the isthmus on electrograms with a constant stimulus electrogram time from the TV annulus to the IVC edge.21

Merely delivering RF energy at a given location does not guarantee a transmural lesion; the lesion-making ability of RF varies according to contact, local blood flow, delivered power, and myocardial thickness. During unidirectional activation in the atrial myocardium (e.g., in the isthmus during typical flutter or during pacing from the low lateral right atrium or from the ostium of the coronary sinus), a local transmural RF lesion of significant size (comparable with the distal bipole) often can be recognized by double potentials separated by an isoelectric interval; the second potential is produced by activation detouring around the lesion.22

The created lesion size depends on tissue heating, which, in turn, is determined by the current density over a given area. RF power, target temperature, or both may need to be manipulated to achieve transmural lesions at each delivery site. Conventional temperature-controlled RF delivery is subject to variations in local convective cooling, which limits the delivered power either by achievement of the target temperature, by coagulum formation and impedance rise, or by both. Irrigating the ablation electrode substantially reduces the electrode temperature and coagulum formation, thus permitting the delivery of desired (and relatively higher) mean RF power, irrespective of variations in convective cooling. This results in consistent electrogram changes, that is, splitting into double potentials. The use of irrigated-tip catheters with a relatively limited power ceiling (40 W) has been shown to be clinically effective and safe for typical flutter cases resistant to conventional catheter ablation and as a first-line strategy. A significant reduction in procedure and fluoroscopy durations is achieved using irrigated-tip catheters compared with conventional 4-mm tip catheters.23

Identification and Ablation of Residual Gaps

Because of variations in isthmus anatomy and the inability to create consistent continuous transmural lesions, isthmus conduction frequently persists despite anatomically sufficient ablation, flutter termination during RF delivery, or both. Locating and ablating residual gaps in the ablation line is therefore necessary. During typical atrial flutter, such residual gaps can be identified with local electrograms, with a single or a fractionated potential centered on or spanning the isoelectric interval of adjacent double potentials (Figure 93-3). This allows targeted ablation of flutter that recurs after previous ablation. The same approach has been used during pacing from either side of the isthmus, targeting single or fractionated potentials adjacent to double potentials and centered on their isoelectric intervals, with the aim of establishing a continuous corridor of double potentials with isoelectric intervals across the full width of the isthmus.21

Assessment of Isthmus Conduction

Termination of flutter during RF delivery is not a sufficient endpoint because in more than 50% of instances, conduction through the cavotricuspid isthmus persists, resulting in frequent recurrence.24 Transient block or conduction slowing within the isthmus (or ectopics) may be enough to terminate flutter without eliminating the substrate. During pacing from one side of the ablation lesion, a delay in activation on the opposite side and an activation sequence demonstrating a 180-degree change in direction of activation on the other side have been used to diagnose isthmus block. This can be documented sequentially by using rove mapping and simultaneously with a duodecapolar electrode catheter during low lateral atrial or coronary sinus pacing. Local electrogram–based criteria (double potentials) have been shown to be highly sensitive markers of block in the cavotricuspid isthmus.21

A high recurrence rate was observed when flutter termination and noninducibility were considered sufficient endpoints, but the demonstration of isthmus block with local electrogram-based criteria—mapping double potentials supplemented with differential pacing—has reduced recurrence rates to less than 5%.

The sensitivity of double potential–based local electrogram criteria derives from being assessed on the linear lesion, and their specificity relates to relative independence from catheter positioning as well as from posterior intercaval conduction. The presence of posterior intercaval conduction may shorten the timing to the second component of double potentials but cannot produce false-positive gap electrograms on the ablation line. The activation detour suggestive of isthmus block can be demonstrated by standard multi-electrode multi-catheter techniques, basket catheters, electroanatomic sequential mapping, and noncontact mapping.

The choice of pacing site is instrumental in maximizing the sensitivity and specificity for complete isthmus block and must be as close as possible to the line of block to maximize the sensitivity for detecting slow conduction through the line and avoid it being concealed by the wavefront going around the lesion with a relatively shorter conduction time. This can be easily assessed by the stimulus to the first potential time (st-DP1) on the line; an st-DP1 of 30 ms or less is considered optimal.25 Isthmus block that produces a change in the surface ECG P wave during pacing is greater when pacing from the low lateral right atrium compared with pacing from the coronary sinus ostium. Left atrial activation from the coronary sinus input acts as a surface ECG amplifier of P-wave change during pacing from the low lateral right atrium.

Differential pacing assesses the response of local electrograms (including double and fractionated potentials) to advancing toward or withdrawing the pacing site away from the ablation line.25 If activation on both sides of the line (indicated by local double potentials) is directly linked by a conducting gap, withdrawing the pacing site will increase the activation time to both sides and by roughly the same magnitude. However, if no conducting gap across the line exists, withdrawing the pacing site will certainly increase the activation time upstream of the line. It will either shorten the activation time on the other side of the line or leave it unchanged, but it will not increase it.

The demonstration of functional linking by changing the pacing sites depends on the relative conduction times to both flanks of the ablation line and therefore will be affected by the pacing position, relative conduction velocities, length of the activation detour, and intervening areas of slow conduction or block that affect only one of the two pacing positions. The pacing catheter should be positioned as close as possible to the lesion line and the magnitude of displacement of the pacing position limited (15 mm) so that the stimulus to the first potential time is approximately 40 ms during distal pacing and 60 ms during proximal pacing. To detect very slow conduction through the isthmus (e.g., ≤0.05 m/s), both pacing sites may need to be even closer to the ablation line, that is, with shorter stimulus to the first potential times. Very slow conduction through the isthmus, however, cannot be ruled out by any technique or criteria.

Differential pacing is best used as a complement to local electrogram assessment to evaluate double or triple potentials or fractionated potentials without having to move the recording ablation catheter from the recording site or perform supplemental mapping. A gap-like electrogram, if validated to represent persistent conduction through the ablation lesion (by differential pacing), requires prompt ablation. Because very slow conduction ultimately cannot be excluded, the gold standard for complete (and stable) isthmus conduction block is only the absence of recurrence of typical atrial flutter.

Stable conduction block is necessary to avoid recurrence. The probability of conduction recovery across a composite lesion can be estimated by the number of constituent lesions (mean of 6 to 10 point lesions) multiplied by the individual probability of recovery. If the latter is estimated to be 2% (based on data from WPW ablation, the so-called point ablation), this works out to 12% to 20%. Recent data have indicated high rates of conduction recovery after the achievement of complete block (as also after the termination of flutter by RF delivery in the cavotricuspid isthmus). This mandates monitoring of the stability of isthmus conduction block after ablation. The exponential reduction in the incidence of recovery with time suggests an empirical cutoff for the duration of the monitoring period. An extended duration of monitoring (15 to 20 minutes at least) is compatible with the current low recurrence rates observed in our laboratory.24

The ablation of typical flutter, as described above, is very well tolerated, and few adverse effects have been reported. As a result, indications have been expanded so that RF catheter ablation is being offered to all patients with symptomatic and atrial flutter refractory to at least one drug. It may even legitimately be considered an alternative first-line therapy.

Indications for catheter ablation include the following:

Atypical Atrial Flutter

Macro–re-entrant ATs other than cavotricuspid isthmus–dependent atrial flutter are termed atypical atrial flutter (Box 93-4).

Barriers in the Atria

Some form of fixed or functional central barrier is a prerequisite for re-entry. The wavefront of typical atrial flutter circulates around the posterior intercaval crista terminalis complex. Other naturally occurring barriers in the right atrium include the IVC and the SVC. Acquired barriers in the right atrium include surgical incisions or patches, as well as “mute” regions devoid of electrical activity of uncertain origin.26,27 The role of functional activation inhomogeneities in the right atrium is unclear, although the crista terminalis region has been shown to permit conduction across it during sinus rhythm (as well as during certain forms of re-entry). Each of the above barriers or inhomogeneities—whether functional or fixed—can potentially support a re-entry circuit around it, provided that an appropriate trigger is present and the conditions of sufficient conduction times (wavelength) around it are met.

The left atrium has little evidence of a naturally present zone of block or slow conduction and certainly lacks the anatomic equivalent of a crista terminalis. Perhaps as a result, left atrial macro–re-entry in the absence of structural heart disease, surgery, or catheter ablation is far less common than typical atrial flutter. Left atrial flutter occurs in predominantly two situations: (1) as a sequel to ablation in the left atrium and (2) in subjects with left-sided structural heart disease.26 Characteristically, most patients have more than one ECG arrhythmia morphology (hence circuit) of re-entry, correlating with the frequent demonstration of multiple loop re-entry. Anatomic obstacles such as the pulmonary veins or the mitral valve, buttressed centrally or laterally by linear ablation lesions, form the core of the circuit. Although these macro–re-entrant circuits are large in diameter (approximately 5 cm), other small (1- to 1.5-cm diameter) re-entrant circuits have been observed around a pulmonary vein ostium or an incomplete lesion at the pulmonary vein ostium.28 In patients without prior left atrial ablation, macro–re-entrant circuits frequently are anchored around relatively large electrically silent areas, in addition to anatomic obstacles such as the pulmonary veins and the mitral valve. Although the exact nature of these “mute zones” remains to be determined, infarction or a myocarditic inflammatory scar have been evoked. Coronary angiography has, however, failed to show an appropriately located coronary occlusion (or lesion), and in the absence of other evidence of myocarditis, other, perhaps hemodynamic (related to elevated pressures), reasons may be operative.

Intracardiac Mapping

In the electrophysiology lab, a stable and sustained arrhythmia allows the sequential acquisition of data—both anatomic and electrical—to enable the reconstruction of a three-dimensional activation map. Entrainment mapping (which is a sequential data acquisition technique) is also very useful. However, an arrhythmia with varying activation sequences, conduction times, or both does not lend itself to sequential analysis; accordingly, data acquired simultaneously from multiple sites (by multi-electrode catheters) must be used to determine the chamber of interest. Detailed three-dimensional electroanatomic mapping can be pursued after cardioversion in sinus rhythm to locate fixed barriers. In patients without inducible arrhythmia, the same strategy may also have to be followed.

Ideally, the aim of mapping is to determine the complete re-entrant circuit. This is defined as the spatially shortest route of unidirectional activation returning to the site of earliest activation and encompassing the complete cycle length of the tachycardia in terms of activation timing. Double-loop or multiple-loop re-entry is defined by the presence of more than one activation front fulfilling the above conditions. Limited mapping may lead to an incomplete loop being mistaken for a complete one because of software interpolation. High-density mapping or entrainment mapping can clarify this situation by documenting wavefront collision and long postpacing intervals, respectively. Multiple fixed barriers produce multiple isthmuses, and recurrence may result from a completely different arrhythmia or transformation to a circuit dependent on another of the multiple isthmuses.

The tachycardia behavior can also provide some clues as to its nature. A single-loop tachycardia with a fixed barrier as its core typically remains stable and unchanged during catheter manipulation and can even be difficult to pace terminate. Mechanical “bump” termination (without extrasystoles) suggests a restricted and relatively fragile isthmus. A change in ECG morphology without change in cycle length may be caused by a transformation of a multi-loop tachycardia by interruption of one loop or by a change in bystander activation sufficient to be visible on the surface ECG or in activation of the same circuit in the opposite direction. A significant change in activation within the circuit distinguishes between circuit transformation or antidromic activation around the same circuit and changes in bystander activation. Similarly, variations in cycle length can suggest variations in activation pathways resulting from circuit transformation or simply changes in conduction time, the latter usually manifesting as cycle length alternans. ECG changes may not occur despite changes in activation sequences because of distance from the recording electrodes or insufficient electrically active tissue.

Critical isthmuses can be identified during sustained stable re-entry by activation mapping supplemented by entrainment mapping. However, it is not possible to identify a critical isthmus during sinus rhythm. Significantly large (in two dimensions) areas devoid of electrical activity can be easily recognized as electrical scars, provided that catheter contact is verified, but narrow lines of block (thinner than the recording field of clinically used bipoles) may easily be missed unless the conduction delay across them is maximized by the appropriate choice of optimal pacing sites. It may be necessary to perform mapping during more than one form of activation, for example, during both proximal coronary sinus pacing and low lateral right atrial pacing to identify the majority of potential isthmuses.

In the absence of sustained stable re-entry (allowing sequential mapping), it is important to evaluate the potential of scars or barriers to support re-entry. In a study of 22 consecutive patients with AT after surgical closure of an atrial septal defect, three-dimensional electroanatomic mapping of the right atrium was performed during stable sustained tachycardia and in sinus rhythm to study the properties of electrical scars.

The characteristics of the line of block resulting from the surgical atriotomy on the right atrial free wall played a significant role in determining the kind of arrhythmia circuit that developed.29 A right atrial free wall peri-atriotomy re-entry circuit was more likely to occur if the scar was relatively long, resulting in a restricted isthmus bounded inferiorly by the IVC, and if the scar was vertical or oblique and relatively anteriorly placed. Nearly all these patients with a free wall circuit also had peri-tricuspid re-entry. Isolated peri-tricuspid re-entry was observed when no electrophysiological evidence of a right atrial free wall atriotomy was found or if this scar was too small or posteriorly placed. If the right atrial free wall atriotomy is long enough to extend to the IVC inferiorly, thus eliminating the inferior isthmus, peri-atriotomy re-entry cannot occur. It is not clear whether the near-universal presence of peri-tricuspid re-entry in this cohort of patients is caused by diffuse substrate alterations or results from the presence of a (posteriorly placed) atriotomy buttressing the often-functional block zone of the crista terminalis.

These data suggest that in a right atrial free wall line of block, even if sustained re-entrant arrhythmias are not inducible in the electrophysiological laboratory, both isthmuses (inferior end of scar to IVC and the cavotricuspid isthmus) should be ablated with the endpoint of complete and stable block to eliminate the occurrence of both peri-tricuspid as well as peri-atriotomy flutter.

Patients with pronounced hemodynamic loads, for example, after corrective or palliative surgery such as the Fontan procedure, have the substrate for re-entry in the form of small channels or isthmuses in the altered milieu of a low-voltage area of the right atrial free wall.30 It may be difficult to distinguish multiple small channels within this low-voltage area without three-dimensional mapping, and mapping during different rhythms may be necessary to detect as many potential isthmuses as possible. Despite the frequent presence of surgical incisions in the septum, re-entry in this anatomic location is infrequent.

The remaining minority of atypical right atrial re-entrant tachycardias include the so-called peri-crista re-entry as well as small and functional re-entry circuits in various parts of the right atrium. It is likely that block within the cavotricuspid isthmus facilitates the occurrence of this form of re-entry. Mapping during re-entry is useful to document the circuit, and RF delivery at the site of conduction across the crista region of block or more medially between the ostium of the coronary sinus and the posterior intercaval region can be effective.

Left Atrial Substrate

Re-entry within the left atrium typically is related to scars from surgery or catheter ablation or is associated with structural heart disease. A detailed account of the surgical or catheter ablation procedure is useful to allow optimal mapping within a particular region of interest. Gaps of persisting conduction across linear or isolating lesions are the typical substrate of isthmuses and can be easily recognized by fractionated “diastolic” electrograms coinciding with synchronous isoelectric intervals in all 12 ECG leads.28 These isthmuses usually support small-diameter (1 to 1.5 cm) re-entrant circuits. Less frequently, large re-entrant circuits depend on such small, slow conducting isthmuses. Large circuits typically are the result of wide isthmus re-entry around anatomic barriers or anatomic plus lesion-based barriers. In our experience, they are typically associated with severe left atrial dilation, linear lesions in the left atrium, or both. Multiple linear lesions in the left atrium frequently are associated with the greatest likelihood of multi-loop re-entry.

Ablation Procedure

Ideally, detailed mapping is necessary to determine the full re-entrant circuit to plan and achieve interruption by ablation. An accurate knowledge of the anatomy is very useful in estimating the anatomic extent of isthmuses. Three-dimensional electroanatomic mapping, which combines electrical activation in an anatomic setting, is the strategy of choice. It is often possible to limit mapping to the detection of the critical isthmus, estimate its width, and target it by ablation to eliminate the tachycardia even without knowledge of the complete circuit. It may be possible to eliminate some circuits by using double potential and entrainment mapping, usually the subset of right atrial free wall macro–re-entries, in which the position and extent of the atriotomy scar are well standardized. Once the circuit has been mapped, the safest, narrowest, and most convenient access to the isthmus is chosen for ablation. Entrainment mapping is helpful in choosing between various isthmuses. The anticipated difficulty of creating a complete conduction block across the chosen site also needs to be considered; areas of catheter instability caused by mechanical contraction (e.g., the right atrial free wall in the region of the tricuspid valve annulus) or regions of thick tissue (particularly in patients with congenital heart disease before or after surgical repair) render ablation that much more difficult.

The width of the targeted isthmus, which significantly affects both the duration of the procedure and its success rate, is commonly estimated by anatomic and electrophysiological landmarks. Local electrograms, including double potentials indicating a line of block and long-duration fractionated electrograms suggesting a protected corridor of slow conduction, can be particularly useful. Single-deflection electrograms suggest a wider and relatively large ablation target. Electrophysiological signals are a necessary supplement to anatomic guidance because the electrically active isthmus may be smaller than an anatomically defined one. High-density mapping can reveal that a given segment of the circuit is, in fact, functionally narrower than the anatomy would suggest by demonstrating the presence of lateral boundaries in the form of zones of block.

With the same principles as for typical atrial flutter, RF energy can be delivered sequentially point by point to span the targeted segment or by dragging during continuous energy administration. Lesion contiguity and continuity depend on the coalescence of multiple transmural lesions, best ensured by documenting the breakdown of the target electrogram (at each site) into double potentials and continuing RF delivery at this point for approximately 30 to 40 seconds more to ensure a stable lesion. An irrigated-tip catheter is preferable to permit the delivery of higher power necessary for consistent transmural lesions and to avoid generating char.

Assessment of Outcome

During the delivery of RF lesions, the tachycardia may terminate, or its cycle length may increase, transiently or permanently. Both phenomena indicate that the delivered lesions have slowed conduction within the re-entry circuit and should be followed by continuation of RF or extension of the lesion to achieve complete conduction block.

The inability to re-induce the original arrhythmia is a clearly desirable endpoint but may reflect only conduction delay or variations in autonomic parameters. Complete stable conduction block within the re-entry path is the most objective endpoint for ablation of large-diameter re-entrant arrhythmias, as for accessory AV connections, or typical atrial flutter. The achievement of complete conduction block clearly correlates with lower rates of recurrence in the population with re-entrant tachycardias in the left or right atrium.

An inversion of the activation sequence downstream of ablation, local electrogram changes, or both are characteristic of conduction block in the targeted isthmus. The sensitivity of assessing conduction block depends on the choice of pacing site as for assessment of cavotricuspid isthmus conduction.

To ensure the stability of the block, re-verification is advisable after a waiting period; because careful three-dimensional mapping may be time consuming, an initial assessment by local electrogram criteria followed approximately 15 to 20 minutes later by mapping is an effective and time-saving strategy.

The subset of small re-entrant circuits requires careful mapping for their recognition and can be difficult to distinguish from non–re-entrant arrhythmias. Because of their characteristic dependence on a fragile, narrow, and slow-conducting isthmus (which gives rise to fractionated low-amplitude electrograms coinciding with 12-lead isoelectric intervals), they are relatively easy to ablate with one or few RF current applications. However, their small circuit dimensions render it difficult to use activation mapping to demonstrate block through the ablated isthmus. Tachycardia noninducibility is the only available endpoint.

The different re-entrant circuits encountered in the right and left atria require ablation to be individually tailored. Multiple isthmuses may require ablation. An empiric linear lesion—Maze-like solution, including multiple lesions to block all anatomic isthmuses—may be necessary, although with current ablation techniques, it is difficult to achieve conduction block across long linear lesions. At present, success rates for ablation of typical atrial flutter remain significantly higher than for atypical flutter. When keeping in mind the complexity of atypical flutter, this is perhaps not so difficult to understand.

Symptomatic and drug-refractory atypical flutters usually are considered for ablation, although the threshold typically is lower in the presence of tachycardiomyopathy, postoperative congenital heart disease, or left ventricular dysfunction. However, the complexity of mapping and ablation means that the experience and success rates of individual centers should be considered.

Non–re-entrant Atrial Tachycardias

This arrhythmia subset must be distinguished from re-entrant AT because the approach to successful ablation is clearly different (Box 93-5).

The major difference is a radial pattern of activation during these tachycardias.31 For practical purposes, the atria, unlike the ventricles, can be considered to be two dimensional. The absence of electrical activity spanning a significant part of the cycle length indicates a diastolic pause characteristic of abnormal impulse generation resulting from triggered activity or abnormal automaticity. This deduction presupposes an adequate and complete exploration of the endocardium, including the great thoracic veins; the coronary sinus should not be considered a surrogate for the left atrium. Mapping reveals radial atrial activation confined to electrical systole, that is, coinciding with the surface ECG P wave. The sequence of activation of the two bipoles of the rove mapping and ablation catheter can be useful; a distal-to-proximal activation persisting with catheter advancement indicates a vector originating from that direction, whereas a proximal-to-distal activation sequence suggests that the wavefront originates from the direction of the proximal bipole. An iterative mapping sequence documenting the transition from the former to the latter activation direction is characteristic of the source of radial activation and requires returning to the initial catheter position.

The inference of a non–re-entrant mechanism is further strengthened by a pattern of arrhythmia bursts with cycle length irregularity, warm-up behavior, or both, unlike a stable re-entrant mechanism (see Figure 93-4). Identical activation sequences for the first and subsequent beats of the tachycardia further support a non–re-entrant mechanism. If the arrhythmia is sustained, demonstration of entrainment is supportive of re-entry. If the re-entry circuit is small, radial activation of the rest of the atria cannot be distinguished from a non–re-entrant mechanism. Therefore recognition of small re-entry circuits depends on the mapping resolution. If the arrhythmia source is confined to an unmapped area (e.g., in the left atrium or within the SVC), the arrhythmia mechanism may be difficult to recognize. Unlike re-entry, pacing maneuvers—typically overdrive—suppress the non–re-entrant tachycardia without demonstrating the classic features of entrainment with progressive fusion. In the experimental laboratory, termination of the tachyarrhythmia by sectioning a critical part of the circuit is a highly specific evidence of re-entry, but clinically, the difficulty in consistently achieving a completely transmural incision such as a lesion by RF delivery reduces the value of this criterion.

The surface ECG with a clearly visible P wave in all 12 leads is highly valuable for localizing the origin of the tachycardia. A clear isoelectric baseline separating individual P waves is compatible with non–re-entrant mechanisms; however, some very rapid non–re-entrant tachycardias may not exhibit a baseline. P-wave polarity in leads V1, I, and aVL is important in assigning the atrium of origin.32 A late positive, dominantly positive, or completely positive P wave in lead V1 indicates left atrial origin.33 Septally originating tachycardias exhibit narrower P waves, and the frontal plane axis is helpful as it indicates a superior or inferior origin. Localization can be further refined with the help of pacemapping by using both the surface ECG as well as patterns of intracardiac activation.

Multi-catheter activation mapping allows quicker assessment of approximate localization, although the selection of an appropriate ablation site still requires careful mapping and sufficient arrhythmia. The optimal ablation site is one with the earliest bipolar and unipolar activation, although the issue of determining the early enough timing remains. A timing preceding the surface ECG onset of the P wave with a QS morphology on the unipolar electrogram and that does not precede bipolar activation at that site is usually necessary (see Figure 93-4).

Sites on or close to the inter-atrial septum as well as the posterior mitral annulus may need to be mapped from more than one side to choose the better site—the former from the right and the left atria and the latter from the endocardium as well as the epicardium (coronary sinus).

In most instances, focal ablation suffices and arrhythmia elimination and noninducibility are the only endpoints. Venous tachycardia (which can be either re-entrant or non–re-entrant) is a specific situation, particularly a pulmonary vein tachycardia, which can even trigger and maintain AF. An expeditious solution is to disconnect the vein from the left atrium (right atrium in case of the SVC), making sure that the disconnection is proximal to the source of the arrhythmia.34 This endpoint is useful even with arrhythmia that is difficult to induce.

The role of three-dimensional mapping systems in understanding and ablating these arrhythmias is unclear. Systems that rely on sequential data acquisition require sufficient reproducible arrhythmia, whereas simultaneous data-acquiring techniques such as multi-electrode basket arrays or the Ensite noncontact system have an advantage if fewer arrhythmias are available for analysis. The resolution of noncontact mapping systems is, however, not good enough at present to allow selection of the ablation site in the absence of ambient arrhythmia.

Indications for catheter ablation include the following35:

Key References

Haissaguerre M, Dartigues JF, Warin JF, et al. Electrogram patterns predictive of successful catheter ablation of accessory pathways. Value of unipolar recording mode. Circulation. 1991;84(1):188-202.

Haissaguerre M, Gaita F, Fischer B, et al. Elimination of atrioventricular nodal reentrant tachycardia using discrete slow potentials to guide application of radiofrequency energy. Circulation. 1992;85(6):655-656.

Jackman WM, Beckman KJ, McClelland JH, et al. Treatment of supraventricular tachycardia due to atrioventricular nodal reentry, by radiofrequency catheter ablation of slow pathway conduction. N Engl J Med. 1992;327(5):313-318.

Jackman WM, Friday KJ, Yeung-Lai-Wah JA, et al. New catheter technique for recording left free wall accessory atrioventricular pathway activation. Identification of pathway fiber orientation. Circulation. 1988;78:598-611.

Jais P, Shah DC, Haissaguerre M, et al. Mapping and ablation of left atrial flutters. Circulation. 2000;101(25):2928-2934.

Klein GJ, Bashore TM, Sellers TD, et al. Ventricular fibrillation in the Wolff-Parkinson-White syndrome. N Engl J Med. 1979;301(20):1080-1085.

Nakagawa H, Shah N, Matsudaira K, et al. Characterization of reentrant circuit in macroreentrant right atrial tachycardia after surgical repair of congenital heart disease: Isolated channels between scars allow “focal” ablation. Circulation. 2001;103(5):699-709.

Sellers TDJr, Gallagher JJ, Cope GD, et al. Retrograde atrial preexcitation following premature ventricular beats during reciprocating tachycardia in the Wolff-Parkinson-White syndrome. Eur J Cardiol. 1976;4:283-294.

Shah DC, Haissaguerre M, Jais P, et al. High density mapping of activation through an incomplete isthmus ablation line. Circulation. 1999;99(2):211-215.

Shah D, Jais P, Takahashi A, et al. Dual loop intra-atrial reentry in humans. Circulation. 2000;101(6):631-639.

Shah DC, Takahashi A, Jais P, et al. Local electrogram based criteria of cavotricuspid isthmus block. J Cardiovasc Electrophysiol. 1999;10(5):662-669.

Tang CW, Scheinman MM, Van Hare GF, et al. Use of P wave configuration during atrial tachycardia to predict site of origin. J Am Coll Cardiol. 1995;26(5):1315-1324.

Tchou P, Lehmann MH, Jazayeri M, Akhtar M. Atriofascicular connection or a nodoventricular Mahaim fiber? Electrophysiological elucidation of the pathway and associated reentrant circuit. Circulation. 1988;77:837-848.

Yamane T, Shah DC, Peng JT, et al. Morphological characteristics of P waves during selective pulmonary vein pacing. J Am Coll Cardiol. 2001;31(5):1505-1510.

References

1 Gallagher JJ, Pritchett ELC, Sealy WC, et al. The preexcitation syndromes. Prog Cardiovasc Dis. 1978;20:285-327.

2 Martinez-Alday JD, Almendral J, Arenal A, et al. Identification of concealed posteroseptal Kent pathways by comparison of ventriculoatrial intervals from apical and posterobasal right ventricular sites. Circulation. 1994;89:1060-1067.

3 Hirao K, Otomo K, Wang X, et al. Para-Hisian pacing. A new method for differentiating retrograde conduction over an accessory AV pathway from conduction over the AV node. Circulation. 1996;94:1027-1035.

4 Sellers TDJr, Gallagher JJ, Cope GD, et al. Retrograde atrial preexcitation following premature ventricular beats during reciprocating tachycardia in the Wolff-Parkinson-White syndrome. Eur J Cardiol. 1976;4:283-294.

5 Klein GJ, Bashore TM, Sellers TD, et al. Ventricular fibrillation in the Wolff-Parkinson-White syndrome. N Engl J Med. 1979;301(20):1080-1085.

6 Haissaguerre M, Dartigues JF, Warin JF, et al. Electrogram patterns predictive of successful catheter ablation of accessory pathways. Value of unipolar recording mode. Circulation. 1991;84(1):188-202.

7 Jackman WM, Friday KJ, Yeung-Lai-Wah JA, et al. New catheter technique for recording left free wall accessory atrioventricular pathway activation. Identification of pathway fiber orientation. Circulation. 1988;78:598-611.

8 Kirsh JA, Gross GJ, O’Connor S, et al. Transcatheter cryoablation of tachyarrhythmias in children: Initial experience from an international registry. J Am Coll Cardiol. 2005;45:133-136.

9 Takahashi A, Shah DC, Jais P, et al. Specific electrocardiographic features of manifest coronary vein posteroseptal accessory pathways. J Cardiovasc Electrophysiol. 1998;9(10):1015-1025.

10 Tchou P, Lehmann MH, Jazayeri M, Akhtar M. Atriofascicular connection or a nodoventricular Mahaim fiber? Electrophysiological elucidation of the pathway and associated reentrant circuit. Circulation. 1988;77:837-848.

11 Gaita F, Haissaguerre M, Scaglione M, et al. Catheter ablation in a patient with a congenital giant right atrial diverticulum presenting as Wolff-Parkinson-White syndrome. Pacing Clin Electrophysiol. 1999;22(2):382-385.

12 Goya M, Takahashi A, Nakagawa H, Iesaka Y. A case of catheter ablation of accessory atrioventricular connection between the right atrial appendage and right ventricle guided by a three dimensional electroanatomic mapping system. J Cardiovasc Electrophysiol. 1999;10(8):1112-1118.

13 Veenhuyzen GD, Coverett K, Quinn FR, et al. Single diagnostic pacing maneuver for supraventricular tachycardia. Heart Rhythm. 2008;5:1152-1158.

14 Wu J, Wu J, Olgin J, et al. Mechanisms underlying the reentrant circuit of atrioventricular nodal reentrant tachycardia in isolated canine atrioventricular nodal preparations using optical mapping. Circ Res. 2001;88(11):1189-1195.

15 Jazayeri MR, Hempe SL, Sra JS, et al. Selective transcatheter ablation of the fast and slow pathways using radiofrequency energy in patients with atrioventricular nodal reentrant tachycardia. Circulation. 1992;85(4):1318-1328.

16 Jackman WM, Beckman KJ, McClelland JH, et al. Treatment of supraventricular tachycardia due to atrioventricular nodal reentry, by radiofrequency catheter ablation of slow pathway conduction. N Engl J Med. 1992;327(5):313-318.

17 McGuire MA, de Bakker JM, Vermeulen JT, et al. Origin and significance of double potentials near the atrioventricular node. Correlation of extracellular potentials, intracellular potentials, and histology. Circulation. 1994;89(5):2351-2360.

18 Haissaguerre M, Gaita F, Fischer B, et al. Elimination of atrioventricular nodal reentrant tachycardia using discrete slow potentials to guide application of radiofrequency energy. Circulation. 1992;85(6):655-656.

19 De Sisti A, Tonet J, Gueffaf F, et al. Effects of inadvertent atrioventricular block on clinical outcomes during cryoablation of the slow pathway in the treatment of atrioventricular nodal re-entrant tachycardia. Europace. 2008;10(12):1421-1427.

20 Shah DC, Jais P, Haissaguerre M, et al. Three dimensional mapping of the common atrial flutter circuit in the right atrium. Circulation. 1997;96(11):3904-3912.

21 Shah DC, Takahashi A, Jais P, et al. Local electrogram based criteria of cavotricuspid isthmus block. J Cardiovasc Electrophysiol. 1999;10(5):662-669.

22 Shah DC, Haissaguerre M, Jais P, et al. High density mapping of activation through an incomplete isthmus ablation line. Circulation. 1999;99(2):211-215.

23 Jais P, Shah DC, Haissaguerre M, et al. Prospective randomized comparison of irrigated tip versus conventional tip catheters for ablation of common flutter. Circulation. 2000;101(7):772-776.

24 Shah DC, Takahashi A, Jais P, et al. Tracking dynamic conduction recovery across the cavotricuspid isthmus. J Am Coll Cardiol. 2000;35(6):1478-1484.

25 Shah D, Haissaguerre M, Takahashi A, et al. Differential pacing for distinguishing block from persistent conduction through an ablation line. Circulation. 2000;102(13):1517-1522.

26 Jais P, Shah DC, Haissaguerre M, et al. Mapping and ablation of left atrial flutters. Circulation. 2000;101(25):2928-2934.

27 Shah D, Jais P, Takahashi A, et al. Dual loop intra-atrial reentry in humans. Circulation. 2000;101(6):631-639.

28 Shah D, Sunthorn H, Burri H, et al. Narrow, slow-conducting isthmus dependent left atrial reentry developing after ablation for atrial fibrillation: ECG characterization and elimination by focal RF ablation. J Cardiovasc Electrophysiol. 2006;17(5):508-515.

29 Shah DC, Jais P, Hocini M, et al. Catheter ablation of atypical right atrial flutter. In Zipes DP, Haissaguerre M, editors: Catheter ablation of arrhythmias, ed 2, Armonk, NY: Futura publications, 2001.

30 Nakagawa H, Shah N, Matsudaira K, et al. Characterization of reentrant circuit in macroreentrant right atrial tachycardia after surgical repair of congenital heart disease: Isolated channels between scars allow “focal” ablation. Circulation. 2001;103(5):699-709.

31 Goldreyer BN, Gallagher JJ, Damato AN. The electrophysiologic demonstration of atrial ectopic tachycardia in man. Am Heart J. 1973;85:205-215.

32 Tang CW, Scheinman MM, Van Hare GF, et al. Use of P wave configuration during atrial tachycardia to predict site of origin. J Am Coll Cardiol. 1995;26(5):1315-1324.

33 Yamane T, Shah DC, Peng JT, et al. Morphological characteristics of P waves during selective pulmonary vein pacing. J Am Coll Cardiol. 2001;31(5):1505-1510.

34 Shah DC, Haissaguerre M, Jais P, Clementy J. High resolution mapping of tachycardia originating from the superior vena cava: Evidence of electrical heterogeneity, slow conduction, and possible circus movement reentry. J Cardiovasc Electrophysiol. 2002;13(4):388-392.

35 Blomström-Lundqvist C, Scheinman MM, Aliot EM, et al. ACC/AHA/ESC guidelines for the management of patients with supraventricular arrhythmias—executive summary: A report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines and the European Society of Cardiology Committee for Practice Guidelines. Circulation. 2003;108(15):1871-1909.