3

How to Ablate Atrial Flutter Postsurgery

Aarti Dalal, DO; George F. Van Hare, MD

Introduction

This chapter will discuss current techniques for mapping and ablating AFL in patients who have previously undergone heart surgery. The unique challenge of catheter ablation in patients with previous cardiac surgery has resulted in guidelines from the Pediatric and Adult Congenital Electrophysiology Society and the Heart Rhythm Society.¹ Catheter ablation, while often challenging, is an attractive option in patients who have undergone surgery for congenital heart disease. Experience has shown that such tachyarrhythmias are very unlikely to disappear spontaneously, and therefore, the need for antiarrhythmic therapy is likely to be lifelong, absent definitive therapy with ablation. There is an increased incidence of sinus node dysfunction in this patient population, and the addition of antiarrhythmic agents may cause a patient with sinus node disease to experience new or more serious symptoms. These symptoms may include syncope, mandating implantation of a permanent pacemaker in order to continue antiarrhythmic therapy. Similarly, this patient population often has coexisting ventricular dysfunction. Many of the most effective agents for the control of tachyarrhythmias have the potential to worsen ventricular dysfunction, especially beta blockers and sotalol.

The most common form of arrhythmia seen in the postoperative congenital heart disease patient population is atrial flutter (AFL), also known as intra-atrial reentrant tachycardia (IART). To understand the techniques used in mapping of large macroreentrant circuits, several concepts need to be considered: the concept of barriers to impulse propagation, and the concept of sites that are “in the circuit” versus sites that are “outside the circuit.” These concepts come originally from classic studies by Waldo, et al,² which were applied by various workers to the mapping and ablation of common AFL in adult patients^3–5 and subsequently extended for use in postoperative patients.^6–9

Initial activation mapping studies of the typical form of AFL showed a “counterclockwise” reentrant activation in the RA,¹⁰ with impulses spreading up the septum and down the RA free wall. From studies using concealed entrainment techniques, described below, as well as techniques for precise placement of ablative lesions, it is now well established that one critical element of the AFL reentrant circuit is the isthmus between the inferior vena cava (IVC) and the tricuspid valve annulus.¹¹ This area of tissue is protected by these two barriers to impulse propagation, which prevent the reentrant wave from circling back and catching the “tail of refractoriness,” and thereby being extinguished. The situation is more complex, however, than simply a small isthmus of tissue between two small barriers. In fact, as shown by Olgin et al and Kalman et al, it is not the IVC per se but actually the crista terminalis and its extension as the Eustachian valve ridge that acts as the barrier to impulse propagation.^3,5 The crista terminalis is formed at the junction between the sinus venosus portion and the true, heavily trabeculated portion of the RA; it runs along the posterolateral aspect of the RA, coursing inferiorly. As it reaches the region of the IVC, it is extended by the Eustachian valve ridge, which courses superiorly to the os of the CS, joining with the valve of the CS to form the tendon of Todaro. In patients with common AFL, the crista terminalis has been shown to act as a long line of intra-atrial block. This block seems to be anatomic and fixed rather than functionally determined in patients with clinical AFL. The tricuspid annulus constitutes the “anterior barrier” in typical flutter. Sites around the tricuspid annulus are activated sequentially and in a counterclockwise direction.

These 2 long barriers to impulse propagation form a “funnel” of conducting tissue in the RA. This funnel forces atrial activation to the narrow isthmus between the tricuspid annulus and the IVC, where, because of the short distance, the reentrant circuit is most amenable to successful ablation.

It is interesting that in the otherwise normal human heart, despite the fact that there are numerous potential barriers to impulse propagation (IVC, superior vena cava [SVC], mouth of coronary sinus [CS], tricuspid and mitral valve annuli, ostia of pulmonary veins, crista terminalis), the vast majority of atrial reentrant arrhythmias are due to common counterclockwise or clockwise AFL. This fact speaks to the importance of the crista terminalis and tricuspid annulus. One would expect these structures to also be important in IART, which is seen following congenital heart disease surgery. The effect of extensive atrial surgery is clearly complex and may involve several sequelae that make IART more likely. First, the creation of a long atriotomy with subsequent suture closure creates a long line of block of impulse propagation, which is superimposed on the existing RA anatomy described above. Second, such an atriotomy may modify the typical flutter circuit by making it longer, thereby lengthening the tachycardia cycle length and slowing the atrial tachycardia rate. Third, the placement of an atriotomy near the crista terminalis or the use of the crista terminalis for anchoring a suture line (as is done in the lateral Fontan modification) may cause the crista terminalis to begin to act as a line of conduction block.12 Finally, extensive atrial surgery may cause slowing of conduction, making reentry more likely. At present, it is not known which of these possible mechanisms is most important. It is clear from clinical experience, however, that slow flutter involving the posterior flutter isthmus is very common in postoperative patients¹³; circuits that do not include the typical flutter zone and so are due to reentry involving incisional suture lines are also frequently seen.¹⁴

Preprocedure Planning

In preparation for mapping a patient with IART, it is important to carefully review the patient’s cardiovascular anatomy, and, in particular, the exact surgical approach that was used and history of vascular obstruction. This will be facilitated by a review of the original operative reports and previous interventional procedures such as EP or cardiac catheterizations. The details of the exact placement of atriotomies, baffles, patches, and conduits will become important in the interpretation of the EP recordings and the results of mapping. If possible, the sites bounded by surgically created and anatomic obstacles to impulse propagation should be identified, and several possible candidate sites for ablation should be determined prior to the study. For example, for patients who have undergone simple surgery, such as repair of an atrial septal defect, such sites might be (a) the typical flutter isthmus, (b) between an atriotomy and the tricuspid annulus, or (c) between an atriotomy and the SVC. The same sites are commonly found in patients with more complex surgery, such as repair of tetralogy of Fallot, because a long atriotomy is often employed in these repairs (Figure 3.1). Similarly, knowledge of vascular anatomy will help the proceduralist determine ahead of time if non-traditional approaches like a transhepatic approach is warranted.¹⁶ The cardiologist must combine an intimate knowledge of the patient’s congenital defect with knowledge of the details of the exact surgical procedure used previously to determine appropriate ablation sites.

Procedure

General Techniques for Mapping

In general, methods for mapping clinical atrial arrhythmias may be classified in three broad categories: single-site roving mapping, simultaneous multisite mapping, and “destructive” mapping. In practice, the typical atrial tachycardia ablation incorporates elements of all three. Single-site mapping involves the use of a single, steerable catheter, which is maneuvered throughout the atrium during tachycardia. Electrograms from various sites are recorded, and the map is constructed from these nonsimultaneous measurements, ideally using a 3-dimensional (3D) mapping system to record timing of activation superimposed on the anatomy of the atrium. Unfortunately, with some patients, the tachycardia mechanism may change in the midst of a map, forcing the operator to stop to re-induce the original rhythm. For this reason, when confronted by a substrate with numerous tachycardia circuits, “substrate mapping” may be considered, in which a voltage map in sinus or paced rhythm is constructed, allowing for the identification of areas of scar, lines of block, suture lines, and other important anatomic details.¹⁵

Simultaneous multisite mapping involves various systems for introducing large numbers of electrodes into or onto the heart. These may include basket catheters or, more commonly, noncontact systems that compute virtual electrograms based on far-field intracavitary systems (e.g., EnSite, St. Jude Medical or CARTO, Biosense Webster). An advantage is the potential to obtain a map on one beat of tachycardia, yet still see the entire circuit. It is limited, however, by the basic inability to introduce electrodes in all parts of both atria in the catheterization laboratory. Also, in large chambers, resolution will not be adequate. Areas that are “in the circuit” may not easily be identified and separated from those that are “out of the circuit” without the ability to perturb the system—for example, by entrainment pacing.

“Destructive” mapping is defined as the direct interruption of an area of conducting myocardium, with subsequent observation to determine whether the target rhythm has been eradicated. This may be done by the delivery of RF or cryoablation lesions during tachycardia. Ideally, such lesions are directed by the use of detailed substrate maps to target these lesions. A successful RF lesion that terminates a tachyarrhythmia is perhaps the best evidence that the site chosen for ablation was critical for maintenance of the arrhythmia. In 1914, Mines17 recognized the limitations of multisite mapping, saying in reference to AFL that “the test for a circulating excitation is to cut through the ring at one point thereby terminating the flutter.” The advantage of this approach is that the lesion may very well be curative. The limitation, of course, is the potential for needless destruction of working myocardium that is not involved in the tachycardia, as well as the potential for lengthening the reentrant circuit, slowing the tachycardia, and making it more incessant.

Identification of Lines of Block

During the EP study, the goal is to identify an isthmus of tissue that is bounded by two long barriers. The identification, for example, of the tricuspid annulus, which often provides one important barrier in such patients, is not challenging, as one has fluoroscopic landmarks as well as local atrial electrogram characteristics. Specifically, on the tricuspid annulus, one normally records both atrial and ventricular electrograms, and these are approximately equivalent in size when the catheter is resting on the annulus. Other sites of conduction block are identified by the presence of double potentials, reflecting conduction up one side of the barrier and down the other side, with the bipolar electrogram recording both waves of atrial activation (Figure 3.2). Such double potentials are easily recorded in patients with common AFL along the crista terminalis and the Eustachian valve ridge. In patients who have undergone atrial surgery, a long atriotomy is often identified along the anterior wall of the atrium and may be followed along the atrial wall for some distance. In patients after the Senning procedure for transposition, a long line of double potentials may be recorded along the edge of the baffle in the systemic venous atrium.⁸

It must be emphasized that the identification of a line of double potentials is not sufficient for the completion of the map, because such lines of double potentials are very common, and often are not associated in any way with the actual reentrant circuit. That is, both, either, or neither of the areas of atrial myocardium on either side of the line of block may be involved in the reentrant circuit. Areas that are uninvolved in the circuit are considered to be “bystander” areas. Confirmation that the line of block is critical for the tachycardia circuit must be obtained by assessment of the entrainment response from areas of viable tissue adjacent to these lines of block.

Current mapping techniques should include assessment of the entrainment response, partly for the confirmation of a reentrant mechanism and partly for identification of candidate sites for ablation. An understanding of these concepts is essential. Waldo described a number of criteria for transient entrainment,2 and others have suggested additional criteria.¹⁸ The demonstration of any one of these criteria establishes that the atrial tachycardia is a macroreentrant rhythm with an excitable gap. The most common of these criteria is the identification of constant fusion: pacing into tachycardia at a rate slightly faster than the AFL rate produces a P-wave morphology that is intermediate between the free-running tachycardia P-wave morphology and the morphology that would have been seen with simple pacing into sinus rhythm. Fusion is due to collision between the orthodromic wavefront and an antidromic wavefront emanating from the pacing site, thereby changing atrial activation sequence. The observation of constant fusion cannot be made in the case of a focal automatic tachycardia, in which fusion either would not be seen or would not be constant. In the situation of postoperative AFL, the distinction is important, as a focal tachycardia can occupy a potential reentrant circuit created by lines of block and masquerade as a reentrant rhythm. Electroanatomic mapping is unable to make the distinction. Therefore, it is always preferred to demonstrate one of the criteria for transient entrainment before proceeding with mapping and ablation.

“Concealed entrainment” has been defined as the inability to demonstrate the usual criteria for entrainment in an atrial rhythm that is otherwise known to be reentrant, despite demonstration of acceleration to the pacing rate and return to the tachycardia rate following pacing. One common cause for concealed entrainment is that of pacing from within a protected zone of slow conduction. There is no fusion, because the retrograde wave of activation collides with the antegrade wave in the protected zone, and the atrial activation sequence changes little. There is latency between the pacing stimulus and the onset of the P wave due to conduction within this protected zone. Furthermore, the degree of latency from the stimulus to the P-wave onset during entrainment pacing is similar to the latency, when not pacing, from the local electrogram at the pacing site to the onset of the P wave. Most importantly for the purposes of entrainment mapping, when one terminates pacing, the time necessary for return of the wave of activation to the pacing site, the postpacing interval (PPI), is the same as the tachycardia cycle length (TCL) (Figures 3.3 and 3.4). As pointed out by Stevenson with respect to ventricular tachycardia, the characteristic of PPI = TCL should be a reliable indication of whether a given site is within or outside of the reentry circuit.²⁰

Stay updated, free articles. Join our Telegram channel

Join