Abstract
Atrial substrate modification is required for a successful outcome in a minority of patients with paroxysmal atrial fibrillation (AF), and in most patients with persistent AF. Substrate modification is considered when AF persists despite effective elimination of pulmonary vein (PV) arrhythmogenicity by extraostial PV isolation (PVI), antral PVI, or wide area circumferential ablation. Substrate modification strategies are linear ablation, ablation guided by complex fractionated atrial electrograms, and ablation of ganglionic plexi. Termination of AF to sinus rhythm or to an atrial tachycardia is considered the most favorable procedural end point for substrate modification. Complete bidirectional conduction block should be confirmed when linear ablation is performed.
Keywords
arrhythmogenic substrate, atrial fibrillation, catheter ablation
Key Points
- •
Atrial substrate modification is required for a successful outcome in a minority of patients with paroxysmal atrial fibrillation (AF) and in most patients with persistent AF.
- •
Substrate modification is considered when AF persists despite effective elimination of pulmonary vein (PV) arrhythmogenicity by extraostial PV isolation (PVI), antral PVI, or wide area circumferential ablation.
- •
Substrate modification strategies are linear ablation, ablation guided by complex fractionated atrial electrograms, and ablation of ganglionic plexi.
- •
Termination of AF to sinus rhythm or to an atrial tachycardia is considered the most favorable procedural end point for substrate modification.
- •
Complete bidirectional conduction block should be confirmed when linear ablation is performed.
Mechanisms of Atrial Fibrillation and Rationale for Substrate Ablation
The pathogenesis of atrial fibrillation (AF) is complex and multifactorial. Pulmonary vein (PV) tachycardias have demonstrated that they play a critical role in both initiation and perpetuation of AF.
Although elimination of PV arrhythmogenicity has been highly effective for paroxysmal AF, it has modest efficacy for persistent AF, suggesting that mechanisms beyond the PVs also contribute to perpetuation of AF in these patients.
AF promotes diffuse electroanatomic remodeling. AF results in a nonhomogeneous decrease in atrial refractoriness and slowing of intraatrial conduction. Histologic examinations of atrial tissue in patients with AF show patchy fibrosis, which may contribute to the nonhomogeneity of conduction. Atrial biopsies from patients undergoing cardiac surgery show an increase in cell size, loss of sarcoplasmic reticulum and atrial myofibrils, changes in mitochondrial shape, accumulation of glycogen granules, alteration in connexin expression, and increase in extracellular matrix. Structural changes in response to AF may be a consequence of a physiologic adaptation to chronic Ca 2+ overload and metabolic stress. Reduction of atrial compliance and contractility during AF may enhance atrial dilation, which may add to the persistence of AF.
In the multiple-wavelet hypothesis proposed by Moe and Abildskov, multiple randomly propagating and self-perpetuating daughter wavelets act as a mechanism for perpetuation of AF. Critical to the multiple-wavelet hypothesis is a minimal left atrial size that can accommodate the wavelength as determined by the product of the conduction velocity and the effective refractory period (ERP). More recently, high-frequency sources (i.e., rotors), as a result of anisotropic reentry, have demonstrated the ability to perpetuate AF in experimental and simulation models. A novel mapping approach targeting focal sources and rotors has been developed and in some studies was shown to improve AF ablation outcomes in patients undergoing PV isolation (PVI); focal impulse and rotor modulation (FIRM) ablation is discussed in Chapter 18 .
Modulation of the autonomic innervation of the atria through ganglionic plexi (GP) has also been suggested to play a role in AF because an increase in vagal tone is associated with a decrease in the ERP and an increase in spontaneous depolarizations from the PVs and elsewhere in the atria. A number of ablation strategies have been proposed, alone or in combination, to target substrate-related mechanisms beyond the PV arrhythmogenicity, particularly in patients with persistent AF ( Table 17.1 ).
| Ablation Strategy | Targets | Mapping | Substrate Altered | End Point |
|---|---|---|---|---|
| PV isolation | PV antrum and encircling tissue | Anatomic with or without 3D mapping | PV arrhythmogenicity CFAE Autonomics Microreentry Rotors Debulking | Complete PV and antral electrical isolation |
| Linear ablation | LA roof Mitral isthmus Posterior wall isolation | Anatomic with or without 3D mapping | Macroreentry CFAEs with or without autonomics Rotors | Conduction block across lines |
| Electrogram-guided ablation | CFAEs Frequency gradient Activation gradient | Electrogram features with or without computerized analysis | Slow conduction Rotors, high-frequency sources Autonomics | AF termination Elimination of CFAE |
| Autonomics | Parasympathetic ganglionic plexi LOM | High-frequency pacing (plexi) Angiography (LOM) | Autonomics with or without CFAE | Absence of vagal response |
Antral Pvi
Pathophysiology
The pathophysiologic basis for PVI is covered in detail in Chapter 14 . The PVs may serve as critical sources of rapid repetitive depolarizations, referred to as intermittent PV tachycardias , as a result of triggered activity, automaticity, or reentry that both initiate and sustain AF. In addition to a direct role of the PVs, the atrial tissue around the PVs may harbor complex myofibril arrays, GP, and areas of slow conduction and fractionated electrical activity, which also contribute to AF. In addition to eliminating PV arrhythmogenicity, PVI may also result in ablation of anchor points for rotors, which are more prevalent in the antral regions of the PVs; debulking of the left atrium (LA); ablation of GP; and ablation of arrhythmogenic foci other than the PVs, such as the ligament of Marshall (LOM) and posterior LA.
Mapping and Ablation
Antral PV isolation involves electrical isolation of the PVs and their respective antra, which often includes most of the posterior left atrial wall, at the anterior aspect of the left-sided PVs, where ablation is performed along the ostial aspect of the ridge between the left atrial appendage and the PVs ( Figs. 17.1 and 17.2 ). By extending the encircling lesions outside the PV into the antral or atrial tissue, residual arrhythmogenic foci and GP may simultaneously be eliminated. Thus PVI is generally considered a cornerstone of current catheter ablative therapy for AF. The end point of ablation is complete electrical isolation of the PV, confirmed preferably with a circular mapping catheter.
Outcomes
The clinical outcomes after PVI are reviewed in Chapter 14 , Chapter 15 . PVI is generally considered an appropriate stand-alone procedure for patients with paroxysmal AF. Patients with nonparoxysmal forms of AF typically require additional ablation, specifically for substrate modification, to achieve maximal benefit from ablation procedures.
Problems and Limitations
PVI is a complex and technically demanding procedure. However advances in catheter technology and energy sources have improved efficacy, efficiency, and safety of the ablation procedure.
An important safety consideration during ablation along the posterior left atrial wall is the risk of inadvertent collateral injury to the esophagus. Atrioesophageal fistula is a rare but often fatal complication. Ingestion of barium paste and esophageal temperature monitoring have both been used to prevent injury to the esophagus during ablation. However, esophageal luminal temperature measurement may underestimate the true esophageal tissue temperature. There are also attempts to move the esophagus away from the target sites by using specially designed steerable probes.
Linear Ablation
Pathophysiology
Catheter ablation for AF initially consisted of linear ablation to emulate the Cox surgical maze procedure. Linear catheter ablation was first limited to the right atrium and had low efficacy. Later linear ablation was performed in the LA. Linear ablation has been performed both as a stand-alone strategy and as an adjunctive strategy to other ablation techniques targeting the PV antrum and complex electrograms. Several studies demonstrated that additional linear ablation improves the clinical efficacy of catheter ablation in patients with paroxysmal and persistent AF. The original intent of linear ablation for AF was to interrupt macroreentrant circuits. Other possible mechanisms by which linear ablation may improve outcomes of AF ablation are interruption of microreentrant circuits, elimination of anchor points for high-frequency sources, and atrial debulking. Complex fractionated atrial electrograms (CFAEs) may also be prevalent along the course of linear lesions such as the septum or the roof. Finally, autonomic ganglia may be eliminated during linear ablation at certain sites.
Linear lesions may be a necessary step in the conversion of AF to sinus rhythm, often through an intermediate step of atrial tachycardia. In a study that used a stepwise ablation strategy including isolation of thoracic veins, ablation of CFAEs, and linear ablation until AF terminated, linear ablation was necessary in more than 80% of the patients with persistent AF for termination.
Mapping and Ablation
Linear ablation has been performed along the roof of the LA between the contralateral superior PVs, along the lateral mitral isthmus between the ostium of the left inferior PV and the lateral mitral annulus, along the left atrial septum, from the anterior aspect of the right PV antrum to the septal mitral annulus, along the posterior mitral annulus parallel to the coronary sinus, anteriorly between a roofline and anterior mitral annulus, and along the right atrial aspect of the interatrial septum from the superior vena cava (SVC) to the inferior vena cava (IVC; Fig. 17.3 ). In addition, a box set of lesions to isolate the posterior LA has been performed with an improvement in efficacy in some studies. At present, the left atrial roof and the mitral isthmus are the most commonly targeted sites. Although completeness of conduction block along a linear lesion has not been uniformly assessed, it is always desirable to confirm complete bidirectional conduction block. Incomplete block with slow conduction promotes reentry and may facilitate proarrhythmia, often in the form of persistent or recurrent atrial flutters. Previous studies have suggested that macroreentrant circuits may be present during AF. Elimination of high-frequency drivers that lead to fibrillatory conduction often results in termination of AF to a macroreentrant tachycardia. Therefore linear ablation may interrupt these macroreentrant circuits that coexist with AF.
Left Atrial Roof Line
The goal of the roof line is to produce a line of block between the left and right superior PVs ( Fig. 17.4 ; also see Figs. 17.2 and 17.3 ). The line should be directed as cranially as possible avoiding the posterior wall where esophageal injury may result. It is efficient to perform this ablation after encircling PVI such that the roofline connects the gap between the PVI lines. A long fixed curve (Daig SL0) or steerable sheath can greatly improve catheter contact and stability. Two techniques for creation of this line have been described. In the first method, the catheter is positioned at the margin of the left superior PV and dragged to the right superior PV. The catheter tip is maintained in a perpendicular orientation to the atrial wall ( Fig. 17.5 ). The sheath extends almost to the distal electrode to support and steer the catheter. Energy is delivered for 30 to 60 seconds at each site, moving the catheter by approximately 5-mm increments between locations. Catheter temperature and impedance should be closely monitored because of the perpendicular electrode orientation that may predispose to tissue overheating, steam pops, and perforation. Introduction of contact force sensing ablation catheters have been extremely helpful to create effective lesions safely. (Please see chapter 3 .)
The second approach positions the catheter at the right superior PV with the catheter retroflexed over the sheath ( Fig. 17.6 ). The electrode is parallel to the tissue with this technique. The sheath and catheter are then advanced, driving the electrode toward the left superior PV. Releasing the catheter deflection will also advance the catheter toward the left vein.
Assessment of Conduction Block
A complete linear lesion should result in widely separated double potentials along the length of the line ( Table 17.2 ). A delay of more than 100 ms is usually indicative of complete block, but reliance on conduction time alone can be misleading. During left atrial appendage pacing, posterior left atrial activation should proceed in a caudal-to-cranial direction. Differential pacing maneuvers are also useful to detect slow residual conduction through the line. As the pacing site moves away from the edge of the line, the conduction time from the stimulus to the electrogram on the opposite side of the line will shorten with complete block, but will prolong in the presence of residual slow conduction through the line ( Fig. 17.7 ).
| Finding/Maneuver | Finding Indicative of Complete Block | Comment |
|---|---|---|
| Widely split electrograms | >100 ms generally indicates block | Absolute time between electrogram components depends on local conduction velocities Block may be present with shorter times and conduction persistent with longer times |
| Differential pacing | Pacing site close to edge line produces longer conduction times from stimulus to opposite side of line than pacing at site away from line | Must pace from immediate edge of line |
| Propagation from pacing adjacent to line | Pacing on either side of line produces two wave fronts, each propagating toward the ablation line from opposite directions | Roofline: pace LAA and confirm that posterior wall is activated from caudal to cranial direction Mitral isthmus line: pacing CS lateral to line produces proximal to distal CS activation and pacing CS medial to the line produces distal-to-proximal CS activation Anterior mitral isthmus line: pacing lateral to line produces atrial septal activation from the posterior and lateral directions Posterior wall isolation: entrance and exit block within box |
Mitral Isthmus Line
The goal is to produce a continuous line of block from the lateral mitral isthmus to the left inferior PV. It is recommended that a multipolar catheter be placed in the coronary sinus spanning the linear lesion to serve as an anatomic reference and to monitor the effects of ablation. Up to 70% of patients require ablation within the distal coronary sinus to achieve conduction block. If ablation within the coronary sinus is not possible or unacceptable to the operator due to high impedance or concern for left circumflex artery injury, the creation of this line should be carefully considered. As with any linear ablation, the creation of incomplete linear lesions is often counterproductive by prolonging procedure time, increasing the risk of complications, and possibly creating a proarrhythmic substrate. Using a long sheath, the ablation electrode is positioned at the lateral mitral annulus to record an atrial-to-ventricular ratio of 1:1 or 1:2 ( Fig. 17.8 ). The electrode can be perpendicular or parallel to the tissue (see Fig. 17.8 ). The site on the annulus can be selected to result in the shortest distance to the left inferior vein and its previously created encircling lesion. Lesions are delivered at each site for 30 to 60 seconds with a power limit of 30 to 35 W. Starting at the mitral annulus, the catheter and sheath are rotated with clockwise torque to move the electrode toward the PV in 5-mm steps. Care should be taken to detect inadvertent catheter displacement abruptly into the left inferior PV or appendage. Progress during ablation is gauged by splitting of the ablation electrogram and by delay in conduction across the lesion on the coronary sinus catheter during pacing proximal or distal to the line. Conduction times longer than 100 ms from the stimulus to the electrogram on the opposite side of the line are often associated with conduction block across the line. A more lateral line from the base of the appendage to the annulus may produce block when the initial line fails. Ablation within the coronary sinus can be required to achieve complete block. Ablation in the coronary sinus should be limited to use of irrigated catheters with maximal power of 20 to 25 W. The catheter should be deflected toward the endocardium to avoid the left circumflex artery. An average of 5 ± 4 minutes of radiofrequency (RF) delivered in the coronary sinus usually completes isthmus block in 84% of patients. Ethanol ablation of the vein of Marshall has been considered to achieve complete mitral isthmus block in patients refractory to RF catheter ablation.
Assessing Conduction Block
Achieving complete conduction block along the mitral isthmus can be challenging. With complete isthmus block, widely separated electrograms, usually by more than 100 ms, should be recorded along the length of the ablation line. Pacing the coronary sinus or appendage distal to the ablation line results in proximal-to-distal coronary sinus activation on the septal side of the line ( Fig. 17.9 ). Likewise, coronary sinus pacing proximal to the line results in distal-to-proximal coronary sinus activation lateral to the line. As the pacing site is moved away from the edge of the ablation line, the conduction time from the stimulus to the electrograms recorded on the opposite side of the line decreases in the setting of conduction block, but increases in the setting of slow conduction across the line ( Fig. 17.10 ). To assess conduction block accurately, it is important to position the catheters as close as possible to the mitral isthmus line.
Anterior Left Atrial Line
The goal of this procedure is to create a line of block from either the right superior PV or roofline to the anterior mitral annulus (see Fig. 17.2 ). This line serves as an alternative to the mitral isthmus line and also interrupts more localized reentry circuits in this area. It is recommended that the line start at the anterior mitral isthmus ( Fig. 17.11 ). The ablation catheter is withdrawn with counterclockwise torque to maintain contact with the anterior or anteroseptal left atrial wall. When the ablation electrode reaches the level of the transseptal puncture, clockwise torque is begun and the catheter advanced to reach the right superior PV or roofline. The anterior line may lead to a significant delay in atrial conduction and left atrial appendage activation and may adversely affect left atrial transport function.
Assessing Conduction Block
With complete conduction block, widely separated electrograms should be recorded along the entire length of the line. In addition, pacing the anterior LA just lateral to the line results in left atrial activation proceeding from lateral and posterior directions to activate the left atrial septum. As the pacing site is moved away from the edge of the ablation line, the conduction time from the stimulus to the electrograms recorded on the opposite side of the line decreases in the setting of conduction block, but increases in the setting of slow conduction across the line.
Inferior Mitral Annulus Line
Ablation along the inferior mitral isthmus parallel to the coronary sinus may be performed to interrupt muscular connections between the atrium and coronary sinus and other foci that may perpetuate AF. This line is generally used within the strategy of ablation of complex atrial electrograms. This line is begun with the catheter forming a loop in the LA and directed back toward the atrial septum ( Fig. 17.12 ). The electrode is then withdrawn from the 7- to 4-o’clock position as viewed in the left anterior oblique projection. Opening the catheter curvature as the tip moves laterally will improve tissue contact. Ablation within the coronary sinus is needed to eliminate all complex atrial activity and to produce maximal slowing of the AF rate (see Fig. 17.12 ). The end point for this ablation line is the elimination and organization of complex atrial electrograms within the coronary sinus or significant slowing of the atrial cycle length.
Posterior Left Atrial Isolation
The goal of this lesion set is complete electrical isolation of the left atrial posterior wall. Two techniques are described (see Fig. 17.3 ). In the left atrial box set, the PVs are isolated by antral encircling ablations, and a roofline is created as described previously. A linear lesion is then created between the left and right inferior PV to complete the set. Alternatively, the continuous single-ring linear lesion extends up the ridge between the left PV and the atrial appendage, across the roof, inferiorly along the interatrial septum between the foramen ovale and right PVs, then inferior to the right PV, and lateral across the atrium to meet the start of the line by ascending lateral to the left inferior PVs. Completion of this line isolates the PV and posterior left atrial wall en bloc ( Fig. 17.13 ).
Assessing Conduction Block
The end point for both approaches is the absence or dissociation of electrograms within the lesion box and exit block during pacing from the posterior wall. The PV should also be isolated with the single-ring approach.
Outcomes
Mitral and left atrial rooflines increase AF cycle length to an extent similar to PVI, and AF cycle length prolongation is associated with improved outcome in some studies. Linear lesions may be a necessary step in conversion of AF to sinus rhythm, often through an intermediate step of atrial tachycardia. In a study that used a stepwise ablation strategy, including isolation of thoracic veins, ablation of CFAEs, and linear ablation until AF was terminated, linear ablation was necessary in more than 80% of the patients for termination of persistent AF. For patients with persistent AF, the addition of linear lesions to PVI at the initial procedure reduces the incidence of subsequent macroreentrant arrhythmias. In patients with paroxysmal AF, the addition of confirmed mitral isthmus ablation reduces the recurrence rate of AF to 13% at 1 year, compared with 31% for those undergoing PVI alone. The addition of mitral isthmus ablation also improves outcomes in patients with persistent AF.
After the single-ring lesion, macroreentrant atrial arrhythmias may occur in 34% of patients after the procedure, and recurrent AF may be noted in 35% of patients. Gaps in the ring lesion were found in all patients undergoing repeat ablation for AF after the single-ring approach. Gaps in the ring lesion and mitral annular flutter are responsible for most atrial flutters. The most frequent site of ring gaps is along the left atrial appendage ridge. In a randomized trial, the completion of the box set did not improve outcomes compared with antral PVI and roofline alone.
Recent randomized trial, however, demonstrated that linear ablation after PV isolation (with or without CFAE ablation) did not improve clinical efficacy in patients with persistent and long-standing persistent AF. In another recent study, linear ablation was not incremental after circumferential PV isolation. A meta analysis also suggested that linear ablation after PV isolation is not incremental and may increase the risk of proarrhythmia because of atrial flutter and tachycardias.
Problems and Limitations
The optimal indications and sites for placement of linear lesions are not known. The greatest difficulty with linear ablation is achieving and confirming complete block across the lesions. This is particularly difficult for the mitral isthmus line, which may require extensive ablation within the coronary sinus to accomplish. Incomplete linear lesions may be proarrhythmic and may lead to refractory atrial flutters that are often highly symptomatic. The additional catheter manipulation and ablation may also increase the risk of procedural complications.
Electrogram-Guided Atrial Ablation
Pathophysiology
An intraoperative epicardial mapping study in humans suggested that areas of CFAEs may indicate sites of slow conduction, conduction block, wave front collision, or anchor points for reentrant circuits that can perpetuate AF. CFAEs may also indicate sites of high-frequency sources (rotors), fibrillatory conduction, and sites of autonomic innervation. CFAEs have been targeted for ablation to eliminate both paroxysmal and persistent AF. CFAEs are quite prevalent and can be found at many sites in both atria. The sites that most commonly harbor CFAEs are the interatrial septum, left atrial roof, left atrial appendage, near the PVs, and along the crista terminalis in the right atrium. Clustering of CFAE sites near the PVs appears to be more pronounced in patients with paroxysmal compared with nonparoxysmal forms of AF. CFAEs may be recorded over 38% to 56% of the left atrial endocardium.
Mapping and Ablation
CFAEs are identified during AF by visual assessment or by automated computerized algorithms ( Fig. 17.14 ). In the original description by Nademanee and coworkers, CFAEs were defined as (1) atrial electrograms that are fractionated and composed of two or more deflections, and/or have perturbation of the baseline with continuous deflections of a prolonged activation complex over a 10-second recording period; or (2) atrial electrograms with a very short cycle length (≤120 ms) averaged over a 10-second period. In addition, these electrograms are usually considered to be of low voltage (<0.15 mV). Other electrogram characteristics that have been targeted for ablation are sites with a large (>70 ms) temporal gradient between activation of the proximal and distal ablation bipoles, sites with continuous electrical activity without isoelectric intervals, sites with a cycle length less than the mean left atrial cycle length, and sites demonstrating centrifugal activation ( Fig. 17.15 ; Table 17.3 ).
