Rationale and Outcomes of Cryoballoon Ablation of the Left Atrial Posterior Wall in Conjunction with Pulmonary Vein Isolation

There is strong evidence in support of pulmonary vein isolation (PVI) with concomitant left atrial (LA) posterior wall (PW) isolation (PWI) for the treatment of patients with persistent atrial fibrillation (persAF). While this may be achieved using surgical and catheter-based strategies, there is growing interest in performing this approach using the cryoballoon. There are several potential advantages to this strategy. First, lesions created using the current-generation cryoballoons are typically large and durable. Second, cryoballoon ablation offers a simple technique to directly ablate and debulk the LAPW. Moreover, some consider cryoenergy a safer modality specifically with regard to collateral structures (ie, the esophagus). Based on the available data, cryoballoon PVI + PWI is associated with greater intraprocedural AF terminations and reductions in long-term AF recurrence (typically by ~20%), as compared to PVI alone in patients with persAF, but with similar rates of adverse events. As such, PVI + PWI has emerged as a significant predictor of freedom from recurrent AF (odds ratio: 3.67, 95% confidence interval: 1.44–9.34; p = 0.006) as well as all atrial arrhythmias (hazard ratio: 2.04, 95% confidence interval: 1.15–3.61; p = 0.015). Adjunct radiofrequency ablation to complete PWI is required in at least one-third of the patients, and this need is highly predicted by the LA size (significantly increased with an LA diameter > 48 mm). LAPW reconnection also seems to be associated with LA dimension, particularly an LA diameter greater than 48 mm (negative predictive value: 89.7%). Nevertheless, based on the analysis of patients who underwent repeat electrophysiology study for arrhythmia recurrences, cryoballoon PVI + PWI yields acceptable long-term durability (> 80%).


Introduction
Pulmonary vein (PV) isolation (PVI) remains an effective treatment strategy for paroxysmal atrial fibrillation (AF). 1 However, PVI by itself is often insufficient for the treatment of patients with persistent AF (persAF), 2,3 with success remaining low despite the various ablation techniques. 4 Observations from surgical ablation of the left atrial (LA) posterior wall (PW) have implicated the region lying between the PVs in patients with persAF, the so-called PV component 5 (Figure 1). This is plausible as persAF is generally not considered a triggered dysrhythmia but a substrate-based arrhythmia. 6 The original Cox maze procedure as well as all its current iterations, which encompass the isolation of this region of the LAPW, have reported improved efficacy even in those with long-standing persAF. 7,8 Likewise, recent catheter ablation studies [9][10][11][12][13][14][15] have found similar benefits in patients with persAF associated with PW isolation (PWI) within the region of the PV component. Anatomically, this region is defined by the LA roof, superiorly, the left and the right PVs, laterally, and the plane extending from the lower borders of the left and the right inferior PVs inferiorly. 5 There is clinical, anatomic, and electrophysiologic evidence to suggest that this region of the LAPW may contribute to the genesis and maintenance of AF, particularly in those with per-sAF. Moreover, targeting this region of the LA inherently represents an extended form of wide-area antral PVI, which too has been shown to be superior to a distal/ostial PVI strategy. 16 Anatomic and embryologic evidence supporting catheter ablation of the pulmonary venous component A visual examination of the PV component and the orientation of its myofibrils suggests direct continuity between this region and the PV antra, as does a gross anatomical assessment of certain LA morphologies (Figure 2), further supporting the notion proposed by some that the PV component may in fact be a direct extension or a constituent of the PV tissue. 5,18 The PV component forms the LA dome with the superior PVs situated anteriorly and the inferior PVs more posteriorly. Thus, what is commonly referred to by many as the LAPW, in reality, represents the PV component, whereas the true PW extends from the lower borders of the inferior PVs to the superior margin of the vestibule-that is, the region surrounding the mitral valve orifice. 5 The PV component shares a common primordial origin with the PVs. 5 The embryologic origin of the four PVs and the PV component can be traced back to the mediastinal myocardium derived from a mid-pharyngeal strand at six weeks of gestation. 19 As such, the cellular lineage of the PVs and the PV component is entirely different from that of the remainder of the myocardium, which originates from the primary heart tube and the systemic venous tissue. 5,19 Electrophysiologic evidence supporting catheter ablation of the pulmonary venous component As discussed, the PV component is derived from tissues other than the primitive cardiac tube. 19,20 As such, it is believed to be related more to the PVs than the atrial tissue. Some studies have suggested that the cells from these tissues share more in common with the sinoatrial nodal myocytes, for instance, displaying higher diastolic calcium contents and propensity to spontaneous depolarization. 21 Moreover, the rationale for ablation of the PV component is further supported by clinical and cellular data, which evoke that this region of the LAPW exhibits significant remodeling believed to play a critical role in the pathogenesis of persAF, 22 as well as conduction abnormalities, 23 a higher incidence of delayed afterdepolarizations and larger late sodium and intracellular and sarcoplasmic reticulum Ca ++ contents, but smaller inward rectifier potassium currents 24,25 and a reduced resting membrane potential. 25,26 The LAPW and the PV myocytes exhibit shorter action-potential durations and slower phase 0 upstroke velocities. 26 Furthermore, the increased duration of AF observed in those with persAF is characterized by reduced wave conduction velocities and complex local activation within the PV component. 27 The region of the PV component is also believed to be the site of potential collision of activation wavefronts as they sweep across the LA dome. 5 Anatomically, there seems to be significant heterogeneity in myocardial fiber orientation within the PV antra and the LAPW, creating nonuniform anisotropic conduction favoring local reentry. 28 Furthermore, the PV component seems to exhibit greater conduction heterogeneity and anisotropy in patients with persAF as compared to those in sinus rhythm. Mapping of fibrillatory waves during cardiac surgery in patients with persAF has revealed simultaneous propagation of longitudinally dissociated fibrillation waves which are separated by continuously changing lines of block. 4 These lines of block are most densely packed within the PV component, leading to the highest degree of block and dissociation and the lowest incidence of   high-frequency driving rotors to persist; and (4) fibrosis that serves as a unidirectional block, which favors rotor formation and acts as an anchor for rotor maintenance. 31 Similarly, mapping in humans often localizes stable rotors or focal sources 32 as well as complex fractionated electrograms 33 in the LAPW and the roof. The PV component has in fact been shown to be a common source of triggers, accounting for approximately 40% of non-PV triggers in patients with AF. 34 The PV component is also the site of the main autonomic ganglionic plexi related to the LA dome (ie, the superior LA ganglionated plexus), which is believed to modulate extrinsic cardiac innervation and facilitate the occurrence of AF in a hyperactive autonomic state. 5,28 Thus, it is believed that catheter ablation of the PV component may greatly attenuate the input of these plexi to the PVs and interrupt the vagosympathetic input to the ligament of Marshall and the inferior left ganglionated plexus, which have also been highly implicated in the pathogenesis of AF. 5,28 In addition to these findings, an elevated LA pressure is believed to disproportionately affect the PV component. Prior studies have implicated acute, rapid LA activation in stretch-related AF with the level of spatiotemporal organization correlating with the degree of pressure elevation. 35 Moreover, areas of the LA exposed to the highest wall stress, particularly those around the PV antra and the PV component, have been correlated with low voltage and electrical scarring in prior mapping studies. 36 Accordingly, pathologic studies have similarly found increased fibrosis within the PV component of the LA and the LAPW of patients with chronic AF and mitral valve disease. 23

Available techniques for the ablation of the pulmonary venous component
A variety of minimally invasive strategies have been employed for ablation/isolation of the LAPW and the PV component. One method involves surgical ablation using endoscopically delivered epicardial lesions ( Figure 3A). However, this typically warrants an electrophysiology study, either concurrently or during follow-up, to carefully examine the LA voltage and/or to complete PVI + PWI, hence, commonly known as a hybrid approach. 37 Several catheter-based methods have also been described.
One technique involves creating large antral lesion sets to include the PV component versus creating a separate LAPW "box" lesion set using point-by-point radiofrequency ablation ( Figure 3B). Though feasible, 38 the theoretical advantages of this approach are sometimes offset by the challenges of creating uninterrupted, linear LAPW radiofrequency lesions. 39,40 As such, in many instances, In the first-generation cryoballoon, the refrigerant distribution comes from four jets placed proximally, yielding a maximal cooling zone that consists of an equatorial band around the balloon's circumference. Accordingly, optimal balloon positioning does not necessarily equate to optimal tissue contact with the balloon's maximal cooling zone (arrow). B: In the design of the current-generation cryoballoons, the number of injection ports has been increased to eight and they are positioned more distally, resulting in a larger and more uniform zone of freezing. This in turn expands the maximal cooling zone to the entire distal half of the balloon surface. As such, proper balloon alignment can ensure optimal tissue contact with the balloon's maximal cooling zone (arrow). LSPV: left superior pulmonary vein; RIPV: right inferior pulmonary vein; RSPV: right superior pulmonary vein. Figure 6: Non-PV occlusive cryoballoon maneuvers for isolation of large-sized PVs. A: A cardiac computed tomography image demonstrating a left common PV with an ostium measuring 36 mm in diameter. B: When approaching such a vein using a 28-mm cryoballoon, the operator should target the PV with a series of nonocclusive applications to avoid a distal-level (ostial) PVI. In this manner, the PV is ablated antrally in a segmental fashion, as shown in steps 1 to 3. A similar strategy can also be adopted in general when treating large-sized PVs, even in the absence of a "true" common PV ostium. LCPV: large-ostia left common pulmonary vein; RIPV: right inferior pulmonary vein; RSPV: right superior pulmonary vein.

A B
additional applications inside the "box" are required to complete PWI. Another approach involves directly ablating the PW/PV component in conjunction with circumferential PVI using point-by-point radiofrequency or cryoballoon ablation ( Figure 3C). The latter can result in significant LA debulking (Figure 4), and seems to offer consistent and incremental clinical benefits in patients with persAF. 11,12 Meanwhile, prior studies have shown that isolating the PV component not only does not compromise the contractile function of the LA 41-43 but, in fact, this approach is associated with LA size reduction and reverse remodeling. 44

Rationale for cryoballoon ablation of the left atrial posterior wall and the pulmonary venous component
Ablation lesions created using the currently available cryoballoons (second-generation or later) are typically large [45][46][47] and durable. 48 As such, these characteristics make these catheters potentially an attractive tool for performing LAPW ablation. Moreover, some have considered the cryoballoon a safer means for this approach, particularly with regard to certain collateral structures such as the esophagus. [49][50][51] Despite the weak level of evidence, there are limited data available from a handful of studies which have examined the effects of cryoballoon versus other ablation modalities on esophageal injury. 50,51 Based on indirect comparisons, the incidence of esophageal injury/ulceration following AF ablation as confirmed by endoscopy may be as high as 48% to 60% with radiofrequency, 52-54 but typically only half as prevalent (0%-22%) with cryoballoon ablation. [55][56][57][58] A comparative study 50 directly evaluating the safety of radiofrequency versus cryoablation on the calf esophagus found that esophageal lesion width and volume were significantly larger with radiofrequency versus cryoablation at 7 days. These authors also detected significant histological differences between the two modalities, including a higher incidence of partial-and full-thickness esophageal ulcerations with radiofrequency versus cryoablation. Furthermore, an interesting study by Cai et al. 59 recently discovered a correlation between esophageal contraction and cryoenergy, suggesting perhaps that this phenomenon might in itself serve as a protective mechanism to the esophagus during cryoablation. But before recognizing the cryoballoon as a suitable tool for LAPW ablation, it is important to first review the rationale for this approach as well as the biophysics of cryoballoon ablation.

The level and extent of posterior vein isolation using cryoballoon ablation
The level of PVI achieved using cryoballoon ablation has been the subject of controversy. Reddy et al. 60 were the first to investigate the level of PVI achieved during cryoballoon ablation of AF. In their study, they examined the precise location of the lesions created using a 23-mm first-generation cryoballoon (Arctic Front; Medtronic, Minneapolis, MN, USA) in a cohort of patients using three-dimensional (3D) electroanatomic mapping. These authors found that isolation had occurred predominantly at the level of the PV ostia, whereas the PV antra were left largely intact/unablated. In a subsequent study, Chierchia et al. 61 also evaluated the extent and level of PVI in a similar fashion using both 23-and the 28-mm first-generation cryoballoons. They noted that PVI occurred more proximally using the latter, resulting in significantly larger areas of atrial tissue ablation using the 28-mm (40.2% ± 3.9%) versus the 23-mm (20.7% ± 2.8%) balloon. Most recently, Kenigsberg et al. 62 reported a more contemporary experience using exclusively a 28-mm second-generation cryoballoon (Arctic Front Advance; Medtronic). These authors found that the level of PVI was much wider and more antral than previously reported. Moreover, they noticed partial isolation and debulking of the LAPW by approximately 70% in a cohort of patients who largely exhibited paroxysmal AF and only mild-to-moderate LA enlargement.

Cryoballoon PVI and Posterior Wall Isolation
The Journal of Innovations in Cardiac Rhythm Management, August 2021 While it is clear that the differences between the study by Reddy et al. 60 and the more recent experiences, in part, stem from the use of the smaller 23-versus the 28-mm cryoballoon, there is little doubt that the observed differences are also attributed to the design improvements of the current-generation cryoballoons. The incorporation of additional injection ports in the design of the second-/later-generation cryoballoons ( Figure 5) has led to more optimal and homogenous delivery of cryothermal energy, in turn yielding significantly larger areas of tissue ablation. Not surprisingly, this has been associated with improvements in procedural and clinical efficacy. 63,64 However, it remains unclear whether the observed clinical improvements are driven by the enhanced ability to create continuous and durable PVI or related to the larger and wider antral areas of circumferential ablation, or perhaps even both.
A study 45 comparing lesion characteristics and clinical outcomes associated with catheter ablation of AF using the hot balloon (SATAKE HotBalloon; Toray Industries, Tokyo, Japan) versus the current-generation cryoballoon (Arctic Front Advance; Medtronic, Inc) in 71 consecutive patients found that lesions created using the cryoballoon were significantly larger (38.2 ± 12.1 cm 2 vs. 24.3 ± 8.0 cm 2 ). Consistent with this observation, fewer instances of focal touch-up radio frequency ablation were required with cryoballoon (31% vs. 53%). Similarly, Perrotta et al. 46 investigated the size of LA isolation following AF ablation using the laser balloon (HeartLight™; CardioFocus, Marlborough, MA, USA) versus the cryoballoon (Arctic Front Advance; Medtronic, Inc) and discovered that total (42 ± 15 vs. 57 ± 14 cm 2 ; p = 0.002) and antral (54% ± 10% vs. 65% ± 8%; p = 0.001) surface areas of isolation were both greater with the cryoballoon. Having said that, it should be mentioned that the laser balloon used in this study represented an earlier iteration of this technology and the second/third-generation ablation catheters are believed to provide improved tissue contact and compliance. As such, it remains unknown whether this could have negatively impacted the results of this study. Lastly, in a study comparing the outcomes of cryoballoon ablation again point-by-point, force-sensing radiofrequency, Okumura et al. 47 showed that the low-voltage areas created using cryoballoon were significantly greater in size and the unexcitable tissue along the ablation line was significantly wider than those created with force-sensing radiofrequency ablation (16.7 ± 5.1 mm vs. 5.3 ± 2.3 mm; p < 0.0001).

The rationale for nonocclusive cryoballoon ablation (NOCA)
Historically, cryoballoon ablation has been guided by PV occlusion. 65 Cryothermal effects on cardiac tissue range from reversible ion channel block (ie, electrical dormancy) to permanent cellular injury and death (ie, cellular nonviability). Although cardiac cells can be rendered electrically dormant at temperatures of +20°C to +25°C, it is generally accepted that temperatures between −20°C and −50°C create lethal effects on cardiac cells. 66,67 Takami et al. 68 previously demonstrated that the magnitude of PV occlusion using the cryoballoon was an important determinant of durable PVI in a canine model. Similarly, in a clinical study 69 investigating the procedural and biophysical predictors associated with durable PVI using the second-generation cryoballoon, PV occlusion was found to highly correlate with the significant markers of PVI durability (ie, time to PVI and thaw time). Nevertheless, PV occlusion in itself is likely not a requirement for creating optimal cryolesions. For instance, this is observed when targeting large-sized PVs (Figure 6), as in the case of common PV ostia using a segmental NOCA approach. 70,71 Although PV occlusion can likely augment the magnitude of the freeze, optimal tissue contact and not necessarily PV occlusion which indirectly implies the same, is believed to be the essential requirement for creating durable cryolesions. This notion is further supported by finite element modeling data. 72 Meanwhile, the author believes that there is an inherent drawback associated with PV occlusion using the currently available cryoballoons when targeting PVs in patients with persAF. Those with persAF frequently exhibit an enlarged LA and dilated PV antra. 73,74 Unlike point-by-point radiofrequency, a cryoablation approach guided by PV occlusion using a fixed-diameter cryoballoon (ie, 28 mm) does not permit the operator to adjust the approach accordingly to allow optimal, antral cryoballoon positioning and ablation of the larger PV antra encountered in these patients. Consequently, this approach is more likely to yield an ostiallevel PVI in patients with persAF. The author believes that this in part accounts for the reason for diminished success associated with PV occlusion-guided cryoballoon ablation in patients with persAF. 3 Consistent with this notion, Güler et al. 75 previously identified PV diameter along with other markers such as persAF and an enlarged LA as significant predictors of AF recurrence following cryoballoon PVI. Along the same lines, Li et al. 76 identified PV diameter as the strongest independent predictor of long-term AF recurrence following cryoballoon ablation. Furthermore, PV diameter emerged as a more powerful predictor than LA diameter, itself.

The safety of nonocclusive cryoballoon ablation (NOCA)
When considering a novel ablation technique, it is important to carefully consider the safety of the approach. Due to the close proximity of LAPW to the esophagus, one has to consider the possibility of increased risk of esophageal injury in conjunction with cryo applications required to achieve PWI. However, to date, no increased risk of atrioesophageal fistula has been detected with cryoballoon PWI. Furthermore, a small study recently evaluating the outcomes of cryoballoon PVI + PWI found that interruption of cryoapplications at a luminal esophageal temperature less than 15°C was associated with the absence of esophageal thermal lesions. 77 As for other potential complications, the only other risk specifically associated with cryoballoon ablation has to do with phrenic nerve injury. It should be emphasized that NOCA or antral cryoballoon applications tend to actually mitigate this risk by avoiding distal placement of the cryoballoon into the PV  ostia. 44 Indeed, this is one of the main reasons behind a higher rate of phrenic nerve injury using balloon-based versus point-by-point radiofrequency ablation strategies.

The clinical experience with cryoballoon pulmonary vein and posterior wall isolation
Using a series of overlapping nonocclusive applications, the cryoballoon offers a simple and facile approach for directly ablating and debulking the LAPW to achieve PVI + PWI (Figure 7). Moreover, this approach can be greatly facilitated through the use of intracardiac echocardio graphy and 3D image integration (Figure 8).
A multicenter, retrospective study previously analyzed the outcomes of PVI + PWI using the second-generation cryoballoon in 222 consecutive patients with symptomatic persAF versus PVI alone. 12 This study found that acute PWI using this method was feasible in more than twothirds of the patients without sequelae, whereas adjunct radiofrequency ablation was required to complete this endpoint in the remaining patients ( Figure 9). Moreover, as compared with PVI alone, PVI + PWI yielded significantly greater LAPW (77.2% vs. 40.6%; p < 0.001) and total LA (53.3% vs. 36.3%; p < 0.001) isolation, as well as a higher incidence of AF termination and/or conversion to atrial flutters ( Figure 10). Adverse events were similar in both groups, whereas recurrence of AF and all atrial arrhythmias was significantly reduced with PVI + PWI at 12 months of follow-up (AF recurrence reduced by ~20%). Moreover, in a Cox regression analysis, PVI + PWI emerged as a significant predictor of freedom from recurrent atrial arrhythmias (hazard ratio: 2.04, 95% confidence interval: 1.15-3.61; p = 0.015). Similar results have been reported in subsequent, retrospective, single-center analyses. For instance, Nishimura et al. 13 studied the safety and feasibility of cryoballoon PVI + PWI as compared to PVI alone in 100 consecutive patients and found that the former was associated with significantly higher rates of sinus rhythm maintenance at 1 year (80.0% vs. 55.1%; p = 0.01). Nordsieck et al. 14 also investigated the outcomes of cryoballoon PVI + PWI versus conventional radiofrequency ablation as well as a hybrid surgical approach. These authors found that cryoballoon PVI + PWI was associated with decreased AF recurrence and need for repeat procedures as compared to conventional radiofrequency ablation. Furthermore, cryoballoon PVI + PWI was found to be superior to a hybrid surgical approach with regard to procedural safety and hospital length of stay. Moreover, Iacopino et al. 15

Cryoballoon PVI and Posterior Wall Isolation
The Journal of Innovations in Cardiac Rhythm Management, August 2021 value of cryoballoon PVI + PWI in a cohort of patients who underwent a redo procedure for recurrent persAF. They found that this approach was safe and feasible and associated with an 85% freedom from recurrent arrhythmias at 1 year.
In a more recent study, 78 the durability of PVI + PWI using the 28-mm cryoballoon was investigated in 81 consecutive patients referred to undergo repeat ablation for arrhythmia recurrence. The analysis showed durable PWI in 67 of 81 patients (82.7%) during 18 ± 4 months of follow-up. It also found that LA diameter (measured in parasternal long-axis view) represented a significant predictor for the need for adjunct radiofrequency ablation, particularly an LA diameter greater than 48 mm. Additionally, those with LAPW reconnection exhibited larger LA diameters, such that, none of the patients with PW reconnection exhibited an LA diameter measuring less than 45 mm, whereas 29% demonstrated an LA diameter of 45 to 48 mm and 71% an LA diameter greater than 48 mm (negative predictive value: 89.7%) (Figure 11). An atypical LAPW/ roof flutter represented the third most common cause of arrhythmia recurrence and virtually every patient with LAPW reconnection exhibited such an arrhythmia (Figure 12). In a more recent multicenter, prospective, randomized-controlled trial ( ClinicalTrials. gov #NCT03057548), once again a 20% reduction in AF recurrence was observed at 12 months of follow-up with PVI + PWI as compared to PVI alone (25.5% vs. 45.5%; p = 0.028), in patients with symptomatic per-sAF. 44 Adverse events were similar, but approximately 45% required adjunct radiofrequency ablation to complete PWI. PVI + PWI was also associated with fewer intraprocedural cardioversions. Furthermore, PVI + PWI yielded a much greater reduction in AF recurrence in patients with versus without LAPW low-voltage areas recorded at baseline (11.8% vs. 35.6%; p < 0.05). Lastly, PVI + PWI emerged as a significant predictor of freedom from recurrent AF (odds ratio: 3.67, 95% confidence interval: 1.44-9.34; p = 0.006). The validity of these studies and their findings will be ultimately tested on a larger scale in the PIVoTAL-IDE (LAPW and PV Isolation Using Cryoballoon for Treatment of persAF) trial (ClinicalTrials.gov #NCT04505163). This study, 79 which is currently underway, is a large, prospective, multicenter, randomized-controlled IDE trial (G190171) with the aim to precisely evaluate the acute and long-term outcomes of PVI + PWI versus PVI alone using the cryoballoon in patients with symptomatic persAF. Undoubtedly, future modifications and evolutions in catheter design, including larger balloons and non-fixed/variable diameters, will help to significantly facilitate and enhance the NOCA approach to catheter ablation and isolation of PVs and extra-PV structures.