Challenging Achievement of Bidirectional Block After Linear Ablation Affects the Rhythm Outcome in Patients With Persistent Atrial Fibrillation

The study protocol adhered to the principles of the Declaration of Helsinki and was approved by the Institutional Review Board at Yonsei University Health System. All patients provided written informed consent for inclusion in the Yonsei AF Ablation Cohort Database (ClinicalTrials.gov Identifier: NCT02138695). Among 1793 patients with AF who underwent catheter ablation, the study population included 398 patients with PeAF (75.6% male; age, 59.8±10.3 years) who underwent a catheter ablation procedure with a consistent ablation protocol of the Dallas lesion set: CPVI; cavotricuspid isthmus (CTI) ablation; posterior wall box lesion (roof line [RL] and posterior inferior line [PIL]); and anterior line (AL).⁶ Exclusion criteria were as follows: (1) permanent AF refractory to electrical cardioversion; (2) AF with valvular disease >grade 2; (3) any form of cardiomyopathy with uncontrolled heart failure or a left ventricular ejection fraction of <45%; (4) congenital heart disease; (5) history of cardiac surgery; and (6) previous ablation procedure. Before all ablation procedures, the anatomy of the LA and PV was visually defined on three‐dimensional (3D) computed tomography (CT) scans (64 Channel, Light Speed Volume CT; Brilliance 63; Philips, Best, The Netherlands). All antiarrhythmic drugs were discontinued for a period of at least 5 half‐lives, and amiodarone was stopped at least 4 weeks before the procedure.

Intracardiac electrograms were recorded using the Prucka CardioLab Electrophysiology system (General Electric Medical Systems, Inc., Milwaukee, WI), and radiofrequency (RF) catheter ablation (RFCA) was performed in all patients using 3D electroanatomical mapping (NavX, St. Jude Medical, Inc., Minnetonka, MN) merged with 3D spiral CT. Double trans‐septal punctures were made and multiview pulmonary venograms were obtained. After securing trans‐septal access, a circumferential PV‐mapping catheter (Lasso; Biosense‐Webster Inc., Diamond Bar, CA) was introduced with a long sheath (Schwartz left 1; St. Jude Medical). Systemic anticoagulation was performed with intravenous heparin to maintain an activated clotting time of 350 to 400 seconds during the procedure. For electroanatomical mapping, 3D geometry of both the left atrium (LA) and PV was generated using the NavX system and then merged with 3D spiral CT images.

Details of the RFCA technique and strategy used in our center are described in our previous study.¹¹ Briefly, we used an open irrigated‐tip catheter (Coolflex; 25–35 W, irrigation rate of 10–15 mL/min; St. Jude Medical) to deliver RF energy for ablation. All patients initially underwent CPVI and CTI ablation. RL, PIL, and AL⁶ were added as the standard lesion set, also known as the “Dallas lesion set” (Figure 1A). To generate the posterior box lesion, linear ablations of RL and PIL were made by connecting both sides of the CPVI at the top and bottom levels, respectively. AL was generated by ablation from the mitral annulus at the 12 o'clock position toward the LA RL.⁶ If atrial tachyarrhythmias could not be terminated by standard lesion set ablation, internal cardioversion and evaluation of BDB state were performed. BDB of RL was confirmed by differential pacing from LA appendage versus LA posterior wall (LAPW) and successful generation of PIL was considered to be achievement of LAPW isolation, which was defined as no endocardial electrogram in the LAPW with a setting of RL block and no capture of isolated LAPW pacing (Figure 1A). BDB of AL was confirmed by differential pacing from LA appendage versus LA septum.⁶ When BDB of linear ablation lines was not achieved, additional ablations were performed to achieve BDB of these lines. However, if BDB could not be achieved after 3 attempts of linear ablation, those lines were kept unblocked to avoid collateral damage. The operators could opt to perform additional ablations in the superior vena cava or non‐PV foci, or conduct complex fractionated electrograms¹² at their discretion. If there were mappable AF triggers or atrial premature beats with isoproterenol infusion (5 μg/min), we carefully mapped and ablated those non‐PV foci as much as possible. All RFCA procedures were conducted according to the above specific protocol by 2 operators with more than 10 years of experience.

Among 398 patients, 95 (23.9%) maintained antiarrhythmic medication before AF recurrence because of a high chance of recurrence with frequent atrial premature beats or short runs of nonsustained atrial tachycardia (AT). Patients visited the outpatient clinic regularly at 1, 3, 6, and 12 months post‐RFCA and then every 6 months thereafter or whenever they experienced symptoms. All patients underwent electrocardiography (ECG) during every visit and 24‐hour Holter recording at 3 and 6 months and then every 6 months, in accord with the 2012 HRS/EHRA/ECAS Expert Consensus Statement guidelines.¹³ However, patients reporting symptoms of palpitations underwent Holter monitor or event monitor recordings and were evaluated for the possibility of arrhythmia recurrence. The primary endpoint was the clinical recurrence of atrial tachyarrhythmia as any episode of AF or AT lasting for at least 30 seconds in duration.¹³ Any ECG documentation of AF/AT recurrence after 3 months of the blanking period was diagnosed as clinical recurrence.¹³ However, AF/AT recurrence in the first 3 months after catheter ablation (blanking period) was counted as early recurrence. Early recurrence was neither classified as clinical AF/AT recurrence nor used in all data analyses.

Statistical analysis was performed using SPSS (Statistical Package for Social Sciences, Chicago, IL) software for Windows (version 20.0). Continuous variables were expressed as mean± SD and were compared using the Student t test. Categorical variables were reported as frequency (percentage) and compared using chi‐square test and Fisher's exact test. Kaplan–Meier analyses with log‐rank tests were used to calculate AF recurrence‐free survival over time and compare recurrence rates across groups. Logistic regression analyses were used to identify predictors of LAPW isolation, and Cox regression analyses were used to assess independent predictors of AF and/or AT recurrence after RFCA. Age, sex, and all other variables with P value of less than 0.1 in the univariate analyses were selected for the multivariate analyses. Once AF/AT recurrence was documented by ECG after 3 months of blanking period, we considered those patients as clinical recurrence of AT/AF. Therefore, the repeated measurements of ECG or Holter recording after clinical recurrence event were not used in statistical analyses. The duration between catheter ablation and first documentation of AF/AT recurrence after 3 months of blanking period was calculated as month and used in Kaplan–Meier analysis and Cox‐regression analysis. A P<0.05 (two‐sided) was considered to be statistically significant.

Baseline characteristics of overall study population with respect to the presence or absence of clinical AF/AT recurrence are shown in Table 1. There were no differences in baseline characteristics between patients with clinical AF/AT recurrence and those without.

Table 2 summarizes the procedural characteristics. In 398 patients with PeAF who achieved 100% CPVI, BDB rates of additional linear ablations were 100% in CTI, 84.7% in RL, 44.7% in PIL (confirmed by LAPW isolation in patients with BDB of RL), and 63.6% in AL. LAPW isolation has a same meaning to achievement of both RL and PIL blocks, and no patient showed PIL block without RL block. There were no differences in the rate of achievement of BDBs for CTI, RL, PIL, and AL between patients with clinical recurrence and those without. Patients with clinical AF/AT recurrence showed longer total procedural time than those without clinical AF/AT recurrence (234.67±56.21 vs 210.70±47.86 minutes; P<0.001), but no difference in ablation time (6424.69±1594.93 vs 6345.24±1217.80 seconds; P=0.586).

Table 3 shows logistic regression analysis of predictors for successful BDB of additional linear ablation lines. There were no significant predictors for BDB achievement of RL and PIL. Older age (odds ratio [OR], 1.03; 95% CI, 1.01–1.05; P=0.016) and greater LA dimension in echocardiography (OR, 1.04; 95% CI, 1.01–1.08; P=0.025) were independently associated with BDB of AL.

During 29.0±18.4 months of follow‐up, 126 of 398 patients (31.7%) experienced clinical AF/AT recurrence. Kaplan–Meier analysis according to BDB of RL (log‐rank, P=0.178; Figure 2A) or AL (log‐rank, P=0.764; Figure 2C) showed no significant benefit for clinical outcome after catheter ablation. However, patients with LAPW isolation showed a significantly lower clinical recurrence rate compared with those without LAPW isolation (log‐rank, P=0.017; Figure 2B). Patients with a larger number of BDB‐achieving linear ablation lines showed no significant benefit compared with those with fewer BDB‐achieving linear lines (2–3 vs 0–1; log‐rank, P=0.073; Figure 2D). In multivariate Cox regression analysis, achievement of LAPW isolation was independently associated with lower clinical recurrence of AF/AT after catheter ablation (hazard ratio [HR], 0.68; 95% CI, 0.47–0.98; P=0.041; Table 4).

Among 126 patients with AF recurrence, 52 (41.3%) underwent repeat procedures at 23.02±16.05 months after the de novo procedure. Achievement rates of LAPW isolation and BDB for linear lines at de novo procedure, redo mapping, and postrepeat ablation are presented in Figure 3, respectively. BDB rates for CTI, RL, LAPW, and AL were 100%, 90.4%, 55.8%, and 65.4% at the de novo procedure, and 94.2%, 63.5%, 34.6%, and 40.4% at the redo mapping, respectively; thus, the BDB maintenance rates for CTI, RL, PIL, and AL were 94.2% (49 of 52), 63.5% (33 of 47), 62.1% (18 of 29), and 61.8% (21 of 34), respectively. After redo procedures, BDB was finally achieved in 100% for CTI, 94.2% for RL, 65.4% for PIL, and 76.9% for AL (Figure 3A). Among patients who achieved LAPW isolation at the de novo procedure (n=29), achievement rates of BDB for linear lines are presented in Figure 3B.

Linear lesion generation was initially suggested and attempted in the surgical treatment of AF, and Cox et al demonstrated the feasibility of curing AF by biatrial linear lesion generation, with a success rate of 93%.¹⁴ Linear ablation in catheter ablation is derived from the concept underlying surgical treatment of AF and modifies atrial substrates and compartmentalizes the atrium into smaller regions to reduce the critical mass of atrial tissue.¹⁵ However, complete conduction block with transmural lesion generation remains challenging with catheter ablation, and durability of conduction block is also an important issue for long‐term successful AF ablation. Even in AF surgery, transmurality of the lesions is reported to be achievable in only up to 75% of cases.¹⁶ Although Rostock et al reported conduction recovery in 90% of cases for mitral isthmus line and 79% for RL at the repeat procedure,¹⁷ there are a lack of data on the initial achievement and durability of BDB rate of linear ablation lesions. In this study, the overall de novo BDB rate for linear ablations, including RL, PIL, and AL, was 64.3%, and only 74.2% of previously confirmed BDBs were maintained at repeat procedure. Although we used a consistent ablation protocol in this study, the patients in whom BDB can be achieved might be somewhat different from others. Recently, Kim et al reported a lower reconnection of previously isolated PV potential in women than in men on the second procedure.¹⁸ It is possible that genetic or intrinsic sex difference results in a thinner LA myocardial layer in women than in men.¹⁹ Therefore, BDB achievement after linear ablation can be affected by myocardial wall thickness, lesion length, or extent of low‐voltage scar tissue. In this study, we found a higher BDB rate of the PIL in females and a smaller LA dimension on univariate logistic regression analyses. On the other hand, old age and a larger LA dimension were independently associated with BDB of the AL, suggesting that LA with advanced remodeling and fibrosis also contributes to the achievement of BDB.

In this study, we demonstrated that achievement of LAPW isolation by additional linear ablations of RL and PIL was associated with a better clinical outcome. Although Tamborero et al²⁰ reported no beneficial effect of LAPW isolation with linear ablations of RL and PIL, their study group consisted of 60% paroxysmal AF cases. Like PVs, LAPW embryologically originates from the cells of the primordial PV²¹ and thus might contribute to the initiation of AF with triggered activity²² and also have a role in perpetuation of AF.^{23, 24} LAPW myocytes are known to have higher arrhythmogenic potential than other parts of the LA because of larger intracellular Ca²⁺ transients, more sarcoplasmic reticulum calcium content, and less protein expression of the Na‐Ca exchanger,²⁵ as shown in a representative example of recurred AT in Figure 1B. In this patient, LAPW isolation was achieved at de novo procedure, but recurred focal AT originated from LAPW conducted through the reconnected RL. Moreover, LA size is an important determinant of AF perpetuation, and therefore LA mass reduction by LAPW isolation might modify the AF substrate and have a beneficial effect on the clinical outcome of catheter ablation—the so‐called critical mass hypothesis.²⁶

The standard lesion set of additional linear ablation post‐CPVI has not yet been clearly defined. Although the most commonly added linear lesions are the mitral isthmus and RL,^{4, 5} some researchers also suggested a lesion set in which LAPW is isolated by performing RL and PIL.^{7, 27} Furthermore, the surgical ablation lesion set reported by Edgerton et al, which is known as the Dallas lesion set, can be generated successfully with the catheter ablation technique.^{6, 28} In this study, we performed Dallas lesion set additional linear ablation with RL and PIL for LAPW isolation in nearly 400 patients with more than 2 years of follow‐up. Our results indicated that although BDB of PIL and consequent LAPW isolation is difficult to achieve compared to RL or AL, it seems to be worth the effort.

Although several researchers recently suggested the beneficial effect of LAPW isolation on clinical outcome of PeAF catheter ablation,^{29, 30} the results are inconsistent²⁰ and the studies used different ablation strategies for LAPW isolation. Whereas our group and others performed additional linear ablation with RL and PIL for LAPW isolation,^{20, 29} Bai et al conducted LAPW isolation with extensive ablation of the entire LAPW and reported an additional benefit over CPVI alone in the PeAF ablation procedure.³⁰ Furthermore, they verified CPVI and LAPW isolation by performing a repeat procedure regardless of arrhythmia recurrence, and therefore assessed the true impact of LAPW isolation on clinical outcome.

If complete, LAPW with linear lesions seems to have a critical mass reduction effect with LA compartmentalization and fewer ablation lesions with less myocardial injury. However, BDB of PIL is hard to achieve and there are safety concerns over collateral damage, including esophagus injury. Although there were no differences in the major complication rates and ablation‐related complication rates between patients with LAPW isolation and those without (Table S1), there was a case of atrioesophageal fistula in each group. The overall incidence of atrioesophageal fistula was 0.098% (2 of 2047) in the Yonsei AF Ablation cohort. Therefore, further studies comparing the safety and complications related to these procedures are warranted. Moreover, incomplete BDB with gap formation might be rather proarrhythmic. However, there were no differences in the recurrence types of tachycardia (AF/AT; Table S2) or AT recurrence rates (Figure S1) between patients with BDB and those without in this study, and an extensive LAPW ablation strategy could result in more myocardial injury and, consequently, might exhibit arrhythmogenicity.^{31, 32} Further prospective, randomized studies with different ablation strategies for LAPW isolation as endpoints are warranted.

This study was an observational cohort study from a single center. Although we confirmed BDB of all linear lesions, some lesions might have conduction recovery, which may have affected the clinical outcome. Whether complete LAPW isolation has a significant beneficial effect on clinical outcome or incomplete BDB has a proarrhythmic effect remain to be elucidated.