Radiofrequency catheter ablation of persistent atrial fibrillation by pulmonary vein isolation with or without left atrial posterior wall isolation: long-term outcomes of the CAPLA trial

Abstract

Background and Aims

Posterior wall isolation (PWI) is commonly incorporated into catheter ablation (CA) strategies for persistent atrial fibrillation (AF) in an attempt to improve outcomes. In the CAPLA randomized study, adjunctive PWI did not improve freedom from atrial arrhythmia at 12 months compared with pulmonary vein isolation (PVI) alone. Whether additional PWI reduces arrhythmia recurrence over the longer term remains unknown.

Methods

In this multi-centre, international, randomized study patients with persistent AF undergoing index CA using radiofrequency were randomized to PVI + PWI vs. PVI alone. Patients underwent regular follow-up including rhythm monitoring for a minimum of 3 years after CA. Atrial fibrillation burden at 3 years after ablation was evaluated with either 28-day continuous ambulatory electrocardiogram (ECG) monitoring, twice daily single-lead ECG or from cardiac implanted device. Evaluated endpoints included freedom from any documented atrial arrhythmia recurrence after a single procedure, AF burden, need for redo CA, rhythm at last clinical follow-up, healthcare utilization metrics, and AF-related quality of life.

Results

Three hundred thirty-three of 338 (98.5%) patients (mean age 64.3 ± 9.4 years, 23% female) completed 3-year follow-up, with 169 patients randomized to PVI + PWI and 164 patients to PVI alone. At a median of 3.62 years after index ablation, freedom from recurrent atrial arrhythmia occurred in 59 patients (35.5%) randomized to PVI + PWI vs. 68 patients (42.1%) randomized to PVI alone (hazard ratio 1.15, 95% confidence interval 0.88–1.51, P = .55). Median time to recurrent atrial arrhythmia was 0.53 years (interquartile range 0.34–1.01 years). Redo ablation was performed in 54 patients (32.0%) in the PVI + PWI group vs. 49 patients (29.9%, P = .68) in the PVI alone group. Pulmonary vein reconnection was present in 54.5% (mean number of reconnected PVs 2.2 ± .9) and posterior wall reconnection in 75%. Median AF burden at 3 years was 0% in both groups (interquartile range 0%–0.85% PVI + PWI vs. 0%–1.43% PVI alone, P = .49). Sinus rhythm at final clinical follow-up was present in 85.1% with PVI + PWI vs. 87.1% with PVI alone (P = .60). Mean AF Effect On Quality-Of-Life (AFEQT) score at 3 years after ablation was 88.0 ± 14.8 with PVI + PWI vs. 88.9 ± 15.4 with PVI alone (P = .63).

Conclusions

In patients with persistent AF, the addition of PWI to PVI alone at index radiofrequency CA did not significantly improve freedom from atrial arrhythmia recurrence at long-term follow-up. Median AF burden remains low and AF quality of life high at 3 years with either ablation strategy.

Structured Graphical Abstract

Atrial fibrillation, Ablation, Posterior wall isolation, CAPLA

See the editorial comment for this article ‘Ablation of persistent atrial fibrillation: lessons learnt from the CAPLA trial', by J.N. Koneru and K.A. Ellenbogen, https://doi.org/10.1093/eurheartj/ehae832.

Introduction

Atrial fibrillation (AF) is the most prevalent sustained arrhythmia, affecting an estimated 60 million adults globally.¹ A rhythm control strategy targeting the maintenance of sinus rhythm has been shown to improve clinical outcomes, including reduced risk of stroke, heart failure hospitalization, and cardiovascular mortality.² Randomized trials demonstrate the superiority of catheter ablation (CA) compared with anti-arrhythmic drug (AAD) therapy with respect to AF prevention, healthcare utilization, and improvement in quality of life.^3–5 However, recurrent atrial arrhythmia after CA remains problematic, particularly in patients with persistent AF (PsAF) compared with paroxysmal AF (PAF).⁶

Various adjunctive ablation techniques beyond pulmonary vein isolation (PVI) alone have been investigated in the management of PsAF; however, no one strategy has been demonstrated to be superior in randomized studies.^7,8 Of these investigational techniques, the most widely adopted is left atrial posterior wall isolation (PWI). In the CAPLA randomized study, the addition of PWI to PVI using radiofrequency (RF) ablation did not improve freedom from recurrent atrial arrhythmia at 12 months after index ablation.⁹ However, RF-based PWI can be procedurally challenging owing to the presence of epicardial connections and risk of oesophageal thermal injury.¹⁰ The advent of pulsed field ablation (PFA) as a non-thermal ablation modality has renewed interest in PWI, with potential improvements in both procedural efficiency and safety. However, a recent registry study again demonstrated no improvement in AF-free survival with adjunctive PWI using PFA at 12 months after ablation.¹¹

Twelve-month follow-up may be insufficient to determine the impact of PWI, as patients with PsAF and more advanced atrial substrate continue to recur in the longer term compared with PAF.¹² Whether adjunctive PWI yields improved outcomes over longer follow-up periods remains unclear. We report findings from the CAPLA randomized trial with a minimum follow-up duration of 3 years.

Methods

Trial design

The CAPLA study protocol and institutional board review have been published in detail previously.^9,13 Briefly, CAPLA was an investigator-initiated, multi-centre randomized controlled trial conducted at 11 cardiac institutions across three countries (Australia, the UK, and Canada). This post hoc extended long-term follow-up analysis was approved by the Alfred Hospital Human Ethics and Research Committee (Melbourne, Australia). The purpose of the CAPLA study was to evaluate the impact of PVI + PWI compared with PVI alone in patients with PsAF undergoing first-time RF-based AF ablation.

Study cohort

We prospectively enrolled adults ≥ 18 years old with symptomatic PsAF refractory to at least one AAD. Persistent AF was defined as the presence of an electrocardiogram (ECG)-documented episode of AF with at least one continuous episode lasting between 7 days and 3 years. We excluded patients with PAF, very long-standing PsAF (continuous AF duration ≥ 3 years) and hypertrophic cardiomyopathy. Full inclusion and exclusion criteria are included in Supplementary data online, Methods S1.

Ablation procedures

The ablation strategy has been described previously.¹⁰ In brief, AF ablation was conducted under general anaesthesia using 3D electroanatomic multi-polar mapping and irrigated, contact force-sensing RF catheters. At index procedure, PVI was achieved through wide antral circumferential ablation around the pulmonary veins, targeting electrical isolation as the endpoint. For patients assigned to additional PWI, this was performed using a ‘box isolation’ approach with 25–40 W at operator’s discretion. This involved creating a left atrial roof line connecting the superior aspects of the superior pulmonary veins and a floor line linking the inferior margins of the inferior pulmonary veins (Figure 1). Additional ablation within the box was performed if linear ablation was not successful in achieving complete PWI. Successful PVI and PWI were confirmed by demonstrating entrance and exit blocks, dissociated posterior wall potentials, or the complete absence of local electrograms without capture at high output pacing. As per the original CAPLA protocol, clinicians and patients were instructed to cease AAD therapy by the completion of the 90-day blanking period after index ablation.

Figure 1

Pulmonary vein isolation alone vs. pulmonary vein isolation plus posterior wall isolation. PVI, pulmonary vein isolation; PWI, posterior wall isolation

At redo procedure, the approach conformed to the participants’ original treatment allocation arm. In the PVI only group, only PV re-isolation was permitted if pulmonary vein reconnection was present. The addition of PWI was permitted if AF recurred in the presence of enduring PVI. In the PVI + PWI group, only re-isolation was permitted for reconnection of pulmonary veins and/or posterior wall. If AF recurred in the presence of enduring PVI and PWI, then further ablation was at the discretion of the operator. This could include non-pulmonary vein trigger ablation, complex fractionated atrial electrogram ablation, or additional linear ablation.

Arrhythmia surveillance

Patients underwent regular clinical review for a minimum of 3 years after ablation. Patients who died prior to the 3-year time point were included in this analysis but censored at the time of their last clinical review. The first recurrence of documented atrial arrhythmia and rhythm at the most recent clinical follow-up were recorded. Outpatient arrhythmia surveillance was performed with one of three modalities: (i) twice-daily plus symptomatic transtelephonic ECG transmissions (TTM) using the Kardia mobile device; (ii) continuous monitoring from a cardiac implanted electronic device (CIED, including implantable loop recorders, pacemaker or implantable defibrillator); or (iii) 28-day continuous ambulatory ECG monitoring performed at the 3-year time point. For patients monitored with Kardia devices, AF burden was evaluated using all available TTM strips sent between 2.5 and 3.5 years after index ablation. The overall AF burden was then calculated as the average daily AF burden for the days where at least one TTM was sent. Data were deemed sufficient for the calculation of overall burden at 3 years if TTM strips across at least 28 days were available. For patients with CIED (either implantable loop recorders or transvenous devices with functioning atrial leads), AF burden was calculated using all available device interrogation reports between the 2.5- and 3.5-year time points. For patients who were non-compliant with Kardia ECG transmissions and without CIED, AF burden was assessed by 28-day continuous ambulatory ECG monitoring performed at the 3-year time point. Further details about the methodology used to calculate AF burden are included in Supplementary data online, Methods S1.

Healthcare utilization and quality of life

Healthcare utilization metrics were collected prospectively, including unplanned hospitalization events (including emergency department visits), direct current cardioversion (DCCV) procedures, and the need for redo ablation procedures for recurrent atrial arrhythmia.^14,15 Disease-specific quality of life was recorded at baseline, 1 year, and 3 years after ablation using the AF Effect On Quality-Of-Life (AFEQT) questionnaire.

Endpoint analyses

We evaluated freedom from documented atrial arrhythmia (AF, atrial tachycardia, or atrial flutter exceeding 30 s) with or without the use of AAD therapy at 3 years after single ablation procedure. Atrial arrhythmia recurrence occurring in the first 90 days after index ablation was blanked in line with previous studies.⁹ Additional endpoints included percentage AF burden, need for redo ablation procedures, AFEQT, healthcare utilization metrics, anti-arrhythmic use, cardiovascular mortality, and all-cause mortality, which were each assessed at 3 years after ablation. The full list of additional endpoints is included in Supplementary data online, Methods S1.

Statistical analysis

The power calculation for the CAPLA study has been described previously.¹³ The study cohort was analysed according to treatment allocation (PVI or PVI + PWI) on an intention-to-treat basis. Normally distributed continuous data were summarized as mean ± standard deviation and analysed by the Student’s t-test. Skewed continuous data are presented as median with interquartile range and were analysed using the Mann–Whitney U test. Categorical variables are presented as frequency (%) and were analysed using χ² test. A two-sided P-value of <.05 was considered statistically significant. The Bonferroni correction for three pairwise comparisons across monitoring strategy cohorts was used to determine the significance of intergroup differences with an adjusted alpha of .017. Time-to-event outcomes were analysed using Kaplan–Meier survival curves in an intention-to-treat analysis, utilizing the log-rank test with 1 degree of freedom. Hazard ratios (HRs) and 95% confidence intervals (CIs) for time-to-event outcomes were estimated using univariate Cox’s proportional hazards modelling. Statistical analyses were performed using SPSS v27 (IBM, Armonk, NY, USA).

Results

Clinical and procedural characteristics

Of the 338 study participants randomized as part of the CAPLA study, long-term follow-up data were available for 333 (98.5%), including 169 randomized to PVI + PWI and 164 randomized to PVI alone (Figure 2). Median long-term follow-up duration was 3.6 years (IQR 3.2–4.3 years) after index ablation. Baseline clinical characteristics were balanced between the study groups (Table 1). Mean age was 64.3 ± 9.4, and 77.2% were male sex. Median duration of the longest AF episode prior to index ablation was 5 months (IQR 2–8 months). Late mortality after ablation occurred in six patients.

Figure 2

Consort diagram. AF, atrial fibrillation; PVI, pulmonary vein isolation; PWI, posterior wall isolation

Table 1

Cohort characteristics

Baseline characteristic	PVI + PWI (n = 169)	PVI only (n = 164)	P-value
Demographics
Age (years)	64.4 (9.3)	64.1 (9.3)	.84
Male sex (n, %)	128 (75.7%)	129 (78.7%)	.53
BMI (kg/m2)	30.1 (5.3)	29.9 (5.3)	.80
Median longest AF duration (months)	6.6 (7.2)	7.1 (7.2)	.59
CHA2DS2-VASC score	2 (1–3)	2 (1–3)	.38
Number of previous cardioversions for AF	1.8 (1.5)	1.8 (1.5)	.88
Long-standing persistent AF (n, %)	29 (17.1%)	27 (16.4%)	.86
Comorbidities
Obesity (BMI > 27 kg/m2) (n, %)	111 (65.7%)	108 (65.9%)	.97
Hypertension (n, %)	85 (50.6%)	73 (44.5%)	.27
Congestive heart failure (n, %)	70 (41.7%)	71 (43.3%)	.76
Ischaemic heart disease (n, %)	85 (50.6%)	73 (44.5%)	.36
Type 2 diabetes (n, %)	17 (10.1%)	17 (10.4%)	.94
Previous stroke (n, %)	11 (6.5%)	7 (4.3%)	.36
Alcohol excess† (n, %)	18 (10.7%)	16 (9.8%)	.79
Echocardiographic parameters
Left ventricular ejection fraction (%)	52.6 (12.1)	51.9 (12.1)	.62
LV end-diastolic diameter (mm)	35.5 (8.4)	37.2 (8.4)	.14
Left atrial diameter (cm)	4.6 (.6)	4.4 (.6)	.16
Left atrial volume index (mL/m2)	49.5 (15.5)	45.4 (15.5)	.05
E/e′	10.3 (4.2)	10.2 (4.2)	.87
AAD therapy at time of enrolment
Flecainide (n, %)	29 (17.2%)	19 (11.6%)	.15
Sotalol (n, %)	45 (26.6%)	34 (20.7%)	.21
Amiodarone (n, %)	51 (30.2%)	55 (33.5%)	.51
Rhythm monitoring strategy for AF burden
Transtelephonic ECG transmissions	89 (59.3%)	79 (63.7%)	.67
28-day continuous ambulatory ECG	44 (29.3%)	24 (19.2%)	.04
Implanted cardiac device	16 (10.7%)	22 (17.8%)	.25

Normally distributed data expressed in mean (SD), non-normally distributed data expressed as median (IQR).

^†Alcohol excess defined as >8 standard drinks/week.

AAD, anti-arrhythmic drug/medication; AF, atrial fibrillation; BMI, body mass index; CHA2DS2-VASc, congestive heart failure, hypertension, age 75 years (doubled), diabetes, stroke/transient ischemic attack/thromboembolism (doubled), vascular disease (prior myocardial infarction, peripheral artery disease, or aortic plaque), age 65–75 years; PVI, pulmonary vein isolation; PWI, posterior wall isolation; SD, standard deviation.

Procedural characteristics at index ablation have been previously published⁹ and are summarized in Table 2.⁹ Of note, electrical isolation of the pulmonary veins was achieved in all patients across both groups, while PWI was acutely achieved in 146/169 (86.4%) patients in the PVI + PWI group. Two patients randomized to a PVI alone strategy received PVI + PWI at operator discretion and were analysed as part of the PVI alone cohort as part of the intention-to-treat analysis. There were no cases of procedural mortality, cerebrovascular events, or atrio-oesophageal fistula.

Table 2

Procedural characteristics

Index procedural characteristics	PVI + PWI (n = 169)	PVI only (n = 164)	P-value
Procedural strategy
Atrial arrhythmia at time of ablation (n, %)	107 (63.7%)	96 (58.9%)	.37
DCCV prior to ablation (n, %)	98 (58.0%)	89 (54.3%)	.49
High power, short duration ablationa (n, %)	50 (29.6%)	46 (28.0%)	.76
Total procedural time (min)	147.4 (47.5)	124.7 (47.5)	<.001*
Total RF time (min)	38.1 (16.2)	30.1 (16.2)	<.001*
Procedural success
Pulmonary vein isolation achieved	169 (100%)	164 (100%)	N/A
Posterior wall isolation achieved	146 (86.4%)	N/A	N/A
Ablation required ‘within the box’ to achieve isolation	89 (26.7%)	N/A	N/A
Procedural complications
All procedural complications	5 (3.0%)	4 (2.4%)	.96
Access site haematoma	1 (.5%)	0 (0%)	1.0
Inferior epigastric nerve injury	1 (.5%)	0 (0%)	1.0
Pericarditis	0 (0%)	1 (.5%)	.49
Phrenic nerve injury	0 (0%)	1 (.6%)	.49
Tamponade	1 (.5%)	0 (0%)	1.0
Congestive heart failure	2 (1.2%)	2 (1.2%)	1.0
Atrio-oesophageal fistula/ulcer	0 (0%)	0 (0%)	N/A

DCCV, direct current cardioversion; PVI, pulmonary vein isolation; PWI, posterior wall isolation; RF, radiofrequency ablation.

^aDefined as ablation power ≥40w on the posterior left atrial wall.

^*Bold value means statistically significant result.

Late mortality occurring remotely to the index ablation procedure occurred in six patients (three in each arm) at a mean of 1.5 ± 1.1 years after ablation. These were due to sudden cardiac death (n = 3), post-operative complications following cardiac surgery (n = 1), sepsis (n = 1), and traumatic intracranial haemorrhage (n = 1).

Arrhythmia recurrence

Endpoint analyses are shown in Table 3. At 3 years after ablation, freedom from documented recurrence of atrial arrhythmia (on or off AAD therapy after a single ablation procedure) occurred in 59 patients (35.5%) randomized to PVI + PWI vs. 68 patients (42.1%) randomized to PVI alone (HR 1.15, 95% CI .88–1.51, log-rank P = .55; Figure 3). Median time to recurrent atrial arrhythmia was 6.9 months (IQR 4.3–12.3 months) in the PVI + PWI group and 5.9 months (IQR 3.8–11.4 months) in the PVI alone group (P = .29).

Figure 3

Time to recurrent atrial arrhythmia. CI, confidence interval; HR, hazard ratio; PVI, pulmonary vein isolation; PWI, posterior wall isolation

Table 3

Outcome data at long-term follow-up

Outcome data	PVI + PWI (n = 169)	PVI only (n = 164)	P-value
Arrhythmia recurrence
Freedom from atrial arrhythmia at long-term follow-up (on or off AAD after a single ablation procedure) (n, %)	59 (35.5%)	68 (42.1%)	.55
Percentage AF burden at 3 years
Overall AF burden (median, IQR)	0% (0%–0.85%)	0% (0%–1.43%)	.49
AF burden in patients with documented AA recurrence (median, IQR)	0% (0%–4.1%)	0% (0%–5.2%)	.75
Sinus rhythm at final clinical follow-up	144 (85.1%)	143 (87.1%)	.60

AF, atrial fibrillation; IQR, interquartile range; N/A, not applicable; PV, pulmonary vein; PW, posterior wall; SD, standard deviation.

Atrial fibrillation burden

Sufficient rhythm data to calculate AF burden at 3 years were available in 274/333 patients (82.3%). Twice-daily plus symptomatic TTM recordings were utilized in 61.3%, with a total of 37 687 TTM ECG recordings analysed (mean of 238.5 recordings per patient). Ambulatory ECG monitoring for 28 days was used in 24.5% and implanted cardiac devices for continuous rhythm surveillance in 13.9%.

Median AF burden at long-term follow-up was 0% in both groups (IQR 0%–0.85% PVI + PWI vs. 0%–1.43% PVI alone, P = .49; Figure 4). Median AF burden was higher amongst patients monitored with CIED (0.2%, IQR 0%–100%) compared with those monitored with Kardia devices (0%, IQR 0%–1.71%) and 28-day continuous monitoring (0%, IQR 0–0.4%, P = .008). In the subset of patients with documented arrhythmia recurrence by the 3-year time point (206 patients, 61.2% of the cohort), the median AF burden at 3 years was also 0% (IQR 0%–4.1% PVI + PWI vs. 0%–5.2% PVI alone, P = .75). Sinus rhythm at the most recent clinical follow-up was present in 85.1% with PVI + PWI vs. 87.1% with PVI alone (P = .60).

Figure 4

Bee-swarm plot of percentage atrial fibrillation burden at 3 years after ablation. AF, atrial fibrillation; IQR, interquartile range; PVI, pulmonary vein isolation; PWI, posterior wall isolation

Redo ablation procedures

The outcomes of redo ablation are presented in Table 4. At long-term follow-up, redo procedures were performed in 104 patients (30.9%), including 54 (32.0%) with PVI + PWI compared with 49 (29.9%) with PVI alone (HR 0.94, 95% CI 0.63–1.42). The median time to redo the ablation procedure was 1.4 years (IQR 0.8–2.8 years). Two patients in each group underwent redo ablation via a surgical (epicardial) approach without electroanatomic mapping as part of cardiac surgery for coronary (n = 1) or valvular heart disease (n = 3). Of the patients undergoing redo endocardial procedures following initial PVI + PWI, posterior wall reconnection was present in 39 (75%). Pulmonary vein reconnection was identified in 54.5% patients overall, with an average of 2.2 ± 0.9 veins reconnected per patient. Rates of pulmonary vein reconnection were similar between groups (P = .60). At redo procedure, there were no patients in the PVI alone group who had inadvertent isolation of the posterior wall. Left atrial macro-reentrant tachycardia (LAMT) was observed in 21.2% in the PVI + PWI group compared with 12.8% in the PVI alone group (P = .27). Peri-mitral flutter accounted for the majority of LAMT, present in 17.3% of patients who initially received PVI + PWI compared with 10.6% of patients who received PVI alone (P = .34). The incidence of cavotricuspid isthmu0073-dependent flutter was similar between the two groups (11.5% with PVI + PWI vs. 10.6% with PVI alone, P = .89).

Table 4

Redo ablation data at long-term follow-up

Redo ablation data	PVI + PWI (n = 169)	PVI only (n = 164)	P-value
Need for ≥1 redo AF ablation (n, %)	54 (32.0%)	49 (29.9%)	.68
Need for ≥2 redo AF ablations (n, %)	5 (3.0%)	5 (3.0%)	.96
1st redo procedure endocardial	52 (30.1%)	47 (28.7%)	.67
Findings at 1st redo endocardial ablation
PV reconnection at redo procedure (n, %)	29/52 (55.8%)	25/47 (53.2%)	.80
Mean number of reconnected veins at redo procedure (mean, SD)	2.3 ± 0.8	2.1 ± 0.9	.60
PW reconnection at redo procedure (n, %)	39/52 (75.0%)	N/A	N/A
LAMT at time of redo	11a/52 (21.2%)	6a/47 (12.8%)	.27
– Peri-mitral flutter	9/52 (17.3%)	5/47 (10.6%)	.34
– Roof-dependent flutter	1/52 (1.9%)	3/47 (6.4%)	.26
– Anterior LA flutter	3/52 (5.8%)	0/47 (0%)	N/A
CTI flutter at the time of redo (n, %)	6/52 (11.5%)	5 (10.6%)	.89
Ablation strategy at redo procedure
PV re-isolation	29/52 (55.8%)	25/47 (53.2%)	.80
De novo PW isolation	N/A	27/47 (57.5%)	N/A
PW re-isolation	39/52 (75.0%)	N/A	N/A
Linear ablationb	25/52 (48.1%)	17/47 (36.1%)	.23
Ablation targeting focal triggers	2/52 (3.8%)	2/47 (4.3%)	.92
Slow pathway ablation/modification	0 (0%)	2 (4.3%)	NA

AF, atrial fibrillation; CAPLA, Catheter Ablation for Persistent Longstanding Atrial Fibrillation; CTI, cavotricuspid isthmus; IQR, interquartile range; LA, left atrium/atrial; LAMT, left atrial macro-reentrant tachycardia; N/A, not applicable; n, number; PAF, paroxysmal atrial fibrillation; PV, pulmonary vein; PW, posterior wall; PWI, posterior wall isolation; SD, standard deviation.

^aDual loop LA flutter was seen in two patients in each group.

^bLinear ablation included CTI, anterior, or lateral mitral isthmus lines.

The ablation strategy at the redo procedure is shown in Table 4. In the 47 patients from the PVI alone group who underwent redo endocardial ablation procedures, pulmonary vein re-isolation was performed in 29 patients (55.8%) and PWI added in 18 (57.5%). Re-isolation of the posterior wall was successfully achieved in all patients from the PVI + PWI group who had reconnection at redo procedure (39 patients, 75.0%). In the eight patients with enduring PVI + PWI at redo procedure, linear ablation at the mitral isthmus was performed in five patients. Multi-procedure success, defined a freedom from recurrent atrial arrhythmia after one or two ablation procedures (on or off AAD therapy), was not significantly different between the two groups (HR 1.04, 95% CI 0.76–1.42, P = .82; Figure 5). Arrhythmia free survival after redo procedure did not vary significantly according to strategy at the redo procedure (P = .15; Supplementary data online, Figure S3).

Figure 5

Time to recurrent atrial arrhythmia after one or two ablation procedures. CI, confidence interval; HR, hazard ratio; PVI, pulmonary vein isolation; PWI, posterior wall isolation

Healthcare utilization and quality of life

Healthcare utilization metrics are displayed in Table 5 and Figure 6. A healthcare utilization event, as defined by unplanned hospitalization, electrical cardioversion, or redo ablation procedure, occurred in 103 patients (60.9%) in the PVI + PWI group vs. 93 patients (56.7%) in the PVI alone group (P = .50). A total of 115 unplanned hospitalization events occurred in the PVI + PWI group compared with 83 events in the PVI alone group. This equated to a mean of 0.7 ± 1.5 events per patient in the PVI + PWI group compared with 0.5 ± 1.5 events per patient in the PVI alone group (P = .20). Symptomatic atrial arrhythmia was the most common reason for hospitalization (64.6%) followed by congestive heart failure (15.2%). The mean number of DCCV procedures performed per patient over long-term follow-up was similar (P = .97). Ongoing AAD use and anticoagulant drug use did not differ significantly between groups at long-term follow-up (P = .31 and P = .43, respectively). Mean AFEQT score at 3-year follow-up was 88.0 ± 14.8 in the PVI + PWI group compared with 88.9 ± 14.8 in the PVI alone group (P = .63). Both groups recorded a substantial improvement in AFEQT score compared with pre-ablation (Δ AFEQT 35.6 ± 21.1 after PVI + PWI vs. 33.8 ± 21.1 after PVI alone), with no significant between-group differences seen in this degree of improvement (P = .85).

Figure 6

Healthcare utilization metrics. Healthcare utilization defined as any one of redo ablation, hospitalization event, or direct current cardioversion event. AAD, anti-arrhythmic therapy; DCCV, direct current cardioversion; PVI, pulmonary vein isolation; PWI, posterior wall isolation

Table 5

Healthcare utilization and quality of life data at long-term follow-up

Outcome data	PVI + PWI (n = 169)	PVI only (n = 164)	P-value
Healthcare utilization eventa	103 (60.9%)	93 (56.7%)	.50
Unplanned hospitalization events per patient (mean, SD)	0.7 ± 1.5	0.5 ± 1.5	.20
DCCV events per patient (mean, SD)	0.6 ± 1.3	0.6 ± 1.3	.97
Anti-arrhythmic drug use	55 (32.5%)	45 (27.4%)	.31
Anticoagulant use	103 (60.9%)	93 (56.7%)	.43
AFEQT score at long-term follow-up	88.0 (14.8)	88.9 (14.8)	.63

AFEQT, Atrial Fibrillation Effect on Quality of Life; AA, atrial arrhythmia; DCCV, direct current cardioversion; SD, standard deviation.

^aHealthcare utilization event defined as any one of unplanned hospitalization, electrical cardioversion, or redo ablation procedure.

Discussion

The central finding of the CAPLA study is that the addition of PWI to standard PVI at the time of index RF ablation procedure for PsAF did not reduce recurrent atrial arrhythmia at a median follow-up of 3.6 years (Structured Graphical Abstract). Irrespective of ablation strategy important additional findings at long-term follow-up include the following: (i) AF burden of 0% in both groups with 86% in sinus rhythm at last review; (ii) in those undergoing redo ablation, pulmonary vein reconnection was present in 55% and posterior wall reconnection in 75%; (iii) no difference in multi-procedure success; and (iv) no difference in healthcare utilization or AF-related quality of life between groups.

Determining the optimal CA strategy for patients with PsAF has long represented a focal point in the field of AF ablation research.¹⁶ Of the proposed targets for adjunctive ablation, the left atrial posterior wall has been widely heralded as a promising option, informed by both biological rationale and observational data.^17–19 Hypothesized advantages of PWI include the reduction of focal AF triggers,²⁰ modification of the septopulmonary bundle and parasympathetic ganglia,^21,22 the targeting of critical atrial substrate,^23,24 PVI reinforcement,²⁵ and mitigation of roof-dependent atrial flutter.²⁶ However, contrary to these anticipated benefits, the 1-year outcomes of the CAPLA study revealed that adjunctive PWI using RF conferred no advantage over standard PVI alone. In general, the CAPLA patient population may have been expected to achieve a favourable response to PVI alone given a relatively short duration of continuous AF with the majority < 12 months, mild to moderate left atrial enlargement, and the presence of sinus rhythm in 40% at the time of index ablation. However, the findings of the present study again reinforce the absence of incremental benefit of empiric PWI observed across any of the studied clinical endpoints at over 3 years of follow-up. Taken together, these results suggest against the empiric adoption of RF-based PWI at index CA for PsAF.

Pulmonary vein and posterior wall reconnection

This study provides valuable insights into the ongoing challenges of achieving durable PVI and PWI with RF ablation. Half of patients with recurrent arrhythmia underwent repeat procedures with pulmonary vein reconnection present in 54% and posterior wall reconnection in 75%. These findings are consistent with contemporary observational studies showing similar rates of pulmonary vein and posterior wall reconnection following initial RF ablation.^27,28 While the overall durability of PWI across the entire PVI + PWI cohort is unknown, these data suggest that the limited efficacy of RF-based PWI in this study may be related to its poor temporal durability with a single ablation procedure.

There are several explanations for the low durability of RF-based PWI. First, linear ablation at the roof line may not adequately penetrate the epicardial layers of the septopulmonary bundle, particularly when the bundle is particularly thick.²⁹ This epicardial layer may have multiple posterior left atrial epi-endocardial connections and thereby necessitate extensive focal ablation within the box despite continuous endocardial block of the roof and inferior lines to achieve eventual PWI.³⁰ Focal RF ablation targeting such epicardial connections was required in 53% of patients randomized to PWI at index procedure in this study, underscoring the prevalence of these connections. The success of PWI is unlikely to be enhanced by 50 W RF ablation, as this higher power setting yields lesions that are broader and more superficial compared with the deeper lesions associated with lower power RF.^31,32 A surgical approach provides epicardial access to isolate the posterior wall. The CASA-AF study randomized 120 patients with long-standing PsAF to a box lesion set (PVI + PWI) using either surgical thorascopic ablation vs. endocardial ablation and reported no difference in arrhythmia outcomes at 12-month and 3-year follow-up.^33,34 Second, oesophageal heating during RF ablation of the posterior left atrium can significantly hinder the formation of transmural lesions, particularly in cases where extensive focal RF is required to achieve PWI.³⁵ These factors underscore inherent challenges encountered when using RF to perform PWI.

Conversely, the introduction of PFA as a relatively cardiomyocyte-specific modality of ablation has the potential to enhance both the initial success and long-term durability of PVI and PWI procedures compared with RF. Initial experience has demonstrated very high rates of acute procedural success, both with focal PFA to perform traditional ‘box’ isolation³⁶ and single-shot systems to eliminate posterior wall signals en bloc.³⁷ While there is great enthusiasm regarding the ease, speed and durability of PFA for performing PVI this have not translated to improved outcomes in a randomized trial.³⁸ Patients in the EU-PORIA registry with recurrent AF undergoing redo ablation following index PFA had pulmonary vein reconnection rates of 62% compared with 54% in the present study using RF.³⁹ A recent observational substudy from the MANIFEST-PF registry found that the addition of PWI using PFA did not improve post-ablation outcomes compared with PVI alone.¹¹ Nonetheless, we await the outcomes of randomized trials comparing the addition of PWI to PVI alone using PFA.

A cautionary note tempering the empiric use of PWI is that additive ablation may inadvertently create further arrhythmic substrate and paradoxically increase the risk of subsequent arrhythmia. In a recent observational study including 1100 patients undergoing redo AF ablation, the incidence of LAMT was significantly higher in patients who had previously undergone PVI + PWI compared with PVI alone.²⁸ A similar trend was seen in the present study, with LAMT observed in twice as many patients following PVI + PWI compared with PVI alone, although not statistically significant. The lack of improved outcomes with the addition of irrigated RF for PWI may be explained by the lack of efficacy of thermal ablation and pro-arrhythmia from more advanced atrial substrate modification or that this simply is not a clinically important target in unselected patients.

Long-term outcomes for catheter ablation for persistent atrial fibrillation

Recent attention has drawn focus to the limited clinical utility of AF recurrence as a parameter of success following AF ablation, particularly in comparison with AF burden.¹⁴ In the present study, while the majority (62%) of patients experienced a binary recurrence, median AF burden at both the 1- and 3-year time points was 0% with either ablation strategy. Median AF burden at 3 years remained at 0% even in the subset of patients who had experienced a recurrence in earlier follow-up. This low AF burden aligns with the observation that most patients experiencing recurrent arrhythmia post-CA for PsAF recur with paroxysmal rather than a persistent form.¹⁵ Furthermore, disease-specific quality of life, as measured by AFEQT scores, was similarly high in both groups at 3 years after ablation. These results underscore the efficacy of CA for PsAF and the importance of evaluating ablation success through a multi-faceted lens, including both AF burden and patient-reported outcomes, rather than relying solely on the recurrence of arrhythmias as a success metric.

Limitations

This study has several limitations. The exclusive use of RF ablation in this study may have impacted PWI durability and hence blunted the potential benefits of adjunctive PWI, which may be achieved with alternate technologies. Second, the detection of recurrent arrhythmia and calculation of AF burden is ideally performed by implantable rhythm monitoring rather than the heterogeneous rhythm surveillance strategies used. In particular, both intermittent transtelephonic monitoring and shorter duration ambulatory ECG monitoring may underestimate AF burden compared with continuous monitoring.⁴⁰ Third, this was a single-blinded study as operators were not blinded to treatment allocation. However, the members of the research team who adjudicated endpoints were blinded to the randomization group.

Conclusions

In patients with PsAF undergoing initial RF CA, the addition of PWI to standard PVI did not improve long-term freedom from atrial arrhythmia recurrence. While documentation of recurrent atrial arrhythmia was common, overall AF burden remained low at long-term follow-up with either ablation strategy.

Supplementary data

Supplementary data are available at European Heart Journal online.

Declarations

Disclosure of Interest

Dr William is supported by a NHF post-graduate PhD scholarship. Dr Crowley is supported by a Baker institute PhD scholarship. Dr Chieng was supported by a co-funded NHMRC/NHF post-graduate PhD scholarship. Dr Segan is support by a co-funded NHMRC/NHF post-graduate PhD scholarship. Dr Ling reported receiving grants from Abbott Australia. Dr Prabhu reported receiving grants from NHMRC, University of Melbourne, and Baker Heart and Diabetes Institute; receiving postdoctoral fellowship support from the Heart Foundation; receiving advisory fees from Biosense Webster; and receiving speaker fees from Abbott Medical. Dr G. Wong reported receiving grants from the National Heart Foundation. Dr Sterns reported receiving personal fees from Biosense Webster. Dr Ginks reported serving on speaker bureaus for Biosense Webster Speaker and Abbott. Dr Sanders reported serving on advisory boards for Medtronic, Abbott Medical, CathRx, Pacemate, and Boston Scientific and reported being supported by a practitioner fellowship from NHMRC and by the National Heart Foundation of Australia. Dr Kalman reported receiving fellowship support from Medtronic and Biosense Webster. Dr Kistler is a recipient of an Investigatorship from the NHMRC of Australia and received grants from Baker Department of Metabolic Health University of Melbourne; speaker fees from Abbott Medical; and serving on an advisory board for Biosense Webster.

Data Availability

The raw data underlying this article cannot be shared publicly due to the privacy of individuals who participated in the study. The data will be shared on reasonable request to the corresponding author.

Funding

This study was supported by seed grant funding from the Baker Heart and Diabetes institute, Melbourne, Australia.

Ethical Approval

Ethical approval was granted by the Alfred Health Office of Ethics and Research Governance (Alfred Health, Melbourne Australia) (ethical approval number 89/17).

Pre-registered Clinical Trial Number

The pre-registered clinical trial number is ACTRN12616001436460.