Extended Ablation Versus Pulmonary Vein Isolation to Treat Persistent Atrial Fibrillation

Overview

The aim is to determine the effect of extended ablation (pre-specified linear PF/RF lesion set) in addition to PVI in symptomatic persistent AF patients on AF/AT/AFL recurrence. The study is designed as multicenter, randomized trial. Eligible patients are patients with persistent AF (but not long-standing persistent AF) planned for a first-ever AF ablation procedure. Patients will be randomly assigned 1:1 either to: i) PVI-only, or ii) PVI-plus arms. Patients in both groups will undergo catheter ablation using an ablation system capable of performing PF and RF ablation (Sphere-9, Affera, Medtronic). Patients randomized to the PVI-only arm will undergo only PVI. Patients randomized to the PVI-plus arm will undergo PVI plus linear lesions (roof, bottom line, lateral or anterior MI line, and septal line in the LA; intercaval line and cavotricuspid line in the RA). The primary endpoint will be freedom from recurrent AF/AT/AFL, assessed as time-to-first recurrence in the period of 12 months after randomization (post 2-month blanking period). Secondary clinical endpoints will be 1) 1) AF/AT/AFL burden during the first 12 months after randomization, 2) 2) Quality of life according to the AFEQT questionnaire score at 12 months, 3) 3) AF/AFL/AT-related outcomes (hospitalization or emergency visits), 4) MACE defined as cardiovascular death, stroke, myocardial infarction, or hospitalization for heart failuree, 5) A composite of pre-specified, procedure-related SAE within 7 days after the procedure, including clinically relevant delayed SAEs adjudicated as related to the index ablation procedure. Secondary endpoints will be evaluated throughout the entire (minimum 12-month) follow-up period.

BACKGROUND The cornerstone of catheter ablation for AF is pulmonary vein (PV) isolation (PVI). PVI alone is highly effective in patients with paroxysmal AF; however, its efficacy is lower in patients with non-paroxysmal AF (persistent and long-standing persistent). For this reason, additional ablation strategies have been developed and investigated in patients with non-paroxysmal AF. However, results from randomized catheter-based studies comparing PVI alone with extended ablation strategies combining PVI with additional linear lesions have been inconsistent. While some studies showed no improvement in sinus rhythm (SR) maintenance with the addition of linear lesions compared with PVI alone, other studies demonstrated superiority of an extensive ablation approach, including linear lesions over PVI alone. A key determinant of success in these studies is the completeness of the ablation lesion set. . In studies using radiofrequency (RF) energy alone, such as the STAR AF II trial, all planned linear ablation lesions were completed in only 74% of patients during the procedure. Moreover, in the CAPLA study, comparing PVI alone with PVI combined with adjunctive left atrial (LA) posterior wall isolation using roof and inferior lines, posterior wall reconnections were present in 75% of patients. However, the results of older catheter ablation studies that relied solely on RF energy cannot be directly extrapolated to contemporary procedures performed using PF energy or combined PF/RF systems that also enable simultaneous endocardial mapping and thus reliable identification of conduction gaps and completion of ablation lesions. The markedly higher efficacy of these newer energy modalities and their combinations raises the question of whether highly effective comprehensive endocardial ablation, including linear lesions, will be associated with superior sinus rhythm maintenance compared with PVI alone. The aim of this study is therefore to compare PVI-alone with an extended ablation strategy (PVI plus linear lesions) in patients with persistent AF, using a technology that enables ablation with both PF and RF energy and 3D mapping. STUDY DESIGN Prospective, randomized controlled trial that will be conducted at 4 sites in the Czechia. STUDY POPULATION The study focuses on patients with persistent AF as defined in recent ESC guidelines (i.e., with ≥2 AF episodes; ≥1 episode with a duration of \>7 days) who are referred the catheter ablation in accordance with the indication criteria outlined in these guidelines. Inclusion criteria: * symptomatic persistent AF * ≥2 episodes of persistent AF in the last 24months, of which ≥1 episode of persistent AF in the last 12 months; or ongoing AF lasting \>7 days at the time of screening * LAd ≥42 mm * signed informed consent Exclusion criteria: * first-manifested AF * paroxysmal AF, long-standing persistent AF, permanent AF * AF of secondary cause (e.g., hyperthyroidism) * any previous LA ablation * severe valvular disease (mitral valve insufficiency ≥3+, moderate or severe aortic stenosis) or history of valvular surgery or intervention * left ventricular ejection fraction ≤40% * pulmonary hypertension (estimated systolic pulmonary artery pressure ≥40 mm Hg) * symptomatic coronary artery disease * pregnancy * LAd ≥55 mm * body mass index ≥40 kg/m2 * age \< 18 years, or ≥80 years * chronic kidney disease stage 3b or higher * general contraindications of catheter ablation * contraindication to anticoagulation * life expectancy \<2 years due to other comorbidities RANDOMIZATION AND MASKING Patients will be randomized to the PVI-plus or PVI-only arm. Operators, personnel involved in the procedure, and staff participating in the patient´s clinical follow-up will not be blinded to the treatment allocation. ECG monitoring will be performed outside the participating centres, and the technicians and physicians analyzing the ECG recordings will be fully blinded to the patient's assignment. ELECTROPHYSIOLOGICAL STUDY AND ABLATION PROCEDURE Ablation will be performed using an ablation system that enables endocardial mapping and delivery of both PF and RF energy (Affera, Medtronic, USA). The aim of ablation therapy in the PVI-only arm will be PVI alone. The objective in the PVI-plus arm will be, in addition to PVI, the deployment of a comprehensive ablation lesion set designed to block major atrial reentry circuits, including the roof and bottom lines, perimitral (lateral or anterior) line, septal line in the left atrium, and CTI line with intercaval line in the right atrium. OUT-PATIENT FOLLOW-UP AND BLANKING PERIOD Starting on the day of randomization, out-patient follow-up visits will be scheduled at 2, 6, and 12 months. The first 2 months will serve as a blanking period. In the event of AF/AT/AFL recurrence during the blanking period, management with AADs and/or electrical cardioversion is permitted. In addition, all patients presenting with AF at the 2-month visit, irrespective of symptom burden, will be referred for early electrical cardioversion, so that the majority of patients enter the rhythm analysis period in SR. Five-day ECG monitoring will be performed at 3, 6, 9, and 12 months starting on the day of randomization, and every 6 months thereafter. ENDPOINTS Primary endpoint The primary outcome is freedom from any documented AF, AT, or AFL without antiarrhythmic drug therapy, cardioversion, or redo ablation, through 12 months after randomization, following the protocol-defined blanking period. Secondary endpoints * AF/AFL/AT burden during the first 12 months after randomization * Quality of life assessment using the AFEQT questionnaire * AF-related clinical outcomes, i.e., emergency visits or hospitalizations due to documented recurrence of AF/AFL/AT during the whole study course * Major Adverse Cardiovascular Events (MACE), a composite of cardiovascular death, stroke, myocardial infarction, or hospitalization for heart failure during the whole study course * A composite of pre-specified procedure-related SAEs within 7 days after the procedure, including clinically relevant delayed SAEs adjudicated as related to the index ablation procedure Endpoint definition Primary endpoint: An episode of AF/ AT/AFL is defined as either (1) a recording demonstrating at least 30 seconds of continuous interpretable signal during Holter monitoring performed at 3, 6, 9, or 12 months, or (2) a 12-lead ECG demonstrating the arrhythmia throughout the entire tracing, with at least 10 seconds of continuous interpretable signal during scheduled or emergency visits. In addition to a documented AF/AT/AFL episode as defined above, treatment failure (a primary endpoint event) will also be considered to have occurred if, after the end of the blanking period, any of the following occurs: (a) initiation or continuation of a Class I or Class III antiarrhythmic drug, or of sotalol, irrespective of the indication; (b) electrical cardioversion for AF/AT/AFL; (c) a repeat (redo) ablation for AF/AT/AFL. Beta-blockers and calcium-channel blockers do not constitute treatment failure. The date of the earliest of these events, or the qualifying documented episode will be considered the date of the primary endpoint. Secondary endpoints AF/AT/AFL burden Overall AF burden will be calculated for each patient as the percentage of total analysable monitoring time across protocol-specified Holter recordings during the first 12 months after randomization spent in AF, AT, or AFL. AF/AT/AFL burden will also be calculated separately at each scheduled Holter assessment at 3, 6, 9, and 12 months. Quality of life: quality of life will be assessed using the standardized AFEQT (Atrial Fibrillation Effect on QualiTy-of-life) questionnaire at the baseline visit and at the 12-month follow-up visit after randomization. The 12-month score is the endpoint and is analysed with the baseline score as a covariate. AF-related clinical outcomes are defined as any emergency visits or hospitalizations due to recurrence of AF/AFL/AT during the entire study period. Importantly, the arrhythmia must be documented by ECG (12-lead ECG or other ECG tracing obtained by emergency medical services). Emergency visits due to palpitations without an ECG-documented arrhythmia will not be considered an endpoint. MACEs are defined as a composite of cardiovascular death, stroke, myocardial infarction, or hospitalization for heart failure during the entire study period. Cardiovascular death will be defined as death resulting from a cardiovascular cause, including myocardial infarction, heart failure, sudden cardiac death, stroke, pulmonary embolism, cardiovascular procedures, or other vascular causes, and includes any death for which a cardiovascular cause cannot be definitively excluded (sudden unwitnessed death). Stroke will be defined as an acute neurological deficit of vascular origin lasting ≥24 hours or resulting in death, confirmed by neuroimaging (CT or MRI) demonstrating cerebral infarction or intracranial haemorrhage, or by clinical assessment when imaging is not available, and not explained by an alternative non-vascular diagnosis. Myocardial infarction will be defined according to ESC criteria as a rise and/or fall in cardiac troponin with at least one value above the 99th percentile upper reference limit, together with clinical, electrocardiographic, imaging, or angiographic evidence of myocardial ischaemia. However, an isolated procedure-related elevation in cardiac troponin following ablation procedure, in the absence of other evidence of myocardial ischaemia, will not be considered myocardial infarction. Hospitalization for heart failure will be defined as an unplanned admission to a hospital with a primary diagnosis of heart failure, characterized by worsening symptoms and/or signs of heart failure, requiring at least an overnight in-hospital stay and necessitating initiation or escalation of heart failure-directed treatment, including but not limited to intravenous diuretics, vasodilators, inotropes, or mechanical circulatory or ventilatory support. Hospitalization for heart failure may, but does not necessarily have to, be associated with recurrence of AF/AT/AFL. Procedure-related SAEs: list of pre-specified procedure-related SAEs is provided below. Clinically relevant SAEs occurring beyond the predefined 7-day safety window (e.g., atrio-oesophageal fistula or delayed cardiac tamponade) will be assessed for their potential relationship to the index ablation procedure. Events adjudicated as procedure-related will also be included in the safety endpoint. All AF-related outcomes, MACEs and procedure-related SAEs will be assessed by the Clinical Event Committee (CEC), whose members will be unaware of treatment allocation; the adjudicated classification will be used in all analyses. Criteria under which initiation of antiarrhythmic therapy, cardioversion or repeat ablation constitutes a primary endpoint event are pre-specified in Endpoint section. Data management A tailor-made website will be developed for the study. Each participating medical centre will have access to a dedicated section of the website (data entry and randomization). Sample size calculation We assume that 55% of patients in the PVI-only group will achieve freedom from AF/AT/AFL during the first year of follow-up. In contrast, freedom from AF/AFL/AT in the PVI-plus group is expected in at least 68% of patients. These proportions correspond to a hazard ratio of 0.645 (log 0.68 / log 0.55) for AF/AT/AFL recurrence in the PVI-plus group as compared with the PVI-only group, assuming proportional hazards over the follow-up period. Because the primary analysis is a time-to-event comparison using the Kaplan-Meier method and the stratified log-rank test, the sample size was derived from this hazard ratio by the method of Schoenfeld. A hazard ratio of 0.645 requires 164 primary-endpoint events to provide 80% power to detect a between-group difference at a two-sided alpha level of 0.05. With 12-month recurrence probabilities of 45% in the PVI-only and 32% in the PVI-plus group (a pooled probability of 38.5%), a total of 426 patients (213 per group) completing 12 months of follow-up are expected to yield those 164 events. Allowing for 5% of patients to be lost before the 12-month assessment and, conservatively, to contribute no events, a total of 450 patients (225 per group) will be randomized. The calculation was performed in R version 4.6.1 (R Foundation for Statistical Computing, Vienna, Austria). Statistical analysis and methods Analyses of the primary and secondary endpoints follow the intention-to-treat principle. All tests are two-sided at a significance level of 0.05 and all confidence intervals are two-sided at the 95% level. The intention-to-treat population comprises all randomized patients, analysed according to the treatment assigned and irrespective of the procedure performed, of adherence to the protocol or of withdrawal from follow-up. It is the primary analysis population for the primary endpoint and for all secondary efficacy endpoints. Primary endpoint analysis Treatment failure is defined in the Endpoint definition section; the date of the earliest qualifying event is the date of the primary endpoint. All components of the primary endpoint will be adjudicated either by the Clinical Events Committee or blinded ECG Holter analysis team without knowledge of treatment allocation, and the adjudicated value will be used in the analysis. Day 0, the end of the blanking period, delayed entry and the handling of patients in whom the protocol-mandated cardioversion fails are defined in the Blanking period section and apply to all analyses. Where AF/AT/AFL is documented by rhythm monitoring initiated at the 12-month assessment, the event date used is the earlier of 365 days after randomization and the actual date of the episode, so that episodes identified at the 12-month assessment are counted even where the assessment took place after day 365 but within its ±14-day window. Patients without a qualifying event are censored at the last date on which all components of the primary endpoint could be ascertained, that is, the later of the last available rhythm assessment and the last clinical contact at which antiarrhythmic drug use, cardioversion and redo ablation status were recorded; no follow-up time is credited beyond that date. Patients who die without a qualifying event are censored at the date of death and their number reported by arm; given the small number of deaths anticipated over 12 months, no competing-risk analysis is planned. Freedom from treatment failure will be estimated in each arm by the Kaplan-Meier method, with standard errors from Greenwood's formula and two-sided 95% confidence intervals based on the log-log transformation. The following hypothesis will be tested by a two-sided log-rank test at α = 0.05, stratified by the three randomization factors (sex, left atrial diameter, age category): H0: S\_PVI-plus(t) = S\_PVI-only(t) for all t between the end of the blanking period and 12 months after randomization Ha: S\_PVI-plus(t) ≠ S\_PVI-only(t) for some t between the end of the blanking period and 12 months after randomization where S(t) denotes the probability of freedom from treatment failure. Because patients enter the risk set at the end of their individual blanking period, the statistic will be obtained as the score test of the corresponding stratified Cox model on left-truncated data, which is numerically identical to the stratified log-rank test. The treatment effect will be estimated as a hazard ratio from the same stratified Cox model with a 95% confidence interval, interpreted as an average effect over the follow-up period. The proportional hazards assumption will be examined on scaled Schoenfeld residuals, and the difference in restricted mean survival time at 12 months reported as a supportive measure of effect. The Kaplan-Meier estimates of freedom from treatment failure at 12 months and their difference with a 95% confidence interval, which is the quantity for which the sample size was calculated, will be reported as the primary descriptive summary; the curves will be displayed from randomization with the numbers at risk at each time point. Supportive analyses comprise the unstratified log-rank test, so that the marginal comparison is also on record, and the primary analysis additionally stratified by study site. The stratification factors used in all stratified analyses are those recorded by the randomization system at the time of allocation. Any discrepancy between these and the final clinical values recorded in the case report form will be reported. Secondary endpoint analyses AF/AT/AFL burden. The overall burden is the proportion of analysable monitoring time spent in AF, AT or AFL across all protocol-specified Holter recordings obtained during the first 12 months after randomization, computed for each patient as the total time in AF/AT/AFL divided by the total analysable monitoring time. A recording contributes only if at least 24 hours of it are analysable, and a patient contributes only if at least 72 hours are analysable across all scheduled recordings; patients below that threshold are treated as missing for this endpoint and their number is reported by arm with reasons. AF burden after ablation is strongly zero-inflated and, in patients whose persistent AF recurs, also has a mass near 100%, so it is neither approximately normal nor a location shift of a common distribution. The primary comparison is therefore the non-parametric relative effect, the probability that a randomly selected patient in the PVI-plus arm has a lower burden than a randomly selected patient in the PVI-only arm, with ties counted at one half. It is reported with a two-sided 95% confidence interval and tested against one half by the Brunner-Munzel test, which does not require the two distributions to have the same shape and tolerates the extensive ties that the zero-inflation produces; if the asymptotic test is in doubt, the p-value is obtained by studentized permutation. Burden is summarized by arm as the median with interquartile range, the mean with standard deviation, and the proportion of patients with 0%, above 0% to 1%, above 1% to 10%, above 10% to 50%, and above 50%. Quality of life The quality-of-life endpoint is the AFEQT overall score at 12 months, analysed using analysis of covariance (ANCOVA) with the 12-month AFEQT score as the dependent variable, treatment group as the main factor, and the baseline AFEQT score as a covariate. The adjusted between-group difference with the corresponding 95% confidence interval will be reported and interpreted against a minimal clinically important difference of 5 points. AF-related clinical outcomes, MACE, and procedure-related SAE AF-related clinical outcomes and MACEs will be analysed using Kaplan-Meier estimates and Cox proportional hazards regression models, with treatment group as a covariate. The proportion of patients experiencing at least one pre-specified procedure-related SAE will be compared between treatment groups using Fisher's exact test. Formal testing of the secondary endpoints is conditional on the primary endpoint reaching statistical significance, and then proceeds in the pre-specified fixed sequence (1) overall AF burden, (2) quality of life, (3) AF-related clinical outcomes, (4) MACEs, stopping at the first endpoint that does not reach the two-sided 0.05 level. The composite of pre-specified, procedure-related SAEs is reported unconditionally with risk differences, 95% confidence intervals and nominal p-values; it is not part of the hierarchy.