Association Between Quality of Life and Procedural Outcome After Catheter Ablation for Atrial Fibrillation: A Secondary Analysis of a Randomized Clinical Trial

Key Points

Question  Is ablation for atrial fibrillation (AF) associated with changes in quality of life in patients with persistent AF?

Findings  In this secondary analysis of the Substrate and Trigger Ablation for Reduction of Atrial Fibrillation–Part II (STAR AF II) randomized clinical trial, among 549 patients, catheter ablation was associated with a significant decrease in AF burden and increase in quality of life at 18 months. Patients with less than 70% reduction in AF burden had a significantly lower quality of life score compared with those with greater than 70% reduction in AF burden.

Meaning  The findings suggest that changes in quality of life are associated with the percentage of AF burden reduction after ablation.

Importance  Catheter ablation is effective in reducing atrial fibrillation (AF), but the association of ablation for AF with quality of life is unclear.

Objective  To evaluate whether the procedural outcome of ablation for AF is associated with quality of life (QOL) measures.

Design, Setting, and Participants  This was a prespecified secondary analysis of the Substrate and Trigger Ablation for Reduction of Atrial Fibrillation–Part II (STAR AF II) prospective randomized clinical trial, which compared 3 strategies for ablation of persistent AF. This analysis included 549 of the 589 patients enrolled in the trial who underwent ablation. Enrollment occurred at 35 centers in Europe, Canada, Australia, China, and Korea from November 2010 to July 2012. Data for the current study were analyzed on December 11, 2019.

Interventions  Patients underwent AF ablation with 1 of 3 ablation strategies: (1) pulmonary vein isolation (PVI), (2) PVI plus complex fractionated electrograms, or (3) PVI plus linear lesions.

Main Outcomes and Measures  Quality of life was assessed at baseline and at 6, 12, and 18 months after ablation for AF using the 36-Item Short Form Health Survey and the EuroQol Health-Related Quality of Life 5-Dimension 3-Level questionnaire. Scores were also converted to a physical health component score (PCS) and a mental health component score (MCS). Individual AF burden was calculated by the total time with AF from Holter monitors and the percentage of transtelephonic monitor recordings showing AF.

Results  Among the 549 patients included in this secondary analysis, QOL was assessed in 466 (85%) at baseline and at 6, 12, and 18 months after ablation for AF. The mean (SD) age of the study population was 60 (9) years; 434 (79%) individuals were men, and 417 (76%) had continuous AF for 6 months or more before ablation. The AF burden significantly decreased from a mean (SD) of 82% (36%) before ablation to 6.6% (23%) after ablation (P < .001). Significant improvements in mean (SD) PCS (68.3 [20.7] to 82.5 [18.6]) and MCS (35.3 [8.6] to 37.5 [7.6]) occurred 18 months after ablation (P < .05 for both). Significant QOL improvement occurred in all 3 study arms and regardless of AF recurrence, defined as AF episodes lasting more than 30 seconds: for no recurrence, mean (SD) PCS increased from 66.5 (20.9) to 79.1 (19.4) and MCS from 35.3 (8.7) to 37.7 (7.7); for recurrence, mean (SD) PCS increased from 70.2 (20.4) to 86.4 (16.8) and MCS from 35.3 (8.6) to 37.1 (7.4) (P < .05 for all). When outcome was defined by AF burden reduction, in patients with less than 70% reduction in AF burden, the increase in PCS was significantly less than in those with greater than 70% reduction, and only 3 of 8 subscales showed significant improvement.

Conclusions and Relevance  In this secondary analysis, decreases in AF burden after ablation for AF were significantly associated with improvements in QOL. Quality of life changes were significantly associated with the percentage of AF burden reduction after ablation.

Trial Registration  ClinicalTrials.gov Identifier: NCT01203748

Atrial fibrillation (AF) is substantially associated with decreased quality of life (QOL) among patients.¹ Interventions for AF, such as rate control and antiarrhythmic drugs, have shown the ability to improve patient QOL.^2,3

Catheter ablation of AF is an important modality to treat AF.⁴ Studies reporting outcomes after ablation have focused on a primary end point of freedom from episodes of AF lasting longer than 30 seconds. With this reference standard, the success of ablation is 50% to 60%. However, many patients experience symptom improvement despite experiencing ongoing brief episodes. The Catheter Ablation vs Antiarrhythmic Drug Therapy in Atrial Fibrillation (CABANA) trial did not show a benefit of ablation in terms of mortality or stroke⁵ but showed a significant improvement in QOL.⁶ The Catheter Ablation Compared With Pharmacological Therapy for Atrial Fibrillation (CAPTAF) trial also showed a significant improvement in QOL after ablation even though approximately 70% of patients experienced ongoing brief recurrences.⁷ The precise relationship between AF burden reduction and QOL remains understudied.

The Substrate and Trigger Ablation for Reduction of Atrial Fibrillation–Part II (STAR AF II) trial tested 3 different strategies for ablation of persistent AF: pulmonary vein isolation (PVI), PVI plus complex fractionated electrograms (CFEs), and PVI plus linear lesions (LL).⁸ At 18 months, there was no significant difference between the study arms in the number of episodes of AF lasting longer than 30 seconds.⁹ The trial also measured QOL. The purpose of this secondary analysis of the STAR AF II trial was to assess the association between ablation treatment for AF and patient QOL.

This was a prespecified secondary analysis of the STAR AF II trial (NCT01203748) (trial protocol in Supplement 1). Enrollment occurred at 35 centers in Europe, Canada, Australia, China, and Korea from November 2010 to July 2012. In the trial, 589 patients were randomized to receive 1 of the 3 aforementioned strategies in a 1:4:4 ratio. A total of 549 patients underwent ablation (21 patients did not receive ablation; 19 dropped out before 3 months) and were included in this analysis. Patients were required to have had symptomatic persistent AF (<3 years) refractory to at least 1 antiarrhythmic agent and no previous ablations. The primary end point for the original trial and for this analysis was freedom from documented episodes of AF lasting longer than 30 seconds after an initial 3-month blanking period in the presence or absence of antiarrhythmic treatment after a single procedure. All patients provided written informed consent, and the protocol was approved by all participating sites’ ethics boards. The current analysis was approved by the institutional review board of St Jude Medical. Further details of the study design,⁸ a CONSORT flow diagram, the trial protocol, and the primary end point results⁹ have been published previously. This study followed the Consolidated Standards of Reporting Trials (CONSORT) reporting guideline.

Follow-up visits occurred 3, 6, 9, 12, and 18 months after ablation. All patients underwent 12-lead electrocardiography and 24-hour Holter monitoring at baseline and at every follow-up visit; patients wore the Holter monitors at home between visits. In addition, patients transmitted electrocardiograms and symptoms with a transtelephonic monitor (TTM) weekly for 18 months. A total of 2334 Holter monitor and 28 798 TTM readings were received for analysis.

Quality of life before and after ablation was analyzed using the 36-Item Short Form Health Survey (SF-36). The SF-36 and EuroQol Health-Related Quality of Life 5-Dimension 3-Level (EQ-5D-3L) questionnaires were administered at baseline and at 6, 12, and 18 months after ablation. Details of the SF-36 and EQ-5D-3L questionnaires are included in the eAppendix in Supplement 2. According to the SF-36 manual, a 5-point difference in score was defined as clinically and socially relevant.¹⁰ The EQ-5D-3L includes a visual assessment scale that grades overall health status from 0 to 100, in which 100 is the best and 0 the worst.^11,12 As part of the prespecified QOL analysis, we sought to assess the changes in QOL from before to after ablation in all patients according to randomization arm, success defined by the trial’s primary end point, and reductions in AF burden (60%-90%).

Because patients did not have an implantable loop recorder for continuous monitoring, AF burden before and after ablation needed to be calculated. Details of the burden calculation have been previously published⁹ and are provided in the eAppendix in Supplement 2.

The data for this secondary analysis were analyzed on December 11, 2019. All data are reported as a mean (SD) or median (interquartile range [IQR]) for continuous variables according to the distribution or as the number (percentage) of patients for categorical variables. Only patients completing the baseline, 12-month, and 18-month surveys were included in the analysis. A paired t test was used for the comparison of the QOL scores and the magnitude of change in QOL scores between baseline and 12 months and baseline and 18 months. If the normality assumption was violated, the equivalent nonparametric method, specifically the Wilcoxon signed rank test, was used. Comparison of QOL scores between the 3 treatment arms was conducted using the Kruskal-Wallis test. For univariable and multivariable analyses, linear regression was used to investigate the association of various factors with the aggregate change in the mental health component score (MCS) and physical health component score (PCS). P < .05 using a 2-sided test was considered statistically significant for all analyses. Statistical calculations were performed using SAS, version 9.2 (SAS Institute).

A total of 549 patients underwent ablation, with 466 (85%) completing the 12-month surveys and 465 (85%) completing the 18-month surveys; 84 patients (15%) did not complete all of the surveys and were excluded from analysis. Baseline demographic characteristics of the patients are provided in Table 1. The mean (SD) age of patients was 60 (9) years; 434 (79%) were men, and 417 (76%) had continuous AF for 6 months or more before ablation. Most patients had a low CHADS₂ (congestive heart failure, hypertension, age >75 years, diabetes, and stroke or transient ischemic attack) risk score (scores range from 0 to 6, with lower scores indicating a lower chance of stroke), with 35% scoring 0 and 47% scoring 1.

The mean (SD) preablation AF burden among patients was 82% (36%), and the median was 100% (IQR, 0.03%-100%), with 76% of the patients considered to have a preablation burden of 100%. The distribution of postablation burden is shown in Figure 1. There was a significant reduction in mean (SD) AF burden from 82% (36%) before ablation to 6.6% (23%) after ablation (median, 0% [IQR, 0%-100%]; P < .001), representing a 92% reduction in AF burden. A total of 247 patients (45%) had a postablation AF burden of 0, with another 214 (39%) having a burden between 0% and 10%. Only 22 patients (4%) had a postablation AF burden greater than 20%.

The change in QOL for each SF-36 subscale, the PCS, and the MCS is detailed in Table 2. There was a statistically significant improvement in all the SF-36 subscales and summary component scores from baseline to 6 months, 12 months, and 18 months after ablation. There was no statistically significant change from the 6-month to the 12-month or 18-month scores.

The magnitude of change in each of the subscales was greater than 5 points even when taking the 95% CI into account, which represents changes that were statistically and clinically significant (Table 2). The change in the PCS score was 13.9 points (95% CI, 12.0-15.8 points). The change in the MCS score was 1.9 points (95% CI, 1.1-2.7 points). The EQ-5D-3L score also showed a significant positive change in QOL after ablation (eTable 3 in Supplement 2).

There was no significant difference among the 3 arms of the trial at 18 months for the primary outcome of recurrence of AF episodes lasting longer than 30 seconds. Patients in all 3 arms experienced a significant improvement in the PCS and MCS from baseline to 18 months (eTable 1 in Supplement 2). The mean change in PCS was greater than 5 points in all 3 arms, but the change in MCS was less than 5 points in all 3 arms. Among the 3 arms, however, there was no statistically significant difference in the magnitude of change for either the PCS or the MCS (eTable 1 in Supplement 2). For the PCS, the magnitude of change was 12.2 points (95% CI, 10.6-13.8 points) for PVI, 14.6 points (95% CI, 12.8-16.4 points) for PVI plus CFE, and 13.7 points (95% CI, 11.7-15.7 points) for PVI plus LL (P = .31). For the MCS, the magnitude of change was 1.3 points (95% CI, 0.5-2.0 points) for PVI, 2.6 points (95% CI, 1.9-3.3 points) for PVI plus CFE, and 1.3 points (95% CI, 0.5-2.1 points) for PVI plus LL (P = .56). The magnitude of change for each of the SF-36 subscales also did not show a statistically significant difference between arms (eTable 1 in Supplement 2). The results of the EQ-5D-3L scale are included in eTable 4 in Supplement 2. A statistically significant improvement in the EuroQol visual assessment scale was seen in all 3 arms.

Patients who did and did not experience a recurrence of AF episodes lasting longer than 30 seconds experienced a statistically significant improvement in the PCS and MCS from baseline to 18 months (Figure 2). For no recurrence, mean (SD) PCS increased from 66.5 (20.9) to 79.1 (19.4) and MCS from 35.3 (8.7) to 37.7 (7.7); for recurrence, mean (SD) PCS increased from 70.2 (20.4) to 86.4 (16.8) and MCS from 35.3 (8.6) to 37.1 (7.4) (P < .05 for all). The EuroQol visual assessment scale also showed a statistically significant improvement in both groups, as did all the scales of the EQ-5D-3L except for self-care (eTable 5 in Supplement 2). Between the recurrence and no recurrence groups, there was no statistically significant difference in the magnitude of change for either the PCS or the MCS. For the PCS, the magnitude of change was 15.5 points (95% CI, 13.7-17.3 points) for patients with AF recurrence and 12.5 points (95% CI, 10.6-14.4 points) for patients without AF recurrence (P = .07). For the MCS, the magnitude of change was 1.4 points (95% CI, 0.7-2.1 points) for patients with AF recurrence and 2.3 points (95% CI, 1.5-3.1 points) for patients without AF recurrence (P = .29). The magnitude of change for each of the SF-36 subscales also did not show a statistically significant difference between the recurrence and no recurrence groups (eTable 2 in Supplement 2).

Patients were analyzed according to whether they had an AF burden reduction above or below the following thresholds: 90%, 80%, 70%, and 60%. If patients had an AF burden reduction above any of the specified thresholds, each of the 8 subscales showed a positive and statistically significant change of greater than 5 points (Figure 3). For patients with less than 90% AF burden reduction, there were still statistically significant positive changes greater than 5 points in 6 of the 8 subscales. For patients with less than 80% AF burden reduction, 4 of 8 scales showed a statistically significant positive change (role limitation owing to physical problems, vitality, mental health, and physical functioning), whereas the others showed no statistically significant change. For patients with less than 70% AF burden reduction, 3 of 8 scales (38%) showed a statistically significant positive change (role limitation owing to physical problems, vitality, and mental health). For patients with less than 60% AF burden reduction, 2 of 8 scales (25%) showed a statistically significant positive change (role limitation owing to physical problems, vitality).

eTable 3 in Supplement 2 lists the univariable factors associated with change in the aggregate MCS plus PCS, the MCS, and the PCS. By univariable analysis, the baseline QOL score was the only factor associated with a change in the aggregate MCS plus PCS and the MCS at 18 months. The baseline QOL score and AF burden reduction less than 70% were significantly associated with change in the PCS at 18 months. By multivariable analysis, AF burden reduction less than 70% was significantly associated with a change in the PCS at 18 months even after correction for the baseline QOL score (β, –4.03; 95% CI, –7.67 to –0.396; P = .03). Reductions in AF burden of less than 80% and less than 90% were also considered separately in the model but were not significantly associated with score changes by univariable or multivariable analysis. Reduction in AF burden of less than 60% was not assessed because of the small number of patients available for such an analysis (n = 9).

In this secondary analysis of the STAR AF II randomized clinical trial, we analyzed the change in QOL after catheter ablation for patients with persistent AF. Ablation was associated with a 92% reduction in AF burden. Patients experienced a statistically significant improvement in QOL from baseline to 18 months. There was no difference in QOL improvement between patients with and without AF recurrence, defined as AF episodes lasting more than 30 seconds. However, when the outcome was defined by AF burden reduction, patients with lower AF burden reductions had fewer QOL subscales showing significant improvement. By multivariable analysis, AF burden reduction of less than 70% was associated with lower QOL scores, and the association remained statistically significant after correction for baseline score.

The primary indication for AF ablation is relief of symptoms.⁴ The CABANA trial failed to show any significant improvement in stroke, bleeding, or mortality after ablation, but there was a sustained and significant improvement in QOL.⁶ This improvement correlated with a significant reduction in AF recurrence after ablation compared with after drug therapy.⁵ The Medical Antiarrhythmic Treatment or Radiofrequency Ablation in Paroxysmal Atrial Fibrillation (MANTRA-PAF) trial^13,14 also showed a statistically significant improvement in QOL after ablation that was superior to antiarrhythmic drugs as first-line treatment and was sustained for 5 years. However, the trial did not assess the relationship of AF burden or procedural success with QOL change. The recently published CAPTAF trial⁷ was unique in that improvement in QOL was the primary end point of the trial and patients also received implantable loop recorders for assessment of AF burden. In CAPTAF, the ablation group showed a nearly 9-point difference in general health measured by SF-36, and 5 of the 7 subscales also showed a significant improvement. Compared with CAPTAF, the results of the current subanalysis showed a larger difference, with a nearly 15-point increase in the combined MCS and PCS. Each subscale also showed a significant improvement of 5 points or more. This could be explained by the higher freedom from AF recurrence seen in STAR AF II (nearly 50%) compared with in the CAPTAF trial (25%).^7,9 The AF burden after ablation was similar in both studies (6.6% [SD, 23%] in STAR AF II vs 5.5% [SD, 18.1%] in CAPTAF), but CAPTAF used continuous monitoring with implantable loop recorders, which better ascertains AF compared with the intermittent monitoring used in STAR AF II.

In patients with and without recurrence of AF episodes lasting more than 30 seconds, a significant improvement in QOL was found. This might suggest that factors independent of AF recurrence contributed to postablation QOL improvement. For example, postablation AF recurrences are more likely to be asymptomatic because of shorter duration and postablation autonomic modulation.^15,16 Response bias (the so-called placebo effect) is also a possibility because patients undergo an interventional procedure. However, improvements in QOL after ablation were sustained for 18 months in the current study and for more than 5 years in CABANA, which would be unlikely from a placebo effect alone. The disconnect between apparently suboptimal success rates of ablation for AF and postablation improvements in QOL may be the way in which success is defined: freedom from recurrence of AF episodes lasting more than 30 seconds. This end point was established by guidelines⁴ to allow for consistent trial reporting, but a 30-second recurrence of AF is rarely clinically significant, particularly if the preablation AF burden is high. When success is measured by reduction in AF burden, the beneficial effect of AF ablation on QOL is more apparent. In STAR AF II, for example, the success rate was 49% when defined as freedom from AF episodes lasting more than 30 seconds. However, this analysis showed that most patients had residual AF burdens of less than 20%. In the recent Cryoballoon vs Irrigated Radiofrequency Catheter Ablation: Double Short vs Standard Exposure Duration (CIRCA DOSE) study, freedom from recurrence of AF episodes lasting more than 30 seconds was only approximately 50% as well,¹⁷ but all patients had implantable loop recorders and AF burden was reduced by more than 99% after ablation.

The current study found that residual AF burden reduction of less than 70% was associated with a change in QOL, more specifically the PCS. The 80% and 90% thresholds for AF burden reduction were not found to be significantly associated with change because even patients with a reduction threshold below 80% or 90% could still have improvement in QOL. Reduction in AF burden was not associated with MCS changes, but mental health measures might be less sensitive to AF recurrences than physical health measures, as has been reported in previous studies.^13,18 The baseline score was consistently associated with all QOL measures (MCS plus PCS and MCS and PCS individually) and is well known to be statistically significantly associated with the final QOL score regardless of the intervention.^16,19,20

The exact cutoff for residual AF burden after ablation that indicates improved QOL is not well known.¹⁹ In a substudy from the STAR AF I trial, patients who experienced up to 2.3 hours of AF per month still had significant improvement in QOL.²⁰ Through the use of continuous AF monitoring, Björkenheim et al²¹ showed that individuals with ongoing AF burden greater than 0.5% experienced a significant improvement in QOL even though their episodes lasted longer than 30 seconds. When catheter ablation was compared with antiarrhythmic drugs in patients with different types of AF, ablation was associated with a consistently greater improvement in QOL despite ongoing AF, albeit with smaller burdens.^6,7

This study has limitations. The estimate of AF burden was based on frequent Holter and TTM monitoring rather than using continuous implantable monitoring. However, we used the highest value of AF burden extracted from clinical reporting forms, TTMs, and Holter monitors to avoid underestimation of the postablation burden. Furthermore, a prior analysis of the Atrial Fibrillation and Congestive Heart Failure (AF-CHF) trial showed that AF burden could be estimated from intermittent 12-lead electrocardiograms.²² Another limitation is the use of generic QOL questionnaires (SF-36 and EQ-5D-3L). These tools have limited resolution to determine symptoms related to AF.^21,23 Scales such as the Atrial Fibrillation Effect on QualiTy of Life questionnaire and the Mayo AF-Specific Symptom Inventory are AF specific and have greater discriminatory power for AF-related QOL, but these were not widely used at the time the STAR AF II trial was conceived. Recent studies have used AF-specific scales that show correlation with generic QOL questionnaires.^13,24 In addition, 84 patients (15%) did not complete all of their QOL assessments and were excluded from analysis. However, the remaining sample size was large, so this did not likely influence the findings. The results are consistent with the previous CAPTAF^7,9 and MANTRA-PAF trials.^13,14

In this study, ablation for AF was associated with improved QOL regardless of the ablation strategy and the success of the procedure when defined as absence of recurring AF episodes lasting more than 30 seconds. Greater reductions in AF burden were associated with more QOL subscales showing improvement. A cutoff of 70% reduction in AF burden was associated with a significant improvement in QOL in both univariable and multivariable analyses.

Accepted for Publication: August 28, 2020.

Published: December 4, 2020. doi:10.1001/jamanetworkopen.2020.25473

Open Access: This is an open access article distributed under the terms of the CC-BY-NC-ND License. © 2020 Terricabras M et al. JAMA Network Open.

Corresponding Author: Atul Verma, MD, Southlake Regional Health Centre, University of Toronto, 602-581 Davis Dr, Newmarket, Ontario L3Y 2P6, Canada (atul.verma@utoronto.ca).

Author Contributions: Drs Terricabras and Verma had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Terricabras, Mantovan, Jiang, Chen, Macle, Weerasooriya, Nardi, Sanders, Verma.

Acquisition, analysis, or interpretation of data: Terricabras, Mantovan, Betts, Chen, Deisenhofer, Macle, Morillo, Haverkamp, Weerasooriya, Albenque, Menardi, Novak, Verma.

Drafting of the manuscript: Terricabras, Macle, Verma.

Critical revision of the manuscript for important intellectual content: Terricabras, Mantovan, Jiang, Betts, Chen, Deisenhofer, Macle, Morillo, Haverkamp, Weerasooriya, Albenque, Nardi, Menardi, Novak, Sanders.

Statistical analysis: Macle, Verma.

Obtained funding: Verma.

Administrative, technical, or material support: Betts, Weerasooriya.

Supervision: Terricabras, Mantovan, Morillo, Albenque, Nardi, Menardi, Sanders, Verma.

Conflict of Interest Disclosures: Dr Betts reported receiving grants from St Jude Medical outside the submitted work. Dr Chen reported receiving grants from Medtronic and personal fees from Johnson & Johnson and Abbott outside the submitted work. Dr Macle reported receiving grants from Abbott and Biosense Webster and personal fees from Biosense Webster and BMS-Pfizer outside the submitted work. Dr Morillo reported serving on the advisory boards of Abbott, Medtronic, Servier, and Bayer; serving as a speaker for Medtronic and Servier; receiving personal fees from Abbott, Medtronic, Servier, and Bayer; and receiving grants from Novartis outside the submitted work. Dr Albenque reported receiving grants and consulting fees from Abbott and grants from Biosense Webster outside the submitted work. Dr Sanders is supported by a practitioner fellowship from the National Health and Medical Research Council of Australia and the National Heart Foundation of Australia and reported receiving a clinical trials contract from Abbott Medical; receiving grants from Microport, Medtronic, Boston Scientific, and Abbott Medical; and serving on the advisory boards of Medtronic, Abbott Medical, CathRx, Boston Scientific, and PaceMate outside the submitted work. Dr Verma reported receiving grants from Medtronic, Biosense Webster, Bayer, and Biotronik and personal fees from Servier outside the submitted work. No other disclosures were reported.

Funding/Support: This research was funded by St Jude Medical, Inc (now Abbott Laboratories).

Role of the Funder/Sponsor: St Jude Medical (Abbott Laboratories) had no role in the design of the study but was responsible for the conduct, data collection, data management, and data analysis of the study and approved the manuscript in addition to the authors.

Group Information: The following are the STAR AF II Investigators: Atul Verma, MD, Southlake Regional Health Centre, Newmarket, Ontario, Canada; Rukshen Weerasooriya, MBBS, University of Western Australia, Perth, Australia; Jonathan Kalman, MBBS, PhD, The Royal Melbourne Hospital and University of Melbourne, Melbourne, Victoria, Australia; Prashanthan Sanders, MBBS, PhD, Centre for Heart Rhythm Disorders, Royal Adelaide Hospital and University of Adelaide, Adelaide, Australia; Stuart Thomas, BMed, PhD, Westmead Hospital, Sydney, Australia; John Hayes, MBBS, St Andrew’s War Memorial Hospital, Queensland, Australia; Helmut Purerfellner, MD, Department of Electrophysiology, Academic Teaching Hospital, Ordensklinikum Linz Elisabethinen, Linz, Austria; Maximo Rivero-Ayerza, MD, PhD, Department of Cardiology, Ziekenhuis Oost-Limburg, Genk, Belgium; Paul Novak, MD, Department of Medicine, Royal Jubilee Hospital, Victoria, British Columbia, Canada; Jean Champagne, MD, Université Laval, Quebec City, Quebec, Canada; Laurent Macle, MD, Montreal Heart Institute, Department of Medicine, Université de Montréal, Montreal, Quebec, Canada; Ratika Parkash, MD, QEII Health Sciences Centre, Dalhousie University, Halifax, Nova Scotia, Canada; Allan Skanes, MD, London Heart Institute, Western University, London, Ontario, Canada; Vidal Essebag, MD, PhD, Division of Cardiology, McGill University Health Center, Montreal, Quebec, Canada; Jean-Francois Roux, MD, Department of Medicine, Centre Hospitalier Universitaire de Sherbrooke, Sherbrooke, Quebec, Canada; Carlos A. Morillo, MD, Population Health Research Institute, McMaster University, Hamilton, Ontario, Canada; Pablo Nery, MD, Division of Cardiology, University of Ottawa Heart Institute, Ottawa, Ontario, Canada; Felix Ayala-Paredes, MD, PhD, Department of Medicine, Centre Hospitalier Universitaire de Sherbrooke, Sherbrooke, Quebec, Canada; Franck Molin, MD, Electrophysiology Division, Institut Universitaire de Cardiologie et de Pneumologie de Québec, Québec, Canada; Jiang Chen-yang, MD, Department of Cardiology, Sir Run Run Shaw Hospital, College of Medicine, Zhejiang University, Hangzhou, China; Chen Minglong, MD, Jiangsu Province Hospital, The First Affiliated Hospital of Nanjing Medical University, Nanjing, China; Shu-lin Wu, MD, Guangdong Cardiovascular Institute, Guangdong General Hospital, Guangzhou, China; Jiang Cao, MD, Department of Cardiovascular Diseases, Changhai Hospital, Second Military Medical University, Shanghai, China; Jean-Paul Albenque, MD, Clinique Pasteur, Toulouse, France; Dominique Babuty, MD, PhD, Service de Cardiologie, Faculté de Médecine, Université François-Rabelais, CHU Trousseau, Tours, France; André Pisapia, MD, Unité de Rythmologie Interventionnelle, Hôpital Saint Joseph Marseille, France; Jean-Pierre Cebron, MD, Nouvelles Cliniques Nantaises, Nantes, France; Franck Halimi, MD, Department of Cardiology, Hospital Prive Parly 2, Le Chesnay, Paris, France; Hervé Poty, MD, Clinique la Protestante, Rhone, France; Thierry Chalvidan, MD, Service de Cardiologie, Hôpital Sainte-Marguerite, CHU Marseille, France; Jean Luc Pasquie, MD, Department of Cardiology, CHU Montpellier, Montpellier, France; Wilhelm Haverkamp, MD, PhD, Department of Cardiology, Charité-Universitaetsmedizin Berlin, Campus Virchow-Klinikum, Berlin, Germany; Isabel Deisenhofer, MD, Deutsches Herzzentrum München, Munich, Germany; Markus Zarse, MD, Märkische Kliniken GmbH, Department of Cardiology and Angiology, Klinikum Luedenscheid, Germany; Thomas Kleemann, MD, Klinikum Ludwigshafen, Medizinische Klinik B, Ludwigshafen, Germany; Armin Sause, MD, Department of Cardiology, HELIOS University Hospital Wuppertal, Wuppertal, Nordrhein-Westfalen, Germany; Malte Kuniss, MD, Department of Cardiology, Kerckhoff Clinic, Bad Nauheim, Germany; Stefano Nardi, MD, Department of Cardiology, Pineta Grande Hospital, Castel Volturno, Italy; Endrj Menardi, MD, Santa Croce e Carle Hospital Cuneo, Cuneo, Italy; Massimo Mantica, MD, Istituto Clinico Sant’Ambrogio, Milan, Italy; Maria Grazia Bongiorni, MD, University Hospital of Pisa, Pisa, Italy; Vittorio Calzolari, MD, Department of Cardiology, Ca’ Foncello, Civil Hospital, Treviso, Italy; Giuseppe De Martino, MD, Cardiocentro Clinica Mediterranea, Napoli, Italy; Fulvio Bellocci, MD, Division of Cardiology, Catholic University of the Sacred Heart, Rome, Italy; Young-Hoon Kim, MD, PhD, Korea University, Seoul, South Korea; Mats Jensen-Urstad, MD, PhD, Department of Cardiology, Karolinska Institutet, Karolinska University Hospital, Stockholm, Sweden; Timothy R Betts, MD, Oxford University Hospitals NHS Trust, Oxford, United Kingdom; Derek Connelly, MD, Department of Cardiology, Leeds General Infirmary, Leeds, United Kingdom; Christopher Pepper, MD, Golden Jubilee National Hospital, Agamemnon Street, Glasgow, United Kingdom.

Data Sharing Statement: See Supplement 3.

Meeting Presentation: This substudy was presented at the Heart Rhythm Society Scientific Sessions; May 13, 2017; Chicago, Illinois.

Additional Contributions: Emily Zhang, MSc, Amanda DeGraw, PhD, and Chris Williams, BSc (St Jude Medical, Inc/Abbott Laboratories), helped with statistical analysis and did not receive compensation outside their usual salaries. We thank the investigators of the STAR AF II trial.