Author + information
- Received May 29, 2018
- Revision received September 24, 2018
- Accepted September 27, 2018
- Published online December 17, 2018.
- Gabor Sandorfi, MDa,
- Moises Rodriguez-Mañero, MD, PhDb,c,d,
- Johan Saenen, MD, PhDa,
- Aurora Baluja, MD, PhDe,
- Wim Bories, MSBME, MSE, CEPSa,
- Wim Huybrechts, MDa,
- Hielko Miljoen, MDa,
- Lien Vandaele, BSa,
- Hein Heidbuchel, MD, PhDa and
- Andrea Sarkozy, MD, PhDa,∗ ()
- aCardiology Department, University Hospital of Antwerp, University of Antwerp, Antwerp, Belgium
- bCardiology Department, Complejo Hospital Universitario de Santiago, Santiago de Compostela, Spain
- cInstituto de Investigación Sanitaria (IDIS), Universidad de Santiago de Compostela, Santiago de Compostela, Spain
- dCentro de Investigación Biomédica en Red de Enfermedades Cardiovasculares (CIBERCV CB16/11/00226-CB16/11/00420), Santiago de Compostela, Spain
- eAnesthesiology Department, Hospital Universitario Santiago de Compostela, Santiago de Compostela, Spain
- ↵∗Address for correspondence:
Dr. Andrea Sarkozy, Cardiology Department, University Hospital Antwerp, University of Antwerp, Wilrijkstraat 10, 2650 Edegem, Belgium.
Objectives This study investigated whether real-world use of contemporary technologies changed pulmonary vein (PV) reconnection and redo pulmonary vein isolation (PVI) procedure frequencies.
Background Previous studies consistently reported that following PVI recurrence of PV conduction is observed in >80% of patients.
Methods Consecutive patients undergoing 529 first and/or redo radiofrequency point-by-point PVI between January 2013 and December 2016 were included.
Results Between 2013 and 2016, redo PVI rate in atrial fibrillation significantly decreased (p < 0.001); in ≤12 months, first redo PVI rate decreased from 19% to 4%. The percentage of patients having PV reconnection at second PVI significantly decreased from 90% to 29% (p = 0.001). One PVI was performed in 393 and >1 in 79 patients. Female sex was associated with >1 PVI (hazard ratio [HR]: 1.86; 95% confidence interval [CI]: 1.10 to 3.13; p = 0.02). Sixty patients underwent first and second PVI in the study period. Female sex (HR: 2.79; 95% CI: 1.67 to 4.64; p < 0.001) and left atrial diameter (HR: 1.05; 95% CI: 1.01 to 1.08; p = 0.01) were associated with more and use of automatic ablation annotation algorithm during first PVI with fewer (HR: 0.54; 95% CI: 0.32 to 0.92; p = 0.02) redo PVI procedures. In 31 of 60 patients, ≥1 PV was reconnected at second PVI. The need for “touch-up” applications at the first PVI was the only predictor of PV reconnection.
Conclusions Redo rate in atrial fibrillation and PV reconnection at redo PVI significantly decreased in recent years. Male sex, left atrial diameter, and use of automatic ablation annotation algorithm at first PVI were associated with fewer redo procedures. First-pass isolation was associated with lower PV reconnection rate at second procedure. Female sex was associated with more redo procedures but not higher PV reconnection frequencies.
The clinical efficacy of pulmonary vein isolation (PVI) for atrial fibrillation (AF) is limited by arrhythmia recurrences due to PV reconnection and/or non-PV triggers and substrate (1). Reconnection in PV conduction indicates a lack of durable contiguous and transmural lesion formation. Contact between catheter tip and tissue, duration, power, impedance, and temperature have all been shown to be an important determinant of lesion size and depth during point-by-point radiofrequency (RF) catheter ablation (1). Incorporation of a contact force (CF) measuring sensor at the distal tip of the ablation catheter in 2012 provided real-time CF feedback and in initial studies led to a significant reduction in procedural, RF ablation, and fluoroscopy times, as well as acute PV reconnection rate (1). In the EFFICAS I (Efficacy Study on Atrial Fibrillation Percutaneous Catheter Ablation With Contact Force Support) study certain CF parameters were highly predictive of gap presence and PV reconnection at the repeat electrophysiological study. On the basis of these results, CF targets were proposed in 2013 (2). In 2015, an automated ablation annotation algorithm was introduced into clinical practice, allowing the placement of continuous effective lesions, leading to higher first-pass isolation and lower acute PV reconnection rate in initial studies (3). In the last 5 years, intravenous adenosine testing has also been proposed to unmask dormant conduction to further reduce PV reconnection (1,4).
In the past decade, studies without CF-sensing catheters have consistently reported that following circumferential PVI among patients undergoing repeat PVI ablation procedure, recurrence of PV conduction is observed in >80% of the patients (5,6). The aim of our study was to investigate whether the introduction of contemporary technologies in a real-word academic single-center practice changed PV reconnection frequencies during the redo procedure following wide circumferential antral PVI with point-by-point RF catheter ablation.
We screened 499 consecutive patients with symptomatic paroxysmal or persistent AF undergoing first and/or redo PVI RF catheter ablation between January 2013 and December 2016 at the University Hospital Antwerp (Figure 1), and 472 patients were included in this retrospective study. All 472 patients undergoing 529 first and/or redo PVI procedures between 2013 and 2016 were followed for redo procedures until December 2017. In 12 patients the first PVI was performed between 2013 and 2016 and the redo PVI in 2017. Additional redo procedures performed in patients with confirmed persistent PVI at a previous redo procedure were excluded from further analysis.
The PVI technique used in the study period has been previously described in detail (7). Briefly, all procedures were performed under general anesthesia. For ablation, irrigated tip ablation catheter (Thermocool NaviStar, Biosense Webster, Diamond Bar, California) with and without CF-sensing technology was used. RF lesions were placed in temperature-limited power control mode at the antrum of the ipsilateral PV (30 W posterior and 35 W anterior wall). The target duration of RF delivery ranged between 30 (posterior left atrium [LA]) and 60 s (anterior PV antrum) but was terminated prematurely if local electrogram attenuation was >80% or >30 Ω impedance drop or >10 Ω increase, or if sudden catheter movement occurred. From January 2013 on, RF applications were performed with the catheter in a stable position and a numeric CF value of ≥10 g. The target numeric CF was 20 g (range 10 g to 30 g). In September 2013, the Thermocool SmartTouch catheter was temporarily withdrawn worldwide from clinical use. In the recall period, irrigated-tip ablation catheter without CF measurement (NaviStar Thermocool) was used as previously described (8). Following circumferential ablation of a unilateral PV pair, entry and exit block of PVs were assessed with a circular mapping catheter. If the PVs were not isolated, initially antral “touch-up” lesions were performed on the basis of entry and exit mapping. More ostial applications were placed if attempts failed to close the remaining gaps at the antrum. Routine adenosine testing was introduced in May 2014. A bolus of intravenous adenosine was administered (12- to 24-mg dose to induce atrioventricular block). In case of adenosine-induced PV reconnection, ablation was performed until entry and exit block was achieved. An automatic ablation annotation algorithm (Visitag, Biosense Webster) was introduced in August 2014. Catheter stability during the study period was set at 2 mm for >7 s and minimum CF of 6 g to 8 g >50% of the application time was used as a standard setting. A 5-mm distance between applications was targeted. Between August 2015 and March 2016, a minimum force-time integral (FTI) of 400 g/s was targeted, and from April 2016 on, at the anterior ridge a minimum FTI of 500 g/s and in the posterior LA segments, a minimum FTI of 300 g/s were targeted. The target duration of RF delivery ranged between 30 and 60 s but was terminated prematurely if no automatic annotation was achieved due to low or unstable contact or if electrogram attenuation was >80% or >30 Ω impedance drop or if >10 Ω impedance rise was observed. Application time was prolonged if necessary to achieve the target minimum FTI value. Details of the repeat PVI procedure and the number of patients undergoing PVI with each new technique are described in the Online Appendix (Online Table 1).
Continuous variables were expressed as mean ± SD and categorical values as frequencies (percentages). The 2-sided unpaired Student t-test was used to compare continuous variables when values were normally distributed or its nonparametric equivalent (Mann-Whitney U test) when variables were not normally distributed. The chi-square or Fisher exact test was used to compare categorical variables. Kaplan-Meier survival analysis was performed to analyze the effect of the year the first PVI was performed to redo PVI–free survival during follow-up. Patients were categorized in 4 groups according to the year of the first PVI between 2013 and 2016 (group 1: first PVI in 2013, group 2: first PVI in 2014, group 3: first PVI in 2015, and group 4: first PVI in 2016). Outcome variable was redo PVI during follow-up. Time was calculated from the first PVI to the first redo PVI or the last follow-up. To assess the contribution of baseline patient and first procedural characteristics to: 1) need for redo PVI procedure and 2) PV reconnection at the time of the redo procedure (outcome of the model was the binary outcome for redo PVI and PV isolation at the time of the redo), multivariable Cox proportional hazard regression analysis was used. Only variables with p < 0.10 in univariable analysis were used for the multivariable analysis. A p value of ≤0.05 was considered to be statistically significant. IBM SPSS Statistics version 22 (SPSS Inc., Chicago, Illinois) was used for all statistical analyses.
Temporal changes in PVI procedural characteristics between 2013 and 2016
Between 2013 and 2016, redo PVI frequencies, PV reconnection at second PVI, and fluoroscopy time significantly decreased and RF application time significantly increased (Table 1). Kaplan-Meier analysis showed significantly shorter first redo PVI–free survival in the patients in group 1 (whose first PVI was performed in 2013) compared with patients in group 2 (first PVI 2014), group 3 (first PVI 2015), and group 4 (first PVI in 2016) (Figure 2). The ≤12-month first redo PVI rate in the whole study population steadily decreased from 19% to 11% to 8% to 4% in group 1, group 2, group 3, and group 4, respectively. The ≤12-month first redo PVI rate in paroxysmal AF decreased from 23% to 6% to 10% to 1% in group 1, group 2, group 3, and group 4, respectively.
A total of 472 consecutive patients were included; 393 patients underwent 1 PVI and 79 patients >1 PVI. The patients’ clinical characteristics are depicted in Table 2. Median time to the second PVI was 12 ± 14 months. The median follow-up for redo PVI procedures was 30 ± 12.8 months. Female sex was the only significant clinical predictor for redo procedures in both univariate (hazard ratio [HR]: 2.10; 95% confidence interval [CI]: 1.34 to 3.30; p = 0.001) and multivariate (HR: 1.86; 95% CI: 1.10 to 3.13; p = 0.02) analyses.
Redo PVI procedural data
Nineteen patients had the initial PVI procedure prior to 2013 and were excluded from first procedural characteristics’ analysis. The remaining 48 patients with >1 PVI underwent both first and second PVI between 2013 and 2016. An additional 12 patients underwent a second PVI procedure during the follow-up in 2017. Comparing 60 patients undergoing >1 PVI to 393 patients undergoing 1 PVI during follow-up, male sex, use of Visitag, and adenosine testing during the first PVI procedure were significantly different (Table 3). In univariate analysis, female sex (HR: 2.46; 95% CI: 1.48 to 4.08; p < 0.001), and absence of Visitag use (HR: 0.56; 95% CI: 0.33 to 0.96; p = 0.03) were significantly associated with redo procedures. In multivariate analysis, female sex (HR: 2.79; 95% CI: 1.67 to 4.64; p < 0.001), left atrial diameter (LAD) (HR: 1.05; 95% CI: 1.01 to 1.08; p = 0.01), and the absence of Visitag use (HR: 0.54; 95% CI: 0.32 to 0.92; p = 0.02) were significant predictors of redo procedures.
In 31 of the 60 patients, ≥1 PV was reconnected and 29 patients presented with persistent PVI at the second PVI. Table 4 shows the comparison of baseline clinical characteristics and first procedural data between patients with and without PV reconnection. In univariate analysis, adenosine reconnection (HR: 3.73; 95% CI: 1.34 to 10.4; p = 0.01) and need for “touch-up” applications (HR: 3.81; 95% CI: 1.80 to 80.5; p < 0.001) were significant predictors of PV reconnection, but absence of Visitag use was not (HR: 0.58; 95% CI: 0.28 to 1.23; p = 0.16). In multivariate analysis, the need for touch-up applications (HR: 4.53; 95% CI: 1.60 to 12.8; p = 0.004) remained a significant predictor of PV reconnection.
Women undergoing PVI were significantly older at the time of the first procedure than men were (62 ± 10 vs. 59 ± 10; p = 0.007) and more frequently had hypertension (76 [53%] of 144 women vs. 127 [39%] of 328 men; p = 0.005). Duration of AF (44 ± 46 vs. 48 ± 54 months; p = 0.453) and the number of antiarrhythmic drugs used prior to ablation (2 ± 0.9 vs. 1.9 ± 0.9; p = 0.142) did not differ significantly between women and men. LAD was significantly smaller in women (37 ± 7 vs. 40 ± 6 mm; p < 0.001), and obstructive sleep apnea syndrome was less frequent (39 of 328 [12%] in men vs. 6 of 144 [4%] in women; p = 0.01). There was no significant difference in the proportion of AF types between women and men (paroxysmal AF: 105 [73%] of 144 women vs. 222 [68%] of 328 men).
Comparing first PVI procedural data, the number of RF applications was similar (women: 76 ± 23 vs. men: 81 ± 26; p = 0.122), but the total RF application time was significantly less in women (44 ± 15 vs. 47 ± 13 min; p = 0.036). Major complications occurred as frequently in women as in men (5 [3.6%] of 139 vs. 8 [2.5%] of 314; p = 0.549).
Comparison of first procedural data with and without the use of Visitag
Visitag was used in 329 (73%) of 453 first PVI procedures. The use of Visitag was associated with significant (35%) increase in RF application time (Online Table 2).
The main findings of our study are the following: 1) between 2013 and 2016, the redo PVI rates significantly decreased in AF; 2) between 2013 and 2016, the finding of PV reconnection at second PVI significantly decreased; 3) male sex, smaller LA size, and the use of Visitag were associated with less redo PVI procedures; 4) first-pass isolation at initial PVI was associated with lower PV reconnection at second procedure; and 5) female sex was associated with more redo procedures but not higher PV reconnection frequencies.
Although studies of patients undergoing redo PVI procedures in the past decade used different measures of reconnection (as reconnection rate per PV or PV pair or patient), they all consistently reported high PV reconnection both in paroxysmal and persistent AF. The majority of these studies reported PV reconnection from 80% up to 95% of patients undergoing repeat procedure for recurrent symptomatic atrial tachyarrhythmias (1,5,6,9,10). For ethical reasons, fewer data are available on PV reconnection in patients without recurrent symptomatic atrial tachyarrhythmias. In 2 recent studies following PVI without the use of CF measurement to guide ablation, all patients underwent repeat assessment of PV conduction at 3 months of follow-up. In the randomized Gap-AF (Gap-Atrial Fibrillation-German Atrial Fibrillation Competence Network 1 Trial) in the complete PVI arm, PV reconnection was observed in 70% of patients (11). In the smaller EFFICAS I study, the reassessment showed PV reconnection in 65% of the patients (2). These studies prove that PV reconnection without the use of a CF-sensing catheter remains high (65% to 95%) in patients undergoing PVI irrespective of the presence of recurrent arrhythmias.
In 2012, CF-sensing catheters were introduced into clinical practice in an attempt to achieve persistent transmural contiguous lesions and persistent PVI in a single procedure. In the nonrandomized EFFICAS II trial with the use of CF catheters and CF targets, all patients underwent a repeat assessment of PV conduction (12). The proposed CF targets included contiguous lesions with a target CF of 20 g and minimum FTI of 400 g/s. In this study, only 38% of the patients were found to have PV reconnection, suggesting that the use of CF-sensing catheters and following the proposed CF targets can decrease PV reconnection. However, the use of the 2 commercially available CF catheters itself without the use of the proposed EFFICAS CF targets did not improve single-procedure PVI success for paroxysmal AF compared with a non-CF-sensing catheter in 2 multicenter randomized U.S./European (no CF target) (13) and U.K. (target CF range between 5 g and 40 g) (14) trials. In our study, a CF-sensing catheter was used in 96% of the patients, and only 16 patients underwent PVI without use of CF catheter.
Following the introduction of CF-sensing catheters, a small clinical study suggested that acute and dormant PV reconnection was dependent not only on a mean CF <10 g and mean FTI <400 g/s, but also on an interlesion distance >5 mm (15). In 2014, an automatic annotation algorithm (Visitag) was introduced into routine clinical practice in Europe to allow better operator control of both CF and FTI target parameters, stability of catheter position during RF application, and interlesion distance. The maximal range of catheter movements in a given time, the minimum CF value, and target FTI value of each application are programmable. Visitag without the use of CF parameters was reported in a preliminary single-center study to increase first-pass isolation and decrease acute reconnection through reducing incidence of gaps (3). In a recent small single-center study, the importance of minimum CF, FTI, and ablation index parameters and automated interlesion distance >5 mm in predicting acute and late PV reconnection was confirmed (16). In another recent single-center study, the use of Visitag with the use of minimum CF and FTI parameters significantly increased 1-year success rate as compared to the control group (17). In all of these studies, different settings of Visitag, minimum CF, and FTI targets were used. In another recent study, patients underwent PVI with the use of Visitag but without the use of minimum target FTI parameters and with a CF target of 5 g to 40 g (18). All patients underwent a repeat study, and in 62% of the patients, PV reconnection was observed. The minimum FTI and ablation index of a given segment were predictive of late reconnection, and higher minimum FTI and ablation index values were required to avoid reconnection in anterior/roof segments than were necessary for posterior/inferior segments. The high reconnection rate in this study together with the results of the randomized CF-sensing catheter studies suggests that the use of Visitag targeting 5-mm interlesion distance, targeting a minimum CF of 10 g, and region-dependent target minimum FTI or ablation index value are all important and necessary to achieve persistent PVI. In the current study, we used Visitag in 73% of patients undergoing first PVI with strict stability, minimum CF, and region-dependent minimum target FTI settings. The use of Visitag with these settings at the initial PVI was associated with fewer redo procedures. Interestingly, the use of Visitag was associated with a significant 35% increase in RF time. The increase is likely due to the prolongation of application times to reach the minimum targeted FTI values and remained consistent throughout the study period. The increased RF time was not associated with a higher complications rate.
Intravenous adenosine in sufficient dose can differentiate between permanent conduction block and dormant conduction (19). Unmasking dormant PV reconnection followed by repeat ablation of these sites was proposed to decrease PV reconnection and increase clinical success in a randomized study (4). In a later, larger randomized study with higher dose of adenosine administered later after PVI, these findings could not be reproduced (20). Importantly, none of these studies used CF catheters for PVI. In the current study, adenosine reconnection at the initial PVI was associated with PV reconnection at the second procedure in univariate but not in multivariate analysis. However, these results should be interpreted with caution as adenosine and Visitag were introduced into our routine practice in the same year. As a consequence, the majority of patients (82%) who underwent Visitag-guided ablation also received adenosine testing.
In our study, the single predictor of PV reconnection at second procedure was the need for touch-up applications following the first circumferential applications. First-pass isolation indicates high quality of the initial PVI applications and may predict permanent contiguous lesion formation. Our study results confirm the clinical usefulness of this less widely used parameter.
The purpose of the current study was to investigate whether with the real-world clinical use of contemporary technologies, PV reconnection frequencies have changed in the last years. Contemporary technologies included the combination of a widely available new automatic ablation annotation algorithm followed by adenosine testing and CF targets. We used as the control group consecutive patients undergoing PVI with CF catheter the years prior to the introduction of Visitag and adenosine testing. Our results show that in the last years an important switch has occurred (Figure 3). Currently, at second PVI procedures we can expect persistent PVI instead of PV reconnection in the majority of patients. This finding has important clinical implications. In the past, at the second procedure as PV reconnection was the rule, especially in paroxysmal AF, reisolation of the PV was usually the chosen straightforward strategy. Today, as in up to 70% of the patients at second procedure persistent PVI can be expected, the operators should carefully plan the redo procedure following informed discussion with the patient. The advantages and disadvantages of alternative strategies, such as non-PV trigger evaluation with provocative tests, rotational and focal activation, and/or substrate mapping-based ablation technologies, should be considered.
PVI redo rate
The clinical success of PVI has consistently been higher for paroxysmal than for persistent AF and recently was estimated at between 50% and 85% in the first years following a single procedure (1). The higher success of PVI is thought to be due to the more important role of the PV in the initiation and maintenance of paroxysmal AF. In persistent AF, non-PV triggers and substrate are thought to have an additional important role in the initiation and maintenance of AF (1). Decrease in PV reconnection rate is accepted to increase single procedural success and decrease redo procedure rates, especially in paroxysmal AF. The need for redo procedures is, from patients’ perspective, an important measure of clinical outcome. In the current study, we report that with the use of contemporary technologies, the redo rate in recent years in AF has significantly decreased to <10%, likely mirroring the lower PV reconnection rate following the initial PVI procedure (Figure 2).
Importance of female sex
In our study, fewer women (31%) than men were referred for PVI procedures. Women were older with smaller LAD, more frequently had hypertension, and less frequently had sleep apnea. These findings are very similar to those of several other large registries (1,21–23). There are several reasons why fewer women undergo PVI. Referral bias is among them. In contrast to the large registries, in our study, late referral does not seem to play a main role, as although women were older, they had the same AF duration, number of antiarrhythmic drugs, proportion of persistent AF and amiodarone use, and smaller LAD as men.
In our study, female sex was associated with lower RF application time and higher number of redo PVI procedures. Shorter RF application time in women has also been reported in some of the large registries (21). Intuitively the smaller body and LA size in women may be associated with thinner LA wall and may lead to shorter applications by the operators to increase safety. Higher recurrence rate of atrial arrhythmias has been reported in some large studies (21) but not in others. Currently, it seems that there is no significant sex-related difference in success of PVI (1). However, in our study, female sex was associated with higher redo frequencies. Interestingly, during the redo procedures, we did not find more PV reconnection in women than in men. These findings suggest that women may more frequently have extra-PV foci initiating AF in the setting of persistent PVI, as described previously (24,25). Alternatively, the older age and higher incidence of hypertension in women may lead to more extensive atrial fibrosis and extra-PV substrate, decreasing the success of the PVI approach.
In the current study, women had femoral access site or cardiac complications at the same frequency as men. This finding is in contrast to some of the large registries in which the complication rate was higher in female subjects but is in line with several other registries reporting similar incidence of procedure-related complications between male and female subjects (21).
In the current study, it was not possible to differentiate which of the applied novel techniques had the most important role in decreasing PV reconnection and redo frequencies. However, due to the very fast pace of development and introduction of new technologies into routine clinical practice and the costs, additional randomized studies comparing the use of CF-sensing catheters with or without specific CF targets or with and without adenosine and the use of CF-sensing catheters with or without automatic annotation algorithm are unlikely. We believe, therefore, similar to the current study, single- and multicenter registries are important to provide insights into current real-word clinical outcomes. We are reporting outcomes of redo procedures. The decision to perform a redo procedure is multifactorial and is dependent on several physician- and patient-dependent individual factors. However, we used as the control group consecutive patients from the same center performed by the same operators to allow comparison of procedures with and without new technologies. In the current study, we included a mixture of paroxysmal and persistent AF patients. Throughout, we use the nomenclature PVI and redo PVI procedure. Technically, a redo procedure (more frequently in persistent AF) in a patient with the finding of persistent PVI is an AF ablation rather than a PVI procedure. However, in the current study, additional third or fourth ablation procedures, mostly in persistent AF patients with proven PVI at a previous procedure, were excluded.
In the current single-center retrospective study, we report that between 2013 and 2016, the redo PVI rate significantly decreased in AF as well as the finding of PV reconnection at the second PVI. Male sex, smaller LA size, and the use of Visitag followed by adenosine testing at the first PVI were associated with fewer redo procedures. First-pass isolation at the initial PVI were associated with lower PV reconnection rate at the second procedure. Female sex was associated with more redo PVI procedures but not higher PV reconnection rate.
COMPETENCY IN MEDICAL KNOWLEDGE: In the past 5 years with the use of contemporary RF catheter ablation technologies, both redo following first PVI and PV reconnection rates at second PVI procedure, have significantly decreased.
TRANSLATIONAL OUTLOOK: Further prospective single- and multicenter studies are necessary to determine the current efficacy and safety of modern catheter ablation technologies in the treatment of AF.
Dr. Sandorfi received a clinical fellowship grant from Biosense Webster Inc. and the European Heart Rhythm Association. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
All authors attest they are in compliance with human studies committees and animal welfare regulations of the authors’ institutions and Food and Drug Administration guidelines, including patient consent where appropriate. For more information, visit the JACC: Clinical Electrophysiology author instructions page.
- Abbreviations and Acronyms
- atrial fibrillation
- contact force
- confidence interval
- force-time integral
- hazard ratio
- left atrium
- left atrial diameter
- pulmonary vein
- pulmonary vein isolation
- Received May 29, 2018.
- Revision received September 24, 2018.
- Accepted September 27, 2018.
- 2018 American College of Cardiology Foundation
- Calkins H.,
- Hindricks G.,
- Cappato R.,
- et al.
- Neuzil P.,
- Reddy V.Y.,
- Kautzner J.,
- et al.
- Verma A.,
- Kilicaslan F.,
- Pisano E.,
- et al.
- Quyang F.,
- Antz M.,
- Ernst S.,
- et al.
- Sarkozy A.,
- Shah D.,
- Saenen J.,
- et al.
- Wolf M.,
- Saenen J.,
- Bories W.,
- et al.
- Quyang F.,
- Tilz R.,
- Chun J.,
- et al.
- Kuck K.H.,
- Hoffmann B.A.,
- Ernst S.,
- et al.,
- for the Gap-AF-AFNET Investigators
- Reddy V.Y.,
- Dukkipati S.R.,
- Neuzil P.,
- et al.
- Ullah W.,
- McLean A.,
- Tayebjee M.H.,
- et al.,
- for the UK Multicentre Trials Group
- El Haddad M.,
- Taghji P.,
- Phlips T.,
- et al.
- Tanaka N.,
- Inoue K.,
- Tanaka K.,
- et al.
- Das M.,
- Loveday J.J.,
- Wynn G.J.,
- et al.
- Kobori A.,
- Shizuta S.,
- Inoue K.,
- et al.,
- for the UNDER-ATP Trial Investigators
- Zylla M.M.,
- Brachmann J.,
- Lewalter T.,
- et al.
- Avgil Tsadok M.,
- Gagnon J.,
- Joza J.,
- et al.
- Patel N.,
- Deshmukh A.,
- Thakkar B.,
- et al.
- Lee S.H.,
- Tai C.T.,
- Hsieh M.H.