Author + information
- Received December 30, 2012
- Revision received March 9, 2013
- Accepted March 26, 2013
- Published online July 2, 2013.
- Daniel Steven, MD∗∗ (, )
- Arian Sultan, MD∗,
- Vivek Reddy, MD†,
- Jakob Luker, MD∗,
- Manuel Altenburg, MD∗,
- Boris Hoffmann, MD∗,
- Thomas Rostock, MD‡,
- Helge Servatius, MD∗,
- William G. Stevenson, MD§,
- Stephan Willems, MD∗ and
- Gregory F. Michaud, MD§
- ∗University Heart Center Hamburg, Department for Cardiac Electrophysiology, Hamburg, Germany
- †Helmsley Electrophysiology Center, Mount Sinai School of Medicine, New York, New York
- ‡Department of Medicine 2, University Medical Center Mainz, Mainz, Germany
- §Brigham and Women's Hospital, Cardiac Arrhythmia Service, Boston, Massachusetts
- ↵∗Reprint requests and correspondence:
Dr. Daniel Steven, University Heart Center Hamburg, Martinistrasse 52, Hamburg, Germany.
Objectives This study was conducted to determine if an additional procedural endpoint of unexcitability (UE) to pacing along the ablation line reduces recurrence of atrial fibrillation (AF) or atrial tachycardia (AT) after radiofrequency catheter ablation.
Background AF/AT recurrence is common after pulmonary vein isolation (PVI).
Methods We included 102 patients from 2 centers (age 63 ± 10 years; 33 women; left atrium 38 ± 7 mm; left ventricular ejection fraction 61 ± 6%) with symptomatic paroxysmal AF. A 3-dimensional mapping system and circumferential mapping catheter were used in all patients for PVI. In group 1 (n = 50), the procedural endpoint was bidirectional block across the ablation line. In group 2 (n = 52), additional UE to bipolar pacing at an output of 10 mA and 2-ms pulse width was required. The primary endpoint was freedom from any AF/AT (>30 s) after discontinuation of antiarrhythmic drugs.
Results Procedural endpoints were successfully achieved in all patients. Procedure duration was significantly longer in group 2 (185 ± 58 min vs. 139 ± 57 min; p < 0.001); however, fluoroscopy times were not different (23 ± 9 min vs. 23 ± 9 min; p = 0.49). After a follow-up of 12 months in all patients, 26 patients (52%) in group 1 versus 43 (82.7%) in group 2 were free from any AF/AT (p = 0.001) after a single procedure. No major complications occurred.
Conclusions The use of pacing to ensure UE along the PVI line markedly improved near-term single-procedure success, compared with demonstration of bidirectional block alone. This additional endpoint significantly improved patient outcomes after PVI. (Unexcitability Along the Ablation as an Endpoint for Atrial Fibrillation Ablation; NCT01724437)
Because ectopy from the pulmonary veins is primarily responsible for initiation and maintenance of paroxysmal atrial fibrillation (PAF), pulmonary vein isolation (PVI) has become the standard treatment of this arrhythmia (1–3). Arrhythmia recurrence is high, however, especially during longer follow-up (4). Data suggest that dormant conduction gaps on the ablation line are the most common reasons for arrhythmia recurrence (5,6). To date, no successful strategy to decrease the likelihood of conduction gaps using point-by-point catheter ablation technology has been reported in a prospective, randomized trial. Previously published results suggest that additional radiofrequency (RF) application guided by atrial tissue excitability may be helpful to identify potential conduction gaps in atrial myocardium (7,8). The present prospective, randomized 2-center trial was conducted to evaluate whether the endpoint of unexcitability along the ablation line is as safe and more effective in preventing arrhythmia recurrence than electrical PVI alone.
The aim of the study was to assess the near-term (12 months) efficacy of PVI using a standard approach compared with application of an additional acute procedural endpoint of unexcitability along the ablation line.
A total of 103 patients (age 63 ± 10 years; 34 women; left atrium [LA] 38 ± 7 mm; left ventricular ejection fraction 61 ± 6%) referred for catheter ablation for symptomatic PAF who had failed at least 1 type I or III antiarrhythmic drug were included in our study. One patient withdrew written informed consent. Ablation was performed in one of 2 participating centers (University Hospital, Hamburg, Germany, or Brigham and Women's Hospital, Boston, Massachusetts) by 1 of 3 surgeons during sinus rhythm in all patients. Randomization to either study arm was conducted before the procedure based on a previously generated computer algorithm (Fig. 1). No patients had evidence of significant structural heart disease (see next section) as assessed by at least 1 imaging study such as echocardiography, computed tomography scan, or magnetic resonance imaging. Baseline parameters are shown in Table 1. There were no statistically significant differences between groups.
Study inclusion/exclusion criteria
Patients were eligible for the study if they: 1) were able and willing to give written, informed consent; 2) had a history of symptomatic PAF (episode duration <48 h; 3) had no prior electrical cardioversion in the year before study inclusion; 4) at least 1 class I or III antiarrhythmic drug had failed; and 5) presented in sinus rhythm. Exclusion criteria consisted of: 1) age <18 years; 2) structural heart disease (hypertrophic cardiomyopathy [septal thickness >13mm], left ventricular ejection fraction ≤35%, significant valvular heart disease, LA size >50 mm); 3) reversible causes of AF; 4) intracardiac thrombus; and 5) inability to take warfarin.
Electrophysiological study and ablation procedure
The patients underwent electrophysiological study and catheter ablation after providing written informed consent. Study protocols were approved by the Brigham and Women's Hospital Human Subject Protection Committee and the Ethical Board Committee of the Ethical Committee of the Medical Association, Hamburg, Germany. All antiarrhythmic drugs except amiodarone were stopped a minimum of 5 half-lives prior to the procedure.
Surface and intracardiac electrocardiograms (ECGs) were digitally recorded and stored (Prucka CardioLab EP system, GE Healthcare, Waukesha, Wisconsin; LabSystem Pro EP recording system, BARD Electrophysiology, Lowell, Massachusetts). Nonfluoroscopic 3-dimensional mapping was performed using the Carto (Biosense Webster, Diamond Bar, California) or Ensite NavX (St. Jude Medical, St. Paul, Minnesota) system at the operator's discretion.
A 7-F multipolar (20-pole) catheter (Daig DuoDeca 2-10-2, St. Jude Medical, or Ismus, Biosense Webster) was used with the distal poles (poles 1 to 10) placed within the coronary sinus and the proximal electrodes (poles 11 to 20) located along the tricuspid annulus in the lateral and inferior right atrium. In some cases, the duodecapolar catheter was introduced through a long guiding sheath (Convoy 55° curve, Boston Scientific, Natick, Massachusetts) to facilitate placement and stability. Alternatively, a 6-F decapolar catheter (St. Jude Medical) was placed within the coronary sinus. A 10- or 20-pole circumferential PV mapping catheter (Optima, Irvine Biomedical, Irvine, California, or Lasso, Biosense Webster) was positioned at the ostium of one of the ipsilateral PVs, usually the superior vein, either before or after PVI. To avoid potential bias with respect to conduction breakthrough prior to PV entrance block, the circumferential mapping catheter (CMC) signals were continuously recorded but not displayed to the operators during deployment of the ablation line or until atrial tissue directly on this line was rendered unexcitable to pacing. The electrogram signals from the CMC were then reviewed, and the catheter was positioned to sequentially assess electrical isolation of the ipsilateral PVs. Entrance block into PVs was confirmed by the complete absence of PV potentials and/or by retrospective review of the PV signals after each ablation lesion to determine when entrance block was achieved. If far-field potentials were seen on the CMC, direct pacing of the suspected far-field source was performed for confirmation. Pacing along the ablation line was performed with careful attention to avoid parallel orientation of the ablation catheter tip to the tissue; this prevented inadvertent pace capture from the proximal electrode of the distal bipole. During LA ablation, heparin was administered intravenously to maintain an activated clotting time of >300 s.
Ablation was performed using an open-irrigated 3.5-mm tip mapping and ablation catheter (Thermocool, Biosense Webster) advanced into the LA via a long sheath (SL0, 8-F, St. Jude Medical) to mechanically support the ablation catheter. Ablation lesions were generated in a power-controlled mode applying 10 to 40 W for 30 to 60 s per lesion during irrigation at a rate of 17 to 30 ml/min. The upper temperature limit was set to 45°C. Esophageal position was monitored fluoroscopically after oral administration of 5 ml of a barium sulfate esophageal cream (E-Z-EM, Lake Success, New York) (8) or with the use of an esophageal temperature probe. RF power was reduced at sites near the esophagus to 10 to 30 W; ablation at sites in direct proximity to the esophagus was avoided, and ablation was terminated before the esophageal temperature exceeded 38.5°C (9). Patients were administered either intravenous conscious sedation (Hamburg) or general anesthesia (Boston). The procedures consisted of PVI only. No infusion of isoproterenol was administered nor were extra PV ectopies observed or ablated in any patient. Patients with documented typical right atrial flutter underwent right atrial isthmus ablation in the same session.
Three-dimensional mapping with registration of a previously performed imaging study (computed tomography scan or magnetic resonance imaging) was used in all cases (Ensite NavX or Carto). Ablation lesions were placed along the antral circumference of the ipsilateral PVs ≥1 cm from the PV ostia. At the anterior portion of the left pulmonary veins, the ablation line was directed at the PV side of the LA appendage (LAA) ridge.
Evaluation of unexcitability along the ablation line
When the anatomic ablation line encircling the ipsilateral veins was completed, pacing was performed during sinus rhythm on the entire ablation line with pacing sites approximately 5 mm apart. Bipolar pacing between the 2 distal electrode pairs of the ablation catheter was performed with a pacing output set to 10 mA at a pulse width of 2 ms (8). Bipolar pacing was chosen based on data from a pre-clinical study that suggested greater predictive value of bipolar versus unipolar loss of pace capture (9).
Catheter stability during pacing was validated by observing the tip position on biplane fluoroscopy and the 3-dimensional mapping system and by monitoring the stability of local electrogram morphology. If atrial tissue was still excitable to pacing at the predetermined output, additional ablation lesions were delivered until unexcitability was achieved at that location. If atrial tissue along the line was still excitable to pacing after RF applications of a duration totaling at least 60 s, RF application was continued at adjacent sites until pacing capture no longer occurred in the region.
Analysis of entrance and exit block from PVs
When possible, the RF lesion associated with PV entrance block during ablation line deployment was determined after encircling RF ablation. PVI was prospectively confirmed in ipsilateral PVs by demonstrating failure of conduction into the veins and local capture of PV potentials with exit block from each PV. The amount of RF energy (Joules) delivered beyond PV entrance block was calculated; these data were not available for 15 of 102 patients.
Bipolar electrograms from the distal and first ring electrode with an interelectrode distance of 1 mm, filtered at 30 to 500 Hz, were recorded from the mapping catheter and stored on digital media (Prucka CardioLab EP or LabSystem Pro EP). Catheter position at each pacing site was marked on the 3-dimensional mapping system. Atrial electrogram width (milliseconds) and amplitude (milliamperes) were measured using electronic calipers and grouped according to excitability or unexcitability at the pre-determined pacing output using the stored procedural data on the recording system in a subset of patients (n = 29).
For identification of sites requiring additional RF applications, the ablation line was schematically divided into 8 segments. RF applications guided by pacing were then assigned to the appropriate segments along the ablation line.
All patients were consistently seen by an electrophysiologist every 3 months after the procedure for at least 12 months. At each visit, 72-h Holter monitoring or 7-day auto-triggered event monitor was provided. After a 3-month blanking period, arrhythmia recurrence was defined as any episode of atrial tachycardia (AT) or AF >30 s according to current expert consensus on AF ablation (10). During the follow-up visits, current antiarrhythmic medications and symptoms of AF such as palpitations and loss of exercise capacity were assessed.
Continuous variables are reported as mean ± SD or as median and range (minimum–maximum) as appropriate. Comparison of continuous variables was performed using the Student t test. Ordinal variables were compared using the Mann-Whitney U test. Time to recurrence and event-free survival curves were calculated using the Kaplan-Meier estimation method. A p value <0.05 was considered statistically significant. Calculations were performed using the statistical software SPSS version 16.0 for Macintosh (SPSS, Chicago, Illinois). Our sample size calculation assumed 65% arrhythmia-free survival in patients assigned to the conventional group and 35% arrhythmia-free survival in patients assigned to the pace-guided group with a sigma (SD) of 0.15. To detect this difference with a power of 0.8 and an alpha level of 0.05, we calculated that 48 patients per arm should be included in the study.
Patients and procedure
The baseline data of the 102 patients included to the study are given in Table 1. Fifty patients (49%) were randomly assigned to conventional PVI (group 1) and 52 (51%) to pace-guided PVI (group 2). Procedure duration was significantly longer in the pace-guided group (185 ± 59 min vs. 139 ± 57 min; p < 0.001); however, fluoroscopy times were not significantly different (25 ± 9 min vs. 23 ± 9 min; p = 0.177). In the conventional group, entrance and exit block in the PV guided by a CMC was reached in all patients. After achievement of unexcitability in the pace-guided group, 49 of 50 (98%) left vein pairs and 48 of 50 (96%) right vein pairs were isolated. The remaining gaps were ablated with CMC guidance. In the pace-guided group, 20 (40%) left pairs of veins showed entrance block after anatomic completion of the ablation line as opposed to 17 (34%) on the right side. Overall, after PV potentials disappeared on the CMC, further ablation to achieve the endpoint of unexcitability in the pace-guided group added 13.7 ± 12.7 kJ of RF energy to the procedure (Table 1). Additional RF application was required in 93 of 104 pairs of veins after PVI and anatomic completion of the line. Areas requiring significantly more RF application consisted of the superior portion of the right pulmonary veins (p = 0.03) and the ridge between LAA and left superior vein (p = 0.02) compared with other areas along the line. Analysis of the signal amplitude of regions that were tested for capture at a pacing output of 10 mV at 2-ms pulse width demonstrated a significantly greater amplitude at capture versus noncapture sites (0.25 ± 0.32 mV vs. 0.17 ± 0.25 mV; p = 0.008); however, voltages overlapped substantially for capture versus noncapture sites (Fig. 2).
Modification of the ablation line was necessary in 5 patients to avoid esophageal temperature rises.
All patients completed at least 12 months of follow-up. After a mean follow-up of 18 ± 6 months, 26 patients (52%) in the conventional versus 43 (82.7%) in the pace-guided group were free from any AF/AT episode lasting longer than 30 s after a single procedure free from class I or III antiarrhythmic medications (p = 0.001). None of the patients had progression to persistent AF during the follow-up. There was 1 stroke in the CMC-guided group that resulted in a midterm comprehension and reading deficit. No other further procedure-related complications occurred. See Figure 3 for comparison of AF/AT-free survival in the 2 groups.
In the present randomized study, we were able to confirm our findings from a pilot feasibility trial that achieving and assessing unexcitability along the ablation line is feasible and may serve as an additional endpoint for PVI procedures. We were furthermore able to show that additional ablation does not translate into higher risk for procedure-related complications. Most remarkably, achievement of unexcitability along the ablation line significantly improved freedom from AF/AT by almost 30% during a mean follow-up of 18 months.
Electrical isolation of the PV has been performed for more than a decade now (1). After initial PVI, recurrences are unfortunately common.
Many studies have suggested that the main reason for AF recurrence in patients with PAF is electrical reconnection of the PVs after initially successful isolation (4,6). By extension, permanent electrical isolation of the PVs at the initial procedure would be expected to decrease arrhythmia recurrence. Different strategies have been used to improve outcomes for patients with PAF, including the application of a longer intraprocedural waiting period after electrical PVI, administration of adenosine to reveal “dormant” conduction, and the use of technological advances such as “single-shot” devices, duty cycling multielectrode ablation catheters, balloon catheters, and robotic navigation systems to improve catheter contact (3,11–13). None of these techniques or technologies have been shown to improve outcomes compared with a “standard” point-by-point catheter ablation guided by electrical PVI.
Recently Eitel et al. (7) reported that pacing during ablation around PVs is safe and feasible to achieve good midterm arrhythmia-free survival. Independently, we were also able to show that rendering the ablation line unexcitable to pacing is feasible and may serve as an endpoint in addition to bidirectional conduction block into and out of PVs (8). However, the study did not address whether additional application of RF energy necessary to achieve unexcitability would translate to an improved clinical outcome.
Comparison with our pilot study
As in our prior study, we found that a smaller electrogram amplitude generally correlated with an unexcitable line; however, electrogram amplitude alone did not reliably distinguish between excitable and unexcitable sites, and a clinically relevant cutpoint was not identifiable in our data. In part, this is related to the presence of significant far-field signals, commonly seen at the LAA/left superior PV ridge and roof of the right superior PV.
Similar to the prior study, we found that once the anatomic lesion set was completed, more than 50% of PV pairs required additional ablation for electrical isolation. Of those that required additional ablation, 97% of PV pairs could be isolated with pace guidance; however, a small number required additional input from a CMC. Also consistent with the prior study, we found that a significant amount of ablation energy, beyond that required for achieving disappearance of PV electrograms, was required to render the PV ablation line unexcitable and that the regions requiring more ablation were often in areas where poor contact has been commonly documented, such as the anterior portion of the left PVs and the LA roof (14).
Poor arrhythmia-free survival in the conventional group
In the present study, we observed an 82% arrhythmia-free survival in patients with the additional endpoint of ablation line unexcitability compared with a 52% arrhythmia-free survival in patients with electrical PVI alone. In comparison, patients with drug-resistant PAF in the Thermocool study showed a 63% arrhythmia-free survival at 9 months in the ablation arm without routine ambulatory ECG monitoring (15). Also included in the Thermocool study were 12% of patients with a second procedure and 7.5% who continued to take antiarrhythmic medication. The relatively disappointing single-procedure, arrhythmia-free survival seen in the conventional group is similar to that seen in the Thermocool study and other recent trials (16,17). In accordance with the current guidelines, the patients in this study underwent 3- to 7-day ECG monitoring every 3 months, such that more recurrences may have been detected as compared with trials with less intensive monitoring (18).
Clinical implications: endpoint of unexcitable line improved arrhythmia-free survival
The striking difference in recurrence rates between the pace-guided and conventional groups was substantial and suggests that PVI is often achieved after suboptimal lesions that are likely to allow recurrent conduction. Additional evidence for inadequate lesion delivery from the present study is the finding that <40% of PV pairs were isolated after an anatomically complete lesion set. In a recent study in swine, ablation lesions were much more likely to be uniform and transmural when rendered unexcitable (9), suggesting that this endpoint provides an indirect assessment of lesion quality. The better single-procedure success in the pace-guided group, therefore, suggests that a more durable lesion set is created by rendering the ablation line unexcitable. Other attempts to prevent conduction gaps include a longer waiting period and administration of adenosine, which has been shown to facilitate “dormant” venoatrial conduction. Yamane et al. (3) were able to show that an additional waiting period of 60 min along with RF application in case of dormant conduction reduced recurrence rates of PAF by 10%. It is reasonable to speculate that a longer waiting period and/or demonstration of persistent PVI with adenosine injection might have similar benefit. Pacing to assess block has some advantages. It can be easily performed with all present systems. It does not rely on additional waiting time. It may help reduce application of ineffective lesions at sites with poor catheter contact because capture likely indicates reasonable catheter contact.
In the study of Yamane et al., the researchers were able to show that the predominant sites of reconnection were around the superior veins, which is consistent with the data of the present study in which the ridge of the LAA/left superior PV and the top of the right superior PV were the sites that required more RF to achieve unexcitability to pacing. Further investigations are warranted to determine the relative contribution of adenosine response, longer waiting periods, and an unexcitable ablation line to a more durable ablation line.
We did not test whether a longer waiting time at the completion of the procedure would have identified sites where local edema or atrial myocardial stunning may have contributed to the endpoint of unexcitability.
The use of pacing to ensure unexcitability along the PVI line markedly improved long-term, single-procedure success, compared with demonstration of bidirectional block alone, with the same low rate of complications using a standard approach. The findings of the present study underscore the perception that PV entrance block alone may not be sufficient to achieve satisfactory clinical results for catheter ablation of PAF. The use of this simple endpoint in addition to PVI has the potential to significantly improve patient outcomes.
Part of the study was financially supported by a limited investigator-initiated study grant from St. Jude Medical, Inc. Dr. Stevenson is co-holder of a patent for needle ablation that has been consigned to Brigham and Women's Hospital. Dr. Michaud has received honoraria and speaker fees from Boston Scientific, Medtronic, and Mediasphere Medical. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- atrial fibrillation
- atrial tachycardia
- circumferential mapping catheter
- left atrial/atrium
- left atrial appendage
- paroxysmal atrial fibrillation
- pulmonary vein isolation
- Received December 30, 2012.
- Revision received March 9, 2013.
- Accepted March 26, 2013.
- American College of Cardiology Foundation
- Arentz T.,
- Weber R.,
- Bürkle G.,
- et al.
- Yamane T.,
- Matsuo S.,
- Date T.,
- et al.
- Ouyang F.,
- Tilz R.,
- Chun J.,
- et al.
- Ouyang F.,
- Antz M.,
- Ernst S.,
- et al.
- Kosmidou I.,
- Houde-Walter H.,
- Foley L.,
- Michaud G.
- Calkins H.,
- Kuck K.H.,
- Cappato R.,
- et al.
- Chun K.R.,
- Schmidt B.,
- Metzner A.,
- et al.
- Scharf C.,
- Boersma L.,
- Davies W.,
- et al.
- Wissner E.,
- Petru J.,
- Metzner A.,
- et al.
- Morillo C, Verma A, Kuck K. Radiofrequency Ablation vs Antiarrhythmic Drugs as First-Line Treatment of Symptomatic Atrial Fibrillation (RAAFT 2): a randomized trial. Paper presented at: Heart Rhythm Society 2012 Scientific Sessions; May 9–12, 2012; Boston, MA. Abstract LB02-1.
- Charitos E.I.,
- Stierle U.,
- Ziegler P.D.,
- et al.