Article Text
Abstract
Objective To examine the long-term clinical outcomes of patients undergoing catheter ablation (CA) for either paroxysmal (PAF) or persistent atrial fibrillation (PsAF) using an extensive pulmonary vein isolation (EPVI)-based strategy.
Methods and results 574 patients (61±9 years; 449 men) with drug-refractory PAF or PsAF underwent CA. Ipsilateral pulmonary veins (PVs) were isolated with extensive antral ablation. A cavotricuspid isthmus line with bidirectional conduction block was created and all non-PV triggers of AF ablated at the index procedure. Left atrial linear ablation was performed in patients with PsAF if AF remained inducible. Patients with recurrent atrial arrhythmias had previous lesions assessed and consolidated. With mean follow up of 27±14 months after the final procedure, 480 patients (83.6%) were in sinus rhythm without antiarrhythmic drugs (1.4±0.6 procedures). The single procedure success rate without antiarrhythmic drugs was 65.5%. A late recurrence (>6 months after the initial procedure) and very late recurrence (>12 months) were seen in 1.6% and 1.7% of patients, respectively. All patients with a late recurrence had PV reconnection, with one patient also demonstrating a non-PV trigger. In comparison, non-PV triggers of AF accounted for 85.7% of very late recurrences with no evidence of PV reconduction.
Conclusions EPVI-based CA is effective in maintaining sinus rhythm in patients with PAF and PsAF over the long term. Recurrent AF after 1 year is mainly due to non-PV triggers, suggestive of an ongoing pathological process.
- Atrial fibrillation
- catheter ablation
- extensive pulmonary vein isolation
- double lasso technique
- long-term follow-up
- atrial arrhythmias
Statistics from Altmetric.com
- Atrial fibrillation
- catheter ablation
- extensive pulmonary vein isolation
- double lasso technique
- long-term follow-up
- atrial arrhythmias
Introduction
The pulmonary veins (PVs) are recognised as the source of the majority of triggers that both initiate and perpetuate paroxysmal atrial fibrillation.1 This key observation led to the development of segmental ostial isolation of each individual PV.2 This technique subsequently evolved to include the whole PV antrum to avoid PV stenosis and to target non-PV triggers.3–5 The addition of left atrial (LA) linear lesions has been shown to improve the outcomes of AF ablation,6–8 and Nademanee et al reported that an ablation strategy targeting areas displaying complex fractionated atrial electrograms (CFAEs) was effective for the treatment of AF.9 However, it is only a decade since the original observation of PV triggered ectopy inducing AF was described and coupled with the evolving understanding and techniques has meant that there are relatively few long-term studies of the effectiveness of AF ablation.
The purpose of this study was to evaluate the long-term efficacy and feasibility of extensive PV isolation (EPVI)-based catheter ablation for paroxysmal and persistent AF. Further, we subsequently investigated the nature of late recurrences following the index procedure.
Methods
Study population
This study consisted of consecutive 574 patients with drug-refractory AF (61±9 years; 449 men; 452 patients with paroxysmal AF (PAF) and 122 with persistent AF (PsAF)) who were observed for at least 12 months following the AF ablation. AF was classified as paroxysmal if episodes terminated spontaneously in <7 days and as persistent if it failed to self-terminate within 7 days. Patients with longstanding PsAF (lasting >12 months) were excluded from this study. All patients gave written informed consent.
Mapping and ablation protocol
All antiarrhythmic drugs (AADs) were discontinued for more than 7 days (amiodarone was discontinued for more than 1 month) before ablation. Patients were effectively anticoagulated for >1 month and transoesophageal echocardiography was performed to exclude atrial thrombi. The surface electrocardiogram (ECG) and bipolar intracardiac electrograms were continuously monitored and stored on a computer-based digital recording system (LabSystem PRO, Bard Electrophysiology, Lowell, Massachusetts, USA). Bipolar electrograms were filtered from 30 to 500 Hz. A 7Fr 20-pole or 14-pol,e two-site mapping catheter (Irvine Biomedical, Irvin, California, USA) was inserted through the right jugular vein and positioned in the coronary sinus (CS) for pacing, recording and internal cardioversion. The electrophysiological study was performed under sedation with propofol or dexmedetomidine.
Extensive pulmonary vein isolation and cavotricuspid isthmus ablation
After trans-septal puncture, a long sheath (SR0, AF Division, St Jude Medical, Minneapolis, Minnesota, USA) and Mullins trans-septal sheath (Medtronic, Minneapolis, Minnesota, USA) were introduced into the left superior pulmonary vein and left inferior pulmonary vein, respectively. Left pulmonary venography and contrast oesophagography were simultaneously performed to obtain the anatomical relationship of the PV ostia relative to the oesophagus. Subsequently, right pulmonary venography was undertaken (figure 1). A 5000 IU intravenous bolus of heparin was administered following trans-septal puncture, with a continuous infusion of 1000 IU/h to maintain an activated clotting time of 200–300 s. Two circular mapping catheters (Lasso, Biosense Webster, Diamond Bar, California, USA) were placed in the superior and inferior PVs, and the left-sided and right-sided ipsilateral PVs were circumferentially and extensively ablated under fluoroscopic and electrophysiological guidance (figure 1).10 The LA posterior wall, at a distance of 1–3 cm from the left-sided or right-sided ostia of the PVs, was anatomically ablated and the distal edges of the anterior aspect of the PVs with early PV potentials or continuous PV and LA potentials were targeted for ablation. Isolation of the left-sided PVs was performed during distal CS pacing and isolation of the right-sided PVs during sinus rhythm (SR). Radiofrequency (RF) current deliveries were applied with an 8 mm tip ablation catheter (Japan Lifeline, Tokyo, Japan) in temperature control mode, with a target temperature of 55°C and maximum power of 35 W on the LA posterior wall and 40 W at the anterior aspect of the PVs. A deflectable 4 mm tip ablation catheter, connected to a thermocouple thermometer (Delta Ohm, Caselle di Selvazzano, Padova, Italy), was used for an oesophageal temperature probe.10 RF energy delivery was applied for 40 s per point on the LA posterior wall; however, if the oesophageal temperature reached 42°C, the application was stopped and restarted when the temperature had reduced to <38°C. The end point was elimination of all PV potentials (figure 2). After completing the EPVI, a 40 mg bolus of adenosine triphosphate (ATP) was injected to unmask dormant PV conduction, with additional RF as required.11
Following EPVI, a cavotricuspid isthmus (CTI) line was performed with an end point of bidirectional conduction block.12 If frequent atrial premature ectopics were present, focal ablation was performed. When ectopic foci within the superior vena cava (SVC) initiated AF, the SVC was isolated.
For patients with PsAF only, SR was restored by internal cardioversion before ablation. If cardioversion failed to restore SR, the ablation was performed during AF, and after completion of EPVI the patient was cardioverted.
Additional ablation following EPVI
AF inducibility was evaluated by burst atrial pacing from the high right atrium at a cycle length of 250 ms, reducing in 10 ms intervals until atrial refractoriness following EPVI and CTI ablation. In patients with PsAF, if AF was sustained for more than 5 min, a LA roof line7 between the isolated left-sided and right-sided PVs was added. If sustained AF was induced after this stage, an inferior line was performed to isolate the LA posterior wall.8 If sustained AF was induced following this, a mitral isthmus line6 between the LIPV and mitral annulus was performed, with an end point of bidirectional conduction block. If sustained AF remained inducible, ablation targeting CFAEs were obtained within the LA.9 Additional ablation was also performed in the same order for the patients with PAF only at the third and fourth procedure.
Repeat procedure
In patients with a recurrence of AF, previous ablation lesions were evaluated. In the presence of conduction recovery, re-isolation of the PVs was performed. If AF occurred spontaneously or was inducible by burst pacing following PV isolation, ablation continued as before. If a stable atrial tachycardia (AT) was present, its mechanism was deduced using activation and entrainment mapping and ablated. In addition, dormant PV conduction was assessed with ATP and eliminated.13
Follow-up
After 3–6 months, in the absence of any AF, anticoagulation was discontinued. No AADs were prescribed in the patients with PAF, but were restarted for 3 months for patients with PsAF after the procedure. If patients had symptomatic premature atrial contractions, AADs were prescribed. Patients had continuous ECG monitoring as an inpatient for 5 days after the procedure. All patients were seen at 2, 6, 10, 14, 24, 36 and 48 weeks after discharge and every 3 months thereafter with Holter monitoring. Patients with palpitations were encouraged to use an event recorder. Recurrent AF was defined according to the patient's symptoms and/or if seen on an ECG. A CT scan examined the morphological changes in the PVs 3 months after the procedure.
Recurrence of AF was defined as AF lasting for >30 s. A blanking period of 1 month was used and a repeat ablation procedure was recommended for patients experiencing AF/AT after this time point.
Statistical analysis
Continuous variables are expressed as mean±SD. Continuous and categorical variables were compared with a Student t test and the χ2 test, respectively. A probability value of p<0.05 indicated statistical significance. A Kaplan–Meier analysis was used to determine the percentage of patients free from AF after the initial and last procedure and differences in the AF-free survival were evaluated using the log-rank test.
Results
Patient characteristics
The mean duration of AF was 55±58 months, mean LA diameter 38.9±5.3 mm and mean ejection fraction 65.8±8.1% determined by echocardiography. Twenty-four patients had cardiomyopathy (10 patients dilated cardiomyopathy, 14 patients hypertrophic cardiomyopathy), 47 coronary artery disease, 33 mitral valve disease and 156 hypertension.
Clinical outcome
Successful EPVI with bidirectional conduction block of the CTI was achieved in all patients. Two patients received EPVI of one side, because they were young and the arrhythmogenic PV was limited to one side. The mean fluoroscopy time for EPVI was 45±27 min.
During 30.1±13.8 months of follow-up after the initial ablation procedure, 399 (69.5%) patients were free of atrial tachyarrhythmia (376 patients (65.5%) without AADs: 67.3% of those with PAF and 59.0% of those with PsAF, p=0.066). One hundred and fifty-seven patients had recurrent AF (PAF in 125 patients and PsAF in 32) and 32 recurrences of AT/atrial flutter (AFL) (paroxysmal AT/AFL in 14 patients and persistent AT/AFL in 18). A repeat ablation procedure was performed in 160 patients and the remaining 15 patients rejected a repeat procedure. The clinical course following each ablation and additional ablation sites are shown in figures 3 and 4 and table 1.
Among the 32 patients who had recurrent AT/AFL, two patients rejected a repeat procedure and no mappable AT/AFL could be induced in four patients. Of 31 elucidated ATs/AFLs in 26 patients, there were 15 macrore-entrant (48%), eight (26%) extra PV foci and eight (26%) PV foci origin ATs/AFLs. A roof-dependent tachycardia was seen in four patients and one (25%) had received a previous roof line ablation in the initial procedure. Perimitral flutter was seen in seven patients and two patients (29%) had received perimitral ablation at the index procedure. A dual-loop tachycardia was seen in one patient without any previous linear ablation. CTI-dependent flutter was seen in two patients with a previous CTI ablation. In total, five ATs were related to a gap in a previous line. The location of the extra PV foci was the ostium of the LA appendage (one), left/right septum (three), high right atrium (one), CS ostium (two) and crista terminalis (one).
At the end of follow-up (27.3±13.8 months from the final ablation procedure), 535 (93.2%) patients (480 patients (83.6%) without AADs: 85.2% in PAF versus 77.9% in PsAF, p=0.052) were arrhythmia free after 1.4±0.6 ablations (figure 5). Among the 39 patients with a recurrence of AF/AFL, two patients underwent an atrioventricular node ablation and one underwent a mitral valve replacement with a Maze operation for mitral valve stenosis.
Late recurrences
Nineteen (3.3%) patients (58±9 years; 13 men; PAF in 10 patients) had a recurrence of AF more than 6 months after the initial procedure. The mean time to recurrent tachyarrhythmia was 14±10 months. Twelve (63%) of 19 patients received a second procedure and the remaining patients rejected a second procedure. Of those 12 patients, five (42%) had AF recurrence due to reconnection of the PVs (one patient had another non-PV focus from the crista terminalis) and became AF free after the repeat procedure without AADs. One patient had a recurrence of AFL and became AF/AFL free after a mitral isthmus line. Three patients had a recurrence of AF owing to firing from the SVC and became AF free after SVC isolation. Three patients had multiple extra PV foci and required AADs to control AF after a repeat procedure.
Among the 12 patients undergoing a repeat procedure, five had a recurrence within 1 year (late recurrence) and seven after 1 year (very late recurrence). The characteristics of patients with either a late or a very late recurrence are shown in table 2.
Among the seven patients who rejected a repeat procedure, two did not have any further AF after the initial recurrence. Four patients maintained SR with AADs and the remaining patient had a further recurrence.
Procedural complications
Among a total of 763 procedures, major complications occurred in 17 (2.2%). Cardiac tamponade was seen in six patients (0.8%), all of whom were treated uneventfully with percutaneous pericardiocentesis. Two patients (0.3%) experienced acute pyloric spasm with gastric hypomotility.14 One patient recovered fully but the other patient required a laparoscopic oesophagojejunal anastomosis, partial gastrectomy and gastroduodenal anastomosis. Two patients (0.3%) experienced transient right coronary air emboli, which resolved spontaneously without complication. One patient (0.1%) experienced a stroke, but full neurological recovery was made within 2 days. One patient (0.1%) developed a pseudoaneurysm at the femoral puncture site that required surgical correction and the patient subsequently experienced a pulmonary embolism, which was treated conventionally. One patient (0.1%) experienced a deep vein thrombosis, which was treated conventionally. One patient (0.1%) experienced retroperitoneal bleeding after the procedure, which was successfully treated with a blood transfusion and catheter intervention. Three patients (0.4%) developed phrenic nerve injury during SVC isolation or right superior PV isolation, which recovered within a week in two patients and by 6 months in the other. No patient had a PV stenosis on CT.
Discussion
EPVI-based catheter ablation
Ouyang et al using three-dimensional mapping and a double-Lasso technique demonstrated complete isolation of the PVs by ablating within the left atrium around both ipsilateral veins.4 When this technique was used a 95% success rate without any AADs was achieved for both PAF and PsAF, with a mean follow-up of 6 and 8 months, respectively. Arentz et al using a non-fluoroscopic navigation system demonstrated that isolation encompassing a large area around both ipsilateral PVs, with confirmed PV isolation was more effective than isolation of individual PVs.5 The success rate without AADs after 15±4 months was 67% and 49%, for patients with PAF and PsAF, respectively. These results suggested that the atrial myocardium surrounding the PVs was involved in the pathophysiology of AF. Arrhythmogenic ostial foci, parasympathetic innervation and rotors have been observed in the PV antrum in different models and have been suggested as possible mechanisms for the initiation and perpetuation of AF.
In this study we performed EPVI and CTI ablation in all patients. If patients remained inducible, focal ablation, including SVC isolation if necessary, was performed. In patients with PsAF, LA linear ablations were performed if AF remained inducible. Previous reports have already demonstrated that non-inducibility of AF after ablation for patients with AF is associated with better long-term clinical suppression of the arrhythmia.15 In contrast to previous studies this study (1) included a relatively large number of patients; (2) the follow-up period was longer; (3) more aggressive LA ablation including linear ablation and focal ablation according to AF inducibility was performed and (4) repeat procedures were routinely performed for recurrences.
This study demonstrates that EPVI-based ablation using fluoroscopic guidance is associated with a good long-term clinical outcome in patients with PAF and PsAF. The difference in the success rate between patients with PAF and PsAF did not reach significance, which may be explained by the difference in strategy and the exclusion of patients with longstanding AF. EPVI may eliminate the AF triggers, including the vein of Marshall, anchor points for rotors or mother waves that drive the AF and modify the substrate for AF perpetuation, reducing the area available for circulating wavelets.
In this study, 32 patients (5.6%) had a recurrence of AT/AFL. About a quarter of mappable AT/AFL was associated with PV reconnection and also extra PV foci. The remaining patients with AT/AFL were macrore-entry and a previous linear ablation had been performed in five (33%) of them. We could not eliminate the possibility that previous linear ablation was associated with recurrent macrore-entrant tachycardia in this subset.
Late recurrence after AF ablation
Previous studies have reported that most recurrences of AF occur within 6 months of the index ablation and that recovery of LA-PV conduction is the major cause of recurrences after pulmonary vein isolation.16 17 A very late recurrence has been defined as a recurrence occurring more than 1 year after the index procedure and has previously been associated with non PV-foci and new PV triggers; however, the follow-up period was relatively short in those reports.
In this study, a late recurrence was almost always associated with PV reconduction, whereas this was never the case with a very late recurrence which was due to a non-PV focus.
Safety
A worldwide survey showed that major complications occurred in 6% of the patients with AF undergoing catheter ablation.18 In this study, major complications occurred in 2.2%. One patient experienced severe gastric hypomotility, probably due to damage to the perioesophageal vagal plexus.14 This occurred before the use of oesophageal temperature monitoring. Following the complications of pulmonary emboli and deep vein thrombosis, graduated compression stockings and intermittent pneumatic compression boots were used and no further episodes occurred. As air embolisms occurred during catheter exchange only when patients were snoring, a milder sedation protocol was used.
Study limitations
AF recurrences may be underestimated as patients were not constantly monitored. However, this study only included highly symptomatic patients, who distinguished their frequent AF episodes. Second, in this series some patients received CFAE ablation. Nademanee et al reported that the end point of CFAE ablation was the complete elimination of all CFAEs in both the atria and CS. However, in this study the CFAE ablation was performed in a limited population of patients only. Third, the fluoroscopy time was relatively long compared with studies using a three-dimensional mapping system. However, fluoroscopy time was similar to previous study.5 Fourth, we cannot rule out the possibility that PV reconnection has an important role in very late recurrence because our sample size of late recurrence is small. Finally, this study is composed of limited group of patients, because patients with longstanding AF were excluded.
Conclusions
This study demonstrates the feasibility, efficacy and safety of EPVI-based catheter ablation over the long-term follow-up period in patients with PAF and PsAF. Careful mapping for extra PV foci and a complete pulmonary vein isolation, including the PV antrum, might reduce the need for repeated procedures and improve clinical outcome.
Acknowledgments
We thank Matthew Wright for help in preparation of the manuscript.
References
Footnotes
See Editorial, p 610
Competing interests None.
Ethics approval This study was conducted with the approval of the ethics committee of Yokosuka Kyosai Hospital.
Provenance and peer review Not commissioned; externally peer reviewed.