Author + information
- Received November 16, 2004
- Revision received March 23, 2005
- Accepted March 29, 2005
- Published online July 19, 2005.
- Hildegard Tanner, MD⁎ (, )
- Gerhard Hindricks, MD,
- Richard Kobza, MD,
- Anja Dorszewski, MD,
- Petra Schirdewahn, MD,
- Christopher Piorkowski, MD,
- Jin-Hong Gerds-Li, MD and
- Hans Kottkamp, MD
- ↵⁎Reprint requests and correspondence:
Dr. Hildegard Tanner, University of Leipzig, Heart Center, Cardiology, Dept. of Electrophysiology, Struempellstrasse 39, D-04289 Leipzig, Germany
Objectives The aim of this study was to analyze trigger activity in the long-term follow-up after left atrial (LA) linear ablation.
Background Interventional strategies for curative treatment of atrial fibrillation (AF) are targeted at the triggers and/or the maintaining substrate. After substrate modification using nonisolating linear lesions, the activity of triggers is unknown.
Methods With the LA linear lesion concept, 129 patients were treated using intraoperative ablation with minimal invasive surgical techniques. Contiguous radiofrequency energy-induced lesion lines involving the mitral annulus and the orifices of the pulmonary veins without isolation were placed under direct vision.
Results After a mean follow-up of 3.6 ± 0.4 years, atrial ectopy, atrial runs, and reoccurrence of AF episodes were analyzed by digital 7-day electrocardiograms in 30 patients. Atrial ectopy was present in all patients. Atrial runs were present in 25 of 30 patients (83%), with a median number of 9 runs per patient/week (range 1 to 321) and a median duration of 1.2 s/run (range 0.7 to 25), without a significant difference in atrial ectopy and atrial runs between patients with former paroxysmal (n = 17) or persistent AF (n = 13). Overall, 87% of all patients were completely free from AF without antiarrhythmic drugs.
Conclusions A detailed rhythm analysis late after specific LA linear lesion ablation shows that trigger activity remains relatively frequent but short and does not induce AF episodes in most patients. The long-term success rate of this concept is high in patients with paroxysmal or persistent AF.
For curative treatment of atrial fibrillation (AF), the initiating triggers and the perpetuating substrate are the main targets for interventional ablation strategies (1–6). A specific left atrial (LA) linear lesion concept targeting the perpetuation of AF and therefore modifying the substrate was tested using intraoperative ablation with minimally invasive surgical techniques in the Intraoperative Radiofrequency Ablation of Atrial Fibrillation (IRAAF) study (7,8). In that study, contiguous radiofrequency (RF) energy-induced lesion lines involving the mitral annulus (MA) and the orifices of the pulmonary veins (PV) were placed under direct vision to prevent anatomically defined LA re-entrant circuits. No PV isolation, surgical mass reduction, or right atrial lesions were performed.
After substrate modification using nonisolating linear lesion lines, the activity of initiating triggers is unknown. Therefore, the aim of the present study was to analyze trigger activity in the IRAAF population, i.e., premature atrial complexes (PAC) and, in particular, atrial runs, and reoccurrence of AF episodes in these patients during long-term follow-up more than three years after the ablation procedure.
Of 129 patients treated for paroxysmal and persistent AF with intraoperative RF ablation as a primary indication for operation using minimally invasive surgical techniques through a right lateral mini-thoracotomy, 30 consecutive patients (27 men, 3 women, mean age 55 ± 11 years) reaching a follow-up of at least 3 years were included in the present analysis (Table 1).Seventeen patients (57%) originally had paroxysmal AF and 13 (43%) persistent AF with an AF history of 8 ± 6 years (range 1 to 25 years). All patients were highly symptomatic before ablation despite multiple antiarrhythmic drug regimens including class I and III antiarrhythmic drugs. All patients with paroxysmal AF had two or more AF episodes per week lasting >12 h.
Lone AF was present in 19 patients (64%), 7 (23%) had a history of arterial hypertension, and 4 (13%) had a history of myocarditis and/or pericarditis. No patient had coronary artery disease or valvular heart disease as assessed with coronary angiography or transthoracic or transesophageal echocardiography.
Five patients had a total of six ablation procedures before study entry: two patients had conventional ablation of the cavo-tricuspid isthmus because of typical atrial flutter, two patients had Lasso-guided PV isolation with AF recurrence, and one patient had both procedures. The institutional review committee approved the study, and all patients gave written informed consent to the investigational nature of the procedure.
Surgical approach and RF catheter ablation
A detailed description of the surgical approach and the ablation technique has been published previously (7,8). Briefly, all patients were operated on a video-assisted minimally invasive surgical approach via a right anterolateral mini-thoracotomy. The LA was incised parallel to the interatrial groove anterior to the right PVs, allowing access to the MA as well as exposure of the orifices of the PVs. Radiofrequency alternating current (500 kHz, modified HAT 200S; Osypka GmbH, Grenzach-Wyhlen, Germany) was delivered in a unipolar mode between the 10-mm T-shaped tip electrode of a specially designed ablation probe (Osypka GmbH) and a 10 × 16 cm external backplate electrode. A drawing of the LA with the RF energy-induced LA linear lesions is given in Figure 1.The first lesion line extended from the left inferior aspect of the MA to the left PVs. The second line connected the left lower and upper PV orifices. From there, a third line coupled the left and right upper PVs. Then the right upper and lower PV orifices were connected. Finally, the line at the LA roof was connected to the surgical incision to prevent “incisional re-entry.”
Postoperative management and follow-up
Daily 12-lead electrocardiograms (ECGs) and two 24-h ECGs were performed within the first 10 postoperative days. In case of early recurrences of AF, electrical cardioversion and institution of an antiarrhythmic drug treatment (class Ic or amiodarone) were followed. According to the study protocol, antiarrhythmic drug treatment was withdrawn three months after the operation. All patients were seen in the outpatient clinic at six weeks, three months, six months, one year, two years, and three years after the procedure. A 12-lead ECG, 24-h ECG, and echocardiography were performed in all patients during each follow-up visit.
In two-thirds of the patients with persistent and in one-third of the patients with paroxysmal AF, transient recurrences of AF were observed in the early postoperative period (8). However, these early relapses of AF directly after operation did not predict long-term treatment outcome, and freedom from AF after 12 months measured >90%.
For the present study, in all patients reaching a follow-up of at least three years, a continuous digital seven-day ECG (Lifecard CF; DelmarReynolds Medical, Inc., Irvine, California) was recorded. One patient was excluded from analysis because of the presence of a dual-chamber pacemaker because of sick sinus syndrome with permanent atrial stimulation.
Detailed rhythm analysis was done by the means of the seven-day ECG recordings. As surrogate for trigger activity, PAC and atrial runs (3 beats up to 30 s) were counted. Frequent PACs were defined as >100 PAC per 7-day ECG. The duration and number of ventricular contractions resulting from atrial runs were measured. In addition, a quantitative analysis of all recurrent AF episodes (episode lasting more than 30 s) was performed.
The analysis of the recordings was done by a single electrophysiologist. The counting was done manually in the first 10 patients because an automatic counting was not possible. Thereafter, the data of these patients were exported to analysis software (Pathfinder, DelmarReynolds Medical Inc., Irvine, California). After analysis by this software, all premature complexes and atrial runs were checked again. There was no difference in manually counting and using the software with respect to the number of atrial runs, and the difference with respect to premature atrial contraction was <5% and not significant. Therefore, the remaining 20 patients received software-assisted counting with manual correction.
Patients were equipped with a symptom log and instructed on how to document duration, quality, and severity of symptoms during the ECG recording and medical treatment (i.e., antiarryhthmic drugs, beta-blockade, and oral anticoagulation therapy). The symptoms (palpitation, dyspnea, dizziness, and chest pain as surrogate of AF) were qualitatively scored in the symptom log as none, minimal, mild, moderate, or severe.
Continuous variables were expressed as mean ± SD or as median and range when variables were not normally distributed.
After a mean follow-up of 3.6 ± 0.4 years (range 3.0 to 4.3 years), 162 ± 10 ECG hours per patient (range 131 to 168 h) were recorded and analyzed. Typical examples of ECG recordings of rare and frequent PACs, short and long atrial runs, and AF in five different patients are given in Figure 2.
Overall, 29 of 30 patients (97%) did not receive any antiarrhythmic drugs; 1 patient was treated with amiodarone. Twenty patients (67%) received beta-blockers and 11 patients (37%) had treatment with an angiotensin-converting enzyme inhibitor. Patients were treated with beta-blockers because of arterial hypertension (n = 7), rate control in case of AF (n = 2), and mild symptoms of PACs and atrial runs (n = 11). No other rate-controlling medication was used. Two patients (7%) with recurrent AF received oral anticoagulation therapy; nine patients (30%) had treatment with aspirin.
Quality of recordings
The intention to monitor was seven days for all patients. Therefore, the overall duration of the assessed single-lead ECG recordings was 5,040 h. From the intended 5,040 h of recordings, 4,857 h (96%) were available for analysis. More than 3% of the missing data were because of part-time removal of the device by the patient. Less than 1% of the recordings were excluded from analysis because of artifacts. Patients were equipped with additional surface electrodes in case of insufficient skin contact. In this case, the patients were alerted by the device. Therefore, the overall recording quality was excellent.
PAC and atrial runs
Premature atrial complexes were present in all patients over a wide range (median 240, range 10 to 62,751) (Fig. 3).Premature atrial complexes in patients with former paroxysmal AF (median 188, range 11 to 32,568) and former persistent AF (median 726, range 10 to 62,751) were comparable. More than 100 PACs per 7-day ECG were found in 21 patients (70%). Frequent PACs in patients with former paroxysmal AF (65%) and in those with former persistent AF (77%) were comparable. Atrial runs were found in 83% of all patients (n = 25). Patients with former paroxysmal AF had slightly more atrial runs (n = 15, 88%) than patients with former persistent AF (n = 10, 77%). The median duration of ventricular contractions resulting from atrial runs measured 1.2 s/run (range 0.7 to 25 s) with a median number of 4 beats (range 3 to 56 beats) per atrial run.
Reoccurrence of AF
Overall, 26 of 30 patients (87%) were completely free from AF and 4 patients (13%) had paroxysmal episodes of AF. Two of them were asymptomatic. Three of these patients had former persistent AF, later presenting with paroxysmal AF, and one patient had former paroxysmal AF. No patient presented with persistent AF. Freedom from AF in patients with former paroxysmal AF was 94% (one patient with antiarrhythmic drug treatment) and 77% in patients with former persistent AF (no patients with antiarrhythmic drug treatment) (Fig. 4).Directly after the procedure, freedom from AF measured 53%, after 6 months 97%, after 1 year 100%, after 2 years 100% as assessed with conventional 24-h ECG, and after 3 years using 7-day ECGs 87% (Fig. 5).
Altogether, 29 AF episodes were found in the 4 patients with recurrent paroxysmal AF. Given the overall number of 1,017 atrial runs, 2.8% of these runs resulted in AF (defined as lasting ≥30 s). Twenty-six of 29 AF episodes (90%) lasted <2 h, one AF episode (3%) lasted between 2 and 24 h, and only 2 AF episodes (7%) lasted more than 24 h. In patients with former paroxysmal AF, no episode lasted more than 2 h. No episode with atrial flutter was detected.
The definition of recurrent AF was arbitrarily set to 30 s. We decided to take this widely used definition to allow comparison with other AF studies. If the definition of AF is changed to >15 s, only one additional patient had recurrence of AF because the vast majority of atrial runs were of short duration (median duration 1.2 s). Therefore, results would not differ substantially if this shorter duration had been chosen.
PAC and atrial runs in patients with and without reoccurrence of AF
Comparing patients without (n = 26) and patients with AF recurrences (n = 4, 3 with former persistent AF), PACs were more frequent in patients with AF recurrences: median 13,560 (range 726 to 62,751) versus median 199 (range 10 to 32,568). Accordingly, patients with AF recurrences had more atrial runs: median 131 (range 7 to 321) versus median 7 (range 0 to 83) in patients without AF recurrence.
Patients with only PACs and atrial runs scored their symptoms as none in 68% (n = 17), as minimal in 24% (n = 6), and as mild in 8% (n = 2). No patient had moderate or severe symptoms. One patient did not state his symptoms on the symptom log. Patients with AF recurrence (n = 4) scored their symptoms as none (n = 1), mild (n = 1), moderate (n = 1), and severe (n = 1). Sixty-one percent of all asymptomatic patients were treated with beta-blockers, whereas 73% of all symptomatic patients (minimal to severe symptoms) had beta-blocker treatment.
The present study shows, for the first time, that trigger activity late after specific linear LA ablation remains present in the majority of patients. Premature atrial complexes were found in all patients. Atrial runs (3 beats up to 30 s) were frequent but of short duration (median 1.2 s).
Role of trigger activity after catheter ablation
For curative treatment of AF, the initiating triggers and the perpetuating substrate are the main targets for interventional ablation strategies (1–6). The majority of foci triggering AF originate from myocardial sleeves within the PVs. Additionally, foci were found in the funnel-shaped transition zone and, in some cases, also in areas of the left or right atrium unrelated to the PVs (1,9–11).
Electrophysiologic studies in patients undergoing repeat PV isolation because of AF recurrences showed that in the majority of cases reinitiating triggers originate from previously isolated PVs, which have regained conduction to the LA (12,13). To date, no reports on the presence of ectopic activity and atrial runs after successful PV isolation targeting trigger elimination are present during long-term follow-up. Accordingly, it is unknown whether trigger activity remains present in the long-term follow-up after LA linear ablation targeting at the perpetuation of AF without PV isolation.
The prevention of the perpetuating re-entrant circuits as substrate modification is the alternative interventional treatment strategy to cure AF.
In the IRAAF study, a specific LA linear lesion concept targeting at the prevention of anatomic LA re-entry was tested using intraoperative ablation with minimally invasive surgical techniques (7,8). In that study, contiguous RF energy-induced lesion lines involving the MA and the orifices of the PVs were placed under direct vision to prevent anatomically defined LA re-entrant circuits.
The present study analyzed the occurrence and role of trigger activity late after substrate modification in patients with paroxysmal and persistent AF treated with the IRAAF concept. Continuous digital seven-day ECGs were used for follow-up. With this approach, significantly more AF episodes could be detected compared with conventional 24-h ECG in a recent study using percutaneous ablation (14).
The quantitative analysis of atrial ectopy and in particular atrial runs, defined as runs of atrial tachycardia/nonsustained AF episodes from 3 beats up to 30 s, showed the presence of PAC in all patients and frequent PAC and atrial runs in the majority of patients. The median duration of ventricular contractions resulting from atrial runs was low, measuring 1.2 s/run with a median number of 4 ventricular contractions per atrial run. Whether all PACs and atrial runs represent potential triggers for induction of AF remains unproven, because their origin was not determined invasively and their role in inducing AF before ablation cannot be quantified. However, despite the high prevalence of atrial ectopy, the rate of AF recurrence was low (4 of 30 patients, 13%) both in patients with paroxysmal AF (1 of 17 patients, 6%) and in those with former persistent AF (3 of 13 patients, 23%). This implies that with this strategy aiming at the perpetuation of AF, the substrate cannot sustain AF in the majority of patients.
Atrial ectopy was found in all patients, and atrial runs were found in 83% of patients. However, the high use of beta-blockers in our study cohort may have influenced the amount and occurrence of PACs and atrial runs, which might have been even higher without this treatment. In addition, beta-blockade may have influenced the high percentage of asymptomatic or only mildly symptomatic patients.
Measurement of trigger activity in one-lead surface ECG is a surrogate of trigger activity measured invasively, and the origin within the atria cannot be determined. However, trigger activity is likely underestimated by seven-day ECG because of AV block and hidden P waves in the T waves and ST-segment, and because of non-conducted ectopic activity from the foci to atrial musculature. Moreover, trigger activity has a wide time-dependent intra-individual and interindividual variability, which underlines the usefulness of long-term recordings such as the continuous seven-day ECG recordings that were used in the present study. Analysis with larger sample size would be desirable to confirm our results.
Dr. Tanner was supported by a grant from the Swiss Foundation for Pacemaker and Electrophysiology.
- Abbreviations and Acronyms
- atrial fibrillation
- Intraoperative Radiofrequency Ablation of Atrial Fibrillation
- left atrium/atrial
- mitral annulus
- premature atrial complex
- pulmonary vein
- Received November 16, 2004.
- Revision received March 23, 2005.
- Accepted March 29, 2005.
- American College of Cardiology Foundation
- Ernst S.,
- Schluter M.,
- Ouyang F.,
- et al.
- Pappone C.,
- Oreto G.,
- Rosanio S.,
- et al.
- Oral H.,
- Scharf C.,
- Chugh A.,
- et al.
- Pappone C.,
- Rosanio S.,
- Augello G.,
- et al.
- Kottkamp H.,
- Hindricks G.,
- Autschbach R.,
- et al.
- Chen S.A.,
- Hsieh M.H.,
- Tai C.T.,
- et al.
- Goya M.,
- Ouyang F.,
- Ernst S.,
- et al.
- Haissaguerre M.,
- Shah D.C.,
- Jais P.,
- et al.
- Kottkamp H.,
- Tanner H.,
- Kobza R.,
- et al.