Abstract
Cryoballoon ablation (CBA) is a well-used technique when performing pulmonary vein (PV) isolation in patients with paroxysmal atrial fibrillation (AF). Our aim is to describe incidence, characteristics, and clinical predictors for developing atrial tachycardias (ATs) after cryoballoon PV isolation in patients with paroxysmal AF.
The study population consisted of 181 consecutive patients undergoing a first CBA. All patients received an event-recorder before cryoablation and transmitted daily electrocardiogram (ECG) during 1 month before ablation and 3 months after. Further follow-up consisted of 24 h Holter monitoring and ECG registration every 3 months and also in patients presenting with symptoms. A mean follow-up period was 497.9 ± 283.9 days, and 175 patients completed follow-up. In 14 (8%) patients regular ATs were registered. In multivariate logistic regression model, the following parameters were independently associated with ATs after ablation: an additional right PV, treatment with beta-blockers, and presence of AT on event-recording strips before ablation. Seven (4%) patients with registered AT underwent a redo procedure. In two (1.1%) patients ATs were originated in reconnected PVs. In other patients no left AT was induced. No macro re-entrant left AT was documented in any patient. During follow-up, after a redo ablation, no AT was registered.
The incidence of left AT after CBA is low, and no left atrial macro re-entrant tachycardia was found. The following independent predictors of AT development have been identified: an additional right PV, regular AT registered before ablation, and the use of beta-blockers.
The incidence of regular atrial tachycardias after cryoballoon ablation is low.
Independent predictors of atrial tachycardias after cryoballoon ablation are described.
Introduction
Since the first demonstration of atrial fibrillation (AF) triggers predominantly in the pulmonary veins (PVs),1 radiofrequency (RF) PV isolation procedure has become an increasingly frequent procedure in many electrophysiological laboratories. However, the levels of PV isolation, the completeness of the circumferential lines, when performed by point-by-point ablation, are operator-dependent. Moreover, non-transmural lesion around PVs is the major risk factor for the development of new atrial tachycardias (ATs) after the ablation procedure.2
Other techniques aiming for PV isolation using a ‘single application’ have been developed. Cryoballoon ablation (CBA) is one of the technologies for PV isolation, which has been widely adopted for the treatment of patients with paroxysmal AF. Acute and mid-term success rates of CBA have been well published.3 However, there is still lack of systematic data regarding ATs after CBA.
The aim of this study was to describe incidence, characteristics, and clinical predictors of ATs after cryoballoon PV isolation in patients with paroxysmal AF.
Methods
The study population consisted of 181 consecutive patients undergoing a first cryothermal balloon PV isolation in Erasmus MC between 2006 and August 2008. All patients had paroxysmal AF, refractory to at least one antiarrhythmic drug (AAD). The study population consisted of patients with left ventricular ejection fraction >55%, and a left atrial (LA) diameter <55 mm. A total of 36 (19.9%) patients had already gone cavotricuspid isthmus (CTI) ablation in the past. None of the patients had structural heart disease and none of them had previously been ablated in the LA. All patients signed informed consent forms. The study complies with the Declaration of Helsinki.
Initial cryoballoon ablation procedure
A PV isolation was performed using a double-lumen cryoballoon (Arctic front, Cryocath, Medtronic). Both femoral veins and the left subclavian vein were used for venous access. A 10 F, intracardiac echocardiography (ICE) catheter (Flexview, EPmed) was introduced through the left femoral vein and positioned in the right atrium. A decapolar catheter was placed in the coronary sinus. After the first 10 cases, a double transseptal puncture was replaced by a single transseptal approach using a Brockenbrough needle and an 8-F non-steerable transseptal sheath, guided by both ICE and fluoroscopy. Intracardiac echocardiography was also used to ensure a posterior transseptal approach. The sheath was exchanged for a 14-F steerable sheath (FlexCath, Medtronic). A 23 or 28 mm 12-F balloon catheter was positioned over an exchange wire to occlude the ostium of each PV. Balloon occlusion was assessed by a manual injection of contrast through the central lumen of the catheter. Conformity of the PV antra was considered if there was full retention of contrast medium without visible outflow or only minor outflow leakage. Cryoenergy was given for 300 s per application, and with at least two applications for each vein. The applications per vein were directed towards the major side branches. Before targeting the right PVs, a quadripolar catheter was positioned in the superior caval vein for continuous phrenic nerve stimulation during cryoapplication. At loss of capture, the ablation was instantaneously terminated. After targeting all PVs, the cryocatheter was exchanged for a circular mapping catheter (Lasso 2515, Biosense Webster) to confirm electrical PV entrance block. If persistence of PV conduction was still present, the cryoballoon was introduced again, trying to maximize the wall contact at the location of the remaining potentials (as guided by the circular catheter, ICE, and fluoroscopy). If after the second ablation attempt a PV was not isolated, a conventional 9-F 8 mm tip cryocatheter (Freezor Max, Medtronic) was used to perform a segmental isolation through the same transseptal puncture. Each cryoapplication using an 8 mm conventional cryocatheter was performed for up to 300 s. At 1 day post-procedure, a transthoracic echocardiogram was made to exclude pericardial effusion and a chest X-ray to exclude pneumothorax and other thoracic complications. All patients were anticoagulated as described previously.4
Until February 2007, cryoballoon diameter size was selected according to the availability of 23 and 28 mm balloons. From February 2007 the 28 mm size was preferred, as analysis of available data revealed a higher prevalence of right phrenic nerve paralysis using the smaller balloon.4
Electrocardiogram screening for tachyarrhythmias
All patients received an event-recorder before cryoablation. The patients were instructed to transmit at least one transtelephonic ECG strip of 30 s duration daily at a fixed hour. When symptoms were experienced, additional strips could be sent. The registration was started 1 month before ablation and continued for 3 months afterwards. The compliance of patients with this follow-up method was monitored, and when no data have been sent, they were reminded to do so.
Post-procedural patient visits were scheduled every 3 months, and included a physical examination, 12-lead rest ECG registration, and a 24 h Holter monitoring.
The received ECGs and strips from the event-recording system were independently reviewed by two physicians and coded as sinus rhythm, regular AT, or AF. An AT was considered if a regular atrial activity with stable P-wave morphology was found throughout the whole ECG recording, a rest ECG, or event-recorder strip. Typical right atrial flutter was considered on a 12-lead ECG, if typical saw-tooth pattern in inferior leads was seen, F-waves were positive in lead V1, and isoelectric or negative in leads V5–V6.
After ablation, AADs were continued for 3 months in the majority of patients. They were terminated when no arrhythmias occurred.
We did not include a blanking period following catheter ablation, since our aim was to assess the total incidence of ATs.
Redo ablation procedures
Redo ablation was proposed to patients with symptomatic sustained tachyarrhythmia recurrence, refractory to 1–2 IC class drugs. Patients underwent a redo procedure if drug-refractory symptomatic recurrence of sustained AT or AF was recorded. Redo procedures were performed in electrophysiological laboratories equipped with either CARTO system (Biosense Webster) or NavX system (St Jude Medical).
If a tachycardia was present, entrainment mapping from a diagnostic coronary sinus catheter was performed first. If a right atrial flutter was suggested, entrainment-mapping of the right atrium was performed using a steerable mapping catheter.
If a LA tachycardia was suggested, PV mapping was performed using a circular 20-pole catheter. If electrical conduction into PVs was present, cryoballoon PV re-isolation was performed using the same method as in the first procedure.
Programmed (basic cycle 600 ms; 1, 2, and 3 extrastimuli) and burst pacing from the LA was used for tachyarrhythmia induction. The follow-up scheme and recurrence detection methods were the same as after the index procedure.
Statistical analysis
Continuous variables were expressed as mean ± standard deviation and were compared using the t-test, if their distribution did not deviate significantly from the normal distribution (tested with the Kolmogorov–Smirnov test). If significant deviation from the normal distribution was found, continuous variables were expressed as median (interquartile range) and were compared using non-parametric tests (the Mann–Whitney U tests). Categorical variables were expressed as percentages and numbers and were compared using Fisher's exact test.
Associations between clinical and procedural parameters and AT after ablation were tested using logistic regression. All parameters showing significant association with AT after ablation in univariate regression model were selected for the multivariate model.
Correlations for normally distributed variables were tested using the linear Pearson correlation test; for non-normally distributed variables the Spearman test was used.
P < 0.05 (two-sided) was considered to indicate statistical significance. All analyses were performed using the Statistica Software Package v. 6.0 (StatSoft Inc., Tulsa, OK, USA).
Results
Patient characteristics
Cryoballoon PV isolation was performed in 181 patients. Follow-up analysis was completed in 175 patients: in 2 patients the ablation procedure was not completed; 4 patients did not send ECG strips. The patients were followed up for a mean period of 497.9 ± 283.9 days. The ECG strips via the event-recording system were obtained during 1 month before ablation and 3 months after CBA. A total number of 10 187 strips were assessed. Sinus rhythm was present on 9276 (91%) rhythm strips, 818 (8%) strips showed AF in 87 patients. Regular ATs were present on 91 (0.9%) strips in 14 (8.0%) patients; in 5 of them we obtained 12-lead ECGs during ATs. In two of these patients 12-lead ECG showed typical CTI-dependent flutter.
The total number of patients without any registered recurrence was 86 (49.1%). Atrial fibrillation as a recurrence was present in 83 (47.4%); AT (including typical right atrial flutter) was registered in 14 (8.0%). In 3 (1.7%) of the 12 patients ECG strips were obtained with both AT and AF. A mean AT cycle length was 238 ± 27 (190–280) ms. Electrocardiogram strips with ATs presented in Figure 1; 12-lead ECGs with ATs are shown in Figure 2.
ECG strips obtained by the event-recording system. A number of each strip corresponds to the number of a patient.
Twelve-lead ECGs with ATs (ECGs with apparently typical atrial flutter are not shown). The number of each strip corresponds to the number of a patient.
In 10 patients (71.4% out of the AT group), AT appeared within the first 3 months of follow-up. In five patients (35.7% out of the AT patients), no further arrhythmia recurrence was recorded during subsequent follow-up, and no symptoms were reported after antiarrhythmic treatment had been restarted or changed. Following the first 3 months ATs were revealed or registered in five (2.9%) patients.
Comparison between patients with and without atrial tachycardia recurrence
Clinical characteristics of 14 patients with ATs and procedural data are shown in Tables 1 and 2. There was no significant difference in time of recurrence between patients with AT (36.7 ± 36.2 days), and patients with AF recurrence (64.9 ± 99.1 days, P = 0.454).
Patient clinical characteristics in group with AT after ablation, and in group without AT
| AT group, N = 14 | No AT group, N = 161 | P | |
|---|---|---|---|
| Age | 55.0 ± 9.1 | 55.2 ± 9.7 | 0.749 |
| Gender, males | 9 (64.3%) | 123 (76.4%) | 0.337 |
| LA diameter | 42.7 ± 4.5 | 42.1 ± 3.6 | 0.942 |
| Hypertension | 4 (28.5%) | 39 (24.22%) | 0.749 |
| Thyroid disease | 1 (8.33%) | 11 (6.83%) | 1.000 |
| COPD | 0 | 7 (4.35%) | 1.000 |
| Diabetes | 1 (8.33%) | 4 (2.48%) | 0.344 |
| Beta-blockers at the moment of recurrence | 9 (64.29%) | 50 (31.06%) | 0.017 |
| AAD at the moment of recurrence (except beta-blockers) | 10 (71.43%) | 118 (73.29%) | 1.000 |
| Amiodarone at the moment of recurrence | 0 | 36 (33.36%) | 0.077 |
| Presence of AT on event strips before ablation | 3 (21.43%) | 8 (4.97%) | 0.046 |
| AT burden before ablation | 2.90 ± 5.92% | 0.50 ± 2.61% | 0.004 |
| LSPV size, mm | 22.02 ± 4.30 | 21.40 ± 3.75 | 0.604 |
| LIPV size, mm | 17.89 ± 1.68 | 18.35 ± 3.1 | 0.681 |
| RSPV size, mm | 19.64 ± 2.47 | 19.58 ± 3.27 | 0.953 |
| RIPV size, mm | 17.63 ± 2.78 | 19.14 ± 3.20 | 0.132 |
| Left PV common ostium | 4 (28.6%) | 22 (13.7%) | 0.230 |
| Right PV common ostium | 0 | 4 (2.48%) | 1.000 |
| Additional right PVa | 2 (14.28%) | 3 (1.86%) | 0.052 |
| AT group, N = 14 | No AT group, N = 161 | P | |
|---|---|---|---|
| Age | 55.0 ± 9.1 | 55.2 ± 9.7 | 0.749 |
| Gender, males | 9 (64.3%) | 123 (76.4%) | 0.337 |
| LA diameter | 42.7 ± 4.5 | 42.1 ± 3.6 | 0.942 |
| Hypertension | 4 (28.5%) | 39 (24.22%) | 0.749 |
| Thyroid disease | 1 (8.33%) | 11 (6.83%) | 1.000 |
| COPD | 0 | 7 (4.35%) | 1.000 |
| Diabetes | 1 (8.33%) | 4 (2.48%) | 0.344 |
| Beta-blockers at the moment of recurrence | 9 (64.29%) | 50 (31.06%) | 0.017 |
| AAD at the moment of recurrence (except beta-blockers) | 10 (71.43%) | 118 (73.29%) | 1.000 |
| Amiodarone at the moment of recurrence | 0 | 36 (33.36%) | 0.077 |
| Presence of AT on event strips before ablation | 3 (21.43%) | 8 (4.97%) | 0.046 |
| AT burden before ablation | 2.90 ± 5.92% | 0.50 ± 2.61% | 0.004 |
| LSPV size, mm | 22.02 ± 4.30 | 21.40 ± 3.75 | 0.604 |
| LIPV size, mm | 17.89 ± 1.68 | 18.35 ± 3.1 | 0.681 |
| RSPV size, mm | 19.64 ± 2.47 | 19.58 ± 3.27 | 0.953 |
| RIPV size, mm | 17.63 ± 2.78 | 19.14 ± 3.20 | 0.132 |
| Left PV common ostium | 4 (28.6%) | 22 (13.7%) | 0.230 |
| Right PV common ostium | 0 | 4 (2.48%) | 1.000 |
| Additional right PVa | 2 (14.28%) | 3 (1.86%) | 0.052 |
LA, left atrium; AT, atrial tachycardia; COPD, chronic obstructive pulmonary disease; AAD, antiarrhythmic drug; PV, pulmonary vein; LIPV, left inferior pulmonary vein; LSPV, left superior pulmonary vein; RIPV, right inferior pulmonary vein; RSPV, right superior pulmonary vein.
aAdditional right PV defined during contrast angiography as a PV with a separate ostium.
Patient clinical characteristics in group with AT after ablation, and in group without AT
| AT group, N = 14 | No AT group, N = 161 | P | |
|---|---|---|---|
| Age | 55.0 ± 9.1 | 55.2 ± 9.7 | 0.749 |
| Gender, males | 9 (64.3%) | 123 (76.4%) | 0.337 |
| LA diameter | 42.7 ± 4.5 | 42.1 ± 3.6 | 0.942 |
| Hypertension | 4 (28.5%) | 39 (24.22%) | 0.749 |
| Thyroid disease | 1 (8.33%) | 11 (6.83%) | 1.000 |
| COPD | 0 | 7 (4.35%) | 1.000 |
| Diabetes | 1 (8.33%) | 4 (2.48%) | 0.344 |
| Beta-blockers at the moment of recurrence | 9 (64.29%) | 50 (31.06%) | 0.017 |
| AAD at the moment of recurrence (except beta-blockers) | 10 (71.43%) | 118 (73.29%) | 1.000 |
| Amiodarone at the moment of recurrence | 0 | 36 (33.36%) | 0.077 |
| Presence of AT on event strips before ablation | 3 (21.43%) | 8 (4.97%) | 0.046 |
| AT burden before ablation | 2.90 ± 5.92% | 0.50 ± 2.61% | 0.004 |
| LSPV size, mm | 22.02 ± 4.30 | 21.40 ± 3.75 | 0.604 |
| LIPV size, mm | 17.89 ± 1.68 | 18.35 ± 3.1 | 0.681 |
| RSPV size, mm | 19.64 ± 2.47 | 19.58 ± 3.27 | 0.953 |
| RIPV size, mm | 17.63 ± 2.78 | 19.14 ± 3.20 | 0.132 |
| Left PV common ostium | 4 (28.6%) | 22 (13.7%) | 0.230 |
| Right PV common ostium | 0 | 4 (2.48%) | 1.000 |
| Additional right PVa | 2 (14.28%) | 3 (1.86%) | 0.052 |
| AT group, N = 14 | No AT group, N = 161 | P | |
|---|---|---|---|
| Age | 55.0 ± 9.1 | 55.2 ± 9.7 | 0.749 |
| Gender, males | 9 (64.3%) | 123 (76.4%) | 0.337 |
| LA diameter | 42.7 ± 4.5 | 42.1 ± 3.6 | 0.942 |
| Hypertension | 4 (28.5%) | 39 (24.22%) | 0.749 |
| Thyroid disease | 1 (8.33%) | 11 (6.83%) | 1.000 |
| COPD | 0 | 7 (4.35%) | 1.000 |
| Diabetes | 1 (8.33%) | 4 (2.48%) | 0.344 |
| Beta-blockers at the moment of recurrence | 9 (64.29%) | 50 (31.06%) | 0.017 |
| AAD at the moment of recurrence (except beta-blockers) | 10 (71.43%) | 118 (73.29%) | 1.000 |
| Amiodarone at the moment of recurrence | 0 | 36 (33.36%) | 0.077 |
| Presence of AT on event strips before ablation | 3 (21.43%) | 8 (4.97%) | 0.046 |
| AT burden before ablation | 2.90 ± 5.92% | 0.50 ± 2.61% | 0.004 |
| LSPV size, mm | 22.02 ± 4.30 | 21.40 ± 3.75 | 0.604 |
| LIPV size, mm | 17.89 ± 1.68 | 18.35 ± 3.1 | 0.681 |
| RSPV size, mm | 19.64 ± 2.47 | 19.58 ± 3.27 | 0.953 |
| RIPV size, mm | 17.63 ± 2.78 | 19.14 ± 3.20 | 0.132 |
| Left PV common ostium | 4 (28.6%) | 22 (13.7%) | 0.230 |
| Right PV common ostium | 0 | 4 (2.48%) | 1.000 |
| Additional right PVa | 2 (14.28%) | 3 (1.86%) | 0.052 |
LA, left atrium; AT, atrial tachycardia; COPD, chronic obstructive pulmonary disease; AAD, antiarrhythmic drug; PV, pulmonary vein; LIPV, left inferior pulmonary vein; LSPV, left superior pulmonary vein; RIPV, right inferior pulmonary vein; RSPV, right superior pulmonary vein.
aAdditional right PV defined during contrast angiography as a PV with a separate ostium.
Procedural data in both groups of patients
| AT group, N = 14 | No AT group, N = 161 | P | |
|---|---|---|---|
| 23 mm balloon | 3 (21.4%) | 37 (23.0%) | 1.000 |
| 28 mm balloon | 11 (78.6%) | 124 (77.0%) | 1.000 |
| Number of applications in LSPV | 3 (IQR: 2–3.5) | 3 (IQR: 2–4) | 0.488 |
| Number of applications in LIPV | 2 (IQR: 2–3) | 2 (IQR: 2–2.75) | 0.640 |
| Number of applications in RSPV | 2 (IQR: 2–2) | 2 (IQR: 1.75–2.25) | 0.889 |
| Number of applications in RIPV | 2 (IQR: 1–2) | 2 (IQR: 1–2) | 0.296 |
| LSPV isolation by balloon | 10 (71.4%) | 135 (83.9%) | 0.265 |
| LIPV isolation by balloon | 7 (50%) | 132 (82.0%) | 0.010 |
| RSPV isolation by balloon | 14 (100%) | 141 (87.6%) | 0.374 |
| RIPV isolation by balloon | 11 (78.6%) | 141 (87.6%) | 0.401 |
| Touch-up applications required, patients | 7 (50%) | 48 (29.8%) | 0.138 |
| Number of PVs with touch-ups | 1 (IQR: 0–1) | 0 (IQR: 0–1) | 0.411 |
| Total number of balloon applications | 9.5 (IQR: 7–11) | 9 (IQR: 7–12) | 0.777 |
| Total cryoapplication time, s | 2494 ± 525 | 3268 ± 806 | 0.061 |
| Total procedure time, min | 232.0 ± 69.5 | 202.2 ± 75.5 | 0.171 |
| Fluoro time, min | 58.9 ± 32.6 | 48.9 ± 26.5 | 0.233 |
| Ablation of CTI | 7 (50%) | 62 (38.5%) | 0.410 |
| AT group, N = 14 | No AT group, N = 161 | P | |
|---|---|---|---|
| 23 mm balloon | 3 (21.4%) | 37 (23.0%) | 1.000 |
| 28 mm balloon | 11 (78.6%) | 124 (77.0%) | 1.000 |
| Number of applications in LSPV | 3 (IQR: 2–3.5) | 3 (IQR: 2–4) | 0.488 |
| Number of applications in LIPV | 2 (IQR: 2–3) | 2 (IQR: 2–2.75) | 0.640 |
| Number of applications in RSPV | 2 (IQR: 2–2) | 2 (IQR: 1.75–2.25) | 0.889 |
| Number of applications in RIPV | 2 (IQR: 1–2) | 2 (IQR: 1–2) | 0.296 |
| LSPV isolation by balloon | 10 (71.4%) | 135 (83.9%) | 0.265 |
| LIPV isolation by balloon | 7 (50%) | 132 (82.0%) | 0.010 |
| RSPV isolation by balloon | 14 (100%) | 141 (87.6%) | 0.374 |
| RIPV isolation by balloon | 11 (78.6%) | 141 (87.6%) | 0.401 |
| Touch-up applications required, patients | 7 (50%) | 48 (29.8%) | 0.138 |
| Number of PVs with touch-ups | 1 (IQR: 0–1) | 0 (IQR: 0–1) | 0.411 |
| Total number of balloon applications | 9.5 (IQR: 7–11) | 9 (IQR: 7–12) | 0.777 |
| Total cryoapplication time, s | 2494 ± 525 | 3268 ± 806 | 0.061 |
| Total procedure time, min | 232.0 ± 69.5 | 202.2 ± 75.5 | 0.171 |
| Fluoro time, min | 58.9 ± 32.6 | 48.9 ± 26.5 | 0.233 |
| Ablation of CTI | 7 (50%) | 62 (38.5%) | 0.410 |
CTI, cavotricuspid isthmus; PV, pulmonary vein; LIPV, left inferior pulmonary vein; LSPV, left superior pulmonary vein; RIPV, right inferior pulmonary vein; RSPV, right superior pulmonary vein; IQR, interquartile range.
Procedural data in both groups of patients
| AT group, N = 14 | No AT group, N = 161 | P | |
|---|---|---|---|
| 23 mm balloon | 3 (21.4%) | 37 (23.0%) | 1.000 |
| 28 mm balloon | 11 (78.6%) | 124 (77.0%) | 1.000 |
| Number of applications in LSPV | 3 (IQR: 2–3.5) | 3 (IQR: 2–4) | 0.488 |
| Number of applications in LIPV | 2 (IQR: 2–3) | 2 (IQR: 2–2.75) | 0.640 |
| Number of applications in RSPV | 2 (IQR: 2–2) | 2 (IQR: 1.75–2.25) | 0.889 |
| Number of applications in RIPV | 2 (IQR: 1–2) | 2 (IQR: 1–2) | 0.296 |
| LSPV isolation by balloon | 10 (71.4%) | 135 (83.9%) | 0.265 |
| LIPV isolation by balloon | 7 (50%) | 132 (82.0%) | 0.010 |
| RSPV isolation by balloon | 14 (100%) | 141 (87.6%) | 0.374 |
| RIPV isolation by balloon | 11 (78.6%) | 141 (87.6%) | 0.401 |
| Touch-up applications required, patients | 7 (50%) | 48 (29.8%) | 0.138 |
| Number of PVs with touch-ups | 1 (IQR: 0–1) | 0 (IQR: 0–1) | 0.411 |
| Total number of balloon applications | 9.5 (IQR: 7–11) | 9 (IQR: 7–12) | 0.777 |
| Total cryoapplication time, s | 2494 ± 525 | 3268 ± 806 | 0.061 |
| Total procedure time, min | 232.0 ± 69.5 | 202.2 ± 75.5 | 0.171 |
| Fluoro time, min | 58.9 ± 32.6 | 48.9 ± 26.5 | 0.233 |
| Ablation of CTI | 7 (50%) | 62 (38.5%) | 0.410 |
| AT group, N = 14 | No AT group, N = 161 | P | |
|---|---|---|---|
| 23 mm balloon | 3 (21.4%) | 37 (23.0%) | 1.000 |
| 28 mm balloon | 11 (78.6%) | 124 (77.0%) | 1.000 |
| Number of applications in LSPV | 3 (IQR: 2–3.5) | 3 (IQR: 2–4) | 0.488 |
| Number of applications in LIPV | 2 (IQR: 2–3) | 2 (IQR: 2–2.75) | 0.640 |
| Number of applications in RSPV | 2 (IQR: 2–2) | 2 (IQR: 1.75–2.25) | 0.889 |
| Number of applications in RIPV | 2 (IQR: 1–2) | 2 (IQR: 1–2) | 0.296 |
| LSPV isolation by balloon | 10 (71.4%) | 135 (83.9%) | 0.265 |
| LIPV isolation by balloon | 7 (50%) | 132 (82.0%) | 0.010 |
| RSPV isolation by balloon | 14 (100%) | 141 (87.6%) | 0.374 |
| RIPV isolation by balloon | 11 (78.6%) | 141 (87.6%) | 0.401 |
| Touch-up applications required, patients | 7 (50%) | 48 (29.8%) | 0.138 |
| Number of PVs with touch-ups | 1 (IQR: 0–1) | 0 (IQR: 0–1) | 0.411 |
| Total number of balloon applications | 9.5 (IQR: 7–11) | 9 (IQR: 7–12) | 0.777 |
| Total cryoapplication time, s | 2494 ± 525 | 3268 ± 806 | 0.061 |
| Total procedure time, min | 232.0 ± 69.5 | 202.2 ± 75.5 | 0.171 |
| Fluoro time, min | 58.9 ± 32.6 | 48.9 ± 26.5 | 0.233 |
| Ablation of CTI | 7 (50%) | 62 (38.5%) | 0.410 |
CTI, cavotricuspid isthmus; PV, pulmonary vein; LIPV, left inferior pulmonary vein; LSPV, left superior pulmonary vein; RIPV, right inferior pulmonary vein; RSPV, right superior pulmonary vein; IQR, interquartile range.
Patients with AT after CBA were more frequently treated with beta-blockers; however, there was no difference in ADD class IC and III treatment. In the AT group patients regular ATs before ablation were registered more frequently on event-recordings. In this group an additional right PV was more frequently encountered, however, with borderline statistical significance (P = 0.052). Touch-up applications were more frequently required to complete left inferior PV isolation in the AT group (Table 2).
We did not find any statistically significant correlation between AT cycle length and LA diameter, age, PV sizes, AAD treatment, or other factors.
Redo ablation procedures in patients with atrial tachycardias
Among 14 patients with registered regular AT, a redo procedure due to symptomatic drug-refractory recurrence was performed in 7 (50.0%) persons. As mentioned previously, five patients had no further recurrence. Two patients refused a redo procedure because of significant improvement to quality of life.
Spontaneous or induced regular tachycardia was present in six patients during the redo procedure. Right atrial activation and entrainment mapping revealed right-atrial CTI-dependent flutter in four patients, and RF catheter ablation of CTI was successfully performed with termination of the tachyarrhythmia. Left AT was diagnosed in two (1.1%) patients. In both cases AT was initiated by a firing source from a focal origin: in one case from the area adjacent to the left superior PV; in the second case the tachycardia originated from the right superior PV. Cryoballoon PV re-isolation using the cryoballoon technique successfully terminated both tachycardias.
Atrial tachycardia, recorded during ambulatory observation, was non-inducible in one patient. Pulmonary vein re-isolation using CBA was performed.
No other tachyarrhythmias were inducible in all seven patients at the end of a redo procedure (after PV re-isolation and CTI ablation, where required). After the second CBA procedure no further AT was registered in any patient from this group.
Predictors of atrial tachycardias
In a multivariate regression model we identified the following parameters associated with any AT after ablation: (i) therapy with beta-blockers—odds ratio (OR) 5.08 (1.26–20.50), confidence interval (CI) 95%, P = 0.022; (ii) AT registered before ablation—OR 5.98 (1.21–29.61), CI 95%, P = 0.029; (iii) an additional right PV—OR 10.47 (1.33–82.30), CI 95%, P = 0.026.
In a model with exclusion of typical atrial flutter, where only 10 patients remained for analysis, the multivariate regression model identified two parameters independently associated with ATs: (i) therapy with beta-blockers—OR 6.27 (1.21–32.38), CI 95%, P = 0.029; (ii) an additional right PV—OR 11.65 (1.44–94.23), CI 95%, P = 0.021.
Discussion
The major finding of this study is that symptomatic, drug-refractory AT is uncommon after CBA and rarely requires a redo procedure. Only seven (4.0%) patients with confirmed AT during daily event monitoring required a redo ablation. After the blanking period ATs were found in five (2.9%) patients. In all of these cases ATs were related to PV conduction recovery or CTI-dependent flutter. No macro re-entrant left ATs developed after CBA.
Factors, independently associated with AT recurrence after initial CBA, were the following: an additional right PV, registered AT before ablation, and the use of beta-blockers.
Atrial tachycardias after radiofrequency pulmonary vein isolation
Since the initial description of segmental, circumferential, and antral PV ablation, RF catheter treatment has been widely accepted as a potentially curative option for AF.1 New regular AT occurring after RF catheter ablation for AF has been well documented. However, there are limited data on ATs occurring after RF circumferential PV isolation and no additional linear lesions in the LA.
As it has been shown by Gerstenfeld et al.,5 after ostial PV isolation in 341 patients with paroxysmal and persistent AF, a new persistent left AT developed in 10 (2.9%) patients, and in 1 patient AT was macro re-entrant. The remaining patients had focal or unknown ATs.
Ouyang et al.6 have found that among 100 patients with paroxysmal and persistent AF after circumferential PV isolation ATs were registered in 21% of cases. In 15% of patients left ATs were verified and ablated, and in two cases left ATs were macro re-entrant.
In a recently published study incorporating wide circumferential PV isolation in 839 patients with both paroxysmal and persistent AF, the authors found that ATs developed only in 4% of patients, and ATs had mainly a macro re-entrant mechanism.7
Other studies, where different ablation strategies were applied, including circumferential PV isolation, showed the occurrence of stable ATs between 10 and 43%.8–10 In several of these studies additional linear lesions were created, and no circular mapping catheter was used in others.
Difference between radiofrequency- and cryoablation
Balloon ablation techniques have been developed to simplify PV isolation procedure, aiming for PV isolation with a single ‘shot’; however, the complexity of the PV anatomy occasionally forces operators to use different sizes of balloons, and additional touch-up applications.
The use of cryothermal energy for PV isolation has several potential advantages over RF energy: presumed better catheter stability due to freeze-mediated catheter adhesion, the creation of well-demarcated homogeneous lesions that are less arrhythmogenic than the ragged indistinct lesions associated with RF ablation.11–13 Thus, the cryolesions, in contrast to RF lesion, are more homogeneous and show fibrotic tissue without any sign of chronic inflammation and sharp borders of demarcation between normal and ablated tissue.14–16
In a study by Reddy et al.,17 it has been shown that the electrical isolation following 23 mm CBA occurs at the level of the PV ostia. However, a previous study by our group18 demonstrated that the PV antra are also isolated after ablation with the 28 mm balloon. Moreover, our findings are supported by other studies, showing that 28 mm balloon isolation area differs significantly from those created by the smaller balloon.19 Thus, the low incidence of ATs after CBA in our study might be related to the more distal PV isolation present only in patients with 23-balloon ablation. We suggest that homogeneous and circumscribed lesions after CBA are more important factors, explaining a low incidence of AT after cryoballoon PV ablation. No patients were found to have a macro re-entrant left AT.
In comparison with previous studies describing the success rates of RF PV isolation, the overall recurrence rate in our study may seem high. It should be noted, that we have incorporated very strict ECG monitoring post-ablation, which could detect more arrhythmic events than intermittent Holter monitoring. Also, we did not include a blanking period in this study; therefore, all recurrences were analysed.
Previous studies with cryoballoon ablation
A number of studies documenting cryoablation to treat AF have been published recently. Results of ostial PV isolation using transcatheter cryoablation have been published by several groups,20–22 but no data regarding ATs development have been provided.
In a study with CBA to treat paroxysmal AF in 27 patients, recurrence documentation was performed using Holter ECG recording at 1, 3, 6, 9, and 12 months post-ablation.23 In addition, all patients were provided with event-monitors and were asked to transmit daily ECG tracings for 6 months after the ablation. Interestingly, during a mean follow-up period of 271 days (147 ± 356) there was no AT.
In a prospective three-centre study on cryoballoon PV isolation,24 in 346 patients left AT was ablated in two (0.6%) patients during the first 3 months of follow-up. However, the mechanisms of these tachyarrhythmias were not specified.
One study compared 28 mm CBA (25 patients) and RF (25 patients) PV isolation for paroxysmal AF. The authors found that none of the patients after CBA developed AT, and one (4%) patient after RF ablation showed macro re-entrant perimitral flutter.25
In another study the authors reported the outcomes of 124 patients, who underwent CBA for paroxysmal and persistent AF. The authors reported no organized ATs over an 18-month follow-up period.26
To the best of our knowledge, no previous study systematically assessed ATs after CBA.
Predictors of atrial tachycardias after cryoballoon pulmonary vein isolation
Among the many factors, associated with ATs after RF ablation, the following have been documented: persistent AF, additional linear lesions in the LA, incomplete PV isolation.2,27
We have identified that an additional right PV is associated with AT development after CBA. We suggest that anatomical variations of PVs can be associated with less success of ablation, probably due to technical difficulties to complete the lesions. Thus, in recently published studies authors found that unusual PV anatomy (number of PVs, angle of PV direction, early PV branches) is associated with worse AF freedom.28,29
Regular ATs, identified on event-recorder strips before ablation, have been found in association with ATs after ablation. Considering the design of our study, we are not able to describe the mechanisms of ATs registered before ablation. Although some patients with AT after ablation did not undergo a redo procedure, we have not found AT sources different from PVs, except typical right atrial flutter.
Another predictor of AT after ablation identified in our study is beta-blocker therapy. We assume, that beta-blockers could help to prevent AF recurrence in this group of patients, thus making AT the only clinical tachyarrhythmia.
Study limitations
In our study the patients had been extensively screened for recurrences during the first 3 months of follow-up; however, further ECG screening was less rigorous. This may partly explain higher incidence of ATs during the blanking period, than afterwards.
We have given a high number of touch-up applications to complete PV isolation. Although some operators pursue multiple cryoballoon applications until PV isolation is achieved, we prefer focal applications for PVs that prove initially difficult to isolate. The following factors are in line with our strategy: the potential increased risk of extracardiac damage, and prolonged procedural time in the LA. It should be noted that intraprocedural success of PV isolation have not been shown to differ between the two approaches.3
The low number of events during follow-up (ATs) limits the model of logistic regression.
Another study limitation may be the use of different ADDs. As many other groups, we routinely continue antiarrhythmics for at least 3 months in the majority of patients after PV isolation. Antiarrhythmic drugs may influence the development and mechanisms of ATs.
Conclusion
The incidence of AT after CBA is low, and no macro re-entrant left AT was found in our study group. Patients requiring redo procedures for left AT are very uncommon. The following independent predictors of AT development have been identified: an additional right PV, regular AT registered before ablation, and the use of beta-blockers.
Acknowledgements
We thank Richard Alloway for language revision of the article.
Conflict of interest: none declared.
