Abstract

Aims

It is not known if successful catheter ablation for atrial fibrillation (AF) improves the patient's long-term cardiovascular outcomes. This study investigated the long-term outcomes and mortality of AF patients at high risk who received antiarrhythmic medication and catheter ablation.

Methods and results

The propensity scores for AF were calculated for each patient and were used to assemble a cohort of 174 AF patients with ablation who were compared with an equal number of AF patients without ablation. Composite cardiovascular end points (major adverse cardiovascular event, MACE), including mortality and vascular events in the medically treated patients representing the control group (group 1), were compared with those in the ablation-treated patients (group 2). The rates of the total mortality (2.95% vs. 0.74% per year; P < 0.01), cardiovascular death (1.77% vs. 0% per year; P = 0.001), and ischaemic stroke/transient ischaemic attack (2.21% vs. 0.59% per year; P = 0.02) were higher in group 1 than group 2, respectively. A multivariate Cox regression analysis of the MACE scores showed that a higher CHA2DS2-VASc score [hazard ratio (HR) = 1.309 per increment of score, 95% confidence interval (CI) = 1.06–1.617; P = 0.01] and the performance of the ablation procedure (HR = 0.225, CI = 0.076–0.671; P = 0.007) were independent predictors of a MACE. In patients who received catheter ablation, recurrence of any atrial arrhythmia was a predictor of vascular events and total mortality (P < 0.05).

Conclusion

In AF patients with CHA2DS2-VASc score ≥1, catheter ablation of AF reduced the risk of the total/cardiovascular mortality and total vascular events. Atrial fibrillation recurrence predicts long-term cardiovascular outcomes, as well as the CHA2DS2-VASc score.

What's new?

• Based on the National Death Registry, this study demonstrated that there is a potential increased survival rate in AF patients with rhythm control achieved by ablation strategy, irrespective of recurrence state. This indiated the potential benefits of decreasing AF burden. Furthermore, only patients with absence of recurrence of any atrial arrhythmia were associated with a lower rate of stroke.

Introduction

Atrial fibrillation (AF) is a common arrhythmia that is an independent risk factor for stroke and associated with increased risk of morbidity and mortality.1–3 In recent years, catheter ablation of AF has proven to be an effective therapy for patients with symptomatic and drug-refractory AF.4,5 However, most of the previous studies on AF ablation had a short-term follow-up period and recruited relatively low-risk patients (unlike the patient population group from the Atrial Fibrillation Follow-up Investigation of Rhythm Management trial).6 Currently, it is not known whether successful catheter ablation of AF improves the long-term outcome (total mortality and cardiovascular events) of the patients. It has been suggested that patients with symptom-free AF may have better cardiovascular outcomes than those treated with antiarrhythmic medications.7 Therefore, we hypothesized that the incidence of cardiovascular events may be lower after a successful AF ablation and without recurrence of any atrial arrhythmias. To confirm this theory, after adjusting for the confounding variables, we investigated the long-term cardiovascular outcomes and mortality of high risk of patients with AF who received antiarrhythmic medications for rhythm control and catheter ablation.

Methods

Study population

Our study population comprised 1262 consecutive patients with symptomatic AF who were referred to the electrophysiology unit of Taipei Veterans General Hospital between January 2003 and December 2009. Low-risk patients with CHA2DS2-VASc score = 0 were excluded in this study to avoid type 2 error which may significantly confound the analysis and results. A total of 953 patients were high-risk patients (CHA2DS2-VASc score ≥1). A non-parsimonious logistic regression model was used to generate propensity scores for the likelihood of each patient receiving an ablation procedure, which was based on 16 demographic and clinical variables, including the age, sex, hypertension, diabetes, previous coronary artery disease, previous admission to clinic due to congestive heart failure, thyroid disease, prior coronary revascularization/stroke/other vascular events, the use of ACE inhibitors, diameter of the left atrium, left ventricular ejection fraction, AF types, and CHADS2 and CHA2DS2-VASc score.

In this study, CHA2DS2-VASc score was calculated for every patient by assigning 1 point for history of hypertension, diabetes mellitus, heart failure, vascular disease (myocardial infarction, complex aortic plaque, or peripheral artery disease), and sex of female, age between 65 and 74 years; 2 points for a history of stroke or transient ischaemic attack (TIA), age ≥75 years.8,9 CHADS2 score was also calculated for each patient by assigning points depending on the presence of the following factors: 1 point for age ≥75 years, hypertension, diabetes mellitus, heart failure, and 2 points for a previous stroke or TIA.10 All the medically controlled patients received at least one antiarrhythmic medication, including amiodarone (45%), propafenone (33%), and flecainide (22%), as well as other drugs.

Our study cohort consisted of patients who had at least one or more risk factor. The exclusion criteria were an age of <18 years or >80 years, sick sinus syndrome, or atrioventricular conduction disturbances with the implantation of a permanent pacemaker during the follow-up. The method of AF treatment (catheter ablation or medically control) was chosen either by the patient or electrophysiologist. All patients who received catheter ablation were symptomatic and refractory to at least one antiarrhythmic medication. The study was approved by the ethics committee of the hospital, and all patients who received catheter ablation provided written informed consent.

Catheter ablation of atrial fibrillation and follow-up

Following the attainment of consent, electrophysiological studies and catheter ablation procedures were conducted in the patients in a fasting non-sedated state, the details of which have been described previously.11–14 In brief, after the left atrial geometry had been completed, continuous circumferential lesions were created using the NavX system with either a conventional 4 mm tip or irrigated-tip catheter to encircle the right and left pulmonary vein (PV) ostia. The aim was to position the radiofrequency lesions at a distance of at least 1–2 cm from the angiographically defined ostia. Successful circumferential PV isolation was demonstrated by the absence of any PV activity or dissociated PV activity. In patients with positive AF inducibility, additional linear ablation was performed with a bidirectional conduction block as the end point.13

If non-paroxysmal AF continued after the PV isolation, linear ablation and/or an additional complex fractionated electrographically guided substrate ablation was performed. Complex fractionated electrographically guided ablation was confined to the continuous complex fractionated sites (>5 s) in the left atrium and proximal coronary sinus.12,15 The end points of the complex fractionated electrographic site ablation were a prolongation of the cycle length, elimination of the complex fractionated electrograms, or elimination of local fractionated potentials (bipolar voltage of <0.05 mV).

After sinus rhythm (SR) had been restored following procedural AF termination or electrical cardioversion, mapping and ablation were applied to only spontaneously initiated focal atrial tachycardias (ATs) or non-PV AF-initiating ectopy. If a non-PV AF-initiating ectopy was identified in the superior vena cava, isolation of the vein was guided by circular catheter recordings from the superior vena cava–atrial junction.

After the catheter ablation, all patients were administered anti-arrhythmic drugs for 8 weeks (amiodarone, 84%; propafenone, 10%; flecainide, 5%; and procainamide, 1%). The strategy for the use of warfarin after ablation has been described previously.16 In brief, if patients were receiving oral anticoagulation therapy before ablation, then the therapy was continued with dose-adjusted warfarin for 3 months to maintain a prothrombin time international normalized ratio (INR) of between 2 and 3. After this period, continuation of warfarin was decided by the physician responsible for the treatment of each patient and was dependent on the individual characteristics of each patient. A follow-up of the patients (2 weeks after catheter ablation, and then every 1–3 months for at least 3 months) was conducted either at our cardiology clinic or with the referring physician. After 3 months, the patients received follow-up every 3 months for at least 1 year and then every 6 months.

During each follow-up, 24 h Holter monitoring and/or the monitoring of cardiac events for duration of 1 week were performed. Atrial arrhythmias or AF recurrence, including atrial tachycardia, atrial flutter, or AF, were defined as episodes that lasted for more than 1 min that were confirmed on electrocardiograms 3 months after ablation (blanking period). The long-term efficacy was assessed from the clinical symptoms, surface 12-lead electrocardiograms at rest, 24 h Holter monitoring, and/or 1-week cardiac event recordings, according to the Heart Rhythm Society consensus document.4

Definitions of clinical end points

The composite cardiovascular end point was the combined occurrence of major adverse cardiovascular events (MACEs), including the total vascular events (ischaemic stroke, TIA, acute coronary artery events, and peripheral vascular events or pulmonary embolism) and death.

Ischaemic stroke was defined as a sudden-onset focal neurological deficit (diagnosed by a neurologist) with a duration of >24 h that was caused by ischaemia. A TIA was defined as a sudden-onset focal neurological deficit (diagnosed by a neurologist) with a duration of <24 h. Acute coronary syndrome was defined as the occurrence of a spontaneous myocardial infarction and unstable angina that required hospitalization or admission to the emergency room. A peripheral embolism was defined as a thromboembolism that did not occur in the brain, heart, eyes, or lungs. A pulmonary embolism diagnosis was established based on computed tomography images of the chest as confirmed by an experienced cardiologist. The causes and dates of death in the patients who had died during the follow-up period after the baseline survey were obtained. This was achieved by linking our database with the National Death Registry through a unique, life-long personal identification number that is provided to every Taiwanese citizen. Subjects not appearing on the National Death Registry database were considered to be alive. The National Death Registry database registers valid information that is based on the certified death certificates, which were coded according to the International Classification of Disease, Ninth Revision (ICD-9). Cardiovascular deaths are represented by the ICD-9 codes 390–459. The accuracy of this coding in the Taiwan's National Death Registry database has been validated.17

Statistical analysis

The data are presented as mean values and standard deviations for normally distributed continuous variables and as proportions for categorical variables. The differences between continuous values were assessed using an unpaired two-tailed t-test for normally distributed continuous variables, a Mann–Whitney test for skewed variables, and a χ2 test for nominal variables. The MACE-free survival curves were plotted using the Kaplan–Meier method, and statistical significances were examined using the log-rank test. A Cox regression analysis was used to identify the factors associated with the composite end point, mortality, cerebrovascular accident (CVA), coronary events, and other vascular events.

In order to minimize the impact of the confounding factors on the clinical characteristics, we employed the propensity analysis and matching technique. We matched pairs one-to-one (ablation vs. control) with identical propensity scores with a 0.01 caliper width. The adequacy of the match was assessed by estimating the standardized differences between the ablation and control patients for all the variables.

Variables with a P value of <0.10 in the univariate models were selected for testing in the multivariate analysis. Two-sided P value of <0.05 was considered statistically significant.

Results

Patient characteristics

In a total of 953 patients with a CHA2DS2-VASc score ≥1, overall 383 patients (40.2%) received catheter ablation. Patients were more likely to receive catheter ablation if they had a history of coronary artery disease and an old stroke, no history of heart failure, a smaller left atrium, and an improved left ventricular function (Table 1). One hundred seventy-four ablation patients (45%) were matched with individuals who received antiarrhythmic medications without catheter ablation (N= 174). The baseline characteristics of the propensity-matched group are shown in Table 2 after an adjustment for 16 confounding factors. The mortality data were documented from the National Registry of Mortality, and all surviving patients in both groups completed the follow-up.

Table 1

Baseline characteristics for non-propensity-matched patients according to the type of rhythm control of atrial fibrillation

Characteristics Patients with medical treatment (N = 570) Patients with catheter ablation (N = 383) P value 
Age, years 55 ± 11 57 ± 11 0.02 
Gender (male, N, %) 282 (49.5) 232 (60.6) 0.001 
Hypertension (N, %) 349 (61.2) 255 (66.6) 0.11 
Diabetes mellitus (N, %) 94 (16.5) 67 (17.5) 0.75 
Previous vascular diseases (N, %) 16 (2.8) 8 (2.1) 0.63 
Coronary artery disease (N, %) 124 (21.8) 125 (32.6) <0.001 
History of heart failure admission (N, %) 111 (19.5) 37(9.7) <0.001 
History of coronary revascularization (N, %) 11 (1.9) 2 (0.5) 0.14 
Thyroid disease (N, %) 59 (10.4) 56 (14.6) 0.06 
Old stroke history (N, %) 18 (3.2) 26 (6.8) 0.01 
ACE inhibitor (N, %) 38 (6.7) 19 (5.0) 0.86 
Left ventricular ejection fraction (%) 55.5 ± 10.0 58.7 ± 8.31 <0.001 
Left atrial diameter (mm) 41.5 ± 10.1 39.5 ± 6.53 <0.01 
Paroxysmal AF patients (N, %) 444 (77.9) 311 (81.2) 0.25 
Mean CHADS2 score 1.09 ± 1.00 1.11 ± 0.88 0.77 
Mean CHA2DS2-VASc score 1.74 ± 1.08 1.77 ± 0.99 0.70 
Characteristics Patients with medical treatment (N = 570) Patients with catheter ablation (N = 383) P value 
Age, years 55 ± 11 57 ± 11 0.02 
Gender (male, N, %) 282 (49.5) 232 (60.6) 0.001 
Hypertension (N, %) 349 (61.2) 255 (66.6) 0.11 
Diabetes mellitus (N, %) 94 (16.5) 67 (17.5) 0.75 
Previous vascular diseases (N, %) 16 (2.8) 8 (2.1) 0.63 
Coronary artery disease (N, %) 124 (21.8) 125 (32.6) <0.001 
History of heart failure admission (N, %) 111 (19.5) 37(9.7) <0.001 
History of coronary revascularization (N, %) 11 (1.9) 2 (0.5) 0.14 
Thyroid disease (N, %) 59 (10.4) 56 (14.6) 0.06 
Old stroke history (N, %) 18 (3.2) 26 (6.8) 0.01 
ACE inhibitor (N, %) 38 (6.7) 19 (5.0) 0.86 
Left ventricular ejection fraction (%) 55.5 ± 10.0 58.7 ± 8.31 <0.001 
Left atrial diameter (mm) 41.5 ± 10.1 39.5 ± 6.53 <0.01 
Paroxysmal AF patients (N, %) 444 (77.9) 311 (81.2) 0.25 
Mean CHADS2 score 1.09 ± 1.00 1.11 ± 0.88 0.77 
Mean CHA2DS2-VASc score 1.74 ± 1.08 1.77 ± 0.99 0.70 
Table 2

Baseline characteristics for propensity-matched patients according to the type of rhythm control of atrial fibrillation

Characteristics Patients with medical treatment (N = 174) Patients with Catheter ablation (N = 174) P value 
Age, years 57 ± 11 57 ± 10 0.82 
Gender (male, N, %) 93 (53.4) 92 (52.9) >0.99 
Hypertension (N, %) 113 (64.9) 122 (70.1) 0.36 
Diabetes mellitus (N, %) 30 (17.2) 31 (17.8) >0.99 
Previous vascular diseases (N, %) 7 (4.0) 5 (2.9) 0.77 
Coronary artery disease (N, %) 55 (31.6) 47 (27.0) 0.41 
History of heart failure admission (N, %) 29 (16.7) 23 (13.2) 0.45 
History of coronary revascularization (N, %) 2 (1.1) 1 (0.6) >0.99 
Thyroid disease (N, %) 16 (9.2) 18 (10.3) 0.86 
Old stroke history (N, %) 8 (4.6) 5 (2.9) 0.57 
ACE inhibitor (N, %) 15 (8.6) 11 (6.3) >0.99 
Long-term use of anti-arrhythmic drugs (N, %) 63 (36.2) 48 (27.6) 0.10 
Long-term use of warfarin (N, %) 18 (10.5) 4 (2.3) <0.01 
Left ventricular ejection fraction (%) 57.0 ± 9.63 57.5 ± 8.89 0.67 
Left atrial diameter (mm) 40.1 ± 9.00 39.4 ± 6.84 0.37 
Paroxysmal AF patients (N, %) 123 (70.7%) 132 (75.9%) 0.33 
Mean CHADS2 score 1.15 ± 1.00 1.10 ± 0.84 0.69 
Mean CHA2DS2-VASc score 1.83 ± 1.13 1.84 ± 0.98 0.92 
Characteristics Patients with medical treatment (N = 174) Patients with Catheter ablation (N = 174) P value 
Age, years 57 ± 11 57 ± 10 0.82 
Gender (male, N, %) 93 (53.4) 92 (52.9) >0.99 
Hypertension (N, %) 113 (64.9) 122 (70.1) 0.36 
Diabetes mellitus (N, %) 30 (17.2) 31 (17.8) >0.99 
Previous vascular diseases (N, %) 7 (4.0) 5 (2.9) 0.77 
Coronary artery disease (N, %) 55 (31.6) 47 (27.0) 0.41 
History of heart failure admission (N, %) 29 (16.7) 23 (13.2) 0.45 
History of coronary revascularization (N, %) 2 (1.1) 1 (0.6) >0.99 
Thyroid disease (N, %) 16 (9.2) 18 (10.3) 0.86 
Old stroke history (N, %) 8 (4.6) 5 (2.9) 0.57 
ACE inhibitor (N, %) 15 (8.6) 11 (6.3) >0.99 
Long-term use of anti-arrhythmic drugs (N, %) 63 (36.2) 48 (27.6) 0.10 
Long-term use of warfarin (N, %) 18 (10.5) 4 (2.3) <0.01 
Left ventricular ejection fraction (%) 57.0 ± 9.63 57.5 ± 8.89 0.67 
Left atrial diameter (mm) 40.1 ± 9.00 39.4 ± 6.84 0.37 
Paroxysmal AF patients (N, %) 123 (70.7%) 132 (75.9%) 0.33 
Mean CHADS2 score 1.15 ± 1.00 1.10 ± 0.84 0.69 
Mean CHA2DS2-VASc score 1.83 ± 1.13 1.84 ± 0.98 0.92 

Catheter ablation results

After a mean follow-up period of 47 ± 23 months, SR maintenance was observed in 39.7% medical control group and 90.2% catheter ablation group (P < 0.001, N = 174 in each propensity-matched group). In patients who received ablation, AF (documented from Holter) lasting for ≥1 min occurred in 56 patients (32.2%) after the index procedure (average ablation number was 1.32 per patient). All patients received systemic screening for PV stenosis (N = 383, group 2, non-propensity matching), which was subsequently observed in three patients (0.8%); one patient (0.3%) received PV angioplasty. No procedure-related thromboembolic event was noted among patients receiving catheter ablation (0%).

Composite cardiovascular end points in the ablation and medicated patients

Composite cardiovascular end points: major adverse cardiovascular events

In Table 3, the MACEs of medically control patients (group 1) are compared with the MACEs of ablation-treated patients without recurrence of atrial arrhythmias (group 2a) and with recurrence (group 2b). The total mortality rates were 11.5, 1.7, and 5.4% (2.94, 0.44, and 1.38% per year; P < 0.01), cardiovascular death rates were 6.9, 0, and 0% (1.77, 0, and 0% per year; P = 0.001), and the total vascular event rates were 12.1, 1.7, and 5.4% (3.1, 0.44, and 1.38% per year; P < 0.01), respectively. A Kaplan–Meier survival analysis indicated that the group 1 patients had a higher cumulative incidence of MACEs compared with the group 2 patients (Figure 1A).

Table 3

Causes of mortality and vascular events in high-risk propensity-matched patients during the follow-up

  Patients with medical treatment (N = 174) Patients with catheter ablation (N = 174) P value Ablation patients
 
P value 
Without recurrence (N = 118) With recurrence (N = 56) 
Mortality   
 Cardiovascular death 12 (6.9%) 0 (0.0%)* 0.001 0 (0.0%)* 0 (0.0%)* 0.001 
 Non-cardiovascular death 8 (4.6%) 5 (2.9%) 0.57 2 (1.7%) 3 (5.4%) 0.33 
 Total mortality 20 (11.5%) 5 (2.9%)* <0.01 2 (1.7%)* 3 (5.4%)* <0.01 
Vascular events   
 TIA/ischaemic stroke 15 (8.6%) 4 (2.3%)* 0.02 1 (0.8%)* 3 (5.4%)* 0.02 
 Coronary events 2 (1.1%) 1 (0.6%) >0.99 1 (0.8%) 0 (0.0%) >0.99 
 Pulmonary embolism 1 (0.6%) 0 (0.0%) >0.99 0 (0.0%) 0 (0.0%) >0.99 
 Peripheral arterial events 2 (1.1%) 0 (0.0%) 0.50 0 (0.0%) 0 (0.0%) 0.66 
 Deep vein thrombosis 1 (0.6%) 0 (0.0%) >0.99 0 (0.0%) 0 (0.0%) >0.99 
 Vascular events except TIA/ischaemic stroke 6 (3.4%) 1 (0.6%) 0.12 1 (0.8%) 0 (0.0%) 0.19 
 Total vascular events 21 (12.1%) 5 (2.9%)* <0.01 2 (1.7%)* 3 (5.4%)* <0.01 
  Patients with medical treatment (N = 174) Patients with catheter ablation (N = 174) P value Ablation patients
 
P value 
Without recurrence (N = 118) With recurrence (N = 56) 
Mortality   
 Cardiovascular death 12 (6.9%) 0 (0.0%)* 0.001 0 (0.0%)* 0 (0.0%)* 0.001 
 Non-cardiovascular death 8 (4.6%) 5 (2.9%) 0.57 2 (1.7%) 3 (5.4%) 0.33 
 Total mortality 20 (11.5%) 5 (2.9%)* <0.01 2 (1.7%)* 3 (5.4%)* <0.01 
Vascular events   
 TIA/ischaemic stroke 15 (8.6%) 4 (2.3%)* 0.02 1 (0.8%)* 3 (5.4%)* 0.02 
 Coronary events 2 (1.1%) 1 (0.6%) >0.99 1 (0.8%) 0 (0.0%) >0.99 
 Pulmonary embolism 1 (0.6%) 0 (0.0%) >0.99 0 (0.0%) 0 (0.0%) >0.99 
 Peripheral arterial events 2 (1.1%) 0 (0.0%) 0.50 0 (0.0%) 0 (0.0%) 0.66 
 Deep vein thrombosis 1 (0.6%) 0 (0.0%) >0.99 0 (0.0%) 0 (0.0%) >0.99 
 Vascular events except TIA/ischaemic stroke 6 (3.4%) 1 (0.6%) 0.12 1 (0.8%) 0 (0.0%) 0.19 
 Total vascular events 21 (12.1%) 5 (2.9%)* <0.01 2 (1.7%)* 3 (5.4%)* <0.01 

*P < 0.05 when compared with the medically treated patients.

Figure 1

Event-free survival from major adverse cardiovascular events (A), total mortality (B), and cardiovascular death (C).

Figure 1

Event-free survival from major adverse cardiovascular events (A), total mortality (B), and cardiovascular death (C).

Figure 2

Event-free survival from vascular events (A) and stroke/transient ischemic attack (B).

Figure 2

Event-free survival from vascular events (A) and stroke/transient ischemic attack (B).

Mortality

In the propensity-matched patients, all-cause mortality occurred in 20 medically controlled patients (cumulative incidence: 11.5%) and five ablation-treated patients (cumulative incidence: 2.9%; P < 0.01). Cardiovascular mortality was noted in 12 patients (all from medically control group), and they were more likely to have non-paroxysmal AF (75% vs. 25%; P < 0.01), a larger diameter of the left atrium (55.8 ± 15.8 mm vs. 39.2 ± 7.01 mm; P < 0.01), and a higher CHA2DS2-VASc score (3.08 ± 1.68 vs. 1.79 ± 1.01; P = 0.02) as compared with the surviving patients. The recorded causes of death were heart failure (4 of 12; 33.3%), stroke (2 of 12; 16.7%), and coronary events (3 of 12; 25%). The Kaplan–Meier survival analysis demonstrated that the rates of total cardiovascular mortality and total mortality were higher in the group 1 patients compared with the group 2 patients (Figure 1A, B). In contrast, AF recurrence after ablation did not affect the cardiovascular mortality rates (Figure 1C).

Vascular events

In the propensity-matched patients, vascular events were observed in 21 of the medically treated patients (cumulative incidence: 12.1%) and in five of the ablation-treated patients (cumulative incidence: 2.9%; P < 0.01). The incidence of an ischaemic stroke/TIA was higher in the medically treated patients compared with the ablation-treated patients (2.21% vs. 0.59% per year; P = 0.02). Other vascular events except strokes/TIAs were statistically similar between the medically and ablation-treated patients (P = 0.12). A Kaplan–Meier survival analysis demonstrated that the ablation-treated patients without an AF recurrence had lower incidences of total vascular events (P = 0.004) and ischaemic strokes/TIAs (P = 0.015) compared with the patients with an AF recurrence or medically treated patients (Figure 2).

Predictors of a major adverse cardiovascular event, mortality, and vascular events

In Table 4, the univariate analysis of the propensity-matched patients showed that the incidence of a MACE during the follow-up correlated with non-paroxysmal AF, lager LA diameter, lower left ventricular ejection fraction, a higher CHA2DS2-VASc score, and the performance of catheter ablation (all P < 0.05). However, the long-term use of anti-arrhythmic drugs and warfarin was not the predictor of MACEs. A multivariate Cox regression analysis showed that the independent predictors of a MACE occurring were a higher CHA2DS2-VASc score [hazard ratio (HR) = 1.309 per increment of score, 95% confidence interval (CI) = 1.06–1.617; P = 0.01] and the performance of an ablation procedure (HR = 0.225, CI = 0.076–0.671; P = 0.007).

Table 4

Cox regression analysis for predictors of major adverse cardiovascular events in propentity-matched patients

Variable Univariate analysis
 
Multivariate analysis
 
Hazard ratio 95% CI P value Hazard ratio 95% CI P value 
AF type (persistent type) 3.471 1.726–6.981 <0.001 1.473 0.610–3.557 0.389 
LAD (>40 mm) 2.963 1.404–6.252 0.004 1.751 0.739–4.149 0.203 
LVEF (≤45%) 4.368 1923–9.924 <0.001 2.269 0.933–5.519 0.071 
CHA2DS2-VASc score (per increment of score) 1.589 1.325–1.906 <0.001 1.309 1.06–1.617 0.012 
Performance of ablation procedure 0.202 0.077–0.525 0.001 0.225 0.076–0.671 0.007 
Long-term use of anti-arrhythmic drugs 1.147 0.509–2.586 0.740 – – – 
Long-term use of warfarin 0.772 0.267–2.235 0.633 – – – 
Variable Univariate analysis
 
Multivariate analysis
 
Hazard ratio 95% CI P value Hazard ratio 95% CI P value 
AF type (persistent type) 3.471 1.726–6.981 <0.001 1.473 0.610–3.557 0.389 
LAD (>40 mm) 2.963 1.404–6.252 0.004 1.751 0.739–4.149 0.203 
LVEF (≤45%) 4.368 1923–9.924 <0.001 2.269 0.933–5.519 0.071 
CHA2DS2-VASc score (per increment of score) 1.589 1.325–1.906 <0.001 1.309 1.06–1.617 0.012 
Performance of ablation procedure 0.202 0.077–0.525 0.001 0.225 0.076–0.671 0.007 
Long-term use of anti-arrhythmic drugs 1.147 0.509–2.586 0.740 – – – 
Long-term use of warfarin 0.772 0.267–2.235 0.633 – – – 

LAD, left atrial diameter; LVEF, left ventricular ejection fraction. Variables with a P value of <0.1 in the univariate analysis were analysed in the multivariate Cox regression analysis.

Subgroup analysis based on CHADS2 and CHA2DS2-VASc score

The total MACE rate for propensity-matched patients based on CHADS2 and CHA2DS2-VASc scores was summarized in Table 5. Ablation patients with higher CHADS2 (≥2) and CHA2DS2-VASc (≥1) score had lower risk of MACE compared with those with medical control. In patients with CHADS2 score = 1, the rate of MACE was similar between two groups.

Table 5

Rate of cardiovascular mortality and total vascular events in propensity-matched patients with catheter ablation and medical control based on different level of CHADS2 and CHA2DS2-VASc score

Scores
 
Ablation
 
Medical control
 
P value 
Event Total N Event Total N  
CHADS2 0 (0%) 37 2 (5.3%) 38 0.12 
 2 (2.2%) 92 9 (9.8%) 92 0.06 
 ≥2 3 (6.7%) 45 16 (36.4%) 44 <0.01 
CHA2DS2-VASc 0 (0%) 81 6 (7.0%) 86 0.03 
 ≥2 5 (5.4%) 93 21 (23.9%) 88 0.001 
Scores
 
Ablation
 
Medical control
 
P value 
Event Total N Event Total N  
CHADS2 0 (0%) 37 2 (5.3%) 38 0.12 
 2 (2.2%) 92 9 (9.8%) 92 0.06 
 ≥2 3 (6.7%) 45 16 (36.4%) 44 <0.01 
CHA2DS2-VASc 0 (0%) 81 6 (7.0%) 86 0.03 
 ≥2 5 (5.4%) 93 21 (23.9%) 88 0.001 

CHADS2 score of zero only provided if CHA2DS2-VASc score ≥1.

Relative risk of events in medically and ablation-treated patients

A subgroup analysis revealed the predictors of a MACE, total mortality, and vascular events in medically treated and ablation-treated patients based on the multivariate Cox regression analysis (Figure 3). In patients with medical control, the significant predictors of these events were a larger diameter of the left atrium (diameter >40 mm), a lower left ventricular ejection fraction (<45%), persistent AF, and a higher CHA2DS2-VASc score (Figure 3A). In ablation-treated patients, a higher CHA2DS2-VASc score had a higher risk of a MACE (HR = 1.48 per increment of score, CI = 1.11–1.98; P = 0.01) or vascular event (HR = 1.44, CI = 1.05–1.98; P = 0.02). Furthermore, AF recurrence also predicted the future occurrence of a MACE (HR = 2.61, CI = 1.15–5.9; P = 0.02) or vascular event (HR = 2.52, CI = 1.05–6.06; P = 0.04) in the ablation-treated patients (Figure 3B).

Figure 3

Total mortality and total vascular events in patients with medical treatment (A) and catheter ablation (B) according to the multivariate Cox regression analysis, using the following categorical variables: the left ventricular ejection fraction (>45% or ≤45%), left atrial diameter (>40 or ≤40 mm), sinus maintenance, atrial fibrillation types (persistent atrial fibrillation or paroxysmal atrial fibrillation), and ordinal variable of the CHA2DS2-VASc score (score 1–9).

Figure 3

Total mortality and total vascular events in patients with medical treatment (A) and catheter ablation (B) according to the multivariate Cox regression analysis, using the following categorical variables: the left ventricular ejection fraction (>45% or ≤45%), left atrial diameter (>40 or ≤40 mm), sinus maintenance, atrial fibrillation types (persistent atrial fibrillation or paroxysmal atrial fibrillation), and ordinal variable of the CHA2DS2-VASc score (score 1–9).

Discussion

Main findings

This study had several significant findings. First, in the patients with at least one risk factor, catheter ablation of AF (irrespective of the recurrence state) reduced the incidence of the total mortality and cardiovascular mortality based on a nation-wide survey. Second, the absence of AF recurrence after catheter ablation appeared to be associated with lower rate of stroke. Third, the independent predictors of a MACE were the performance of an ablation procedure and a higher CHA2DS2-VASc score. These results were demonstrated repeatedly, despite multiple adjustments with a propensity analysis and subgroup analysis.

Long-term outcome in ablation-treated patients

In previous studies on patients treated with traditional antiarrhythmic medications, SR maintenance was not associated with the clinical outcome.6 New-generation class III antiarrhythmic medications have provided enhanced control of AF rhythm, which is associated with a lower risk of strokes and cardiovascular mortality.18 However, it is not yet known if AF burden modifies the risk of a cardiovascular mortality and vascular events. Currently, catheter ablation is the standard therapy for patients with drug-refractory AF, and the rate of symptom-free AF survival of >1 year following this procedure is more than 70%, according to a worldwide survey.4 Pappone's group used a registry to compare the clinical outcomes of 589 ablation-treated patients with 582 medically controlled patients.19 Stroke and heart failure mortality were reduced by 50% in patients who received catheter ablation. Nademanee et al.7 examined the outcomes of 674 high-risk AF patients who received substrate ablation and demonstrated that the maintenance of SR after AF ablation was associated with a relatively low mortality and stroke risk. In that study, SR was maintained in 81% of patients, and mortality was reduced by 12%. However, these previous studies have not adequately assessed the effects of AF recurrence after catheter ablation on the clinical outcomes.4,7 Recently, Hunter et al.20 demonstrated that freedom from AF by catheter ablation was associated with lower rates of stroke and death compared with patients treated medically. However, the event rate among the patients receiving catheter ablation was compared with that of another AF cohort treated medically in the Euro Heart Survey, rather than from the same study population. In contrast, we compared the cardiovascular risks between two different strategies for managing AF patients directly using a propensity-matched method.

The results of this study demonstrated that catheter ablation could reduce the all-cause mortality (11.5% vs. 2.9%; P<0.01). In the patients with medical treatment, heart failure was the cause of cardiovascular death in 33.3% patients, and CVA was in 16.7% patients. Therefore, a reduction in the incidences of heart failure and strokes would reduce the cardiovascular mortality. A lower rate of atrial arrhythmia recurrence in the high-risk ablation-treated patients compared with those patients with medical treatment may explain the better clinic outcome of the former group. A multivariate analysis showed that the higher CHA2DS2-VASc score and ablation procedure were independent predictors of vascular events (Table 4). A subgroup analysis showed that recurrence of any atrial arrhythmia predicted the vascular events and a MACE in patients who received catheter ablation. However, catheter ablation reduces the MACE in patients with higher risk (CHADS2 score ≥2 and CHA2DS2-VASc score ≥1, Table 5).

Effect of atrial arrhythmia recurrence on the outcome

Previously, the Atrial Fibrillation Clopidogrel Trial with Irbesartan for Prevention of Vascular Events (ACTIVE-W) study demonstrated that the incidences of a stroke or systemic embolism were not associated with paroxysmal or persistent types of AF.21 However, the TRENDS study and other device trial demonstrated the quantitative relationship between the burden of a short-duration atrial arrhythmia and the risk of stroke.22–24 A previous study showed that recurrence of AF after catheter ablation usually had a short duration.25 This study demonstrated the risk of individual vascular events according to the ablation status and AF recurrence. First, 75% of the total vascular events in all patients were strokes and/or TIAs. In Figure 2B, the risk of a stroke in ablation-treated patients with symptomatic AF recurrence was similar to the risk in patients with medical treatment (P > 0.05). By contrast, patients who had received a successful ablation had a lower risk of a stroke. In addition to the CHA2DS2-VASc score, this study provides additional clinical evidence that the risk of a vascular event was increased in patients with atrial arrhythmias with duration of >1 min. In addition, recurrence of atrial arrhythmias is an independent risk factor for vascular events.

Survival after catheter ablation

This study demonstrated that catheter ablation of AF could improve the cardiovascular mortality, which was mostly related to the reduced risk of strokes or TIAs. Although this study was not a randomized trial, we used the propensity-matched method to minimize selection bias between patients with and without catheter ablation. Currently, no randomized controlled trials are available to examine this issue, even in the era of catheter ablation. In fact, the rate of stroke and mortality in AF patients is very low in predominantly low-risk patients who have been selected for catheter ablation. This study showed that the total mortality rate in ablation-treated patients was 0.74% per year. Owing to this low rate, more follow-up patient-years are required to show the difference in the incidences of events. This study used the National Death Registry database to demonstrate the difference in mortality rates. The results of this study suggest that there is a potential increased survival rate in patients with rhythm control achieved using an ablation strategy, irrespective of AF recurrence, indicating the potential benefits of decreasing AF burden, because the documented AF duration was at least 1 min in this study. A future randomized study is warranted to confirm this finding.

Limitations

There were several limitations of this study. First, it was a retrospective study and further prospective and randomized trials are required to confirm whether the incidence of cardiovascular events is reduced in the catheter ablation-treated patients compared with medically controlled patients. Second, we admitted the underuse of warfarin in our cohort, which is a common issue in managing AF patients in daily practice.26–28 However, the use of warfarin was not a significant factor associated with adverse events in the Cox regression analysis. Therefore, the issue may not confound the results of this study. Third, this study did not enroll any patients who were receiving newer class III drugs (dronedarone), which may affect the outcome. Finally, we did not investigate whether AF ablation could reduce more cardiovascular risks than medical treatment among the patients with CHADS2 or CHA2DS2-VASc score of zero (the patients of CHADS2 score zero were only analysed if CHA2DS2-VASc score ≥1). As the event rate is very low in these low-risk patients, a further study with a much larger sample size is necessary to answer this question.

Conclusion

In the AF patients with high risk (CHA2DS2-VASc score ≥1), the rates of thomboembolic vascular events and the total mortality are lower in AF patients treated with catheter ablation than those receiving medical treatment. In patients who received catheter ablation, the recurrence of any atrial arrhythmia predicted the long-term cardiovascular outcome in addition to the CHA2DS2-VASc score.

Acknowledgements

This work was supported by National Science Council (NSC; Taiwan, ROC), Grant No. NSC99-2628-B-075-007-MY3.

Conflicts of interest: none declared.

References

1
Benjamin
EJ
Wolf
PA
D'Agostino
RB
Silbershatz
H
Kannel
WB
Levy
D
Impact of atrial fibrillation on the risk of death: the Framingham Heart Study
Circulation
 , 
1998
, vol. 
98
 (pg. 
946
-
52
)
2
Chugh
SS
Blackshear
JL
Shen
WK
Hammill
SC
Gersh
BJ
Epidemiology and natural history of atrial fibrillation: clinical implications
J Am Coll Cardiol
 , 
2001
, vol. 
37
 (pg. 
371
-
8
)
3
Wolf
PA
Mitchell
JB
Baker
CS
Kannel
WB
D'Agostino
RB
Impact of atrial fibrillation on mortality, stroke, and medical costs
Arch Intern Med
 , 
1998
, vol. 
158
 (pg. 
229
-
34
)
4
Calkins
H
Brugada
J
Packer
DL
Cappato
R
Chen
SA
Crijns
HJ
, et al.  . 
HRS/EHRA/ECAS expert Consensus Statement on catheter and surgical ablation of atrial fibrillation: recommendations for personnel, policy, procedures and follow-up.
Europace
 , 
2007
, vol. 
9
 (pg. 
335
-
79
)
5
Cappato
R
Calkins
H
Chen
SA
Davies
W
Iesaka
Y
Kalman
J
, et al.  . 
Updated worldwide survey on the methods, efficacy, and safety of catheter ablation for human atrial fibrillation
Circ Arrhythm Electrophysiol
 , 
2010
, vol. 
3
 (pg. 
32
-
8
)
6
Wyse
DG
Waldo
AL
DiMarco
JP
Domanski
MJ
Rosenberg
Y
Schron
EB
, et al.  . 
A comparison of rate control and rhythm control in patients with atrial fibrillation
N Engl J Med
 , 
2002
, vol. 
347
 (pg. 
1825
-
33
)
7
Nademanee
K
Schwab
MC
Kosar
EM
Karwecki
M
Moran
MD
Visessook
N
, et al.  . 
Clinical outcomes of catheter substrate ablation for high-risk patients with atrial fibrillation
J Am Coll Cardiol
 , 
2008
, vol. 
51
 (pg. 
843
-
9
)
8
Lip
GY
Nieuwlaat
R
Pisters
R
Lane
DA
Crijns
HJ
Refining clinical risk stratification for predicting stroke and thromboembolism in atrial fibrillation using a novel risk factor-based approach: the euro heart survey on atrial fibrillation
Chest
 , 
2010
, vol. 
137
 (pg. 
263
-
72
)
9
Camm
AJ
Kirchhof
P
Lip
GY
Schotten
U
Savelieva
I
Ernst
S
, et al.  . 
Guidelines for the management of atrial fibrillation: the Task Force for the Management of Atrial Fibrillation of the European Society of Cardiology (ESC)
Europace
 , 
2010
, vol. 
12
 (pg. 
1360
-
420
)
10
Gage
BF
Waterman
AD
Shannon
W
Boechler
M
Rich
MW
Radford
MJ
Validation of clinical classification schemes for predicting stroke: results from the National Registry of Atrial Fibrillation
JAMA
 , 
2001
, vol. 
285
 (pg. 
2864
-
70
)
11
Lin
YJ
Tai
CT
Kao
T
Tso
HW
Higa
S
Tsao
HM
, et al.  . 
Frequency analysis in different types of atrial fibrillation
J Am Coll Cardiol
 , 
2006
, vol. 
47
 (pg. 
1401
-
7
)
12
Lin
YJ
Tai
CT
Chang
SL
Lo
LW
Tuan
TC
Wongcharoen
W
, et al.  . 
Efficacy of additional ablation of complex fractionated atrial electrograms for catheter ablation of nonparoxysmal atrial fibrillation
J Cardiovasc Electrophysiol
 , 
2009
, vol. 
20
 (pg. 
607
-
15
)
13
Chang
SL
Tai
CT
Lin
YJ
Wongcharoen
W
Lo
LW
Tuan
TC
Udyavar
AR
, et al.  . 
The efficacy of inducibility and circumferential ablation with pulmonary vein isolation in patients with paroxysmal atrial fibrillation
J Cardiovasc Electrophysiol
 , 
2007
, vol. 
18
 (pg. 
607
-
11
)
14
Chen
SA
Hsieh
MH
Tai
CT
Tsai
CF
Prakash
VS
Yu
WC
, et al.  . 
Initiation of atrial fibrillation by ectopic beats originating from the pulmonary veins: electrophysiological characteristics, pharmacological responses, and effects of radiofrequency ablation
Circulation
 , 
1999
, vol. 
100
 (pg. 
1879
-
86
)
15
Lin
YJ
Tai
CT
Kao
T
Chang
SL
Lo
LW
Tuan
TC
, et al.  . 
Spatiotemporal Organization of the Left Atrial Substrate after Circumferential Pulmonary Vein Isolation of Atrial Fibrillation
Circ Arrhythmia Electrophysiol
 , 
2009
, vol. 
2
 (pg. 
233
-
41
)
16
Chao
TF
Lin
YJ
Tsao
HM
Tsai
CF
Lin
WS
Chang
SL
, et al.  . 
CHADS(2) and CHA(2)DS(2)-VASc scores in the prediction of clinical outcomes in patients with atrial fibrillation after catheter ablation
J Am Coll Cardiol
 , 
2011
, vol. 
58
 (pg. 
2380
-
5
)
17
Lu
TH
Lee
MC
Chou
MC
Accuracy of cause-of-death coding in Taiwan: types of miscoding and effects on mortality statistics
Int J Epidemiol
 , 
2000
, vol. 
29
 (pg. 
336
-
43
)
18
Hohnloser
SH
Crijns
HJ
van Eickels
M
Gaudin
C
Page
RL
Torp-Pedersen
C
, et al.  . 
ATHENA Investigators: effect of dronedarone on cardiovascular events in atrial fibrillation
N Engl J Med
 , 
2009
, vol. 
360
 (pg. 
668
-
78
)
19
Pappone
C
Rosanio
S
Augello
G
Gallus
G
Vicedomini
G
Mazzone
P
, et al.  . 
Mortality, morbidity, and quality of life after circumferential pulmonary vein ablation for atrial fibrillation: outcomes from a controlled nonrandomized long-term study
J Am Coll Cardiol
 , 
2003
, vol. 
42
 (pg. 
185
-
97
)
20
Hunter
RJ
McCready
J
Diab
I
Page
SP
Finlay
M
Richmond
L
, et al.  . 
Maintenance of sinus rhythm with an ablation strategy in patients with atrial fibrillation is associated with a lower risk of stroke and death
Heart
 , 
2012
, vol. 
98
 (pg. 
48
-
53
)
21
De Caterina
R
Connolly
SJ
Pogue
J
Chrolavicius
S
Budaj
A
Morais
J
, et al.  . 
Mortality predictors and effects of antithrombotic therapies in atrial fibrillation: insights from ACTIVE-W
Eur Heart J
 , 
2010
, vol. 
31
 (pg. 
2133
-
40
)
22
Glotzer
TV
Daoud
EG
Wyse
DG
Singer
DE
Ezekowitz
MD
Hilker
C
, et al.  . 
The relationship between daily atrial tachyarrhythmia burden from implantable device diagnostics and stroke risk: the TRENDS study
Circ Arrhythmia Electrophysiol
 , 
2009
, vol. 
2
 (pg. 
474
-
80
)
23
Daoud
EG
Glotzer
TV
Wyse
DG
Ezekowitz
MD
Hilker
C
Koehler
J
, et al.  . 
Temporal relationship of atrial tachyarrhythmias, cerebrovascular events, and systemic emboli based on stored device data: a subgroup analysis of TRENDS
Heart Rhythm
 , 
2011
, vol. 
8
 (pg. 
1416
-
23
)
24
Capucci
A
Santini
M
Padeletti
L
Gulizia
M
Botto
G
Boriani
G
, et al.  . 
Monitored atrial fibrillation duration predicts arterial embolic events in patients suffering from bradycardia and atrial fibrillation implanted with antitachycardia pacemakers
J Am Coll Cardiol
 , 
2005
, vol. 
46
 (pg. 
1913
-
20
)
25
Steven
D
Rostock
T
Lutomsky
B
Klemm
H
Servatius
H
Drewitz
I
, et al.  . 
What is the real atrial fibrillation burden after catheter ablation of atrial fibrillation? A prospective rhythm analysis in pacemaker patients with continuous atrial monitoring
Eur Heart J
 , 
2008
, vol. 
29
 (pg. 
1037
-
42
)
26
Ogilvie
IM
Newton
N
Welner
SA
Cowell
W
Lip
GY
Underuse of oral anticoagulants in atrial fibrillation: a systematic review
Am J Med
 , 
2010
, vol. 
123
 (pg. 
638
-
45
)
27
Chan
PS
Maddox
TM
Tang
F
Spinler
S
Spertus
JA
Practice-level variation in warfarin use among outpatients with atrial fibrillation (from the NCDR PINNACLE program)
Am J Cardiol
 , 
2011
, vol. 
108
 (pg. 
1136
-
40
)
28
Lin
LJ
Cheng
MH
Lee
CH
Wung
DC
Cheng
CL
Kao Yang
YH
Compliance with antithrombotic prescribing guidelines for patients with atrial fibrillation–a nationwide descriptive study in Taiwan
Clin Ther
 , 
2008
, vol. 
3
 (pg. 
1726
-
36
)