Antiarrhythmic drug therapy after catheter ablation for atrial fibrillation—Insights from the German Ablation Registry

Abstract Data on the optimal treatment strategy for antiarrhythmic drug therapy (AAD) after catheter ablation for atrial fibrillation (AF) are inconsistent. The present study investigates whether postinterventional AAD leads to an improved long‐term outcome. Patients from the prospective German Ablation Registry (n = 3275) discharged with or without AAD after catheter ablation for AF were compared regarding the rates of recurrences, reablations and cardiovascular events as well as patient reported outcomes during 12 months follow‐up. In patients with paroxysmal AF (n = 2138) the recurrence rate did not differ when discharged with (n = 1051) or without (n = 1087) AAD (adjusted odds ratio (OR) 1.13, 95% confidence interval (CI) [0.95–1.35]). The reablation rate was higher and reduced treatment satisfaction was reported more often in those discharged with AAD (reablation: OR 1.30, 95% CI [1.05–1.61]; reduced treatment satisfaction: OR 1.76, 95% CI [1.20–2.58]). Similar rates of recurrences, reablations and treatment satisfaction were found in patients with persistent AF (n = 1137) discharged with (n = 641) or without (n = 496) AAD (recurrence: OR 1.22, 95% CI [0.95–1.56]; reablation: OR 1.21, 95% CI [0.91–1.61]; treatment satisfaction: OR 1.24, 95% CI [0.74–2.08]). The incidence of cardiovascular events and mortality did not differ at follow‐up in patients discharged with or without AAD. In conclusion, the rates of recurrences, cardiovascular events and mortality did not differ between patients discharged with or without AAD after AF catheter ablation. However, AAD should be considered carefully in patients with paroxysmal AF, in whom it was associated with a higher reablation rate and reduced treatment satisfaction. Clinical trial registration: The trial has been registered under the number NCT01197638.


| INTRODUC TI ON
Recurrences of atrial fibrillation (AF) after catheter ablation are frustrating for patients and physicians as they go along with a reduced quality of life and an increased number of rehospitalizations. 1 In addition to electrical pulmonary vein reconnection, 2 a pathophysiological explanation for recurrences might be an early inflammatory reaction of the left atrial perimyocardial structures in response to the ablation procedure, leading to a temporary increase in atrial stretch and pressure. 3 Previous studies have shown that arrhythmia recurrences during the first year after ablation may be a predictor for a worse long-term outcome. 4 A combined interventional and long-term drug-based antiarrhythmic therapy might have the potential to further reduce the AF burden. 5,6 The evidence regarding postinterventional antiarrhythmic drug therapy (AAD) after catheter ablation is yet sparse and incongruent, leaving clinicians without a clear recommendation. [7][8][9] Our present study's aim was to analyze the effect of prescription of AAD at discharge after AF catheter ablation on the recurrence and reablation rate, mortality and patient reported outcomes during a 12-month follow-up period.

| German Ablation Registry
The German Ablation Registry (NCT01197638) is a prospective multicenter registry that is managed by the "Institut für Herzinfarktforschung" (IHF). Patients >18 years of age were enrolled in 55 participating centers in Germany between January 2007 and January 2010. Trial development, data acquisition, and clinical monitoring were organized by the IHF.

| Patient selection and study design
Patients from the German Ablation Registry presenting for their first AF catheter ablation were included into the analysis (see Figure 1).
The long-term outcome of individuals with prescription AAD of Vaughan Williams class I or class III (AAD group) at the timepoint of discharge after ablation was compared to the outcome of individuals without prescription of specific antiarrhythmic drugs (no AAD group). Betablockers were accepted as a baseline therapy in both groups. Patients with AV-node ablation, arrhythmia recurrence before discharge or patients taking class IV AAD or a combination of class I/class III AAD were excluded from analysis.

| Ablation procedure
Patients were treated according to the respective institutional standards and the contemporary guidelines as described before. 9 In brief, patients underwent the procedure in the fasting state under sedation using midazolam or propofol and fentanyl or sufentanil in most cases. Oral anticoagulation with vitamin K antagonists was stopped a few days before the ablation procedure and substituted with low-molecular heparin. During procedures intravenous heparin was administered aiming at an activated clotting time of 250-300 s.
An electroanatomical 3D-mapping system such as Carto ® (Biosense Webster) or EnSite NavX™ (St. Jude) and the energy source (radiofrequency or cryothermal energy) was used at the operator's discretion.

| Analysis of follow-up data
In addition to the clinical follow-up which was performed according to local standards of the respective participating institution, a F I G U R E 1 Flow-chart of the study design. Patients from the German Ablation Registry presenting for their first catheter ablation for atrial fibrillation were included into the analysis. Patients with paroxysmal or persistent atrial fibrillation were analyzed separately. The long-term outcome of individuals taking antiarrhythmic drug therapy of Vaughan Williams class I or class III at discharge after ablation was compared to the outcome of individuals without prescription of specific antiarrhythmic drugs. Abbreviations: AF, atrial fibrillation  Table S1) was reported on the standardized questionnaire and classified as moderate or severe. The endpoints mortality, MACE (major adverse cardiac event; including death, myocardial infarction), MACCE (major adverse cardiac and cerebrovascular event; including death, myocardial infarction, ischemic stroke) and a quadruple safety endpoint (death, myocardial infarction, ischemic stroke, major bleeding) were analyzed separately.

| Statistics
Descriptive statistics are presented as count and percentage for categorical variables and as mean ± standard deviation or median

| Baseline parameters
In total, 3275 patients with AF were included into the study. Baseline parameters are shown in Table 1 for patients with paroxysmal AF (pAF, n = 2138) and in Table 2 for patients with persistent AF (persAF, n = 1137).
In patients with persAF, additional linear lesions besides PVI were ablated in 22.9% (147/641) of patients of the AAD group and 17.7% (88/496) of patients in the no AAD group (p = .032, see Table 4).
The mean follow-up duration was 477 ± 106 days. Peripheral artery disease a 0.6% 1.0% .63 Pacemaker/ICD 6.8% 5.7% .31 Note: Values are presented as percent (%) of available data sets or mean ± standard deviation. p < .05 is considered statistically significant.
Abbreviations: AAD, antiarrhythmic drugs; COPD, chronic obstructive pulmonary disease; ICD, implanted cardioverter-defibrillator; ICM, ischemic cardiomyopathy; LV, left-ventricular. Antiarrhythmic drugs include class I/III agents. a Data available for 12%-17% of patients due to later inclusion of the variable into the study.
Patients discharged with AAD were still on medication at last follow-up in 40.7% (234/547) of cases while 20.5% (86/448) of patients that were initially discharged without AAD were on medication at last follow-up.  Table 5). Death during follow-up occurred in 1.1% in the AAD group versus 1.2% in the no AAD group (p = .85, see Table 6).

| Adverse events during follow-up
Further details for all groups are shown in Table S1.

| Symptom status at follow-up
After catheter ablation for pAF 81.5% (785/963) of the AAD group versus 84.2% (837/994) of the no AAD group reported either no remaining symptoms or reduced symptoms (p = .11; see Figure 3A).
The multivariable regression analysis of patients with improvement of symptoms showed no significant differences between groups in pAF and persAF patients, respectively (see Figure 2).

| Patients' treatment satisfaction
Patients with pAF discharged with AAD were significantly less often fully satisfied with their treatment (treatment "successful" in 61.1%   Figure 3B).
Data on treatment satisfaction were not available for all patients as this parameter was included into the trial during a protocol amendment.

| DISCUSS ION
The main findings of this study are: 1. In patients with pAF the recurrence and rehospitalization rates after catheter ablation did not differ between individuals discharged with or without postinterventional AAD in the adjusted analysis. However, the reablation rate was higher in patients discharged with AAD.
2. Patients with pAF discharged with AAD rated the treatment as "not successful" more often than patients discharged without AAD.
3. In patients with persAF the rates of recurrences, rehospitalizations and reablations did not differ between individuals discharged with or without AAD in the adjusted analysis.

No matter if discharged with or without postinterventional AAD,
the rates of cardiovascular events did not differ in patients with pAF and persAF during long-term follow-up.

| Recurrence and reablation rates
For many AF patients, the freedom from AAD is one of the de- Note: Values are presented as percent (%) of available data sets or mean ± standard deviation. p < .05 is considered statistically significant.
Abbreviations: AAD, antiarrhythmic drugs; MACCE, major adverse cardiac and cerebrovascular event; MACE, major adverse cardiac event; MI, myocardial infarction. Details on the types of adverse events are shown in Table S1b. a Kaplan-Meier estimates at 366 days after index discharge, compared by log-rank test.
TA B L E 5 12-Months clinical outcomepatients with paroxysmal atrial fibrillation 12-months period after ablation we found no reduction in recurrences and rehospitalizations in patients with pAF and persAF when discharged with AAD. Thus, our data stand in line with experiences from the EAST-AF, AMIO-CAT, and also the 5A trial, which investigated postinterventional AAD. 7,8,13,14 Despite less recurrences during the limited "blanking period," none of those trials showed an overall reduction of recurrences during long-term follow-up. A possible reason might be, that besides the temporary postinterventional inflammatory triggers of AF, 15 gaps in the circular PVI lesion or unsuccessful ablation of triggers unrelated to pulmonary veins are potential causes of longterm recurrences. 16 On the contrary, the POWDER AF study did report less recurrences during long-term follow-up for pAF patients with continued AAD; however, it followed a different approach, including only individuals without recurrences during the "blanking period." 10 The above-mentioned considerations of long-term AAD might not fit for all AF patients. While reablation rates in persAF patients in the present study were not reduced by postinterventional AAD, pAF patients had a 1.3 times higher rate of reablations when AAD was prescribed at discharge. This effect was observed, regardless of similar recurrence rates in patients discharged with or without AAD. Hypothetically, while the recurrences themselves were tolerable for the patients, the perspective of a prolongation of AAD, which was probably suggested as an alternative option to reablation in patients discharged with AAD, motivated the relatively healthy pAF patients to undergo another ablation procedure.

| Patient satisfaction with treatment
Patient reported outcomes are an important part of integrated AF management as the patients' experience of arrhythmia and treatment burden are often subjective and not always represented in the usual clinical endpoints. 9,17 In our present analysis, pAF patients discharged without AAD had a significantly higher level of satisfaction with treatment than all other groups (pAF discharged with AAD, persAF discharged with/without AAD). A detailed analysis by Brachmann et al., 18 including all patients with supraventricular arrhythmias from the German Ablation Registry showed that treatment satisfaction is often influenced by arrhythmia recurrences. At least in patients with pAF this seems not to be the case, as recurrences did not differ significantly between groups.
However, pAF patients discharged with AAD had a significantly higher rate of AF reablations as compared to patients without AAD use, which might influence treatment satisfaction. A study by Kany et al. found that patient satisfaction was similar after primary and reablation procedures. 19 This leads to the conclusion Note: Values are presented as percent (%) of available data sets or mean ± standard deviation. p < .05 is considered statistically significant.
Abbreviations: AAD, antiarrhythmic drugs; MACCE, major adverse cardiac and cerebrovascular event; MACE, major adverse cardiac event; MI, myocardial infarction. Details on the types of adverse events are shown in Table S1b. a Kaplan-Meier estimates at 366 days after index discharge, compared by log-rank test.
TA B L E 6 12-months clinical outcomepatients with persistent atrial fibrillation that at least in the group of patients with pAF postinterventional AAD might reduce treatment satisfaction. Patients with persAF are possibly more used to medication use (due to a higher number of comorbidities) and were not so much bothered by additional drug treatment. treatment approach using catheter ablation and AAD reduces event rates more than one of the options alone. In our analysis rates of cardiovascular events were low and the adjusted analysis showed no significant differences in the occurrence of the combined endpoint "MACCE" between patients discharged with or without AAD. This was observed in both patients with pAF and persAF.

| CON CLUS ION
Patients discharged with AAD after AF catheter ablation had similar rates of recurrences, rehospitalizations and cardiovascular events during follow-up as patients discharged without AAD. However, in pAF patients AAD at discharge was associated with reduced treatment satisfaction and a higher reablation rate. Therefore, postinterventional prescription of AAD has to be carefully considered, especially after ablation of pAF.

E TH I C S A PPROVA L /CO N S E NT TO PA RTI CI PATE
The study has been approved by the local ethics committee and has therefore been performed in accordance with the ethical standards laid down in the 1964 Declaration of Helsinki and its later amendments. All individual participants included into the study gave written informed consent. Patients signed informed consent regarding publishing their data.

DATA AVA I L A B I L I T Y S TAT E M E N T
The underlying data are available upon reasonable request.