Comparison of uninterrupted direct oral anticoagulation with vitamin‐K antagonists during AF‐ablation in the clinical routine. A single center register

Abstract Background Uninterrupted direct oral anticoagulation (DOAC) in AF‐ablation is recommended, proven by randomized trials. The outcome and the periinterventional differences between DOACs and VKA in the real world clinical practice are discussed controversial. Hypothesis To investigate efficiency and safety of uninterrupted DOAC therapy compared to VKA during AF‐Ablation in real world setting with a focus on periinterventional heparin dosage.


| INTRODUCTION
Thromboembolic events on the one hand and bleeding complications on the other hand represent the most commonly feared complications of AF-ablation. 1 The efficacy and safety of continuous periinterventional VKA or DOAC therapy with regard to thromboembolic prophylaxis and bleeding risk has been confirmed in randomized trials. [2][3][4][5] Based on these studies, the guidelines also support AF-ablation under ongoing oral anticoagulation either with DOACs or with VKA. 6 Due to the relatively small number of cases in the randomized trials, further clinical registers are desirable as additional confirmation of the randomized study data in everyday routine. Therefore in the present clinical register the efficiency and safety of uninterrupted DOAC therapy was compared to continued VKA in a real world setting.
In addition, the current study focuses on the different response of heparin to periinterventional ACT in dependence on the ongoing oral anticoagulation.

| Study design
The current study was a single-centre register performed in compliance with the guidelines for good clinical practice and the Declaration of Helsinki. The study was approved by the institutional ethical review board. All data were collected, managed and analyzed at the Heart Centre, University of Dresden (ethics approval: University of Dresden: The primary endpoint of this study was the occurrence of periinterventional complications during AF ablation procedures (pericardial effusion, transient ischemic attack, stroke, access complications) using two different anticoagulation regimes (VKA or DOACs).
The secondary endpoints were the change of hemoglobin levels as marker of periinterventional blood loss, the doses of acquired heparin to reach the target ACT and the mean ACT during the procedure using VKA or DOACs.

| Study population and protocol
Eligible subjects were consecutive male or females >18 years of age suffering from symptomatic drug resistant atrial fibrillation (AF) and requiring catheter ablation. No other inclusion criteria were necessary.
Exclusion criteria were an interrupted anticoagulation for more than 12 h before the ablation procedure or a bridging of the VKA with heparin. If a patient fulfilled all inclusion criteria and none of the exclusion criteria, the clinical data, intraprocedural data, and complications were analyzed. Five different experienced operators performed all ablation procedures.
The DOAC group consisted of all consecutive patients, which were ablated under uninterrupted anticoagulation with direct oral anticoagulants (DOAC). In these patients the morning dose of the DOAC was given to the patient prior to the procedure. Also the evening dose of apixaban and dabigatran was administered to the patient on the day of the ablation procedure.
The VKA group included all consecutive patients, ablated under continuous administration of vitamin K antagonists (VKA). A statistical based matching of both groups was not performed.

| Ablation procedure
If the patients were ≥ 3 weeks on therapeutic OAC no preinterventional transesophageal echocardiogram (TOE) was required.
The AF ablation procedure was performed under sedation utilizing midazolam fentanyl and propofol. One quadripolar diagnostic catheter was advanced into right ventricular apex and one decapolar catheter into the coronary sinus under fluoroscopic guidance. A single transseptal puncture was performed with a steerable intra-cardiac sheath were removed in the EP laboratory, the access sites were compressed manually and a pressure bandage was applied for 6 h.

| Follow-up
Directly after the ablation procedure and 24 h after the ablation a neurological examination was conducted. Additionally, within 24 h after the ablation procedure the access sites were clinically examined.
In case of suspect local findings (hematoma, occurrence of new bruits, painful groin) a doppler ultrasound examination of the access site was performed to rule out pseudoaneurysm or AV-fistula. Further, a transthoracic echocardiography to rule out a pericardial effusion was carried out.

| Statistical analysis
Data were tested for normal distribution. Results of continuous variables are expressed as means ± standard deviation. Statistical analyses were done using the 2 tailed, unpaired Student's t test. Level of significance was set to p < 0.05. Categorical variables are presented as total number with comparison using chi-square statistics and Fisher exact test. If more than 2 groups were analyzed, a one-way ANOVA test was performed. Post hoc analyses have been applied using Bonferroni method. Significance level was set to p < 0.05.  (Table 1). Of note the patients in the VKA group were significantly older (67 ± 8 years) in comparison to the DOAC group (64 ± 9 years). Caused by the higher age in the VKA group the patients in the VKA group had a significantly higher CHA 2 DS 2 -VASC Score (VKA: 2.7 vs. DOAC: 2.5) and a significantly higher HAS-BLED-Score (VKA: 1.3 vs. DOAC: 1.0). The patients in the DOAC group suffered significantly more from diabetes compared to the VKA patients (DOAC: 17.5% vs VKA: 9.7%). There were no relevant differences in gender, co-morbidities (except diabetes) and concomitant medications.

| Periinterventional complications (primary endpoint)
The periinterventional complications are illustrated in Table 2

| Change of periinterventional hemoglobin levels (secondary endpoint)
The comparison of the change of the hemoglobin levels as marker for periinterventional blood loss between both groups is illustrated in

| Mean ACT and heparin doses during AF ablation (secondary endpoint)
The comparison of the acquired doses of heparin to reach the target ACT is shown in Figure 2. It could be revealed, that patients treated with direct factor Xa inhibitors apixaban ( Figure 3. Further, the quotient of the used heparin dose and the resulting ACT was calculated to get a better view of the relative heparin sensitivity in the different anticoagulation groups. Also this heparin/ACT quotient confirmed the significantly higher heparin dose per ACT change in the patients treated with direct Xa inhibitors compared to VKA or dabigatran (results shown in Figure S1).

| DISCUSSION
The findings of the present real world register suggest, that AF-ablation during AF-ablation. There were minor differences in baseline characteristics between the current registry and the randomized trials. In the randomized trials, the mean CHA2DS2-VASc Scores were between 1.5 and 2.4. 3,7 In contrast, in the current study the mean CHA2DS2-VASc Score was 2.5 in the DOAC and 2.7 in the VKA group. In light of the aging population with rising co-morbidities these data support the uninterrupted oral anticoagulation even in patients with higher thromboembolic and bleeding risk. In the current study 1 (0.4%) transient ischemic attack (TIA) in the DOAC group and 1 TIA and 1 stroke (1.0%) were registered in the VKA group. This thrombembolic complication rate was similar to the randomized trials. Thus, in the VENTURE-AF trial 0.8% of patients and in the AXAFA -AFNET 5 trial 0.3% of patients had thrombembolic events. 5,7 But in this context it should be noted that the occurrence of intraprocedural strokes and TIAs is not solely depended on preformed thrombi caused by ineffective preinterventional oral anticoagulation, but also on air emboli or ineffective heparinization during the ablation procedure. In the current register all three patients, suffered from stroke or TIA have had a sufficient periprocedural ACT >300 s and air embolization were not noticed during the procedure.
Therefore the exact underlying mechanism of the periinterventional thromboembolic events cannot be answered validly in this register.  Thus, in our study we registered 1.2% pericardial tamponades in the DOAC group and 1.5% in the VKA group. This number of pericardial tamponades is completely in line with those of other large AF-ablation studies irrespective of the anticoagulation strategy. 1,6,8 This finding suggests, that perioperative caution, rather than the anticoagulation strategy, has the greatest impact on tamponade rates. In the case of pericardial tamponade, complication management is very important for the outcome. 10 In this context, only one case in the VKA group required cardiac surgery in the current study. All other tamponades could be treated conservatively using pericardial drainage with patient retransfusion of the blood without heparin antagonization to avoid clotting in the pericardial space or in the drainage.
One of the most important findings of this register, that every interventionalist should know, is the different response of heparin to periinterventional ACT in dependence on the ongoing oral anticoagulation. In this context, we observed a significantly increased heparin dose requirement in patients treated with apixaban or rivaroxaban to reach the aim ACT compared to patients treated with VKA or dabigatran. This effect was also observed in the randomized trials. [3][4][5]11 It is known that ACT-guided perioperative anticoagulation with heparin helps to prevent thrombembolic complications without the increase of clinical relevant bleeding. Therefore a periinterventional ACT between 300 and 350 s is generally aimed. 12 Under VKA 100 U/kg should be administered to achieve this ACT value. In contrast, under oral anticoagulation with Xa-inhibitors the initial dose of heparin should be between 120 and 130 U/kg. 13

| CONCLUSIONS
The use of uninterrupted DOACs during AF-ablation is safe and the amount of thrombembolic and bleeding events was very low and similar to that for continuing VKA in the clinical routine.
Of note, under the factor Xa inhibitors, there is a significant lower anticoagulatory effect of heparin, as measured by the activated clotting time, requiring higher periinterventional heparin doses. These findings should be translated to other left-sited cardiovascular interventions.

ACKNOWLEDGMENT
We thank all staff of the EP-laboratory, Heart Center Dresden for supporting this register.