Effect of uninterrupted dabigatran or rivaroxaban on achieving ideal activated clotting time to heparin response during catheter ablation in patients with atrial fibrillation

Atrial fibrillation (AF) is the most common supraventricular arrhythmia, and current estimates reveal that more than 33 million individuals worldwide suffer from AF.^[1] The AF catheter ablation is generally recommended as class I rhythm control therapy for patients with symptomatic paroxysmal AF refractory or intolerant to at least one class I or III antiarrhythmic medication, and as class IIa for persistent AF.^[1] The AF ablation returning to normal rhythm can improve cardiac function and quality of life.^[2] However, AF ablation tends to form a prethrombotic state during and post catheter ablation of AF, which increases the risk of thromboembolic complications. Therefore, current guidelines suggest uninterrupted use of oral anticoagulants to lower periprocedural thromboembolic events.^[1]

Oral anticoagulants for patients with AF undergoing catheter ablation include traditional vitamin K antagonist (warfarin) and new oral anticoagulants (NOACs), both of which have a potential risk for bleeding and thromboembolic complications.^[3] In the last few years, more and more AF patients continue to take NOACs agents, posing a management challenge for periprocedural anticoagulation in patients undergoing AF ablation. Unfractionated heparin has been recommended for patients who uninterrupted take NOACs agents during AF ablation. Monitoring intraoperative activated clotting time (ACT) in the range of 300–400 s during AF catheter ablation is considered to be the key to reduce the thromboembolic complications and the risks of bleeding,^[4] whereas these data have been obtained mostly from patients treated with warfarin. Although a large number of studies have focused on the feasibility and safety of uninterrupted NOACs for periprocedural anticoagulation in patients undergoing AF ablation,^[3–6] only a few studies reported the evaluation of NOACs for intraoperative anticoagulation. Previous study showed that patients taking rivaroxaban needed more unfractionated heparin doses for ACT > 300 s compared with the patients taking warfarin, and the mean ACT values was lower in patients treated with rivaroxaban than those treated with warfarin.^[7] Moreover, a retrospective analysis demonstrated that the average time for achieving a target ACT > 300 s was significantly longer in the dabigatran group (DG) than that in the warfarin group, and the unfractionated heparin dose for the procedure was higher in the DG than that in the warfarin group.^[8] However, few comparative analyses reported on the use of NOACs as intraoperative anticoagulation. In addition, the effect of NOACs on ACT to heparin response during AF ablation has not been studied in Chinese population.

To address this issue, we investigated the effect of dabigatran or rivaroxaban on intraoperative ACT value to heparin response in Chinese patients undergoing AF ablation.

METHODS

Study Population and Follow-up

A single-center study was conducted, which included 173 non-valvular AF patients who received dabigatran or rivaroxaban and underwent AF ablation from October 2018 to December 2021. Exclusion criteria included valvular heart diseases, pulmonary hypertension, acute coronary syndrome, hypertrophic cardiomyopathy, heart failure with left ventricular ejection fraction < 40%, chronic kidney disease stage 4–5, and left atrial size > 50 mm. According to dabigatran and rivaroxaban clinical trials in Caucasian patients and the manufacturer’s instructions, dabigatran 150 mg was administered twice a day or rivaroxaban 15–20 mg was administered once a day in the morning. A low dose of dabigatran (110 mg twice daily) or rivaroxaban (5 mg twice daily) was used for patients with following conditions: advanced age (> 75 years), lower body weight (< 60 kg), moderate renal dysfunction (creatinine clearance rate: 30–49 mL/min), a history of gastrointestinal hemorrhage, or coadministration of antiarrhythmic drugs (amiodarone or verapamil). Two NOACs were used without interruption through the ablation procedure. Dabigatran was given twice daily at 7:00 and 19:00, whereas rivaroxaban was administered once daily at 7:00 (some small dose of rivaroxaban 5 mg was administered twice daily at 7:00 and 19:00). All NOACs were started at least three weeks before the ablation and were used uninterruptedly during the periprocedural period. The anticoagulation therapy was continued for eight weeks after AF ablation for patients with a CHA₂DS₂-VASc score of 0–1 (male) or 0–2 (female), and those with scores ≥ 2 (male) or ≥ 3 (female) were given long-term anticoagulant treatment. All patients were followed up in the Department of Cardiology, the Sixth Medical Centre, Chinese PLA General Hospital, Beijing, China for 30 days after the AF ablation because most of the periprocedural complications occurred during this time period.

Catheter Ablation for AF

Before the procedure, all patients were assessed cardiac function and left atrial size and the absence of atrial thrombus by transthoracic and transesophageal echocardiography. The morning and afternoon sessions were begun at 9:00 and 12:00, and 13:00 and 16:00, respectively. Vascular access was achieved through the right femoral vein. Steerable decapolar electrode was advanced in the coronary sinus. After the insertion of all sheaths, initial heparin boluses were administered at 70–80 U/kg for the patients with normal weight [body mass index (BMI): 18.5–23.9 kg/m²]. As for patients with overweight (BMI: 24.0–27.9 kg/m²) and obesity (BMI ≥ 28.0 kg/m²), the initial dose of heparin was determined as 80–120 U/kg for those according to the operator’s experience. The ACT was monitored every 30 min after the heparin bolus administration, and additional heparin boluses were administered to maintain target ACT in the range of 250–350 s. After the successful puncture of atrial septum under fluoroscopy, the left and right pulmonary venography was performed. The PentaRay high-precision mapping electrode catheter (Biosense Webster, California, USA) was advanced into the left atrium, and the three-dimensional geometry of the left atrium was reconstructed under the guidance of CARTO. Encircling linear pulmonary vein (PV) isolation was performed with a smart-touch large-tip ablation catheter, using power mode under the guidance of ablation index (AI): the anterior wall power was 35 W, with an AI of 450 per point; and the posterior wall power of 30 W, with an AI of 350 per point. The power of the top and bottom walls was 30 W, with an AI of 400 per point. After ablating all the PVs, direct current cardioversion was performed if AF continued. Sending the PentaRay catheter into the PVs on both sides showed the absence of PV potential, and the high-output pacing within the PV revealed the exit block, indicating that the PVs were completely electrically isolated. After successful isolation of all the PVs, the patients were administered isoproterenol (5–20 μg/min), and were given burst atrial pacing to determine that AF or other atrial arrhythmias were not induced. If AF was induced and sustained, additional procedures were performed, including ablation targeting of complex fractionated atrial electrograms, linear ablation of left atrium and isolation of superior vena cava. If AF persisted despite all the above procedures, external cardioversion was performed. Cavotricuspid isthmus ablation was performed as required.

Complications

Cardiac tamponade, pericardial effusion, intracranial hemorrhage, gastrointestinal hemorrhage, and hematomas of the puncture site were defined as bleeding complications. Major bleeding complications were classified as any bleeding that required blood transfusion, surgical intervention, and pericardial effusion with drainage. Minor bleeding complications included hematomas of the puncture site and pericardial effusions not requiring drainage. Stroke, transient ischemic attack (TIA), peripheral arterial embolism, and pulmonary embolism were considered thromboembolic complications. The diagnosis of cerebral hemorrhage or cerebral infarction should be confirmed by brain computed tomography and magnetic resonance imaging. All aforementioned complications received prompt and appropriate treatments, and no death or disability events occurred.

Statistical Analysis

Continuous variables were presented as mean ± SD, and categorical variables were expressed as counts (percentages). Unpaired Student’s t-test and Mann-Whitney U test were used to compare continuous variables when appropriate, and the Pearson’s chi-squared test or Fisher’s exact probability test was used to compare categorical variables. Two-sided P-value < 0.05 were considered statistically significant. All statistical analyses were performed with SPSS 18.0 (SPSS Inc., IBM, Armonk, IL, USA).

RESULTS

Patient Characteristics and Procedural Parameters

A total of 173 patients who underwent the ablation procedure were divided into the DG and the rivaroxaban group (RG). The subjects consisted of 101 patients treated with dabigatran and 72 patients treated with rivaroxaban. Baseline characteristics of the subjects were summarized in Table 1. No significant differences were found in the baseline characteristics of age, BMI, clinical history, AF status, echocardiographic parameters and medications between the two groups; whereas the RG had more female, and the DG had longer activated partial thromboplastin time. Procedural values, such as procedure time did not differ among the DG and the RG.

Table  1.  Baseline characteristics of the patients.

Characteristics	Dabigatran group (n = 101)	Rivaroxaban group (n = 72)	P-value
Age, yrs	60 ± 11	62 ± 10	0.180
Female	24 (24%)	28 (39%)	0.032
Body mass index, kg/m2	26 ± 4	26 ± 3	0.369
Paroxysmal atrial fibrillation	65 (64%)	49 (68%)	0.613
Current smoking	40 (40%)	24 (33%)	0.400
Alcohol abuse	43 (43%)	27 (38%)	0.503
History
Coronary artery disease	37 (37%)	23 (31%)	0.523
History of heart failure	4 (4%)	2 (3%)	1.000
Hypertension	57 (56%)	43 (60%)	0.666
Diabetes mellitus	17 (17%)	18 (25%)	0.187
Hyperlipidemia	25 (25%)	27 (38%)	0.071
Peripheral vascular disease	13 (13%)	13 (18%)	0.347
Prior stroke/Transient ischemic attack	8 (8%)	9 (13%)	0.319
Chronic obstructive pulmonary disease	3 (3%)	1 (1%)	0.642
CHA2DS2-VASc score	1.8 ± 1.5	2.2 ± 1.7	0.070
Echocardiography data
Left atrium size, mm	38 ± 5	39 ± 5	0.132
Left ventricular ejection fraction, %	60 ± 6	60 ± 5	0.887
Laboratory data
Glomerular filtration rate, mL/min	93 ± 28	86 ± 19	0.056
Brain natriuretic peptide, pg/mL	263 (104–706)*	267 (80–604)*	0.755
Prothrombin time, s	14 ± 1	14 ± 3	0.566
Activated partial thromboplastin time, s	39 ± 10	34 ± 6	0.000
Medication
Antiplatelet drugs	20 (20%)	10 (14%)	0.311
Proton pump inhibitor/Histamine H2-receptor antagonist	95 (94%)	63 (88%)	0.131
Nonsteroidal anti-inflammatory drug	8 (8%)	1 (1%)	0.082
Antiarrhythmic drug	26 (26%)	23 (32%)	0.372
Anticoagulation therapy
Dabigatran (110 mg twice daily)/Rivaroxaban (5 mg twice daily)	41 (41%)	6 (8%)	< 0.01
Procedural values
Procedure time, min	178 ± 44	172 ± 54	0.422
Morning session	52 (52%)	30 (42%)	0.202
Cardioversion	40 (40%)	22 (31%)	0.221
Data are presented as means ± SD or n (%). *Presented as median (interquartile range).

 | Show Table

DownLoad: CSV

Intraprocedural ACT Values and Heparin Requirements

The requirements for initial heparin and total heparin for the procedure were higher in the DG than in the RG. Additionally, mean ACT and percentage of therapeutic ACTs (250–350 s) were higher in the RG than in the DG, as shown in Figure 1. Similarly, subgroup analysis showed that there were significant differences between the two groups in age < 65 years, men, glomerular filtration rate > 90 mL/min, and no history of diabetes mellitus or stroke (Table 2). The mean ACT values monitored every 30 min during operation in the RG were relatively higher than that in the DG, as shown in Figure 2. The average time required to achieve the target ACT (250–350 s) was significantly longer in the DG than in the RG, as shown in Figure 3. With regard to the aforementioned results, there was no difference between morning and afternoon in the two groups, while the percentage of ACTs within the therapeutic range was higher in the morning than in the afternoon in the DG (Figure 4).

Figure  1.  Comparison of requirements for initial heparin, total heparin, mean procedural ACT, and percentage of ACTs in therapeutic range between the DG and the RG.

Error bars indicate mean and standard deviations. ACT: activated clotting time; DG: dabigatran group; RG: rivaroxaban group.

DownLoad: Full-Size Img PowerPoint

Table  2.  Comparison of requirements for initial heparin and total heparin, mean ACT and ACT TTR between the DG and the RG in different variables.

	Initial heparin dosing, U/kg				Total heparin dosing, U/kg				Mean ACT, s				ACT TTR, %
	DG	RG	P-value		DG	RG	P-value		DG	RG	P-value		DG	RG	P-value
Sex, %
Male	89.9 ± 16.2	81.1 ± 24.9	0.02		142.9 (117.7–164.1)*	90.5 (70.9–116.1)*	< 0.01		264.0 ± 29.4	285.3 ± 37.2	< 0.01		60.0 (33.3–80.0)*	75.0 (66.7–100.0)*	< 0.01
Female	82.2 (72.5–104.7)*	73.9 (50.8–89.2)*	0.22		117.1 (83.3–154.3)*	92.4 (63.7–139.0)*	0.17		268.4 ± 30.4	272.2 ± 32.1	0.67		66.7 (50.0–100.0)*	77.5 (50.0–92.9)*	0.61
Age, yrs
< 65	90.7 ± 14.0	78.6 ± 22.5	< 0.01		142.7 (116.7–160.7)*	89.3 (69.4–116.8)*	< 0.01		261.2 ± 25.5	284.9 ± 41.6	< 0.01		50.0 (33.3–75.0)*	75.0 (66.7–100.0)*	< 0.01
≥ 65	89.3 (65.3–105.3)*	74.7 (59.1–91.7)*	0.13		125.0 (98.2–166.7)*	91.7 (70.3–141.7)*	0.02		270.1 ± 34.0	276.2 ± 29.3	0.40		66.7 (33.3–100.0)*	80.0 (50.0–100.0)*	0.09
Smoking, %
No	88.2 (76.8–101.1)*	80.4 (60.4–96.2)*	0.03		138.2 (107.1–159.1)*	98.2 (68.2–139.8)*	< 0.01		264.3 ± 28.8	275.7 ± 32.4	0.06		50.0 (33.3–80.0)*	75.0 (58.4–100.0)*	< 0.01
Yes	90.1 ± 17.6	78.8 ± 24.8	0.01		139.3 (111.2–171.8)*	78.7 (70.3–103.0)*	< 0.01		266.1 ± 30.9	289.5 ± 40.6	0.01		66.7 (41.7–90.0)*	80.0 (66.7–100.0)*	0.01
Drinking, %
No	88.8 (75.8–101.1)*	79.4 (60.8–89.3)*	0.01		133.9 (101.6–153.2)*	95.5 (65.0–135.5)*	< 0.01		264.5 ± 27.6	271.6 ± 29.0	0.21		66.7 (33.3–100.0)*	75.0 (50.0–85.7)*	0.07
Yes	90.1 ± 17.3	81.3 ± 25.7	0.09		143.8 (116.7–171.9)*	78.1 (70.3–125.0)*	< 0.01		265.6 ± 32.3	294.5 ± 41.3	< 0.01		60.00 (33.3–80.0)*	100.0 (66.7–100.0)*	< 0.01
Hypertension, %
No	88.0 (75.2–100.0)*	83.3 (60.8–101.9)*	0.52		147.5 (112.5–173.4)*	98.7 (72.5–136.4)*	< 0.01		259.7 ± 26.7	284.7 ± 41.0	< 0.01		55.0 (41.7–69.1)*	80.0 (50.0–100.0)*	0.01
Yes	91.9 (80.4–105.3)*	71.4 (62.5–87.5)*	< 0.01		133.9 (110.3–149.7)*	87.9 (69.4–116.8)*	< 0.01		269.1 ± 31.2	277.5 ± 32.0	0.19		66.7 (33.3–100.0)*	75.0 (66.7–100.0)*	0.04
Diabetes mellitus, %
No	90.6 (77.0–103.6)*	73.7 (60.8–89.3)*	< 0.01		142.7 (110.7–166.0)*	88.6 (69.4–125.0)*	< 0.01		262.1 ± 27.7	280.2 ± 38.3	< 0.01		55.0 (33.3–75.0)*	77.5 (66.7–100.0)*	< 0.01
Yes	87.0 (74.6–105.3)*	83.9 (58.3–101.9)*	0.51		131.6 (101.6–141.7)*	100.5 (74.7–148.4)*	0.26		279.4 ± 34.8	280.6 ± 27.5	0.91		100.0 (40.0–100.0)*	70.9 (66.7–100.0)*	0.57
Stroke, %
No	90.3 (77.1–104.2)*	75.0 (60.8–96.2)*	< 0.01		138.9 (110.3–164.1)*	89.3 (69.4–132.6)*	< 0.01		264.0 ± 29.3	280.9 ± 36.4	< 0.01		60.0 (33.3–80.0)*	75.0 (66.7–100.0)*	< 0.01
Yes	89.3 ± 15.2	76.4 ± 21.0	0.17		128.1 (107.6–153.1)*	99.6 (87.9–125.0)*	0.25		276.4 ± 31.6	276.8 ± 32.2	0.98		73.4 (36.7–100.0)*	66.7 (66.7–100.0)*	0.96
Glomerular filtration rate, mL/min
< 90	89.3 (75.8–98.2)*	84.1 (64.7–101.9)*	0.42		139.7 (111.3–160.7)*	101.9 (79.9–141.7)*	< 0.01		268.3 ± 33.0	280.5 ± 37.4	0.09		66.7 (33.3–80.0)*	80.0 (50.0–100.0)*	< 0.01
≥ 90	91.3 ± 16.0	71.5 ± 21.3	< 0.01		138.2 (108.7–166.0)*	77.4 (60.8–99.6)*	< 0.01		260.7 ± 24.0	280.2 ± 33.8	< 0.01		50.0 (33.3–91.7)*	75.0 (66.7–100.0)*	0.02
Anti-platelet therapy, %
No	89.3 (76.9–100.6)*	76.2 (60.8–96.2)*	< 0.01		138.9 (110.3–164.1)*	93.6 (70.3–132.6)*	< 0.01		263.1 ± 27.6	279.4 ± 35.7	< 0.01		60.0 (33.3–80.0)*	75.0 (66.7–100.0)*	< 0.01
Yes	94.2 (76.5–105.3)*	79.0 (53.6–89.3)*	0.09		130.3 (114.4–154.5)*	82.01 (70.3–116.3)*	0.02		272.9 ± 36.0	286.1 ± 36.8	0.35		66.7 (36.7–90.0)*	81.7 (66.7–100.0)*	0.16
Data are presented as means ± SD. *Presented as median (interquartile range). ACT: activated clotting time; DG: dabigatran group; RG: rivaroxaban group; TTR: target therapeutic range.

 | Show Table

DownLoad: CSV

Figure  2.  Changes in ACT in each anticoagulant group.

ACT: activated clotting time; DG: dabigatran group; RG: rivaroxaban group.

DownLoad: Full-Size Img PowerPoint

Figure  3.  Comparison of mean time required to reach target ACT > 250 s after initial heparin bolus between the DG and the RG.

Y-axis represents the proportion of patients who reached the target ACT in each time point after initial heparin bolus. ACT: activated clotting time; DG: dabigatran group; RG: rivaroxaban group.

DownLoad: Full-Size Img PowerPoint

Figure  4.  Requirements for initial heparin, total heparin, mean procedural ACT, and percentage of ACTs in therapeutic range between the morning session and the afternoon session.

Error bars indicate mean and standard deviations. ACT: activated clotting time; DG: dabigatran group; RG: rivaroxaban group.

DownLoad: Full-Size Img PowerPoint

Complications

No significant differences in complications were found between the two groups, as shown in Table 3. Pericardial effusion occurred in 3 of 101 patients in the DG (3%) and 1 of 72 patients in the RG (1%). In the DG, one patient with cardiac tamponade required pericardiocentesis and received red cell suspension and apheresis platelet as blood products, which was performed successfully without any hemodynamic disturbance. The NOACs treatment was stopped in the aforementioned patient until no effusion on echocardiography. One patient in the DG developed limited lower gastrointestinal bleeding (1%) without nausea, vomiting, abdominal pain, and diarrhea. Abdominal enhanced computed tomography showed no obvious intestinal abnormalities. The symptoms disappeared on the third day after stopping oral anticoagulant and no blood transfusion was needed. Incidence of groin hematoma was no different between the DG and the RG (3% vs. 3%, P = 1.00). All the above patients did not need surgical intervention, and the perioperative anticoagulant treatment was uninterrupted. One thromboembolic event occurred in the DG, and the patient suffered a TIA on the procedural day. Brain computed tomography showed no evidence of bleeding. Magnetic resonance imaging excluded acute cerebral infarction.

Table  3.  Comparison of complications between the dabigatran group and the rivaroxaban group.

	Dabigatran group (n = 101)	Rivaroxaban group (n = 72)	P-value
Death	0	0	N/A
Bleeding complications
Cardiac tamponade	1 (1%)	1 (1%)	1.000
Pericardial effusion	3 (3%)	0	1.000
Intracranial hemorrhage	0	0	N/A
Gastrointestinal hemorrhage	1 (1%)	0	1.000
Hematomas of puncture site	3 (3%)	2 (3%)	1.000
Thromboembolic complications
Stroke	0	0	N/A
Transient ischemic attack	1 (1%)	0	1.000
Peripheral arterial embolism	0	0	N/A
Data are presented as n (%). N/A: not available.

 | Show Table

DownLoad: CSV

DISCUSSION

This research was a comparative study of the ACT values in response to heparin during AF ablation among Chinese patients using the two NOACs (dabigatran and rivaroxaban). The present study found that rivaroxaban was a better response to heparin administration compared with dabigatran with regard to ACT values, and made the intraprocedural ACT values more stable. But, no difference in the incidence of complications was found between the two coagulant agents.

Patients undergoing catheter ablation of AF have higher risk of thromboembolism during the perioperative period, and the incidence of thromboembolism was 0.9%–5%.^[9] The mechanism underlying thrombosis associated with ablation may be involved in as following: (1) the AF ablation energy led to left atrial endothelial injury and inflammatory reaction, which induced prethrombotic state and increased the risk of thrombosis; (2) abnormal blood constituents including inflammatory cytokines, growth factors released by damaged endothelial cells also promote hypercoagulable state during AF ablation; and (3) after AF conversion to sinus rhythm, the change of atrial blood flow can lead to prethrombotic state and induce thrombosis.^[10] Therefore, peri-ablation anticoagulant therapy is critical to prevent from the thromboembolic complications.

Perioperative uninterrupted anticoagulation with warfarin was associated with lower risk of periprocedural thromboembolic events after AF ablation.^[11–14] However, by virtue of having several advantages, including less interaction with other drugs and foods and no need for monitoring the international normalized ratio,^[15] the uninterrupted use of NOACs was increasingly applied to patients undergoing AF ablation. However, few comparative studies were reported on NOACs for intraprocedural anticoagulation. Our study showed that patients treated with dabigatran took longer to achieve the target ACT than those treated with rivaroxaban, and had lower mean ACT values, and less percentage of ACTs within the therapeutic range (250–350 s) than those taking rivaroxaban. These findings were consistent with the results of a Japanese study.^[16] In addition, we found that initial and total heparin requirements were significantly higher in patients on uninterrupted dabigatran compared to uninterrupted rivaroxaban. Dabigatran can bind to free and fibrin-bound thrombin to prevent thrombin-mediated effects, including fibrinogen splitting into fibrin and activation of factors V, VIII, XIII and XI. Unfractionated heparin is an indirect anticoagulant that can bind to and activate antithrombin (AT), which then binds to and inactivates coagulation factors such as factor Xa and clot-bound thrombin. Long-term exposure to exogenous thrombin inhibitors such as dabigatran may lead to down-regulation of endogenous thrombin inhibitor AT expression. This will result in the inhibition of AT level and may provide a reasonable mechanism explanation for the impairment of heparin sensitivity. The expression of prothrombin may also be compensated up-regulated, which may attenuate the effect of heparin. In addition, dabigatran may interfere with the effect of heparin by competing with heparin/AT complex to bind to thrombin.^[17] All these effects may cause the different in ACT to heparin response between dabigatran and rivaroxaban.

NOACs have specific properties, including its rapid onset and short half-lives. For example, the peak level of rivaroxaban occurred 2–4 h after oral administration, and had a half-life of 6–13 h depending on the dose and age.^[18] This variability in drug concentration might lead to a wide range of periprocedural ACT values. Considering NOACs onset time and half-lives, we evaluated the ACT values related to the administration time in patients receiving uninterrupted NOACs during the ablation procedure. In the present study, we demonstrated no differences in mean ACT between the morning and the afternoon sessions in each group, while the percentage of ACTs within the therapeutic range was higher in the morning session than in the afternoon session in the DG. A previous study has suggested that dabigatran has a flat dose-response curve at higher concentration. However, a low dose of dabigatran (110 mg twice daily) was more common in our study, which may contribute to the observed difference in the percentage of ACTs within the therapeutic range with the session start time in the DG.

Although the administration of heparin to maintain ACT in the range of 300–400 s is accepted as a routine approach in most centers,^[4] some studies reported that a shorter ACT of 210–225 s is equally safe in the ablation procedure. In the present study, ACT level of 250–350 s was targeted according to operator’s experience. However, the incidence of minor bleeding complications was relatively low and only one patient experienced the TIA. Additionally, though uninterrupted rivaroxaban therapy appeared to be more stable and efficient by measuring ACT to heparin response during AF ablation, the incidence of periprocedural bleeding and thromboembolic complications was no difference between the two groups. However, inconsistent findings showed that ACT could not be regarded as a reliable indicator of the intraprocedural anticoagulation state for uninterrupted NOAC therapy.^[17,19,20] Several reasons might explain the conflicting results among these studies. Firstly, the sensitivity to ACTs differed strongly among the different NOACs. Secondly, in vitro study confirmed that dabigatran interfered more with ACT than anti-Xa agents.^[21] Thirdly, pharmacokinetics may differ among different individuals treated with NOACs. Therefore, perioperative anticoagulant status guided by ACT indicators needed urgent research in the future, so as to more accurately prevent from thromboembolism and bleeding complications.

Our centers target a lower ACTs in the range of 250–350 s, then lower initial heparin doses might be sufficient. In our study, the initial dose of heparin was administrated as 100–120 U/kg for the patients with obesity, which was slightly lower in comparison with a recent report showing a larger initial heparin dose of 120–130 U/kg to achieve the target range of 250–350 s.^[21] Moreover, the complications of thrombotic events did not increase comparing with other similar studies.

LIMITATIONS

In our study, AF ablation was performed by the same operators and assistants, which avoided the result unbiased caused by surgical operation. However, there are still several limitations that must be noted. Firstly, the present study is a single center observational study. Secondly, despite our study included more patients treated with dabigatran than rivaroxaban, it reflected the true situation of anticoagulation therapy for AF in our hospital. Last but not least, this study included a relatively small sample size of patients, but the subjects were consecutive perioperative patients who received continuous anticoagulant therapy with rivaroxaban and dabigatran. Therefore, our results need to be confirmed in a prospective, randomized multicenter study with a large sample population.

CONCLUSIONS

In conclusion, the anticoagulant effect of uninterrupted rivaroxaban therapy appears to be more stable and efficient to heparin response during AF ablation evaluated by measuring ACT. Therefore, the anticoagulation status evaluated by ACT during the ablation may be an ideal approach for prevention from thromboembolism complications.