Author + information
- Received August 13, 2015
- Revision received October 26, 2015
- Accepted November 19, 2015
- Published online June 1, 2016.
- Tae-Seok Kim, MDa,
- Sung-Hwan Kim, MD, PhDb,∗∗ (, )
- Bo-Kyung Kim, RNb,
- Ju Youn Kim, MDb,
- Ji-Hoon Kim, MD, PhDc,
- Sung-Won Jang, MD, PhDd,
- Man Young Lee, MD, PhDe,
- Tai-Ho Rho, MD, PhDd and
- Yong-Seog Oh, MD, PhDb,∗ ()
- aDivision of Cardiology, Department of Internal Medicine, College of Medicine, The Catholic University of Korea, Daejeon St. Mary’s Hospital, Daejon, Republic of Korea
- bDivision of Cardiology, Department of Internal Medicine, College of Medicine, The Catholic University of Korea, Seoul St. Mary’s Hospital, Seoul, Republic of Korea
- cDivision of Cardiology, Department of Internal Medicine, College of Medicine, The Catholic University of Korea, St. Vincent’s Hospital, Suwon, Republic of Korea
- dDivision of Cardiology, Department of Internal Medicine, College of Medicine, The Catholic University of Korea, St. Paul’s Hospital, Seoul, Republic of Korea
- eDivision of Cardiology, Department of Internal Medicine, College of Medicine, The Catholic University of Korea, Yeouido St. Mary’s Hospital, Yeouido, Republic of Korea
- ↵∗Reprint requests and correspondence:
Dr. Yong-Seog Oh, Division of Cardiology, Department of Internal Medicine, College of Medicine, Seoul St. Mary’s Hospital, The Catholic University of Korea, Banpo-daero 222, Seocho-gu, Seoul 137-701, Republic of Korea.
- ↵∗∗Dr. Sung-Hwan Kim, Division of Cardiology, Department of Internal Medicine, College of Medicine, Seoul St. Mary’s Hospital, The Catholic University of Korea, Banpo-daero 222, Seocho-gu, Seoul 137-701, Republic of Korea.
Objectives This study tested the hypothesis that continuous heparin infusions would be favorable for maintaining heparin concentrations during radiofrequency catheter ablation (RFCA) of atrial fibrillation (AF).
Background Heparin infusions are essential for RFCA of AF. There is a paucity of data on the details for the optimal heparin infusion during RFCA of AF.
Methods A total of 333 patients undergoing AF ablation were consecutively enrolled and randomized to intermittent or continuous heparin infusion. A heparin bolus of 100 U/kg was injected just prior to transseptal puncture. The heparin concentration necessary to maintain an optimal activated clotting time (ACT) (300 to 400 s) was determined and checked every 30 min during the procedure. The primary endpoint of the study was the frequency of the maintenance of an optimal intraprocedural ACT.
Results The frequency of an optimal ACT in the continuous group was significantly higher than that in the intermittent group (64.0% vs. 57.6%, respectively, p < 0.01), whereas the total heparin level was significantly lower in the continuous group (13,162 ± 4,634 U vs. 15,837 ± 5,243 U, respectively, p < 0.01). The standard deviation of the ACT was significantly smaller in the continuous group than in the intermittent group (49 ± 30 vs. 33 ± 18, respectively, p < 0.01). Ninety-six patients had new oral anticoagulants (NOACs) before the procedure, and an optimal ACT at the first ACT check was less frequent than in patients taking warfarin (12.5% vs. 59.1%, respectively, p < 0.01). There were no significant differences in periprocedural bleeding or thromboembolic complications between the groups.
Conclusions During AF ablation, a continuous heparin infusion was superior to an intermittent heparin infusion for maintaining an optimal ACT range. (Randomized Comparison of Continuous and Intermittent Heparin Infusion During Catheter Ablation of Atrial Fibrillation [COHERE]; NCT01935557)
Radiofrequency catheter ablation (RFCA) of atrial fibrillation (AF) is an effective therapeutic option for the treatment of symptomatic, drug-refractory AF. RFCA of AF is technically challenging and associated with the risk of periprocedural complications, including thromboembolism. The incidence of periprocedural thromboembolic events is reported to be as high as 0.5% to 4.0%, despite adequate periprocedural anticoagulation and preprocedural imaging studies (1–3). Moreover, intracardiac echocardiography demonstrated a 10.3% thrombus formation in the left atrium (LA) during RFCA of AF (4). Therefore, it is important to maintain optimal anticoagulation during the procedure, with close attention paid to the therapeutic target dosage. According to current guidelines, unfractionated heparin (UFH) should be administered prior to or immediately following transseptal puncture during AF ablation procedures and should be adjusted to maintain a target activated clotting time (ACT) of 300 to 400 s (5). The consensus of the writing group was that ACT should be checked at 10- to 15-min intervals until therapeutic anticoagulation is achieved and then at 15- to 30-min intervals for the duration of the procedure (5). Current guidelines recommend that heparin dose should be adjusted to maintain an ACT of at least 300 to 350 s throughout the procedure, but data for controlling intraprocedural ACTs are lacking. UFH has a half-life of approximately 1 to 2 h post-infusion, and lower doses of UFH have a much shorter half-life (6). The apparent biological half-life of UFH increases from 30 min after an intravenous bolus of 25 U/kg to 60 min after an intravenous bolus of 100 U/kg (6). The pharmacokinetics of UFH make it difficult to achieve an acceptable intraprocedural ACT with minimal fluctuations. Therefore, we hypothesized that a continuous heparin infusion would be favorable for the maintenance of heparin concentration during RFCA of AF. The main purpose of this study was to compare intermittent heparin infusions with continuous heparin infusions for maintaining an optimal ACT during RFCA of AF.
The COHERE (COntinuous HEparin infusion REferring to ablation of atrial fibrillation) trial was a prospective, randomized, open-label, single-blind, controlled trial (NCT01935557) conducted to assess the role of a continuous heparin infusion for maintenance of an optimal ACT during RFCA of AF. The study protocol was approved by the Institutional Review Board, and all patients enrolled in the study provided written informed consent. Inclusion criteria were ≥18 years of age and patients undergoing AF ablation. Patients who refused to be enrolled in the study and who had absolute contraindications to anticoagulation therapy were excluded. Patients were randomized in a 1:1 manner to the intermittent heparin infusion group or continuous heparin infusion group, using sealed envelopes containing a computer-generated randomization sequence. We identified 339 patients who were considered potentially eligible for participation in this study. All subjects underwent transthoracic echocardiography, transesophageal echocardiography, and cardiac computed tomography (CT) in preparation for the ablation procedure.
The primary endpoint of the study was frequency of the maintenance of an optimal intraprocedural ACT. An optimal ACT was defined as the target ACT of 300 to 400 s. Secondary endpoints included establishing the average standard deviation (SD) of ACT. Safety endpoints were major or minor bleeding and thromboembolic complications during the periprocedural period. Major bleeding was defined as the occurrence of cardiac tamponade or a hemopericardium requiring intervention, causing symptoms, or requiring a transfusion; hematomas requiring intervention; massive hemoptysis; hemothorax; and retroperitoneal bleeding. Minor bleeding complications were defined as the occurrence of a hematoma or any bleeding that did not require any intervention or did not cause any symptoms (7). Stroke was defined as the onset of a new neurological deficit that occurred anytime during or within 48 h of the procedure (8). A transient ischemic attack (TIA) was defined as the occurrence of a neurological deficit lasting <24 h (8).
All enrolled patients were treated with anticoagulation therapy with warfarin or new oral anticoagulants (NOACs) for at least 4 weeks before the ablation procedure. Outpatient monitoring of the international normalized ratio (INR) was performed at least 1 to 2 weeks prior to the procedure to ensure a therapeutic level of anticoagulation (INR between 2 and 3). Warfarin therapy was uninterrupted and was administered the night of the procedure according to the patient’s scheduled dose. Those taking NOACs held 2 doses (in the case of dabigatran and apixaban) or 1 dose (in the case of ribaroxaban) immediately prior to the procedure, and the NOAC administration was resumed on the evening of the procedure. An intravenous UFH bolus of 100 U/kg was administrated just prior to transseptal puncture regardless of the group (loading dose). The dose of heparin was determined in order to maintain an optimal ACT (300 to 400 s) and was checked every 30 min during the procedure. ACTs were measured with a commercially available analyzer (Hemochron Response, ITC, Edison, New Jersey). After the first ACT measurement, subsequent UFH boluses were administered every 30 min to maintain an optimal ACT in the intermittent infusion group, and UFH dosage was determined based on previous experience. In the continuous infusion group, the initial maintenance UFH dosage was administrated continuously at approximately 25 to 100 U/kg/h after the initial ACT measurement, and the infusion rate was subsequently adjusted based on the ACT measurement (Figure 1). Usually, additional heparin boluses were given in doses of approximately 1,000 to 4,000 U in the intermittent group, and adjustments to the infusion rate were either up or down by 500 to 2,000 U/h in the continuous group. All data for the timing and dose of the heparin and ACT measurements were collected prospectively on an anonymized data sheet.
Details of pulmonary vein (PV) isolation performed in the present study are as follows: intracardiac electrograms were recorded using the Prucka Cardio Lab electrophysiology system (General Electric Medical Systems, Milwaukee, Wisconsin). A duodecapolar catheter (Livewire, St. Jude Medical, Minneapolis, Minnesota) was inserted into the left femoral vein to map the right atrium (RA) and coronary sinus, and a His bundle-right ventricle (RV) fixed catheter (St. Jude Medical) was advanced and placed in the His-RV region. A double transseptal puncture approach was taken to access the LA, and biplane pulmonary venograms were obtained. Subsequently, a long sheath (Schwartz left 1; St. Jude Medical) circumferential PV mapping catheter (Inquiry Optima, St. Jude Medical) was also inserted into the LA. A 3-dimensional (3D) electroanatomical map (NavX, St. Jude Medical) was generated by merging the NavX system-generated 3D geometry of the LA and PVs with the corresponding 3D spiral CT images. Circumferential ablation lines for the PV isolation (PVI) were created around the left- and right-sided ipsilateral PVs by using a 4-mm irrigated tip catheter (Therapy Cool Flex, St. Jude Medical). Only a PVI with bidirectional blockages was performed in patients with paroxysmal AF. In patients with persistent and long-standing persistent AF, additional substrate modification was also performed at the operator’s discretion. In selected patients, the cavotricuspid isthmus was ablated with confirmation of bidirectional conduction block. Heparin was reversed with intravenous protamine sulfate (25 to 50 mg), and the catheters/sheaths were removed at the end of the procedure.
We estimated that a sample of 280 patients would provide 80% power to detect a significant difference with regard to the primary endpoint at a 2-sided significance level of 0.05, assuming that the maintenance of optimal ACT in the intermittent group would be 60% and in continuous group 70%. These rates were based on our preliminary data. Moreover, considering a dropout rate of approximately 15% for randomized patients, a total of 322 patients (161 per treatment group) needed to be randomized to achieve the required number of patients for the efficacy analysis. Continuous variables were compared using an unpaired Student t test or Wilcoxon rank sum model where necessary, and categorical variables were compared using either the chi-square test or Fisher exact test as appropriate. Baseline demographics, heparin dosages, ACT profiles, and periprocedural complications in the intermittent group were compared to those in the continuous infusion group. A p value of <0.05 (2-sided) was considered statistically significant. The SD was calculated to evaluate the variability of the ACTs measured. All statistical analyses were performed with SPSS version 18.0 software (SPSS, Inc., Chicago, Illinois) for Windows (Microsoft, Redmond, Washington).
A total of 333 patients were included in the study between July 2012 and June 2015. The flow chart of the patients through the study is presented in Figure 1. Mean age of the study population was age 61 ± 11 years, with 227 males (68.2%) males and 174 patients (52.3%) who had paroxysmal AF with no differences between the groups as shown in Table 1. There were no differences in mean body surface areas, prevalence of advanced age (≥75 years), hypertension, diabetes, prior stroke and/or TIA, LA sizes and volumes, and left ventricular ejection fractions between the groups. The mean CHA2DS2-VASc score (1.84 ± 1.42) did not differ between the 2 groups. NOACs were administered to 96 patients, and 51 and 45 patients were allocated to the intermittent group and continuous group, respectively. Before the ablation procedure, the INR values in the patients receiving warfarin were the same in both groups (2.27 ± 0.41 and 2.29 ± 0.38, respectively).
Intraprocedural ACT and heparin requirements
Figures 2 and 3 show summaries of the intraprocedural ACT profiles and heparin requirements of the patients grouped by heparin infusion methods. ACT measurements were made 787 times (4.7 times per patient) in the intermittent group and 776 times (4.7 times per patient) in the continuous group. The frequency of maintaining an optimal ACT was 453 of 787 (57.6%) in the intermittent group and 497 of 776 (64.0%) in the continuous group (p < 0.01) (Figure 2). The total dose of heparin was less in the continuous group than in the intermittent group. There were no differences in heparin infusion times between the 2 groups. Although the mean ACT was significantly prolonged in the intermittent group compared with that in the continuous group, the mean ACTs were within the optimal range in both groups. The SD of ACT in the continuous group was significantly smaller than that in the intermittent group, which indicated that the ACTs measured were clustered closely around the mean and showed a minimal fluctuation (Figure 3D). Table 2 shows the intraprocedural ACTs between the warfarin and NOAC groups. The mean duration of achieving the optimal ACT after loading with UFH was 88.1 ± 33.1 min in 96 patients taking NOACs and 51.4 ± 31.8 min (p < 0.01) in those taking warfarin. In addition, only 12 patients (12.5%) in the NOAC group showed optimal ACT at the initial ACT check (140 of the 237 patients [59.1%] in the warfarin group, p < 0.01).
Bleeding and thrombosis events
A total of 19 patients demonstrated bleeding complications (Table 3). There were no thromboembolic complications in either group. Five patients in the continuous infusion group and 4 in the intermittent group underwent an emergent percutaneous pericardial drainage without any sequelae. None of the patients required surgical drainage. Six patients presented a groin hematoma without a pseudoaneurysm or arteriovenous fistula. Four patients exhibited self-limiting gross hematuria during AF ablation. There were no significant differences between complications in the intermittent group and those in the continuous infusion group. No clinical strokes were observed during the follow-up period (402 ± 284 days).
This is the first randomized study to show that a continuous heparin infusion method is more useful for maintaining an optimal ACT range with a lower dose of heparin and smaller ACT fluctuations than the intermittent heparin infusion method, during RFCA of AF. The main findings of the study were as follows: 1) the frequency of an optimal ACT during the RFCA of AF was only approximately 58% in the intermittent group; 2) the frequency of an optimal ACT increased to 64% with a lower dosage of total heparin in the continuous group; 3) there was minimal ACT fluctuation in continuous infusion group, which meant that the ACT levels were more stable during the RFCA of AF with the continuous heparin infusion method than the intermittent heparin bolus infusion method; and 4) the patients with preprocedual NOACs had a tendancy to have delayed and lower levels of anticoagulation as measured by the intraprocedural ACT.
Pharmacokinetics of UFH and intraprocedural ACT levels
The half-life of UFH is dose-dependent. The half-lives of 25, 100, and 400 U/kg UFH are 30, 60, and 150 min, respectively. At lower doses, the half-life is very short because most of the infused heparin binds directly to macrophages and endothelial cells, where it gets depolymerized. This mechanism becomes saturated at higher doses, and the elimination becomes slower in a dose-dependent manner (6). During RFCA of AF, typically, 25 to 100 U/kg UFH are required per 30 min, leading to the short half-life of infused heparin. In theory, a continuous administration could be associated with less fluctuation in drug plasma concentrations. Therefore, a continuous infusion can provide more constant ACT values during the RFCA of AF, resulting in a safe and stable procedure. Currently, the intraprocedural use of UFH to maintain the ACT within a range of 300 to 400 s is a routine approach in almost all laboratories regardless of the periprocedural anticoagulation (5). Our study showed favorable intraprocedural ACT values with minimal fluctuation, represented by a significantly smaller SD in the continuous infusion group than in the intermittent infusion group. Moreover, the continuous infusion has an advantage of discontinuation of the administration. If the ACT levels are >400 s, it is possible to stop the UFH administration immediately and to resume the administration as required. However, it is difficult to adjust the dose of heparin and to decide the timing of a restart of an UFH bolus injection in the case of a heparin overdose. We suggest a new heparin nomogram during RFCA of AF (Table 4), which was based on our study data, to adjust the intraprocedural ACT in continuous infusion group in the latter half of the enrolled patients.
NOACs and intraprocedural ACT levels
Another emerging anticoagulation strategy involves the use of NOACs, including the direct thrombin inhibitor dabigatran and the factor Xa inhibitors rivaroxaban and apixaban, all of which are approved for systemic anticoagulation in patients with nonvalvular AF. Recently, there have been various studies of the feasibility and safety of NOACs for periprocedural anticoagulation.(9–12) Most of these studies focused on periprocedural bleeding and thromboembolic complications, and each NOAC has shown safe and efficacious results for the RFCA of AF. However, there are limited data about intraprocedural UFH adjustments in patients taking NOACs. Konduru et al. (13) reported that a conventional intraprocedural heparin administration resulted in delayed and lower ACT levels in patients treated with dabigatran than those in patients taking uninterrupted warfarin. Our study also showed that the mean duration for achieving an optimal ACT after the UFH loading was significantly longer in the NOAC group than in the warfarin group (88.1 ± 33.1 min vs. 51.4 ± 31.8 min, respectively, p < 0.01). The percentage of patients reaching the optimal ACT value by the first ACT check after loading with heparin was just 12.5% in the patients taking NOACs (59.1% in the warfarin group). The mechanism of this phenomenon is unknown but is suggested to be the down-regulation of the expression of the endogenous thrombin inhibitor antithrombin (AT), resulting from prolonged exposure to an exogenous thrombin inhibitor such as dabigatran (13). Decreased AT levels may develop an impaired sensitivity to heparin. Seen differently, it could be warfarin’s effect on the ACT. Chang et al. (14) reported a linear increase in ACT in patients taking warfarin. Hamam et al. (15) demonstrated baseline INR as predictor of achieving a first ACT of ≥300 s. Further investigations are warranted to elucidate the mechanism of the decreased responsiveness to UFH and to establish guideline for the UFH infusion method in patients taking NOACs.
Recently, there has been growing interest in post-ablation silent cerebral ischemia (SCI), which is defined as an acute new magnetic resonance imaging (MRI)-detected cerebral ischemic lesion in a patient without any apparent neurological deficit. Several studies (16–18) have reported an SCI incidence of up to 50% after RFCA of AF. One study showed a significant association between just 1 single measurement of a periprocedural ACT of <300 s during a procedure and a 3-fold increased risk for SCI (19). Although SCI was not evaluated in our study, there was the possibility of SCI despite proper intraprocedural anticoagulation. However, an intraprocedural heparin administration with steady ACT levels may reduce the risk of SCI. Further study is necessary to evaluate the relationship between steady ACT levels and SCI.
We did not check the intraprocedural ACT more frequently (per 15 min) until a therapeutic ACT was obtained. The operators were not blinded to intraprocedural anticoagulation management and the dose, frequency, and timing of the UFH boluses were left to the clinical discretion of the operator. Our study did not include the pre- and post-procedural brain MRI findings for detecting of SCI. And our nomogram (Table 4) was not validated.
The continuous infusion method was useful for maintaining an optimal ACT range with a lower dose of heparin and minimal ACT fluctuations during the RFCA of AF. Patients with preprocedural NOACs had a tendancy to have delayed and lower levels of anticoagulation as measured by the intraprocedural ACT. Future larger, randomized trials are warranted to confirm these findings.
COMPETENCY IN MEDICAL KNOWLEDGE: Adequate periprocedural anticoagulation is essential for prevention of thromboembolisms during catheter ablation of AF. However, there is a paucity of data on the optimal anticoagulation method using heparin during catheter ablation of AF.
TRANSLATIONAL OUTLOOK: This study demonstrated the efficacy of continuous heparin infusion in maintaining an optimal ACT during catheter ablation of AF. Further studies will be needed to evaluate the association between post-ablation silent cerebral ischemia and this method.
The authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- activated clotting time
- atrial fibrillation
- left atrium
- new oral anticoagulant
- pulmonary vein
- pulmonary vein isolation
- right atrium
- radiofrequency catheter ablation
- silent cerebral ischemia
- transient ischemic attack
- unfractionated heparin
- Received August 13, 2015.
- Revision received October 26, 2015.
- Accepted November 19, 2015.
- American College of Cardiology Foundation
- Ren J.F.,
- Marchlinski F.E.,
- Callans D.J.
- Calkins H.,
- Kuck K.H.,
- Cappato R.,
- et al.
- Hirsh J.,
- Anand S.S.,
- Halperin J.L.,
- Fuster V.
- Mehran R.,
- Rao S.V.,
- Bhatt D.L.,
- et al.
- Di Biase L.,
- Burkhardt J.D.,
- Santangeli P.,
- et al.
- Lakkireddy D.,
- Reddy Y.M.,
- Di Biase L.,
- et al.
- Bassiouny M.,
- Saliba W.,
- Rickard J.,
- et al.
- Winkle R.A.,
- Mead R.H.,
- Engel G.,
- Kong M.H.,
- Patrawala R.A.
- Konduru S.V.,
- Cheema A.A.,
- Jones P.,
- Li Y.,
- Ramza B.,
- Wimmer A.P.
- Hamam I.,
- Daoud E.G.,
- Zhang J.,
- et al.
- Herrera Siklody C.,
- Deneke T.,
- Hocini M.,
- et al.
- Gaita F.,
- Caponi D.,
- Pianelli M.,
- et al.
- Di Biase L.,
- Gaita F.,
- Toso E.,
- et al.