2022 Volume 86 Issue 2 Pages 280-286
Background: The effects of catheter ablation (CA) energy sources on myocardial injury and coagulation biomarkers among Japanese non-valvular atrial fibrillation patients receiving uninterrupted periprocedural edoxaban are unclear. This KYU-RABLE exploratory subanalysis compared the effects of CA using radiofrequency energy vs. cryoballoon on: (1) myocardial injury; and (2) plasma edoxaban and coagulation biomarker concentrations measured before and after CA.
Methods and Results: Plasma creatine kinase (CK), edoxaban, D-dimer, and prothrombin fragment 1+2 (F1+2) concentrations within 1 h before CA were compared with concentrations the day after. All biomarkers increased after CA, regardless of the energy source, but especially with cryoballoon. Significantly higher increases in CK concentrations from before to the day after CA were seen with cryoballoon compared with radiofrequency energy (P<0.0001). Edoxaban concentrations were similar in both groups. Concentrations of D-dimer and F1+2 increased in both groups, but were significantly higher in the cryoballoon group (P<0.0001 and P=0.006, respectively). There were no significant between-group differences in the incidence of thrombotic or bleeding events.
Conclusions: Uninterrupted edoxaban concentrations were similar in both groups. Both myocardial injury and coagulation biomarkers increased after CA, especially with cryoballoon, but there was no difference in the incidence of thrombotic or bleeding events. These findings suggest the efficacy of uninterrupted edoxaban, regardless of the CA energy source. Periprocedural anticoagulation, particularly with cryoballoon, should be undertaken with care.
Catheter ablation (CA) is an established treatment of non-valvular atrial fibrillation (NVAF),1 with the isolation of pulmonary veins (PV) as an essential component of the process. Different technologies have been developed for CA, including hyperthermic energy sources (e.g., radiofrequency [RF], microwave, laser, or ultrasound) and hypothermic energy sources (e.g., cryoablation). The most used approach is point-by-point RF ablation. Cryoballoon (CB) ablation is an emerging technique with several benefits over RF ablation: CB ablation requires a shorter procedure time and has an extremely low risk of cardiac perforation. Thus, CB ablation tends to be indicated for elderly patients, patients with heart failure, patients with higher bleeding risk, or those undergoing a first-line ablation.2
Editorial p 287
Because of the myocardial necrosis induced by the CA procedure to achieve PV isolation, several biomarkers of endothelial damage, myocardial injury, inflammatory response, and thrombogenesis are released.3–5 A recent study reported that concentrations of troponin I, a biomarker of myocardial injury, were higher after CB than RF ablation. Among patients who underwent CB ablation, high troponin I concentrations were correlated with lower recurrence of atrial fibrillation (AF).5 Further, CA with RF energy seems to be associated with a higher risk of thrombus formation than CB ablation.6 Some studies have found no differences in biomarkers of myocardial injury (i.e., creatine kinase [CK]) and coagulation (i.e., prothrombin fragment 1+2 [F1+2], D-dimer, soluble fibrin monomer complex, and thrombin-antithrombin complex) between CB and RF ablation.3–5 In patients with paroxysmal AF, CB ablation seems to cause more significant myocardial injury (higher troponin I and CK concentrations),3,4 but less endothelial damage than RF ablation.4 However, such changes in biomarkers have not been well studied in patients undergoing CA while receiving uninterrupted periprocedural direct oral anticoagulants (DOACs).
Uninterrupted DOACs have been reported to be effective and safe in preventing periprocedural thromboembolism and major bleeding compared with vitamin K antagonists.7 It has also been reported that there are some differences in the rates and types of events between uninterrupted and interrupted DOAC administration.8 In the ELIMINATE-AF (a Prospective, Randomized, Open-Label, Blinded Endpoint Evaluation Parallel Group Study Comparing Edoxaban versus VKA in Subjects Undergoing Catheter Ablation of Non-valvular Atrial Fibrillation) trial, uninterrupted edoxaban resulted in a lower incidence of the composite endpoint of all-cause death, stroke, or major bleeding compared with vitamin K antagonists.9
The KYU-RABLE (multicenter study associated with KYU-shu to evaluate the efficacy and safety of edoxaban in patients with non-valvulaR Atrial fiBriLlation undergoing cathEter ablation) study investigated the efficacy and safety of uninterrupted periprocedural edoxaban administered for NVAF patients undergoing CA in Japan, and showed that the coagulation status remained unchanged despite delaying 1 dose of edoxaban until after CA.10 Further, it was shown that the coagulation status remained unchanged regardless of the CHADS2 score or AF type.11 The number of events reported in the KYU-RABLE study was low; 1 event of cardiac tamponade (major bleeding event) was observed, and there were no events of stroke, systemic embolism, or death.10 Given the low number of events, we hypothesized that differences in clinical events due to different CA energy sources could not be determined. Nonetheless, biomarker concentrations may differ depending on the CA energy source, and the effect of the CA energy source on levels of myocardial injury and coagulation biomarkers among Japanese NVAF patients receiving uninterrupted periprocedural edoxaban remains unclear. Thus, the present exploratory subanalysis of the KYU-RABLE study aimed to determine the effects of CA using RF as energy compared with CB ablation on: (1) myocardial injury; and (2) plasma concentrations of edoxaban and coagulation biomarkers measured immediately before CA and the day after CA.
Details regarding the study design and the eligibility criteria for the KYU-RABLE study have been published elsewhere.10 Briefly, KYU-RABLE was a prospective multicenter single-arm interventional study conducted from December 2017 to September 2018. Patients aged ≥20 years who were scheduled to undergo CA for AF were recruited at 23 institutions in Japan.
The main exclusion criteria were contraindications for treatment with edoxaban or contraindications for the procedure, impaired renal function, and a history of thromboembolism, myocardial infarction, or minor or major bleeding. The present subanalysis focused on patients who underwent either RF or CB ablation. Those who underwent both procedures were excluded.
Investigators at each site selected the type of CA procedure and energy source, as well as the use of heparin and the dosage to maintain an activated clotting time >300 s during the procedure. The study treatment consisted of edoxaban 60 mg (a reduced dose of 30 mg was given if patients met the dose adjustment criteria) administered once daily in the morning for ≥4 weeks before CA. On the day of CA, edoxaban was administered after confirmation of hemostasis following sheath removal and not before CA. On the day after CA, edoxaban was administered at least 12 h after the dose given on the day of CA. Thereafter, edoxaban was administered in the morning for 4 weeks (±7 days).10
The study protocol and associated documents were approved by the Ethics Committee of Oita University Faculty of Medicine (Approval no. B17-021, December 12, 2017). The study conformed to the Declaration of Helsinki and Good Clinical Practice Guidelines. All patients provided written informed consent. The study was registered in the University Hospital Medical Information Network (UMIN) Clinical Trials Registry under the identifier UMIN000029693.
Study OutcomesThe primary endpoint of the KYU-RABLE study was a composite of stroke/systemic embolic event and major bleeding. Secondary endpoints were thromboembolism, cardiovascular events, major bleeding, clinically relevant non-major bleeding, and minor bleeding. In this subanalysis, we estimated changes in the myocardial injury marker CK, edoxaban concentrations, and coagulation biomarkers (D-dimer and prothrombin F1+2) between the day of CA (1 h before CA) and the day after CA.
Measurement MethodsAs described previously,10 solid-phase extraction using a liquid chromatography-tandem mass spectrometry system (SCIEX, Framingham, MA, USA) was used to measure plasma edoxaban concentrations. Assays were conducted at Shin Nippon Biomedical Laboratories (Wakayama, Japan). D-dimer levels were measured using a latex immunoturbidimetric assay using LATECLE D-dimer reagent (KAINOS Laboratories, Tokyo, Japan). F1+2 was measured by ELISA using the Enzygnost F1+2 monoclonal antibody (Siemens Healthineers, Erlangen, Germany), conducted at SRL Medisearch (Tokyo, Japan). CK concentrations were measured separately at each facility and were not centralized.
Statistical AnalysisClinical events were analyzed in the full analysis set, and the plasma concentrations of edoxaban and coagulation biomarkers were analyzed in the per-protocol set. Student’s t-test was used for continuous values, a Chi-squared-test was used for categorical values, and a Mann-Whitney test was used to compare the pharmacokinetic and pharmacodynamic values. The significance level was 5% (2-tailed) and 95% confidence intervals (CIs) were calculated. Changes in plasma concentrations of edoxaban and coagulation biomarkers against the time from the last edoxaban administration to the CA procedure were calculated and verified using the Mann-Whitney test (2-tailed). All statistical analyses were performed using SAS version 9.4 (SAS Institute, Cary, NC, USA).
Of the 477 patients in the per-protocol set, 356 (74.6%) underwent RF ablation and 104 (21.8%) underwent CB ablation. The remaining 17 patients were excluded from this analysis because they underwent CA with a hot balloon (n=5), laser balloon (n=8), and both RF and CB (n=4).
Among patients who underwent RF ablation, the mean age was 64.2 years, 71.3% were male, and the mean weight was 67.1 kg. Among patients who underwent CB ablation, the mean age was 66.4 years, 68.3% were male, and the mean weight was 65.7 kg. Compared with the RF group, in the CB group significantly more patients had a HAS-BLED score ≥3 (2.4% vs. 7.3%, respectively; P=0.0208), significantly more patients had paroxysmal AF (nearly 2-fold difference; P<0.0001), and a significantly lower proportion of patients had congestive heart failure (9.3% vs. 2.9%, respectively; P=0.0329; Table 1).
| RF | CB | P value | |
|---|---|---|---|
| n (%) | 356 (77.4)A | 104 (22.6)A | – |
| Age (years) | 64.2±10.2 | 66.4±8.4 | 0.0497 |
| Male sex | 254 (71.3) | 71 (68.3) | 0.5441 |
| Weight (kg) | 67.1±13.5 | 65.7±12.9 | 0.3624 |
| BMI (kg/m2) | 24.6±4.1 | 24.4±3.9 | 0.8049 |
| CrCl (mL/min) | 82.5±26.1 | 82.2±26.6 | 0.8941 |
| CHADS2 score | 1.2±0.9 | 1.1±0.9 | 0.2802 |
| ≥2 | 105 (29.5) | 25 (24.0) | 0.2770 |
| CHA2DS2-VASc score | 2.0±1.3 | 2.0±1.3 | 0.9437 |
| ≥2 | 219 (61.5) | 66 (63.5) | 0.7193 |
| HAS-BLED score | 1.2±0.7B | 1.4±0.7C | 0.0633 |
| ≥3 | 8 (2.4)B | 7 (7.3)C | 0.0208 |
| Type of AF | |||
| Paroxysmal | 176 (49.4) | 88 (84.6) | <0.0001 |
| Persistent | 130 (36.5) | 15 (14.4) | |
| Long-standing persistent + permanent | 50 (14.0) | 1 (1.0) | |
| ComorbiditiesD | |||
| Hypertension | 212 (59.6) | 55 (52.9) | 0.2256 |
| Hyperlipidemia | 112 (31.5) | 33 (31.7) | 0.9584 |
| Diabetes | 52 (14.6) | 18 (17.3) | 0.4999 |
| Arrhythmia other than AF | 39 (11.0) | 9 (8.7) | 0.4995 |
| Congestive heart failure | 33 (9.3) | 3 (2.9) | 0.0329 |
| Ulcer or reflux esophagitis | 21 (5.9) | 8 (7.7) | 0.5080 |
| Angina pectoris | 18 (5.1) | 8 (7.7) | 0.3058 |
| Renal disease | 17 (4.8) | 7 (6.7) | 0.4302 |
| Treatment of AF | 220 (61.8) | 67 (64.4) | 0.6268 |
| Rate control | 139 (39.0) | 33 (31.7) | 0.1750 |
| Rhythm control | 122 (34.3) | 46 (44.2) | 0.0635 |
| Previous treatment of ablation | 67 (18.9)E | 4 (3.8) | 0.0002 |
| The starting time of CA (AM/PM) | |||
| AM | 205 (57.6) | 70 (67.3) | 0.0752 |
| PM | 151 (42.4) | 34 (32.7) | |
| Total procedure duration (min) | 179.0±65.9 | 146.3±63.3 | <0.0001 |
| Mean total heparin dose (units) | 9,564.8±5,086.1E | 8,483.2±3,948.8F | 0.0471 |
| Activated clotting time (s) | 357.1±74.6G | 334.5±58.6H | 0.0053 |
| Time from CA procedure to restarting edoxaban (h) | 6.2±4.1I | 7.0±3.5F | 0.0668 |
| Time from last edoxaban dose before CA to restarting edoxaban after CA (h) |
36.3±4.6I | 35.8±4.4F | 0.3394 |
Unless stated otherwise, data are presented as the mean±SD. A460 patients were used as a denominator. Bn=335. Cn=96. DComplications with a proportion of ≥5% in the entire study population are listed. En=355. Fn=103. Gn=344. Hn=102. In=354. AF, atrial fibrillation; AM, morning; BMI, body mass index; CA, catheter ablation; CB, cryoballoon; CrCl, creatinine clearance; PM, afternoon; RF, radiofrequency.
The mean time after CA to restarting edoxaban (6.2 vs. 7.0 h in the RF and CB groups, respectively) and the mean time from the last edoxaban administration before CA to when edoxaban was restarted after CA (36.3 vs. 35.8 h in the RF and CB groups, respectively) were similar in the 2 groups. Significantly more patients in the RF group had undergone a previous ablation treatment than in the CB group (18.9% vs. 3.8%, respectively; P=0.0002). Regarding current CA treatment, the total procedure time was significantly longer in the RF than CB group (mean±SD 179.0±65.9 vs. 146.3±63.3 min, respectively; P<0.0001). In addition, the RF group had a significantly higher mean total heparin dose (9,564.8 vs. 8,483.2 units; P=0.0471) and a significantly greater activated clotting time (357.1 vs. 334.5 s; P=0.0053) than the CB group (Table 1).
Study OutcomesThe primary results of the KYU-RABLE study have been reported elsewhere.10 Briefly, no patients undergoing RF or CB ablation presented with thromboembolism, and no deaths occurred (Table 2). Bleeding events were numerically more predominant in the RF than CB group.
| RF (n=380) |
CB (n=112) |
|
|---|---|---|
| Primary endpoint | ||
| Composite of stroke/SEE and major bleeding | 1 (0.3) | 0 (0.0) |
| Secondary endpointA | ||
| Thromboembolism | 0 (0.0) | 0 (0.0) |
| Cardiovascular event | 0 (0.0) | 1 (0.9) |
| Major bleedingB | 1 (0.3) | 0 (0.0) |
| Clinically relevant non-major bleeding | 6 (1.6) | 1 (0.9) |
| Minor bleeding | 9 (2.4) | 1 (0.9) |
Data are presented as n (%). AThere were no deaths or stroke/systemic embolic events (SEE). BCardiac tamponade. CA, catheter ablation; CB, cryoballoon; RF, radiofrequency.
In the present exploratory subanalysis, no significant differences were noted for edoxaban, CK, and D-dimer plasma concentrations immediately before CA between the RF and CB groups. Compared with the edoxaban concentration immediately before CA, the plasma concentration of edoxaban was higher on the day after CA in both groups; however, there were no significant differences between the RF and CB groups (Table 3).
| Edoxaban (ng/mL) | CK (IU/L) | D-dimer (μg/mL) | F1+2 (pmol/L) | |||||
|---|---|---|---|---|---|---|---|---|
| RF | CB | RF | CB | RF | CB | RF | CB | |
| Immediately before CA | ||||||||
| n | 356 | 104 | 279 | 68 | 356 | 104 | 356 | 104 |
| Mean±SD | 14.6±11.5 | 16.5±21.0 | 87.3±60.5 | 141.8±387.0 | 0.26±0.44 | 0.21±0.16 | 153.4±111.1 | 192.7±188.2 |
| Median | 12.1 | 11.2 | 72.0 | 78.0 | 0.18 | 0.18 | 128.5 | 137.0 |
| Range | 1, 138 | 3, 175 | 22, 674 | 30, 3,247 | 0.11, 8.00 | 0.11, 1.39 | 54, 1,050 | 62, 1,190 |
| IQR | 7.3, 18.7 | 8.2, 17.3 | 54.0, 104.0 | 58.0, 116.5 | 0.12, 0.30 | 0.13, 0.24 | 99.0, 165.0 | 113.5, 205.5 |
| P value | 0.915 | 0.125 | 0.404 | 0.014 | ||||
| Day after CA | ||||||||
| n | 316 | 95 | 349 | 98 | 316 | 95 | 316 | 95 |
| Mean±SD | 81.5±52.7 | 87.4±58.9 | 145.4±159.5 | 383.5±240.1 | 0.37±0.54 | 0.46±0.27 | 178.2±123.7 | 194.6±106.6 |
| Median | 70.6 | 71.0 | 120.0 | 337.5 | 0.30 | 0.39 | 156.0 | 175.0 |
| Range | 1, 310 | 5, 421 | 44, 2,496 | 135, 2,238 | 0.11, 9.20 | 0.11, 1.54 | 47, 1,190 | 86, 984 |
| IQR | 44.2, 111.6 | 53.6, 113.3 | 89.0, 164.0 | 262.0, 444.0 | 0.20, 0.44 | 0.26, 0.63 | 125.5, 194.0 | 142.0, 215.0 |
| P value | 0.399 | <0.0001 | <0.0001 | 0.006 | ||||
| Differences in plasma concentrations from immediately before to the day after CA | ||||||||
| n | 316 | 95 | 277 | 67 | 316 | 95 | 316 | 95 |
| Mean±SD | 66.8±51.8 | 72.2±59.9 | 62.8±162.9 | 235.9±197.4 | 0.13±0.15 | 0.25±0.26 | 29.6±129.6 | 24.9±89.4 |
| Median | 57.9 | 57.6 | 39.0 | 218.0 | 0.10 | 0.17 | 24.0 | 33.0 |
| Range | −31, 297 | −80, 405 | −181, 2,432 | −1,009, 598 | −0.24, 1.20 | −0.37, 1.43 | −799, 1,020 | −410, 320 |
| IQR | 30.4, 95.4 | 35.4, 96.1 | 17.0, 74.0 | 160.0, 333.0 | 0.03, 0.20 | 0.10, 0.34 | −4.0, 57.0 | 9.0, 67.0 |
| P value | 0.420 | <0.0001 | <0.0001 | 0.093 | ||||
The Mann-Whitney test was used for continuous variables. CA, catheter ablation; CB, cryoballoon; CK, creatine kinase; F1+2, prothrombin fragment 1+2; RF, radiofrequency.
CK concentrations increased on the day after CA in both the RF and CB groups (P<0.0001; data not shown), and the CK concentrations in the CB group were significantly higher than those in the RF group (P<0.0001; Figure). Increases in mean CK plasma concentrations from immediately before CA to the day after CA were 3-fold greater in the CB than RF group (mean±SD 235.9±197.4 vs. 62.8±162.9 IU/L, respectively; P<0.0001; Table 3).
Plasma concentrations of (A) edoxaban, (B) creatine kinase (CK), (C) D-dimer, and (D) prothrombin fragment 1+2 (F1+2) in the per protocol set. Data are presented as box-and-whisker plots, in which the box represents the interquartile range, with the median indicated by the horizontal line, the whiskers represent the 90th and 10th percentiles, and the open symbols represent the 95th and 5th percentiles. The significance of differences in the concentrations of edoxaban and myocardial injury (CK) and coagulation (D-dimer and F1+2) biomarkers was verified using the Mann-Whitney test (2-tailed). CA, catheter ablation; CB, cryoballoon; RF, radiofrequency.
D-dimer concentrations increased the day after CA in both the RF and CB groups (P<0.0001; data not shown), and the D-dimer concentrations in the CB group were significantly higher than those in the RF group (P<0.0001; Figure). Immediately before CA to the day after CA, the mean (±SD) differences in D-dimer plasma concentrations were approximately 2-fold greater in the CB group than the RF group (0.25±0.26 vs. 0.13±0.15 μg/mL, respectively; P<0.0001; Table 3).
Concentrations of F1+2 were significantly higher in the CB than RF group both before CA (P=0.014) and on the day after CA (P=0.006; Figure). Although F1+2 concentrations increased in both groups on the day after CA, there were no significant differences in the changes in F1+2 from immediately before CA to the day after CA between the RF and CB groups (P=0.093; Table 3).
Using data from the KYU-RABLE study,10 we conducted an exploratory subanalysis to determine the effects of CA energy (RF vs. CB) on myocardial injury, as well as plasma concentrations of edoxaban and coagulation biomarkers measured immediately before and the day after CA. We observed higher plasma concentrations of edoxaban in both the RF and CB groups on the day after CA. Regarding the higher edoxaban concentration on the day after CA, it is important to note the time from the end of the CA procedure to the first edoxaban dose after concluding CA (6.2 and 7.0 h in the RF and CB groups, respectively) and the time from the last edoxaban dose before CA to when edoxaban was restarted after CA (36.3 and 35.8 h in the RF and CB groups, respectively); the difference between these times, which is approximately 30 h, accounts for the time from the final administration of edoxaban immediately before CA to the time of blood sampling. Thus, the blood concentration of edoxaban was higher on the day after CA based on the time elapsed from edoxaban administration to blood sample collection. These results are consistent with those of the main KYU-RABLE study,10 and with previously reported data from AF patients with normal renal function or mild renal impairment in which the median (minimum–maximum) concentrations were 9.7 (3.9–23.2) ng/mL with edoxaban 30 mg and 18.2 (6.9–29.5) ng/mL with edoxaban 60 mg.12 The plasma concentration of edoxaban was not significantly different between the CB and RF groups; this suggests the efficacy of uninterrupted edoxaban, regardless of the type of energy source used for CA.
Significantly higher increases in CK concentrations from immediately before to the day after CA were observed in the CB group than the RF group (P<0.0001). This finding is consistent with the mechanism of myocardial injury described for CB ablation,13 as well as the results of previous studies of patients with paroxysmal AF undergoing PV isolation by CB and RF ablation.4,5 RF and CB ablation cause tissue injury by different mechanisms.14,15 CB ablation causes more significant myocardial injury and is considered to have a wider PV isolation range.4 These CB ablation characteristics are attributed to good contact of the balloon with myocardial tissue, which increases the lesion size and increases levels of myocardial injury biomarkers. Increased levels of myocardial injury biomarkers in the CB group may be indicative of a more effective CA.5
Levels of D-dimer and F1+2 increased in both groups, and the levels of both biomarkers were significantly higher in the CB group than the RF group after CA (P<0.0001 and P=0.006, respectively). The increase in D-dimer in the CB group was nearly 2-fold that in the RF group. This difference between the groups may be attributed to the different characteristics of the CB and RF ablation techniques, as well as differences in patient characteristics.15,16 For F1+2, the mean perioperative levels were slightly different between the RF and CB groups before CA, possibly due to a single patient in the CB group with an extremely high F1+2 concentration (3,247 pmol/L). However, there were no significant differences in the changes from immediately before to the day after CA between the RF and CB groups. F1+2 is generated upstream of D-dimer in the coagulation/fibrinolysis system. F1+2 are peptides released when prothrombin is activated by Factor Xa to yield thrombin. It is possible that the effect of edoxaban, a DOAC that specifically inhibits Factor Xa,17 contributed to the lack of differences in F1+2, not in D-dimer, between the RF and CB groups. Further, it has been reported that the risk of blood clot formation increases when CA is performed,6 but the mechanism remains unclear. It is possible that physical and both high- and low-temperature stimulation, such as that during CA with RF or CB, respectively, may enhance blood coagulation. It has been reported that blood coagulation activity was enhanced by a decrease in blood temperature in the right atrium due to CB ablation.18 In contrast, the present results differed from those of other studies that reported significant increases in prothrombotic markers of a similar magnitude after ablation with RF and CB.3,4 Nevertheless, the changes in plasma concentrations of D-dimer and F1+2 in the present study are in line with those observed in the main KYU-RABLE study,10 as well as in other studies.18,19
All biomarkers assessed were increased on the day after CA, regardless of the energy source. Although both myocardial injury and coagulation biomarkers were significantly higher in the CB than RF group, there were no significant differences in the incidence of thrombotic or bleeding events, suggesting that edoxaban can be safely used during CA with RF or CB. Concentrations of D-dimer and prothrombin fragment levels in the plasma indicate the activation of blood coagulation and fibrinolysis. D-dimer and prothrombin fragments are well-known predictors of thromboembolism,20 and both these biomarkers were significantly higher on the day after CA in the CB compared with RF group in the present study. The present findings suggest that perioperative anticoagulation therapy should be performed carefully in patients who are to undergo either RF or CB ablation, but especially CB ablation.
The 2018 Japanese Circulation Society/Japanese Heart Rhythm Society guidelines for the non-pharmacotherapy management of arrhythmias recommend both CB and RF ablation for the management of AF.16 The present subanalysis of the KYU-RABLE study10 revealed that the frequency of CA procedures with RF was more than 3-fold that of CA with CB. Ultimately, the choice of the energy source for a CA procedure depends on the judgment of the treating physician and the availability of trained personnel and equipment at each facility. However, the use of CB ablation in Japan has been growing rapidly since 2014, at which time the insurance reimbursement for CB ablation was established,21 and is expected to continue to increase.
The present exploratory study has some limitations. Intraoperative heparin concentrations and activated clotting time were lower in the CB group than the RF group, which may have contributed to a higher level of coagulation biomarkers in the CB than RF ablation group. In contrast, the longer total procedure time for RF compared with CB ablation may have affected coagulation biomarker and CK concentrations. These effects are not understood in detail, but could have resulted in higher coagulation biomarker and CK concentrations after CA with CB than with RF. Moreover, some measurement methods (e.g., those used to assess CK concentrations) were not centralized and were measured separately at each facility, meaning that reported CK values varied.
Among Japanese NVAF patients receiving uninterrupted periprocedural edoxaban, plasma edoxaban concentrations were similar between the CB and RF groups. Both myocardial injury and coagulation biomarkers increased after CA, especially in the CB group, but no difference in the incidence of thrombotic or bleeding events was noted. The present findings suggest that edoxaban can be used safely during CA with RF or CB, without increasing periprocedural complications. The present findings also indicate that periprocedural anticoagulation is needed for both RF and CB ablation and that the management of anticoagulation should be handled with care, especially when CB is used for CA.
The authors thank Keyra Martinez Dunn, MD, of Edanz Pharma, for providing medical writing support, which was funded by Daiichi Sankyo Co., Ltd.
This research and manuscript development were funded by Daiichi Sankyo Co., Ltd.
This trial has been registered with the University Hospital Medical Information Network (UMIN) Clinical Trials Registry (ID: UMIN000029693).
T.S. and H.O. have no conflicts of interest to disclose. N.T. has received remuneration from Daiichi Sankyo, Bristol-Myers Squibb, and Pfizer Japan and research funding from Ono Pharmaceutical, and is a member of Circulation Journal’s Editorial Team. Y.M. has received remuneration from Daiichi Sankyo, Bristol-Myers Squibb, and Pfizer Japan, and research funding from Bristol-Myers Squibb. T.K., K.Y., and A.T. are employees of Daiichi Sankyo. K.O. has received remuneration from Nippon Boehringer Ingelheim, Daiichi Sankyo, Johnson & Johnson, and Medtronic.
This study was approved by the Ethics Committee of Oita University Faculty of Medicine (Reference no. B17-021, November 12, 2017).
The deidentified participant data will be shared upon request basis to the corresponding author. Deidentified data supporting the results presented in this article, as well as the study protocol, will be shared. The data will become available from submission up to 36 months after article publication. Researchers requesting access to the will need to provide a methodologically sound proposal, which may be reviewed by a committee led by Daiichi-Sankyo Co., Ltd. The shared data can be used for any purpose, and proposals should be directed to the corresponding author. To gain access to the data, those requesting access will need to sign an access data agreement.
