Yamaji 2019.

Study characteristics
Methods	Study design: randomised controlled trial Study grouping: parallel group Total duration of the study: January 2017 to October 2018 Duration of follow‐up: data obtained over 120 days (from 30 days before ablation to 90 days after ablation). Initial follow‐up visit was scheduled 2 weeks after AF ablation Number of study centres and location: single centre in Japan
Participants	Inclusion criteria: people with paroxysmal AF, persistent and long‐standing AF, or atrial tachycardia; people who underwent their first AF ablation between January 2017 and October 2018 Exclusion criteria: decreased renal function (CrCl rate < 30 mL/minute)   Total number of participants: 584 Number of randomised participants: 584 Number lost to follow‐up/withdrawn: 0 Number of analysed participants: 584 Number of participants in each treatment group: minimally interrupted DOAC: 307; uninterrupted DOAC: 277 Baseline characteristics Interrupted anticoagulation Age, years, mean: 65.0 (SD 10.5) Men, n (%): 212 (69) CHADS2 score, mean: 0.90 (SD 0.87) CHA2DS2‐VASc score, mean: 1.88 (SD 1.37) HAS‐BLED score, mean: 1.35 (SD 1.05) Type of AF (n): paroxysmal AF: 199 persistent AF: 61 long‐standing persistent AF: 40 Comorbidities, n (%): diabetes mellitus: 34 (11) hypertension: 144 (47) prior stroke or TIA: 13 (4) CHF: 22 (7) Concomitant medications, (n): antiplatelet agents: 10 dabigatran: 56 rivaroxaban: 109 apixaban: 59 edoxaban: 83 Uninterrupted anticoagulation Age, years, mean: 66.4 (SD 10.3) Men, n (%): 211 (76.17) CHADS2 score, mean: 0.97 (SD 0.97) CHA2DS2‐VASc score, mean: 1.88 (SD 1.37) HAS‐BLED score, mean: 1.42 (SD 1.04) Type of AF, (n): paroxysmal AF: 171 persistent AF: 65 long‐standing persistent AF: 38 Comorbidities, n (%): diabetes mellitus: 39 (14) hypertension: 134 (48) prior stroke or TIA: 16 (6) CHF: 6 (2) Concomitant medications, (n): antiplatelet agents: 7 dabigatran: 83 rivaroxaban: 65 apixaban: 58 edoxaban: 71 Group differences: no significant differences in clinical and echocardiogram parameters and thromboembolic and bleeding risk scores between groups, except for AF duration, which was about 1 year shorter in the minimally interrupted group than the uninterrupted group. Significant difference in proportion of participants using dabigatran and rivaroxaban between groups. More participants with heart failure in minimally interrupted group
Interventions	Procedural anticoagulation Anticoagulants used: dabigatran, rivaroxaban, edoxaban, and apixaban Dose: dabigatran and apixaban twice a day (morning and evening). Rivaroxaban and edoxaban once a day (morning) Duration of anticoagulant therapy in trial: initiated 30 days before ablation Intensity of anticoagulation or dose adjustment: not reported Adherence to anticoagulant treatment: rivaroxaban administered once a day in the morning, rather than in the evening, to maintain a sufficient adherence rate Time of interruption of the anticoagulants prior to procedure: holding the morning dose of DOAC on the day of ablation Heparin bridge therapy: not used Time and strategy of resumption of interrupted OAC after procedure: single dose of apixaban or dabigatran resumed in evening of the day of ablation or 4 hours after the completion of the postmeridiem ablation session, with confirmation of haemostasis. Rivaroxaban and edoxaban administrations resumed in the morning of the day after ablation. DOAC therapy continued for ≥ 3 months after TOE: not reported  Ablation procedure Type of ablation: superior vena cava isolation/cavotricuspid isthmus ablation Ablation energy source: not reported Intraprocedural anticoagulant Heparin bolus administered just before septal puncture, based on age, sex, and bodyweight. If pre‐ACT value 120–130 U/Kg for a pre‐ACT ≥ 150 seconds and 140–150 U/kg for a pre‐ACT < 150 seconds. IV infusion heparin 400 U/hour to maintain an ACT of 300–400 seconds
Outcomes	Primary outcome Pre‐ACT Primary safety outcome Composite of bleeding and thromboembolic complications (yielding the bleeding and thromboembolic risk score) Secondary outcome Thromboembolic and bleeding complications Thromboembolic complications: cerebrovascular accidents and TIAs once intracranial haemorrhage had been ruled out. Pulmonary embolism and deep venous embolism Major bleeding complications defined as cardiac tamponade, retroperitoneal bleeding, and groin haematoma requiring blood transfusion Cardiac tamponade defined by characteristic clinical features and the presence of a considerable pericardial effusion requiring drainage. Late cardiac tamponades were those occurring greater than 48 hours after the procedure Minor bleeding complications: pericardial effusion reduced haemoglobin without blood transfusion, and haematuria defined as minor complications. Pericardial effusion defined as an effusion identified in the pericardial space by routine follow‐up echocardiography, without haemodynamic disturbance (non‐tamponade)
Notes	Investigators' conflicts of interest: none Funding: not reported Country: Japan Setting: Okayama Heart Clinic Comments: selection and doses of DOACs were not randomised and were at the discretion of each treating physician, considering the participant's characteristics (including renal function) and drug manufacturer's directions Author's name: Hirosuke Yamaji Institution: Heart Rhythm Center, Okayama Heart Clinic, Okayama University Email: yamaji2@mac.com Address: Heart Rhythm Center, Okayama Heart Clinic, Okayama University, Takeda 54‐1, Naka‐Ku, Okayama 703‐8251, Japan
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	High risk	Randomisation method not reported. Also, they mentioned occurrence of randomisation error. Quote: "although patients were randomly allocated to the min‐Int and Unint DOAC therapy groups, due to unintended randomized error, the number of patients was not completely equivalent between these two anticoagulation strategy groups."
Allocation concealment (selection bias)	Unclear risk	Methods of allocation concealment not reported.
Blinding of participants and personnel (performance bias) all outcomes	Unclear risk	No mention if participants and personnel were aware of intervention.
Blinding of outcome assessment (detection bias) all outcomes	Unclear risk	Methods of blinding of outcome assessors not reported.
Incomplete outcome data (attrition bias) all outcomes	Unclear risk	Participant flow chart is not reported, no mention of excluded participants. Insufficient information to judge.
Selective reporting (reporting bias)	Unclear risk	No protocol available.