Risk Factors for Delayed Bleeding after Therapeutic Gastrointestinal Endoscopy in Patients Receiving Oral Anticoagulants: A Multicenter Retrospective Study

Background/Aims: Delayed bleeding is among the adverse events associated with therapeutic gastrointestinal endoscopy. The aim of this study was to evaluate risk factors for delayed bleeding after gastrointestinal endoscopic resection in patients receiving oral anticoagulants as well as to compare the rates of occurrence of delayed bleeding between the oral anticoagulants used. Methods: We retrospectively analyzed a total of 772 patients receiving anticoagulants. Of these, 389 and 383 patients were receiving direct oral anticoagulants (DOACs) and warfarin, respectively. Therapeutic endoscopic procedures performed included endoscopic submucosal dissection (ESD), endoscopic mucosal resection, polypectomy, and cold polypectomy. Results: Delayed bleeding occurred in 90 patients (11.7%) with no significant difference between the DOAC and warfarin groups (9.5 and 13.8%, respectively). Delayed bleeding occurred significantly more frequently with apixaban than with rivaroxaban (13.5 vs. 6.4%; p < 0.05). A multivariate analysis identified continued anticoagulant therapy (OR 2.29), anticoagulant withdrawal with heparin bridging therapy (HBT; OR 2.18), anticoagulant therapy combined with 1 antiplatelet drug (OR 1.72), and ESD (OR 3.87) as risk factors for delayed bleeding. Conclusion: This study identified continued anticoagulant therapy, anticoagulant withdrawal with HBT, anticoagulant therapy combined with 1 antiplatelet drug, and ESD as risk factors for delayed bleeding after therapeutic endoscopy in patients receiving oral anticoagulants. Delayed bleeding rates were not significantly different between those receiving DOACs and warfarin. It was also suggested that the occurrence of delayed bleeding may vary between different DOACs and that oral anticoagulant withdrawal should be minimized during therapeutic gastrointestinal endoscopy, given the thromboembolic risk involved.