Cheng Han Ng1, Lincoln Kai En Lim1, Gwyneth Kong1, Zachariah Gene Wing Ow1, Yip Han Chin1, Xiao Jieling1, Webber Chan Pak Wo6, Mark Y. Chan1,3, Choon Seng Chong1,2, Nicholas WS Chew3, Khek Yu Ho1,4, Mark Dhinesh Muthiah1,4,5

1Yong Loo Lin School of Medicine, National University of Singapore, Singapore, Singapore, 2Division of Colorectal Surgery, National University Health System, Singapore, 3Division of Cardiology, Department of Medicine, National University Hospital, Singapore, Singapore, 4Division of Gastroenterology and Hepatology, Department of Medicine, National University Hospital, Singapore, Singapore, 5National University Centre for Organ Transplantation, National University Health System, Singapore, 6Division of Gastroenterology, Department of Medicine, Singapore General Hospital, Singapore, Singapore

Background/Aims: The use of anticoagulation has been associated with increased risk of post-procedural bleeding (PPB) in polypectomy, resulting in concerns and debates regarding thrombosis and anticoagulant use in colorectal polypectomy. However, recent studies have found varying safety profile with different anticoagulants. This study aims to compare bleeding rates between different methods of anticoagulation through a Bayesian Network meta-analysis.

Methods: Medline and Embase databases were searched to identify articles relating to the use of uninterrupted warfarin, Heparin Bridging Therapy (HBT) or direct oral anticoagulant (DOACs) in colorectal polypectomy along with a control with patients undergoing polypectomy after stopping anticoagulation. Bayesian network analysis was conducted to summarize the evidence and the surface under the cumulative ranking curve (SUCRA) was used to rank treatments.

Results: 12 studies with 10,739 patients were included and only one individual experience a thromboembolic event. The rate of overall bleeding was 1.90% (CI:1.02%-3.52%) with no anticoagulant, 4.67% (CI:1.56%-12.92%) with uninterrupted warfarin, 9.13% (CI:6.88%-12.01%) with DOACs and 14.56% (11.66%- 8.03%) with HBT. SUCRA analysis found the no anticoagulant use has the lowest likelihood of overall bleeding (0.996), followed by uninterrupted warfarin (0.687), DOACs stopped at 24 hours (0.489), DOACs stopped in the morning (0.480), HBT (0.198) and uninterrupted DOACs (0.147). There was 10 times increase in likelihood of bleeding between heparin and no anticoagulant (Crl: 4.9-21). Uninterrupted warfarin was found to be the safest anticoagulant with lowest likelihood of both overall (0.687) and delayed bleeding (0.644). No thromboembolism event was reported in warfarin, DOACs (24hrs), DOACs (morning) and uninterrupted DOACs. There was one event each in HBT and no use of anticoagulant.

Conclusion: This meta-analysis summarizes the likelihood of bleeding from different anticoagulation therapies. Uninterrupted warfarin was found to be the safest anticoagulant therapy. Heparin bridge had a surprisingly high rate of bleeding.

Keywords: Anticoagulant, PPB, Bleeding, Endoscopy