Abstract
Background and study aims Delayed bleeding and thrombotic events are uncontrolled adverse events that are hard to balance in patients receiving anticoagulants after endoscopic resection. The present study aims to assess the clinical effect of warfarin, when compared to direct oral anticoagulants (DOACs), in terms of delayed bleeding and thrombotic events. Methods A comprehensive electronic literature search was conducted for eligible literature. Pairwise meta-analyses were performed on outcomes of delayed bleeding and thrombotic events. Two networks within the Bayesian framework were established based on the management of anticoagulants and type of DOAC. Results Eight cohort studies with 2,046 patients were eligible for inclusion, including 1,176 patients treated with warfarin and 870 with DOACs. There was no significant difference between warfarin and DOACs, in terms of delayed bleeding (OR = 1.29, 95 % CI [0.99-1.69]) and thromboembolism (OR = 2.0, 95 % CI [0.32-12.39]). In the network meta-analyses for delayed bleeding, the rank probabilities revealed that the safest management was discontinuous warfarin without heparin bridge therapy (HBT). Rank probabilities for the types of DOACs demonstrated that the safest drug was dabigatran. Conclusions There was no significant difference in delayed bleeding and thromboembolism between warfarin and DOACs in patients receiving endoscopic treatment. In terms of delayed bleeding, discontinuous warfarin without HBT was suggested as the best management, and dabigatran was recommended as the best type of DOAC.