BACKGROUND AND AIM: Anticoagulants are used to prevent thromboembolic events. Direct oral anticoagulants (DOAC) are our new choice; however, their effect on bleeding risk for endoscopic treatment has not been reported. We aimed to assess the clinical effect of DOAC compared to warfarin for gastric endoscopic submucosal dissection (ESD). METHODS: We retrospectively studied 97 patients on anticoagulants and treated 108 gastric neoplasms with ESD in three referral institutes. Twenty-four patients were taking DOAC, including dabigatran (12), rivaroxaban (11), and apixaban (one) and 73 were taking warfarin. RESULTS: In the DOAC group, delayed bleeding rate was significantly higher in patients on rivaroxaban than in patients on dabigatran (45% vs 0%, P < 0.05) without relation to heparin bridge therapy (HBT). In the warfarin group, 78% of patients underwent HBT, and delayed bleeding rate was significantly higher in patients with HBT than in those without (36% vs 0%, P < 0.05). Delayed bleeding rate increased as intake of antithrombotic agents increased (P < 0.05). HBT period was shorter (P < 0.05) in DOAC because DOAC achieve the maximum effect quicker, and hospitalization period was shorter (P < 0.05), compared with warfarin. Multivariate analysis showed that HBT (OR, 10.7), rivaroxaban (OR, 6.00) and multiple antithrombotic agents (OR, 4.35) were independent delayed bleeding risk factors. CONCLUSIONS: The DOAC effect differs in each agent. Dabigatran is a feasible alternative to warfarin for shortening the hospitalization period and decreasing delayed bleeding rate, although rivaroxaban has a significantly higher delayed bleeding risk.
BACKGROUND AND AIM: Anticoagulants are used to prevent thromboembolic events. Direct oral anticoagulants (DOAC) are our new choice; however, their effect on bleeding risk for endoscopic treatment has not been reported. We aimed to assess the clinical effect of DOAC compared to warfarin for gastric endoscopic submucosal dissection (ESD). METHODS: We retrospectively studied 97 patients on anticoagulants and treated 108 gastric neoplasms with ESD in three referral institutes. Twenty-four patients were taking DOAC, including dabigatran (12), rivaroxaban (11), and apixaban (one) and 73 were taking warfarin. RESULTS: In the DOAC group, delayed bleeding rate was significantly higher in patients on rivaroxaban than in patients on dabigatran (45% vs 0%, P < 0.05) without relation to heparin bridge therapy (HBT). In the warfarin group, 78% of patients underwent HBT, and delayed bleeding rate was significantly higher in patients with HBT than in those without (36% vs 0%, P < 0.05). Delayed bleeding rate increased as intake of antithrombotic agents increased (P < 0.05). HBT period was shorter (P < 0.05) in DOAC because DOAC achieve the maximum effect quicker, and hospitalization period was shorter (P < 0.05), compared with warfarin. Multivariate analysis showed that HBT (OR, 10.7), rivaroxaban (OR, 6.00) and multiple antithrombotic agents (OR, 4.35) were independent delayed bleeding risk factors. CONCLUSIONS: The DOAC effect differs in each agent. Dabigatran is a feasible alternative to warfarin for shortening the hospitalization period and decreasing delayed bleeding rate, although rivaroxaban has a significantly higher delayed bleeding risk.