Bleeding during and after ERCP? A real world study evaluating clinical practice guidelines for anticoagulant and antiplatelet drugs

Aims Anticoagulants according to guidelines play a role in the development of bleeding during and following Endoscopic Retrograde Cholangiopancreatography (ERCP). At the same time it is recommended that antiplatelets with the exception of aspirin should be also discontinued before ERCP. Aim of our study was to evaluate the impact of following these guidelines in a real world setting. Methods We analysed prospectively documented clinical, epidemiological and ERCP data from consecutive patients (n=3147, mean age:68±16, Females:51%), with an intact papilla who were submitted to an ERCP during a 20 years-period. Data analysis and collection was executed on 2025. In patients submitted to ERCP from 2000 until 2011 (period A n=1521, ) all anticoagulants and antiplatelet drugs were arrested before ERCP while in patients submitted to ERCP from 2012 until 2020 (period B, n=1626) aspirin was not arrested before ERCP. Results Antiplatelet use did not differ between the two periods (7.7% respectively, p=1.000) while anticoagulants were used numerically but not statistically significantly more often in the period B (6.5% vs 5%, p=0.096, συνολικά 5.8%). Bleeding during ERCP was observed in total in 33.3% and was observed more commonly in period B Β (40.9% vs 25.9%, p<0.001), upon periampullary diverticulum existence (37.9%), when antiplatelet (if yes 43.4% vs 32.5% if not used, p<0.001) or anticoagulant drugs were used (if yes 42.0% vs 32.8% if not used, p=0.012) and if needle knife (NK) was used (if yes 41.5% vs 28.6% if not used, p<0.001). All the above mentioned factors were independently correlated with intraprocedural bleeding during ERCP. (PAD: OR:1.354, p=0.002, antiplatelet use: OR:1.323, p=0.005, anticoagulant use: OR:1.275, p=0.049, NK use: OR:1.610, p<0.001) The rate of clinical significant bleeding post ERCP was in total 1.2% without difference between the two periods (Β:1.2% vs A:1.3%, p=0.749), while it was observed more commonly if intraprocedural bleeding occurred (if yes 2.7% vs 0.5% if not used, p<0.001) as also upon anticoagulant use (if yes 6.7% vs 0.9% if not used, p<0.001) but not antiplatelet use (if yes 1.7% vs 1.2% if not used, p=0.544), needle knife use (if yes 1.5% vs 1.1% if not used, p=0.320) or the existence of a periampullary diverticulum (if yes 1.3% vs 1.2% if not used, p=0.998). Both intraprocedural bleeding and anticoagulant usage were independent factors correlated with clinical significant bleeding post ERCP (OR:4.842, p<0.001 and OR:1.855, p<0.001 respectively). Conclusions The policy of not arresting aspirin before ERCP is correlated with increased rate of intraprocedural bleeding but is not a factor predisposing patients to clinical significant bleeding. Anticoagulant use is both correlated with intraprocedural and clinical significant bleeding. Article published online: 05 June 2026 © 2026. European Society of Gastrointestinal Endoscopy. All rights reserved. Georg Thieme Verlag KG Oswald-Hesse-Straße 50, 70469 Stuttgart, Germany