A total of 399 consecutive patients undergoing 598 ERCPs (endoscopic retrograde cholangiopancreatographies), including 88 pre-cut papillotomies and 206 conventional papillotomies, are described in a retrospective study. Clotting parameters, haemoglobin levels, indications for pre-cut and/or conventional papillotomy and the use of drugs assumed to interfere with blood clotting (anticoagulants, platelet-aggregation inhibitors, low-molecular-weight heparin) were evaluated in order to detect risk factors for ERCP-associated bleeding. The overall incidence of ERCP-associated bleeding was 18/598 (3.0%). The incidence of bleeding in the group without papillotomy was 7/346 (2.0%). This group consisted of patients who underwent only a diagnostic ERCP, patients who had undergone papillotomy previously, patients in whom a renewed attempt was made to extract biliary stones, and patients in whom removal or change of a stent was necessary. The incidence of papillotomy-associated bleeding was 11/252 (4.4%). Pre-cut papillotomy did and conventional papillotomy did not significantly increase the incidence of bleeding: 15.2% (P < 0.001) and 1.9% (P= 1.00) respectively. The incidence of ERCP-associated bleeding in the group not using any drugs interfering with blood clotting was 2.5%. The use of low-molecular-weight heparin (10.3%) during ERCP significantly increased the risk of bleeding (P= 0.01). However, the use of platelet aggregation inhibitors (2.4%) did not (P= 1.00). As the incidence of bleeding in patients with normal clotting parameters, including the patients with abnormal parameters which were well corrected (4.3%), was higher than in patients with abnormal haemostatic screens (2.7%), abnormal coagulation tests did not predict ERCP-associated bleeding.