Heparin resistance (HR) is a common event in cardiac operations. At present, no clear recognition of the risk factors for HR has been reached. The aim of this study was to determine a predictive model for HR, based on the preoperative patient's profile. Two hundred consecutive patients scheduled for elective coronary artery bypass operations were enrolled in a prospective trial. Demographics, type of preoperative anticoagulation therapy and preoperative coagulation profile were collected and statistically analysed with respect to the evidence of a HR. Heparin resistance was defined as at least one activated clotting time < 400 s after heparinization and/or the need for purified antithrombin III (AT-III) administration. With a multivariate analysis we could identify five predictors for HR: AT-III < or = 60%; preoperative subcutaneous heparin therapy; intravenous heparin therapy; platelet count > or = 300000 cells/mm3; age > or = 65 years. We conclude that HR is a predictable event. In the presence of all the risk factors, the likelihood of HR is 99%; in the absence of all of them, it is 10%. Predicting HR allows us to apply many possible therapeutic strategies.