Aims We conducted a phase I placebo-controlled trial with two i.v. doses (0.5 mg h x1 and 3 mg h x1 ) of S 18326, a selective thrombin inhibitor that interacts with the catalytic site of thrombin, with the aim to study the relationships between increasing plasma levels of S 18326 and changes in coagulation tests and thrombin generation markers. Methods Thirty-six healthy male volunteers were divided into three groups. In each group, 10 volunteers were randomly assigned to receive S 18326 and two to receive a placebo. Following a bolus of 4.5 mg, doses were 0.5 mg h x1 in the ®rst group and 3 mg h x1 in the two other groups, administered as an i.v. infusion for 24 h. Blood was drawn repeatedly up to 36 h after the bolus, and tested for the activated clotting time (ACT) and activated partial thromboplastin time (APTT). The APTT reagent was chosen among ®ve commercial reagents to yield a linear increase in the clotting time among possible therapeutic S 18326 concentrations in vitro. To accurately measure the thrombin-inhibiting effects of low doses of S 18326 (<0.5 mM), we developed a speci®c chromogenic assay. We also measured F1+2 prothrombin fragment levels to assess the effect of S 18326 on thrombin generation in vivo. Results A two-compartment pharmacokinetic model was ®tted to the S 18326 plasma concentration vs time data by using population pharmacokinetic methods. Results of the pharmacodynamic-pharmacokinetic relationships showed that both the ACT and APTT methods yielded a linear increase according to the S 18326 concentration measured using a highly sensitive analytical method. At the end of infusion, ACT was prolonged 1.20 and 1.95-fold in the 0.5 mg h x1 and the 3 mg h x1 groups, respectively, and APTT was prolonged 1.27 and 2.75-fold. Thrombin inhibition plateaued above 0.5 mM of S 18326 according to an Emax model, con®rming that the test was highly sensitive. F1+2 levels fell signi®cantly after the 24 h S 18326 infusion (0.83 nM to 0.6 nM and 0.80 nM to 0.44 nM in the 0.5 mg h x1 and the 3 mg h x1 groups, respectively), but remained stable after the placebo infusion. Conclusions Our results support speci®c monitoring of the thrombin inhibitor S 18326 with ACT and APTT to establish the safety range of the drug in further studies. Moreover, the fall in F1+2 prothrombin fragments suggests that S 18326 effectively reduces the retroactivation of factors V and VIII by thrombin.