The synthetic direct thrombin inhibitor argatroban has a rapid onset and offset of anticoagulation. However, there are no data about the pharmacokinetic-pharmacodynamic (PK-PD) relationship of argatroban in patients undergoing contemporary percutaneous coronary intervention (PCI) and no data about other coagulation parameters than activated clotting time (ACT) in this setting. In the ARG-E04-trial, 140 patients were randomly assigned to argatroban (250, 300, or 350 µg/kg as bolus before PCI, followed by 15, 20, or 25 µg/kg/min infusion) or unfractionated heparin (70-100 IU/kg bolus). A 2-compartment model with first-order elimination adequately described the pharmacokinetic profile of argatroban over all 3 dosing groups. Clearance (CL) and distribution volumes (V1 and V2) were 21 L/h, 9.2 L, and 6.6 L, respectively. A significant sigmoidal E(max) relationship was established between the argatroban plasma concentration and the response in ACT and the endogenous thrombin potential (ETP), whereas the response in activated partial thromoplastin time (aPTT), ecarin time (ECA-T), and prothrombinase-induced clotting time (PiCT) could be described by a nonsigmoidal E(max) model. This study proves a relatively small interindividual variability of both PK and PK-PD properties of argatroban even at high doses and supports the profile of argatroban as a drug with a predictive dose-effect relationship and therefore good controllability.