Delayed concentration effect models for dabigatran anticoagulation.

ABSTRACT

INTRODUCTION: Dabigatran is an anticoagulant with potential use during cardiopulmonary bypass in children and adults. The pharmacokinetic-pharmacodynamic relationship for dabigatran anticoagulation effect was investigated in an intact animal model using rabbits. METHODS: Ten male New Zealand white rabbits were given a novel preparation of intravenous dabigatran 15 mg.kg-1 . Blood samples were collected for activated clotting time, thromboelastometric reaction time, and drug assay at 5, 15, 30, 60, 120, 180, 300, and 420 min. Plasma dabigatran concentrations and coagulation measures were analyzed using an integrated pharmacokinetic-pharmacodynamic model using nonlinear mixed effects. Effects (activated clotting and thromboelastometric reaction times) were described using a sigmoidal EMAX model. Pharmacokinetic parameters were scaled using allometry and standardized to a 70 kg size standard. Pharmacodynamics were investigated using both an effect compartment model and an indirect response (turnover) model. RESULTS: A two-compartment model described dabigatran pharmacokinetics with a clearance (CL 0.135 L.min-1 .70 kg-1 ), intercompartment clearance (Q 0.33 L.min-1 .70 kg-1 ), central volume of distribution (V1 12.3 L.70 kg-1 ), and peripheral volume of distribution (V2 30.1 L.70 kg-1 ). The effect compartment model estimates for a sigmoid EMAX model with activated clotting time had an effect site concentration (Ce50 20.1 mg.L-1 ) eliciting half of the maximal effect (EMAX 899 s) and a Hill coefficient (N 0.66). The equilibration half time (T1/2 keo) was 1.4 min. Results for the reaction time were plasma concentration (Cp50 65.3 mg.L-1 ), EMAX 34 min, N 0.80 with a baseline thromboelastometric reaction time of 0.4 min. The equilibration half time (T1/2 keo) was 2.04 min. CONCLUSIONS: Dabigatran reversibly binds to the active site on the thrombin molecule, preventing thrombin-mediated activation of coagulation factors. The effect compartment model performed slightly better than the turnover model and was able to adequately capture pharmacodynamics for both activated clotting and thromboelastometric reaction times. The equilibration half time was short (<2 min). These data can be used to inform future animal preclinical studies for those undergoing cardiopulmonary bypass. These preclinical data also demonstrate the magnitude of parameter values for a delayed effect compartment model that are applicable to humans.