1. Steady-state plasma concentrations of anticoagulants and the time since the previous administration in mainly outpatients with atrial fibrillation administered standard or reduced doses were analyzed for 110 elderly Japanese subjects (mean age, 76 years) treated with apixaban (2.5 or 5.0 mg twice daily), dabigatran etexilate (110 or 150 mg twice daily), edoxaban (30 or 60 mg once daily), or rivaroxaban (10 or 15 mg once daily) at one general hospital. 2. The pharmacokinetics in patients treated with standard and reduced doses of the four anticoagulants using liquid chromatography-tandem mass spectrometry were compared with the concentration ranges estimated using physiologically based pharmacokinetic modelling. Reduced doses of anticoagulants resulted in relatively small pharmacokinetic variations compared with the standard dose. 3. Statistical analyses revealed that renal impairment is likely not the sole determinant factor for high plasma concentrations of apixaban, dabigatran, edoxaban, and rivaroxaban. 4. Patients with atrial fibrillation should be treated with the correct doses of oral anticoagulants as specified in the package inserts (e.g., reduced doses for elderly patients, patients with low body weights, and in combination with P-glycoprotein inhibitor drugs) to avoid excessive or insufficient doses of direct oral anticoagulants.

1. Benzydamine is used clinically as a nonsteroidal anti-inflammatory drug in oral rinses and is employed in preclinical research as a flavin-containing monooxygenase (FMO) probe substrate. In this study, plasma concentrations of benzydamine and its primary N-oxide and N-demethylated metabolites were investigated in control TK-NOG mice, in humanized-liver mice, and in mice whose liver cells had been ablated with ganciclovir. 2. Following oral administration of benzydamine (10 mg/kg) in humanized-liver TK-NOG mice, plasma concentrations of benzydamine N-oxide were slightly higher than those of demethyl benzydamine. In contrast, in control and ganciclovir-treated TK-NOG mice, concentrations of demethyl benzydamine were slightly higher than those of benzydamine N-oxide. 3. Simulations of human plasma concentrations of benzydamine and its N-oxide were achieved using simplified physiologically based pharmacokinetic models based on data from control TK-NOG mice and from reported benzydamine concentrations after low-dose administration in humans. Estimated clearance rates based on data from humanized-liver and ganciclovir-treated TK-NOG mice were two orders magnitude high. 4. The pharmacokinetic profiles of benzydamine were different for control and humanized-liver TK-NOG mice. Humanized-liver mice are generally accepted human models; however, drug oxidation in mouse kidney might need to be considered when probe substrates undergo FMO-dependent drug oxidation in mouse liver and kidney.