Abstract: OBJECTIVE To establish an UPLC-MS/MS method for the determination of rivaroxaban in Beagle dog plasma. METHODS The plasma samples were treated with protein precipitated by methyl tert-butyl ether-dichloromethane (2:1), the deuterium substituted rivaroxaban(rivaroxaban-d₄) was selected as internal standard. The chromatographic separation was performed with a Thermo BDS Hypersil C₁₈ column (100 mm×4.6 mm, 2.4 μm) and gradient mobile phase (mobile phase A was methanol solution containing 0.015 mol·L^-1 ammonium formate and 0.1% formic acid; mobile phase B was water-methanol (90:10) solution containing 0.015 mol·L^-1 ammonium formate and 0.1% formic acid.). Mass spectrometry conditions were ESI⁺ with multiple reaction monitoring(MRM) mode. The ion reactions used for quantitative detection wereM+H⁺ m/z 436@30 eV→ m/z 144.900 (rivaroxaban) andM+H⁺ m/z 440@24eV→m/z 144.900 (rivaroxaban-d₄). RESULTS Good linearity was obtained for rivaroxaban in the range of 0.5-400 ng·mL^-1(r=0.996 1), the lower quantification of current method was 0.5 ng·mL^-1 for rivaroxaban. The intra- and inter-batch precision were <13%. The absolute recoveries ranged from 89.4% to 96.5% and the matrix effects were 94.7%-97.2%. The plasma samples were stored frozen (-80℃) for 33 d with good stability. After three freeze-thaw cycles of the samples and at room temperature for 12 h after extraction, the change in samples concentration were not obvious. It was applied in dog pharmacokinetic study after oral administration of rivaroxaban tablets. The main pharmacokinetic parameters were as follows:T_max (2.11±1.133)h, C_max (179.3±50.65)ng·mL^-1, t_1/2(10.05±4.34)h, AUC_0-t (1161.6±339.74)h·ng·mL^-1. CONCLUSION The trial results show that the method is simple in operation, high in sensitivity, good in reproducibility, accurate and reliable, and can be used to determine the concentration of rivaroxaban in Beagle dog plasma and in vivo pharmacokinetic studies in Beagle dog.