The Influence of Assay Selection on Prothrombin Time Measured in Patients Treated With Rivaroxaban for Nonvalvular Atrial Fibrillation

BACKGROUND: Prothrombin time (PT) can provide a qualitative assessment of the relative intensity of anticoagulation by rivaroxaban. More than ten types of assay are available for the measurement of PT in clinical settings, but it is not yet fully understood whether their interactions with rivaroxaban are uniform or inconsistent. METHODS: We examined 139 blood samples from patients taking rivaroxaban. We measured PT using five different commercially available assays. We also evaluated the estimated rivaroxaban concentration using a chromogenic anti-factor Xa assay. RESULTS: The median estimated concentration of rivaroxaban was 192 ng/ml (interquartile range 85-284 ng/ml). The correlation coefficient (r) between PT and the estimated concentrations of rivaroxaban was as follows: Thromborel S, r = 0.768; Thrombocheck PT, r = 0.861; Coagpia PT-N, r = 0.909; Neoplastin Plus, r = 0.882; and Triniclot PT Excel S, r = 0.870. The gradients of the regression plots differed more than fourfold, and the standard deviation of the regression line ranged from 1.001 to 2.980, which tended to be higher for the assays with the higher regression slope gradients. CONCLUSION: The estimated concentration of rivaroxaban varied greatly depending on the assay, so the PT measured in patients taking rivaroxaban should be interpreted with caution.