Determination of an international sensitivity index of thromboplastin reagents using a WHO thromboplastin as calibrator for plasma spiked with rivaroxaban

Rivaroxaban and other direct factor Xa inhibitors are used at fixed doses without drug monitoring and dose adjustment. Patients may require determination of the anticoagulant effect during treatment. The aim of this study was to develop a method to reduce the differences between thromboplastin reagents and coagulation analysers for determination of the anticoagulant effect of rivaroxaban in human plasma. Purity of rivaroxaban extracted from commercially available drug was confirmed by mass spectrometry, elemental analysis and 1H-NMR spectroscopy. Coagulation times of pooled human plasma spiked with 50-900 ng/ml rivaroxaban were analysed. Thromboplastin reagents, WHO RBT/90, Innovin, RecombiPlasTin 2G, STA Neoplastin Plus, Technoclot PT Plus and Thromborel S, the manual Kolle-Hook method and the KC10 analyser were used. An international sensitivity index (ISI) was determined for each reagent and coagulation method using the RBT/90 thromboplastin reagent as reference. The orthogonal, used for the determination of the ISI of coumarin plasmas, and ordinary regression analyses were compared. The results showed than increasing concentrations of rivaroxaban prolonged coagulation values of all thromboplastin assays linearly (r (2)= 0.96 and r(2) = 0.99, respectively). The coefficient of variation between the slopes of the dilution curves and the ratios of the thromboplastin reagents were reduced using the international normalized ratio (INR) and ISI calculated for rivaroxaban. The ISIs of the thromboplastin reagents ranged from 0.73 to 1.67 as compared with the WHO reagent using the manual technique. The coefficient of variations between the thromboplastin reagents comparing the orthogonal and the ordinary regression analysis were 6.8 versus 3.7% (Kolle-Hook method, P = 0.0011) and 8.5 versus 4.8% (KC10 method, P < 0.0001). Using ISI for vitamin-K antagonist and rivaroxaban, the INRs for the rivaroxaban-containing samples were significantly different for one of five commercial thromboplastin reagents. In conclusion, the determination of an ISI for rivaroxaban using a WHO thromboplastin reagent is required for commercial thromboplastin reagents. The manual Kolle-Hook method and an ordinary linear regression analysis should be adopted.