Abstract Backgrounds: Rivaroxaban, a direct oral anticoagulant (DOAC) targeting Factor Xa, is widely used without routine monitoring. However, the absence of reliable drug level measurement may hinder clinical decision-making. The study evaluated the correlation between a commercial rivaroxaban-calibrated anti-Xa assay (Riva anti-Xa) and plasma drug levels measured by HPLC-MS/MS (Riva LC-MS), and assess its diagnostic accuracy for clinically relevant thresholds. Methods A spiking experiment and a prospective multi-center study were conducted, using Riva LC-MS as the reference. Correlation and agreement were assessed via Spearman analysis, Bland-Altman plots, and Deming regression. Sensitivity and specificity were calculated of 30, 50, and 100 ng/mL. Results The Riva anti-Xa assay showed excellent correlation with Riva LC-MS (Spearman r = 0.982, 95% CI: 0.970–0.988) and maintained acceptable bias in the presence of interfering substances. Clinical samples showed high concordance with LC-MS results. For concentrations ≥ 30, 50, and 100 ng/mL, the assay yielded sensitivities of 100.0%, 100.0%, and 92.74%, and specificities of 99.58%, 99.49%, and 99.30%, respectively. ROC-AUC values were 0.99 across all thresholds. Conclusion The Riva anti-Xa assay is a reliable, rapid alternative for estimating rivaroxaban levels, with strong correlation to LC-MS and excellent diagnostic accuracy, supporting its use in urgent clinical settings.
Zhao et al. (Thu,) studied this question.