Clin Appl Thromb Hemost. 2026 Jan-Dec;32:10760296261469286. doi: 10.1177/10760296261469286. Epub 2026 Jul 14.
ABSTRACT
BackgroundRapid assessment of rivaroxaban exposure is clinically critical for older patients with atrial fibrillation in scenarios such as acute bleeding, urgent invasive procedures, or thrombolysis decision-making. This study evaluated rivaroxaban-calibrated anti-factor Xa activity (anti-FXa activity), prothrombin time (PT), and activated partial thromboplastin time (APTT) for identifying clinically relevant rivaroxaban concentrations in older atrial fibrillation patients.MethodsIn this prospective multicenter diagnostic accuracy study, older patients with atrial fibrillation receiving rivaroxaban were enrolled. Liquid chromatography-tandem mass spectrometry (LC-MS/MS) was used as the reference standard, with ≥50 ng/mL defined as the primary diagnostic threshold. Anti-FXa activity, PT, and APTT were evaluated as index tests. Diagnostic performance was assessed using receiver-operating-characteristic analysis, Youden-derived cut-offs, and patient-level clustered bootstrap resampling.ResultsA total of 183 patients were included, contributing 350 LC-MS/MS samples. Anti-FXa activity showed the highest discrimination for identifying LC-MS/MS-measured rivaroxaban concentrations ≥50 ng/mL, with an area under the receiver-operating characteristic curve (AUC) of 0.980 (95% confidence interval [CI], 0.960-0.993), followed by PT (0.904; 95% CI, 0.869-0.938) and APTT (0.825; 95% CI, 0.779-0.869). At Youden-derived cut-offs, anti-FXa activity had sensitivity of 94.0% and specificity of 96.1%, compared with 84.8% and 85.4% for PT, and 74.4% and 81.0% for APTT. Spearman correlations with LC-MS/MS were strongest for anti-FXa activity, followed by PT and APTT.ConclusionIn this cohort of older patients with AF, anti-factor Xa activity showed the highest diagnostic accuracy for identifying rivaroxaban concentrations ≥50 ng/mL, whereas PT provided moderate discrimination and APTT was less informative.
PMID:42446159 | DOI:10.1177/10760296261469286