Detection of Abnormally High DOAC Plasma Concentrations using PT/INR in three different reagents in Frail Older Patients: the DOAC-FRAIL study.

BACKGROUND DOACs are prescribed for the prevention and treatment of thromboembolism. Frail older patients represent a vulnerable population with altered pharmacokinetics. This may result in elevated DOAC plasma concentrations with an increased bleeding risk. DOAC plasma concentrations are not routinely measured. PT and INR are widely available, but diagnostic value for detecting elevated DOAC plasma concentrations is uncertain and may be reagent-dependent. METHODS We evaluated the correlation between INR and DOAC plasma concentrations using spiking- and plasma samples from acutely admitted frail older patients. INR was measured using three different reagents (Neoplastin, RecombiPlasTin and Innovin). Diagnostic performance was assessed using predefined INR thresholds. Sensitivity, specificity, predictive values, and diagnostic accuracy were calculated. RESULTS Neoplastin demonstrated the strongest correlation with rivaroxaban plasma concentrations. An INR cut-off ≥1.8 showed a sensitivity of 100% and specificity of 66.7% for detecting rivaroxaban concentrations ≥300 ng/mL (PPV 82.4%, NPV 100%). RecombiPlasTin showed high sensitivity but lower specificity. Innovin demonstrated low sensitivity but high specificity. For apixaban, RecombiPlasTin was evaluated due to lower linearity of the other reagents. An INR cut-off ≥1.6 yielded a sensitivity of 62.5% and specificity of 75.0% for detecting apixaban concentrations ≥300 ng/mL (PPV 33.3%, NPV 90.9%). Sensitivity increased for apixaban concentrations ≥350 ng/mL, reaching 100% across all INR thresholds. CONCLUSION The INR performance for detecting elevated DOAC plasma concentrations varies substantially between reagents and DOAC-type. Routine coagulation assays may offer a screening tool for detecting elevated DOAC plasma concentrations in frail older patients when specific DOAC measurements are not available.