Resolving Interference of Direct Oral Anticoagulants (DOACs) on Routine Coagulation, Thrombophilia, and Lupus Anticoagulant Testing Using Activated Carbon

Background: Direct (anti-IIa and anti-Xa) oral anticoagulants (DOACs) have a significant impact on various coagulation test results. As the correct interpretation of these tests is mandatory to prevent misclassification and subsequent clinical consequences, withholding treatment could be necessary, with the associated risk of thrombosis.

Aims: To evaluate the performance of the activated carbon DOAC-Remove (5-Diagnostics, Basel, Switzerland) in extracting DOACs from plasma samples and its effect on various routine and esoteric coagulation test results.

Patients, Materials and Methods: Left-over plasmas from patients treated or not with DOAC obtained in the routine laboratory workload were evaluated. Briefly, 0.8 mL of plasma sample was incubated with 1 tablet of the activated carbon for 10 min at room temperature using a rotating shaker. After a 2 min-centrifugation at 2000x g and room temperature, the supernatant was pipetted before being analyzed. Tests were performed before and after such incubation. DOACs were measured using either a specific direct thrombin inhibitor or an anti-Xa assay with specific calibrations. Routine coagulation tests (PT, aPTT, and factor(F)V, FVIII:C, FIX, fibrinogen, and D-dimer) were performed, as well as thrombophilia panel [antithrombin, protein C (PC, chromogenic and clotting assays), PS (free PS antigen and clotting assays)], and lupus anticoagulant (LA) panel [silica clotting time (SCT), and dilute Russell venom clotting time (dRVVT)]. All assays were performed using reagents from Werfen (Bedford, MA, USA) on the ACL TOP 700 analyzer. As the distribution of data was not normal, test results were compared using non-parametric tests.

Results: We evaluated a total of 756 plasma samples, obtained from patients treated with dabigatran (n=139, median concentration 129 ng/mL [range:18-905]), rivaroxaban (n=157, median concentration 159 ng/mL [range:19-815]), or apixaban (n=155, median concentration=154 ng/mL (range:11-510), and from 305 patients with various disease states and not on DOAC including 35 patients on coumadin and 32 patients on heparin (UFH: n=18 ; LMWH: n=14).

In untreated patients, the DOAC-Remove had no significant impact on test results (n>30 for each parameter), except for D-dimer, free PS and FVIII:C. However, the mean biases, evaluated according to Bland-Altman, were below the accepted limits, and changes would not have had any clinical relevance.

In the plasma from patients treated on DOAC, the DOAC-Remove eliminated all the three DOACs, with levels far below the detection limit of the techniques after a 10 min-incubation, leading to a dramatic correction of the DOAC-induced prolongations of PT and APTT, as well as of their lowering effect on FV, FVIII and FIX activities. The same applied to the DOAC-induced elevation of PC and PS anticoagulant activities (clotting assays), and antithrombin activity in the plasma from patients on rivaroxaban and apixaban.

DRVVT screen/confirm ratios, which were above the normal ranges in 45% of the patients on dabigatran (n=31), and 78% of the patients on rivaroxaban (n=32), were normalized in most samples after a 10 min-incubation with the carbon and remained positive in only 10% of the patients on dabigatran and 12% of the patients on rivaroxaban. The impact of the DOACs on the SCT screen/confirm ratio was less noticeable with baseline positive test results in only one of the samples from patients on dabigatran (n=31), and in 2 patients on rivaroxaban (n=32). Treatment with the carbon leaded to a correction in the positive sample from patient on dabigatran and in one of the two positive samples from patients on rivaroxaban. Both dRVVT and SCT test results were within the normal range in the plasma from 32 patients on apixaban before and after treatment with the activated carbon.

As expected, fibrinogen, PC activity evaluated using a chromogenic assay, free PS antigen concentration, VWF:RCo and VWF:Ag were not affected by any of the 3 DOACs, and test results were not significantly different before and after incubation with the activated carbon.

Conclusion: The DOAC-Remove had no effect on test results obtained in the plasma from untreated patients and from patients on traditional anticoagulants. It effectively removed all 3 tested DOACs from the plasma of treated patients, allowing an accurate measurement of routine and esoteric coagulation tests in DOAC treated patients without withholding the treatment.