The French Working Group on Perioperative Haemostasis (GIHP) published in March 2013 proposals for the management of major bleeding complications and emergency surgery in patients treated with direct oral anticoagulants, and especially the Rivaroxaban. This Direct Oral Anticoagulant (DOA) induces most of clotting tests, including Prothrombin Time (PT) and APTT. This led the GIHP to a degraded proposition for the institutions who does not propose the specific dosage of the molecule. This proposal assumes that the normality of the PT and aPTT exclude most patients with Rivaroxaban level >30 ng/ml. The aim of our evaluation is to verify the real-life practicability of this proposal.

Samples from Rivaroxaban treated patients have been tested with 4 PT reagents: STA Neoplastine R (NeoR), STA Neoplastine CI+, Recombiplastine and Innovin. Four APTT reagents were also tested: STA-PTT automate, STA Cephascreen, Synthasil, Pathrontin SL. All reagents were tested on their specific analysers (STA-R, ACL TOP500, BCS XP). PT reagents were calibrated following manufacturer’s instructions (Stago, Instrumentation Laboratory, Siemens).

Rivaroxaban prolongs PT but reagent’s sensitivity varies considerably and we observed a major individual variability, and this even at low Rivaroxaban concentrations. The NeoR presents the greatest sensitivity, the Innovin the lowest. For a normal PT (>70%), the NeoR alone can validly identify patients with Rivaroxaban level > 30ng/ml (sensitivity 93%). With the other reagents, from 42 to 51% of our patients with normal PT has Rivaroxaban level > 30 ng/ml. Two patients with Rivaroxaban > 150 ng/ml had normal PT with these 3 reagents. None of the APTT reagents has sufficient sensitivity to validly exclude these patients.

In conclusion, none of the PT reagents can estimate Rivaroxaban concentration. The NeoR is the only reagent that allows distinguishing patients with residual Rivaroxaban level. For all other situations, specific test seems mandatory and the development of point-of-care system should be encouraged.

Disclosures

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution