Point-of-care testing of coagulation in patients treated with edoxaban


The first systematic evaluation of coagulation point-of-care testing in edoxaban treated patients has been carried out. The results were presented by Florian Härtig, University of Tuebingen, Germany, at the International Stroke Conference (ISC; 24–26 January 2018, Los Angeles, USA)

Currently there is no means of measuring the anti-coagulant effect of edoxaban in an emergency setting, meaning that thrombolytic drugs cannot be given.

In the case of vitamin K antagonists like warfarin, blood is taken from the patient and applied to a point-of-care test device, and an INR (international normalised ratio) reading is given. An INR of 1.7 is endorsed in the AHA/ASA guidelines from 2014, and according to this threshold thrombolysis can be applied or antidotes can be given in cases of bleeding. For direct oral anticoagulents (DOACs) it is not known which test should be used and which threshold applies. Because of this it is not known what device should be used for emergency coagulation testing in the cases of DOAC treatment.

Härtig and his group have looked at the different coagulation tests and shown that it is possible to get some idea of DOAC levels from nonspecific coagulation tests like PT/ INR (prothrombin time), and PTT (activated partial thromboplastin time).

“This led us to two small studies we conducted in recent years, one of which was published in Stroke in 2015, where we used the ROCHE CoaguChek INR test device to see whether we can use readings from patients who have been treated with rivaroxaban,” said Härtig. “We showed that even though the correlation is not perfect there is increased INR ratings with increasing levels of rivaroxaban in the blood sample.”

For point-of-care emergency evaluation of coagulation in patients (POCT) with suspected stroke, this research led to different assessment paths. People on warfarin will have a normal point-of-care evaluation and based on that they will receive thrombolysis or in case of bleeding complications receive four-factor concentrate. For rivaroxaban the same can be done within limits, and with dabigatran the Hemochron device can be used to see whether thrombolysis should be performed or in cases of bleeding apply praxbind. There have been some suggestions that this could be bypassed completely and praxbind should be given and then thrombolysis, but this is not a well evaluated procedure. For apixaban no correlation with any point of care tests has been found.

To look at edoxaban, a DOAC that has been available for a relatively short period of time, the group designed a study that included patients that were started on edoxaban but had not taken any other oral anticoagulants or heparin. Patients with abnormal baseline coagulation were excluded. Blood samples were taken just before the dose and then at 30 minutes, one, two, eight and 24 hours after dose and all the point of care tests including CoaguChek INR, Hemochron-INR, HC-aPTT (Hemochron- activated partial thromboplastin time), HC-ACT+ (Hemochron activated clotting time) and HC-ACT-LR were performed. Mass-spectrometry was carried out at every time point to find out edoxaban plasma concentrations.

“We found that if you perform Hemochron-INR you have very good linear correlation between INR reading and edoxaban plasma concentration,” said Härtig. “It was not so good for the ACT+, ACT-LR (low range activated clotting time), and the aPTT. With the CoaguChek INR we found very similar readings to the first CoaguChek trail with rivaroxaban”.

The group also calculated receiver operating characteristic and the HC-INR was very good, meaning it was possible to retrospectively define a cut off INR of 1.5 with the Hemochron device, which differs from the INR of the CoaguChek. This was a retrospectively defined cut off so a prospective trial needs to be carried out to confirm these findings.

(Visited 36 times, 1 visits today)


Please enter your comment!
Please enter your name here

This site uses Akismet to reduce spam. Learn how your comment data is processed.