Comparing Anticoagulant Effect of Edoxaban in Patients With Cirrhosis Against Healthy Controls

Edoxaban, an anti-factor Xa targeting direct oral anticoagulant (DOAC), is a less effective inhibitor of coagulation in patients with cirrhosis than in healthy individuals, despite similar drug plasma levels between the two groups, according to a study published in Blood. The FDA-approved edoxaban package insert current cautions against using the drug in patients with moderate or severe hepatic impairment (Child-Pugh class B or C) but does not recommend a dose alteration in patients with mild (Child-Pugh class A) hepatic impairment.

Study author Ton Lisman, PhD, of the University Medical Center Groningen in the Netherlands, told ASH Clinical News that the new findings suggest standard dosing of edoxaban could result in undertreatment of patients with cirrhosis. “The undesirable effect of undertreatment would be suboptimal prevention or treatment of thrombosis,” said Dr. Lisman. “In general, anticoagulants are given to treat or prevent thrombosis, but too little of a dose does not offer enough protection against thrombosis or results in ineffective treatment.” He explained that dose increases in patients with cirrhosis may be required for optimal thrombosis management, but additional research is needed, since too high a dose will result in a high risk for bleeding.

A total of 16 patients with a confirmed cirrhosis diagnosis from an outpatient hepatology clinic of an academic medical center in the Netherlands were enrolled in the study. Another 16 healthy individuals were enrolled as a control group. Treatment consisted of edoxaban 60 mg administered once daily for 7 consecutive days. Levels of edoxaban were examined in blood samples taken twice on day 1 and once on days 3 and 7, approximately 2 hours following ingestion of the drug.

The median ages in the patient and control groups were 60 years and 49 years, respectively. Approximately 38% of patients with cirrhosis and 44% of patients in the control group were women. Patients with cirrhosis were classified based on the etiology of their liver disease:

  • cholestatic liver disease (n=4)
  • nonalcoholic steatohepatitis (n=2)
  • alcoholic steatohepatitis (n=4)
  • autoimmune hepatitis (n=2)
  • hepatitis C virus (n=1)
  • hemochromatosis (n=1)
  • overlap syndromes (n=2)

A higher proportion of patients with cirrhosis had diabetes (25% vs. 6%) and cardiovascular disease (31% vs. 13%). Patients with cirrhosis also had a higher median D-dimer value at baseline compared with controls (517 vs. 185 ng/ml, respectively). Compared with baseline, the D-dimer levels in controls decreased significantly over time (p<0.05). The D-dimer levels in patients with cirrhosis remained relatively unchanged during the study.

The baseline endogenous thrombin potential (ETP) – an in vitro measure of coagulability – was higher in patients with cirrhosis compared with controls (707 nM IIa*min vs. 403 nM IIa*min, respectively). Although the ETP decreased in patients and controls on days 1, 3, and 7, it remained higher in patients with cirrhosis versus healthy volunteers.

The anti-Xa levels were consistent during the 1-week treatment period in both patients and healthy participants. Overall, the median plasma levels of anti-Xa were approximately 200 ng/mL. At day 3, a total of 2 patients had low plasma levels of edoxaban, which the investigators suggest was attributable to low compliance. A minor nosebleed was reported in 2 patients, and 1 of these patients decided to discontinue from the study following this adverse event. Mild bruising was reported in 4 patients and 5 healthy volunteers.

“Anticoagulant treatment in patients with cirrhosis is complex because the liver is the site of synthesis of most proteins involved in hemostasis, and patients with cirrhosis acquire complex hemostatic disorders,” said Dr. Lisman. He added that patients with mild cirrhosis can ultimately experience both bleeding and thrombotic complications, but treatment and prevention of thrombosis in cirrhosis with traditional drugs remains a challenge. “DOACs are a class of drugs that lack the ‘liver disease-specific’ disadvantages of vitamin K antagonists and heparins,” Dr. Lisman said, “but a potential issue of DOACs in patients with liver disease is clearance, with accumulation of drug being a concern.” Most DOACs have predominantly hepatic clearance; estimated hepatic clearance for apixaban is about 75%, that for rivaroxaban 65%, edoxaban 50%, and dabigatran 20%.

The study was limited by its enrollment of a small number of patients and inclusion of only patients with mild hepatic disease. Dr. Lisman suggests future studies should examine other types of DOACs, as well as edoxaban, in a larger and sicker patient population. “Our study was not meant to change clinical practice but to inform future clinical studies,” he said. “The findings are unique in that we are the first ones to study drug levels and anticoagulant effects after more than 1 dose of DOAC.”

Several study authors report relationships with Daiichi Sankyo, which sponsored the trial.


Bos S, Schreuder T, Blokzijl H, et al. Anticoagulant activity of edoxaban in patients with cirrhosis. Blood. 2020 May 26. [Epub ahead of print]