Context: Cirrhosis impacts the coagulation cascade since the first stage of the liver disease. For years, a procoagulant imbalance, related to the severity of cirrhosis, has been demonstrated. This procoagulant imbalance is detected when coagulation assays are sensibilised to the protein C pathway. A resistance to thrombomodulin (TM) is consistently found in thrombin generation assays. TM and endothelial protein C receptor (EPCR) are two transmembrane proteins involved in the anticoagulant protein C pathway. A soluble form of EPCR (sEPCR) is release of the endothelial surface upon cleavage by mettaloprotease. Soluble EPCR have a procoagulant and proinflammatory effects: bound to PC or activated PC (APC), PC is unavailable for activation by thrombin-TM complex and APC is unable to inactivate FVa and FVIIIa. The aim of this study was to evaluate the plasma levels of sEPCR in non-complicated cirrhotic patients as a potential marker for procoagulant imbalance, compared to healthy controls.

Materials and Methods: Patients prospectively included were confirmed cirrhotic patients (prothrombin time <70% and/or liver dysmorphia and/or fibroscan > 20 kPa and/or histology and/or association of portal hypertension and liver failure). Patients were free of hepatocellular carcinoma and were not anticoagulated. Patients with on-going infection or inflammatory complication were excluded. None of them had a thromboembolic event or a familial history of thromboembolism. Controls were free of anticoagulation, coagulation disorders and without oral contraceptives. All ethical requirments were obtained and patients and controls gave their written informed consent. Kruskal-Wallis test and Dunn post test were used for statistical analysis. Results are expressed as mean (Q1 - Q3) values of sEPCR.

Results: Fifty two cirrhotic patients (24 Child-Pugh A, 21 Child-Pugh B and 7 Child-Pugh C) and 32 healthy controls were prospectively included in our hospital. The sEPCR values for healthy controls were 100 ng/mL (81 - 135 ng/mL). Comparatively to controls, Child-Pugh A patients and Child-Pugh B patients had statistically significant higher values of sEPCR : 169 ng/mL (117-225 ng/mL, p < 0.01) and 176 (135 - 230 ng/mL, p < 0.001) respectively (Figure 1). Compared to healthy controls, Child-Pugh C patients had an increased level of sEPCR : 174 ng/mL (78 - 254 ng/mL, not significant but probably underpowered). These data are in lines with the thrombin generation profiles showing a resistance to TM in cirrhotic patients increasing from Child-Pugh A to C. No patient have developed thrombosis during the follow-up period.

Conclusion: Patients with Child-Pugh A and B cirrhosis had significantly higher values of sEPCR compared to healthy controls (p < 0.01 for both). The putative role of sEPCR as a marker of the procoagulant imbalance associated to cirrhosis remains to be accurately evaluated.

Figure 1.

sEPCR levels in plasma of cirrhotic patients.

Figure 1.

sEPCR levels in plasma of cirrhotic patients.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.