Pharmacologic Prophylaxis Against Venous Thromboembolism in Hospitalized Patients With Cirrhosis and Associated Coagulopathies

Lindsey Koliscak; Lena Maynor


Am J Health Syst Pharm. 2012;69(8):658-663. 

In This Article

Abstract and Introduction


Purpose Published evidence on the incidence and predictors of venous thromboembolism (VTE) in patients with cirrhosis of the liver is reviewed.
Summary The frequently observed phenomenon of elevated International Normalized Ratio (INR) values in patients with cirrhosis has led to a theory of "autoanticoagulation" and the assertion that such patients may not benefit from the VTE risk-reduction therapies routinely used in other groups of hospitalized patients. A literature search identified six reports specifically addressing the issue of VTE risk in patients with cirrhosis. Reported rates of VTE development in such patients vary widely (0.5–8.2%) as a result of investigators' use of varying study methods and endpoints. The results of three studies (including two studies of longitudinal data on about 100,000 and nearly 450,000 patients) found no significant correlation of INR values and VTE risk. With regard to potential clinical markers of VTE risk in the context of cirrhosis, data from two studies suggested that serum albumin might serve as a reliable marker of coagulation status and, therefore, VTE risk. The results of other studies indicated that independent predictors of VTE in patients with cirrhosis include malnutrition and significant comorbidities such as chronic kidney disease and congestive heart failure. In aggregate, the available evidence does not support the autoanticoagulation theory.
Conclusion In hospitalized patients with cirrhosis who have elevated INR values, pharmacologic VTE prophylaxis should be strongly considered if there is no active or recent bleeding and if more than one risk factor for VTE is present.


Cirrhosis of the liver causes significant morbidity and mortality. In 2007, cirrhosis was the 12th leading cause of mortality in the United States, accounting for 29,165 deaths.[1] Cirrhosis is associated with multiple complications, including ascites, spontaneous bacterial peritonitis, portal hypertension, variceal bleeding, hepatic encephalopathy, and hepatorenal syndrome.[2] Patients with cirrhosis also may experience hematologic abnormalities and coagulation disturbances. Thus, the question has arisen as to whether routine pharmacologic prophylaxis of venous thromboembolism (VTE) is appropriate during the hospitalization of patients with cirrhosis-related coagulopathies.

Coagulation disturbances occur due to several complex mechanisms. Various proteins relevant to coagulation are synthesized in the liver, including albumin, proteins C and S, and clotting factors.[3,4] In patients with cirrhosis, the synthesis of clotting factors is decreased.[3–6] These clotting factors are important in the generation of thrombin, which converts fibrinogen to fibrin, an essential protein in the formation of blood clots.[4,7] The International Normalized Ratio (INR) is used to evaluate a patient's coagulation status during treatment with vitamin K antagonists. When used in this manner, the INR standardizes prothrombin time (PT) in patients receiving such therapies, and elevated INR values correspond to a higher state of anticoagulation. Patients with cirrhosis often have elevated PT and INR values in the absence of the use of pharmacologic anticoagulants. This phenomenon has led to the notion that some process of hypercoagulability, or "autoanticoagulation," in patients with liver disease might protect against the development of VTE.[3–10] Additionally, patients with cirrhosis commonly have reduced platelet counts due to increased platelet destruction and sequestration in the spleen and liver, further leading clinicians to believe these patients are at increased risk for bleeding.[11]

However, the INR is not intended as a gauge of the coagulation status of patients not receiving pharmacologic anticoagulation therapy.[4] PT (and therefore INR) is measured by sensing procoagulants, including clotting factors I, II, V, VII, and X.[4,5,7,11,12] However, the assessment of PT does not reflect levels of endogenous anticoagulants (proteins C and S and antithrombin), factor VIII, and von Willebrand factor.[4,7]

During clot formation, protein C is activated by a thrombin–thrombomodulin complex, which (with protein S) provides negative feedback, resulting in reduced thrombin production. Antithrombin also serves to reduce thrombin formation.[4] Levels of two procoagulants, factor VIII and von Willebrand factor, are actually increased in patients with cirrhosis; as von Willebrand factor promotes platelet migration and attachment to the site of injury, this increase could potentially offset the anticoagulative effects of reduced platelet counts and enable patients to have efficient clotting.[4,11,13] Von Willebrand factor may also be involved in maintaining elevated factor VIII levels by indirectly reducing the breakdown of factor VIII, allowing it to maintain its procoagulative properties.[4]

Thus, although the reduced production of most clotting factors in patients with cirrhosis increases the risk of bleeding, the reduced production of proteins C and S and antithrombin, coupled with increased von Willebrand factor and factor VIII levels, increases the risk of coagulopathy. Taking all of these points into consideration, the balance between endogenous procoagulants and anticoagulants is the crucial factor. If that balance is disrupted, unfavorable bleeding or clotting could occur. An in-depth review of coagulopathy in the context of cirrhosis can be found elsewhere.[4]

It has been suggested that measuring thrombin generation (rather than INR) may serve as a reliable test for the assessment of bleeding or clotting risk among patients with cirrhosis.[4,7,11] Some research has focused on thrombin generation as a more balanced marker of coagulation status in patients with cirrhosis. In a study by Tripodi et al.,[11] no difference in thrombin generation was noted in patients with cirrhosis relative to controls. Gatt et al.[7] found an increased velocity of thrombin generation in patients with cirrhosis (especially those with a mean INR of 1.2–2.0) compared with controls; this actually suggests that a hypercoagulable state may be present in patients with cirrhosis.

For the reasons mentioned above, pharmacologic VTE prophylaxis in hospitalized patients with cirrhosis is a topic of debate. The development of VTE in hospitalized patients is known to extend their length of stay, increase health care costs, and increase mortality, and both pharmacologic prevention and mechanical prevention have been shown to reduce their risk of developing VTE.[14] Well-documented risk factors for VTE development include surgery, trauma, immobility, cancer, increasing age, obesity, and acute medical illness, and many hospitalized patients have at least one of these risk factors. The use of low-molecular-weight heparin (LMWH), low-dose unfractionated heparin, or fondaparinux is recommended for VTE prophylaxis in acutely ill nonsurgical patients. However, clinical practice guidelines do not specifically address the issue of VTE prophylaxis in patients with cirrhosis and elevated INR values.

There are limited data on the risk of VTE in patients with cirrhosis. The objectives of this article are to evaluate available published studies and provide a recommendation regarding the use of pharmacologic thromboprophylaxis in patients with cirrhosis and associated coagulopathies.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.