Splanchnic and Extrasplanchnic Thrombosis in Cirrhosis: Prophylaxis vs Treatment

Venous thromboembolism (deep vein thrombosis and pulmonary embolism) and portal vein thrombosis (PVT) occur in up to 6.3 % and 15.9 % of patients with cirrhosis, respectively. There is recent evidence that a procoagulable prothrombotic state is related to cirrhosis despite the reduced levels of many coagulation factors, and decreased platelet counts. Indeed, (i) the combination of high levels of factor VIII, with low levels of protein C and antithrombin induces a procoagulant state in vitro; while (ii) increased levels of von Willebrand factor and decreased ADAMTS 13 activity can compensate for decreased platelet counts. PVT is partial in a majority of patients in whom it develops and may spontaneously resolve in some of them. Although PVT is associated with features of more severe liver disease, it is uncertain whether it plays a causal role in the decompensation of cirrhosis. In patients listed for liver transplantation, PVT may make the procedure difficult or impossible. Pre-transplant PVT is associated with increased post-transplant mortality rates. Studies evaluating clinical outcome of anticoagulation therapy for splanchnic or extrasplanchnic venous thrombosis are scarce. Anticoagulation therapy, given to patients with cirrhosis of intermediate severity before PVT occurrence, in prophylactic doses, appears to decrease decompensation and mortality rate. Interestingly, this improvement is out of proportion of the prophylaxis of extrahepatic portal vein thrombosis. The risk of bleeding does not seem to be increased in patients with cirrhosis receiving anticoagulation therapy, once prophylaxis for bleeding related to portal hypertension has been implemented. Overall, the room for anticoagulation therapy is probably larger than previously recognized, and may be of particular benefit in patients without portal vein thrombosis. However, clinical trials remain to be done before the benefit risk ratio of anticoagulation therapy is properly evaluated.