One of the most common, and feared, bleeding complications in liver disease is bleeding esophageal varices; however, variceal bleeding is largely related to local vascular abnormalities, including vessel radius, thickness, and pressure, rather than hemostatic disturbances. Vessel pressure is predominantly dictated by splanchnic blood pressure, which is often increased due to hypervolemia, a common problem in liver disease [11]. Other features of bleeding in liver dysfunction include ecchymosis, epistaxis, oral mucosal bleeding, and gastrointestinal bleeding. Further, bleeding can be precipitated by invasive procedures.

Previously, it was assumed liver disease provided protection against thrombosis given the prolonged PT on routine laboratory tests. This now appears to be incorrect. The rate of deep venous thrombosis and pulmonary embolism is anywhere between 0.5 and 8.1% [12]. A more common complication is portal venous thrombosis (PVT), which has a reported prevalence of 11–36% [13]. Portal venous stasis appears to be the major change in liver disease contributing to the increased risk for PVT [14]. Hypercoagulability (i.e., increased FVIII, decreased protein C, etc.) likely plays a role in clotting when the hemostatic balance is tipped in the favor of thrombosis. The prevention and treatment bleeding and thrombotic complications in chronic liver disease will be discussed later.

No one test can accurately predict the risk of bleeding in liver disease (Table 2.2). Two of the most commonly used tests are the PT and aPTT, which measure the time to formation of a fibrin clot. While inexpensive and widely available, both gauge just one aspect of coagulation and are not predictive of bleeding in chronic liver disease [15]. Similarly, obtaining a platelet count is another common test to evaluate bleeding risk. A platelet count less than 50,000/μL confers an increased risk of bleeding with invasive procedures in liver disease; however, higher platelet counts do not appear to predict bleeding risk [16]. Various other less commonly used laboratory tests are employed. Fibrinogen is a measure of the fibrinolytic system and decreased levels are indicative of fibrinolysis; however, fibrinogen is not correlated with bleeding risk in liver disease [16].

A major disadvantage of the above tests is their inability to assess more than a single aspect of the hemostatic system, which is less than ideal in chronic liver disease, a disorder with multiple perturbations of hemostasis. Global tests of hemostasis, such as thromboelastography (TEG), are methods of measuring whole-blood coagulation. TEG is often used perioperatively by surgeons and anesthesiologists. Given the numerous abnormalities present in liver disease, many acting in opposition to one another, tests such as TEG, may provide a more accurate assessment of bleeding risk. Indeed, TEG has been shown to be useful in detecting coagulopathy in liver disease [17]. Another global measure of hemostasis, the thrombin generation assay measures thrombin production and may be beneficial when evaluating coagulopathy in liver dysfunction. Thrombin generation is often normal or increased in liver disease, which highlights the concept of rebalanced hemostasis previously mentioned [18]. Thrombin generation assays are still experimental and may provide a more accurate measure of bleeding and thrombotic risk in chronic liver disease but further study is needed.

A variety of options are available for the treatment and prevention of bleeding in chronic liver disease (Table 2.3). Prevention of bleeding is a concern in certain high-risk patients and prior to invasive procedures.

One of the most commonly encountered bleeding complications experienced in liver disease is esophageal variceal bleeding. As previously mentioned, the etiology of variceal bleeding is related to local vascular abnormalities, such as splanchnic blood pressure, rather than abnormalities of hemostasis. Thus, treatment is not necessarily directed at correcting hemostatic abnormalities; however, as is the case with all potentially life-threatening bleeding events, volume resuscitation with red blood cells is critical. The goal hemoglobin concentration is 7–8 mg/dL [19]. It is important to avoid excessive transfusion since excess volume can increase splanchic portal pressure and further exacerbate bleeding. The key treatment modality in acute variceal bleeding is endoscopic variceal banding or ligation. While not the mainstay of treatment, correction of hemostatic defects is frequently attempted prior to invasive procedures, such as endoscopic therapy, to prevent worsened bleeding. Other potential bleeding complications that may arise in chronic liver disease include portal hypertensive gastropathy or gastric vascular ectasia-related bleeding and bleeding associated with invasive procedures. Commonly performed invasive procedures in liver dysfunction include percutaneous or transjugular liver biopsy, abdominal paracentesis, and accessing vascular sites (i.e., central venous catheter placement), among others.

Various blood products and hemostatic agents are administered for the treatment and prevention of bleeding in liver disease. Red blood transfusions to replace blood loss have already been discussed. The others are aimed at improving underlying hemostatic defects. Fresh frozen plasma (FFP) contains both pro- and anticoagulation factors and can be administered to replace deficiencies of either. FFP is most commonly administered to correct a prolonged PT. The efficacy of FFP to prevent bleeding has never been demonstrated [20]. Moreover, the volume of FFP necessary to correct coagulation factor deficiencies is large – 20–40 mL/kg – and complete correction is seldom accomplished [21, 22]. Potential adverse effects include pulmonary edema and increased portal venous blood pressure, among others. Therefore, in chronic liver disease, FFP is not recommend for the prevention of bleeding in patients with a prolonged PT prior to invasive procedures, and its use in actively bleeding patients is questionable. Platelet transfusions are often administered for thrombocytopenia. Adequate thrombin production occurs with a platelet count greater than 50,000/μL. Transfusion to obtain this value is warranted in active bleeding and should be considered for prophylaxis prior to invasive procedures [23–25]. In some instances, it may be difficult to achieve a platelet count of 50,000/μL or greater due to splenic sequestration of platelets in portal hypertension-induced splenomegaly, often present in liver disease. Cryoprecipitate, which contains fibrinogen and coagulation factors V and VIII, should be administered in bleeding patients with hypofibrinogenemia until fibrinogen levels normalize [26]. Its use as a prophylactic agent to prevent hemorrhage is not well studied. Similarly, when hyperfibrinolysis is a concern, antifibrinolytic agents, such as tranexamic acid, may be used. Last, recombinant factor VIIa (rFVIIa) and prothrombin complex concentrates (PCC) represent low-volume prohemostatic alternatives to FFP. Recombinant FVIIa has not been shown to be beneficial in bleeding esophageal varices or with prophylactic use prior to liver transplantation [27]. Therefore, routine use is not recommended except during very high-risk procedures, such as intracranial pressure monitor placement and rescue therapy for refractory, life-threatening bleeding. There are limited data regarding the use of PCCs in similar situations, so it cannot be recommended for routine use either. Adverse effects of both therapies include thrombogenicity, high expense, and need for frequent therapy.