Liver diseases have traditionally been associated with bleeding disorders; however, current evidence suggests a significant risk of thrombotic complications.1 Historically, cirrhotic patients were thought to be “auto-anticoagulated” due to deficiencies in coagulation factors and abnormalities in conventional laboratory tests.2 Nonetheless, several arguments challenge this assumption.

The balance between procoagulant and anticoagulant factors in these patients is significantly altered, creating a complex hemostatic environment,3,4 which affects both the intrinsic and extrinsic coagulation pathways.

Both primary and secondary hemostasis are compromised by the decreased hepatic synthesis of several coagulation factors—including factors II, V, VII, IX, X, and XI. Despite this reduction, levels of von Willebrand factor (vWF) and factor VIII are typically elevated, contributing to a procoagulant state.2

These changes can lead to a condition in which the risk of bleeding or thrombosis increases depending on individual circumstances,2,4 giving rise to the concept of "rebalanced coagulation," which suggests that the decrease in procoagulant factors is offset by a concomitant reduction in anticoagulant proteins such as protein C, protein S, antithrombin, heparin cofactor II, and α2-macroglobulin, all synthesized in the liver and found to be decreased in patients with liver disease.5

This rebalancing is further affected by thrombocytopenia and platelet function defects, which may be compensated by elevated vWF levels (300% in stable cirrhosis, 360% in acute decompensated cirrhosis, and up to 700% in acute-on-chronic liver failure) and decreased levels of its regulator, ADAMTS13 (which can drop to 90% below normal), impacting the overall coagulation profile.2,6

Thrombocytopenia is multifactorial: firstly, due to splenic sequestration (a congestive spleen related to portal hypertension); however, not all cirrhotic patients with thrombocytopenia exhibit splenomegaly, and reducing portal pressure does not always normalize platelet counts.6 Secondly, due to "hypersplenism," resulting from increased immunomediated macrophage activity leading to platelet destruction, especially relevant in patients with hepatitis C and primary biliary cirrhosis.3 Lastly, due to insufficient hepatic production of thrombopoietin (TPO), whose levels vary depending on the stage of liver disease. Additionally, bone marrow suppression caused by antiviral therapy, alcohol, or folate deficiency also contributes to thrombocytopenia in these patients. Regardless of cause or disease severity, this thrombocytopenia is compensated by an increase in the release of reticulated platelets (RePLT), which have greater prothrombotic and procoagulant potential than mature platelets, sometimes exceeding 2%, a value that appears to predict higher risk of hepatic decompensation.7

Changes in platelet function and vWF are useful for stratifying the prognosis of cirrhosis, with increased platelet aggregation associated with a higher risk of portal vein thrombosis and hepatic decompensation.8

Regarding fibrinolysis, all the proteins involved are synthesized by the liver, except for tissue plasminogen activator (tPA) and plasminogen activator inhibitor-1 (PAI-1). Cirrhotic patients show reduced levels of plasminogen, plasmin inhibitor, thrombin-activatable fibrinolysis inhibitor (TAFI), and factor XIII in both acute and chronic liver failure. In contrast, plasma levels of profibrinolytic and antifibrinolytic proteins are often elevated, likely due to increased endothelial cell activation or reduced clearance.5 Consequently, these patients may exhibit either hyperfibrinolysis or hypofibrinolysis, with sepsis being one of the main causes of the latter.9,10

These antithrombotic changes, "compensated" by prothrombotic alterations, occur simultaneously, resulting in a "new" hemostatic state characterized by a delicate rebalancing that displays both hypo- and hypercoagulable phenomena (Fig. 1), which is extremely fragile and may become destabilized by situations such as infections, renal failure, or acute episodes of decompensation.6,10,11

Figure 1.

Coagulation abnormalities and their rebalancing.

In addition to all of the above, there are special situations that pose a greater thrombotic risk and may be common in these patients, such as malignancy (hepatocellular carcinoma), sedentary lifestyle (ascites, sarcopenia, encephalopathy), advanced age, or elevated estrogen levels.

Coagulation test results vary and are intrinsically linked to the degree of hepatic dysfunction. Patients may develop either a hypocoagulable or hypercoagulable state.12 Therefore, constant monitoring and evaluation of coagulation is essential in the clinical management of these patients, as the liver plays a central role in coagulation factor synthesis, and its compromised function can impair the body’s ability to maintain adequate hemostasis. This makes it necessary to assess coagulation not only with classical tests (Table 1) but also with advanced tests such as viscoelastic testing.13

Advanced coagulation tests: viscoelastic testing13

Viscoelastic tests (VET)—thromboelastography (TEG) and rotational Thromboelastometry (ROTEM)—are specialized tools that provide a more detailed and accurate view of the hemostatic and thrombotic systems. These analyses are especially useful when classical coagulation tests do not fully explain the patient's clinical picture. In fact, one of their main advantages is their ability to integrate coagulation testing with platelet function analysis, offering a more global understanding of hemostatic physiology. They are dynamic tests that assess clot formation and stability, as well as fibrinolysis, analyzing various aspects of coagulation, platelet function, and the fibrinolytic system. Although not useful in predicting bleeding or reducing mortality in cirrhotic patients, their pre- or perioperative application may reduce the need for plasma and platelet transfusions.12,13

The procedure involves introducing a whole blood sample into a cuvette containing a cylindrical pin, with a processor analyzing the rotational movements between them. Coagulation and lysis cause torsional changes that are analyzed by the processor and reflected in a graph (Table 2, Fig. 2, and Supplementary data).14–16

Table 2.

Types of viscoelastic tests, characteristics, and reference values.14–17,20

Viscoelastic Test		What It Measures	Interpretation	Normal Values
TEG Thromboelastography	R (Reaction time)	Time until the first fibrin strands form (2 mm amplitude).	• Prolonged: deficiency in fibrin formation, platelet activity, clotting factors, or presence of anticoagulants. • Shortened: hypercoagulability.	4.5–7.5 min
Measures clot formation and stabilization	K (Coagulation constant)	Time until the clot reaches 20 mm amplitude.	• Prolonged: impaired fibrin formation. • Shortened: hypercoagulability.	1–3.5 min
	α Angle	Clot formation velocity, which ranges from R to the point where the formation rate slows and the curve begins	• Decreased: slow clot formation due to platelet or fibrinogen deficiency/dysfunction. • Increased: rapid fibrin formation (platelet hypercoagulability).	50–80°
	A10	Clot amplitude at 10 min after R.	Similar to MA, but preferred in acute patients.	> 50 mm
	MA (Maximum Amplitude)	Maximal clot strength, mostly platelet function.	• Low: weak clot due to thrombocytopenia or platelet dysfunction. • High: hypercoagulability.	50–73 mm
	LY30	% of clot lysis 30 min after MA.	• Increased: hyperfibrinolysis. • Decreased: hypofibrinolysis or prothrombotic state.	15%
ROTEM Rotational Thromboelastometry - This measures the clot's viscosity as it forms and stabilizes. -It uses a panel of reagents that activate different phases of the coagulation cascade, evaluating how these reagents affect clot dynamics. Parameters: CT: Clotting time (equivalent to R) CFT: Clot formation time (equivalent to K) Angle (α) A10: Amplitude 10 min after CFT MCF: Maximum clot firmness (equivalent to MA) LI30: Lysis index at 30 min from MCF (equivalent to LY30) ML: Maximum lysis Its combined use aids in differential diagnosis.	EXTEM -Extrinsic pathway	-Reagent: Tissue Factor (TF). -Similar to PT. -Evaluates clot formation activated by the extrinsic pathway.	-Prolonged EXTEM: deficiency in extrinsic pathway (e.g., factor VII deficiency).	CT: 38–80 s A10: 43–65 mm MCF: 50–75 mm ML: < 15% of MCF
	INTEM -Intrinsic pathway, similar to aPTT FIBTEM -Uses platelet blockers APTEM HEPTEM -Heparin neutralization	-Reagent: Ellagic acid. -Similar to aPTT. -Evaluates coagulation intrinsic pathway. - Reagents: TF + platelet inhibitor (cytochalasin D or abciximab). -Specifically measures the contribution of fibrinogen in clot formation. Reagents: TF + tranexamic acid or aprotinin. -Fibrinolysis inhibitor test -Reagents: ellagic acid + heparinase -Detects heparin-induced coagulation changes	-Prolonged INTEM: Suggests a deficiency in intrinsic factors. -Low FIBTEM: Indicates fibrinogen deficiency. -Improvement in APTEM parameters vs EXTEM points to hyperfibrinolysis. -Prolonged HEPTEM: Indicates an impairment in clot formation, which can be due to: -Fibrinogen deficiency -Platelet dysfunction -Coagulation factor deficiency	CT: 100–240 s A10: 44–66 mm MCF: 50–75 mm ML: < 15% of MCF

Figure 2.

Diagram of thromboelastography (TEG) and rotational thromboelastometry (ROTEM).

According to the clinical scenario, the risk of bleeding varies varies, as described below21:

• •

  Patients with stable cirrhosis without decompensation: Coagulation tests are not recommended prior to most low-risk procedures. Prophylactic and routine correction of INR22 is also not indicated, as no direct relationship has been demonstrated between elevated INR values and bleeding episodes. Depending on the risk associated with the procedure, platelet level correction may be necessary.

• •

  Patients with decompensated cirrhosis: These patients present with hyperfibrinolysis and coagulopathy due to platelet dysfunction. However, the key factor in bleeding events is increased portal hypertension.21 Therefore, a restrictive transfusion strategy is required, which includes avoiding fresh frozen plasma and red blood cell concentrates.

• •

  Patients undergoing liver transplantation: These patients typically present a hypercoagulable state, generally due to tissue factor exposure and the progressive recovery of hepatic synthesis in the liver graft, which gradually increases procoagulant factors23 except in cases of overt surgical bleeding, which contributes to consumptive or dilutional coagulopathy.

• •

  Patients with acute liver failure: Despite significant alterations in conventional coagulation tests, these patients are characterized by a hypercoagulable state. In general, bleeding episodes are infrequent in this context, and prophylactic correction of coagulation parameters is not necessary.24,25

Hospitalized cirrhotic patients have a twofold increased risk of thromboembolic disease vs non-cirrhotic hospitalized patients, and a higher incidence rate of portal vein thrombosis, ranging from 0.6% to 26% among patients with chronic liver disease, increasing in proportion to disease severity.6 Therefore, classic coagulation tests and an elevated INR do not correlate with a greater bleeding tendency nor do they provide protection vs venous thromboembolism.47

In clinical practice, anticoagulation is often avoided in cirrhotic patients due to concerns about inducing serious bleeding, especially in those with prolonged INR or thrombocytopenia.4,12 However, this approach is shifting as the pathophysiology of the hemostatic balance is better understood.48 Current clinical practice guidelines from major hepatology societies, such as the American Association for the Study of Liver Diseases (AASLD) and the European Association for the Study of the Liver (EASL), recognize the importance of thromboprophylaxis in hospitalized cirrhotic patients who are at high risk of thrombosis and low risk of bleeding.4,28

In its 2022 clinical practice guidelines, the EASL28 states that the estimated risk of deep vein thrombosis (DVT) and pulmonary embolism in cirrhotic patients is at least comparable to that of the general population—particularly in those with nonalcoholic steatohepatitis (NASH), which constitutes an independent risk factor for venous thromboembolism.49 The reported prevalence of DVT ranges from 1.2% to 7%,50–52 and several studies and meta-analyses describe a 1.7-fold increased relative risk (RR) of DVT, especially in relation to the lack of pharmacologic thromboprophylaxis.53–56

The EASL recommends the use of thromboprophylaxis in hospitalized cirrhotic patients with platelet counts >50 × 10⁹/L using low molecular weight heparin (LMWH), considering this a safe practice with no significant increase in bleeding risk.28 This also applies to patients with Child-Pugh class A and B cirrhosis. In patients with Child-Pugh class C or with platelet counts between 20 and 50 × 10⁹/L, evidence is currently insufficient to support its use.

For portal vein thrombosis, the VII Baveno Consensus57 recommends initiating anticoagulation in cirrhotic patients with a diagnosis (confirmed by Doppler ultrasound, contrast-enhanced CT, or MRA) of asymptomatic portal vein thrombosis within the first 6 months—whether total or partial occlusion (>50%)—regardless of extension to the superior mesenteric vein. Anticoagulation is also advised in symptomatic patients and/or liver transplant candidates regardless of the extent of thrombosis. If the diagnosis is made via Doppler ultrasound, confirmation by contrast-enhanced CT or MRA is recommended.44

Systemic heparin infusion is recommended for the treatment of symptomatic portal vein, mesenteric, and deep vein thrombosis, although the optimal laboratory parameter for monitoring remains unresolved; both anti-Xa levels and activated partial thromboplastin time (aPTT) are currently under evaluation.27

The paradigm surrounding liver disease patients has changed. The long-held belief in a predominant hypocoagulable state has given way to the concept of rebalanced hemostasis, wherein the patient may be in a procoagulant or anticoagulant state depending on the clinical context. Conventional hemostasis tests have limited utility, while viscoelastic tests may provide more precise information. In general, patients with liver disease do not require correction of coagulation abnormalities unless there is active bleeding or a high-risk invasive procedure. These patients should be considered for pharmacological thromboprophylaxis similarly to the general population.

All authors contributed equally to the development of this manuscript and approved the final version.

No generative AI tools were used in the preparation of this manuscript.

None declared.