Coagulation Disturbances in Liver Disease
Assessment Before Invasive Procedures
In patients with stable cirrhosis undergoing invasive procedures, routine preprocedural coagulation testing (PT/INR, platelet count) should NOT be performed, and prophylactic correction with blood products is NOT recommended. 1, 2
Key Principle: Rebalanced Hemostasis
Patients with cirrhosis exist in a state of "rebalanced hemostasis" where both procoagulant and anticoagulant factors are simultaneously reduced. Traditional coagulation tests (PT/INR, aPTT) only measure procoagulant factors and do not predict bleeding risk in liver disease 2. These tests are insensitive to anticoagulant protein levels and fail to capture the complete hemostatic picture 2.
Testing Strategy by Procedure Risk
Low-Risk Procedures
(paracentesis, thoracentesis, variceal banding, colonoscopy with polypectomy, ERCP, liver biopsy)
- No preprocedural coagulation testing is indicated 2
- Even with known baseline abnormal coagulation parameters, do not repeat PT/INR or platelet counts before these procedures 1
- Bleeding rates are extremely low (<1.5%) and unrelated to coagulation test abnormalities 2
High-Risk Procedures
(major surgery, procedures where local hemostasis is impossible)
- Laboratory testing is generally not indicated to predict bleeding 2
- Testing may serve only to establish a baseline for comparison if bleeding occurs 2
- Technical factors and complications of liver disease (portal hypertension, renal failure, sepsis) are better predictors of bleeding than coagulation tests 2
Correction of Coagulation Abnormalities
Fresh Frozen Plasma (FFP)
Do NOT use FFP to correct prolonged INR before procedures 2
Rationale:
- FFP contains both pro- and anticoagulant proteins in balanced proportions, providing minimal improvement in thrombin generation 2
- FFP frequently fails to normalize PT/INR in cirrhosis 2
- No studies demonstrate efficacy in preventing procedural bleeding 2
- Significant risks: increases portal pressure (worsening bleeding risk), transfusion-related acute lung injury (TACO), volume overload, allergic reactions 2
Platelet Transfusion and TPO-Receptor Agonists
Routine use is NOT recommended 1, 2
Specific thresholds:
- Platelet count >50 × 10⁹/L: No intervention needed for any procedure 2
- Platelet count 20-50 × 10⁹/L:
- Platelet count <20 × 10⁹/L:
- For high-risk procedures where local hemostasis is impossible: Consider platelet transfusion or TPO-agonists on case-by-case basis 2
Prothrombin Complex Concentrates (PCCs)
Routine use is discouraged 2
Other Products
- Fibrinogen replacement: Routine correction is discouraged 2
- Tranexamic acid: Routine use is discouraged 2
- Red blood cell transfusion: Not recommended for bleeding prophylaxis (only treat anemia with iron, folate, B12) 2
Role of Viscoelastic Testing
No general recommendation can be made for routine use of viscoelastic tests (TEG/ROTEM) to predict bleeding 1, 2
- Evidence is insufficient and represents a knowledge gap 1
- May have utility in specific subgroups (patients with sepsis, renal failure, acute-on-chronic liver failure) where abnormal viscoelastic parameters might identify relative hypocoagulability 2
- Can guide transfusion decisions during active bleeding 2
Special Considerations: High-Risk Patients
Identify patients at truly increased bleeding risk based on:
- Renal failure/acute kidney injury - independent predictor of bleeding 2
- Active infection/sepsis - associated with increased bleeding risk 2
- Severe portal hypertension - mechanical cause of bleeding unrelated to coagulation 2
- Technical/procedural factors - operator experience, number of interventions 2
In these patients with severe decompensation, decisions about prophylactic interventions should involve multidisciplinary consultation (hematology), but still avoid routine blood product transfusion 1.
Critical Pitfalls to Avoid
Do not reflexively transfuse based on INR or platelet count alone - these do not predict bleeding and transfusion carries significant risks 2
Do not delay necessary procedures to "correct" coagulation parameters - this exposes patients to transfusion risks without benefit 1, 2
Recognize that bleeding in cirrhosis is usually mechanical (portal hypertension, vessel injury) rather than coagulopathic 2
Remember the prothrombotic risk - patients with cirrhosis can develop thrombosis; excessive "correction" may tip the balance toward thrombosis 2, 3
Evidence Strength
These recommendations are based on strong consensus from two major 2022 guidelines (EASL 2 and AGA 2021 1), representing the highest quality evidence available. Both guidelines emphasize that while evidence quality is low to very low (due to the impracticality of large RCTs requiring >3,000 patients), the combination of biochemical rationale, lack of clinical benefit, and significant transfusion-related harms supports these strong recommendations against routine correction.