S2389 Risk-Benefit Analysis of Therapeutic Anticoagulation in Liver Cirrhosis: A Large-Scale Cohort Study

Introduction: Therapeutic anticoagulation (AC) in patients with cirrhosis presents a challenging clinical dilemma. While variceal hemorrhage contributes significantly to morbidity and mortality, concurrent thrombotic conditions may necessitate anticoagulation, raising significant safety concerns. This study evaluates anticoagulation outcomes in cirrhotic patients with venous thromboembolism (VTE), portal vein thrombosis (PVT), or atrial fibrillation/flutter (AF) using a large national dataset. Methods: We accessed data through a large-scale, de-identified health research network encompassing 102 healthcare organizations across the United States, representing over 150 million patients (TriNetX, Cambridge, MA). We analyzed AC safety in patients with cirrhosis and VTE, PVT, or AF. Patients were stratified into 2 groups: (1) those receiving AC within 1 month of VTE, PVT, or AF diagnosis and (2) those not receiving AC. Specific International Classification of Diseases, Tenth Revision (ICD-10) codes defined cirrhosis, VTE, PVT, and AF, requiring at least 2 documented cirrhosis diagnoses. Patients with hepatocellular carcinoma and bleeding disorders were excluded. Propensity score matching balanced patient cohorts based on demographics, comorbidities, and medications. Outcomes were assessed at 90 days. Results: After matching, both groups were well-balanced with average age of 64 years, predominantly man (59%), and similar comorbidity profiles including hypertension (54–56%), diabetes (34%), and heart failure (24%). At 90 days, the AC group demonstrated lower variceal hemorrhage incidence (0.6% vs 0.9%, P = 0.032) but higher overall gastrointestinal bleeding (3.7% vs 3.1%, P = 0.013). Significantly more blood transfusions were required in the AC group (7.4% vs 3.8%, P < 0.001), with higher rates of severe anemia (9.1% vs 5.2%, P < 0.001) and spontaneous bacterial peritonitis (1.1% vs 0.8%, P = 0.011). No significant difference in 90-day mortality was observed (12.5% vs 12.7%, P = 0.505). Among AC patients, INR ≥2.0 was associated with significantly higher 90-day mortality versus INR <2.0 (21.6% vs 16.8%, P = 0.040). Conclusion: AC in cirrhotic patients requires careful risk-benefit assessment. While AC patients experienced reduced variceal bleeding, they faced increased transfusion requirements, severe anemia, and infectious complications. Most concerning, INR ≥2.0 was associated with significantly higher mortality, emphasizing the need for individualized decision-making and enhanced monitoring strategies.