Abstract

Background: Anticoagulation (AC) is the mainstay of thromboprophylaxis for stroke prevention in atrial fibrillation (AF) and is recommended. Cirrhosis is a risk factor for AF development; hence, AF is common in patients with cirrhosis. The hemostatic pathways in cirrhosis are imbalanced which makes their response to anticoagulation unpredictable. While Direct-Oral Anticoagulants (DOACs) are shown to be safe and effective in patients with AF without cirrhosis, they are hardly studied in patients with cirrhosis. Therefore, this study aims at comparing safety and evaluating different outcomes in patients with Child C cirrhosis receiving anticoagulation for AF.

Methods: We queried the Global Collaborative Network TriNetX- a de-identified database of healthcare organizations from four different countries. Patients with Child-Pugh class C cirrhosis and atrial fibrillation were divided into three cohorts; the first cohort included patients on DOACs (apixaban or rivaroxaban), second cohort included patients not on any anticoagulation, third cohort included patients on warfarin. Two well-matched cohorts were created using 1:1 propensity-scored matching model between cohorts. Three study arms were created after propensity matching. We compared the rates of intracranial hemorrhage, embolic stroke, gastrointestinal (GI) bleed, mortality, and transplant status.

Results: A total of 16,029 patients met the inclusion criteria. Of those, 20.2% (n=3,235) were on DOACs, 47.1% (n=7,552) were not on any anticoagulation, and 32.7% (n=5,242) were on warfarin. In the first study arm comparing patients on AC vs no AC, a statistically significant benefit was identified in terms of 3-year mortality risk (47% vs 71%, p<0.0001) and the transplant status (17% vs 5%, p<0.0001) among patients on AC. However, no significant difference was identified regarding intracranial hemorrhage (3.1% vs 2.7%, p=0.19) and GI bleeding risk 18.8% vs 19.5 %, p=0.3). In the second arm we compared patients on DOACs vs no AC, there was again identified a mortality benefit (40% vs 72%, p<0.0001) and a higher transplant rate (9% vs 3.2%, p<0.0001) with DOACs. The rates of intracranial hemorrhage (6% vs 4%, p=0.03) were higher in patients on DOACs compared to no AC. In the third arm we compared patients on DOAC’s vs Warfarin, a statistically significant lower risk of intracranial hemorrhage (6.6% vs 8.7%, p= 0.004) and GI bleed (2% vs 2.4%, p<0.0001) was identified in patients on DOACs with no difference in the mortality rate (42% vs 43.7%, p=0.2) or the transplant status (8.3% vs 7.1%, p=0.9). (Please see table)

Conclusion: Anticoagulation is safe in patients with Child-Pugh class C cirrhosis with atrial fibrillation and may provide a mortality benefit. DOACs are a safer alternative to warfarin with a lower risk of intracranial hemorrhage and GI bleed. The higher rates of embolic stroke in patients on AC is likely a selection bias as patients with stroke are more likely to be placed on AC.