A network meta-analysis of direct oral anticoagulants for portal vein thrombosis in cirrhosis

Background Current guidelines have limited consensus on the approach to portal venous thrombosis (PVT) in cirrhotic patients. While there is rising interest in direct oral anticoagulants (DOACs) use for PVT, current evidence is limited by small sample size and lack of comparisons to traditional anticoagulants. Thus, a network meta-analysis was conducted to compare the use of DOACs with traditional anticoagulants. Methods Medline and Embase were searched for articles about anticoagulation use in cirrhotic patients with nontumorous PVT for articles on DOACs, warfarin, low-molecular weight heparin (LMWH) or antithrombin III. A network analysis was conducted using risk ratios (RR) with surface under the cumulative ranking curve (SUCRA). A single-arm meta-analysis was used to summarize the outcomes of DOAC treatment. Results A total of 10 articles were included in the study. 79.5% (CI 38.8-95.9) of DOACs patients achieved complete or partial recanalization and 9.80% (CI 4.50-20.0) experienced a bleeding event. DOACs were superior to LMWH (RR 2.299, CI 1.037-5.093, p = 0.040), warfarin (RR 1.762, CI 1.017-3.053, p = 0.043) and no treatment (RR 3.489, CI 1.394-8.733, p = 0.008) in complete recanalization. For partial recanalization, while DOACs were not superior to any treatment, they had the highest probability in achieving partial recanalization in SUCRA analysis. Bleeding risk and mortality were similar compared to other treatments. Conclusion The network analysis supports the use of DOACs in cirrhotic patients, with significant rates of complete recanalization compared to other treatments without increasing bleeding risk. DOACs can potentially be considered for nontumorous PVT in cirrhosis.