Walking the tightrope: toward clinical equipoise in portal vein thrombosis

Walking the tightrope: toward clinical equipoise in portal vein thrombosis P ortal vein thrombosis (PVT) is common in patients with advanced cirrhosis of the liver, occurring in about 16% of patients on the waiting list for liver transplantation. 1 It also occurs in about 11% of patients with less advanced Child-Pugh class A and B cirrhosis. 2 Despite this high frequency and the potential for serious complications in these patients, only limited and incomplete data are available to direct the care of these patients.The article by Volk, Ogola, and Northup in this issue of Baylor University Medical Center Proceedings provides real-world data from a large health care system about the occurrence of PVT in patients with cirrhosis and about the effects of anticoagulation on clinical outcomes. 3These data add to our knowledge and provide us with insight about the management of PVT in cirrhotic patients.These investigators identified 48,596 patients with cirrhosis treated from 2016 to 2023 at Baylor Scott & White Health, a large healthcare system in Texas.After excluding patients who underwent liver transplantation, and who had hepatocellular carcinoma or previous placement of a transjugular intrahepatic portosystemic shunt, they were left with 1153 evaluable patients with PVT and a matched cohort of 3440 cirrhotic patients without evidence of PVT for comparison.Matching criteria included time of first diagnosis of PVT, age, sex, race, Charlson comorbidity index, Model for End-Stage Liver Disease score version 3.0, hemoglobin, platelet count, and presence of ascites, varices, or hepatic encephalopathy. 3ey findings included impaired survival, a higher risk of gastrointestinal bleeding, and a higher rate of hospitalization among the group with PVT as compared to those without PVT. 3Among patients with PVT, those receiving anticoagulation had lower mortality than patients who were not anticoagulated.(This trend persisted after propensity score adjustment, but no longer reached statistical significance: hazard ratio: 0.80, 95% confidence interval 0.62-1.02,P 0.075.)PVT patients who were anticoagulated for >90 days survived better than those receiving treatment for 90 days.On the downside, anticoagulation was associated with a higher risk of gastrointestinal bleeding and a higher rate of hospitalization.