Do you routinely start anticoagulation for a patient with newly diagnosed hepatocellular carcinoma presenting with a portal venous thrombosis?

No. Anti-coagulation is generally not indicated. Anti-coagulation is usually only indicated for acute PVT causing symptoms. This is more common with underlying thrombophilia. PVT is very common in cirrhosis and anti-coagulation is not required. PV thrombus from tumor similarly is common and anti-coagulation is not indicated.

In HCC, portal vein thrombosis can be due to tumor or bland thrombus, or both. Prior to anticoagulating anyone, I usually confirm with the radiologist whether the thrombus is most likely to be tumor thrombus (more likely in HCC) or bland (more likely in other diseases). PVT from HCC enhances similarly to the intrahepatic tumor. If PVT is felt to be tumor-based, I will not anticoagulate as I do not believe there is any role for this in terms of outcomes. If there is clear evidence that this is bl...

This used to be my philosophy, but this article has made me wonder if we shouldn't be more aggressive. Valeriani et al., PMID 32911539. Also, we will soon start anticoagulating outpatient cancer patients with high risk for VTE with prophylactic anticoagulation per recent changes in guidelines. Lyman et al., PMID 33570602. Taken together, I may soon approach incidental splanchnic thrombosis like the more common scenario of sub-segmental PE. Try my best to prove the clot is real or not (talk to ra...