Van Ha et al. have recently described their experience regarding the safety and efficacy of transjugular intrahepatic portosystemic shunt (TIPS) in patients with liver cirrhosis and portal vein thrombosis (PVT) [1]. Amongst 15 patients, including 4 with acute PVT, the overall success rate was 87%, but was less than 75% in 4 patients with long-standing PVT and cavernous transformation. The 30 day mortality was 13%. No patients with non-cirrhotic PVT were included. We have recently reported on 28 patients referred for TIPS who had PVT, which is the largest series in the literature. The overall success rate was 73%, but among 9 patients with cavernous transformation of the portal vein the success rate was similar (6/9; Table 1) [2]. In our series the overall mortality was 10%, but only 4% in the successful TIPS group. At first we thought that the age of the thrombus, the presence of cavernous transformation, or the extension of thrombus into the superior mesenteric vein would be predictive of failure of the procedure, but this was not the case. The only factor which was predictive of failure was the presence of a visible intrahepatic portal vein branch at ultrasound before and during the procedure. Our series included 12 patients with non-cirrhotic PVT for whom the indication for TIPS was recurrent bleeding in 80%. In this group, the success rate was similar to that in the cirrhotic group (9/12, 75%). We agree with Van Ha et al. that the indication for a TIPS procedure for non-cirrhotic PVT remains controversial, but we believe that PVT per se could be an indication. TIPS might not modify survival in patients with non-cirrhotic PVT as they have a limited life expectancy, but it could improve their quality of life. Moreover TIPS should be the preferred therapeutic option in patients with PVT in association with Budd-Chiari syndrome, because it restores blood flow in the sinusoids and improves survival in this group of patients that are characterized by a dreadful prognosis [3]. Van Ha et al. correctly did not reach a firm conclusion on the indication for anticoagulation in this group of patients. In our group we anticoagulated 53% of patients, without complications. We believe that when not contraindicated anticoagulation should be used, especially in non-cirrhotic PVT, given that at least 30% of patients have prothrombotic defects and remembering that not all thrombophilic conditions can be diagnosed [4]. Patients with liver cirrhosis may benefit from anticoagulation in terms of stent patency, when PVT is present before TIPS placement, but this has not been demonstrated conclusively. In our opinion, if no contraindications are present, anticoagulation should be commenced in all patients with PVT. PVT can be an indication for TIPS regardless of the presence of cirrhosis. Cavernous transformation is not a contraindication, even if it increases the technical challenge and possibility of failure, and we believe TIPS should be attempted.