Letter to the editor: discussing the place of TIPS in noncirrhotic patients with chronic extrahepatic portal vein occlusion (EHPVO)

We read with great interest the study published by Knight et al,1 reporting the first monocentric large series on the treatment by TIPS associated or not with portal vein recanalization (PVR) of extrahepatic portal vein obstruction and chronic mesenteric vein thrombosis in patients without cirrhosis. In this study gathering patients considered as refractory to standard of care therapy, 39 patients were included, technical success was obtained in 100% and was associated with an improvement of the symptoms in 87% of the 30 patients having more than 6 months of follow-up. Primary and overall TIPS patency was observed in 63% and 81% of the patients at 36 months, respectively. Three patients (8%) developed hepatic encephalopathy. We recently published our monocentric experience2 reporting long-term favorable outcomes following portal PVR without TIPS insertion in a similar patient population. Thirty-one patients were included, technical success was obtained 87% and associated with the resolution of symptoms in 78% at 5 years. 5-year primary and secondary patencies were, respectively, 73% and 76% in per-protocol analyses. No hepatic encephalopathy has occurred. The overall similar results obtained using different techniques call for an in-depth comparison of the results of the 2 studies to propose patients the best option in the future. First, based on our data, we identified two situations where combining PVR to TIPS could be useful: (a) in patients with an extensive intrahepatic portal vein stenosis or occlusion (type 3 of Marot classification) since there is not enough outflow after recanalisation to allow for long-term stent patency; (b) in patients with chronic/refractory abdominal pain as the main indication for treatment, since 6/7 of these patients have not reached 5-year primary patency. We would be interested in knowing the results obtained by Knight and colleagues in these specific situations. Second, decision for placing a TIPS on top of a PVR could be influenced by the presence of an increased intrahepatic vascular resistance after PVR. It would thus be interesting to know if Knight and colleagues had had the possibility to measure portal caval gradient after recanalization when they used the splenic route. Finally, we observed in our cohort, however on a limited sample size (n=15), that PVR without TIPS was associated with an improved muscle mass (P=0.007) and decreased spleen volume (P=0.01) at 1 year. It would be therefore interesting to know if similar results were obtained using PVR combined with TIPS. In any case, the publication of these 2 series with 2 different approaches are bringing hope to improve the treatment of a severe and chronic condition affecting young patients still recently considered as in a dead-end situation.