Percutaneous transhepatic interventional therapy of portal vein stenosis in paediatric liver transplantation: A systematic review of efficacy and safety.

PURPOSE OF THE SYSTEMATIC REVIEW To determine the efficacy and safety of percutaneous trans-hepatic balloon and/or stent angioplasty (PTA) in the management of portal vein (PV) stenosis following paediatric liver transplantation. METHODS Articles were included from a systematic search of Medline, Embase, Cochrane CENTRAL, ClinicalTrials.gov and International Clinical Trials Registry from inception to the 29th of August 2020. RESULTS There were 213 paediatric liver recipients who underwent PTA for PV stenosis in 19 included studies published between 1991 and 2019. Balloon angioplasty was the initial treatment in the majority (n = 153). Primary stent placement (n = 34) was performed for elastic recoil, intimal tears and PV kinks and rescue stent placement (n = 14) for recurrent PV stenosis following primary balloon angioplasty. The technical success was 97.6%-100% overall, 97.6%-100% for balloon-angioplasty-only and 100% for primary stenting. The clinical success was 50%-100% overall, 50%-100% for balloon-angioplasty-only and 100% for primary stenting. Long-term PV patency was 50%-100% overall, 37.5%-100% for balloon-angioplasty-only and 100% for primary stenting. Primary balloon angioplasty was successful in 78% of the cases. Of the recurrent PV stenoses, 9% resolved with stent placement and one required a meso-Rex shunt. There was one re-transplantation without stenting. The complication rate was 2.6% for balloon-angioplasty-only (bleeding, liver abscess, 2 PV thromboses) and 5.9% for primary stenting (bleeding, stent-fracture). There was no procedure-related mortality. CONCLUSION Percutaneous transhepatic balloon angioplasty may be the initial management of portal vein stenosis in paediatric liver recipients. Stent placement may be a primary option in selected cases and a reliable rescue option for recurrent portal vein stenosis following balloon-angioplasty-only.