Liver transplantation in patients with non-neoplastic portal vein thrombosis: 20 years of experience in a single center

Background: The Yerdel classification is widely used for describing the severity of portal vein thrombosis (PVT) in liver transplant (LT) candidates, but might not accurately predict transplant outcome. Methods: We retrospectively analyzed data regarding 97 adult patients with PVT who underwent LT, investigating whether the complexity of portal reconstruction could better correlate with transplant outcome than the site and extent of the thrombosis. Results: 79/97 (80%) patients underwent thrombectomy and anatomical anastomosis (TAA), 18/97 (20%) patients underwent non-anatomical physiological reconstructions (non-TAA). PVT Yerdel grade was 1–2 in 72/97 (74%) patients, and 3–4 in 25/97 (26%) patients. Univariate analysis revealed higher 30-day mortality, 90-day mortality, 1-year mortality, and a higher rate of severe early complications in the non-TAA group than in the TAA group (p =.018,.001,.014,.009, respectively). In the model adjusted for PVT Yerdel grade, non-TAA remained independently associated with higher 30-day, 90-day, and 1-year mortality (p =.021,.007, and.015, respectively). The portal vein re-thrombosis and overall patient and graft survival rates were similar. Discussion: In our experience, the complexity of portal reconstruction better correlated with transplant outcome than the Yerdel classification, which did not even appear to be a reliable predictor of the surgical complexity and technique.