Albumin-Bilirubin and Platelet-Albumin-Bilirubin Grades Do Not Predict Survival After Transjugular Intrahepatic Portosystemic Shunt Creation.

Abstract

To evaluate the capability of albumin-bilirubin (ALBI) and platelet-albumin-bilirubin (PALBI) grades in predicting transplant-free survival (TFS) in patients undergoing transjugular intrahepatic portosystemic shunt (TIPS) creation. This single-center retrospective study included 342 ALBI and 337 PALBI patients (62% men; age 53-54years) with cirrhosis (median MELD 15) and portal hypertension complications (variceal bleeding, 55%; ascites, 35%; other, 10%) who underwent TIPS between 1998 and 2017. Serum albumin, bilirubin, and platelet levels within 24h prior to TIPS were used to calculate ALBI and PALBI grades. The influence of ALBI and PALBI grade on 30-day, 90-day, and overall post-TIPS TFS was assessed using C-indices, binary logistic regression, and the Cox proportional model, adjusting for Child-Pugh (CP) and MELD scores. The cohort spanned 110 (32%) and 232 (68%) ALBI grades 2 and 3 patients, and 40 (12%) and 297 (88%) PALBI grades 2 and 3 patients. While there were no differences in 30-day survival between ALBI and PALBI grades 2/3 (P > 0.05), 90-day and overall TFS showed statistically significant differences in survival between ALBI and PALBI grades 2/3 (P < 0.05). Nonetheless, using univariate logistic regression, ALBI-PALBI C-indices (0.55-0.58) were inferior to the MELD score (0.81-0.84). Moreover, ALBI-PALBI did not associate with TFS on multivariable models adjusting for CP and MELD. Only MELD independently associated with TFS (P < 0.001). ALBI and PALBI grades do not stratify survival outcomes beyond MELD score following TIPS. MELD score remains the most robust metric for predicting post-TIPS survival outcomes.