A novel online calculator based on noninvasive markers (ALBI and APRI) for predicting post-hepatectomy liver failure in patients with hepatocellular carcinoma

Background and aim: Post-hepatectomy liver failure (PHLF) remains the primary cause of in-hospital mortality after hepatectomy. Identifying predictors of PHLF is important to improve surgical safety. We sought to identify the predictive accuracy of two noninvasive markers, albumin-bilirubin (ALBI) and aspartate aminotransferase to platelet count ratio index (APRI), to predict PHLF among patients with hepatocellular carcinoma (HCC), and to build up an online prediction calculator.

Methods: Patients who underwent resection for HCC between 2013 and 2016 at 6 Chinese hospitals were retrospectively analyzed. The independent predictors of PHLF were identified using univariate and multivariate analyses; derivative data were used to construct preoperative and postoperative nomogram models. Receiver operating characteristic (ROC) curves for the two predictive models, and ALBI, APRI, Child-Pugh, model for end-stage liver disease (MELD) scores were compared relative to predictive accuracy for PHLF.

Results: Among the 767 patients in the analytic cohort, 102 (13.3%) experienced PHLF. Multivariable logistic regression analysis identified high ALBI grade (>-2.6) and high APRI grade (>1.5) as independent risk factors associated with PHLF in both the preoperative and postoperative models. Two nomogram predictive models and corresponding web-based calculators were subsequently constructed. The areas under the ROC curves for the postoperative and preoperative models, APRI, ALBI, MELD and Child-Pugh scores in predicting PHLF were 0.844, 0.789, 0.626, 0.609, 0.569, and 0.560, respectively.

Conclusions: ALBI and APRI demonstrated more accurate ability to predict PHLF than Child-Pugh and MELD. Two online calculators that combined ALBI and APRI were proposed as useful preoperative and postoperative tools for individually predicting the occurrence of PHLF among patients with HCC.