Albumin-indocyanine green evaluation of future liver remnant predicts liver failure after anatomical hepatectomy for hepatocellular carcinoma: A dual-center retrospective study

AimThe albumin-indocyanine green evaluation (ALICE) score is a useful predictor of post-hepatectomy liver failure (PHLF); however, its usefulness in combination with future liver remnant (FLR), measured by 3-D volumetry, has not been investigated. This study aimed to investigate the relationship between the ALICE of the FLR (ALICE-FLR) score and severe PHLF.MethodsThe clinical data of 215 patients who underwent anatomical hepatectomy for hepatocellular carcinoma without portal vein embolization at two institutes between January 2010 and December 2021 were analyzed retrospectively. PHLF occurrence and severity were determined according to the International Study Group of Liver Surgery's definition. Grades B and C PHLF were defined as severe PHLF. The ALICE-FLR, ALICE scores, and indocyanine green clearance of FLR (ICGK-FLR) were evaluated for severe PHLF prediction.ResultsSevere PHLF was observed in 40 patients (18.6%). The areas under the curve (AUCs) for the ALICE-FLR, ALICE scores, ICGK-FLR, and FLR were 0.76, 0.64, 0.73, and 0.69, respectively. The AUC of the ALICE-FLR score was significantly higher than that of the ALICE score. The ALICE-FLR score was identified as an independent predictor of severe PHLF (the odds ratio for every 0.01 increment in the ALICE-FLR score was 1.24; 95% confidence interval, 1.070-1.453; p = 0.004). Among patients with severe PHLF, the ALICE-FLR score was significantly higher in the grade C than in the grade B PHLF group.ConclusionThe combination of liver function models, including indocyanine green, albumin, and FLR is considered compatible for predicting severe PHLF. The albumin-indocyanine green clearance of future liver remnant (ALICE-FLR) score showed a superior predictive value than the ALICE score alone for post-hepatectomy liver failure (PHLF) occurrence and severity. The combination of liver function models, including indocyanine green clearance, albumin, and FLR, measured by 3-D volumetry, is considered compatible for predicting severe PHLF.image