Early predictor of mortality due to irreversible posthepatectomy liver failure in patients with hepatocellular carcinoma

Background Although mortality after liver resection has declined, posthepatectomy liver failure (PHLF) remains a major cause of operative mortality. To date there is not consensus on a definition for PHLF. However, there have been many efforts to define PHLF causing operative mortality. In the present study we sought to identify early predictors of death from irreversible PHLF. Materials and methods We retrospectively analyzed the medical records of 359 patients with hepatocellular carcinoma who underwent liver resection between March 2000 and December 2010. Various biochemical parameters from postoperative days (POD) 1, 3, 5, and 7 were analyzed and compared with the "50?50" criterion. Results Operative mortality was 4.7 %. Prothrombin time (PT)\65 % and bilirubin C38 lmol/L on POD 5 showed the only significant difference as compared with "50?50" criterion. The new combination of bilirubin level and the international normalized ratio showed higher sensitivity, area under the curve, as well as similar accuracy (sensitivity 78.6 vs. 28.6 %; p = 0.002; area under the curve 0.8402 vs. 0.6396; p = 0.00176; accuracy 88.6 vs. 93.4 %; p = 0.090). Multivariate analysis revealed the combination of PT\65 % and bilirubin C38 lmol/L on POD 5 to be the only independent predictive factor of mortality (odds ratio, 82.29; 95 % confidence interval 8.69-779.64; p\0.001). Conclusions In patients with chronic liver disease who will undergo liver resection the combination of PT\65 % and bilirubin C38 lmol/L on POD 5 may be a more sensitive predictor than the "50?50" criterion of mortality from PHLF. Although it needs to validated by prospective study, this measure may be applied to select patients receiving artificial liver supports or liver transplantation.