Perioperative changes in serum transaminases levels predicts long-term survival following liver resection of hepatocellular carcinoma

Abstract

BACKGROUND: We sought to define whether and how hepatic ischemia/reperfusion (I/R) as manifested by perioperative aspartate aminotransferase (AST) and alanine aminotransaminase (ALT) levels impact long-term outcomes after curative-intent resection of hepatocellular carcinoma (HCC). PATIENTS AND METHODS: Intrasplenic injection of HCC cells was used to establish a murine model of HCC recurrence with versus without I/R injury. Patients who underwent curative resection for HCC were identified from a multi-institutional derivative cohort (DC) and separate external validation (VC) cohort. Perioperative changes of transaminase levels were examined relative to the recurrence-free (RFS) and overall survival (OS) among patients following HCC resection. RESULTS: Mice exposed to hepatic I/R injury were more likely to experience tumor recurrence, as well as higher luminescence signal intensity (all p < 0.05) versus mice with no I/R injury. Relative changes between AST and ALT (sum of AST/ALT ratios, SAAR) on postoperative day (POD) 1 and POD 3 AST1ALT1andAST3ALT3 were calculated using the formula: SAAR=12AST1ALT1+AST3ALT3 via Fourier transform theory. Among 734 patients in DC, the median SAAR was 2.1. After adjusting for other competing risk factors, SAAR >= 2.0 remained strongly associated with risk of postoperative recurrence (ref. SAAR < 2.0, HR 1.32, p = 0.03), whereas SAAR >= 3.5 was associated with risk of postoperative mortality (ref. SAAR < 3.5, HR 1.86, p < 0.01). SAAR demonstrated good accuracy to predict postoperative recurrence (c-index 0.724, 0.731) and mortality (c-index 0.655, 0.765) in DC and VC, respectively. CONCLUSIONS: Use of routine labs such as AST and ALT can help identify patients at high risk of recurrence and mortality following HCC resection.