Detailed Abstract
[Liver Oral Presentation 2]
[LV OP 2-6] Serum metabolic biomarkers of liver failure after liver resection
Kyung Chul YOON1, Hyung DO KWON2, Hye-Sung JO1, Yoon Young CHOI1, Jin-I SEOK1, Do Yup LEE3, Dong-Sik KIM*1
1Department of Surgery, Division of HBP Surgery and Liver Transplantation, Korea University Medical Center, Korea University Medical College, Korea
2Department of Bio and Fermentation Convergence Technology, BK21 PLUS Program,, Kookmin University, Korea
3Department of Agricultural Biotechnology, Center for Food and Bioconvergence, Research Institute for Agricultural and Life Sciences, Seoul National University, Korea
Introduction : Even the combination of PT and total bilirubin, known for being the most reliable predictive conventional markers shows low sensitivity. Therefore we need to find more relevant biomarkers of liver failure after liver resection focused on the liver metabolites
Methods : A total of 20 pigs were divided into 3 groups sham operation (n=6), 70% hepatectomy group (n=7), and 90% hepatectomy group (n=7). Blood sampling was performed preoperatively and at 1, 6, 14, 30, 38, and 48 hours after the operation we systematically profiled 129 primary metabolites based on gas-chromatography time-of-flight mass spectrometry.
Results : Orthogonal projection to latent structures-discriminant analysis revealed that central carbon metabolism was the most significant factor in the 90% liver-resection group in contrast to the 70% and sham groups. Subsequent binary logistic regression analysis was used to develop a predictive model for the risk of mortality following hepatectomy. The recommended variables were malic acid, methionine, tryptophan, glucose, and γ-aminobutyric acid. The AUC of the linear combination of 5 metabolites was 1.000 (95% confidence interval: 0.940–1.000, sensitivity: 100.0, specificity: 94.87)
Conclusions : Systematic prioritization based on OPLS-DA and binary logistic regression analysis proposed robust biomarker panels that can accurately predict the risk of mortality associated with hepatectomy.
Methods : A total of 20 pigs were divided into 3 groups sham operation (n=6), 70% hepatectomy group (n=7), and 90% hepatectomy group (n=7). Blood sampling was performed preoperatively and at 1, 6, 14, 30, 38, and 48 hours after the operation we systematically profiled 129 primary metabolites based on gas-chromatography time-of-flight mass spectrometry.
Results : Orthogonal projection to latent structures-discriminant analysis revealed that central carbon metabolism was the most significant factor in the 90% liver-resection group in contrast to the 70% and sham groups. Subsequent binary logistic regression analysis was used to develop a predictive model for the risk of mortality following hepatectomy. The recommended variables were malic acid, methionine, tryptophan, glucose, and γ-aminobutyric acid. The AUC of the linear combination of 5 metabolites was 1.000 (95% confidence interval: 0.940–1.000, sensitivity: 100.0, specificity: 94.87)
Conclusions : Systematic prioritization based on OPLS-DA and binary logistic regression analysis proposed robust biomarker panels that can accurately predict the risk of mortality associated with hepatectomy.
SESSION
Liver Oral Presentation 2
Room A 7/27/2020 2:03 PM - 2:10 PM