Skip to main content
. 2016 May;64(5):1058–1067. doi: 10.1016/j.jhep.2016.01.003

Fig. 3.

Fig. 3

Analysis of markers of cell death in peripheral blood, across the liver and in correlation with metabolic profiling. (A–C) Comparison of markers of cell death in peripheral blood on day 1. DC, Decompensated cirrhosis; CLD, chronic stable liver disease; HC, healthy control. Kruskall Wallis test with Dunns test for multiple comparisons (∗p <0.05, ∗∗p <0.01, ∗∗∗p <0.001, ∗∗∗∗p <0.0001) data are expressed as mean(SEM). (D–F) Total and caspase cleaved cytokeratin-18 levels measured by ELISA from samples taken from the hepatic vein (HV), portal vein (PV) and systemic artery (ART) at the time of liver transplant prior to hepatectomy in n = 7 patients. (D) M30 levels demonstrating no gradient (E) M65 levels demonstrating positive gradient between portal and hepatic vein (E) M30/M65 ratio demonstrating negative ratio between portal and hepatic veins. All M30 and M65 levels are in IU/L (logarithmic scale for D-F, Paired t test n.s., non-significant ∗p <0.05). (G-I) OPLS modelling of 1H NMR plasma of patients with DC and M65 levels (G) OPLS model (1 + 1 + 0 components, R2X = 0.39 R2Y = 0.56 Q2 = 0.27 CV-ANOVA p = 0.004) (H) permutation test demonstrating validity of model (I) OPLS S-line loading plot demonstrating correlation with M65 levels and 1H NMR measured metabolites in particular LDL, PC (negative correlation), alanine, methionine, phenylalanine and tyrosine (positive correlation; peak annotations as for Fig. 1 and Table 2). (This figure appears in colour on the web.)