Glycodeoxycholic acid levels as prognostic biomarker in acetaminophen-induced acute liver failure patients

Abstract

Acetaminophen (APAP)-induced acute liver failure (ALF) remains a major clinical problem. Although a majority of patients recovers after severe liver injury, a subpopulation of patients proceeds to ALF. Bile acids are generated in the liver and accumulate in blood during liver injury, and as such, have been proposed as biomarkers for liver injury and dysfunction. The goal of this study was to determine whether individual bile acid levels could determine outcome in patients with APAP-induced ALF (AALF). Serum bile acid levels were measured in AALF patients using mass spectrometry. Bile acid levels were elevated 5-80-fold above control values in injured patients on day 1 after the overdose and decreased over the course of hospital stay. Interestingly, glycodeoxycholic acid (GDCA) was significantly increased in non-surviving AALF patients compared with survivors. GDCA values obtained at peak alanine aminotransferase (ALT) and from day 1 of admission indicated GDCA could predict survival in these patients by receiver-operating characteristic analysis (AUC = 0.70 for day 1, AUC = 0.68 for peak ALT). Of note, AALF patients also had significantly higher levels of serum bile acids than patients with active cholestatic liver injury. These data suggest measurements of GDCA in this patient cohort modestly predicted outcome and may serve as a prognostic biomarker. Furthermore, accumulation of bile acids in serum or plasma may be a result of liver cell dysfunction and not cholestasis, suggesting elevation of circulating bile acid levels may be a consequence and not a cause of liver injury.

Keywords: acetaminophen; acute liver failure; bile acid; cholestasis; hepatotoxicity; serum biomarker.

© The Author 2014. Published by Oxford University Press on behalf of the Society of Toxicology.All rights reserved. For Permissions, please e-mail: journals.permissions@oup.com.

Figures

FIG. 1.

Plasma bile acid levels were measured in patients with acetaminophen (APAP)-induced liver injury, control patients (Ctrl), and in patients with APAP overdose but normal liver transaminases (NLT). Total bile acid (TBA) values (A) were defined as the combined concentrations of TDCA, GDCA, GCDCA, TCDCA, TCA, GCA, CA, CDCA, UDCA, LCA, CA, and TLCA. A time course was established post first blood draw for individual glycine conjugated bile acids with plasma ALT time course for the overdose patient group (B), and individual bile acid levels for taurine conjugated bile acids (C). Control patients n = 6, normal LT and 24–96 h patients n = 9–12, day 6 patients n = 3. In A, *P < 0.05 versus control; #P < 0.05 versus NLT patients. In B and C, *P < 0.05 compared with control for all bile acids at that time point.

FIG. 2.

Individual serum bile acid levels were measured in control patients and in patients with APAP-induced ALF at peak ALT time for each patient (A), which were grouped into survivors (S) and non-survivors (NS). GDCA was found to be significantly increased in non-survivors (B) and was associated with a moderate predictive value by ROC analysis (C). Glycine conjugates and taurine conjugates of each bile acid were totaled and averaged for each patient (D). N for each group is indicated in Table 1. *P < 0.05 versus matched pair.

FIG. 3.

Serum bile acid levels were measured in control patients, as well as patients with cholestatic liver injury (CLI) or benign mild biliary dysfunction (UI). A, Individual bile acid levels in UI and CLI patients compared with controls. *P < 0.05 (compared with matched pair). B, TBA levels and (C) ALT activities of patients with cholestasis (UI, CLI) or APAP overdose (S, survivors; NS, non-survivors) at the time of peak ALT levels. *P < 0.05 (compared with UI patient group). #P < 0.05 (compared with CLI patient group). N for each group is indicated in Table 2.