Article Text


P02 Apoptosis in paracetamol-induced acute liver failure: importance of extra-hepatic organ dysfunction
  1. L A Possamai1,
  2. A Quaglia2,
  3. C G Antoniades1,
  4. V Zingarelli3,
  5. M J McPhail1,
  6. G Auzinger2,
  7. W Bernal2,
  8. N D Heaton2,
  9. R Hughes3,
  10. Y Ma3,
  11. M Thursz1,
  12. J A Wendon3,
  13. R R Mitry3
  1. 1Imperial College, London, UK
  2. 2King's College Hospital
  3. 3King's College, London, UK


Introduction Both necrotic and apoptotic hepatocyte death pathways have been implicated in paracetamol induced ALF (PALF). Elevated circulating levels of M30, a marker of caspase-3 activation, have been demonstrated in this condition and are postulated to reflect hepatocellular apoptosis. This novel marker has also been reported to have clinical utility as a prognostic indicator in ALF. However, published results are conflicting and it remains unclear whether significant hepatocyte apoptosis is clinically relevant in PALF.

Aim To investigate the role of hepatocyte apoptosis in PALF.

Method Serum M30 levels were quantified by ELISA (Peviva, Bromma, Sweden) in 62 patients with PALF (34 spontaneous survivors (S) and 28 OLT or died (D)). Control groups of 10 healthy volunteers and 20 chronic HCV patients were used for comparison. Clinical outcome measures in PALF were correlated with M30 levels. In four patients undergoing transplantation for PALF, blood was sampled from the hepatic vein (HV), portal vein (PV) and a systemic artery and serum levels of M30 quantified as above. A focused proteome apoptosis array was performed on liver homogenates from 4 PALF cases and 4 controls. Protein was quantified using a modified Lowry assay and concentrations normalised. Array films were scanned and analysed in ImageJ. H&E stained liver sections from AALF patients were examined for histological evidence of apoptosis.

Results Serum M30 levels were significantly elevated in PALF (3970 IU/l) compared with both healthy volunteers (144 IU/l) and HCV patients (170 IU/l) (p<0.001). Serum levels of M30 were significantly elevated in PALF-D compared to PALF-S (5199 vs 2357 IU/l; p<0.0005). There was a transportal gradient of M30 in all patients tested, with highest levels in the PV and lowest in the HV (p<0.03). Systemic arterial M30 levels were intermediate between the PV and HV levels (Abstract P02 figure 1). Analysis of apoptosis arrays showed significant downregulation of a number of apoptosis-associated proteins, including Bax, HIF-1, FADD and SMAC/Diablo (p<0.04) (Abstract P02 figure 2). Catalase was markedly elevated compared to controls in three of the four AALF patients. Histological evaluation revealed confluent hepatocellular loss, epithelial regenerative activity and an absence of apoptotic bodies.

Abstract P02 Figure 1

Regional levels of M30 (marker of cleaved cytokeratin-18) in four cases of Paracetamol-induced acute liver failure.

Abstract P02 Figure 2

Downregulation of apoptosis-associated proteins in AALF patients (n=4) compared with controls (n=4).

Conclusion In this large cohort of PALF patients we have demonstrated the presence of elevated M30 levels and a correlation between caspase three activation and poor clinical outcome. The transhepatic M30 gradient, down regulation of apoptosis-associated proteins and histological appearances indicate that hepatocellular apoptosis might not be the major source of circulating M30. Our data also indicate that in established PALF, apoptosis in non-hepatic epithelial tissues may predominate and is likely to reflect incipient multi-organ failure, with resulting poor outcomes.

Abstract P02 Figure 3

Examples of apoptosis array for case and control.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.