Journal of Hepatology
Volume 34, Issue 6 , Pages 940-942, June 2001

The inflammatory basis of intracranial hypertension in acute liver failure

Institute of Hepatology, University College London Medical School, 69–75 Chenies Mews, London WC1E 6HX, UK

See Article, pages 825–831

Article Outline

 

The encephalopathy that is associated with acute liver failure (ALF) [1] is characterized by the occurrence of cerebral oedema which results in elevation in intracranial pressure (ICP). The pathogenesis of cerebral oedema remains elusive and the current hypotheses suggest derangements in osmotic balance induced by detoxification of ammonia to glutamine in the astrocytes, alterations in cerebral blood flow autoregulation and the yet unidentified products of the necrotic liver as the potential causes (reviewed in detail by Larsen and Blei [2]). The cell in the brain that is central in mediating these cerebral changes is the astrocyte which belongs to the microglial lineage and has extensive repertoire to modulate inflammatory responses. Patients with ALF are also known to be highly susceptible to both bacterial and fungal infections with up to 30% of patients dying as a result, and a recent study on systemic inflammatory response syndrome (SIRS) shows good correlation with severity and progress of encephalopathy (see later) [3]. Although disturbed neutrophil chemotaxis has been blamed as a cause of this enhanced susceptibility to infection, defects in immune regulation can be identified at multiple levels [3].

In the present issue of Journal of Hepatology, Takada et al. [4] report the development of a porcine model of ALF following the intraportal administration of amatoxin and lipopolysaccharide (LPS). They studied four groups of animals in a non-randomized protocol who were administered either saline (group 1), LPS (group 2), amatoxin (group 3) or a combination of amatoxin and LPS (group 4). Their results show that the animals in group 4 had a 100% mortality whereas the mortality in group 3 was only 60%. All the animals in groups 1 and 2 survived. Interestingly, they made two other important observations: (1) the severity of liver injury was significantly greater in the animals where both were given, and (2) the severity of intracranial hypertension was also more marked and occurred earlier in this group of animals. Three pigs underwent liver transplantation following amatoxin and lipopolysaccharide, with prompt recovery of liver function suggesting that the manifestations of ALF were consequent entirely on liver injury. The authors suggest that this model using a combination of amatoxin and LPS is robust and reliable, and would be useful in developing newer therapies for ALF.

Back to Article Outline

1. Severity of liver injury 

This study is potentially important because it does start to put into perspective some of clinical data that exist in the literature. As already referred to, Rolando et al. [3] have recently described the SIRS in a series of 887 patients with ALF who presented to their unit over an 11-year period. In agreement with the findings of the study by Takada et al. [4], the mortality rate from ALF in those who were either overtly infected or had evidence of SIRS (>2 components) was about twice that in those that did not have evidence of either. In addition, they showed clearly that those that had higher SIRS or were overtly infected, had markedly greater severity of encephalopathy and were also more likely to develop intracranial hypertension [3]. Although severe sepsis can rarely produce a syndrome that resembles acute liver failure [5], LPS administration alone in the pigs produced only very mild derangements in liver function and none of the animals died suggesting that the enhanced hepatotoxicity and mortality was due to the synergistic effect of the amatoxin and LPS in inducing hepatic necrosis.

LPS is a potent inducer of the release of tumour necrosis factor (TNF) and its hepatocyte apoptotic effect has been shown to be enhanced both in vitro and in vivo when sensitizing concentrations of a hepatotoxin are present [6]. A neutralizing antibody to LPS has been shown to reduce the extent of hepatocellular damage during administration of hepatotoxin in rats [7]. Experiments using anti-TNF antiserum have suggested that that the mechanism of the hepatotoxin mediated injury may be modulated through an indirect mechanism involving sensitization of the hepatocytes towards the endogenously produced TNF [8]. These indirect mechanisms may involve the effects of LPS on the microvasculature that is mediated through alterations in induction of nitric oxide synthases and the generation of free radical mediated injury. These may set up an inflammatory cascade that inflicts upon essential cellular function such as respiration.

Back to Article Outline

2. Intracranial hypertension 

Why the cerebral oedema was worse in the animals treated with amatoxin and LPS is more difficult to understand. Because the ammonia levels in groups 3 and 4 were similar, it is unlikely that the osmotic influences of ammonia and astrocytic glutamine is the operative mechanism. There are three mechanisms whereby peripheral cytokines can affect their deleterious effects on the brain: (1) direct signalling to the brain via peripheral or autonomic nerves that innervate the peripheral tissue such as the liver; (2) signalling through the brain vasculature through production of endothelial factors such as nitric oxide or prostanoids; (3) direct action of the cytokines on the brain after crossing the blood–brain barrier [9]. No data are provided in the paper by Takada et al. that indicates which of these mechanisms may be operative. It is likely that either or both of the first two mechanisms may be operative because previous studies have indicated that the blood–brain barrier is preserved in patients with ALF [10]. Evidence for an NO-mediated mechanism exists in studies of animal models of ALF where enhanced brain flux of NO has been demonstrated and also increased expression of neuronal nitric oxide synthase [11], [12]. The suggestion that prostaglandins may be involved in the signalling process has been confirmed in animal models of ALF and also through the demonstration that increased ICP may be controlled successfully by the administration of bolus doses of a cyclo-oxygenase inhibitor such as indomethacin [13], [14]. Previous reports have suggested that better control of intracranial hypertension following hepatectomy in patients with ALF is consistent with the importance of the necrotic liver in modulating the greater intracranial hypertension that was observed in the group 4 animals [15], [16]. The mechanism why hepatectomy controlled intracranial hypertension may be by limiting the production of cytokines from the necrotic liver. The clinical relevance of the circulating cytokines was addressed in a preliminary communication where the authors noted higher levels of circulating pro-inflammatory cytokines, interleukin (IL)-1β, IL-6 and soluble TNF receptors in patients with ALF that had worse episodes of encephalopathy [17].

Back to Article Outline

3. Implications for treatment 

The above hypotheses would suggest that prevention or control of infection with mitigation of SIRS and therefore endotoxaemia would reduce the degree of hepatic failure and cerebral oedema. Indeed, controlled clinical trials have clearly shown that the incidence of infection was significantly higher in those patients that developed encephalopathy compared with those that did not. Patients who on admission were not encephalopathic, progressed to coma if they became septic. However, the use of antibiotics was not protective against either cerebral oedema or mortality [18], [19]. There may be two reasons for this. The first is possibly due to administration of antibiotics relatively late in the disease process and the second is a primary defect in the neutrophil function introduced by liver failure. The roles of early enteral nutrition, probiotics and very early administration of antibiotics need to be evaluated. It may be possible to modulate intrahepatic synthesis of pro-inflammatory cytokines by inhibiting the transcription factor NF-κB by pentoxifylline, thalidomide or N-acetylcysteine. Other approaches may involve the specific use of a monoclonal antibody directed against pro-inflammatory cytokines.

In a preliminary communication, moderate hypothermia has been shown to reduce intracranial hypertension [20]. Clearly, hypothermia may reduce ICP through a number of mechanisms but was shown to significantly reduce the levels of circulating IL-1β, suggesting that slowing down liver and whole body metabolism possibly reduces the production of these pro-inflammatory cytokines [21]. In another preliminary communication, reduced hepatocyte necrosis was demonstrated in response to pro-apoptotic signal. These suggest a critical role for modulation of end-organ injury in ALF through activation of pro-inflammatory cytokines which may be exacerbated by concomitant sepsis [22].

Back to Article Outline

4. Conclusions 

In conclusion, the study by Takada et al. is extremely valuable because it provides further evidence that links toxin-induced liver failure and sepsis as synergistic in producing exaggerated liver injury and also end-organ damage. In addition, their data provide a robust model which can be used to test new therapeutic approaches and understand the mechanisms involved in the synergistic effect of a toxic liver injury and sepsis.

Back to Article Outline

References 

  1. Trey C, Davidson CS. The management of fulminant hepatic failure. In:  Popper H,  Schaffner F editor. Progress in liver disease. New York: Grune and Stratton; 1970;p. 282–298
  2. Blei AT, Larsen FS. Pathophysiology of cerebral edema in fulminant hepatic failure. J Hepatol. 1999;31:771–776
  3. Rolando N, Wade J, Davalos M, Wendon J, Philpott-Howard J, Williams R. The systemic inflammatory response syndrome in acute liver failure. Hepatology. 2000;32:734–739
  4. Takada Y, Ishiguro S, Fukunaga K, Gu M, Tanugichi H, Seino K-I, et al.  Increased intracranial pressure in a porcine model of fulminant hepatic failure using amatoxin and endotoxin. J Hepatol. 2001;34:825–831
  5. Dirix LY, Polson RJ, Richardson A, Williams R. Primary sepsis presenting as fulminant hepatic failure. Q J Med. 1989;73:1037–1043
  6. Galanos C, Freudenberg MA, Reutter W. Galactosamine-induced sensitization to the lethal effects of endotoxin. Proc Natl Acad Sci USA. 1979;76:5939–5943
  7. Czaja MJ, Xu J, Ju Y, Alt E, Schmiedeberg P. Lipopolysaccharide-neutralizing antibody reduces hepatocyte injury from acute hepatotoxin administration. Hepatology. 1994;19:1282–1289
  8. Leist M, Gantner F, Naumann H, Bluethmann H, Vogt K, Brigelius-Flohe R, et al.  Tumor necrosis factor-induced apoptosis during the poisoning of mice with hepatotoxins. Gastroenterology. 1997;112:923–934
  9. Licinio J, Wong ML. Pathways and mechanisms for cytokine signaling of the central nervous system. J Clin Invest. 1997;100:2941–2947
  10. Kato M, Hughes RD, Keays RT, Williams R. Electron microscopic study of brain capillaries in cerebral edema from fulminant hepatic failure. Hepatology. 1992;15:1060–1066
  11. Master S, Gottstein J, Blei AT. Cerebral blood flow and the development of ammonia-induced brain edema in rats after portacaval anastomosis. Hepatology. 1999;30:876–880
  12. Rao VL, Audet RM, Butterworth RF. Increased neuronal nitric oxide synthase expression in brain following portacaval anastomosis. Brain Res. 1997;765:169–172
  13. Chung C, Gottstein J, Blei AT. Indomethacin prevents the development of brain edema in experimental ammonia-induced cerebral swelling (abstract). Hepatology. 2000;32:298A
  14. Clemmesen JO, Hansen BA, Larsen FS. Indomethacin normalizes intracranial pressure in acute liver failure: a twenty-three-year-old woman treated with indomethacin. Hepatology. 1997;26:1423–1425
  15. Ringe B, Lubbe N, Kuse E, Frei U, Pichlmayr R. Total hepatectomy and liver transplantation as two-stage procedure. Ann Surg. 1993;218:3–9
  16. Rozga J, Podesta L, LePage E, Hoffman A, Morsiani E, Sher L, et al.  Control of cerebral oedema by total hepatectomy and extracorporeal liver support in fulminant hepatic failure. Lancet. 1993;342:898–899
  17. Rolando N, Ellis T, DeGroote D, Wendon J, Williams R. IL-6 and coma after paracetamol hepatotoxicity. In: Proceedings of the IX ISA: Recent advances in hepatic encephalopathy and metabolism. 1996;p. 51;(abstract)
  18. Rolando N, Wade JJ, Stangou A, Gimson AE, Wendon J, Philpott-Howard J, et al.  Prospective study comparing the efficacy of prophylactic parenteral antimicrobials, with or without enteral decontamination, in patients with acute liver failure. Liver Transpl Surg. 1996;2:8–13
  19. Rolando N, Gimson A, Wade J, Philpott-Howard J, Casewell M, Williams R. Prospective controlled trial of selective parenteral and enteral antimicrobial regimen in fulminant liver failure. Hepatology. 1993;17:196–201
  20. Jalan R, Damink SW, Deutz NE, Lee A, Hayes PC. Moderate hypothermia for uncontrolled intracranial hypertension in acute liver failure. Lancet. 1999;354:1164–1168
  21. Jalan R, Olde Damink SWM, Hayes PC, Lee A. Moderate hypothermia prevents cerebral hyperemia and increase in intracranial pressure in patients undergoing liver transplantation for acute liver failure (abstract). Hepatology. 2000;32:339A
  22. Fu T, Blei AT, Tesu L, Takamura N, Guo D, O'Gorman RG, et al.  Hypothermia suppresses apoptosis of mouse hepatocytes in culture induced by soluble Fas ligand, tumour necrosis factor alpha or thapsigargin (abstract). Hepatology. 2000;32:78A

PII: S0168-8278(01)00038-1

Journal of Hepatology
Volume 34, Issue 6 , Pages 940-942, June 2001

Article Tools