Journal of Hepatology
Volume 47, Issue 2 , Pages 168-171, August 2007

Induction of cerebral hyperemia by ammonia plus endotoxin: Does hyperammonemia unlock the blood–brain barrier?☆

  • Rajiv Jalan

      Affiliations

    • Liver Failure Group, Institute of Hepatology, University College London, 69-75 Chenies Mews, London WCIE 6HX, UK
    • ,
  • Jacques Bernuau

      Affiliations

    • APHP, Hôpital Beaujon, Pôle des Maladies de l’Appareil Digestif, Hépatologie, 92118 Clichy Cedex, France
    • Corresponding Author InformationCorresponding author. Tel.: +33 1 40 87 57 46; fax: +33 1 47 37 05 33.

published online 25 May 2007.

Article Outline

 

Longitudinal studies in patients with acute liver failure (ALF) have clearly shown that the presence of inflammation and/or infection is associated with progression to severe encephalopathy [1], [2]. More recently, this synergistic relationship between ammonia and inflammation in the pathogenesis of increased cerebral blood flow (CBF) and intracranial pressure (ICP) was confirmed in a prospective study in which progression to an ICP greater than 20mmHg was associated with an increase in systemic inflammatory response [3]. Additionally, in the same study, TNFα levels were shown to correlate directly with CBF [3]. This observation was further substantiated with the demonstration of production of pro-inflammatory cytokines from the brain in ALF patients who developed uncontrolled intracranial hypertension [4]. The important study by Pedersen et al. [5], reported in this issue of the Journal, confirms the previous clinical observations of a synergy between hyperammonemia and inflammation in the pathogenesis of cerebral hyperemia and intracranial hypertension. Furthermore, this study provides novel insights into the mechanisms that may underlie this synergy.

Pedersen and colleagues [5] studied (over a 60–65min period) systemic and cerebral hemodynamics in naı¨ve rats (normal liver and no surgical portacaval shunt) that were given intravenous infusion of ammonium acetate and both intraperitoneal and intravenous injections of lipopolysaccharide (LPS), alone or in combination. Moreover, some groups of these animals were pre-treated by amiloride, a non specific inhibitor of the Na+/H+ exchanger (its inhibition induces vasoconstriction), or 5-N-methyl-N-isobutyl-amiloride (MIA) which is a more specific inhibitor of Na+/H+ exchanger (known to be present in the cerebral endothelial cells). The Danish authors [5] essentially showed that : (a) ammonia infusion alone induced no increase in ICP and CBF; (b) LPS alone induced a significant increase in ICP, but not in CBF; (c) combined administration of LPS and ammonia induced a significant increase in both ICP (still more significantly increased than after LPS alone) and in CBF; (d) pre-treatment with amiloride, but not MIA, attenuated the increase in ICP and CBF in animals treated with both ammonia and LPS and this dual improvement occurred independently of changes in plasma ammonia (which increased significantly in the amiloride treated animals), brain glutamine and brain glutamate; (e) treatment with amiloride as well as with MIA was associated with reduced plasma TNFα compared with the corresponding control groups.

The rapidity of administration of ammonia and the degree of hyperammonemia correlate with severity of acute encephalopathy in experimental animals [6] and, in patients with ALF, the degree of hyperammonemia correlates with the outcome [7], [8], [9]. It is therefore noteworthy that, in the naive rats with central temperature maintained at 37°C studied by Pedersen et al., arterial hyperammonemia was above 1mmol/L after the 60-65min ammonia infusion alone [5]. Such marked, and very rapid increase in arterial ammonia – reaching so quickly a level higher than those reported in rats with recent portacaval shunting and either a 6–8h arterial hepatic devascularization [10], [11] or a 60min ammonia infusion at a rate lower than that used by Pedersen et al. [12] – indicates an actual acute ammonia overload/toxicity in these naı¨ve rats given i.v. ammonia alone. Since these animals have preserved cerebral hemodynamics despite high levels of brain glutamine, it would be of great interest to evaluate whether they have early brain edema, the cytotoxic origin of which could be hypothesized. Moreover, from previous studies in hyperammonemic rats [10], one can speculate that these animals with >1mmol/L plasma ammonia, would they not be anesthetized, likely should have alterations in consciousness.

However, the main result of the Pedersen et al’s study is the finding, in naı¨ve rats that were given both LPS and i.v. ammonia infusion, of a noteworthy alteration of cerebral hemodynamics (also observed within 60–65min) with significant increase in both ICP and CBF. This finding elegantly duplicates similar effects observed after simultaneous intraportal administration of α-amanitin and LPS in pigs [13] and experimentally reproduces noteworthy clinical cases such as that of coma with brain edema due to the simultaneous occurrence of acute hepatitis A and paratyphoid infection, a disease associated with endotoxemia [14]. Additionally, Pedersen et al’s findings have other parallels in the existing literature which may further substantiate them. Blei and colleagues [15] have consistently demonstrated that the creation of a portacaval shunt prior to ammonia infusion results in an increase in CBF which is associated with marked brain swelling. Also, creation of a transjugular intrahepatic portasystemic shunt in cirrhotic patients has been shown to be associated with an increase in CBF and intracranial hypertension in susceptible patients [16], [17]. Conceivably, these earlier studies might be re-interpreted as indicating subliminal endotoxemia induced by portacaval shunt with superimposed hyperammonemia, thus offering close similarity with the experiences conducted by Pedersen et al. The fact that cerebral hemodynamic aggravation occurred with no additional increase in arterial ammonia and in cortex glutamine above the levels observed after ammonia infusion alone deserves further comments. One might hypothesize that the main mechanism(s) of this hemodynamic aggravation was (were) not entirely related to the ammonia/glutamine hypothesis alone. However, the significant increase in CBF with no decrease in arterial ammonia suggests that the cerebral flux of ammonia actually did increase [18]. Thus, in these animals, the significant increase in ICP – a feature reflecting, at least in part, brain edema [19] – could result from an increased cerebral uptake of ammonia [18], which was previously emphasized as an important factor of fatal ICP in patients with ALF [8]. Accordingly, lower arterial ammonia which is advantageous in humans with ALF [8] should also be beneficial in rats given both ammonia and LPS. Whether a threshold of arterial ammonia does exist below which the addition of LPS is not followed by acute deterioration of cerebral hemodynamics could be studied in the experimental conditions similar to those used by Pedersen et al.

Ammonia is detoxified by the astrocytes in the brain to glutamine [20]. During acute hyperammonemia, these cells have been shown to swell from the osmotic effect of accumulation of glutamine and partially account for increased ICP [21]. Recent electron microscopic studies in experimental animals have shown that the astrocytes are significantly more swollen in hyperammonemic rats treated with LPS compared with controls and this swelling of the astrocyte occurs in an environment of an anatomically intact blood–brain barrier (BBB), indicating a functional disturbance [22]. The recent elegant experiments demonstrating that astrocytes (which contribute to the BBB) regulate CBF may help to clarify the possible mechanism(s) defining the synergy of interaction between ammonia and LPS in producing cerebral hyperemia and intracranial hypertension [23], [24]. It was shown that the astrocytes are the critical cells that mediate both vasoconstriction and also vasodilation through a mechanism in which the arachidonic acid pathway plays a pivotal role. Accordingly, it might be that during hyperammonemia, the swollen astrocytes are ‘activated’ and then become sensitized to the second hit provided by the LPS. Having said this, an interesting and important observation is the increase in ICP observed in the LPS alone treated animals without any associated cerebral hyperemia (normal CBF) and lack of effect of amiloride (see below). Intriguingly, the cell type in the brain that is swollen following LPS alone (thus with normal ammonia) is also the astrocyte, indicating that the mechanism of astrocytic swelling may involve mechanisms other than the ammonia-glutamine hypothesis [21].

It is also important to note that (a) the dose of the amiloride pre-treatment required for attenuating the increase in ICP and CBF in animals treated with both ammonia and LPS was nearly 200-fold higher than that used for diuresis and (b) amiloride given at lower doses was ineffective [5]. The results must be interpreted in the light of this. As the authors hypothesize, the requirement of the high doses of amiloride and the lack of effect of MIA suggest that the synergy between ammonia and LPS is likely to be mediated, at least in part, through the amiloride sensitive non-selective cation channels. Total brain/ brain microdialysate ammonia measurements may have helped to clarify this issue further. Additionally, the reduction in plasma TNFα observed after both amiloride and MIA is striking and therefore it remains uncertain whether TNFα is playing a role. An alternative hypothesis may well be that amiloride reduces brain ammonia (through effects on amiloride sensitive non-selective cation channels) and, on this background, a reduction in TNFα reduces ICP. The increase in arterial ammonia with the doses of amiloride used reduces its clinical applicability and further supports the important role of the kidneys in the regulation of ammonia metabolism [25].

In the last 30 years, increased permeability of the BBB to many substances was demonstrated in several models of ALF in rats [26], [27]. Recently, ammonia-induced increase of the expression of GLUT1 (the glucose transporter across the BBB) was demonstrated in rats with experimental ALF [28]. Today, the results reported by Pedersen et al. (a) could be interpreted as providing new neurophysiological evidence for the concept of hyperammonemia-induced, functional, non specific increased permeability of BBB, and (b) suggest that hyperammonemia could ‘unlock’ the otherwise histologically intact BBB. This hypothesis is consistent with both the cerebral hemodynamic effects of the combined administration of ammonia and LPS in naı¨ve rats reported by Pedersen et al. [5] and the well known cerebral hypersensitivity to neurotropic drugs in patients with severe liver diseases, such as benzodiazepines in patients with cirrhosis [29] or metoclopramide in patients with acute hyperammonemia due to Reye’s syndrome [30].

What are the messages for the clinician hepatologist in order to reduce, or even obviate, ammonia-induced brain edema? All preventive and curative measures targeting brain edema were masterfully reviewed recently in the Journal [31]. It remains that hepatologists should increase their efforts to teach more efficiently to those who take initial care of patients with acute liver disease – most often non hepatologists – to use as early as possible all the preventive and curative measures aiming at reducing the rapid extension of acute liver lesions, thus reducing the risk of severe, even life-threatening, hyperammonemia and infection/inflammation.

Back to Article Outline

References 

  1. 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
  2. Vaquero J, Polson J, Chung C, Helenowski I, Schiodt FV, Reisch J, et al. Infection and the progression of hepatic encephalopathy in acute liver failure. Gastroenterology. 2003;125:755–764
  3. Jalan R, Olde Damink SW, Hayes PC, Seutz NE, Lee A. Pathogenesis of intracranial hypertension in acute liver failure: inflammation, ammonia and cerebral blood flow. J Hepatol. 2004;41:613–620
  4. Jalan R, Olde Damink SW, Hayes PC, Seutz NE, Lee A. Moderate hypothermia in patients with acute liver failure and uncontrolled intracranial hypertension. Gastroenterology. 2004;127:1338–1346
  5. Pedersen HR, Ring-Larsen H, Olsen NV, Larsen FS. Hyperammonemia acts synergistically with lipopolysaccharide in inducing changes in cerebral hemodynamics in rats anaesthetised with pentobarbital. J Hepatol. 2007;47:245–252
  6. Butterworth RF. Effects of hyperammonemia on brain function. J Inher Metab Dis. 1998;21(Suppl 1):6–20
  7. Clemmesen JO, Larsen FS, Kondrup J, Hansen BA, Ott P. Cerebral herniation in patients with acute liver failure is correlated with arterial ammonia concentration. Hepatology. 1999;29:648–653
  8. Strauss GI, Knudsen GM, Kondrup J, Moller K, Larsen FS. Cerebral metabolism of ammonia and amino acids in patients with fulminant hepatic failure. Gastroenterology. 2001;121:1109–1119
  9. Bhatia V, Singh R, Acharya SK. Predictive value of arterial ammonia for complications and outcome in acute liver failure. Gut. 2006;55:98–104
  10. Bosman DK, Deutz NEP, De Graaf AA, VD Hulst RWN, Van Eijk HMH, Bovée WMMJ, et al. Changes in brain metabolism during hyperammonemia and acute liver failure : results of a comparative 1H NMR spectroscopy and biochemical investigation. Hepatology. 1990;12:281–290
  11. Chatauret N, Desjardins P, Zwingmann C, Rose C, Rama Rao KV, Butterworth RF. Direct molecular and spectroscopic evidence for increased ammonia removal capacity of skeletal muscle in acute liver failure. J Hepatol. 2006;44:1083–1088
  12. Larsen FS, Gottstein J, Blei AT. Cerebral hyperemia and nitric oxide synthase in rats with ammonia-induced brain edema. J Hepatol. 2001;34:548–554
  13. Takada Y, Ishiguro S, Fukunaga K, Gu M, Taniguchi 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
  14. El-Khoury M, Naoushi H, Sawaya R, Aoun E, Nassar NT, Sharara AI. Reversible encephalopathy secondary to paratyphoid infection and concomitant acute hepatitis A. South Med J. 2005;6:723–725
  15. 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
  16. Jalan R, Olde Damink SWM, Redhead DN, Lee A, Hayes PC, Deutz NEP. Increased cerebral and peripheral vasodilation, and whole body nitric oxide production after insertion of a transjugular intrahepatic portal-systemic stent in patients with cirrhosis. In:  Anthony Jones E,  Meijer Alfred J,  Chamuleau Robert AFM editor. Encephalopathy and nitrogen metabolism in liver failure. Dordrecht: Kluwer Academic Publishers; 2003;p. 21–29
  17. Jalan R, Dabos K, Redhead DN, Lee A, Hayes PC. Elevation of intracranial pressure following transjugular intrahepatic portosystemic stent shunt for variceal haemorrhage. J Hepatol. 1997;27:928–933
  18. Ott P, Larsen FS. Blood–brain barrier permeability to ammonia in liver failure: a critical reappraisal. Neurochem Int. 2004;44:185–198
  19. Blei AT. Cerebral edema and intracranial hypertension in acute liver failure: distinct aspects of the same problem. Hepatology. 1991;13:376–379
  20. Martinez-Hernandez A, Bell KP, Norenberg MD. Glutamine synthetase: glial localization in brain. Science. 1977;195:1365–1368
  21. 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
  22. Wright G, Davies NA, Shawcross D, Hodges SJ, Harry D, Mani AR, et al Endotoxaemia induces brain swelling and produces coma in bile duct ligated rats. Hepatology. 2007;45:1517–1526
  23. Takano T, Tian GF, Peng W, Lou N, Libionka W, Han X, et al. Astrocyte-mediated control of cerebral blood flow. Nat Neurosci. 2006;9:260–267
  24. Mulligan SJ, MacVicar BA. Calcium transients in astrocyte endfeet cause cerebrovascular constrictions. Nature. 2004;431:195–199
  25. Olde Damink SW, Dejong CH, Deutz NE, Redhead DN, Hayes PC, Soeters PB, et al. Kidney plays a major role in ammonia homeostasis after portasystemic shunting in patients with cirrhosis. Am J Physiol Gastrointest Liver Physiol. 2006;291:G189–G194
  26. Livingstone AS, Potvin M, Goresky CA, Finlayson MH, Hinchey EJ. Changes in the blood–brain barrier in hepatic coma after hepatectomy in the rat. Gastroenterology. 1977;73:697–704
  27. Zaki AEO, Ede RJ, Davis M, Williams R. Experimental studies of blood brain barrier permeability in acute hepatic failure. Hepatology. 1984;4:359–363
  28. Belanger M, Desjardins P, Chatauret N, Butterworth RF. Selectively increased expression of the astrocytic/endothelial glucose transporter protein GLUT1 in acute liver failure. Glia. 2006;53:557–562
  29. Bakti G, Fisch HU, Karlaganis G, Minder C, Bircher J. Mechanism of the excessive sedative response of cirrhotics to benzodiazepines: model experiments with triazolam. Hepatology. 1987;7:629–638
  30. Bernuau JR, Ichai P, Das V, Mas A, Cazals-Hatem D, Bruguera M, et al. Post-aspirin Reye’s syndrome in young adults: frequent aggravation of encephalopathy by neurotropic drugs and spontaneous recovery. J Hepatol. 2007;46(Suppl 1):S63;[abstract]
  31. Blei AT. Brain edema in acute liver failure: Can it be prevented? Can it be treated?. J Hepatol. 2007;46:564–569

☆ The authors declare that they do not have anything to disclose regarding funding or conflict of interest with respect to this manuscript.

PII: S0168-8278(07)00292-9

doi:10.1016/j.jhep.2007.05.003

Journal of Hepatology
Volume 47, Issue 2 , Pages 168-171, August 2007

Article Tools