Ammonia – an enduring foe – What evaluating whole body ammonia metabolism can teach us about cirrhosis and therapies treating hepatic encephalopathy

Clearance and production of ammonia quantified in humans by constant ammonia infusion – the effects of cirrhosis and ammonia-targeting treatmentsJournal of HepatologyVol. 79Issue 2PreviewHyperammonaemia is a key pathological feature of liver disease and the primary driver of hepatic encephalopathy (HE). However, the relative roles of increased ammonia production and reduced clearance are poorly understood as is the action of ammonia-targeting drugs for HE. We aimed to quantify whole-body ammonia metabolism in healthy persons and patients with cirrhosis and to validate our method by examining the effects of glycerol phenylbutyrate and lactulose + rifaximin treatment. Full-Text PDF Open AccessFurther on ammonia concentration, -clearance, -removal, and -productionJournal of HepatologyPreviewFirst, thanks to Drs. Shawcross, Thabut, and Amodio for their editorial,1 putting our ammonia metabolism method study2 wonderfully into scientific, clinical, and not least historical perspective. We together hope the method can illuminate some of the traditional mysticism surrounding ammonia metabolism. Full-Text PDF See Article, pages 340–348 See Article, pages 340–348 Ammonia is not a new player in the field of hepatology and in 1893, Hahn, Messen, Nencki and Pavlov described the neurocognitive consequences of creating a portacaval shunt in dogs, a surgical procedure first described by Eck in 1879 (“Eck’s fistula”).[1]Hahn M. Massen O. Nencki M. Pavlov I. Die Eck’sche fistel zwischen der unteren hohlvene und der pfortader und ihre folgen Fur den organismus.Arch Exp Pathol Pharm. 1893; 32: 161-210Crossref Scopus (145) Google Scholar Nitrogen balance studies led them to discover that there was an excess of ammonia salts in the urine leading to the first clinical description of the syndrome of hepatic encephalopathy (HE) associated with elevated blood and brain levels of ammonia; they subsequently demonstrated that ammonia is metabolized to urea in the liver. Sheila Sherlock, who almost single-handedly established hepatology as a speciality in the United Kingdom, was convinced there was a relationship between the development of HE and the ammonia level;[2]Sherlock S. Summerskill W.H. White L.P. Phear E.A. Portal-systemic encephalopathy; neurological complications of liver disease.Lancet. 1954; 267: 454-457PubMed Google Scholar in the late 1940s, she performed lumbar punctures to measure ammonia levels in the cerebrospinal fluid of her patients, such was her conviction. In 1954, she published the results of oral ammonia tolerance testing (3 g of ammonium chloride) in healthy individuals and in 51 patients with cirrhosis, in whom she observed the development of elevated blood ammonia.[3]White L.P. Phear E.A. Summerskill W.H. Sherlock S. Ammonium tolerance in liver disease: observations based on catheterization of the hepatic veins.J Clin Invest. 1955; 34: 158-168Crossref PubMed Scopus (29) Google Scholar This highlighted the role of portal-systemic shunting in the pathophysiology of the increased uptake of ammonia from the gut into the systemic circulation (Fig. 1). The study by Peter Lykke Eriksen et al. in this issue of the Journal of Hepatology[4]Eriksen P.L. Djernes L. Vilstrup H. Ott P. Clearance and production of ammonia quantified in humans by constant ammonia infusion - the effects of cirrhosis and ammonia targeting treatments.J Hepatol. 2023; 79: 340-348Abstract Full Text Full Text PDF PubMed Scopus (1) Google Scholar provides a further seminal step in our understanding by showing a new approach to quantifying systemic ammonia metabolism, and confirming that patients with cirrhosis are characterised by a threefold influx of ammonia from the splanchnic vascular bed, while reduced clearance of ammonia (20%) plays only a minor role. The technique described by the authors, which involves a 90-minute constant ammonia infusion to reach steady state, is simple and well suited for animal studies and human studies, where local regulatory authorities have approved its use. However, the technique requires further tuning, since steady state was not reached in all participants during ammonia infusion, supporting the view that the amount of infused ammonia might have been excessive. Indeed, one patient with Child-Pugh B9 cirrhosis developed grade 2 HE with an arterial ammonia concentration of 348 μmol/L following ammonia infusion, although this was linked to a synchronous gastrointestinal bleed. Insight into the dynamics of ammonia and nitrogen disposal in cirrhosis have previously been published in studies of urea synthesis. These have shown: i) functional urea synthesis following alanine infusion is reduced and less sensitive to the increase in amino acid levels;[5]Vilstrup H. Synthesis of urea after stimulation with amino acids: relation to liver function.Gut. 1980; 21: 990-995Crossref PubMed Scopus (113) Google Scholar ii) its modulation by glucose is lost, because of increased glucagon levels.[6]Hamberg O. Regulation of urea synthesis by diet protein and carbohydrate in normal man and in patients with cirrhosis. Relationship to glucagon and insulin.Dan Med Bull. 1997; 44: 225-241PubMed Google Scholar Furthermore, if fibrosis regresses, as occurs in hepatitis C-related cirrhosis following treatment with direct-acting antivirals, urea synthesis capacity improves.[7]Laursen T.L. Sandahl T.D. Kazankov K. Eriksen P.L. Kristensen L.H. Holmboe C.H. et al.Early normalization of reduced urea synthesis capacity after direct-acting antiviral therapy in hepatitis C cirrhosis.Am J Physiol Gastrointest Liver Physiol. 2020; 319: G151-G156Crossref PubMed Scopus (3) Google Scholar The reduction in urea synthesis results from a reduction in the effective hepatocyte mass[5]Vilstrup H. Synthesis of urea after stimulation with amino acids: relation to liver function.Gut. 1980; 21: 990-995Crossref PubMed Scopus (113) Google Scholar and reduced activity of the enzymes implicated in urea synthesis.[8]Khatra B.S. Smith R.B. Millikan W.J. Sewell C.W. Warren W.D. Rudman D. Activities of Krebs-Henseleit enzymes in normal and cirrhotic human liver.J Lab Clin Med. 1974; 84: 708-715PubMed Google Scholar Clearance of gut-derived ammonia is also hampered by perfusion mismatch between portal blood flow and the hepatic acinar structures devoted to ammonia disposal. Ammonia disposal requires precise compartmentalisation based on the liver architecture: glutaminase activity and urea synthesis occur in periportal areas, whereas glutamine synthesis occurs in the perivenous hepatocytes.[9]Häussinger D. Regulation of hepatic ammonia metabolism: the intercellular glutamine cycle.Adv Enzyme Regul. 1986; 25: 159-180Crossref PubMed Google Scholar Structural alterations in the liver architecture in cirrhosis create a blood flow/hepatocyte mismatch, as well as intrahepatic shunts that reduce urea synthesis. Improving urea synthesis in patients with cirrhosis by treating the cirrhosis per se, or by augmenting the urea cycle with L-ornithine L-aspartate[10]Kircheis G. Lüth S. Pharmacokinetic and pharmacodynamic properties of L-ornithine L-aspartate (LOLA) in hepatic encephalopathy.Drugs. 2019; 79: 23-29Crossref PubMed Scopus (29) Google Scholar or zinc,[11]Marchesini G. Fabbri A. Bianchi G. Brizi M. Zoli M. Zinc supplementation and amino acid-nitrogen metabolism in patients with advanced cirrhosis.Hepatology. 1996; 23: 1084-1092Crossref PubMed Google Scholar for example, is thus a reasonable approach. Eriksen et al.[4]Eriksen P.L. Djernes L. Vilstrup H. Ott P. Clearance and production of ammonia quantified in humans by constant ammonia infusion - the effects of cirrhosis and ammonia targeting treatments.J Hepatol. 2023; 79: 340-348Abstract Full Text Full Text PDF PubMed Scopus (1) Google Scholar show that the large influx of ammonia into the systemic circulation results entirely from portal-systemic shunting and this has also previously been demonstrated in stable patients with cirrhosis and a TIPS (transjugular intrahepatic portosystemic shunt), although renal ammonia production was also shown to be important. This means urine ammonia excretion and urea synthesis are the only pathways to remove ammonia from the body.[12]Olde Damink S.W. Dejong C.H. Deutz N.E. Redhead D.N. Hayes P.C. Soeters P.B. 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-G194Crossref PubMed Scopus (30) Google Scholar In health, the hepatic extraction efficiency of ammonia from the portal vein is approximately 93%.[13]Cooper A.J. Nieves E. Coleman A.E. Filc-DeRicco S. Gelbard A.S. Short-term metabolic fate of [13N]ammonia in rat liver in vivo.J Biol Chem. 1987; 262: 1073-1080Abstract Full Text PDF PubMed Google Scholar Thus, little, if any, ammonia enters the systemic circulation from the portal vein under normal conditions, and the liver maintains the level of circulating ammonia at relatively low levels. Ammonia extraction is reduced in the presence of portosystemic shunting.[14]Nomura F. Ohnishi K. Terabayashi H. Nakai T. Isobe K. Takekoshi K. et al.Effect of intrahepatic portal-systemic shunting on hepatic ammonia extraction in patients with cirrhosis.Hepatology. 1994; 20: 1478-1481Crossref PubMed Scopus (31) Google Scholar In cirrhosis, ammonia is not the only gut-derived substance to increase systemically and there is increasing evidence to suggest that a predominance of pathobionts within the gut microbiome that produce more ammonia is a contributing factor irrespective of the degree of portosystemic shunting.[15]Bajaj J.S. Heuman D.M. Hylemon P.B. Sanyal A.J. White M.B. Monteith P. et al.Altered profile of human gut microbiome is associated with cirrhosis and its complications.J Hepatol. 2014; 60: 940-947Abstract Full Text Full Text PDF PubMed Scopus (717) Google Scholar Furthermore, increased small intestinal glutaminase activity, which is induced by diabetes,[16]Ampuero J. Ranchal I. del Mar Díaz-Herrero M. del Campo J.A. Bautista J.D. Romero-Gómez M. Role of diabetes mellitus on hepatic encephalopathy.Metab Brain Dis. 2013; 28: 277-279Crossref PubMed Scopus (26) Google Scholar and small intestinal bacterial overgrowth also contribute to the splanchnic ammonia load in patients with cirrhosis. One unique and major finding of this study is that assessing whole-body ammonia metabolism in health and in patients with cirrhosis offers an opportunity to understand and quantify the impact of therapies used to treat HE if the presumption is they are reducing ammonia production and/or increasing ammonia excretion. In this study, ammonia clearance was lower and ammonia production was higher in patients with cirrhosis. Phenylbutyrate was shown to increase ammonia clearance by 11% in seven healthy individuals and the combination of lactulose and rifaximin for 14 days reduced ammonia production by 20% in eight patients with cirrhosis.[4]Eriksen P.L. Djernes L. Vilstrup H. Ott P. Clearance and production of ammonia quantified in humans by constant ammonia infusion - the effects of cirrhosis and ammonia targeting treatments.J Hepatol. 2023; 79: 340-348Abstract Full Text Full Text PDF PubMed Scopus (1) Google Scholar Phenylbutyrate lowers ammonia by providing an alternate pathway for nitrogen excretion in the form of phenylacetyl glutamine, which is excreted in urine. However, this could in part explain why a randomised-controlled trial of ornithine phenylacetate in patients with cirrhosis and HE was negative[17]Rahimi R.S. Safadi R. Thabut D. Bhamidimarri K.R. Pyrsopoulos N. Potthoff A. et al.Efficacy and safety of ornithine phenylacetate for treating overt hepatic encephalopathy in a randomized trial.Clin Gastroenterol Hepatol. 2021; 19: 2626-26235 e7Abstract Full Text Full Text PDF PubMed Scopus (23) Google Scholar if this compound works by increasing renal ammonia clearance, which was not dramatically impaired at baseline, rather than altering production. That said, this trial was criticised because many of the recruited patients did not have markedly elevated levels of ammonia at baseline, making it difficult to demonstrate an impact of the drug. This differs from urea cycle disorders, where fasting ammonia correlates strongly and positively with symptoms and patients may benefit from tight ammonia control, ammonia being used as a biomarker to monitor therapy. In patients with cirrhosis, measuring a single ammonia level during acute presentation with HE and in outpatients in steady state is likely to have clinical utility. However, a recent prospective multicentre study has confirmed that ammonia, included in a prognostic model, can identify outpatients at highest risk of developing liver-related complications and mortality.[18]Tranah T.H. Ballester M.P. Carbonell-Asins J.A. Ampuero J. Alexandrino G. Caracostea A. et al.Plasma ammonia levels predict hospitalisation with liver-related complications and mortality in clinically stable outpatients with cirrhosis.J Hepatol. 2022; 77: 1554-1563Abstract Full Text Full Text PDF PubMed Scopus (26) Google Scholar Thus, studying ammonia production and monitoring changes in response to therapy could have clinical utility but the method used in the present study is of course not translatable into clinical practice and could raise ethical concerns for some. Notably, this study demonstrates that a combination of lactulose and rifaximin reduced gut ammonia production by 20%. Unfortunately, the impact of lactulose and rifaximin were not studied individually, but these data support previously published studies showing that lactulose[19]Ratnaike R.N. O'Halloran M.W. Hislop I.G. Lactulose lowers arterial ammonia in hepatic failure.Am J Gastroenterol. 1972; 58: 626-632PubMed Google Scholar and, to a lesser extent, rifaximin lower blood ammonia. Whilst only studied in small numbers, Eriksen et al. illustrate the impact of these drugs on gut ammonia production (since the extent of portal-systemic blood shunting could not change) although they have little impact on basal plasma ammonia levels. This elegant study therefore confirms that the use of lactulose and rifaximin may, by reducing the systemic influx of gut-derived ammonia, have real value for the prevention HE, irrespective of the potential additional benefit they may offer in terms of restoring gut eubiosis and gut barrier integrity.[20]Patel V.C. Lee S. McPhail M.J.W. Da Silva K. Guilly S. Zamalloa A. et al.Rifaximin-alpha reduces gut-derived inflammation and mucin degradation in cirrhosis and encephalopathy: RIFSYS randomised controlled trial.J Hepatol. 2021; Google Scholar In summary, this meticulously planned study has sought to quantify whole-body clearance and production of ammonia in 10 healthy male controls and 10 patients with stable cirrhosis. It takes previously published studies of glutamine and amino acid challenge to the next level, with accurate calculations of ammonia clearance and production in both healthy individuals and patients with cirrhosis. Furthermore, they have extended the study by examining the impact of glycerol phenylbutyrate on ammonia clearance in seven healthy individuals and additionally following 14 days of rifaximin and lactulose treatment on ammonia production and clearance in eight patients with cirrhosis. Whilst being on the cusp of what some might consider ethically acceptable, such a study has not been undertaken in patients with cirrhosis for 35 years and will likely go down in the history books as a seminal study in hepatology. The authors received no financial support to produce this manuscript. DLS has undertaken consultancy for Norgine Pharmaceuticals Ltd and EnteroBiotix, and paid lectures for Norgine Pharmaceuticals Ltd. DT and PA have no conflicts to declare. Please refer to the accompanying ICMJE disclosure forms for further details. PA wrote the first draft of the manuscript which was then added to by DT and then extensively revised by DLS. The following are the supplementary data to this article: Download .pdf (.46 MB) Help with pdf files Multimedia component 1