Efficacy of oral L-ornithine L-aspartate in cirrhotic …sunwavepharma.com/studii/gastro/2015_01_04_HEPAID_AMINO.pdfS57 L-ornithine L-aspartate in cirrhosis with hyperammonemic hepatic

S55L-ornithine L-aspartate in cirrhosis with hyperammonemic hepatic encephalopathy. , 2011; 10 (Suppl.2): S55-S59

Efficacy of oral L-ornithine L-aspartate in cirrhoticpatients with hyperammonemic hepatic encephalopathy

Claudia Isabel Blanco Vela,* Jorge Luis Poo Ramírez**

* Servicio de Gastroenterología. Hospital Universitario Dr. José Eleuterio González, Facultad de Medicina. Universidad Autónoma de Nuevo León.Monterrey, Nuevo León, México. ** Departamento de Farmacología, Fundación Clínica Médica Sur. México, D.F., México.

Correspondence and reprint request: Jorge Luis Poo, M.D.Pharmacology Unit. Medica Sur Hospital and Clinical FoundationPuente de Piedra No. 150, Planta baja. C.P. 14050, TlalpanMexico City, Mexico. Tel.: (52 55) 5424-6885. Fax (52 22) 5606-1174E-mail: [email protected]

Manuscript received: May 05, 2010.Manuscript accepted: May 06, 2010.

Vol. 10 Suppl.2, 2011: S55-S59

MODULE ?

INTRODUCTION

Although the association between neurological ab-normalities and liver disease was recognized by Hippo-crates in the 5th century B.C., the pathophysiologicalmechanism responsible for these abnormalities remai-ned obscure until the late 19th century. Liver dysfunc-tion limits the metabolism of substances that areneurotoxic at high concentrations. The neurologicalmanifestations of liver disease range from subtle chan-ges in mental status to coma. This set of neurologicalchanges is called HE.

HE and its association with decompensated liverdisease is important because it has a negative im-pact on patient prognosis and quality of life, andtreatment is costly. This syndrome affects up to 50%of cirrhotic patients with advanced disease.1 Afterthe first episode of HE, the 1-year survival rate is42% and the 3-year survival rate is 23%.2 In 2003,

the total hospital cost associated with HE in theUnited States was estimated to be $900 million.3

Ammonia is the neurotoxin that triggers HE syn-drome. To date, treatment modalities have focusedon reducing ammonia produced by the bowel ratherthan on manipulating mechanisms involved in thesystemic production of ammonia.

Despite the limited number of clinical trials withhigh internal validity that support the use of nonab-sorbable disaccharides (NADs) for treating HE,NADs are still considered the drugs of choice.4

LOLA decreases ammonia concentration by stimula-ting the urea cycle and expression of glutaminesynthetase (GS). In cirrhotic patients, the synthesisof Gln via GS represents an alternative pathway fordetoxification of ammonia.

PATHOGENESIS OFHEPATIC ENCEPHALOPATHY

For many years, it was thought that most of theammonia in the body was produced by degradationof nitrogenous products by the colonic flora. Howe-ver, current evidence indicates that 85% of the in-testinal production of ammonia is the result ofphosphate-activated glutaminase activity in the

ABSTRACT

Hyperammonemia and associated cerebral edema cause neurological abnormalities in liver disease patients.Although only 15% of ammonia production originates in the colon, management strategies for hepatic ence-phalopathy (HE) have focused on reducing ammonia generation from the bowel rather than on manipulatingsystemic mechanisms involved in ammonia metabolism. Administration of L-ornithine L-aspartate (LOLA) im-proves mental status and decreases serum and spinal fluid ammonia levels by stimulating both the ureacycle and glutamine (Gln) synthesis, which are key metabolic pathways in ammonia detoxification. LOLAwas shown to be superior to a placebo for management of HE, and the results of several clinical trials sug-gest that its effectiveness could be higher with the more severe grades of this syndrome. Compared withthe standard treatment, LOLA is effective not only in reducing hyperammonemia and the severity of thisdisease, but also in improving the patient’s perceived quality of life. Therefore, LOLA is a promising alter-native for the management of HE.

Key words. Hepatic encephalopathy. Hepatic coma. Cirrhosis. L-ornithine L-aspartate. Ornithine.

Blanco Vela CI, et al. , 2011; 10 (Suppl.2): S55-S59S56

small intestine.5,6 The kidney also plays an importantrole in the generation of ammonia, and may contri-bute one-third or more of the ammonia released intothe splanchnic circulation.7

In healthy subjects, the main mechanisms invol-ved in maintaining blood ammonia concentrationwithin nontoxic limits are the urea production viathe Krebs cycle and GS-mediated synthesis of Glnin the liver.8 In cirrhotic patients, a reduction in thehepatocellular function and the spontaneous genera-tion of portosystemic shunts are responsible for hy-perammonemia.9 In this situation, the processing ofammonia and its conversion into a nontoxic com-pound are carried out through Gln synthesis in theliver, brain and muscles.8,10

Ammonia passes freely through the blood–brainbarrier (BBB), and is required for the production ofGln in astrocytes. The uptake of glutamate by theastrocyte and the production of Gln from glutamateand ammonia prevent excessive neuronal activationin healthy subjects.11,12 In liver disease, neurologicalabnormalities are associated with low-grade cerebraledema, which is secondary to hyperammonemia.The brain and spinal fluid of cirrhotic patients withhyperammonemia contain excessively high levels ofGln. Accumulation of ammonia and Gln in astro-cytes results in oxidative stress, free radical forma-tion and mitochondrial and sodium channeldysfunction, which ultimately increase intracellularosmolarity, causing edema and cerebral malfunc-tion.13-15 Consequently, brain GS does not contributeto the detoxification of ammonia. However, the con-tribution of muscle GS to ammonia detoxification is

substantial because this tissue constitutes a signifi-cant proportion of total body mass, and the Gln soproduced is the major substrate for the generationof ammonia in the kidney.10,16 Hyperammonemia re-duces the release of ammonia from the kidneys tothe splanchnic circulation and increases urinary ex-cretion of ammonia by up to 70%; that is, it beco-mes an ammonia-excreting organ and, therefore, isresponsible for the systemic removal of this neuro-toxin.10,17

MECHANISM OF ACTION OFL-ORNITHINE L-ASPARTATE

Clinical studies supporting the use of LOLA inhumans for treating HE began in Germany almost40 years ago. LOLA is a salt of the natural aminoacids ornithine and aspartic acid, and provides keysubstrates to metabolic pathways involved in the de-toxification of ammonia.18,19 The administration ofLOLA improves mental status and decreases ammo-nia levels in serum and spinal fluid by stimulatingthe urea cycle and the synthesis of Gln. After admi-nistration of LOLA, normalization of plasma ammo-nia levels is concomitant with a decrease in brainwater content, which delays the onset of neurologi-cal symptoms.20

Ornithine stimulates the activity of carbamoylphosphate synthetase I and aspartate stimulates theactivity of arginase by donating nitrogen (Figure1). Both of these enzymes are necessary for thesynthesis of urea. The administration of LOLA de-creases the plasma concentration of ammonia

Figure 1. Urea cycle.

S57L-ornithine L-aspartate in cirrhosis with hyperammonemic hepatic encephalopathy. , 2011; 10 (Suppl.2): S55-S59

and increases the plasma concentration of urea,which proves that LOLA increases the activity ofthe Krebs cycle.21,22 When liver function is impaired,ammonia that cannot be metabolized by the liver isconverted into Gln in the muscle. Thus, Gln functio-ns as a nontoxic ammonia transporter in the circu-lation.7,9,10 After administration of LOLA, serum Glnlevels increase because of the activity of muscle GS.However, levels of Gln and lactate in spinal fluidare not increased, which prevents the onset of cere-bral edema. This supports that LOLA increases Glnsynthesis in the periphery.20 The primary mecha-nism of detoxification of ammonia in cirrhotic pa-tients is the uptake of ammonia that escapes the liverby muscle and its subsequent conversion to Gln inmuscle.7,16 LOLA enhances the action of ornithineand aspartate transaminases to produce glutamate,which then promotes the synthesis of Gln by GS21

(Figure 2).Ornithine passes through the BBB, suggesting

that the central nervous system is a target for orni-thine, but the mechanism by which it exerts effects isunknown. The improvement in mental status aftertherapy with LOLA is the result not of a direct effectof ornithine on the central nervous system, but of adecrease in exposure of the brain to ammonia secon-dary to a decrease in serum ammonia level.22

EFFICACY

As HE is diagnosed clinically, mental status isconsidered the primary outcome measure for esta-blishing treatment efficacy in clinical trials. Objecti-ve clinical evaluation of HE is accomplished byanalyzing the dominant frequency of an electroence-

phalogram, P300 auditory evoked potentials and theportosystemic encephalopathy index (PSEI). This in-dex comprises electroencephalogram results, numberconnection test (NCT) results, the degree of aste-rixis, serum ammonia levels and the results of men-tal status evaluation. As minimal hepaticencephalopathy (MHE) has no apparent manifesta-tions to clinical observers, the primary outcomemeasure is based on the results of neuropsychologi-cal and neurophysiological tests.23

HE is in some cases episodic and resolved by re-moving the triggering event, making it essentialthat a drug show efficacy against a placebo beforeits effects are compared with those of an active con-trol. The superiority of LOLA over placebo as anoral treatment for the management of HE has beendemonstrated in human models.24-26

The bioavailability of LOLA when administeredp.o. is 82.2 ± 28%,27 and the efficacy of its adminis-tration by this route was supported by Stauch, etal.25 This group of researchers compared the effecti-veness of p.o. LOLA administered with a placebo inpreventing hyperammonemia induced by a high-pro-tein diet in 66 patients with chronic grade I or gra-de II HE or MHE according to the West HavenCriteria. Eighteen grams of LOLA or 10 g of fructo-se was given daily for 14 consecutive days . The pri-mary outcome measures were postprandial ammoniaconcentration and NCT performance, and the secon-dary outcome measures were mental status andPSEI. LOLA was superior to the placebo in reducingpostprandial ammonia level (p < 0.05) and in im-proving NCT performance (p < 0.01), mental status(p < 0.05) and PSEI (p < 0.01). Subgroup analysisshowed that the most pronounced clinical response

Figure 2. LOLA enhan-ces the action of ornithineand aspartate transamina-ses in brain and peripheraltissues to produce gluta-mate, which promotes thesynthesis of Gln by GS.

Blanco Vela CI, et al. , 2011; 10 (Suppl.2): S55-S59S58

to LOLA occurred in patients with grade II HE(57% of patients in LOLA group, whereas only 18%in the placebo group responded). In contrast, only9% of patients with MHE showed a response toLOLA, suggesting that the more severe the impair-ment of mental status, the greater is the effect ofLOLA. It is likely that the lack of response in theMHE group was a function of the method used fordiagnosis of this condition. The use of a set of standar-dized neuropsychological tests or neuropsychologicaltests in conjunction with P300 or electroencephalo-gram dominant frequency tests, which are capable ofdetecting subtle neurological changes in the early sta-ges of the disease, is recommended.

Kircheis, et al.19 compared the effect of intrave-nous LOLA with that of a placebo in 126 cirrhoticpatients with hyperammonemia and MHE or grade Ior II HE according to the West Haven Criteria.LOLA was superior to the placebo with respect toNCT performance (p < 0.001), postprandial ammo-nia level (p < 0.001), preprandial ammonia level (p< 0.01), mental status (p < 0.001) and PSEI (p <0.01). The differences were more pronounced in pa-tients with grade I or grade II HE than in patientswith MHE, and, unlike Stauch, et al., they usedboth NCT and electroencephalography to diagnoseMHE. This confirms the finding by Stauch that thegreater the impairment in mental status, the greaterthe impact of LOLA.

NADs are considered the drugs of first choice fortreating HE. NADs act on the colon, shortening in-testinal transit time and decreasing colon pH, whichreduces the absorption of nonionized ammonia andincreases the assimilation of ammonia by bacteria.28

As the evidence shows that only 15% of ammoniaproduction originates from the colon, the contribu-tion of NADs to the reduction of hyperammonemiais limited.

To compare the efficacy of oral LOLA with that oflactulose, 20 patients with grade I or grade II HEwere randomized by Po, et al., to receive 30 mL oflactulose or 9 mg of LOLA orally for 2 weeks. Thedoses could be adjusted up to 60 mL of lactulose andup to 18 mg of LOLA, according to researcher’s opi-nion.24 The baseline ammonia concentration decrea-sed from 120.4 ± 8.1 to 91.4 ± 10 µg/dL (p < 0.05)in the lactulose group, and from 141.6 ± 9.1 to 96.9± 9.3 µg/dL (p < 0.05) in the LOLA group. Al-though lactulose decreased baseline ammonia con-centration, it did not affect the other variables usedin the PSEI. In contrast, after administration ofLOLA, a significant improvement in mental status,NCT, asterixis, and in the electroencephalogram ac-

tivity, was observed (p < 0.05). The difference bet-ween PSEI values at baseline and those measured af-ter 2 weeks in the LOLA group reach statisticalsignificance (0.44 ± 0.03 and 0.28 ± 0.04, respecti-vely; p < 0.05). Improvement in the quality of lifewas assessed using a EuroQol visual analog scale.The EuroQol index improved after both interventio-ns (p < 0.05), the baseline and final indexes were51.1 ± 24.1 and 61.5 ± 15.8, for the lactulose groupand 56.5 ± 24.5 and 70 ± 19.4, for the LOLA group.

A meta-analysis designed to determine the effica-cy and safety of LOLA for management of HE29 in-cluded the studies of Kircheis, Stauch, and Poo andanalyzed data from 212 patients.19,24,25 According tothis report, compared with the placebo, LOLA resul-ted in clinical improvement of HE (RR, 1.89; 95% CI,1.32-2.71; p = 0.0005). Subgroup analysis showedthat LOLA was superior to the placebo in grade Iand grade II HE (RR = 1.87; 95% CI, 1.30-2.68; p <0.0007), but had no significant effect in patientswith MHE (RR = 1.69; 95% CI, 0.72-3.94; p = 0.23).

CONCLUSIONS

Currently, HE therapy is focused solely on redu-cing the amount of ammonia produced in the colon.However, evidence suggests that only 15% of am-monia of intestinal origin is produced in the colon.Therefore, there is no valid basis for using NADsas monotherapy for HE. LOLA promotes activationof the main ammonia detoxification routes and am-monia storage in the muscles as glutamine, a non-toxic carrier compound. Therefore, LOLA isinvolved in the systemic elimination of ammonia,making it an excellent therapeutic alternative.

The bioavailability of p.o. LOLA is 82.2 ± 28%,27

enabling it to be administered by this route withoutsacrificing efficacy. There is enough evidence thatsupports the superiority of p.o. LOLA over placebofor the management of HE. The studies of Stauchand Kircheis suggest that LOLA efficacy is greaterin the more severe forms of the syndrome andwarrants further examination of the effects of LOLAin grades III and IV of HE.

Controlled clinical trials comparing p.o. LOLAwith standard therapy show that LOLA is as effecti-ve as NADs in the management of HE. The main im-pact of HE is not the treatment cost, but decreasedsurvival and quality of life. The administration ofLOLA has been proven effective, not only in redu-cing hyperammonemia and the severity of this disea-se, but also in improving the patient’s perceivedquality of life.

S59L-ornithine L-aspartate in cirrhosis with hyperammonemic hepatic encephalopathy. , 2011; 10 (Suppl.2): S55-S59

ABBREVIATIONS

• HE:HE:HE:HE:HE: Hepatic encephalopathy.• LOLA:LOLA:LOLA:LOLA:LOLA: L-ornithine L-aspartate.• NADs:NADs:NADs:NADs:NADs: Nonabsorbable disaccharides.• GS:GS:GS:GS:GS: Glutamine synthetase.• Gln:Gln:Gln:Gln:Gln: Glutamine.• PSEI:PSEI:PSEI:PSEI:PSEI: Portosystemic encephalopathy index.• NCT:NCT:NCT:NCT:NCT: Number connection test.• MHE:MHE:MHE:MHE:MHE: Minimal hepatic encephalopathy.

REFERENCES

1. Das A, Dhiman RK, Saraswat VA, Verma M, Naik SR. Preva-lence and natural history of subclinical hepatic encephalo-pathy in cirrhosis. J Gastroenterol Hepatol 2001; 16:531-5.

2. Bustamante J, Rimola A, Ventura PJ, et al. Prognostic sig-nificance of hepatic encephalopathy in patients with cirr-hosis. J Hepatol 1999; 30: 890-5.

3. Poordad FF. The burden of hepatic encephalopathy. Ali-ment Pharmacol Ther 2007; 25(1): 3-9.

4. Als-Nielsen B, Gluud LL, Gluud C. Nonabsorbable disaccha-rides for hepatic encephalopathy: systematic review ofrandomized trials. BMJ 2004; 328(7447): 1046.

5. James LA, Lunn PG, Middleton S, Elia M. Distribution of glu-taminase and glutamine synthetase activities in the hu-man gastrointestinal tract. Clin Sci 1998; 94(3): 313-9.

6. Romero-Gómez M, Jover M, Galán JJ, Ruiz A. Gut ammoniaproduction and its modulation. Metab Brain Dis 2009;24(1): 147-57.

7. Olde DSW, Jalan R, Redhead DN, Hayes PC, Deutz NE,Soeters PB. Interorgan ammonia and amino acid metabo-lism in metabolically stable patients with cirrhosis and aTIPSS. Hepatology 2002; 36(5): 1163-71.

8. Haüssinger D. Nitrogen metabolism in liver: structural andfunctional organization and physiological relevance. Bio-chem J 1990; 267(2): 281-90.

9. Stahl J. Studies of the blood ammonia in liver disease. Itsdiagnostic, prognostic, and therapeutic significance. AnnIntern Med 1963; 58: 1-24.

10. Dejong CH, Deutz NE, Soeters PB. Renal ammonia and glu-tamine metabolism during liver insufficiency-induced hype-rammonemia in the rat. J Clin Invest 1993; 92(6): 2834-40.

11. Lockwood AH, McDonald JM, Reiman RE, Gelbard AS, Laug-hlin JS, Duffy TE, Plum F. The dynamics of ammonia metabo-lism in man. Effects of liver disease and hyperammonemia.J Clin Invest 1979; 63(3): 449-60.

12. Hazell A, Butterworth R. Hepatic encephalopathy: an up-date of pathophysiologic mechanisms. Exp Biol Med 1999;222: 99-112.

13. Norenberg MD, Rao KV, Jayakumar AR. Mechanisms of am-monia-induced astrocyte swelling. Metab Brain Dis 2005;20(4): 303-18.

14. Ranjan P, Mishra A, Kale R, Saraswat V, Gupta R. Cyto-toxic edema is responsible for raised intracranial pressurein fulminant hepatic failure: in vivo demonstration using

diffusion-weighted MRI in human subjects. Metab BrainDis 2005; 20(3): 181-92.

15. Gibson JB. Encephalopathy after portacaval shunt. Br MedJ 1963; 1: 1652-55.

16. Morgan KZ, Binks W, Brues AM, Cipriani AJ, Langham WH,Marinelli LD, et al. Report of International SubcommitteeII on permissible dose for internal radiation. Br J RadiolSuppl 1955; 6: 25.

17. Jalan R, Kapoor D. Enhanced renal ammonia excretion fo-llowing volume expansion in patients with well compensa-ted cirrhosis of the liver. Gut 2003; 52(7): 1041-5.

18. Leonhardt H, Bungert HJ. Therapie der schweren Hypera-mmoniämie. Med Klin 1972; 67: 1052-6.

19. Kircheis G, Nilius R, Held C, Berndt H, Buchner M, Görtel-meyer R, et al. Therapeutic efficacy of L-ornithine-L-as-partate infusions in patients with cirrhosis and hepaticencephalopathy: results of a placebo-controlled, double-blind study. Hepatology 1997; 25(6): 1351-60.

20. Rose C, Michalak A, Rao KV, Quack G, Kircheis G, But-terworth RF. L-ornithine-L-aspartate lowers plasma andspinal fluid ammonia and prevents brain edema in rats withacute liver failure. Hepatology 1999; 30(3): 636-40.

21. Gebhardt R, Beckers G, Gaunitz F, Haupt W, Jonitza D,Klein S, et al. Treatment of cirrhotic rats with L-ornithine-L-aspartate enhances urea synthesis and lowers serumammonia levels. J Pharmacol Exp Ther 1997; 283(1): 1-6.

22. Rose C, Michalak A, Pannunzio P, Therrien G, Quack G, Kir-cheis G, Butterworth RF. L-ornithine-L-aspartate in expe-rimental portal-systemic encephalopathy: therapeuticefficacy and mechanism of action. Metab Brain Dis 1998;13(2): 147-57.

23. Ferenci P, Lockwood A, Mullen K, Tarter R, Weissenborn K,Blei AT. Hepatic encephalopathy-definition, nomenclature,diagnosis, and quantification: final report of the workingparty at the 11th World Congresses of Gastroenterology,Vienna, 1998. Hepatology 2002; 35(3): 716-21.

24. Poo JL, Góngora J, Sánchez-Avila F, Aguilar-Castillo S, Gar-cía-Ramos G, Fernández-Zertuche M, et al. Efficacy of oralL-ornithine-L-aspartate in cirrhotic patients with hyperam-monemic hepatic encephalopathy. Results of a randomized,lactulose-controlled study. Ann Hepatol 2006; 5(4): 281-8.

25. Stauch S, Kircheis G, Adler G, Beckh K, Ditschuneit H,Görtelmeyer R, et al. Oral L-ornithine-L-aspartate thera-py of chronic hepatic encephalopathy: results of a place-bo-controlled double-blind study. J Hepatol 1998; 28(5):856-64.

26. Kircheis G, Wettstein M, Von Dahl S, Häussinger D. Clinicalefficacy of L-ornithine–L-aspartate in the management ofhepatic encephalopathy. Metab Brain Dis 2002; 17(4):453-62.

27. Kircheis G, Metz M, Frey S, Seiller E. Correlation betweenpharmacokinetic aspects and clinical efficacy of L-ornithi-ne–L-aspartate (OA): results of randomized, controlled cli-nical trials. J Hepatol 1996; 25(Suppl. 1): 131 (abstract).

28. Kersh E, Rifkin H. Lactulose enemas. Ann Intern Med1973; 78(1): 81-4.

29. Jiang Q, Jiang XH, Zheng MH, Chen YP. L-Ornithine-L-as-partate in the management of hepatic encephalopathy: ameta-analysis. J Gastroenterol Hepatol 2008; 24(1): 9-14.

Chavez-Tapia et al. Nutrition Journal 2013, 12:74http://www.nutritionj.com/content/12/1/74

REVIEW Open Access

A systematic review and meta-analysis of the useof oral zinc in the treatment of hepaticencephalopathyNorberto C Chavez-Tapia1*, Asunción Cesar-Arce1, Tonatiuh Barrientos-Gutiérrez2,3, Francisco A Villegas-López1,Nahum Méndez-Sanchez1 and Misael Uribe1

Abstract

Background and aim: Because low serum zinc levels precipitate hepatic encephalopathy, zinc supplementation isconsidered a potential therapeutic option. The aim of this study was to assess the effects of oral zincsupplementation in the treatment of hepatic encephalopathy.

Methods: For this systematic review and meta-analysis, data sources included electronic databases (CENTRAL,MEDLINE, EMBASE) and manual searching. Randomized clinical trials of adult patients diagnosed with liver cirrhosisand hepatic encephalopathy were included. The types of interventions considered were any oral zincsupplementation versus no intervention, placebo, or other interventions for the management of hepaticencephalopathy. The data were analyzed by calculating the RR for each trial and expressing the uncertainty as 95%CI. Continuous data were analyzed by calculating the standard mean differences (SMD) between groups withineach trial and their 95% CI. Statistical heterogeneity was defined as a P-value > 0.10 (χ2) or I2 > 25%.

Results: Four trials with a total of 233 patients were included. Oral zinc supplementation was associated with asignificant improvement in performance on the number connection test (SMD –0.62; 95% CI –1.12 to –0.11) reported inthree trials (n = 189), but not with encephalopathy recurrence reduction (RR 0.64; 95% CI 0.26 to 1.59) reported in twotrials (n = 169). Other clinically significant outcomes (mortality, liver related morbidity, quality of life) were not reported.

Conclusion: Oral zinc supplementation improved performance on the number connection test, but no evidence aboutother clinical or biochemical outcomes was available.

Keywords: Therapy, Liver cirrhosis, Evidence-based medicine

IntroductionHepatic encephalopathy is a neuropsychiatric complica-tion of liver disease that affects 20 to 30% of the patientswith cirrhosis [1,2], reducing health-related quality of lifeand causing a reversible decline in cognitive function.Previous studies have demonstrated that a reduction inblood ammonia levels improves hepatic encephalopathy,neuropsychological test performance, cognitive function,and health-related quality of life [3]. Lactulose, an am-monia absorption minimizer, has been successfully usedto reduce blood ammonia levels in minimal hepatic

* Correspondence: [email protected] and Digestive Diseases Unit, Medica Sur Clinic & Foundation,Mexico City, MexicoFull list of author information is available at the end of the article

© 2013 Chavez-Tapia et al.; licensee BioMed CCreative Commons Attribution License (http:/distribution, and reproduction in any medium

encephalopathy. However, lactulose has no ammoniadetoxification effect, rendering it ineffective to treatadvanced hepatic encephalopathy [4-6].Two major organs are involved in the metabolism of

ammonia: the liver, in which ammonia is converted tourea via ornithine transcarbamylase, and the skeletalmuscle, where ammonia is metabolized to glutamic acidvia glutamine synthetase [5]. Zinc is a critical cofactor inthese enzymatic reactions. Animal models have shownzinc deficiency decreases the activity of ornithine tran-scarbamylase, while zinc supplementation markedly in-creases hepatic ornithine transcarbamylase activity. Zincdeficiency has also been reported to impair the activity ofmuscle glutamine synthetase, which leads to hypera-mmonemia [6-8].

entral Ltd. This is an Open Access article distributed under the terms of the/creativecommons.org/licenses/by/2.0), which permits unrestricted use,, provided the original work is properly cited.

Chavez-Tapia et al. Nutrition Journal 2013, 12:74 Page 2 of 6http://www.nutritionj.com/content/12/1/74

Zinc deficiency is observed frequently in patients withcirrhosis and hepatic encephalopathy [9]. Poor nutritionalintake caused by a protein-restricted diet, impaired intes-tinal absorption, and excessive urinary loss are all poten-tial causes of a low serum zinc levels in patients withadvanced cirrhosis [5]. Short-term oral zinc supplementa-tion may improve hepatic encephalopathy by correctingthe zinc deficiency that compromises the conversion ofammonia to urea [10]. Bresci et al. reported better psycho-metric test performance in a zinc-supplemented groupthan in a standard therapy group, although the differencewas not significant [11]. Similarly, oral zinc supplementa-tion can improve hepatic encephalopathy in patients fail-ing to respond to protein restriction and lactulose [2,6-8].Zinc supplementation, in addition to standard therapies,

may increase the hepatic conversion of amino acids intourea, decrease serum ammonia level, and consequentlyimprove health-related quality of life. The effect of long-term oral zinc supplementation in addition to standardtherapy on recurrent hepatic encephalopathy has not beenestablished [7,8,12]. Despite the low cost and infrequentside effects of zinc supplementation, there is littleevidence-based information about the effects of zinc sup-plementation on hepatic encephalopathy. The aim of thismeta-analysis was to assess the effects of oral zinc supple-mentation in the treatment of hepatic encephalopathy.

MethodsTypes of studiesProspective randomized clinical trials that compared theeffects of zinc supplementation with those of no inter-vention, placebo, or standard therapy on hepatic enceph-alopathy in patients with liver cirrhosis were included.Trials were included irrespective of publication status,year of publication, or language.

Types of participantsAll adult patients diagnosed with liver cirrhosis using acombination of biochemical and clinical data, regardlessof the etiology and treatment, diagnosed with hypera-mmonemia and hepatic encephalopathy were included.

Types of interventionsStudies that compared oral zinc supplementation withno intervention, placebo, or other interventions for themanagement of hepatic encephalopathy were included.

Types of outcome measuresThe primary outcome measures were all-cause mortality,disease-specific mortality (mortality secondary to compli-cations of liver cirrhosis), and severity of encephalopathyas assessed by performance on neuropsychometric tests orrecurrence.

The secondary outcome measures were adverse events(all types of adverse events) and quality of life score(measured by any scale).

Search methods for identification of studiesElectronic searchesRelevant randomized trials were identified by searchingin CENTRAL, MEDLINE, and EMBASE.

Searching other resourcesThe references in all identified studies were inspected toidentify other trials. The first or corresponding author ofeach included trial, as well as active researchers in thefield were contacted for information about unpublishedtrials and additional information from their own trials.

Selection of studiesTwo authors independently inspected each identified ref-erence and applied the inclusion criteria. For potentiallyrelevant articles or in cases of disagreement between thetwo reviewers, the full-text article was obtained andinspected independently; if the disagreement could not besolved, a third author reviewed the article. Justification forstudy exclusion was documented.

Data extraction and managementTwo authors independently extracted the data from theincluded trials. In cases of disagreement, a third authorextracted the data. Extracted data were discussed and thisdiscussion was documented; when necessary, the authors ofthe original studies were contacted. Justification for studyexclusion was documented. Trials were identified by thelast name of the first author and the year of publication.

Assessment of risk of bias in included studiesTwo authors independently assessed risk of bias in thetrials without masking the trial names. Assessment wasconducted according to the Cochrane Handbook forSystematic Reviews of Interventions [13].

Measures of treatment effect and data analysisRevMan Analyses software was used for the statisticalanalysis [14]. Dichotomous data were synthesized andanalyzed by calculating the RR and 95% CI for each trial.Continuous data were synthesized and analyzed by cal-culating the standard mean difference (SMD) betweengroups for each trial and its 95% CI.

Assessment of heterogeneityWe checked the heterogeneity of effects across trials by vis-ual inspection of the forest plots and χ2 and I2 tests for het-erogeneity. Statistical heterogeneity was defined as P > 0.10(χ2) or I2 > 25%. Whenever heterogeneity was detected,

Chavez-Tapia et al. Nutrition Journal 2013, 12:74 Page 3 of 6http://www.nutritionj.com/content/12/1/74

subgroup analysis was performed to assess the effect ofpotential sources of heterogeneity on the main results.

Assessment of reporting biasesA funnel plot estimating the precision of trials (plot oflogarithm of the RR against the sample size) was used toevaluate asymmetry and detect potential publicationbias. In addition, Egger´s test was used to quantify thebias captured by the funnel plot [15].

Sensitivity analysisWe analyzed the data using both fixed and random-effectmodels. When both models produced similar estimates,the fixed-effect result was reported; otherwise, we re-ported the results from both analyses (Additional file 1:Figure S1). Outcomes were analyzed as reported in thetrial, either per protocol or as an intention-to-treat.

ResultsStudy selectionA total of 65 potential references were retrieved: 36 werenarrative reviews, 15 were nonrandomized studies, fourwere symposium reviews, one was a systematic review ofdifferent target trials, one was a clinical trial in animals,

Figure 1 Flow diagram of trial selection.

one was a trial in children, one was a clinical guideline,and one was a book chapter. Finally, five randomizedcontrolled trials were included in the first analysis, butone study was excluded after a second evaluation of theinclusion criteria (Figure 1).

Study characteristicsWe included four randomized controlled trials designedto evaluate oral zinc supplementation in the treatment ofhepatic encephalopathy. The number of patients who re-ceived oral zinc supplementation ranged from 20 to 90. Atotal of 233 patients from three countries, Belgium [10],Italy [11], and Japan [1,16], were included. All studies in-volved patients with cirrhosis and different stages of en-cephalopathy. The doses used were zinc sulfate 600 mg/d[16], zinc acetate 600 mg/d [10,11], or polaprezinc225 mg/d [1] (containing 51 mg of zinc and 174 mgof L-carnosine). All studies were randomized, double-blind, placebo-controlled trials (Table 1).

Risk of bias within studiesThe risk of bias was unclear in all trials. Lack of infor-mation precluded a proper evaluation of the risk of biasfor all studies.

Chavez-Tapia et al. Nutrition Journal 2013, 12:74 Page 4 of 6http://www.nutritionj.com/content/12/1/74

Synthesis of resultsGiven the large heterogeneity of outcomes across studies,the meta-analysis was restricted to two primary outcomes:number connection test performance and rate of enceph-alopathy recurrence. Patients treated with oral zinc sup-plementation experienced a significant improvement inthe number connection test performance (SMD –0.62;95% CI –1.12 to –0.11) compared with patients in theplacebo or no supplementation groups (Figure 2). Someheterogeneity of effects (I2 = 50%) was observed, andstratified analyses were conducted by year of the studyand sample size, but no change in the direction or signifi-cance of the effect was observed (data not shown). Thefunnel plot shows no evidence of publication bias(Additional file 2: Figure S2). No reduction was observedin the encephalopathy recurrence rate (RR 0.64; 95% CI0.26 to 1.59) (Figure 3).Reding et al. [10] studied the use of oral zinc supple-

mentation in a double-blind randomized trial involving22 patients with chronic encephalopathy. The zinc groupreceived zinc acetate 600 mg/d. Compared to placebo,the zinc group showed improved performance in thenumber connection test (56 ± 25.4 and 42.12 ± 16.2seconds, respectively).

Table 1 Characteristics of trials included in this systematic re

Characteristic Reding [10] Bresci [11]

Place Brussels, Belgium Pisa, Italy

Study Randomized,double-blind,

placebo-controlledtrial

Randomized,double-blind,

placebo-controlledtrial

Ran

Patients (n) 22 90

Sex (male/female) 15/7 56/34

Age (y) (placebo/zinc)* 52.7 ± 13.4/52.1 ± 9.9 49 ± 9/51 ± 9

Cause of cirrhosis(viral/alcoholic/other)

ND 50/30/10

Child–Pughclassification (A/B/C)

2/17/3 0/65/25

Hepaticencephalopathy grade(1/2)

22/0 90/0

Baseline serum zinclevels (μg/dL)[placebo/zinc]*

60.3 ± 17.9/64.5 ± 21 52 ± 5/56 ± 6

Baseline NCT(placebo/zinc)

56.2 ± 25.4/55.5 ±18.9

62 ± 10/66 ± 9

Intervention Zinc acetate 600mg/d for 7 d

Zinc acetate 600mg/d for 6 mo

Zinc<7.6

Comparison Placebo Standard therapy

Outcome reported Effect of oral zinc onHE

Effect of oral zinc onHE

Effect

ND No data, NCT Number connection test, HRQOL Health-related quality of life, HE H

Bresci et al. [11] assessed the effect of long-term zincsupplementation in 90 patients with cirrhosis with stablerecurrent hepatic encephalopathy. Oral zinc supplemen-tation (zinc acetate 600 mg/d) in addition to standardtherapy normalized the serum zinc levels. Performance inthe number connection test (40 ± 8 vs. 50 ±12 seconds),as well as in the portal systemic encephalopathy index im-proved in the treated compared to placebo group (0.15 vs.0.19). The treated group experienced less recurrence ofencephalopathy; after six months 88.6% of patients in thetreated group had no detectable signs of hepatic enceph-alopathy, compared to 86% in the placebo group.Hayashi et al. [16] reported improved nitrogen metabol-

ism in patients with liver cirrhosis after administration ofbranched-chain amino acids and zinc. Forty patients withliver cirrhosis, low serum albumin, and low zinc levelswere randomized to receive either branched-chain aminoacids alone or a combination of branched-chain aminoacids and zinc supplements. Blood ammonia levels tendedto increase in the amino acid group, while it decreased inthe supplemented zinc group (post/pre change ratio ofblood ammonia 1.22 ± 0.38 and 0.87 ± 0.26, P = 0.003, re-spectively). The Fischer ratio increased in both groups,but showed a sharper increase in the zinc-supplemented

view and meta-analysis

Hayashi [16] Takuma [1]

Osaka, Japan Okayama, Japan

domized double-blind, placebo-controlled trial

Randomized, not blinded, placebo-controlled trial

45 79

23/17 40/39

65.1 ± 11.3/66.0 ± 9.9 66.5 ± 7.4/66.5 ± 5.7

38/0/2 58/13/8

ND 1 15/49/15

ND 1 49/30

60.2 ± 9/58.4 ± 9.2 51.6 ± 13.3/48.9 ± 9.3

ND A: 72.6 ± 30.5/78.8 ± 27B: 141.6 ±31.3/145.8 ± 30.4

sulfate 600 mg/d (blood zinc levelmol/L) and 200 mg/d (blood zinc

level 7.6–10.37 mol/L)

Polaprezinc 225 mg (containing 51mg of zinc and 174 mg of L-

carnosine)/d for 6 mo

Branched-chain amino acids Standard therapy

of oral zinc in patients with hepaticcirrhosis

Effect of oral zinc on HRQOL andHE in patients with liver cirrhosis

epatic encephalopathy. *mean ± SD.

Figure 2 Forest plot for the number connection test results.

Chavez-Tapia et al. Nutrition Journal 2013, 12:74 Page 5 of 6http://www.nutritionj.com/content/12/1/74

group (post/pre change ratio 1.45 ± 0.48 and 1.67 ± 0.5,P = 0.093, respectively).Takuma et al. [1] found that zinc supplementation was

effective in treating hepatic encephalopathy and improv-ing health-related quality of life (particular physical func-tioning, role-physical, and physical component scale).Seventy-nine patients with cirrhosis and hepatic en-cephalopathy were randomized to receive 225 mg ofpolaprezinc in addition to standard therapies. After sixmonths zinc supplementation improved the PhysicalComponent Scale score (P = 0.04) and the Child–Pughscore (7.8 ± 1.6 vs. 7.2 ± 1.4, P = 0.04), and significantly de-creased hepatic encephalopathy grade (1.3 ± 0.9 vs. 0.9 ±0.9, P = 0.03) and blood ammonia levels (112.0 μg/dL ±56.3 vs. 90.4 μg/dL ± 33.4, P = 0.01). In this study one pa-tient discontinued the treatment due to an adverse event(nausea and vomiting).

DiscussionIn this meta-analysis, we included four randomized con-trolled trials evaluating the effect of oral zinc supplemen-tation over hepatic encephalopathy. Three studiesreported data on number connection test; all threeshowed an improvement in performance in the zinc groupcompared to placebo or standard therapy. This improve-ment suggests a beneficial effect of oral zinc in ence-phalopathy patients. Two studies reported data onencephalopathy recurrence rate. Both studies observedlower recurrence rates in the zinc groups, suggesting abeneficial effect of zinc; however, given the small samplesize, confidence intervals were wide and failed to reachstatistical significance.Hepatic encephalopathy is characterized at the neuro-

physiological level by disturbed corticocortical and

Figure 3 Forest plot for recurrence of hepatic encephalopathy.

corticomuscular coupling, and at the cellular level by pri-mary gliopathy [2,5,17,18]. Ammonia is a key patho-physiological factor in hepatic encephalopathy [18,19]. Inthe brain, ammonia is detoxified by astrocytes through areaction catalyzed by glutamine synthetase; an increasedbrain glutamine/glutamate ratio is associated with de-creased myoinositol, reflecting compensation for glialedema [20-23]. Swollen astrocytes predispose to neuronaldysfunction by impairing their regulatory activity againstthe increase in protein tyrosine nitration and the forma-tion of reactive oxygen and nitrogen oxide speciesincluding nitric oxide. If not counteracted, these reac-tions promote RNA oxidation, which prompts geneexpression and the transcription of altered proteins[2,5,6,18,19,21,24].Cytokines or lipopolysaccharides could induce the for-

mation of nitrogen oxide species and trigger zinc releasefrom metallothioneins, the principal zinc storage protein.A fluctuation in intracellular zinc levels modulates signaltransduction, transcription factor activity, and gene ex-pression, causing hepatic encephalopathy symptoms. Zincdeficiency is associated with disturbances in learning,memory, and emotional stability and is accompanied byhyperammonemia. Zinc supplementation has shown to re-duce ammonia levels in experimental animals and humansthrough hepatic urea synthesis stimulation and glutaminesynthesis in skeletal muscle [2,6-8,12,18,19,21,25].The present meta-analysis is limited by the small num-

ber and poor quality of trials included. Available trialsstudied heterogeneous outcomes and failed to measurecritical outcomes such as quality of life. This hinders theability to draw conclusions about the value of oral zincsupplementation in the treatment of hepatic encephalop-athy. Additionally, little information regarding the clinical

Chavez-Tapia et al. Nutrition Journal 2013, 12:74 Page 6 of 6http://www.nutritionj.com/content/12/1/74

importance of the different zinc formulations used in thetrials was available.In conclusion, oral zinc supplementation improved

performance on the number connection test, but thereis no clear evidence that supplementation improvesencephalopathy or encephalopathy-related quality of life.More trials are needed to evaluate the use of oral zincsupplementation in patients with liver cirrhosis andhepatic encephalopathy.

Additional files

Additional file 1: Fixed model.

Additional file 2: Funnel plot.

Competing interestsThe authors declare that no competing interests exist.

Authors’ contributionsNCC-T: protocol writing, searching, trial selection, data extraction, reportwriting, drafting the article, and final approval of the manuscript. AC-A:protocol writing, searching, trial selection, data extraction, report writing,drafting the article, and final approval of the manuscript. TB-G: report writing,drafting the article, and final approval of the manuscript. NM-S: reportwriting, drafting the article, and final approval of the manuscript. MU\: reportwriting, drafting the article, and final approval of the manuscript. All authorsread and approved the final manuscript.

Financial disclosureThis manuscript was partially supported by an educational grant fromMedica Sur Clinic & Foundation. The funders had no role in study design,data collection and analysis, decision to publish, or preparation of themanuscript.

Author details1Obesity and Digestive Diseases Unit, Medica Sur Clinic & Foundation,Mexico City, Mexico. 2National Institute of Public Health, Mexico City, Mexico.3Center for Integrative Approaches to Health Disparities, University ofMichigan, Ann Arbor, USA.

Received: 29 October 2012 Accepted: 4 June 2013Published: 6 June 2013

References1. Takuma Y, Nouso K, Makino Y, Hayashi M, Takahashi H: Clinical trial: oral

zinc in hepatic encephalopathy. Alimen Pharmacol Therap 2010,32:1080–1090.

2. Schliess F, Gorg B, Haussinger D: RNA oxidation and zinc in hepaticencephalopathy and hyperammonemia. Metab Brain Dis 2009,24:119–134.

3. Kobashi-Margain RA, Gavilanes-Espinar JG, Gutierrez-Grobe Y, Gutierrez-Jimenez AA, Chavez-Tapia N, Ponciano-Rodriguez G, Uribe M, MendezSanchez N: Albumin dialysis with molecular adsorbent recirculatingsystem (MARS) for the treatment of hepatic encephalopathy in liverfailure. Ann Hepatol 2011, 10(Suppl 2):S70–S76.

4. Sharma P, Sharma BC: Lactulose for minimal hepatic encephalopathy inpatients with extrahepatic portal vein obstruction. Saudi J GastroenterolJ Saudi Gastroenterol Assoc 2012, 18:168–172.

5. Haussinger D, Schliess F: Pathogenetic mechanisms of hepaticencephalopathy. Gut 2008, 57:1156–1165.

6. Brusilow SW, Traystman R: Hepatic encephalopathy. N Engl J Med 1986,314:786–787. author reply 787.

7. Marchesini G, Fabbri A, Bianchi G, Brizi M, Zoli M: Zinc supplementationand amino acid-nitrogen metabolism in patients with advancedcirrhosis. Hepatology 1996, 23:1084–1092.

8. Tuerk MJ, Fazel N: Zinc deficiency. Cur Opin Gastroenterol 2009, 25:136–143.

9. Poo JL, Rosas-Romero R, Rodriguez F, Silencio JL, Munoz R, Bourges H, UribeM: Serum zinc concentrations in two cohorts of 153 healthy subjectsand 100 cirrhotic patients from Mexico City. Digest Dis 1995, 13:136–142.

10. Reding P, Duchateau J, Bataille C: Oral zinc supplementation improveshepatic encephalopathy. Results of a randomised controlled trial. Lancet1984, 2:493–495.

11. Bresci G, Parisi G, Banti S: Management of hepatic encephalopathy withoral zinc supplementation: a long-term treatment. Eur J Med 1993,2:414–416.

12. Takeda A: Zinc homeostasis and functions of zinc in the brain. Biomet IntJ Role Metal Ions Biol Biochem Med 2001, 14:343–351.

13. Cochrane Handbook for Systematic Reviews of Interventions Version 5.0.2.www.cochrane-handbook.org.

14. Centre TNC: Review Manager (RevMan). 50th edition. Copenhagen: TheCochrane Collaboration; 2010.

15. Egger M, Davey Smith G, Schneider M, Minder C: Bias in meta-analysisdetected by a simple, graphical test. BMJ 1997, 315:629–634.

16. Hayashi M, Ikezawa K, Ono A, Okabayashi S, Hayashi Y, Shimizu S, Mizuno T,Maeda K, Akasaka T, Naito M, et al: Evaluation of the effects ofcombination therapy with branched-chain amino acid and zincsupplements on nitrogen metabolism in liver cirrhosis. Hepatol ResJ Japan Soc Hepatol 2007, 37:615–619.

17. Binesh N, Huda A, Thomas MA, Wyckoff N, Bugbee M, Han S, Rasgon N,Davanzo P, Sayre J, Guze B, et al: Hepatic encephalopathy: aneurochemical, neuroanatomical, and neuropsychological study. J AppClin Med Physics/Am Coll Med Physics 2006, 7:86–96.

18. Butterworth RF: Neurotransmitter dysfunction in hepatic encephalopathy:new approaches and new findings. Metab Brain Dis 2001, 16:55–65.

19. Norenberg MD: Astrocytic-ammonia interactions in hepaticencephalopathy. Sem Liver Dis 1996, 16:245–253.

20. Haussinger D, Laubenberger J, Vom Dahl S, Ernst T, Bayer S, Langer M,Gerok W, Hennig J: Proton magnetic resonance spectroscopy studies onhuman brain myo-inositol in hypo-osmolarity and hepaticencephalopathy. Gastroenterol 1994, 107:1475–1480.

21. Haussinger D: Low grade cerebral edema and the pathogenesis ofhepatic encephalopathy in cirrhosis. Hepatology 2006, 43:1187–1190.

22. Morgan MY: Cerebral magnetic resonance imaging in patients withchronic liver disease. Metab Brain Dis 1998, 13:273–290.

23. Ross BD, Jacobson S, Villamil F, Korula J, Kreis R, Ernst T, Shonk T, Moats RA:Subclinical hepatic encephalopathy: proton MR spectroscopicabnormalities. Radiology 1994, 193:457–463.

24. Aschner M: The functional significance of brain metallothioneins. FASEB J1996, 10:1129–1136.

25. Kruczek C, Gorg B, Keitel V, Pirev E, Kroncke KD, Schliess F, Haussinger D:Hypoosmotic swelling affects zinc homeostasis in cultured rat astrocytes.Glia 2009, 57:79–92.

doi:10.1186/1475-2891-12-74Cite this article as: Chavez-Tapia et al.: A systematic review and meta-analysis of the use of oral zinc in the treatment of hepaticencephalopathy. Nutrition Journal 2013 12:74.

Submit your next manuscript to BioMed Centraland take full advantage of:

• Convenient online submission

• Thorough peer review

• No space constraints or color figure charges

• Immediate publication on acceptance

• Inclusion in PubMed, CAS, Scopus and Google Scholar

• Research which is freely available for redistribution

Submit your manuscript at www.biomedcentral.com/submit