Gut Supplement, 1991, S29-S33

Nutritional support in liver disease

D B A Silk, S J D O'Keefe, C Wicks

It has been recognised for many years thatpatients with chronic parenchymal liver diseaseare malnourished. Dr Roger Williams and hiscolleagues at King's were one of the first toformally document nutritional status in theirpatients when O'Keefe et al' reported that sig-nificant numbers of their patients had clinical,biochemical, haematological and immunologicalvariables thought to be indicative of proteincalorie malnutrition. Furthermore their datashowed links between malnutrition, anergy,sepsis, and mortality, suggesting that effortsto improve nutritional status might improveprognosis by reducing the known high incidenceof infective complications. Subsequently otherworkers have confirmed these findings.2 AtCentral Middlesex Hospital, Keohane et alfound a marked similarity between the nutri-tional status of patients admitted for manage-ment of complications of cirrhosis and thosepatients without liver disease who in the opinionof the nutritional support team required nutri-tional support.4

Impaired dietary intake is almost certainly oneofthe principal causes for nutritional deficienciesin patients with chronic liver disease."4 Docu-mented mean protein and caloric intake was aslow as 47 g/d and 1320 kcal/d2I respectively.Reduced nutrient intake may arise as a result ofassociated anorexia and nausea and also becausedietary protein intake is often restricted fortherapeutic reasons as part of the managementof hepatic encephalopathy.'

Evidence is accumulating which suggests thatnutritional deficiencies also arise because ofimpaired digestion and absorption of nutrients.Exocrine pancreatic insufficiency has beenshown in patients with cirrhosis7 as well asmalabsorption of D-xylose, thiamine, folic acid,and fat.8 Because of the central role of the liverin nutrient metabolism, requirement levels formost nutrients change in liver failure and in someinstances, non-essential nutrients that are norm-ally synthesised or activated by the liver canbecome essential, such as choline and vitaminD.9 As the liver also acts as a store for manyvitamins, increased urinary losses contribute todeficiencies.By virtue of its oxidative and synthetic func-

tions, the liver plays a regulatory role in aminoacid homeostasis. Before his arrival at King's,O'Keefe and his colleagues showed by isotopictracer techniques that the 'catabolic' loss of bodynitrogen after elective surgery was more a conse-quence of reduced protein synthesis, secondaryto inadequate dietary intake, than the classicalview of accelerated protein catabolism that wasderived from nitrogen balance studies.'" Similarmethodology was then applied to the study ofprotein turnover in patients with cirrhosis andfulminant hepatic failure. Tyrosine, an aromatic

amino acid, and whole body protein turnoverwas measured by a constant eight hour intra-venous infusion of 14C labelled tyrosine tracer."In the fasting state, the increase in aromaticamino acid accumulation in blood was shown tobe directly related to increased endogenous pro-tein breakdown. Calculations showed that theinput of amino acids into the plasma fromendogenous protein breakdown was five timeshigher in cirrhosis, and 12 times higher inpatients with fulminant hepatic failure, thanthe normal input from dietary protein. Conse-quently, efforts to suppress protein catabolismwould be more effective than dietary restrictionin preventing the possible toxic accumulationof aromatic amino acids in the blood stream.Furthermore, direct measurements of aromaticamino acid oxidation showed that althoughhepatic oxidative clearance was reduced, it con-tinued even in fulminant hepatic failure suchthat a dietary intake of at least 60 g protein wouldbe required to maintain body protein equilib-rium. In similar studies, Swart and colleaguesused a single oral dose of '"N labelled glycine tomeasure nitrogen kinetics in stable cirrhoticpatients in the fed and fasting state.'2 Contrary totheir hypothesis and expectation, dietary proteinappeared to be more efficiently handled in cir-rhotic patients than in controls, being associatedwith increased protein synthesis rates and not theexpected increase in amino acid oxidation rates.This led them to further hypothesise thatpatients with cirrhosis have higher proteinrequirements because of an accelerated rate oftransition from the fed to fasting state becauseof diminished liver glycogen stores. Thus morerapid transition to glyconeogenesis from aminoacids would fill the energy gap resulting from lowglycogen stores. They have subsequently pre-sented evidence supporting this concept showingthat the provision of a late evening meal im-proved the efficiency of nitrogen metabolism incirrhotic patients.'3 It should be noted, however,that these studies were uncontrolled and it islikely that the same benefits will be seen innormal non-cirrhotic individuals. Furthermoreit should be noted that no kinetic study to date,including those of O'Keefe and Swart describedabove and those of Millikan'4 and Mullen,"5 hasclearly shown increased oxidative losses of pro-tein in patients with liver disease. Consequently,the high incidence of protein deficiency seen inpatients with liver disease is almost certainly aconsequence of a decreased dietary intake result-ing from the anorexia of chronic illness. Theseobservations therefore support the conclusionthat (a) efforts should be made to maintainnormal dietary intakes in patients with stablechronic liver disease and, (b) patients with moresevere liver dysfunction will require dietarymodification and nutritional supplementation.

Department ofGastroenterology andNutrition, CentralMiddlesex Hospital,LondonD B A Silk

Division ofGastroenterology, MayoClinic, Rochester,Minnesota, USAS J D O'Keefe

Kings College Hospital,Denmark Hill,London SE5C WicksCorrespondence to:Dr D B A Silk,Department ofGastroenterology andNutrition, Central MiddlesexHospital, Acton Lane,London NWIO.

S29

on August 17, 2020 by guest. P

rotected by copyright.http://gut.bm

j.com/

Gut: first published as 10.1136/gut.32.S

uppl.S29 on 1 S

eptember 1991. D

ownloaded from

Silk, O'Keefe, Wicks

Nutritional support in the presence ofhepatic encephalopathy: the dilemmaIt is a feature of the natural history of the diseasethat many patients with cirrhosis develop hepaticencephalopathy and coma at some stage duringthe course of their illness. As defective hepaticmetabolism of protein, with the accumulation ofnitrogenous compounds in the bloodstream, hasbeen implicated in the aetiology of encephal-opathy, it has been customary to restrict dietaryprotein intake even in malnourished patients.This might well exacerbate the problem in thelong term, however, as protein deficiency leadsto infections'6 and infection is one of themost common precipitating causes of encephal-opathy. 17During the last 15 years much research has

been directed towards examining the role ofaltered amino acid metabolism in the patho-genesis of hepatic encephalopathy, in particularthe relation between imbalances ofplasma aminoacid profiles and brain false neurotransmittersynthesis. New approaches to the managementof malnutrition and the treatment of encephal-opathy in patients with cirrhosis has emergedbased on the use of branched chain enrichedamino acid formulations.

NUTRITIONAL SUPPORT STRATEGIESThe literature pertaining to nutritional supportof patients with liver disease is not easy tocomprehend, because of the heterogeneity of thepatient groups studied, differences in the aimsand goals of the studies, differences in nutri-tional regimens prescribed and disagreementsabout end point objectives. The ensuing text willattempt to categorise results of work concernedwith the nutritional support ofpatients with liverdisease who are predominantly non-encephal-opathic, those with alcoholic hepatitis, thosewith cirrhosis with evidence ofcoexisting hepaticencephalopathy, those patients with fulminanthepatic failure, and in those with end stage liverdisease coming to human orthotopic liver trans-plantation.

NUTRITIONAL SUPPORT IN THE ABSENCE OFHEPATIC ENCEPHALOPATHYWhen there is no impairment of mental functionthe nutritional state of malnourished patientswith parenchymal liver disease can be improvedby standard means. The current philosophy ofthe Liver Unit at King's College Hospital is toachieve an increased nutritional intake wheneverpossible by either encouraging the patients to eatmore normal food with or without the use of oralsupplements or by instituting enteral nutritionusing standard polymeric enteral diets. Theadoption of such a philosophy is supported bythe results oftwo recent controlled clinical trials,one of which shows that nutritional intake inhospitalised patients with alcoholic liver diseasecan be increased by means of oral supplements.2In the other the clinical state and inpatienthospital mortality of enterally fed cirrhoticpatients were both significantly better comparedwith those in controls who were fed the normalward diet.3

ALCOHOLIC HEPATITISAs emphasised by the King's group6 acute alco-holic hepatitis is a potentially reversible condi-tion, but one with an overall inpatient mortalityofup to 60%. As shown by the King's group2 andby others'8 the majority of these patients are atleast moderately if not severely malnourishedcreating the rationale for nutritional therapy inthis condition. The results have been criticallyreviewed by Achord.'8 The trial by the King'sgroup6 and three others'"2' are worthy of men-tion. In the first Nasrallah and Galambos"9showed that 28 days of peripheral amino acidinfusion resulted in significant improvements ofserum albumin, bilirubin, and mortality in thetreatment group as compared with the controlgroup. In the Veterans Administration coopera-tive study20 those patients randomised to receivesupplemental enteral alimentation containingbranched chain amino acids showed an improve-ment in objective assessment of nutritionalstatus, but not mortality as compared with thosepatients randomised to an ad libitum hospitaldiet. In a further study 73 patients with alcoholichepatitis whose oral diet was supplemented withcentral venous alimentation showed no improve-ment in nutritional status nor mortality.2' Theresults of the King's study involving 64 patients6who were randomised to receive a control dietalone, or with added enteral or intravenoussupplements, no differences were noted in nitro-gen balance, nutritional status or mortality. Likeothers, the King's group are not discouraged bytheir results, they continue to appreciate theimportance of nutritional support as a means ofpreventing deterioration in nutritional status inalcoholic hepatitis while other forms of therapycan be implemented. Moreover they recognisethe imperfection ofstudy design, thus in order todetect a 25% difference within 95% confidencelimits an estimated 45 patients in each group willbe required.'8 A number not contained in their6or the other''2' prospectively randomisedstudies.

NUTRITIONAL SUPPORT IN THE PRESENCE OFHEPATIC ENCEPHALOPATHYOne of the continuing interests of the King'sgroup are the biochemical mechanisms involvedin the pathogenesis of hepatic encephalopathy,which they have reviewed'822 on more than oneoccasion. Of particular relevance to the nutri-tional support of encephalopathic patients hasbeen the demonstration of increases in plasmaconcentrations of the aromatic amino acids,phenylalanine and tyrosine, as well as methi-onine and tryptophan, these changes have beenmirrored by reductions in plasma concentrationsofthe branched chain amino acids valine, leucineand isoleucine. Much discussion has taken placeas to the cause of these changes,522 to which theKing's group have made major contributions.As discussed earlier O'Keefe and colleagues"showed that the increased aromatic amino acidconcentrations in blood were more related toincreased endogenous protein breakdown thandietary intake, plus reduced hepatic oxidativeclearance. The explanation for reducedbranched chain amino acid concentrations is less

S30

on August 17, 2020 by guest. P

rotected by copyright.http://gut.bm

j.com/

Gut: first published as 10.1136/gut.32.S

uppl.S29 on 1 S

eptember 1991. D

ownloaded from

Nutritional support in liver disease

clear. Initial studies of Munro and colleaguessuggested that the increased insulin concentra-tions that resulted from defective hepatic clear-ance, promoted the uptake of branched chainamino acids into peripheral tissues.23 Subsequentstudies by Marchesini and colleagues, however,using a euglycemic insulin infusion, failed touphold this view.24 A more likely explanation isthat the branched chain amino acids form auseful alternative energy source in patients withdefective hepatic metabolism.The relationship of the disturbed plasma

amino acid profiles to the pathogenesis of hepaticencephalopathy and the development of the falseneurotransmitter hypothesis has been reviewedin detail by the King's group52225-27 and therationale underlying the use of branched chainamino acid therapy in patients with hepaticencephalopathy originally proposed by Fischeret al2330 has been fully documented.529 Fischerand his colleagues were the first to create aspecially formulated amino acid solution forintravenous use that contained considerablydecreased amounts of phenylalanine, tyrosine,tryptophan, threonine, and glycine and in-creased amounts of the branched chain aminoacids, leucine, isoleucine and valine, as wellas arginine. After showing that survival inencephalopathic dogs receiving the formulationwas appreciably increased compared with dogsreceiving a standard intravenous formulation ofamino acids25 clinical studies were commenced.The results were encouraging for not only did theinfusion of the amino acid mixture correct theplasma amino acid profiles of patients, but theirneuropsychiatric state seemed to improve.28 Inpassing they found that positive nitrogen balancewas achieved. The innovative studies of Fischerand colleagues have stimulated much furtherclinical research into the use of amino acidmixtures enriched with branched chain aminoacids and deficient in aromatic amino acids.Horst et al3' pointed out how difficult it is to drawconclusions from the data, particularly thosederived from uncontrolled trials. The controlledtrials that have been performed have, however,evaluated different amino acid mixtures ad-ministered in different ways using differentend points in patients with different types anddegrees of encephalopathy. Moreover, thereseems little doubt that Fischer's original conceptwas concerned with the use of these speciallyformulated amino acid solutions as primarytherapy for the hepatic encephalopathy. Asappreciated by the King's group, the emphasisof this approach has shifted from primary treat-ment of the encephalopathy to safe provision ofnutritional support, which at the same time doesnot have the expected deleterious effect on thepatients neuropsychiatric state, as expressed in arecent review.32 Furthermore, from a theoreticalpoint of view, protein synthesis will be moreefficient if the composition of the precursoramino acid pool is balanced rather than contain-ing an excess of aromatic and a depletion ofbranched chain amino acids.33

Silk has recently undertaken a critical analysisof the therapeutic efficacies of branched chainamino acids in cirrhotic patients with hepaticencephalopathy29 and has emphasised the need

to consider separate therapy in those with latent,chronic, and acute forms of encephalopathy.

Latent (sub-clinical) portal systemic encephalopathyEgberts etal32 are the only group to have restrictedtheir study to patients who have never beenclinically encephalopathic and who have neverreceived dietary restriction as part oftheir clinicalmanagement. Taking into account the crossoverdesign, appreciable improvements attributable tobranched chain amino acid treatment could beshown in psychomotor function, attention, andpractical intelligence. The study lasted only aweek, however, so the benefits should be consid-ered at best to be very short term.

CHRONIC PORTAL SYSTEMIC ENCEPHALOPATHYThree groups have addressed the question ofwhether branched chain amino acids supple-mentation has a beneficial effect on the mentalstate on patients with cirrhosis and chronicportal systemic encephalopathy.3'3536 In neitherof the two small and short term crossovertrials3536 was there a question that branchedchain amino acids supplementation had anyappreciable effect on the mental state of patientsstudied. In the longer term and larger trial ofHorst et al3' the branched chain amino acidsupplemented diet, Hepaticaid seemed to have adistinct advantage. This diet was as effective ininducing positive nitrogen balance as was anequivalent amount of protein, and encephal-opathy at the same time was induced far lessoften during the four weeks of trial. As outlinedpreviously, if these observations are confirmedthen the potential role of branched chain acidsin chronic portal systemic encephalopathy willhave shifted from that of a primary therapeuticagent of encephalopathy to a means of moresafety improving the nutritional state of thesepatients.

ACUTE PORTAL SYSTEMIC ENCEPHALOPATHYThree simple randomised controlled clinicaltrials have addressed themselves to the efficacyof branched chain amino acids administeredparenterally to patients with acute portalsystemic encephalopathy.37139 In two of thetrials3738 no significant benefit of branchedchain amino acids was seen in respect of eitherencephalopathy or mortality. In the trial of Cerraet al39 promising results were obtained. A sig-nificant beneficial effect of parenterally adminis-tered branched chain amino acid mixture onencephalopathy and mortality was obtained.Only one study has compared the effect ofbranched chain enriched formulations to stan-dard formulations, however, and it is importantto note that in this study survival was actuallylonger in those given a standard formulation.'Thus, although for theoretical reasons it wouldseem better to use a specialised amino acidformula, cost effectiveness still needs to beproved, particularly as O'Keefe and colleaguesfailed to show any exacerbation of hepaticencephalopathy in patients given standard amino

S31

on August 17, 2020 by guest. P

rotected by copyright.http://gut.bm

j.com/

Gut: first published as 10.1136/gut.32.S

uppl.S29 on 1 S

eptember 1991. D

ownloaded from

S32 Silk, O'Keefe, Wicks

acid infusions.4' There is thus a clear need forfurther controlled studies in this group.

FULMINANT HEPATIC FAILUREThe King's Unit continues to have a majorinterest in the management of patients withfulminant hepatic failure and in recent yearshave addressed themselves to the effect of theillness per se and the various therapeuticmanoeuvres on the nutritional status of thesepatients. In the past, emphasis has been onmaintaining protein free intravenous glucoseinfusions in order to prevent neuroglycopeniaand reduce ammonia production, but studies atKing's have shown that despite massive hepaticnecrosis, amino acid oxidative losses continue"and may be exacerbated by artificial liver supportsystems,42 suggesting a continuing need foramino acids. As amino acids are required forhepatic regeneration, substrate supply must bemaintained. Isotope incorporation studies alsoshowed that, in contrast with the situation inpatients with chronic liver disease, the lowalbumin concentrations seen in patients withfulminant hepatic failure were a consequenceof decreased albumin synthesis. Furthermore,infusions of 14C labelled phenylalanine showedan association between reduced phenylalanineoxidation, reduced incorporation into albuminand mortality, such that no individual with anoxidation rate below 005 mmol/h and a reducedincorporation into albumin below 0*3%/d sur-vived.43 It is very doubtful that nutritional sup-port, in contrast with metabolic support, wouldhave any role in these individuals and the onlyhope would be timely liver transplantation. Onthe other hand, complete recovery was seen insome of the patients with less suppressed ratessuggesting that close attention to nutritionalsupport might form a decisive factor in theiroutcome.Most recently the King's group have esti-

mated the nitrogen requirements of thesepatients and as a prelude to establishing anutritional support programme have shown thatin contrast with previous opinions, patients withliver failure tolerate lipid emulsions well."

ORTHOTOPIC LIVER TRANSPLANTATIONAs outlined earlier in this supplement the King'sgroup have made very great strides in the area oforthotopic liver transplantation. They were thefirst to document that successful liver trans-plantation is associated with a severe state ofprotein catabolism indicating the need for earlypostoperative feeding.45 The Boston group havesubsequently measured aromatic and branchedchain amino acid metabolic rates before and afterliver transplantation and have shown that livertransplantation rapidly normalises aromaticamino acid clearance and that branched chainoxidation increases above normal."6 Thus, thereappears to be a little need for specialised aminoacid formulations after successful liver trans-plantation, but the overall requirement fordietary protein is high. Few studies haveaddressed the question of energy requirementsbut preliminary investigations from the Mayo

Clinic, based on indirect calorimetry, have failedto show increased energy expenditure during thefirst week after liver transplantation.47 TheKing's Unit are currently turning their attentionto assessing nutritional status in their patientswith end stage liver disease undergoing trans-plantation. This has been an important step totake. Recent studies do highlight how malnutri-tion is present preoperatively in these patients.48Aggressive nutritional support will probablybenefit outcome, and preliminary data certainlyfrom America already indicate how the length ofstay in the Intensive Care Unit can be shortenedby instituting early postoperative nutrition.49

1 O'Keefe SJD, El-Zayadi A, Carraher T, Davies M, WilliamsR. Malnutrition and immuno-competence in patients withliver disease. Lancet 1980; ii: 615-7.

2 Bunout D, Aicardi V, Hirsch S, et al. Nutritional support inhospitalised patients with alcoholic liver disease. EurJ ClinNutr 1989; 43: 615-21.

3 Cabre E, Gonzalez-Huiz F, Abad-Lacruz A, et al. Effect oftotal enteral nutrition on the short term outcome of severelymalnourished cirrhotics. Gastroenterology 1990; 98: 715-20.

4 Keohane PP, Attrill H, Grimble G, Spiller R, Frost P, SilkDBA. Enteral nutrition in malnourished patients withhepatic cirrhosis and acute encephalopathy. J Parent EntNutr 1983; 7: 346-50.

5 Silk DBA. Nutrition. In: Jewell PB, Mahida YR, eds. Topicsand gastroenterology 15. Oxford: Blackwell Scientific Publica-tions, 1987: 41-56.

6 Calvey H, Davies M, Williams R. Controlled trial of nutri-tional supplementation, with and without branched chainamino acid enrichment in treatment with acute alcoholichepatitis. JHepatol 1985; 1: 141-51.

7 Van Goidsenhoven GE, Henke WJ, Vacca JB, Knight WA.Pancreatic function in cirrhosis of the liver. Am J Dig Dis1%3; 8: 160-73.

8 Mezey G. Intestinal function in chronic alcoholism. Ann NYAcad Sci 1975; 252: 215-27.

9 Chowla RK, Wolf DC, Kutner MH, Bonkovsky HL. Cholinemay be an essential nutrient in malnourished patients withcirrhosis. Gastroenterology 1989; 97: 1514-20.

10 O'Keefe SJD, Sender PM, James WPT. 'Catabolic' loss ofbody nitrogen in response to surgery. Lancet 1974; 00:1034-7.

11 O'Keefe SJD, Abraham R, Zayadi A, Marshall W, Davies M,Williams R. Increased plasma tyrosine concentrations inpatients with cirrhosis and fulminant hepatic failure asso-ciated with increased plasma tyrosine flux and reducedhepatic oxidation a capacity. Gastroenterology 1981; 81:1017-24.

12 Swart GR, Van den Berg JWO, Wattimena JLD, Rietveld T,Van Vuure JK, Fraenkel M. Elevated protein requirementsin cirrhosis of the liver, investigated by whole body proteinturnover studies. Clin Sci 1988; 75: 101-7.

13 Swart GR, Zillikens MC, van Vuure JK, van den Berg TWO.Effect of a late evening meal on nitrogen balance in patientswith cirrhosis of the liver. BrMedJr 1989; 299: 1202-3.

14 Millikan WJ, Henderson JM, Galloway JR, et al. In vivomeasurement of leucine metabolism with stable isotopes innormal subjects and in those with cirrhosis fed conventionaland branch amino acid enriched diets. Surgery 1985; 98:405-12.

15 Mullen KD, Denne SC, McCullough AJ, et al. Leucinemetabolism in stable cirrhosis. Hepatology 0000; 6: 622-30.

16 Scrimshaw NS. Protein deficiency and infective disease. In:Munroe HN, Allisson JB, eds. Mamalian protein mnetabolism.New York and London: Academic Press: 1974; vol II,chap 23.

17 Lanthier PL, Morgan MY. Lactitol in the treatment of chronichepatic encephalopathy: on open comparison with lactulose.Gut 1985; 26:415-20.

18 Achord JL. Malnutrition and the role of nutritional support inalcoholic liver disease. AmJ Gastroenterol 1987; 82: 1-7.

19 Nasrallah SM, Galambos JT. Amino acid therapy of alcoholichepatitis. Lancet 1980; ii: 1276-8.

20 Mendenhall C, Bongiovanni G, Goldberg S, et al. VAco-operative study on alcoholic hepatitis. III: Changes inprotein calorie malnutrition associated with 30 days ofhospitalization with and without enteral nutrition therapy.ParentEntNutr 1985; 9: 590-6.

21 Naveau S, Pelletier G, Poynard T. A randomized clinical trialof supplementary parenteral nutrition in jaundiced alcoholiccirrhotic patients. Hepatology 1986; 6: 270-4.

22 Crossley IR, Wardall EN, Williams R. Biochemical mech-anisms of hepatic encephalopathy. Clin Sci 1983; 64:247-52.

23 Munro HN, Fernstrom JD, Wurtman RJ. Insulin, plasmaamino acid imbalance and hepatic comas. Lancet 1975; 00:986.

24 Marchesini G, Forlani G, Zollizetto-Eli M, et al. Effect ofeuglycemic insulin infusion on plasma levels of branch chainamino acids in cirrhosis. Hepatology 1983; 32: 184-7.

25 James JH, Ziparo V, Jeppsson B, Fischer JE. Hyperamino-semis, plasma amino acid inbalance, and blood brain amino

on August 17, 2020 by guest. P

rotected by copyright.http://gut.bm

j.com/

Gut: first published as 10.1136/gut.32.S

uppl.S29 on 1 S

eptember 1991. D

ownloaded from

Nutritional support in liverdisease S33

acid transport: a unified theory of portal/systemic encephal-opathy. Lancet ii: 772-5.

26 McCullough AJ, Mullen KD, Tavill AS. Branched chainamino acids as nutritional therapy in liver disease: death orsurfeit. Hepatology 1983; 3: 269-71.

27 Fischer JE. Amino acidsin hepatic coma. DigDisSci 1982; 27:97-102.

28 Fischer JE, Rosen HM, Ebeid AM, James JH, Keane JM,Soeters PB. The effect of normalisation of plasma aminoacids on hepatic encephalopathy in man. Surgery 1976; 80:77-91.

29 Silk DBA. Branched chain amino acids in liver disease: fact orfantasy? Gut 1986; 27: 103-10.

30 Fischer JE, Funovics JM, Aguirre A, et al. The role of plasmaamino acids in hepatic encephalopathy. Surgery 1975; 70:276-90.

31 Horst D, Grace ND, Conn HO, et al. Comparison of dietaryprotein with an oral, branched chain-enriched amino acidsupplement in chronic portal systemic encephalopathy. Arandomised controlled trial. Hepatology 1984; 4: 279-87.

32 Blackburn GL, O'Keefe SJD. Editorial. Nutrition in liverfailure. Gastroenterology 1989; 97(4): 1049-51.

33 O'Keefe SJD, Ogden J, Dicker J. Enteral and parenteralbranched chain amino acid-supplemented nutritional sup-port in patients with encephalopathy due to alcoholic liverdisease. JParent Ent Nutr 1987; 11(5): 447-53.

34 Egbergts EH, Schomerus H, Hamster W, Jurgens P.Branched chain amino acids in the treatment of latentportal systemic encephalopathy. A double-blind placebo-controlled crossover study. Gastroenterology 1985; 88:887-95.

35 Eriksson LS, Person A, Wahren J. Branched chain aminoacids in the treatment of chronic hepatic encephalopathy.Gut 1983; 23: 801-93.

36 McGhee A, Henderson JM, Milhikan WJ, et al. Comparison ofthe effects of hepatic aid and a case in modular diet onencephalopathy, plasma amino acids and nitrogen balance incirrhotic patients. Ann Surg 1983; 197: 288-93.

37 Rossi-Fanelli F, Riggio 0, Cangiano C, et al. Branched chainamino acids vs latulose in the treatment of hepatic coma:a controlled study. Dig Dis Sci 1982; 27: 929-35.

38 Wahren J, Denis J, Desurmont P, et al. Is intravenousadministration ofbranched chain amino acids effective in the

treatment of hepatic encephalopathy? A multicenter study.Hepatology 1983; 3: 475-80.

39 Cerra FB, Cheung NK, Fischer JE, et al. Disease-specificamino acid infusion (FO80) in hepatic encephalopathy:A prospective randomized, double blind controlled trial.J Parent Ent Nutr 1985; 9: 288-95.

40 Michel H, Pomier-Layrargues G, Aubin JP, et al. Treatmentof hepatic encephalopathy by infusion of a modified aminoacid solution: results of a controlled study in 47 cirrhoticpatients. In: Capocaccia L, Fischer JE, Rossi-Fanelli F, eds.Hepatic encephalopathy in chronic liver failure. New York:Plenum, 1984: 301-10.

41 O'Keefe SJD, Ogden J, Ramjee G, Moldawer LL. Short-termeffects of an intravenous infusion of a nutrient solutioncontaining amino acids, glucose and insulin on leucineturnover and amino acid metabolism in patients with liverfailure. J Hepatol 1988; 6(1): 101-8.

42 Chase RA, Davies M, Trewby PN, Silk DBA, Williams R.Plasma amino acid levels in patients with fulminant hepaticfailure treated with polyacrylonitrile haemodialysis. Gastro-enterology 1978; 75: 1033-40.

43 O'Keefe SJD, Ogden J, Rund J. The use of '4C labelledphenylalanine to trace deranged aromatic amino acid meta-bolism in liver failure: A functional indicator with prog-nostic potential. Gastroenterology 1989; %(5): A641.

44 Forbes A, Wicks C. Fulminant hepatic failure nutrition and fatclearance. RecentAdvances in Nutrition 1990; 1: 67-9A.

45 O'Keefe SJD, Abraham RR, Davis M, Williams R. Proteinturnover in acute and chronic liver disease. Acta Clin Scand1981; suppl 507:91-101.

46 Shanbhogue RLK, Bristrian BR, Lakshman K, et al. Wholebody leucine phenylalanine and tyrosine kinetics in endstage liver disease before and after hepatic liver transplant-ation. Metabolism 1987; 36(11): 1047-53.

47 Di Cecco SR, Plevak DJ, Wiesner RH, O'Keefe SJD, et al. Dowe accurately estimate caloric and nitrogen requirements inthe liver transplant patient? Hepatology 1990; 12(4): 978.

48 Di Cecco SR, Wieners EJ, Wiesner RH, Southorn PA, PlevakDJ, Krom RAF. Assessment of nutritional status of patientswith N-stage liver disease undergoing liver transplantation.Mayo Clin Proc 1989; 64: 95-102.

49 Reilly J, Mehta R, Teperman L et al. Nutritional support afterliver transplantation: a randomised prospective study.J3ParentEntNutr 1990; 14: 386-91.

on August 17, 2020 by guest. P

rotected by copyright.http://gut.bm

j.com/

Gut: first published as 10.1136/gut.32.S

uppl.S29 on 1 S

eptember 1991. D

ownloaded from