The role of amino acids and synthetic dipeptides

ARTICLE IN PRESS Clinical Nutrition (2003) Supplement 2: S23–S28 r 2003 ESPEN. Published by Elsevier Ltd. All rights reserved. doi:10.1016/S0261-5614(03)00147-X

The role of amino acids and synthetic dipeptides P. FUºRST University of Hohenheim, Institute for Biological Chemistry and Nutrition, Stuttgart, Germany

reports. The likely requirements for intravenous amino acids were clearly spelt out and numerous commercial preparations (mostly for adults) became available in Europe. The composition of these preparations, however, was mainly based on oral requirement data in healthy man. The major nutrition efforts were directed to improve tolerance of the nitrogen load in the presence of illness rather than to provide specific nutrients for individual organs or tissues. At most instances, the efficacy of an amino acid therapy was monitored in blood (plasma). However, as shown over the past 20 years, changes in plasma-free amino acid concentrations may parallel those in intracellular compartment, but they frequently differ either quantitatively or qualitatively. The importance of this area was recognised early in the history of ESPEN when in the 1984 Arvid Wretlind lecture, P. Fu¨rst discussed this subject ‘Regulation of intracellular metabolism of amino acids’ (4). It could be clearly demonstrated that the intracellular pattern of stress is little influenced by the available nutritional measures (Askanazi, Bergstro¨m, Cynober, Fu¨rst, Roth, Vinnars, Wernerman). It was concluded that nutritional imbalances and antagonisms might explain the poor efficacy of available nutrition preparations. Indeed, the imbalance of cellular amino acids during catabolic stress is one of the fundamental inadequacies in clinical nutrition. The combined, results over 20 years suggest that the amount and proportion of amino acids in the available preparations are not optimally composed for trauma, injury and sepsis (for reference cf 5). Measures to develop suitable preparations on the basis of intracellular amino acid concentrations for the catabolic malnourished, chronically or critically ill patient have been proposed (5). The traditional classification of essentiality of an amino acid has to be modified. In the Sir Cuthbertson Lecture in 1988, Al Harper brought forward new definitions concerning indispensable, dispensable and conditionally indispensable amino acids (6). As shown in subsequent ESPEN meetings, this new approach recognises functional and physiological properties of a given substrate under various pathological states (Millward, G. Grimble, Reeds, V. Young). It is proposed that, regardless of the definition used, a final judgement of the usefulness of an essential new substrate, will be on

The birth of ESPEN Since the work of Cuthbertson in the 1930s (1), increased urinary excretion of nitrogen and enhanced demand of energy have been considered hallmarks of physical injury. Depending on the severity of the catabolic insult, the loss of nitrogen ranged from 2 to 35 g/day. The source of increased urinary nitrogen excretion was almost exclusively the body cell mass. The profound loss of metabolically active protein certainly contributed to the development of serious, organ mal- or dysfunction and resulted in impaired host defence, thereby limiting the efficacy of nutritional and therapeutical efforts. It has been repeatedly postulated that provision of nutrients, especially an adequate supply of amino acids, helps to prevent or minimise the loss of protein stores and assists in the restoration of wasted tissue. That was the scenery of the late 1970s; increased knowledge in nutritional science enabled novel ideas and also implications of clinical trials with nutrients. Many investigators have proceeded on the assumption that ‘tailor-made solutions’ will increase the benefits of intravenous nutrition for specific patient groups. Thus, specific amino acid mixtures have been developed for treatment of renal and liver disease, or to optimise the growth of young infants. Other investigators still hoped for the use of solutions enriched in branched-chain amino acids (BCAAs) (for reference cf. (2, 3)). In 1979, ESPEN probably served as a perfect umbrella for all new developments and trials to systemise actual achievements in the clinical setting. Indeed, with the discovery of common and specific mechanisms for alterations in substrate metabolism, unique opportunities arised to intervene in the disease process. Undoubtedly, the efficacy of providing substrates to the injured, immunocompromised and/or malnourished host has caused a new birth and awakening in the clinical application of dietary intervention in the treatment and prevention of disease.

Requirements—amino acid imbalances Ammo acid requirements in premature infants in uremic state and in adults were the scope of early ESPEN S23

ARTICLE IN PRESS S24

AMINO ACIDS AND SYNTHETIC DIPEPTIDES

the grounds of clinical and nutritional efficacy. According to a more general position, ‘a possible and useful direction might put more emphasis on metabolic control and its regulation of tissue and organ function and nutritional status’ (2, 7). This definition offers suggestions as to how certain metabolic characteristics shared by some substances might be used to differentiate the various nutritionally significant substrates This also would mean that the dietary essentiality of a given substrate is dependent on the ratio of supply to demand; the distinction between ‘essential’ and ‘non-essential’ largely disappears because it is dependent on conditions.

The branched-chain amino acid story The recognition of the specific biochemical/physiological effect of certain amino acids was the start signal for the clinical implication of BCAA. Some of these ideas and applications were appropriate and they serve as a basis for current therapeutical measures. Others, however, turned out to be less valuable. The major hypothesis—partly doubtful—during the 1980s claimed that BCAA-enriched formulas arc beneficial for treatment of encephalophathy, sepsis and cancer. It could be demonstrated that BCAAs competitively inhibit the transport of automatic amino acids precursors of false neurotransmitters in the brain. Despite a great number of reports from Denmark, Israel, Italy, Holland, Sweden and USA, the superiority of BCAA-based preparations in conventional amino acid mixtures or alimentary proteins was not definitively proved (DeFronzo, Eriksson, Holm, Leverve, Mucsari, Plauth, Rossi-Fanelli, Takala), because BCAAs are important fuels in catabolic conditions, it was speculated that use of BCAAenrichcd preparations would be of value in stress, sepsis and especially in MOF. It was repeatedly postulated that BCAA might serve as a preferential substrate during critical illness, when glucose and lipid oxidations are decreased. Nevertheless, the beneficial use of BCAAenriched preparations still remains controversial (8). The stimulatory effect of BCAAs on the central respiratory drive probably accounts for the increased ventilatory response to CO2 as shown in presentations from Finland and USA. It is still doubtful whether this above effect of BCAAs is beneficial or not (Askanazi et al., Takkala). Certainly in patients with respiratory muscle weakness, this could be deleterious, by increasing the risk of respiratory muscle fatigue (Roussos). Attempts to use BCAA-enriched preparations in clinical practise were accompanied by important biochemical and physiological research in order to elucidate basic mechanisms of these amino acids. Valuable information about regulation (stable isotopes) (Bier, Matthews, Renee), interorgan exchange (Wahren), intracellular BCAA (Fu¨rst), protein turnover (Garlick, V, Young, Waterlow), BCAA–glutaminc interrelation-

ship (Darmaun, Rennie, Roth, Wernerman, Vinnars, Wernerrman) was reported during the years.

Glutamine depletion—a hallmark of the response to stress In 1975, Vinnars et al. reported that operative trauma was associated with marked depletion of the muscle free glutamine pool (9). This finding was also reported at the first ESPEN congress and was then confirmed by others. Interestingly, the original hypothesis—presented at ESPEN 1979 and 1981—was completely erroneous. It stated that the fate of muscular glutamine is to supply hepatic processes with carbon and nitrogen and any deficit could limit these processes during catabolic stress (Askanazi, Fu¨rst, Vinnars). Over the past 20 years, increased understanding of glutamine (patho)-physiology and (patho)-biochemistry resulted in a revised hypothesis; Glutamine may primarily serve as a nutrient necessary for support of mucosa, immune cells and probably lung. Increased intestinal requirement of metabolic fuel might be marked by an enhanced muscle glutamine efflux resulting in an intracellular depletion in this tissue (Calder, Fu¨rst, Roth, Souba, Wernerman, Wilmore). This revised hypothesis is underlined with numerous studies presented at ESPEN during the years. Stress induced alterations in the interorgan flow (Darmaun, Rennie, Roth, Saito, Soeters). Muscle and presumably lung glutamine efflux are accelerated in order to provide substitute for the gut, immune cells and the kidneys (Rennie, Souba, Wilmore) explaining the profound muscle free glutamine depletion. In 1984 and 1985, a striking direct correlation between muscle free glutamine concentration and the rate of protein synthesis was presented by British scientists (10, 11). This means that maintenance of the intracellular glutamine pool may promote conservation of muscle protein during catabolic stress. Thus, glutamine supplements might be beneficial in the treatment of stressed and malnourished patients. Numerous experimental studies performed between 1988 and up to date support this hypothesis. As critically evaluated at the meeting in Milano, two specific chemical/physical properties prevent the inclusion of free glutamine in commercially available amino acid preparations: the quantitative decomposition of aqueous glutamine to the cyclic product pyroglutamic acid associated with ammonia liberation and the limited solubility in water (Fu¨rst, Stehle). Therefore, glutaminecontaining TPN solutions must be prepared freshly under strict aseptic conditions and stored for a maximum of 2 days at 41C. To diminish the risk of precipitation, the glutamine concentrations in such solutions should not exceed 2.5% (Hardy, Khan). This means that provision of adequate amounts of glutamine to injured or critically ill patients with such a low concentrated solution represents a severe burden,

ARTICLE IN PRESS CLINICAL NUTRITION

especially in volume-restricted situations. Consequently, the parenteral use of free glutamine is reserved for controlled clinical trials only; such studies presented during 1989–1998 showed improved nitrogen balance compared with control groups (Essen, Hammarqvist, Pettersson, Wilmore, Ziegler). Many studies showing better clinical outcome with supplemental glutamine were presented during the last 15 years. At the ESPEN, Amsterdam, 1997, a sensational clinical study showed reduced 6-month mortality in critically ill patients, with a decrease in treatment costs comparing glutamineenriched parenteral nutrition with isonitrogenous, isocaloric controls (12). At present, there is much controversy concerning the benefit of enteral glutamine (dipeptide) nutrition (13). In many presented studies, no convincing biological or clinical results were obtained in traumatised or intensive care unit (ICU) patients despite considerable supply of glutamine (15–30 g/day) (Bozetti, Darmaun Elia, Jebb, Jensen, Jones) contrasting with a recent Dutch randomised study (Hondijk, van Leeuwen) demonstrating reduced pneumonia, bacteraemia and severe sepsis. The results of this fascinating study request urgent confirmation. There is some evidence that the body glutamine pool is slower to recover with the same dose of glutamine given enterally (orally) 03 parenterally (fish). The enteral route may be ideal when given early to the non-infected patient to improve gut-associated lymphoid tissue (GALT) function and the immune defence against infection. For the already severely stressed or infected patient in the ICU, enteral supplements alone may be inadequate, and parallel parenteral support is likely to be required: It has been clearly demonstrated that, during intensive care, the patient’s parenteral supplementation of enteral nutrition does not increase the risk to the patient and may even ensure a better overall outcome (14). The appraisal of enteral glutamine will certainly be accomplished in the coming ESPEN meetings.

New substrates in protein nutrition According to the classical statement of Wretlind: ‘Thc general approach to the nutritional care of critically ill, malnourished or stressed patients involves delivery of a balanced diet, including an adequate amount of protein or a suitable amino acid preparation that reflects a high biological value like egg protein’ (15). This approach, however, is not feasible in clinical practise today because poor solubility and/or instability prevents inclusion of glutamine, tyrosine and cysteine into the presently available parenteral amino acid preparations. These obvious limitations have initiated an intensive search for alternative substrates. Indeed, dipeptides are perfect candidates. In 1983, two new synthetic-stable and highly

S25

soluble tyrosine and glutamine-containing dipeptides were presented. Both dipeptides were perfectly suitable to be used in clinical nutrition (Fu¨rst, Stehle). At the same meeting, Wilmore brought convincing experimental evidence forward that glutamine supplementation might preserve body glutamine pools. In the following years, dipeptide research was a constant component of the meetings. In 1985, a dipeptide symposium was organised by Adibi and Fu¨rst; data in humans, baboons, dogs and rats were presented (Lochs, Neuha¨user, Roth, Stehle, Vasques). In 1986, the first two reports were delivered in healthy man as a result of a cooperation study from Germany and Sweden (Fu¨rst, Stehle, Wernerman and Vinnars) and in Leipzig, 1988, the first clinical study with glutamine and tyrosine containing dipeptides was presented (Fu¨rst, Mertes, Puchstein, Stehle). In 1995– 1996, commercial preparations with glutamine and tyrosine containing dipeptides became available. Tyrosine nutrition in renal failure was initiated by Austrian scientists in 1985 (Druml, Hu¨bl and Roth). Indeed, it could be shown that supplemental dipeptides are of value in the treatment of acute and chronic renal failure. Therapy with the new tyrosine-dipeptide-containing solution corrected tyrosine depletion and normalised the phenylalanine/tyrosine ratio. In 1998, the first multicentre trial with a glutamine dipeptide (Dipeptivens) involving 16 European centres was presented; reduced hospital stay and PINI score as well as better nitrogen balance and maintained plasma-free amino acid concentrations were found in the dipeptide group (Fu¨rst, Mertes).

Future perspectives with glutamine (dipeptide) nutrition The major issue for the future might, be tissue glutathione synthesis as a crucial factor in causing reversal of the clinical, biochemical signs of critical illness. The biochemical basis for this measure tests in the fact that the highly charged glutamic-acid molecule, one of the direct precursors of glutathione, is poorly transported across the cell membrane, whereas glutamine is readily taken up by the cell. Glutamine is then deaminated and thus can serve as glutamic-acid precursor. New methods to measure glutathione (Kuhn, Luo, Wernerman) enabled to recognise muscle glutathione depletion in the critically ill and in surgical trauma (Luo, Wernerman). At future ESPEN meetings we might hear about how glutamine (dipeptides) preserve and/or restore tissue glutathione concentrations. A second, newly discovered topic of interest is modification of the endogenous inflammatory response (Calder, Deutz, Newsholme, Roth, Wilmore); the attenuation of the elaboration of proinflammatory mediators and upregulation of antiinflammatory

ARTICLE IN PRESS S26

AMINO ACIDS AND SYNTHETIC DIPEPTIDES

factors, with glutamine dipeptides being of particular interest. Supplemental glutamine (dipeptides) decrease cytotoxicity and the synthesis of tumour necrosis factora and interleukin 8, whereas they enhance the ability to express the antiinflammatory interleukin-10 (de Beaux, Morlion). An encouraging perspective is the novel finding that provision of alanyl-glutamine promotes the synthesis of cysteinyl-leukotrienes and that the capacity to generate cysteinyl-leukotrienes was associated with outcome (Morlion). Provocatively, this ability in critical illness is proposed to be a biomarker for survival. The question might be raised as to whether the combined deficiency of glutathione and glutamine is intrinsically related to the leukotriene C4-synthetising capacity of the sick cell (16). Future implications of glutamine (dipeptide) therapy are full of promise. A consistent observation is that glutamine-enriched parenteral feeding attenuates the expansion of extracellular and total body water as presented by Scheltinga and Wilmore at 1991 ESPEN. This interesting finding suggests that provision of glutamine (dipeptide) may influence stress-induced accumulation of extracellular fluid by affecting membrane function and thereby changing the cellular hydration state. This fascinating hypothesis was the scope of the 1994 Wretlind lecture by D. Ha¨usinger (17). Another fascinating approach first presented in 1997 proposes glutamine (dipeptide) as a suitable cardioprotective and rescue agent (Rennie). The mechanism(s) through which glutamine exerts its beneficial effects may include maintenance of myocardial glutathione—change of the ratio reduced oxidised glutathione (Fu¨rst, Rennie) and myocardial high-energy phosphates as well as prevention of myocardial lactate accumulation. The implication of these results could include the possible use of glutamine to support the heart during reperfusion initiated by thrombolysis or coronary angioplasty in patients with acute myocardial infarction. The challenge for the future will be to put the complex knowledge gained from available investigations into the clinical context.

Alternative sources N-acetylated amino acids Only 1 year after the introduction of dipeptides Nacetylated amino acids N-acetyl-cysteine, N-acetyl-tyrosine and N-acetyl-glutamine have been suggested as alternative source of cysteine, tyrosine and glutamine. It could be shown that in the rat they are rapidly taken up and hydrolysed by acylases after their parenteral administration (Ba¨ssler, Neuha˚user). The human use of N-acetylated amino acids is a remarkable story, starting with an erroneous conclusion in 1984 (Ekman, Kihlberg), followed by still erroneous reports in healthy man in 1987 (Magnusson, Wahren). Finally, in 1987,

carefully conducted studies from Swedish and Austrian scientists showed clearly that infusion of the three Nacetylated amino acids resulted in their accumulation in body fluids, the corresponding plasma concentrations were not influenced and about 50% of the infused acetylatcd amino acids were excreted in the urine (Druml, Ekman, Lochs, Magnusson, Roth, Wahren, Wernerman). It was concluded that N, acetylated amino acids are poorly utilised in humans due to restricted acylase capacity (except in kidneys) in this species (for reference cf. 2). Short-chain protein hydrolysates Purified short-chain protein hydrolysates (67% di- and tri-peptides, 10% free amino acids) have been first proposed at the Leipzig meeting in 1988 as a low osmotality alternative to free amino acids and dipeptides (G. Grimble). Subsequently, it could be shown in healthy man that intravenous infusion of the shortchain hydrolysate was associated with peptide excretion corresponding to 6% of total nitrogen. Marked differences between infused and excreted peptide profiles indicated that utilisation of peptides from the hydrolysate was sequence specific (G. Grimble, Silk). Ornithine a-ketoglutarate (OKG) OKG was introduced at the ESPEN in 1987 (Cynober). This salt of ornithine and a-ketoglutarate might increase the synthesis of arginine, proline and polyamines. Whether OKG is a true precursor for glutamine is questionable. In the rat, altered concentrations of tissue amino acids were found after provision of OKG (Cynober, Jevandum) and in burned rats and patients protein status was improved (Cynober, Donali).

Pharmacological immunonutrition The interrelationship between nutrition and immune system has become at ESPEN the focus of ever increasing attention. Increasing number of substrates have been identified to which an immune-modulating function could be attributed. Amino acids like glutamine, arginine, cysteine, taurine were considered as important immune-modulating substrates. Indeed, the substitution of immune-modulating amino acids clearly exceeds the amount used in a simple prevention of deficits. However, pharmacological ‘immunonutrition’ should simultaneously satisfy both the metabolic and immunological needs of the patient. At present, some enteral formulae are available for clinical setting enriched with selected immune-modulating nutrients; the amino acid constituents are glutamine and arginine (Daly, Evoy, Kemen, Soeters, Suchner). The major targets consider mucosal barrier structure and function,

ARTICLE IN PRESS CLINICAL NUTRITION

the cellular defense function and the local or systemic inflammatory response. Obviously, glutamine-mediated glutathione synthesis might be one of the most important factors in the systemic inflammatory response. It is to propose that tissue glutathione synthesis is a crucial factor in causing reversal of the clinical biochemical signs of critical illness (other glutamine-related results vide supra). It is notable that until the early 1990s, parenteral arginine was considered a novel and valuable tool to improve immunity and to beneficially influence metabolism and pathophysiology in cancer and trauma (18). Remarkably, in the last 5 years the intravenous arginine approach is almost absent, whereas great emphasis has been placed on enteral arginine nutrition. Presumably, the reportedly highlighted drawbacks and disadvantages with large amounts of parenteral arginine have been slowly recognised and considered. One potential complication of arginine administration is its known competition with the essential amino acid lysine for tubular reabsorption. Thus, large doses of’ arginine may theoretically induce lysine deficiency by increasing its renal excretion (2). In addition, the role of arginine as a substrate for nitric-oxide synthesis and the observations implicating nitric oxide in the pathogenesis of septic-shock syndrome suggest that excess arginine supplementation could be hazardous in severely ill patients (19). Clinical studies administering arginine enterally have demonstrated moderate net nitrogen retention and protein synthesis compared with isonitrogenous diets in critically ill and injured patients; after surgery for certain malignancies in elderly postoperative patients, supplemental arginine (25 g/d) enhanced immunity (Barbul, Daly, Kemen). Interestingly, insulin like growth factor-1 levels were about 50% higher, reflecting the growth-hormone secretion induced by arginine supplementation. High load of oral arginine (30 g/d) improved wound healing (Barbul). It is probable that the observed beneficial effects of these substrates were due to improved function of the immune system rather than to improved gut-barrier function. Nevertheless, available clinical trials failed to demonstrate improvements in patient outcome (20). There is also some concern that arginine may enhance the systemic inflammatory response due to an enhanced N release in patients with severe SIRS or sepsis. This would mean a negative iono- and chronotropism of the myocardium, impaired, coagulation and vascular dilatation leading to refractory hypotension (Lo¨venstein, Lorente, Moncada, Palmer, Suchner). Apparently, NO may exert cytotoxic effects as a non-specific effector inhibiting growth or killing off cells in untargeted fashion (Fontecare, Koch, Lopoivre, Lo¨venstein, Wink). On the other hand, according to current knowledge, NO synthase and NO-mediated immunofactors as well as intracellular arginase are restricted to distinct compartments; thus, supplemental arginine may not affect extracellular NO

S27

concentration (Moncada) except in critical illness (21). It is notable that in critical illness, mucosal barrier and cellular defense are impaired and a reinforcement with complex enteral immunonutrition is desirable, while local or systemic inflammatory response should be downregulated by nutritional interventions. Improvement in outcome, however, was only seen when critical amounts of the immune-modulating formula were tolerated in patients stratified malnourished. However, in other patients with severe sepsis, shock and organ failure, no benefit or even disadvantages from immunonutrition were reported. In such severe conditions, one may hypothesise that systemic inflammation is undesirably intensified by arginine and unsaturated fatty acids directly affecting cellular defense and inflammatory response. Consequently, in patients suffering from SIRS great caution should be exercised when implicating immune-enhancing substrates which concurrently may aggravate systemic inflammation (21). One of the major tasks in the 21st century ESPEN will be to critically scrutinise the role of immuno-enhancing amino acids.

Conclusion This compilation may illustrate the fascinating 20 years of ESPEN and demonstrate how far we have advanced in our knowledge of amino acid and dipeptide metabolism and their clinical implications. Attempts were made to highlight the important areas that hold promise for the use of conditionally indispensable amino acids (dipeptides) in future patient care. There is little question that efforts to modify the response to disease by nutritional means with amino acids and dipeptides will be reworded with improved patient survival. Surprising and exciting medical progress of yesterday belongs today and tomorrow to the common daily medical exercise.

References 1. Cuthbertson D P. Observations on the disturbances of metabolism produced by injury to the limbs. Q J Med 1932; 1: 223–230 2. Fu¨rst P, Stehle P. Parenteral nutrition substrates. In: Payne-James J, Grimble G, Silk D, (eds). Artificial Nutrition Support in Clinical Practice. London: Edward Arnold, 1994; 301–322 3. Fu¨rst P. Old, new substrates in clinical nutrition. J Nutr 1998; 128: 789–796 4. Fu¨rst P. Regulation of intracellular metabolism of amino acids. Sir Arvid Wretlind Lecture. In: Bozzetti F, Diongi P, (eds). Nutrition in Cancer, Trauma and Sepsis. Karger: Basel, 1985; 21–53 5. Fu¨rst P, Stehle P. Are intravenous amino acid solutions unbalanced? New Horizons 1994; 2: 215–223 6. Harper AE. Some concluding comments on emerging aspects of amino acid metabolism. J Nutr 1994; 124: 1529S–1532S 7. Young V E, El-Khoury A E. The notion of the nutritional essentiality of amino acids, revisited, with a note on the indispensable amino acid requirements in adults. In: Cynober L. (ed). Amino Acid Metabolism and Therapy in Health and Nutritional Disease, Boca Raton, FL: CRC Press, 1995; 191–232

ARTICLE IN PRESS S28

AMINO ACIDS AND SYNTHETIC DIPEPTIDES

8. Brennan M F, Cerra F, Daly J M, et al. Report of a research workshop: branched-chain amino acids in stress and injury. J Parenter Enteral Nutr 1986; 10: 446–452 9. Vinnars E, Bergstro¨m J, Fu¨rst P. Influence of the postoperative state on the intracellular free amino acids in human muscle tissue. Ann Surg 1975; 182: 665–671 10. MacLennan P, Smith K, Weryk B, Watt P W, Rennie M J. Inhibition of protein breakdown by glutamine in perfused rat skeletal muscle. FEBS Lett 1988; 237: 133–136 11. Jepson M M, Bates P C, Broadbent P, Pell J M, Millward D J. Relationship between glutamine concentration and protein synthesis in rat skeletal muscle. Am J Physiol 1988; 255: E166–EI72 12. Griffiths R D, Jones C, Palmer T E A. Six-month outcome of critically ill patients given glutamine-supplemented parenteral nutrition. Nutrition 1997; 13: 295–302 13. Fu¨rst P. Conditionally indispensable amino acids (glutamine. cyst(e)ine, tyrosine, arginine, ornithine, taurine) in enteral feeding and the dipeptide concept. Nestle´ Nutr Workshop Ser Clin Perform Programme 2000; 3: 199–219

14. Bauer P, Charpentier C, Bouchet C, Raffy F, Gaconnet N, Larcan A. Short parenteral nutrition coupled with early enteral nutrition in the critically ill. Intensive Care Med 1998; 24(suppl): 123 15. Wretlind A. Parenteral nutrition. Nutr Rev 1981; 39: 257–265 16. Moriion B J, Torwesten E. Kuhn K S, Lessire II, Puchstein C, Fu¨rst P. Cysteinyl-leucotriene generation as a biomarker for survival in the critically ill. Crit Care Med 2000; 28: 3655–3658 17. Ha˚ussinger D, Roth E, Lang F, Gerok W. Cellular hydration state: an important determinant of protein catabolism in health and disease. Lancet 1993; 341: 1330–1332 18. Barbul A. Arginine, immune function. Nutrition 1990; 6: 53–58 19. Cynober L. Can arginine, ornithine support gut function? Gut 1994; 35: 542 20. Lin E, Concalves J A, Lowry S F. Efficacy of nutritional pharmacology in surgical patients. Curr Opin Clin Nutr Met Care 1998; 1: 41, 50 21. Suchner U, Kuhn K S, Fu¨rst P. The scientific basis of immunonutrition Proc Nutr Soc 2000; 59: 553–563