Portal de Periódicos da CAPES

Zinc: An Essential Trace Element for Parenteral Nutrition

2009; Elsevier BV; Volume: 137; Issue: 5 Linguagem: Inglês

10.1053/j.gastro.2009.08.014

1528-0012

Khursheed N. Jeejeebhoy,

Trace Elements in Health

Zinc is an essential trace element for human nutrition that is an integral part of many enzyme systems, including DNA polymerase complex. Zinc deficiency has been associated with stunting of growth and sexual immaturity. In children, deficiency causes a fatal condition called acrodermatitis enteropathica. The same syndrome has been observed in patients on total parenteral nutrition (TPN) who do not receive zinc. In TPN the requirements have been estimated by balance studies to be 3 mg/d in patients without gastrointestinal losses and a mean of 12 mg/d in patients with diarrhea and fistula losses. Zinc is an essential trace element for human nutrition that is an integral part of many enzyme systems, including DNA polymerase complex. Zinc deficiency has been associated with stunting of growth and sexual immaturity. In children, deficiency causes a fatal condition called acrodermatitis enteropathica. The same syndrome has been observed in patients on total parenteral nutrition (TPN) who do not receive zinc. In TPN the requirements have been estimated by balance studies to be 3 mg/d in patients without gastrointestinal losses and a mean of 12 mg/d in patients with diarrhea and fistula losses. Patients relying on parenteral nutrition (PN) receive most of their nutrients in the form of amino acids, glucose, and lipid. However, for the optimal utilization of these nutrients, other substances called micronutrients are necessary. These substances are required in minute quantities, which are disproportionately low as compared with their influence on metabolism and health. These micronutrients are of 2 types, trace elements and vitamins. Trace elements are inorganic elements, whereas vitamins are complex organic molecules. Trace elements are essential elements which are an integral part of metabolically active organic complexes such as enzymes. Cotzias defined an essential trace element as one with the following characteristics:1Present in healthy tissues of all living things.2Constant tissue concentration.3Withdrawal leads to reproducible functional abnormalities4Addition of the element prevents abnormalities5The biochemical change is prevented or cured with the clinical abnormality. Using the Cotzias criteria, 7 elements—iron, zinc, copper, chromium, selenium, iodine, and cobalt (in vitamin B12)—are necessary for human health. In this monograph, one of these elements, namely zinc and its requirement in oral and PN are presented. Zinc as an essential nutrient was first recognized for the growth of plant life in the 1860s; however, the demonstration of zinc deficiency as a cause of clinical disease was first shown in swine as causing parakeratosis.1Tucker H.F. Salmon W.D. Parakeratosis or zinc deficiency disease in the pig.Proc Soc Exp Biol Med. 1955; 88: 613-616Crossref PubMed Scopus (147) Google Scholar In humans, a rare genetic disorder called acrodermatitis enteropathica resulted in skin parakeratosis and diarrhea in infants fed cows' milk but not human milk. Studies by Moynahan2Moynahan E.J. Acrodermatitis enteropathica: a lethal inherited human zinc deficiency disorder.Lancet. 1974; 304: 399-400Abstract Scopus (417) Google Scholar showed that a specific peptide existed in cow but not human milk, which chelated zinc. The genetic disorder was the lack of an enzyme that hydrolyzed the peptide and released the zinc. Feeding zinc cured the syndrome in these infants who otherwise died. On the other hand, in otherwise normal persons, dietary zinc deficiency does not occur, mainly because zinc is widely distributed in food. Therefore, it was not clear if there was a mandatory requirement for zinc in humans. In 1961, Prasad et al3Prasad A.S. Halstead J.A. Nadami M. The syndrome of iron deficiency anemia, hepatosplenomegaly, hypogonadism, dwarfism and geophagia.Am J Med. 1961; 31: 532Abstract Full Text PDF PubMed Scopus (537) Google Scholar described a syndrome of hypogonadism, anemia, and stunting in humans who were geophagic. Supplementation with zinc restored growth and sexual maturation, demonstrating the essentiality for dietary zinc in humans.3Prasad A.S. Halstead J.A. Nadami M. The syndrome of iron deficiency anemia, hepatosplenomegaly, hypogonadism, dwarfism and geophagia.Am J Med. 1961; 31: 532Abstract Full Text PDF PubMed Scopus (537) Google Scholar The use of PN created a unique situation in which it was possible to feed individuals with purified diets specifically deficient in trace elements such as zinc. In addition, for the first time it was possible to create a demand for nutrients by inducing anabolism. The anabolic state is necessary to demonstrate a requirement for a nutrient because clinical deficiency only occurs if there is a need for the nutrient caused by anabolism, but no intake for the specific nutrient. Kay and Tasman–Jones4Kay R.G. Tasman-Jones C. Acute zinc deficiency in man during Intravenous Alimentation.Aust NZ J Surg. 1975; 45: 325-330Crossref PubMed Scopus (39) Google Scholar described 4 patients on PN who developed a syndrome comparable to acrodermatitis enteropathica with marked reduction of plasma zinc and corrected specifically by feeding/infusing zinc. This paper demonstrated that zinc was essential for patients receiving PN. Zinc is a widely distributed element in foodstuffs (shellfish, liver, milk, and wheat bran). In the human body, zinc is widely distributed in many tissues, blood cells, bone, and teeth.5Tipton I.H. Cook M.J. Trace elements in human tissue. II. Adult subjects from the United States.Health Phys. 1963; 9: 103-145Crossref PubMed Scopus (287) Google Scholar, 6Underwood E.J. Zinc.in: Trace elements in human and animal nutrition. 4th ed. Academic Press, New York1977: 196-242Google Scholar However, zinc at these sites is firmly bound to protein, and during deficiency and refeeding the concentrations of zinc in tissues (with the exception of blood, milk, hair, and liver) do not change significantly.7Kirchgessner M. Roth H.P. Weigand E. Biochemical changes in zinc deficiency.in: Prasad A.S. Trace elements in human health and disease. Vol. 1. Zinc and copper. Academic Press, New York1976: 189-225Google Scholar Endogenous stores of zinc are mobilized in the fasting state, but do not meet metabolic needs during anabolism, because the net movement of zinc is into tissues and circulating zinc is reduced. Although 86% is in skeletal muscle, there are certain areas where zinc concentration is especially high and may represent functional importance. They are the prostate, hippocampus, pancreas, and kidney cortex. Zinc has been identified as a part of about 120 enzymes.8Vallee B.L. Falchuk K.H. Zinc and gene expression.Phil Trans R Soc Lond B. 1981; 294: 185-197Crossref PubMed Scopus (92) Google Scholar Among them are carbonic anhydrase, carboxypeptidase, alkaline phosphatase, oxidoreductases, transferases, ligases, hydrolases, lyases, and isomerases. Although the syndrome of zinc deficiency cannot be identified with the dearth of any 1 enzyme, zinc deficiency does have a pronounced effect on nucleic acid metabolism, thus influencing protein and amino acid metabolism. Zinc is an integral constituent of DNA polymerase, reverse transcriptase, RNA polymerase, tRNA synthetase, and the protein chain elongation factor.5Tipton I.H. Cook M.J. Trace elements in human tissue. II. Adult subjects from the United States.Health Phys. 1963; 9: 103-145Crossref PubMed Scopus (287) Google Scholar Thus, zinc deficiency can alter protein synthesis at a number of different points, and it is not surprising that in the absence of zinc growth arrest occurs.9Prasad A.S. Zinc in human nutrition.in: CRC Press, Boca Raton, FL1979: 1-80Google Scholar Furthermore, zinc deficiency is teratogenic as determined by animal studies and observations in patients with untreated acrodermatitis enteropathica.10Hambidge K.M. Neldner K.H. Walravens P.A. Zinc, Acrodermatitis Enteropathica and congenital malformations.Lancet. 1975; 1: 577-578Abstract Scopus (118) Google Scholar This finding suggests that zinc deficiency may affect gene expression. In experimental studies in unicellular organisms, it has been shown that zinc deficiency changes the nature of RNA polymerase and the base composition of mRNA. The translated peptides contain a preponderance of arginine-rich peptides that can bind to anions such as phosphate groups in nucleic acids and alter their action. Such an alteration could affect the synthesis of histones, proteins that are known to reduce the activity of DNA as a template.5Tipton I.H. Cook M.J. Trace elements in human tissue. II. Adult subjects from the United States.Health Phys. 1963; 9: 103-145Crossref PubMed Scopus (287) Google Scholar These experimental findings about zinc and nucleic acids are interesting in view of the clinical observation that a number of functions dependent on protein synthesis are suppressed by zinc deficiency. These include growth,6Underwood E.J. Zinc.in: Trace elements in human and animal nutrition. 4th ed. Academic Press, New York1977: 196-242Google Scholar cellular immunity,11Golden M.H.N. Golden B.E. Harland P.S.E.G. et al.Zinc and immunocompetence in protein energy malnutrition.Lancet. 1978; 1: 1226-1227Abstract PubMed Scopus (150) Google Scholar, 12Fernandes G. Nair M. Onoe K. et al.Impairment of cell mediated immunity functions by dietary zinc deficiency in mice.Proc Natl Acad Sci U S A. 1979; 76: 457-461Crossref PubMed Scopus (250) Google Scholar fertility,6Underwood E.J. Zinc.in: Trace elements in human and animal nutrition. 4th ed. Academic Press, New York1977: 196-242Google Scholar hair growth, wound healing,13Golden M.H.N. Golden B.E. Jackson A.A. Skin breakdown in kwashiorkor responds to zinc.Lancet. 1980; 1: 1256Abstract PubMed Scopus (38) Google Scholar and plasma protein levels. Thus, it is obvious that zinc deficiency leads to profound disturbances of protein synthesis. In addition, in volunteer studies, experimental mild zinc deficiency reduced thymulin levels and the CD4/CD8 ratio.14Prasad A.S. Meftah S. Abdallah J. et al.Serum thymulin in human zinc deficiency.J Clin Invest. 1988; 82: 1202-1210Crossref PubMed Scopus (262) Google Scholar Zinc absorption has a significant effect on body zinc. Zinc is absorbed by a process that involves binding to a surface receptor, followed by uptake into the enterocyte.15Davies N.T. Studies on the absorption of zinc by rat intestine.Br J Nutr. 1980; 43: 189-203Crossref PubMed Scopus (88) Google Scholar The process is saturable and the efficiency of absorption decreases at high zinc intakes. From the enterocyte, some zinc is removed by albumin or an alpha-2 macroglobulin and carried to the liver, the remainder being bound to a metallothionein,16Richards M.P. Cousins R.J. Isolation of an intestinal metallothionein induced by parenteral zinc.Biochem Biophys Res Commun. 1977; 75: 286-294Crossref PubMed Scopus (66) Google Scholar the proportion bound depending on the metallothionein content of the enterocyte. Because zinc is not transferred to plasma when so bound, this binding action inhibits absorption. Subsequently, such bound zinc returns to the bowel lumen when the enterocyte is shed. When body zinc is high, there is a stimulation of metallothionein synthesis, thus inhibiting absorption. In addition, absorption is influenced by the following factors:1Binding to a ligand secreted by the pancreas enhances absorption.2Luminal amino acids bind zinc and prevent its precipitation by substances such as phosphates and phytates.3Pregnancy, corticosteroids, and endotoxin all enhance absorption.4Phytates, phosphates, iron, copper, lead, and calcium inhibit absorption. Zinc is excreted mainly in the feces, with a smaller amount in the urine.17Robinson M.F. McKenzie J.M. Thomson C.D. et al.Metabolic balance of zinc, copper, cadmium, iron, molybdenum and selenium in young New Zealand women.Br J Nutr. 1973; 30: 195-205Crossref PubMed Scopus (57) Google Scholar The fecal losses rise with increased intake, because they consist mainly of unabsorbed zinc shed with enterocytes. In contrast, urinary excretion is not influenced by intake. Significant losses may occur in sweat in the tropics, but such losses diminish with deficiency.18Prasad A.S. Schulert A.R. Sandstead H.H. et al.Zinc, iron and nitrogen content of sweat in normal and deficient subjects.J Lab Clin Med. 1963; 62: 84-89PubMed Google Scholar Wolman et al19Wolman S.L. Anderson G.H. Marliss E.B. et al.Zinc in total parenteral nutrition Requirements and metabolic effects.Gastroenterology. 1979; 76: 458-467Abstract Full Text PDF PubMed Scopus (193) Google Scholar showed that diarrhea and stomal and fistula losses were the major sites of enhanced abnormal losses of zinc from endogenous sources in patients kept nil per os (NPO). Increased losses also occurred in urine in hypercatabolic individuals. Amino acid infusions also increase urinary zinc losses. In the kidney, zinc infusions enhance distal reabsorption of zinc, and amino acid infusion increases proximal secretion.20Abu Hamdan D.K. Migdal S.D. Whitehouse A.S. et al.Disparate urinary zinc (ZN) handling in response to ZN infusion and amino acids.Kidney Int. 1979; 16: 818Google Scholar As noted, zinc reaching the circulation is bound to albumin and an alpha-2 macroglobulin.21Smith K.T. Cousins R.J. Quantitative aspects of zinc absorption by isolated, vascularly perfused rat intestine.J Nutr. 1980; 110: 316-323Crossref PubMed Scopus (36) Google Scholar From the circulation it is taken up by the liver and other tissues. Infection results in increased uptake of zinc by the liver.22Beisel W.R. Pekarek R.S. Wannemacher Jr, R.W. The impact of infectious disease on trace element metabolism of the host.in: Hoekstra W.G. Suttie J.W. Ganther H.E. Trace element metabolism in animals. Vol. 2. University Park Press, Baltimore1974: 217-240Google Scholar Enhanced uptake of zinc into the liver reduces plasma concentrations, and so circulating zinc concentrations may be reduced by factors other than deficiency.23Talbot T.R. Ross J.F. The zinc content of plasma erythrocytes of patients with pernicious anemia, sickle cell anemia, polycythemia vera, leukemia and neoplastic disease.Lab Invest. 1960; 9: 174-184PubMed Google Scholar, 24Vallee B.L. Wacker W.E.C. Bartholmay A.F. et al.Zinc metabolism in hepatic dysfunction.Ann Intern Med. 1959; 50: 1077-1091Crossref PubMed Scopus (49) Google Scholar, 25Vikbladh I. Studies on zinc in blood.Scand J Clin Lab Invest. 1950; 2: 143-148Crossref PubMed Scopus (27) Google Scholar There is no reliable way of assessing zinc status. Plasma zinc levels are normally regulated to between 80 and 120 μg/l or 12–18 μmol/l. Although circulating zinc levels fall in the deficient state, there are other causes of low circulating zinc levels that make this measurement unreliable. Hair zinc levels are low when there is low-grade chronic deficiency, but in acute deficiency hair does not grow, and with profound deficiency hair loss occurs and the remaining hair may have normal zinc concentrations.26Hambidge K.M. Zinc deficiency in man: its origins and effects.Phil Trans R Soc Lond B. 1981; 294: 129-144Crossref PubMed Scopus (37) Google Scholar It has recently been shown that leukocyte zinc levels are a reliable indicator of zinc deficiency, but this is not an easy measurement to perform.27Whitehouse R.C. Prasad A.S. Rabbani P.I. et al.Zinc in plasma, neutrophils, lymphocytes, and erythrocytes as determined by flameless atomic absorption spectrophotometry.Clin Chem. 1982; 28: 475-480Crossref PubMed Scopus (170) Google Scholar Currently, the best way of assessing zinc status and requirements is through multiple clinical parameters. Abnormal gastrointestinal losses, hypercatabolism, or amino acid infusions raise the need for zinc supplementation. The clinical syndrome of acrodermatitis enteropathica confirms the need for zinc supplementation. This syndrome consists of scaly, red, desquamating lesions involving the nasolabial folds and hands. In severe cases it extends to the trunk, resulting in extensive exfoliation and secondary skin infection. There is often loss of hair. The relationship of intake to absorption is asymptomatic and the fraction of intake absorbed falls as the amount taken rises. The endogenous losses also rises with increased intake. The point at which absorbed and losses are equal is used to estimate dietary zinc requirements. On this basis, the requirements in adults is between 8 and 11 mg/d. Zinc intake during pregnancy and lactation are estimated to be 12–13 mg/d. In patients receiving PN, zinc requirements were estimated by performing metabolic balance of intake versus losses in stool, urine, and drainage during a randomly assigned intake in the infusate. Using this technique Wolman et al19Wolman S.L. Anderson G.H. Marliss E.B. et al.Zinc in total parenteral nutrition Requirements and metabolic effects.Gastroenterology. 1979; 76: 458-467Abstract Full Text PDF PubMed Scopus (193) Google Scholar found that about 2.5 mg/d was required for balance in patients without diarrhea. Requirements increased with increased catabolism and gastrointestinal losses by 12 mg/L of small intestinal fluid and 17 mg/L of stool measured in the NPO state. In patients with burns, it has been shown that infusing about 36 mg/d of zinc in patients with burns increased skin levels of zinc and reduced infectious complications. In addition to replacing losses, infants need zinc for growth. This is especially true of preterm infants, because two thirds of the infant's zinc is transferred from the mother during the last 10–12 weeks of normal gestation. It has been estimated that 0.50–0.75 mg of zinc is taken up per day during the last 3 weeks of gestation and the first 3 weeks of postnatal life in babies gaining 1.5 kg, making the requirements about 300–500 μg/kg per day.28Widdowson E.M. Dauncey J. Shaw J.C.L. Trace elements in foetal and early postnatal development.Proc Nutr Soc. 1974; 33: 275-284Crossref PubMed Scopus (156) Google Scholar During PN, James and MacMahon29James B.E. MacMahon R.A. Balance studies of 9 elements during complete intravenous feeding of small premature infants.Aust Pediatr J. 1976; 12: 154-162PubMed Google Scholar found that infants required 300 μg/kg per day to maintain balance. In older children, 50 μg/kg per day maintained normal serum levels and for growth it is recommended that 100 μg/kg per day be given as a safe intake.30Ricour C. Duhamel J.F. Gros J. et al.Estimates of trace element requirements of children receiving total parenteral nutrition.Arch Fr Pediatr. 1977; 34: 92-100Google Scholar In addition, supplementation is required for abnormal gastrointestinal losses, but these have not been determined experimentally. In stable patients without abnormal gastrointestinal losses, between 3 and 4 mg/d meets requirements. In patients with diarrhea while NPO and gastrointestinal losses from fistulae, stomas, and diarrhea, 12 mg of zinc should be added per liter of losses. In patients with burns, large amounts of zinc supplements has been shown to reduce the rate of infection. Up to 36 mg/d has been added and shown to benefit without toxicity. •Zinc is an essential trace element and must be added to all PN mixtures.•In patients without gastrointestinal losses, 3–4 mg should be given daily.•In patients with fistula, diarrhea, and intestinal drainage, 12 mg of zinc should be added for each liter of loss.•In patients with burns, addition of about 36 mg/d of zinc may reduce infectious complications. •Methods to evaluate zinc status.•Zinc requirements in critically ill patients. Does additional zinc reduce infectious complications?•Benefits and risks of long-term zinc infusions in patients on home PN. DR HOWARD: Let me take the chairman's privilege of asking the first question. Your important balance studies were done as you described with the short bowel patients NPO. But if you're trying to avoid the hepatic complications of PN, you really encourage them to eat. Because we don't know whether food zinc leads to net absorption or net loss in short bowel patients, do you think we need to repeat those balance studies and include the enteral component or do you think we can look separately at absorption and losses of zinc and other divalent cations? So my question is, do we need to repeat those balance studies with the subjects eating? DR JEEJEEBHOY: That's an interesting question. We had patients of all types. Some of them had fistula losses and short bowel. We demonstrated increased GI zinc losses in people who had diarrheal disease. If they are absorbing some zinc by mouth, obviously their requirements may drop. Because people have very different lengths of bowel, I think a study like this would be almost impossible to do. But what is the risk of giving them extra zinc, you can see that tolerance is enormous, so I think that we should err on the side of giving more zinc than giving less zinc, because there's no evidence of toxicity unless you give huge amounts. DR SHULMAN: The 24-hour urinary collections could be helpful to assess zinc status. What are your thoughts on that? DR JEEJEEBHOY: I think the 24-hour urine tells you more about the catabolism and anabolism of zinc rather than whether the person is deficient or not, it depends on when in their course you collect it. It could tell you if the patient is likely to become zinc deficient if they are losing a lot of zinc in their urine. If you want to get a handle on zinc status, it involves looking at the clinical picture—are they catabolic or are they in the repair phase? You can add a 24-hour urine; you can add metallothionein and zinc plasma levels. But the whole picture has to be put together to come to some conclusion as to what to do. DR SHENKIN: I might follow up on Lyn's question about how much extra zinc. If zinc is taken orally, how much of this is absorbed in patients who have short bowel and are on long-term PN? I wonder if we can expand the question a little bit more. Is there any toxicity to chronic long-term overprovision of zinc because that is what you're recommending? As you've pointed out, the body has its own natural defense mechanism for blocking zinc absorption and increasing zinc excretion when you over provide zinc enterally. But when you give zinc parenterally, we bypass the absorption brake and if you give a lot of zinc parenterally, let's say we move toward your higher intakes and the patients at the same time are eating and maybe absorbing some enterally, have the studies been done which would show that this is safe? I make the point especially in the light of the paper which Lyn and I published recently, on the tissue concentrations of patients who lived long-term on PN and who were following the recommendations that you suggested. Many of these patients had very high liver zinc concentrations, and therefore were accumulating zinc, which could be harmful. I'm wondering before we just jump and say this is fine, this is safe, should we be a little more cautious? DR JEEJEEBHOY: The risks of toxicity are extremely small, as far as zinc is concerned. In tissues, 90% as I mentioned is either in muscle, or in the liver (17%). In regard to toxicity, I would like to turn the question around and ask, if there's accumulation in tissue, in what way would it cause toxicity? What is the evidence that there's any toxicity? In contrast, there is good evidence that severe zinc deficiency is associated with death. So I think that we are over doing this toxicity issue with regard to zinc. Patients on PN who have had these zinc amounts for years and years have in fact done extremely well. DR BUCHMAN: Khursh, in a patient with systemic inflammation, for example pancreatitis or Crohn's disease, who also has fistula losses or diarrhea and may be third-spacing fluid, how does one interpret the plasma zinc concentrations? DR JEEJEEBHOY: I don't think you can use the plasma zinc with any degree of accuracy in such a situation. If you measured the metallothionein levels, you would find them to be high because it is an acute phase reactant. Also, you've got albumin losses and cytokine effects, you've got so many interactions I'm not sure you can come to any interpretation of plasma zinc levels. DR SHIKE: I think we need to make a distinction between the hospitalized patient who gets PN for a few days to a few weeks, and the long-term home patient. It is in the long-term home PN patients that we have potential for encountering trace element toxicity. These home patients tend to be more stable on the whole and therefore assessment parameters maybe a little more relevant. DR JEEJEEBHOY: I agree with you in the sense that if you have a home patient who is not otherwise sick and the problem is stable short bowel syndrome, measurements of levels maybe more meaningful. But again, many long-term patients have low albumin levels and other abnormal parameters. The point I'm trying to make is that in the present state of the art, you have to use a total gestalt. You have to look at the whole picture and then come to a decision as to what you want to do. It's more the art of medicine than science. A big problem with home PN patients is that you have patients with different degrees of short bowel and the issue comes up how are you going to assess their absorption of a particular element? Maybe we should ask the question can we recognize zinc toxicity. We do know high levels have effects on the kidneys and on the lungs. Perhaps we should be screening patients for these toxic effects? DR BERGER: Home PN patients are 1 group of patients where there are legitimate concerns about accumulation, and potential toxicity. I myself am an intensive care doctor where the PN requirements are for a period of 1–2 months, also in the critically ill there may be justification for pharmacologic amounts of certain nutrients. I think we must make separate recommendations for stable home PN patients and for critically ill patients. DR BUCHMAN: In terms of the use of zinc in critically ill patients, what kind of outcomes should we be evaluating? There are data to indicate zinc supplements can prevent some viral infections. Is there a role for extra zinc in the treatment or prevention of bacterial infections? For example in burn patients? DR JEEJEEBHOY: I think this is an area for important research. To my mind there are 2 areas where zinc might be important in infection. One is of course its anti-infective or immune-stimulating properties. The other is that zinc affects insulin secretion rates. It's been shown that bloodstream glucose control is one of the more important things in critically ill patients. These 2 areas could be profitably studied in patients in intensive care units. DR COMPHER: Dr Jeejeebhoy, I'm intrigued by your comments about the immune effects of zinc, and aware that in the home PN population the most common complication is catheter-related bloodstream infections. We know with infection, serum zinc levels are hard to interpret. We don't have a good explanation for why some patients go through 20 years with no infections, and other patients have many infections. If we wanted to evaluate a role of zinc status, in describing the risk of bloodstream infection, how should we go about that? DR JEEJEEBHOY: You need to identify these patients with recurrent infections and in a multicenter, randomized trial determine whether they improve if they are given zinc. DR CHAMBRIER (Lyons, France): I'm concerned about the long-term PN patient. Initially we gave 12 mg/d of zinc and all patients had a high zinc plasma so now we give only 10 mg/d of zinc. DR JEEJEEBHOY: I've not said that all patients should receive 12 mg/d. What I've said is that if the patient has 1 liter of constant losses from a fistula, then 12 mg should be added to their basic requirement of 3–4 mg/d. DR CHAMBRIER: But every patient has >1 liter of high bowel losses. DR JEEJEEBHOY: We are talking about the NPO state. They shouldn't be eating anything and still put out 1 liter. There's a big difference. You have to stop them eating. What you're describing is really the postprandial patient. It doesn't apply to this bowel loss. DR CHAMBRIER: Thank you very much. MS. SABINO (St. Francis Hospital, Hartford, CT): Can you comment on whether there are excess losses of zinc in head injury patients in acute care settings. How much would you supplement these patients? Also, I understand medications like propofol increase zinc losses. Are there other medications that deplete zinc? DR JEEJEEBHOY: I've seen those studies. The problem is that propofol is often used in patients in the ICU who are catabolic. These patients, as I mentioned, are breaking down muscle and excreting endogenous zinc. With a head injury, you have hypercatabolism. I'm not aware of any systematic studies that have measured how much zinc is lost in these catabolic individuals and how they can be effectively supplemented. I think this is an area for research. One would have to look at the urinary zinc losses and estimate what the total losses are. There are huge reserves that people have of zinc, so if you have losses for 3 or 4 days, it doesn't mean too much. On the other hand, if you have continued losses for a month or so then appropriate supplementation could be important. DR SEIDNER: Khursh, we haven't discussed your point on research into methods of evaluating zinc status after we begin to replete these patients. In my practice, I measure 24-hour urines to see if calcium and magnesium status is adequate. Has that ever been done for zinc in patients on home PN? DR JEEJEEBHOY: I'm not aware of any systematic studies of losses being done in the way you are talking about, but the trouble is that urine losses of zinc are not like magnesium losses. Magnesium losses are very dependent on the magnesium status of the patient. If you become magnesium deficient, the urine magnesium almost gets wiped out, and it's a very good index. Unfortunately, with zinc, what happens is any kind of catabolism of any sort is associated with increased losses. Low insulin levels, insulin resistance, and infections—all these things increase zinc losses. So I think it's going to be extremely difficult to use 24-hour urine zinc as an index without major qualifications. DR SEIDNER: So how are we going to monitor zinc status? DR JEEJEEBHOY: At the moment, I'm not aware of any way of readily assessing zinc status. MS. BEGANY (Children's Hospital, Philadelphia, PA): I'm just wondering if you could comment on zinc's relationship with copper and the potential when using the higher dosing of zinc of seeing negative effects on copper. DR JEEJEEBHOY: Essentially, you're talking about the oral interaction of zinc and copper. When you have high oral zinc levels, you induce metallothionein, which binds oral copper, taking it out of the body. I'm not aware that parenteral zinc particularly effects copper status. DR ZALOGA (Baxter, Deerfield, IL): Just to go back to the propofol question for a moment. Propofol is available from a number of companies. One of the companies uses EDTA as a preservative and there's Intensive Care Medicine publications that looked at urinary excretion of zinc. It's quite considerable, it can go up 10-fold when you use the EDTA containing propofol. The label for that product has a warning on zinc deficiency. Other propofol products use sodium metabisulfite as a preservative and then zinc losses are not as high. DR HOWARD: Dr Shenkin, you want to get in 1 last hit? DR SHENKIN: Yes, I want to come back to 2 interrelated points, one to do with the effects of infection on zinc, and the other how to monitor improvements in zinc status. In regard to the first point, I think measuring metallothionein is not possible in most clinical settings. Many of us instead use C-reactive protein (CRP) in exactly the same way as metallothionein. CRP is much cheaper and everybody can measure it, and in fact when CRP goes up, metallothionein goes up, and as infection settles, metallothionein comes down and CRP comes down, and zinc concentration ought to rise. If it doesn't, then maybe that's an indication that zinc status is not adequate. This deserves research confirmation I think. I'm aware of 2 conflicting animal studies in relation to zinc supplementation and infection, one of which when zinc was given, the febrile response of rabbits was enhanced. If there was zinc depletion the febrile response was less. In another study, I think also in rabbits, and currently in press, giving zinc reduced the proinflammatory cytokine response. Now, these seem to be totally conflicting studies, one suggesting increased cytokine response to zinc, the other suggesting a reduced cytokine response. I think as far as trying to get a handle on does zinc cause damage, does it encourage inflammation, does it suppress inflammation, focusing on these cytokine responses might well be helpful.