Physical Growth Response to Zinc Supplementation

In contrast to studies on the effects of low-dose zinc supplements on clinical features (e.g., pneumonia, diarrhea [Bhutta et al., 1999]) and on nonspecific laboratory functional tests of zinc status (e.g., tests of neuro-cognitive function [Sandstead et al., 1998]) or immune status (Shankar and Prasad, 1998), studies of the effects of zinc supplementation on physical growth velocity in children are useful in evaluating dietary zinc requirements for several reasons. First, confirmation of the effect of zinc supplements on growth velocity (linear growth and weight) in children with varying degrees of growth retardation has been shown in a number of studies from many countries (Brown et al., 1998; Umeta et al., 2000). Second, because a sufficient number of these studies have been undertaken in North America, growth is applicable as a functional/clinical indicator of zinc requirement in North American children (Gibson et al., 1989; Walravens and Hambidge, 1976; Walravens et al., 1983, 1989). Third, baseline dietary data typically included in these studies are adequate to use for group analyses.

Size and Turnover Rates of Zinc Pools

Strong positive correlations have been observed between dietary zinc content, especially the amount of absorbed zinc, and estimates of the size of the combined pools of zinc that exchange with zinc in plasma (Miller et al., 1994; Sian et al., 1996). Once links to clinical, biochemical, or molecular effects of zinc deficiency have been achieved and appropriate cut-off levels for different age groups and gender have been defined, pool size and turnover measurements may be of value in future refinements of EARs. Even simpler models involving the measurement of plasma zinc clearance may be useful in assessing zinc deficiency, but dietary data derived by such a method are not available at this time (Kaji et al., 1998; Nakamura et al., 1993; Yokoi et al., 1994). More detailed model-based compartmental analyses, when specifically applied to the evaluation of dietary requirements, also have the potential to contribute to a more precise understanding of zinc requirements (Miller et al., 1998; Wastney et al., 1986).

Plasma and Serum Zinc Concentration

While both plasma zinc concentration and serum zinc concentration are used as indicators of zinc status, plasma zinc concentration is preferable because of the lack of contamination of zinc from the erythrocyte. Homeostatic mechanisms are effective in maintaining plasma zinc concentrations for many weeks of even severe dietary zinc restriction (Johnson et al., 1993; Wada et al., 1985). A number of studies have reported no association between dietary zinc intake and plasma or serum zinc concentration (Artacho et al., 1997; Kant et al., 1989; Neggers et al., 1997; Thomas et al., 1988). Payette and Gray-Donald (1991) did observe a significant correlation between dietary zinc intake and serum zinc concentration in noninstitutionalized elderly; however, the correlation was positive for men and negative for women. Discernible relationships have been reported between plasma zinc concentration and habitual dietary zinc intake, even within the range typically occurring in North America. These relationships are of some utility in providing a supportive indicator of zinc requirements. For example, serum zinc concentrations of Canadian adolescent girls aged 14 to 19 years vary inversely with phytate:zinc molar ratios, and a greater percentage of lactoovo-vegetarians have serum zinc values below 70 μg/dL than do omnivores (Donovan and Gibson, 1995). Cut-off concentrations for lower limits have been established and depend on the time of day at which collections are made because of the substantial and cumulative effects of meals in lowering concentrations (King et al., 1994). The cut-off concentrations for prebreakfast samples is 70 μg/dL. Different cut-off concentrations are not considered necessary for different age groups or genders.

Insufficient and inconsistent data exist for plasma or serum zinc concentrations in apparently normal subjects whose habitual dietary zinc intakes straddle the vicinity of the average requirement, and therefore use of those concentrations for estimating an average requirement is limited. Furthermore, plasma and serum zinc concentrations do not seem to be sufficiently sensitive to serve as a subsidiary indicator.

Zinc Concentration in Erythrocytes

Erythrocyte zinc concentration is depressed at moderately severe levels of dietary zinc restriction (Thomas et al., 1992), but the sensitivity of this assay is inadequate to provide more than a secondary supportive indicator of dietary zinc requirements. Sample preparation may account for some of the lack of sensitivity. Results from experimental depletion studies (Baer and King, 1984; Bales et al., 1994; Grider et al., 1990; Ruz et al., 1992; Thomas et al., 1992) have been mixed, and the value of erythrocyte zinc concentrations as an indicator of zinc nutritional status is not well defined.

Zinc Concentration in Hair

Associations between low zinc concentration in hair and poor growth have been documented (Ferguson et al., 1993; Gibson et al., 1989; Hambidge et al., 1972; Walravens et al., 1983). In three of these studies, low zinc concentration in hair was used as a criterion for zinc supplementation in children and resulted in increased growth velocity. Low zinc concentrations in hair have been reported in Canadian children with low meat consumption (Smit-Vanderkooy and Gibson, 1987). Subjects whose habitual diets are high in phytate or who have very high phytate:zinc molar ratios have also been noted to have relatively low zinc concentrations in hair. However, there is a lack of uniformity in apparently low zinc concentrations in hair, and no lower cut-off values have been defined clearly for any age group or either gender. The use of zinc in hair as a supportive indicator for establishing zinc requirements needs further research.

Activity of Zinc-Dependent Enzymes

With the large number of zinc-dependent enzymes that have been identified, it is perhaps remarkable that no single zinc-dependent enzyme has found broad acceptance as an indicator of zinc status or requirement. This state of affairs is attributable to a number of factors, including the homeostatic processes that maintain zinc occupancy of the catalytic sites of these enzymes and the lack of consistency in findings between studies. Other factors include a lack of sensitivity, the inaccessibility of optimal tissues to assay, or, simply, inadequate research. The lack of baseline dietary data also negates the potential value of some reports. Given these limitations, limited dose-response data, and inconsistent responses to dietary zinc (Bales et al., 1994; Davis et al., 2000; Paik et al., 1999; Ruz et al., 1992; Samman et al., 1996), the activities of zinc-dependent enzymes, including alkaline phosphatase, copper-zinc superoxide dismutase, and lymphocyte 5′-nucleotidase, can at most serve as supportive indicators of dietary zinc requirements at this time. Although it is not consistently responsive to zinc intake, the activity of plasma 5′-nucleotidase (Beck et al., 1997a), which is derived from the CD73 cell surface markers of B and T cells, merits specific recognition as a potential marker of zinc status (Failla, 1999).

Metallothionein and Zinc-Regulated Gene Markers

Erythrocyte metallothionein concentrations have been reported to be responsive to both increased and restricted dietary zinc (Grider et al., 1990; Thomas et al., 1992), but the sensitivity and precision of this index has not been thoroughly evaluated. Monocyte metallothionein messenger RNA responds rapidly to in vivo zinc supplementation (Sullivan et al., 1998) and merits additional research. Moreover, this approach points the way for future exploration of molecular markers of zinc status including, for example, a whole family of zinc transporters that are now being identified (Failla, 1999; McMahon and Cousins, 1998).

Indexes of Immune Status

Zinc is essential for the integrity of the immune system, and inadequate zinc intake has many adverse effects (Shankar and Prasad, 1998). Though the immune system, which is thought to underlie several of the most important sequelae of mild zinc deficiency, is sensitive to even mild zinc deficiency, the effects on functional indexes of zinc status are not specific. At this time, therefore, changes in indexes of immune status with manipulation of dietary zinc can serve only as a limited indicator for dietary zinc requirements.

Hormones

The biology of zinc is linked extensively to hormone metabolism. Notable examples are the zinc finger motifs of regulatory proteins required for hormonal signals to regulate gene transcription (Cousins, 1994; Klug and Schwabe, 1995). Zinc has been reported to have roles in the synthesis, transport, and peripheral action of hormones. Low dietary zinc status has been associated with low circulating concentrations of several hormones including testosterone (Prasad et al., 1996), free T4 (Wada and King, 1986), and IGF-1 (Ninh et al., 1996). Zinc supplementation has been associated with an increase in both circulating IGF-1 concentration and growth velocity (Ninh et al., 1996). However, no studies have directly related hormone concentrations to decreases or increases in zinc intake.

Circulating Hepatic Proteins

Reductions in retinol binding protein, albumin, and pre-albumin concentrations have been reported with moderate dietary zinc restriction (Wada and King, 1986). Serum zinc and retinol binding protein concentrations are significantly correlated in zinc-deficient Thai children (Udomkesmalee et al., 1990). Changes in circulating concentrations of these proteins with changes in dietary zinc may serve as minor supportive indicators. The relationship of such indicators to general malnutrition or to dietary deficiency that is not related to zinc status supports their being minor indicators for zinc requirements.

Iron

Daily intake of iron at levels such as those found in some supplements could decrease zinc absorption (O'Brien et al., 2000; Solomons and Jacob, 1981; Valberg et al., 1984). This relationship is of some concern in management of iron supplementation during pregnancy and lactation (Fung et al., 1997). Recent studies of the mechanism of nonheme iron absorption suggest that upregulation of an iron transport protein occurs in iron deficiency (Gunshin et al., 1997). The comparable affinity of this transporter for zinc suggests that, during low iron intake, zinc absorption may be stimulated and suggests one possible locus for a zinc-iron interaction. The influence of heme iron on zinc absorption has not received much attention. The activity of other divalent metal transporters may also affect zinc absorption.

Calcium and Phosphorus

The importance of calcium in the diet and the mass of the element that must be consumed daily to maintain maximum bone density suggest that special attention should be given to its potential inhibitory effect on zinc absorption. Nutrition experiments with swine have shown conclusively that excess dietary calcium produces a decrease in zinc absorption, which leads to a skin condition called parakeratosis. Experiments in humans have been equivocal, with calcium phosphate (1,360 mg/day of calcium) decreasing zinc absorption (Wood and Zheng, 1997) and calcium as the citrate-malate complex (1,000 mg/day of calcium) having no statistically significant effect on zinc absorption (McKenna et al., 1997). Differences could be related to the calcium sources, techniques used, and the extent of luminal zinc solubility. At present, data suggest consumption of a calcium-rich diet does not have a major effect on zinc absorption at an adequate intake level of the nutrient. Calcium effects at low dietary zinc intakes have not been adequately investigated. Dietary phosphorus-containing salts over an extensive intake range have not been shown to influence zinc balance (Greger and Snedeker, 1980; Spencer et al., 1984). Other dietary sources of phosphorus include phytate and phosphorus-rich proteins, for example, milk casein and nucleic acids, all of which bind zinc tenaciously and decrease zinc absorption.

Copper

Large-scale studies on the influence of dietary copper intake on zinc absorption and utilization have not been carried out with human subjects. Various experimental approaches with animals have not revealed a uniform influence of copper on intestinal zinc uptake (Cousins, 1985; Sandstrom and Lonnerdal, 1989). Rather, evidence for an interaction derives from the therapeutic effect of zinc in reducing copper absorption in patients with Wilson's disease (Yuzbasiyan-Gurkan et al., 1992). This action includes the induction of intestinal metallothionein by zinc and the subsequent binding of excess copper by this metalloprotein, which may limit transcellular copper absorption. The relationship may have relevance in situations where zinc supplements are consumed with marginal dietary copper intake.

Folate

Folate bioavailability is enhanced when polyglutamate folate is hydrolyzed by the zinc-dependent enzyme, polyglutamate hydrolase, to the monoglutamate. This occurrence suggests a possible point of interaction. Some studies have shown a relationship between folate and zinc (Milne et al., 1984), with low zinc intake decreasing folate absorption/status. More recent evidence does not support any effect of low zinc intake on folate utilization and shows that folate supplementation does not adversely affect zinc status (Kauwell et al., 1995). Extensive studies on this potential relationship have not been carried out in women. Given that these nutrients have important functions in both fetal and postnatal development, the relationship requires further study.

Protein

Zinc binds tenaciously to proteins at near neutral pH. Consequently, the amount of protein in the diet is a factor contributing to the efficiency of zinc absorption. As protein digestion proceeds, zinc becomes more accessible for zinc transport mechanisms of intestinal cells. The relative abundance of zinc as small molecular weight complexes of low binding affinity enhances the process. Small changes in protein digestion may produce significant changes in zinc absorption (Sandstrom and Lonnerdal, 1989). These changes in absorption may explain the correlation between zinc deficiency symptoms and certain malabsorption disorders (Cousins, 1996). The markedly greater bioavailability of zinc from human milk than from cow's milk is an example of how protein digestibility, which is much lower in casein-rich cow's milk than in human milk, influences zinc absorption (Roth and Kirchgessner, 1985). In general, zinc absorption from a diet high in animal protein will be greater than from a diet rich in proteins of plant origin such as soy (King and Keen, 1999).

Phytic Acid

Plants contain phytic acid (myo-inositol hexaphosphate) for use as a storage form of phosphorus. Consequently, plant-based foods, particularly grains and legumes, have a significant phytic acid content. Enzymatic action of yeast during the leavening of bread and other fermentations reduce phytate levels, whereas extrusion processes (used in preparation of some breakfast cereals), may not (Williams and Erdman, 1999). In Caco-2 cells, the metal binding property of phytic acid decreases proportionally as fewer than six phosphate groups are bound to each inositol molecule (Han et al., 1994). Phytate binding of zinc has been demonstrated as a contributing factor for the zinc deficiency related to consumption of unleavened bread seen in certain population groups in the Middle East (Prasad, 1991). The overall effect of phytate is to reduce zinc absorption from the gastrointestinal tract through complexation and precipitation (Oberleas et al., 1966). These chemical effects appear to be enhanced by simultaneous binding of calcium. Phytate binding in the intestinal lumen includes zinc of both food origin and endogenous origin. Since zinc homeostasis is controlled in part by endogenous secretions, consumption of phytate-rich foods may be of practical importance as a factor that limits absorption and maintenance of zinc balance. While high-fiber-containing foods tend also to be phytate-rich, fiber alone may not have a major effect on zinc absorption.

Picolinic Acid

A metabolite of tryptophan metabolism, picolinic acid has a high metal binding affinity. Picolinate complexes of zinc and chromium are not formed in nature in appreciable amounts, but are sold commercially as dietary supplements. Zinc picolinate as a zinc source for humans has not received extensive investigation. In an animal model, picolinic acid supplementation promoted negative zinc balance (Seal and Heaton, 1985), presumably by promoting urinary excretion.

Method Used to Set the Adequate Intake

Using the method described in Chapter 2, an AI has been used as the goal for intake during the first 6 months of life. The AI is based on the maternal zinc supply to the infant exclusively fed human milk.

There is an unusually rapid physiologic decline in the zinc concentration of human milk and consequently in the zinc supplied to infants fed human milk during the first 6 months of lactation (Krebs et al., 1985, 1994, 1995; Moser and Reynolds, 1983) (Figure 12-3). Concentrations of zinc in human milk decline from approximately 4 mg/L at 2 weeks to 3 mg/L at 1 month, 2 mg/L at 2 months, 1.5 mg/L at 3 months, and 1.2 mg/L at 6 months postpartum (Krebs et al., 1995; see Table 12-1). With a standard volume of intake of 0.78 L/day (Chapter 2), calculated zinc intakes are 2.15 mg/day at 1 month, 1.56 mg/day at 2 months, 1.15 mg/day at 3 months, and 0.94 mg/day at 6 months (Table 12-1). Measured zinc intake of infants fed human milk was 2.3 mg/day at 2 weeks and 1 mg/day at 3 months (Krebs et al., 1994).

FIGURE 12-3

Average zinc consumption from human milk during the first 12 months of lactation. SOURCE: Table 12-1.

TABLE 12-1

Zinc Concentration in Human Milk.

In order to match the zinc intake of the infant in early weeks (Figure 12-3), the AI is set at 2.0 mg/day (2.5 mg/L × 0.78 L/day). This amount appears to be generous at ages 4 to 6 months when evaluated by zinc intake from human milk at this age, and human milk has been shown to result in weight gain and body lengths similar to those of infants provided complementary foods at 4 to 6 months (Dewey et al., 1999). A positive association between zinc content of human milk at 5 months and changes in the weight-forage Z scores for the 5- to 7-month interval have, however, been documented (Krebs et al., 1994). There is also some evidence, however, that growth-limiting zinc deficiency can occur in infants principally fed human milk after the age of 4 months (Walravens et al., 1992).

Factorial estimates of requirements (i.e., 2.1 mg/day at 1 month and 1.54 mg/day at 5 months) are consistent with this AI for infants ages 0 through 6 months. These factorial estimates are based on measurements of zinc intake of infants fed human milk, fractional absorption, and endogenous losses (Krebs et al., 1996). Integumental and urine losses are from published calculations (Krebs and Hambidge, 1986). Also consistent with this AI is an earlier report that physical growth of male infants fed a zinc-fortified cow milk formula (5.8 mg/L) was greater than that of infants receiving the same formula but with a zinc concentration of 1.8 mg/L, which provided about 1.4 mg/day of zinc (Walravens and Hambidge, 1976).

Zinc AI Summary, Ages 0 through 6 Months

Special Considerations

The zinc concentration in cow milk ranges from 3 to 5 mg/L (Lonnerdal et al., 1981) which is greater than the average concentration in human milk (Table 12-1). Singh and coworkers (1989) reported that approximately 32 percent of zinc in cow milk is bound to casein and the majority of the remaining zinc (63 percent) is bound to colloidal calcium phosphate. The absorption of zinc from human milk is higher than from cow milk-based infant formula and cow milk (Lonnerdal et al., 1988; Sandstrom et al., 1983). The zinc bioavailability from soy formulas is significantly lower than from milk-based formulas (Lonnerdal et al., 1988; Sandstrom et al., 1983).

Evidence Considered in Estimating the Average Requirement

Intake from Human Milk. Zinc nutriture in later infancy is quite different from that in the younger infant. It is likely that neonatal hepatic stores, which may contribute to metabolically usable zinc pools in early postnatal life, have been dissipated (Zlotkin and Cherian, 1988). Human milk provides only 0.5 mg/day of zinc by 7 months postpartum (Krebs et al., 1994), and the concentration declines even further by 12 months (Casey et al., 1989). It is apparent, therefore, that human milk alone is an inadequate source of zinc after the first 6 months. As a result, extrapolation from human milk intake during the 0 through 6 months postpartum period, which yields 2.4 mg/day, does not reflect adequate zinc intake during the second 6 months.

Intake from Human Milk and Complementary Foods. Data from the Third National Health and Nutrition Examination Survey indicate that the median intake of zinc from complementary foods is 1.48 mg/day (n = 45) for older infants consuming human milk. Thus, the average zinc intake from human milk and complementary foods is estimated to be approximately 2 mg/day (0.5 + 1.48).

Factorial Analysis. Excretion of endogenous zinc is used to estimate the physiological requirement of zinc in older infants and young children. The Estimated Average Requirement (EAR) for zinc is determined by dividing the physiological requirement by the fractional zinc absorption. Apart from some data on excretion of zinc in the urine (Alexander et al., 1974; Cheek et al., 1968; Ziegler et al., 1978), direct measurements of endogenous zinc excretion are not available for older infants, children, or adolescents. These endogenous zinc losses (intestinal, urinary, and integumental), therefore, are estimated by extrapolation from measured values for either adults (see “Adults Ages 19 Years and Older”) or younger infants. These extrapolations have been based on a reference weight.

Intestinal losses vary directly with the quantity of zinc absorbed (see “Adults Ages 19 Years and Older”). The average intestinal excretion of endogenous zinc in infants aged 2 to 4 months who receive human milk is approximately 50 μg/kg/day (Krebs et al., 1996). There is a “critical” level of intestinal excretion of endogenous zinc in adults at which the quantity of absorbed zinc is equal to the total endogenous zinc losses. This critical level, derived from all available sets of data for adult men, yields an average excretion of 34 μg/kg/ day of zinc and is used for children beyond 1 year of age and adolescents. Therefore, 50 μg/kg/day is used for older infants and 34 μg/ kg/day for children aged 1 through 3 years. It is recognized that this is an approximation, not only because of the extrapolation of values but also because intestinal excretion of endogenous zinc is strongly correlated with zinc absorption.

Urinary losses of zinc are approximately 7.5 μg/kg/day for both men and women (see “Adults Ages 19 Years and Older”). After early infancy, excretion rates for children on a body weight basis seem to differ very little from adult values (Krebs and Hambidge, 1986). No data are available on the integumental losses in children, so estimates for children are derived from data in adult men (Johnson et al., 1993), which provide an estimate of 14 μg/kg/day of zinc. Therefore, the estimated total endogenous excretion of zinc is 64 μg/kg/ day for older infants and 48 μg/kg/day for children aged 1 through 3 years.

Requirements for Growth. These requirements have been estimated from chemical analyses of infants and adults, which give an average concentration of 20 μg/g wet weight of zinc (Widdowson and Dickerson, 1964). It is assumed that each gram of new lean and adipose tissue requires this amount of zinc. The average amount of new tissue accreted for older infants and young children is 13 and 6 g/ day, respectively (Kuczmarski et al., 2000).

With the estimates above, the total amount of absorbed zinc required for infants ages 7 through 12 months is 836 μg/day (Table 12-2). The corresponding value for children ages 1 through 3 years is 744 μg/day (Table 12-3).

TABLE 12-2

Requirement for Absorbed Zinc for Infants Aged 7 through 12 Months.

TABLE 12-3

Requirement for Absorbed Zinc for Children Aged 1 through 3 Years.

Fractional Absorption of Dietary Zinc. Fractional absorption probably has the greatest variation of any of the above physiological factors, depending as it does on numerous factors including quantity of ingested zinc, nutritional status, and bioavailability. Although a “critical” average fractional absorption of 0.4 has been derived from the data sets used for adult men (see “Adults Ages 19 Years and Older”), a more conservative value of 0.3 is used for preadolescent children. This value is based on studies of infants and young children reported by Fairweather-Tait and coworkers (1995) and Davidsson and coworkers (1996). To calculate the dietary zinc requirement based on the fractional zinc absorption, it is assumed that the older infant continues to be fed human milk between 7 and 12 months of age along with complementary foods. The fractional absorption of zinc from human milk continues to approximate 0.5 (Abrams et al., 1997). Based on an average intake of 500 μg/day from human milk and a fractional absorption of 0.5, the amount of zinc ingested from milk is approximately 250 μg/day. Therefore the estimated absorbed zinc required from complementary foods is 586 μg/day (836 – 250). Applying a fractional absorption of 0.3, zinc intake required from complementary foods is 1.95 mg/day (586 ÷ 0.3). Therefore, the EAR for infants ages 7 through 12 months is 2.5 mg/day (0.5 + 1.95). For children ages 1 through 3 years, a fractional absorption of 0.3 is used to estimate the required dietary zinc resulting in an EAR of 2.5 mg/day (744 ÷ 0.3), after rounding.

Extrapolation from Adults. An average requirement of 2.3 and 3.0 mg/day for older infants and young children, respectively, is calculated with use of the method described in Chapter 2 that extrapolates from the adult EAR based on body size.

Growth. Limited dietary zinc data are available for children in this age group. In a 6-month, placebo-controlled, randomized zinc supplementation study (Walravens et al., 1989), a major criterion for inclusion was a weight-for-age less than the tenth percentile in apparently healthy young children with no organic disease and no detectable family dynamic issues that might explain failure to thrive. Compared with placebo-treated control subjects, the zinc-supplemented children had a significantly greater increase in mean weight-for-age Z-scores. Inspection of the individual data points indicated that 87.5 percent of zinc-supplemented subjects had an increase in weight-for-age Z-scores compared with 52 percent of control subjects. These results indicate that 35.5 percent of 10 percent, or 3.6 percent of the overall population in this age group, had growth-limiting zinc deficiency. The calculated mean dietary intake at baseline for the placebo-treated children was 4.1 ± 0.8 mg/day (standard deviation [SD]) of zinc. Subtraction of two SDs from this population mean gives an EAR of 2.5 mg/day. It is likely that this calculation errs on the low side because of the variability associated with 24-hour recall dietary information and because some children with weight-for-age greater than the tenth percentile are also likely to have mild growth-limiting zinc deficiency. Hence this value corresponds reasonably well with the EAR determined from the factorial approach.

Zinc EAR and RDA Summary, Ages 7 Months through 3 Years

The Recommended Dietary Allowance (RDA) for zinc is set by using a coefficient of variation (CV) of 10 percent (see Chapter 1) because information is not available on the standard deviation of the requirement. The RDA is defined as equal to the EAR plus twice the CV to cover the needs of 97 to 98 percent of the individuals in the group (therefore, the zinc RDA is 120 percent of the EAR). The calculated RDA is rounded to the nearest 1 mg.

Evidence Considered in Estimating the Average Requirement

Factorial Analysis. Factorial analysis is used to determine the EAR for children ages 4 through 8 years. The nonintestinal endogenous losses and requirement for growth are based on data previously discussed (see “Infants and Children Ages 7 Months through 3 Years”). For this age group, the average intestinal losses are 34 μg/ kg/day of zinc and the amount of new tissue accreted is 7 g/day (Kuczmarski et al., 2000). Based on the summation of zinc losses and requirements for growth, the required amount of absorbed zinc for this age group is approximately 1.2 mg/day (Table 12-4). With a fractional absorption of 0.3 based on studies in infants and young children (Davidsson et al., 1996; Fairweather-Tait et al., 1995), the EAR is 4.0 mg/day of zinc.

TABLE 12-4

Requirement for Absorbed Zinc for Children Aged 4 through 8 Years.

Extrapolation from Adults. The average requirement for zinc is 4 mg/day as determined by the method described in Chapter 2, which extrapolates from the adult EAR.

Growth. Some dietary data are available from children aged 4 through 8 years whose growth percentiles were at the lower end of the normal range and who were subjects in placebo-controlled, randomized trials of dietary zinc supplementation. In each of two studies, one in Canada (Gibson et al., 1989) and the other in the United States (Walravens et al., 1983), zinc supplementation was associated with greater linear growth gain. Mean dietary intakes of the placebo-treated controls in the Canadian and U.S. studies were 6.4 and 4.6 mg/day of zinc, respectively. No growth response was observed with zinc supplementation of healthy children of either gender, unselected for growth, whose average calculated zinc intake was 6.3 mg/day (Hambidge et al., 1979a). The SDs were too large (likely attributable to methodological limitations) to use these data with any confidence in setting an EAR. However, these data are consistent with the EAR derived from a factorial approach.

Zinc EAR and RDA Summary, Ages 4 through 8 Years

The RDA for zinc is set by using a CV of 10 percent (see Chapter 1) because information is not available on the standard deviation of the requirement. The RDA is defined as equal to the EAR plus twice the CV to cover the needs of 97 to 98 percent of the individuals in the group (therefore, the zinc RDA is 120 percent of the EAR). The calculated RDA was rounded to the nearest 1 mg.

Evidence Considered in Estimating the Average Requirement

Factorial Analysis. Estimates used for factorial analysis are similar for boys and girls, and therefore calculations are used to estimate a single average requirement for both genders. With use of the same values as for younger children, an average accretion of 10 g/day of new tissue (Kuczmarski et al., 2000), and a reference weight of 40 kg, the required amount of absorbed zinc is 2.1 mg/day (Table 12-5). Based on a fractional absorption of 0.3 observed in infants and young children (Davidsson et al., 1996; Fairweather-Tait et al., 1995), the EAR is 7 mg/day.

TABLE 12-5

Requirement for Absorbed Zinc for Children Aged 9 through 13 Years.

Extrapolation from Adults. As determined by the extrapolation method described in Chapter 2, the average requirement for boys and girls is 6.7 and 5.6 mg/day of zinc, respectively.

Zinc EAR and RDA Summary, Ages 9 through 13 Years

The RDA for zinc is set by using a CV of 10 percent (see Chapter 1) because information is not available on the standard deviation of the requirement. The RDA is defined as equal to the EAR plus twice the CV to cover the needs of 97 to 98 percent of the individuals in the group (therefore, the zinc RDA is 120 percent of the EAR). The calculated RDA is rounded to the nearest 1 mg.

Evidence Considered in Estimating the Average Requirement

Factorial Analysis. Endogenous losses are calculated as for younger age groups by using the reference weights (see Chapter 2) with the addition of 100 μg/day of zinc to allow for calculated average semen or menstrual losses (see “Adults Ages 19 Years and Older”, which follows). For this age group, a fractional absorption of 0.4 is used; it corresponds to the average “critical” value for adult men from the data sets used in estimating adult requirements (see below). Gender differences are sufficient at this age for boys and girls requirements to be calculated separately. As determined by the summation of average zinc losses and the zinc requirement for growth (Kuczmarski et al., 2000; Widdowson and Dickerson, 1964), the amount of absorbed zinc that is required for boys and girls is approximately 3.4 and 3.0 mg/day, respectively (Table 12-6). On the basis of a fractional zinc absorption of 0.4 that was derived for men (see below), the EARs for adolescent boys and girls are calculated to be 8.5 and 7.3 mg/day of zinc, respectively.

TABLE 12-6

Requirement for Absorbed Zinc for Adolescent Boys and Girls Aged 14 through 18 Years.

Extrapolation from Adults. Based on the extrapolation method described in Chapter 2, the average requirement for adolescent boys and girls is 9.5 and 6.4 mg/day, respectively.

Zinc EAR and RDA Summary, Ages 14 through 18 Years

The RDA for zinc is set by using a CV of 10 percent (see Chapter 1) because information is not available on the standard deviation of the requirement. The RDA is defined as equal to the EAR plus twice the CV to cover the needs of 97 to 98 percent of the individuals in the group (therefore, the zinc RDA is 120 percent of the EAR). The calculated RDA is rounded up to the nearest 1 mg.

Evidence Considered in Estimating the Average Requirement

As discussed earlier, there are no adequately documented functional or simple laboratory indexes of zinc nutriture that can provide a principal indicator of zinc requirements in adults. However, sufficient data are now available to apply a factorial approach to determine the EAR for adults. With this approach, the principal indicator selected is the minimal quantity of absorbed zinc that is adequate to replace endogenous zinc losses. The EAR is the average zinc intake that provides this quantity of absorbed zinc. An outline of these calculations follows.

Step 1: Calculation of Endogenous Losses of Zinc via Routes Other than the Intestine. Urinary zinc excretion declines only with extreme dietary zinc restriction and is not correlated with zinc ingested by young adult men over a range of 4 to 25 mg zinc/day (Baer and King, 1984; Behall et al., 1987; Coudray et al., 1997; Hallfrisch et al., 1987; Holbrook et al., 1989; Hunt JR et al., 1992; Jackson et al., 1984; Johnson et al., 1982, 1993; Lee et al., 1993; Mahalko et al., 1983; Milne et al., 1983; Snedeker et al., 1982; Spencer et al., 1979; Turnlund et al., 1984, 1986; Wada et al., 1985). In men, therefore, zinc excretion via the kidney should be regarded as a constant in calculating zinc requirements, the average excretion being 0.63 mg/ day. Though fewer data are available, the same constancy appears to be true for combined integumental and sweat losses (Johnson et al., 1993) and losses in semen (Hunt CD et al., 1992; Johnson et al., 1993) for which the zinc losses average 0.54 and 0.1 mg/day, respectively. Therefore, losses of endogenous zinc via routes other than the intestine can be regarded as a constant over the range of dietary zinc intake that encompasses zinc requirements. This average constant for men has been calculated to be 1.27 mg/day (0.63 + 0.54 + 0.1) of zinc. An equal quantity of zinc must be absorbed to match this loss.

In 10 studies, the mean urinary loss of zinc from women was 0.44 mg/day (Colin et al., 1983; Greger et al., 1978; Hallfrisch et al., 1987; Hunt JR et al., 1992, 1998; Miller et al., 1998; Swanson and King, 1982; Taper et al., 1980; Turnlund et al., 1991; Wisker et al., 1991). Reported integumental losses for men are multiplied by 0.86 to adjust for the different average surface area of women, and accordingly the average total zinc endogenous losses are 0.46 mg/ day for women. Menstrual zinc losses are assumed to average 0.1 mg/day (Hess et al., 1977). Therefore, the calculated total loss of endogenous zinc for women via routes other than the intestine is 1.0 mg/day (0.44 + 0.46 + 0.10).

Step 2: Relationship Between Excretion of Endogenous Zinc via the Intestine and Quantity of Zinc Absorbed. In contrast to other endogenous zinc losses, the quantity of endogenous zinc excreted via the intestine is positively correlated with the quantity of zinc absorbed over a wide range. This correlation is shown in Figure 12-1. This figure is based on 10 sets of balance data from seven studies (Hunt JR et al., 1992; Jackson et al., 1984; Lee et al., 1993; Taylor et al., 1991; Turnlund et al., 1984, 1986; Wada et al., 1985) of healthy young men, which also included isotopic tracer measurements of fractional zinc absorption. This correlation, in turn, allows for the quantification of daily zinc absorption and intestinal excretion of endogenous zinc. Importantly, this linear relationship, which indicates that for each milligram of zinc absorbed the intestine excretes approximately 0.6 mg/day of endogenous zinc, has been demonstrated only for zinc absorption ranging from 0.8 to 5.5 mg/day. It is also noted that most of these data were relatively short-term, and these variables were not examined while the participants were consuming habitual diets. However, the studies did extend as long as 6 months, a duration that suggests the observed relationship between absorption and endogenous losses via the intestine is a long-term phenomenon. Therefore, in contrast to other endogenous losses of zinc, losses from the intestine cannot be treated as a constant.

To achieve balance, absorption must match the sum of nonintestinal and intestinal endogenous zinc losses. The minimum amount of zinc that must be absorbed before absorption matches the losses is determined in step 3 below.

Corresponding data for women are both limited and divergent (Hunt JR et al., 1992, 1998; Sian et al., 1996; Turnlund et al., 1991). It has therefore been assumed that there are no significant gender differences for this relationship between absorbed zinc and intestinal excretion of endogenous zinc.

Step 3: Determination of Minimal Zinc Absorption Required to Replace Total Endogenous Zinc Excretion. The sum of nonintestinal endogenous zinc losses (1.27 mg/day for men and 1.0 mg/day for women) is added to the linear regression line for excretion of endogenous zinc in the feces versus absorbed zinc (Figure 12-1). These “adjusted” lines depict the quantitative relationship between absorbed zinc and total endogenous zinc losses for men and women.

The intercept between the dashed line (line of equality for absorbed zinc) and the gender-specific lines is then used to determine the minimal quantity of absorbed zinc required to replace endogenous zinc losses.

With this approach, the calculated average total minimal quantity of absorbed zinc required for the men in these studies is 3.84 mg/ day (1.27 mg to match endogenous zinc losses from nonintestinal sources and, therefore, 2.57 mg/day to match intestinal endogenous zinc losses). The corresponding value for women is 3.3 mg/ day (1.0 mg/day to match endogenous zinc losses from nonintestinal sources and, therefore, 2.3 mg/day to match intestinal endogenous zinc losses).

These calculated average minimal values for absorbed zinc are then used as the principal indicator for establishing an EAR in step 4.

Step 4: Determination of the Average Zinc Intake Required to Achieve Absorption of the Quantity of Zinc Necessary to Match Total Endogenous Losses. The EAR is determined from the asymptotic regression of absorbed zinc on zinc intake (Figure 12-2) that was derived from the same data sets used for Figure 12-1. Thus, if 3.84 mg/day of absorbed zinc is required for men, the amount of ingested zinc, and therefore the EAR, is 9.4 mg/day. When this approach is used for women, the EAR is 6.8 mg/day. This value corresponds to average fractional absorptions of 0.41 and 0.48 for men and women, respectively. A similar fractional absorption of 0.4 was observed for adult men fed experimental diets from which zinc bioavailability is likely to be favorable (August et al., 1989).

Other Criteria for Men. Zinc deficiency has not been documented in healthy adult men in North America with the assessment methods currently in use. Some supportive data have been derived from one of the studies included in the factorial approach outlined above (Wada et al., 1985). This study included six men who received a diet containing 5.5 mg/day of zinc for an 8-week period. At the end of this period, several zinc-responsive biochemical changes had occurred, including declines in serum retinol binding protein, albumin, prealbumin, and thyroxin concentrations (Wada and King, 1986).

Other data from experimental zinc depletion studies are also consistent but at lower levels of intake (zinc intakes of 3 to 5 mg/day). These data include decreased erythrocyte metallothionein (Grider et al., 1990; Thomas et al., 1992) and zinc concentrations, decreased 5′-nucleotidase activity (Beck et al., 1997a), and various abnormalities of laboratory indexes of immune status (Beck et al., 1997b).

Other Criteria for Women. Twenty-six percent of a group of apparently healthy Canadian omnivore women had prebreakfast serum zinc concentrations below the cut-off of 70 μg/dL (Gibson et al., 2000). The zinc intake of these subjects averaged 7.3 mg/day, which by this criterion is slightly above the EAR. These data are consistent with an EAR of 6.8 mg/day.

Elderly. Reported values on the fractional absorption of zinc in the elderly have been quite variable (Couzy et al., 1993; Hunt et al., 1995; Turnlund et al., 1982, 1986), and no consistent evidence indicates that aging affects absorption adversely. Results of balance studies are again, predictably, variable (Bunker et al., 1982; Hallfrisch et al., 1987; Wood and Zheng, 1997). No evidence suggests that the zinc requirements of the elderly are higher than those of younger adults, but possible differences in zinc metabolism (Wastney et al., 1986) merit further investigation.

Other Criteria for the Elderly. Zinc supplementation of 53 elderly men and women whose diet contained an average of 9.2 mg/day of zinc was not associated with any detectable benefits (Swanson et al., 1988). Specifically, there were no changes in circulating protein or immunoglobulin concentrations. In contrast, dietary zinc was positively correlated with serum albumin in a group of 82 elderly Canadians whose zinc intakes averaged 5 mg/day for women and 6.5 mg/day for men (Payette and Gray-Donald, 1991). Several studies in which improvements in laboratory indexes of zinc status with zinc supplementation were reported did not, unfortunately, include information on habitual zinc intake (Boukaiba et al., 1993; Cakman et al., 1997; Duchateau et al., 1981; Fortes et al., 1998). Fifteen older men and women whose habitual dietary zinc averaged 8.8 mg/day had a significant decline in the activity of 5′-nucleotidase activity after a 2-week period during which zinc intake was restricted to 4 mg/day (Bales et al., 1994). Subsequently, a 6-day supplementation period in which total zinc intake averaged 28 mg/day was associated with a significant increase in 5′-nucleotidase activity, but not beyond baseline levels. In 119 elderly women, serum IGF-1 concentration was weakly correlated with dietary zinc over a range of 5 to 17 mg/day (Devine et al., 1998). A nonplacebo controlled study of zinc supplementation in 13 elderly subjects, part of a larger group of 180 subjects whose average calculated zinc intake was 9 mg/day, was reported to result in normalization of zinc in granulocytes and lymphocytes and improvement in various immune parameters (Prasad et al., 1993). Ethanol tolerance tests indicated a change in ethanol metabolism when dietary zinc intake of postmenopausal women was restricted to 2.6 mg/day (Milne et al., 1987).

Zinc EAR and RDA Summary, Ages 19 Years and Older

The RDA for zinc is set by using a CV of 10 percent (see Chapter 1) because information is not available on the standard deviation of the requirement. The RDA is defined as equal to the EAR plus twice the CV to cover the needs of 97 to 98 percent of the individuals in the group (therefore, for zinc the RDA is 120 percent of the EAR). The calculated RDA was rounded to the nearest 1 mg.

Evidence Considered in Estimating the Average Requirement

Factorial Approach. The average daily rates of zinc accumulation by maternal and embryonic/fetal tissues during the four quarters of pregnancy are 0.08, 0.24, 0.53, and 0.73 mg (Swanson and King, 1987). On the assumption of no compensatory change in intestinal excretion of endogenous zinc, it is concluded that increasing daily zinc absorption by these amounts is desirable.

The average fractional absorption of zinc was 27 percent for nonpregnant women from eight studies in which dietary zinc averaged 10 mg/day (Fung et al., 1997; Hunt JR et al., 1992, 1998; Miller et al., 1998; Sian et al., 1996; Turnlund et al., 1991). Increases in fractional absorption during pregnancy have been reported to be nonsignificant (Fung et al., 1997), but this outcome may reflect inadequate power of the study design. Therefore, increases in dietary zinc requirements during pregnancy are calculated to be the following:

To set a single EAR for pregnant women, the EAR is based on the additional requirement during the fourth quarter (2.7 mg/day) of pregnancy plus the EAR for nonpregnant adolescent girls and women. It should be noted, however, that the zinc requirement during the first quarter of pregnancy is only minimally greater than the preconceptional requirement.

Other Criteria. Dietary supplementation reduced the decline in plasma/serum zinc concentration across pregnancy in a large cohort of Peruvian women whose dietary zinc intake was estimated to be 7 mg/day (Caulfield et al., 1999a), but not in North American women whose dietary zinc intake averaged 11 mg/day (Hambidge et al., 1983). Correlations observed between maternal biochemical indexes of zinc status and complications of pregnancy, delivery, and fetal development have been inconsistent.

Gravid women with a zinc intake of 6 mg/day or less were found to have a high incidence of premature deliveries (Scholl et al., 1993). Increased gestational age at delivery and increased birth size have been reported to result from zinc supplementation of pregnant African-American women whose baseline dietary zinc intake was calculated to be 13 mg/day (Goldenberg et al., 1995). This calculated dietary zinc intake is notably high in comparison with other data for African-American women (Mares-Perlman et al., 1995). Without additional supporting documentation, it is difficult to reconcile the implications of the results of this study (with respect to dietary zinc requirements during pregnancy) with the EARs derived from a factorial approach. Nor is it easy to reconcile these findings with the results of other intervention studies. For example, no effect of zinc supplements on birth size was observed in a recent large-scale study of Peruvian women whose dietary zinc intake was estimated to be 7 mg/day (Caulfield et al., 1999b). There was, however, evidence of improved fetal neurobehavioral development (Merialdi et al., 1998).

A report that zinc intakes of less than 7.5 mg/day during the third trimester are associated with lower zinc concentrations in human milk is consistent with the EAR (Ortega et al., 1997).

Zinc EAR and RDA Summary, Pregnancy

The RDA for zinc is set by using a CV of 10 percent (see Chapter 1) because information is not available on the standard deviation of the requirement. The RDA is defined as equal to the EAR plus twice the CV to cover the needs of 97 to 98 percent of the individuals in the group (therefore, for zinc the RDA is 120 percent of the EAR). The calculated RDA is rounded to the nearest 1 mg.

Evidence Considered in Estimating the Average Requirement

Losses in Human Milk. Average concentrations of zinc in human milk decline physiologically from approximately 4 mg/L at 2 weeks postpartum to 3 mg/L at 4 weeks, 2 mg/L at 8 weeks, 1.5 mg/L at 12 weeks, and 1.2 mg/L at 24 weeks (Krebs et al., 1995; Moser-Veillon and Reynolds, 1990). With use of a standard volume of 0.78 L/day of human milk secreted per day (Chapter 2), calculated zinc losses via the mammary gland are 2.15 mg/day at 4 weeks, 1.56 mg/ day at 8 weeks, 1.17 mg/day at 12 weeks, and 0.94 mg/day at 24 weeks.

Postpartum involution of the uterus and decreased maternal blood volume should release approximately 30 mg of zinc that has been accumulated during pregnancy (King and Turnlund, 1989); that is, an average of approximately 1 mg/day for the first month. It is reasonable to assume that this endogenous zinc is available for reutilization. Thus, 1 mg/day is subtracted from the amount of zinc lost during the first 4 weeks of lactation. The loss of zinc for weeks 8, 12 and 24 are averaged:

Week 4: (2.15 – 1.0) = 1.15 mg/day of zinc

Week 8: (2.15 + 1.56) ÷ 2 = 1.85 mg/day of zinc

Week 12: (1.56 + 1.17) ÷ 2 = 1.36 mg/day of zinc

Weeks 12–24: (1.17 + 0.94) ÷ 2 = 1.05 mg/day of zinc

The average calculated increased requirement for absorbed zinc during lactation is 1.35 mg/day.

Reported values for fractional absorption of zinc for adult women outside the reproductive cycle averages 27 percent (Fung et al., 1997; Hunt JR et al., 1992, 1998; Sian et al., 1996; Turnlund et al., 1991). If this value were applied to the calculation of increased dietary zinc required during lactation (1.35 ÷ 0.27), the average dietary requirement would increase by 5 mg/day. However, the fractional absorption of zinc increases during lactation by 0.107 (Fung et al., 1997). Therefore, the fractional absorption would be increased to 0.716 (0.27 ÷ 0.377) to give an additional requirement of 3.6 mg/day (5 × 0.716). This value is added to the EAR for adolescent girls and women to set the EAR during lactation.

Other Criteria. Typically, human milk zinc concentrations are not increased by the administration of a daily zinc supplement across lactation (Kirksey et al., 1979; Krebs et al., 1995; Moser-Veillon and Reynolds, 1990). In one study, however, a modest but statistically significant reduced rate of decline in zinc concentrations in milk across lactation was observed with a zinc supplement (Krebs et al., 1985). The dietary zinc in the placebo group averaged 10.7 mg/ day. In a subsequent study, in which the average dietary zinc was higher at 13.0 mg/day, there was no evidence of an effect of zinc supplementation on zinc concentration in milk.

Zinc EAR and RDA Summary, Lactation

The RDA for zinc is set by using a CV of 10 percent (see Chapter 1) because information is not available on the standard deviation of the requirement. The RDA is defined as equal to the EAR plus twice the CV to cover the needs of 97 to 98 percent of the individuals in the group (therefore, for zinc the RDA is 120 percent of the EAR). The calculated RDA is rounded to the nearest 1 mg.

Vegetarianism

Cereals are the primary source of dietary zinc for vegetarians (Gibson, 1994). The bioavailability of zinc in vegetarian diets is reduced if the phytate content in the diet is high (Gibson, 1994), and this may result in low zinc status (Freeland-Graves et al., 1980b). Absorption of zinc from vegetarian diets is lower than from nonvegetarian diets (Hunt et al., 1998; Kies, 1988); however, relatively minor changes to the diet can improve zinc absorption (Gibson et al., 1997; Harland et al., 1988). Vegetarian diets rich in calcium may negatively affect zinc bioavailability (Ellis et al., 1987).

Zinc intake from vegetarian diets has been found to be both similar to intake from nonvegetarian diets (Alexander et al., 1994; Berglund et al., 1994; Donovan and Gibson, 1996; Johansson and Widerstrom, 1994; Kelsay et al., 1988; Levin et al., 1986; Srikumar et al., 1992) and lower than intake from nonvegetarian diets (Faber et al., 1986; Freeland-Graves et al., 1980a; Harland and Peterson, 1978; Hunt et al., 1998; Janelle and Barr, 1995). In most older adult and elderly populations, vegetarians have lower zinc intakes than nonvegetarians (Brants et al., 1990; Hunt et al., 1988; Lowik et al., 1990). Among vegetarians, zinc concentrations in serum, plasma, hair, urine, and saliva are either the same as or lower than those of nonvegetarians (Anderson et al., 1981; Freeland-Graves et al., 1980a, 1980b; Hunt et al., 1998; Kadrabova et al., 1995; King et al., 1981; Krajcovicova-Kudlackova et al., 1995; Levin et al., 1986; Srikumar et al., 1992). The variations in these status indicators are most likely due to the amount of phytate, fiber, calcium, or other inhibitors of zinc absorption in the vegetarian diets. Individuals consuming vegetarian diets were found to be in positive zinc balance (Ganapathy et al., 1981; Hunt et al., 1998).

The requirement for dietary zinc may be as much as 50 percent greater for vegetarians and particularly for strict vegetarians whose major food staples are grains and legumes and whose dietary phytate:zinc molar ratio exceeds 15:1. At this time there are not sufficient data to set algorithms for establishing dietary requirements for zinc on the basis of the presence and concentration of other nutrients and food components.

Alcohol

Long-term alcohol consumption is associated with impaired zinc absorption and increased urinary zinc excretion. Low zinc status is observed in approximately 30 to 50 percent of alcoholics. Thus, with long-term alcohol consumption, the daily requirement for zinc will be greater than that estimated via the factorial approach.

Adverse Effects

Acute Effects. Acute adverse effects of excess zinc have been reported. These include epigastric pain, nausea, vomiting, loss of appetite, abdominal cramps, diarrhea, and headaches (Prasad, 1976; Samman and Roberts, 1987). Fosmire (1990) estimated that an emetic dose of zinc sulfate was approximately 1 to 2 g of the salt (225 to 450 mg of zinc). Gastrointestinal distress has been reported at doses of 50 to 150 mg/day of supplemental zinc (Freeland-Graves et al., 1982).

Immunological Response. Intake of 300 mg/day of supplemental zinc as the sulfate for 6 weeks has been shown to cause some functional impairment in immunological response as well as significantly decreased concentrations of HDL cholesterol (Chandra, 1984).

Lipoprotein and Cholesterol. Two studies (Black et al., 1988; Hooper et al., 1980) have found that zinc at doses between 50 and 160 mg/ day decreased serum lipoprotein and cholesterol concentrations in men. Samman and Roberts (1988), however, reported no depression of HDL concentrations in men at 150 mg/day of zinc and found some indication of a depression of low-density lipoproteins (LDL) in women. The different response to excess zinc in women was supported by an earlier study by Freeland-Graves and coworkers (1982). The reduction in HDL cholesterol concentration was shown to be transient and not dose related.

Reduced Copper Status. Reduced copper status has been associated with increased zinc intake (Boukaiba et al., 1993; Burke et al., 1981; Festa et al., 1985; Fischer et al., 1984; Prasad et al., 1978; Samman and Roberts, 1988; Yadrick et al., 1989) (Table 12-7). In all studies in which the interaction of excess zinc and copper was measured, there was a consistent decrease in erythrocyte copper-zinc superoxide dismutase (ESOD) activity, an erythrocyte enzyme indicative of copper status. Yadrick and coworkers (1989) reported this effect after total zinc intakes of about 60 mg/day (50-mg supplement plus 10 mg of dietary zinc) for up to 10 weeks. Although the clinical significance of the depressed ESOD activity is unknown, this marker enzyme is known to be a sensitive indicator of the effect of high zinc levels on copper homeostasis.

TABLE 12-7

Effect of Increasing Doses of Zinc (Zn) Intake on Copper (Cu) Status.

Zinc-Iron Interactions. Zinc and iron are known to interact, and Whittaker (1998) has reviewed the available studies (also see “Factors Affecting the Zinc Requirement”). The primary effect appears to be a decreased absorption of zinc at an iron:zinc ratio of 3:1 when the iron was administered in water. However, when iron was administered during a meal, no such effect was found. Similarly, when iron was present as heme iron, no effect was noted. One study found a 56 percent decline in iron absorption when the zinc:iron ratio was 5:1 and was administered in water (Rossander-Hulten et al., 1991). However, when this ratio of zinc and iron was administered in a hamburger meal, no effect on iron absorption was noted.

Other Endpoints. No evidence was found of reproductive effects in humans from zinc intake. There is one case report of three premature deliveries and one stillborn infant after excess zinc intake during pregnancy (Kumar, 1976). Because details on other contributing factors were not provided, interpretation of these results is limited. There is insufficient evidence of carcinogenicity from human or animal studies.

Summary

Although there are no data indicating adverse interactions between zinc and other nutrients when zinc is found in food, adverse nutrient interactions are present after feeding zinc in the form of dietary supplements. The adverse effect of excess zinc on copper metabolism (i.e., reduced copper status) was chosen as the critical effect on which to base a UL for total daily intake of zinc from food, water, and supplements in humans. This selection is based on (1) the consistency of findings from studies measuring the interaction of zinc and copper (Fischer et al., 1984; Samman and Roberts, 1988; Yadrick et al., 1989), (2) the sensitivity of ESOD activity as a marker for this effect, and (3) the quality and completeness of the database for this endpoint. The data on the effects of zinc on HDL cholesterol concentration were not consistent from study to study and therefore were not used to derive a UL.

Adults

Data Selection. Data on reduced copper status in humans were used to derive a UL for zinc (Table 12-7). Studies measuring ESOD activity (which is a sensitive indicator of copper status) or other indicators of copper status (such as ceruloplasmin or serum copper concentration) were considered optimal for the dose-response assessment.

Identification of a No-Observed-Adverse-Effect Level (NOAEL) and Lowest-Observed-Adverse-Effect Level (LOAEL). A LOAEL of 60 mg/day is based on the study of Yadrick and coworkers (1989) who evaluated copper status after supplemental intake of 50 mg/day as zinc gluconate in 18 healthy women (aged 25 to 40 years) for 10 weeks. ESOD activity was significantly lower than pretreatment values. Although no dietary zinc or copper intakes were reported, a level of dietary zinc can be estimated at approximately 10 mg/day for women (aged 19 to 50 years) from the 1988–1994 Third National Health and Nutrition Examination Survey (Appendix Table C-26). A LOAEL of 60 mg/day was calculated by adding the supplemental intake of 50 mg/day with the rounded estimate of dietary intake, 10 mg/day. Support for a LOAEL of 60 mg/day is provided by other studies showing altered copper balance after zinc supplementation (Fischer et al., 1984) (Table 12-7).

Uncertainty Assessment. An uncertainty factor (UF) of 1.5 was selected to account for interindividual variability in sensitivity and for extrapolation from a LOAEL to a NOAEL. Because reduced copper status is rare in humans, a higher UF was not justified.

Derivation of a UL. A LOAEL of 60 mg/day was divided by a UF of 1.5 to derive a UL of 40 mg/day for total intake of zinc from food, water, and supplements.

Zinc UL Summary, Ages 19 Years and Older

Infants, Children, and Adolescents

Data Selection. There is only one case report of zinc-induced copper deficiency anemia in a young child (Botash et al., 1992): a 13-month-old girl was given 16 mg/day of zinc for 6 months followed by 24 mg/day for 1 month. There are no reports on the adverse effects of zinc on copper status in children or adolescents. The UL values for infants are based on a study by Walravens and Hambidge (1976).

Identification of a NOAEL. Walravens and Hambidge (1976) fed 68 healthy, full-term infants either formula containing 1.8 mg/L of zinc (control) or the same formula supplemented with an additional 4 mg/L (total of 5.8 mg/L) of zinc for 6 months. No effects of zinc on serum copper or cholesterol concentrations or other adverse effects were found. Thus, 5.8 mg/L is the NOAEL selected. Multiplying the NOAEL for infants 0 through 6 months of age by the estimated average intake of human milk of 0.78 L/day (Allen et al., 1991; Butte et al., 1984; Heinig et al., 1993) results in a NOAEL of 4.5 mg/day.

Uncertainty Assessment. The length of the study by Walravens and Hambidge (1976) and the high number of infants justifies a UF of 1.0, given that there is no evidence that intakes from formula of 5.8 mg/L of zinc result in infant toxicity.

Derivation of a UL. The NOAEL of 4.5 mg/day was divided by a UF of 1.0 to obtain a UL of 4 mg/day (rounded down) for infants ages 0 through 6 months. No adverse effects of zinc in children and adolescents could be found. Due to a dearth of information, the UL for young infants was adjusted for older infants, children, and adolescents on the basis of relative body weight as described in Chapter 2 and using reference weights from Chapter 1 (Table 1-1). Values have been rounded down.

Zinc UL Summary, Ages 0 through 18 Years

Pregnancy and Lactation

Because the UL is based on reduced copper status and because there are inadequate data to justify a different UL for pregnant and lactating women, the UL for pregnant and lactating women is the same as that for nonpregnant and nonlactating women.

Zinc UL Summary, Pregnancy and Lactation

Special Considerations

Individuals with Menke's disease may be distinctly susceptible to the adverse effects of excess zinc intake. Since Menke's disease is a defect in the ATPase involved in copper efflux from enterocytes, supplying extra zinc will likely further limit copper absorption (Yuzbasiyan-Gurkan et al., 1992). Brewer and coworkers (1993) demonstrated the effectiveness of zinc therapy in reducing copper accumulation in individuals with Wilson's disease. The UL is not meant to apply to individuals who are being treated with zinc under close medical supervision.