Longitudinal measurements of zinc absorption in Peruvian children consuming wheat products fortified with iron only or iron and 1 of 2 amoun
the Program in International Nutrition and the Department of Nutrition, University of California, Davis (DLdR, JMP, and KHB)
the Instituto de Investigación Nutricional, Lima, Perú (DLdR, MS, and MEP)
the Section of Nutrition, Department of Pediatrics, University of Colorado, Denver (KMH and NFK).
ABSTRACT
Background:Information is needed on the fractional absorption of zinc (FAZ) and absorbed zinc (AZ) during prolonged exposure to zinc-fortified foods.
Objective:The objective was to measure FAZ and AZ from diets fortified with different amounts of zinc and to determine whether zinc absorption changes over 7 wk.
Design:Forty-one stunted, moderately anemic children received daily, at breakfast and lunch, 100 g wheat products fortified with 3 mg Fe (ferrous sulfate) and 0 (group Zn-0), 3 (group Zn-3), or 9 (group Zn-9) mg Zn (zinc sulfate) per 100 g flour. FAZ was measured on days 2–3 and 51–52; meal-specific AZs were calculated as the product of FAZ and zinc intake.
Results:For the breakfast and lunch meals combined, mean total zinc intakes were 2.14, 4.72, and 10.04 mg/d in groups Zn-0, Zn-3, and Zn-9, respectively, during the initial absorption studies; mean (±SD) FAZ values were 0.341 ± 0.111, 0.237 ± 0.052, and 0.133 ± 0.041, respectively, on days 2–3 (P < 0.001) and did not change significantly on days 51–52 in the subset of 31 children studied twice. Mean initial AZ was positively related to zinc intake (0.71 ± 0.18, 1.11 ± 0.21, and 1.34 ± 0.47 mg/d, respectively; P < 0.001); final values did not differ significantly from the initial values.
Conclusions:AZ from meals containing zinc-fortified wheat products increases in young children relative to the level of fortification and changes only slightly during 7-wk periods of consumption. Although consumption of zinc-fortified foods may reduce FAZ, zinc fortification at the levels studied positively affects total daily zinc absorption, even after nearly 2 mo of exposure to zinc-fortified diets.
Key Words: Zinc iron wheat fortification absorption stable isotopes children
INTRODUCTION
Zinc deficiency is associated with poor growth (1), depressed immune function (2), increased susceptibility to and severity of infections (3-5), adverse outcomes of pregnancy (6), and neurobehavior abnormalities (7). In many lower-income countries, the diets are composed primarily of cereals and legumes, which contain substantial amounts of phytate (myo-inositol hexaphosphate), a compound known to inhibit zinc absorption. These diets contain few animal-source foods (muscle meats or organs), which are rich sources of zinc and are free of phytate (8). Recent analyses of national food balance data suggest that 20% of the world's population is at risk of inadequate zinc intake (9). Moreover, approximately one-third of children in low-income countries have a low height-for-age relative to international reference data. Zinc supplementation increases linear growth in stunted children (1), which suggests that these high rates of stunting may be due, in part, to zinc deficiency. Thus, zinc deficiency appears to be widespread in lower-income countries, and intervention programs are needed to improve zinc status in these high-risk populations.
Food fortification is considered to be one of the most cost-effective strategies for improving micronutrient status (10). Cereal products, such as wheat flour, are currently being fortified with iron in many developing countries, and simultaneous fortification with zinc could be achieved with a small incremental cost (9). However, to design rational zinc-fortification programs, additional information is needed, such as 1) the amount of zinc that is absorbed from foods fortified with different amounts of zinc, 2) the possible influence of zinc status on zinc absorption, 3) the effect of zinc fortification on zinc absorption from nonfortified foods or meals, and 4) the effect of zinc fortification on the absorption of other minerals from the diet.
Whole-body zinc homeostasis is achieved primarily by means of the balance between the absorption of dietary zinc and the fecal excretion of endogenous zinc (11, 12). Although it is known that the amount of zinc in a meal affects zinc absorption from that meal (13, 14), quantitative data from human studies are limited, especially in children. Also, there is little information on the effects of zinc intake in one meal on zinc absorption from other meals consumed during the day. Furthermore, although it has been postulated that differences in zinc intake that lead to changes in zinc status may affect the longer-term efficiency of zinc absorption (15), there is limited information on the relation between an individual's zinc status and intestinal absorption of zinc. Thus, the objectives of the present study were to 1) measure zinc absorption from meals containing zinc-fortified products and from other meals in relation to the level of zinc fortification of iron-fortified wheat flour, and 2) determine whether zinc absorption changes after a more prolonged exposure to zinc-fortified foods.
SUBJECTS AND METHODS
Study design
Fifty-eight children 3–4 y of age, who were residing in a poor community on the periphery of Lima, Peru and who were considered to be at high risk of zinc deficiency, were randomly assigned to receive in 2 meals/d a total of 100 g wheat products fortified with 3 mg Fe as ferrous sulfate and either 0 (group Zn-0), 3 (group Zn-3) or 9 (group Zn-9) mg Zn as zinc sulfate per 100 g flour for 70 d. Fractional absorption of zinc (FAZ) was assessed by measuring urinary ratios of different stable isotopic tracers administered orally and intravenously on study days 2–3 and 51–52 for the 2 sets of absorption studies, and AZ was calculated as the meal-specific FAZ multiplied by the zinc intake from the respective meal or meals.
Assessment of usual zinc intakes in the study community
To develop locally appropriate diets for the absorption studies, we completed preliminary observations of the dietary intakes of 50 children aged 3–4 y in the study community. For these background dietary studies, food intakes were measured quantitatively during 12-h daytime observations in the children's homes by subtracting the amounts of any leftovers from the amounts served. Any additional foods consumed during the preceding 12 h were elicited from the children's parents. The food intakes were converted to nutrient intakes by using data from a Peruvian food-composition table (16) and additional information on the phytate contents of these foods (17). These background dietary studies indicated that the children in the study area had slightly low intakes of zinc and iron and moderately high phytate:zinc ratios (Table 1). The major sources of zinc were rice, chicken, milk, wheat bread, potato, oatmeal, wheat noodles, and beans. The diets used for the subsequent absorption studies were designed to mimic the characteristics of the usual home diets, except that additional wheat products were substituted for other cereals to permit consumption of the desired amounts of zinc from the fortified wheat products.
Twenty-five of the children underwent a second round of dietary assessments to determine intraindividual variations in zinc intake and to estimate the number of days of observation that would be required to characterize an individual's usual zinc intake, as described by Beaton et al (18). The results of these latter studies indicated that a total of 7 24-h dietary recalls would provide estimates of individual children's usual zinc intake within ±25% with 95% confidence. Thus, we attempted to schedule 7 sets of 24-h dietary recall histories of the absorption study subjects both before they initiated the first round of the tracer studies and between the 2 rounds of studies.
Absorption study subjects
A total of 765 children aged 3–4 y were screened to identify children who were deemed to have a high risk of zinc deficiency because of the presence of stunting (defined as height-for-age z score <–2.0 SD with respect to international reference data; 19) and moderate anemia (hemoglobin <110 and >90 g/L). Anemia was used as a screening factor for possible zinc deficiency because iron deficiency is the most common cause of anemia in this population and many of the same dietary factors that affect iron status would be expected to influence zinc status similarly. Whereas hemoglobin could be measured with a portable device in the field, information on plasma zinc concentration was not available immediately.
Fifty-eight children identified as having a high risk of zinc deficiency were enrolled in the subsequent absorption studies to examine the effects of different levels of zinc fortification of iron-fortified wheat products on zinc absorption (Figure 1). The absorption studies were repeated after a period of 7 wk during which the children received the same fortified foods each day to determine whether their efficiency of zinc absorption changed during this period of time. In this article, we present the results from 41 of these children for whom absorption data were collected successfully at the initial time point and from the subset of 31 children for whom absorption data were obtained at both time points. The reasons that absorption data were not available from the remaining children are described below and summarized in Figure 1.
Parents of all participating children provided written informed consent at the beginning of the study. The study protocol and consent forms were approved by the institutional review boards of the University of California, Davis, and the Instituto de Investigación Nutricional in Lima.
Study diets
With reference to the results of the background dietary studies, a 7-d rotating menu was designed to provide the study subjects with basal diets composed of similar types of foods and the same average amounts of energy, protein, zinc, phytate, and calcium as consumed by the children who were studied previously in the same community. The wheat products included in the breakfast and lunch meals were fortified with iron and the assigned level of zinc.
The children enrolled in the absorption studies were stratified by sex and then randomly assigned to 1 of the 3 treatment groups. Children in group Zn-0 received biscuits for breakfast and noodles for lunch, both of which were prepared from a total of 100 g wheat flour fortified only with 30 mg Fe as ferrous sulfate per kg flour. Children in group Zn-3 received the same iron-fortified wheat products, but the products were also fortified with 30 mg Zn as zinc sulfate per kg flour. Children in group Zn-9 received the same iron-fortified wheat products, but the products were also fortified with 90 mg Zn as zinc sulfate per kg flour. The wheat products were prepared at the Institute of Agro-Industrial Development of the National Agrarian University, La Molina, under the supervision of the study personnel. Samples of each of the fortified foods were analyzed periodically by atomic absorption spectrophotometry to confirm appropriate levels of fortification.
Absorption studies
As many as 4 children were enrolled in the absorption studies at one time. Before breakfast on the morning of study day 1, the fasting children were transported from their homes to the study unit, which was located in a building near the study community. The children's height and weight were recorded, and a 7-mL venous blood sample was collected to measure baseline hemoglobin and plasma zinc and ferritin concentrations. On each of the 2 following days (study days 2 and 3), the children were taken again to the study unit, where they received the assigned study diets for breakfast, lunch, and dinner between 0700 and 0800, 1200 and 1300, and 1700 and 1800, respectively. The composition of each of the meals provided during the first 2 d of each absorption study is shown in Table 2. All foods were individually weighed before serving, and any leftovers were weighed to calculate actual intakes. On both days, small sips of aqueous solutions containing oral zinc tracer doses of 0.15 mg 70Zn/meal were given along with both breakfast and lunch, beginning mid-way through the meals; and 0.3 mg of the 68Zn tracer was provided with each of the dinners to measure zinc absorption separately from the latter meals, which did not contain any fortified foods. The rationale for labeling meals for 2 d was to limit the total amount of tracer added to the meals each day. A total of 0.6 mg 70Zn and 0.6 mg 68Zn was provided over the 2 d. A single 1 mg dose of the 67Zn tracer was administered intravenously on day 3, a few minutes after the children had finished lunch. This time was selected on the basis of simulation studies utilizing Miller's compartmental model of zinc metabolism (20).
For the remainder of the metabolic period (study days 4–10), a nursing aide stayed with each of the children at their homes for 12 h each day to record their dietary intakes by direct weighing, as described above, and to assist with specimen collection. During these days, the same study meals (without isotopic tracers) were delivered to the children's homes. Beginning on study day 5, 2 separate midstream urine samples were collected each day for the next 6 d (ie, on study days 5–10). A 24-h urine collection was also performed on study day 6. Urine samples were stored in zinc-free, acid-washed 50-mL plastic tubes at –20°C until analysis.
After completing the first set of metabolic studies, the children continued to receive the same study meals for breakfast and lunch, which were delivered to their homes each day during the following 6 wk. They consumed their usual home diets in the evenings, and a total of 7 24-h dietary recall histories were collected, approximately once weekly (generally on weekdays), to estimate their total daily iron and zinc intakes during this period between the 2 sets of absorption studies. Energy, zinc, and iron intakes were expressed both as the total amounts consumed per day and as the amounts consumed divided by the children's body weights.
A second round of absorption studies was performed during study days 51–60, using the same procedures as described for the first round of studies.
Isotope preparation
Accurately weighed quantities of zinc oxide enriched with 67Zn, 68Zn or 70Zn (Trace Sciences International, Richmond Hill, Ontario, Canada) were dissolved in 0.5 M H2SO4 to prepare stock solutions. For preparation of the orally administered zinc isotopes, 68Zn and 70Zn, the stock solutions were diluted with triply deionized water and titrated to pH 5.0 with metal-free ammonium hydroxide. For preparation of the intravenously administered zinc isotope, 67Zn, the pH of the stock solution was adjusted to 6.0 with ammonium hydroxide and diluted with sterile isotonic sodium chloride to a zinc concentration of 1.5 mmol/L. Oral and intravenous solutions were filtered through a 0.2 μm filter, and the zinc concentrations of these solutions were measured by atomic absorption spectrophotometry with mass correction factors applied (21). Accurately weighed quantities were stored in plastic tubes (oral doses) or sealed sterile vials (intravenous doses). The intravenous doses were tested for pyrogens immediately before being stored.
Isotope analyses and other biochemical assays
Urinary zinc isotopic ratios were analyzed by using a modification of the method of Veillon et al (22). Urine samples were transferred to Pyrex beakers (Corning Inc, Reston, VA), dried at 100°C, and ashed for 24 h at 450°C in a muffle furnace. A few drops of concentrated nitric acid (Fisher Scientific, Pittsburgh) were added to the samples, which were then redried on a hot plate and ashed in the muffle furnace for a second 24-h period. Each sample was then dissolved in 2.5 mL ammonium acetate buffer (pH 5.6), and zinc was extracted into hexane after the addition of pyridine (Fisher Scientific). The solvents were then evaporated, and any remaining organic material was digested by adding concentrated nitric acid and 30% (by vol) hydrogen peroxide and heating at 100°C on a heating block. Finally, zinc was dissolved in 4 mL 2% (by vol) nitric acid distilled at subboiling temperature (Optima; Fischer Scientific) for subsequent isotopic ratio analysis. Glass tubes used in the extraction were washed in aqua regia and rinsed in Milli-Q triply deionized water (Millipore, Billerica, MA) before use.
For the analyses of zinc isotopic ratios (67Zn/66Zn, 68Zn/66Zn, and 70Zn/66Zn), 8-mL aliquots of the samples were prepared in 2% (by vol) nitric acid to achieve a final concentration of 50 ppb. The resulting solutions were introduced into an inductively coupled plasma mass spectrometer (PlasmaQuad 3; VG Elemental, Cheshire, United Kingdom), using an auto-sampler (ASX-500, Model 510, CETAC, Omaha) and peristaltic pump (Perimax 12, CPETEC, Erding, Germany). Instrument parameters were: argon gas flow: 13L/m; intermediate gas flow: 1.4 L/m; nebulizer gas flow: 0.82 L/m; forward power: 1350W; temperature of pneumatic nebulizer: 4°C; sample flow rate: 1 mL/min. Data acquisition parameters were set as follows: dwell times were 3 milliseconds for 66Zn, 4 milliseconds for 67Zn, 3 milliseconds for 68Zn and 5 milliseconds for 70Zn, one point per peak, 1800 sweeps, 10 replicates and 50 ns dead time. A natural zinc standard was run every 6 samples and 2% HNO3 every 12 samples. Counts per second of nitric acid were subtracted from counts per second of all samples. The precision of this method is <0.3% RSD for 67Zn/66Zn and 68Zn/66Zn and <0.6% RSD for 70Zn/66Zn. All isotope analyses were done at the Pediatric Nutrition Laboratory at the University of Colorado Heath Sciences Center. A full description of the analytic method has been published (23).
Isotopic enrichments were calculated from the measured isotopic ratios. Calculations of isotopic enrichment data accounted for the presence of the other isotopes. For each particular isotopic label used, enrichment was defined as all zinc in the sample from an isotopically enriched source divided by the total amount of zinc in the sample. FAZ was determined from the ratio of urinary enrichment with the oral doses to urinary enrichment with the intravenous dose (24). Meal-specific estimates of the quantities of zinc absorbed were calculated by multiplying the FAZs by the amounts of zinc in the respective meal(s). FAZb,l and AZb,l refer to absorption of zinc from breakfast and lunch, whereas FAZd and AZd refer to absorption of zinc from dinner and AZb,l.d refers to AZ from all meals.
The blood samples drawn by finger prick during screening were analyzed for hemoglobin concentration with a portable hemoglobin photometer (Hemocue; AB Leo Diagnostics, Helsingborg, Sweden). The blood samples collected during the metabolic periods were analyzed for hemoglobin concentrations using a specific, colorimetric assay kit (Wiener Labs; Rosario, Argentina). Plasma ferritin concentrations were determined by immunoradiometry (Diagnostics Products Corporation; Los Angeles). Plasma zinc concentrations were determined by inductively coupled plasma mass spectrometer after wet ashing (25).
Sample size estimate and statistical analyses
A sample size of 20 children per group was estimated to be sufficient to detect a difference in FAZ of 0.10 between groups, assuming a baseline mean FAZ of 0.10 based on previous unpublished studies in Peruvian infants, and an SD of ±0.09 based on previous studies in Chinese women (26) and allowing for 20% attrition.
A two-factor analysis of variance (ANOVA) was used to compare the 3 groups at baseline (SPSS for WINDOWS, version 10.0; SPSS Inc, Chicago). Repeated-measures ANOVA, with zinc group as the between-subject variable, time (days 1–2 or days 51–52) as the within-subject variable, and a group-by-time interaction, was used to compare differences in dietary zinc intake, FAZ, and AZ by study groups and metabolic period. Analysis of covariance (ANCOVA) was used to compare differences in the mean change in plasma zinc concentration, with control for initial values. All comparisons were done at the 5% level of significance.
RESULTS
Characteristics of study subjects
Of the 58 children originally assigned to one of the treatment groups, results of the absorption studies were available from 41 children at the initial time point and from 31 children at both time points; these 2 sets of information form the basis of the current report. The reasons for the missing results were as follows: 1) we were unable to administer the tracers successfully to 17 children during the first round of absorption studies (because of the failure to place the intravenous line or spillage of the oral tracer), 2) we lost 8 children to follow-up between the first and second rounds of absorption tests, and 3) we failed to administer the tracers to 2 children in the second round. There were no significant differences in the rates of incomplete studies by treatment group. The children with complete absorption data at both time points differed from those without complete data only with regard to sex distribution (P = 0.019) and initial hemoglobin concentration (P = 0.011), but neither of these variables was associated with the main outcomes of the absorption studies. Thus, the results of the absorption studies completed in the subset of children with longitudinal measurements can be considered to be generally representative of the whole group of high-risk children in the study community.
The baseline characteristics of the 41 children included in at least the first set of absorption studies are presented in Table 3. The treatment groups did not differ with regard to age, percentage males, initial mean body weight and height, and initial mean hemoglobin and plasma zinc and ferritin concentrations. Children in group Zn-9 had significantly lower height-for-age z scores than did those in group Zn-0, but there were no group-related differences in weight-for-height or weight-for-age. As per the selection criteria, all children were anemic at baseline (hemoglobin concentration <110 g/L). Eight (20.5%) of the 39 children with acceptable specimens had low fasting plasma zinc concentrations (<65 μg/dL) at baseline.
Dietary zinc intake at baseline and between absorption studies
The mean baseline (preintervention) usual dietary intakes of energy, protein, iron, zinc, and phytate of the children included in the absorption studies are shown by study group in Table 4. The mean baseline energy intakes of the study subjects were 6% less than those of the children included in the background studies, presumably because the children enrolled in the absorption studies were selected on the basis of their low height-for-age. The study subjects' mean total zinc intakes of 5 mg/d were similar to the amounts observed during the background studies, as were the mean phytate:zinc molar ratios (13).
The mean dietary intakes and body weights during the period between the absorption studies are also presented in Table 4 for the 31 subjects who participated in both rounds of absorption studies. Their body weights increased significantly during the period of study, but there were no significant changes in their mean total daily energy intakes, expressed as either kcal/d or kcal · kg body wt–1 · d–1. Iron intakes increased significantly during the interim period because the children were provided with iron-fortified wheat products, and zinc intakes increased in accordance with the level of zinc fortification. There were no changes in phytate intake from baseline, so the phytate:zinc molar ratios varied in relation to the amount of zinc intake.
Zinc intake and zinc absorption during tracer studies
For both sets of absorption studies, the same oral isotopic tracer was used to measure zinc absorption from the breakfast and lunch meals, both of which contained iron- or iron- and zinc-fortified wheat products. Thus, combined results are presented for zinc intake and absorption from these 2 meals. Dinner meals did not contain any fortified foods, and a different oral isotopic tracer was used to measure zinc absorption, so results for the dinners are presented separately, as are the combined results for AZ for all meals.
Initial studies
During the first 2 d of the first round of absorption studies, the children consumed 80 ± 8% of the energy offered at breakfast and lunch; the mean amounts of energy consumed did not differ significantly by treatment group (Table 5). As expected, zinc intakes from breakfast and lunch (and total daily zinc intakes from all meals combined) were directly related to the level of zinc fortification. Although the mean FAZb,l (from breakfast and lunch combined) was inversely related to zinc intake from these meals (P < 0.001), the mean AZb,l was progressively greater in children who received the higher intakes (P < 0.001). The mean energy and zinc intakes from the dinner meal did not differ significantly by study group, but the mean FAZ and AZ values from dinner (FAZd and AZd) were significantly lower in the group that received the highest level of zinc fortification at breakfast and lunch than in the group that did not receive any zinc-fortified foods (ANCOVA: P = 0.015 for FAZd and P = 0.012 for AZd). Despite the lower FAZs in children who received the zinc-fortified foods, AZs from all meals combined were greater in those who received the zinc-fortified foods (P = 0.01).
Interestingly, a curvilinear model provided a better fit to the data relating the individual means for FAZb,l to mean zinc intake from breakfast and lunch than did a linear model (departure from linearity: P = 0.006; Figure 2). In other words, the decline in FAZb,l per unit intake was more pronounced at the lower than at the higher end of the range of zinc intake. Notably, inspection of the relation between zinc intake and AZb,l suggests that there were diminishing returns as the level of fortification increased (Figure 2), although the quadratic term in this model was not statistically significant (P = 0.42), perhaps because of one outlying value, as shown in the figure. If this value is eliminated from consideration, the quadratic term in this model is marginally significant as well (P = 0.078).
Longitudinal studies
Longitudinal data for dietary intake and zinc absorption from the subset of 31 children with complete data at both time points are presented in Table 6. Because the children gained weight between the initial and final absorption studies, the major variables for zinc intake and absorption are presented both before and after being divided by body weight. Final mean zinc intakes at breakfast and lunch, expressed as mg/d, did not change significantly from those observed during the initial studies; however, the mean zinc intakes decreased slightly in relation to body weight in all treatment groups (P = 0.029). The mean FAZs from the zinc-fortified (breakfast and lunch) meals did not change significantly during the period of observation, and the inverse relation between mean zinc intake and FAZb,l that was observed at baseline remained significant in the final studies. The mean AZb,l, expressed as mg/d, decreased slightly, but not significantly, from baseline; although these changes from the initial values were significant when expressed in relation to body weight (P = 0.007).
Total zinc intakes from dinner increased significantly from the initial to the final studies, although these differences disappeared when the intakes were expressed in relation to body weight. Mean FAZd and AZd values decreased significantly compared with the initial results, possibly because of the increased zinc intakes, although the group-wise differences that were observed at baseline were no longer statistically significant during the second round of studies.
The mean AZs from all meals combined tended to decrease slightly between the 2 rounds of studies (P = 0.054); these changes were significant when expressed in relation to body weight (P < 0.001). Nevertheless, the mean AZb,l,d was still positively related to the level of fortification and, hence, to the total amount of zinc consumed. Although the relative group-wise differences in AZb,l,d were still present when the data were expressed in relation to body weight, these latter results were not statistically significant, possibly because of the greater variability in these data.
Change in weight, height, and biochemical indicators of zinc and iron status
The children in all groups combined gained 1.28 ± 1.29 kg and 2.94 ± 1.43 cm during the course of the study and increased their height-for-age, weight-for-height, and weight-for-age z scores, although the magnitude of these changes did not differ by study group (Table 3).
The mean plasma zinc concentrations did not change significantly in any of the treatment groups during the course of the study, although the percentage of children with plasma zinc concentrations <65 μg/dL decreased significantly from 20.5% to 3.3% (P = 0.046; Table 3). All children were anemic initially (hemoglobin <110 g/L). Mean hemoglobin concentrations increased significantly during the study in all treatment groups, and there were no significant group-wise differences in hemoglobin concentration at the end of the study period, when only 8.8% still had low hemoglobin concentrations. There were no significant changes in geometric mean plasma ferritin concentrations from the initial to the final measurements.
DISCUSSION
The results of this study indicate that, despite the lower FAZ that occurs when young children's dietary zinc intakes are increased by fortifying a common staple food, total AZ increases as more zinc is consumed. The mechanism for the effect of increased zinc intake on FAZ is not certain; this may represent simple saturation kinetics of the zinc transporters that are responsible for the uptake of zinc by the enterocyte (15) or down-regulation of the number of transporters or specific mucosal receptors (27). In the case of the latter explanation, however, the specific transporters or receptors that were affected would have had to respond to the dietary changes within 24–48 h, because the differences in absorption were already present during the first set of tracer studies, which were completed within 1-2 d of introduction of the study diets. Notably, the relation between zinc intake and AZ suggests that with increasingly higher zinc intakes there is a diminishing return on total zinc absorption, as was described previously (13).
In general, the relations between zinc intake and zinc absorption remained unchanged during the 7-wk period of continued consumption of zinc-fortified foods, although there is some suggestion that the efficiency of zinc absorption may begin to decline after continued exposure to zinc-fortified foods. In particular, during the interval from the initial to the final absorption studies, there was a reduction of 9% in FAZ from the fortified breakfast and lunch meals (from 0.23 to 0.21, all groups combined), which was not statistically significant (P = 0.12), and there was a significant decline of 15% in FAZ from the nonfortified dinner meals (from 0.45 to 0.38; P < 0.001). It is uncertain, however, whether the small observed decreases in FAZ were due to the cumulative effect of increased AZb,l,d on the zinc status of the children who received the zinc-fortified foods or because the children's zinc intakes increased slightly as they increased in age and body size. Notably, there was no interaction between treatment group and time, which suggests that the time-related changes in FAZ may not have been due to zinc fortification per se, although the small sample size limits the statistical power to assess this interaction.
A novel finding of the current study was the fact that those children who consumed more zinc from the zinc-fortified breakfast and lunch meals absorbed less zinc from the nonfortified dinners during the initial absorption studies. These results are consistent with recent observations in one adult volunteer whose FAZ at breakfast was reduced on the morning after consumption of a single 20-mg supplement of zinc (KM Hambidge, unpublished observations, 2004). Because the lunch and dinner meals in the present study were separated by 4–5 h, it is unlikely that very much of the zinc that was consumed during the earlier meals would have remained in the small intestinal lumen at the time that the dinners were consumed. However, it is conceivable that the zinc transporters or receptors involved in zinc uptake may have remained saturated for a longer period of time by zinc consumed at earlier meals.
The fact that the intervention had only a small effect on serum zinc concentration was unexpected because earlier studies in young children have consistently found a moderately large response to zinc supplementation with doses ranging from 3 to 20 mg/d (1). This difference between the serum responses to zinc fortification and to zinc supplementation may be due to the lower efficiency of zinc absorption that occurs when zinc is delivered with food than when an aqueous supplement is provided between meals (15). Alternatively, it is possible that the duration of the study was too short for plasma zinc concentrations to be affected or that the sample size was inadequate to detect such changes. A sample size of 32 per group would have been necessary to detect a statistically significant change of 0.8 SD, which was the magnitude of change observed in the meta-analysis of earlier supplementation trials (1).
It is notable that the children's hemoglobin concentrations increased in response to the dietary treatments with iron-fortified foods, even though these foods delivered a net increase of only 1 mg Fe/d. Although the increase in mean hemoglobin concentration tended to be less in the groups that received higher levels of zinc fortification, these differences were not statistically significant, possibly because of the relatively small sample size. Nevertheless, only 3 children were still mildly anemic at the end of the study, despite the concomitant provision of zinc in the fortified foods. Thus, the levels of zinc fortification provided in this study would not be likely to affect hematologic status adversely if provided continuously over longer periods of time along with iron fortification.
It is worth reemphasizing that the current studies were conducted in children who were selected because of their presumptive high risk of current or prior zinc deficiency. Specifically, the children had a low height-for-age, which may be related to poor zinc status (1), anemia, and low plasma ferritin concentrations, which suggests that their dietary intake of bioavailable iron (and possibly, therefore, of bioavailable zinc) was less than adequate. The children's mean baseline dietary zinc intakes were 5 mg/d, and 5 of the 41 children (12%) had usual zinc intakes that were less than their age-specific estimated average requirements (3 mg/d for children aged 1–3 y and 4 mg/d for children aged 4–8 y; 13). Moreover, the moderately high phytate:zinc ratio of the diet might have limited zinc absorption to some extent (23). However, only 20% of the children had low plasma zinc concentrations at baseline, which suggests that the population from which they were drawn was only marginally at risk for zinc deficiency (9). Thus, it is not certain whether these children were truly zinc deficient at the time of the absorption studies or whether this might have affected the study outcomes.
One of the underlying objectives of this study was to determine for this population the appropriate level of zinc fortification, which depends on the children's usual dietary zinc intake, the amount of the fortified food vehicle that is consumed, and the total zinc absorption from the mixed diet and fortified foods in relation to the age-specific requirements for AZ. Several expert groups have developed estimates of the physiologic requirements for AZ. For example, the recent publication on dietary reference intakes (DRIs) for North America (13) suggests that children 1–3 y of age need to absorb 0.74 mg Zn/d and children 4–8 y of age need to absorb 1.20 mg Zn/d. By contrast, the International Zinc Nutrition Consultative Group (IZiNCG) concluded that children in these respective age ranges need to absorb 0.53 and 0.83 mg Zn/d (9). Notably, in the present study, 10–30% of the children who did not receive any zinc-fortified foods had total daily AZs less than the DRI estimates of adequate absorption (depending on whether round 1 or round 2 is considered) and 10% had AZs less than the IZiNCG estimates. By contrast, 11–22% of the children in group Zn-3 and none of those in group Zn-9 had total AZs less than the DRI estimates, and none of the children in either of these groups absorbed less than the IZiNCG estimates. Thus, for children in this population, fortification at a level slightly higher than 30 mg Zn/kg wheat flour would be desirable.
Among the current group of 3–4-y-old children included in the absorption studies, wheat intakes averaged 150 g/d when wheat products were provided by the study. However, other children in the study community consumed only 70 g wheat/d when no additional wheat products were provided by the project team. Thus, if the consumption of wheat products is assumed to remain at the baseline (preintervention) level, it might be necessary to fortify these products at even higher levels to achieve the same AZs observed among the study subjects. Alternatively, a zinc-fortification program might be linked with other efforts to promote increased consumption of the fortified food products. It is important to recognize that the highest level of zinc fortification (90 mg/kg wheat) used in the current study might be excessive for a universal fortification program because any adults consuming as much as 400 g wheat/d, which is within the range of conceivable intakes, would receive 36 mg Zn. This amount of fortificant zinc in addition to zinc provided by other dietary sources would be likely to exceed the currently proposed safe upper limit of 40 mg/d for adults (9, 13). Thus, the highest level of fortification used in the current project would be appropriate only for wheat products specifically targeted to young children.
In summary, we found that despite the reduction in FAZ that occurs when zinc intake is increased by fortifying a common staple food, fortification allows for greater total absorption of zinc from the diet. Moreover, the increased level of zinc absorption is sustained during 2 mo of continued consumption of the fortified foods. In this population, fortification at a level 30 mg Zn/kg wheat would be desirable to ensure that zinc absorption exceeds the physiologic requirements of almost all children in the age range that was studied. Because of the apparent diminishing returns for zinc absorption as the level of zinc fortification is increased further, it is unlikely that there would be any practical benefit of increasing levels of fortification beyond 70–80 mg/kg wheat flour. Importantly, almost all of the previously anemic children enrolled in the study recovered from anemia, so zinc fortification did not prevent utilization of the iron in the doubly fortified foods.
ACKNOWLEDGMENTS
We appreciate the contributions of Sian Lei, who completed the stable-isotope analyses, and Jamie Westcott and Leland Miller, who assisted with the quality control and analysis of the isotope data.
DLdR, KMH, NFK, and KHB were responsible for the conceptualization and the design of the study. DLdR and MEP implemented the clinical procedures and supervised the data collection. MS completed the laboratory analyses. JMP assisted with the data analysis and statistical modeling procedures. DLdR and KHB drafted the manuscript, which was reviewed by all coauthors. None of the authors had any financial conflicts of interest.
REFERENCES
Brown KH, Peerson JM, Rivera J, Allen LH. Effect of supplemental zinc on the growth and serum zinc concentrations of prepubertal children: a meta-analysis of randomized controlled trials. Am J Clin Nutr 2002;75:1062–71.
Fraker PJ, King LE. Reprogramming of the immune system during zinc deficiency. Annu Rev Nutr 2004;24:277–98.
Bhutta ZA, Black RE, Brown KH, et al. Prevention of diarrhea and pneumonia by zinc supplementation in children in developing countries: pooled analysis of randomized controlled trials. Zinc Investigators' Collaborative Group. J Pediatr 1999;135:689–97.
Bhutta ZA, Bird SM, Black RE, et al. Therapeutic effects of oral zinc in acute and persistent diarrhea in children in developing countries: pooled analysis of randomized controlled trials. Am J Clin Nutr 2000;72:1516–22.
Sazawal S, Black RE, Menon VP, et al. Zinc supplementation in infants born small for gestational age reduces mortality: a prospective, randomized, controlled trial. Pediatrics 2001;108:1280–6.
Caulfield LE, Zavaleta N, Shankar AH, Merialdi M. Potential contribution of maternal zinc supplementation during pregnancy to maternal and child survival. Am J Clin Nutr 1998;68(suppl):499S–508S.
Black MM. Zinc deficiency and child development. Am J Clin Nutr 1998;68(suppl):464S–9S.
Gibson RS. Zinc nutrition in developing countries. Nutr Res Rev 1994;7:151–73.
International Zinc Nutrition Consultative Group. Assessment of the risk of zinc deficiency in populations and options for its control. Food Nutr Bull 2004;25(suppl):S94–204. (IZiNCG technical document 1.)
Lotfi M, Venkatesh Mannar MG, Merx RJHM, Naber-van den Heuvel P. Micronutrient fortification of foods. Current practices, research, and opportunities. Ottawa, Canada: The Micronutrient Initiative, 1996.
Jackson MJ, Jones DA, Edwards RH, Swainbank IG, Coleman ML. Zinc homeostasis in man: studies using a new stable isotope-dilution technique. Br J Nutr 1984;51:199–208.
Lee DY, Prasad AS, Hydrick-Adair C, Brewer G, Johnson PE. Homeostasis of zinc in marginal human zinc deficiency: role of absorption and endogenous excretion of zinc. J Lab Clin Med 1993;122:549–56.
Food and Nutrition Board, Institute of Medicine. Dietary reference intakes of vitamin A, vitamin K, arsenic, boron, chromium, copper, iodine, iron, manganese, molybdenum, nickel, silicon, vanadium, and zinc. Washington, DC: National Academy Press, 2001.
Jalla S, Westcott J, Steirn M, Miller LV, Bell M, Krebs NF. Zinc absorption and exchangeable zinc pool sizes in breastfed infants fed meat or cereal as first complementary food. J Pediatr Gastroenterol Nutr 2002;34:35–41.
Lnnerdal B. Dietary factors influencing zinc absorption. J Nutr 2000;130(suppl):1378S–83S.
Instituto de Investigacion Nutricional. Tabla de composición de alimentos Peruanos. (Peruvian food-composition table.) Lima, Peru: Instituto de Investigacion Nutricional, 1996.
WorldFood Dietary Assessment System. Internet: http://www.fao.org/infoods/software_worldfood_en.stm (accessed 8 October 2004).
Beaton GH, Milner J, Corey P, et al. Sources of variance in 24 hour recall data: implications for nutrition study design and interpretation. Am J Clin Nutr 1979;32:2546–9.
Ogden CL, Kuczmarski RJ, Flegal KM, et al. Centers for Disease Control and Prevention 2000 growth charts for the United States: improvements to the 1977 National Center for Health Statistics version. Pediatrics 2002;109:45–60.
Miller LV, Krebs NF, Hambidge KM. Development of a compartmental model of human zinc metabolism: idetifiability and multiple studies analyses. Am J Physiol Regul Integr Comp Physiol 2000;279:R1671–84.
Peirce PL, Hambidge KM, Goss CH, Miller LV, Fennessey PV. Fast atom bombardment mass spectrometry for the determination of zinc stable isotopes in biological samples. Anal Chem 1987;59:2034–37.
Veillon C, Patterson KY, Moser-Veillon PB. Digestion and extraction of biological materials for zinc stable isotope determination by inductively coupled plasma mass spectrometry. J Anal Atomic Spectrom 1996;11:727–30.
Hambidge KM, Huffer JW, Raboy V, et al. Zinc absorption from low-phytate hybrids of maize and their wild-type isohybrids. Am J Clin Nutr 2004;79:1053–9.
Friel JK, Naake VL Jr, Miller LV, Fennessey PV, Hambidge KM. The analysis of stable isotopes in urine to determine the fractional absorption of zinc. Am J Clin Nutr 1992;55:473–7.
Clegg MS, Keen CL, Lonnerdal B, Hurley LS. Influence of ashing techniques on the analysis of trace elements in animal tissue. I. Wet ashing. Biol Trace Elem Res 1981;3:107–15.
Sian L, Mingyan X, Miller LV, Tong L, Krebs NF, Hambidge KM. Zinc absorption and intestinal losses of endogenous zinc in young Chinese women with marginal zinc intakes. Am J Clin Nutr 1996;63:348–53.
Cousins RJ, McMahon RJ. Integrative aspects of zinc transporters. J Nutr 2000;130(suppl):1384S–7S., http://www.100md.com(Daniel López de Romaa, Ma)
the Instituto de Investigación Nutricional, Lima, Perú (DLdR, MS, and MEP)
the Section of Nutrition, Department of Pediatrics, University of Colorado, Denver (KMH and NFK).
ABSTRACT
Background:Information is needed on the fractional absorption of zinc (FAZ) and absorbed zinc (AZ) during prolonged exposure to zinc-fortified foods.
Objective:The objective was to measure FAZ and AZ from diets fortified with different amounts of zinc and to determine whether zinc absorption changes over 7 wk.
Design:Forty-one stunted, moderately anemic children received daily, at breakfast and lunch, 100 g wheat products fortified with 3 mg Fe (ferrous sulfate) and 0 (group Zn-0), 3 (group Zn-3), or 9 (group Zn-9) mg Zn (zinc sulfate) per 100 g flour. FAZ was measured on days 2–3 and 51–52; meal-specific AZs were calculated as the product of FAZ and zinc intake.
Results:For the breakfast and lunch meals combined, mean total zinc intakes were 2.14, 4.72, and 10.04 mg/d in groups Zn-0, Zn-3, and Zn-9, respectively, during the initial absorption studies; mean (±SD) FAZ values were 0.341 ± 0.111, 0.237 ± 0.052, and 0.133 ± 0.041, respectively, on days 2–3 (P < 0.001) and did not change significantly on days 51–52 in the subset of 31 children studied twice. Mean initial AZ was positively related to zinc intake (0.71 ± 0.18, 1.11 ± 0.21, and 1.34 ± 0.47 mg/d, respectively; P < 0.001); final values did not differ significantly from the initial values.
Conclusions:AZ from meals containing zinc-fortified wheat products increases in young children relative to the level of fortification and changes only slightly during 7-wk periods of consumption. Although consumption of zinc-fortified foods may reduce FAZ, zinc fortification at the levels studied positively affects total daily zinc absorption, even after nearly 2 mo of exposure to zinc-fortified diets.
Key Words: Zinc iron wheat fortification absorption stable isotopes children
INTRODUCTION
Zinc deficiency is associated with poor growth (1), depressed immune function (2), increased susceptibility to and severity of infections (3-5), adverse outcomes of pregnancy (6), and neurobehavior abnormalities (7). In many lower-income countries, the diets are composed primarily of cereals and legumes, which contain substantial amounts of phytate (myo-inositol hexaphosphate), a compound known to inhibit zinc absorption. These diets contain few animal-source foods (muscle meats or organs), which are rich sources of zinc and are free of phytate (8). Recent analyses of national food balance data suggest that 20% of the world's population is at risk of inadequate zinc intake (9). Moreover, approximately one-third of children in low-income countries have a low height-for-age relative to international reference data. Zinc supplementation increases linear growth in stunted children (1), which suggests that these high rates of stunting may be due, in part, to zinc deficiency. Thus, zinc deficiency appears to be widespread in lower-income countries, and intervention programs are needed to improve zinc status in these high-risk populations.
Food fortification is considered to be one of the most cost-effective strategies for improving micronutrient status (10). Cereal products, such as wheat flour, are currently being fortified with iron in many developing countries, and simultaneous fortification with zinc could be achieved with a small incremental cost (9). However, to design rational zinc-fortification programs, additional information is needed, such as 1) the amount of zinc that is absorbed from foods fortified with different amounts of zinc, 2) the possible influence of zinc status on zinc absorption, 3) the effect of zinc fortification on zinc absorption from nonfortified foods or meals, and 4) the effect of zinc fortification on the absorption of other minerals from the diet.
Whole-body zinc homeostasis is achieved primarily by means of the balance between the absorption of dietary zinc and the fecal excretion of endogenous zinc (11, 12). Although it is known that the amount of zinc in a meal affects zinc absorption from that meal (13, 14), quantitative data from human studies are limited, especially in children. Also, there is little information on the effects of zinc intake in one meal on zinc absorption from other meals consumed during the day. Furthermore, although it has been postulated that differences in zinc intake that lead to changes in zinc status may affect the longer-term efficiency of zinc absorption (15), there is limited information on the relation between an individual's zinc status and intestinal absorption of zinc. Thus, the objectives of the present study were to 1) measure zinc absorption from meals containing zinc-fortified products and from other meals in relation to the level of zinc fortification of iron-fortified wheat flour, and 2) determine whether zinc absorption changes after a more prolonged exposure to zinc-fortified foods.
SUBJECTS AND METHODS
Study design
Fifty-eight children 3–4 y of age, who were residing in a poor community on the periphery of Lima, Peru and who were considered to be at high risk of zinc deficiency, were randomly assigned to receive in 2 meals/d a total of 100 g wheat products fortified with 3 mg Fe as ferrous sulfate and either 0 (group Zn-0), 3 (group Zn-3) or 9 (group Zn-9) mg Zn as zinc sulfate per 100 g flour for 70 d. Fractional absorption of zinc (FAZ) was assessed by measuring urinary ratios of different stable isotopic tracers administered orally and intravenously on study days 2–3 and 51–52 for the 2 sets of absorption studies, and AZ was calculated as the meal-specific FAZ multiplied by the zinc intake from the respective meal or meals.
Assessment of usual zinc intakes in the study community
To develop locally appropriate diets for the absorption studies, we completed preliminary observations of the dietary intakes of 50 children aged 3–4 y in the study community. For these background dietary studies, food intakes were measured quantitatively during 12-h daytime observations in the children's homes by subtracting the amounts of any leftovers from the amounts served. Any additional foods consumed during the preceding 12 h were elicited from the children's parents. The food intakes were converted to nutrient intakes by using data from a Peruvian food-composition table (16) and additional information on the phytate contents of these foods (17). These background dietary studies indicated that the children in the study area had slightly low intakes of zinc and iron and moderately high phytate:zinc ratios (Table 1). The major sources of zinc were rice, chicken, milk, wheat bread, potato, oatmeal, wheat noodles, and beans. The diets used for the subsequent absorption studies were designed to mimic the characteristics of the usual home diets, except that additional wheat products were substituted for other cereals to permit consumption of the desired amounts of zinc from the fortified wheat products.
Twenty-five of the children underwent a second round of dietary assessments to determine intraindividual variations in zinc intake and to estimate the number of days of observation that would be required to characterize an individual's usual zinc intake, as described by Beaton et al (18). The results of these latter studies indicated that a total of 7 24-h dietary recalls would provide estimates of individual children's usual zinc intake within ±25% with 95% confidence. Thus, we attempted to schedule 7 sets of 24-h dietary recall histories of the absorption study subjects both before they initiated the first round of the tracer studies and between the 2 rounds of studies.
Absorption study subjects
A total of 765 children aged 3–4 y were screened to identify children who were deemed to have a high risk of zinc deficiency because of the presence of stunting (defined as height-for-age z score <–2.0 SD with respect to international reference data; 19) and moderate anemia (hemoglobin <110 and >90 g/L). Anemia was used as a screening factor for possible zinc deficiency because iron deficiency is the most common cause of anemia in this population and many of the same dietary factors that affect iron status would be expected to influence zinc status similarly. Whereas hemoglobin could be measured with a portable device in the field, information on plasma zinc concentration was not available immediately.
Fifty-eight children identified as having a high risk of zinc deficiency were enrolled in the subsequent absorption studies to examine the effects of different levels of zinc fortification of iron-fortified wheat products on zinc absorption (Figure 1). The absorption studies were repeated after a period of 7 wk during which the children received the same fortified foods each day to determine whether their efficiency of zinc absorption changed during this period of time. In this article, we present the results from 41 of these children for whom absorption data were collected successfully at the initial time point and from the subset of 31 children for whom absorption data were obtained at both time points. The reasons that absorption data were not available from the remaining children are described below and summarized in Figure 1.
Parents of all participating children provided written informed consent at the beginning of the study. The study protocol and consent forms were approved by the institutional review boards of the University of California, Davis, and the Instituto de Investigación Nutricional in Lima.
Study diets
With reference to the results of the background dietary studies, a 7-d rotating menu was designed to provide the study subjects with basal diets composed of similar types of foods and the same average amounts of energy, protein, zinc, phytate, and calcium as consumed by the children who were studied previously in the same community. The wheat products included in the breakfast and lunch meals were fortified with iron and the assigned level of zinc.
The children enrolled in the absorption studies were stratified by sex and then randomly assigned to 1 of the 3 treatment groups. Children in group Zn-0 received biscuits for breakfast and noodles for lunch, both of which were prepared from a total of 100 g wheat flour fortified only with 30 mg Fe as ferrous sulfate per kg flour. Children in group Zn-3 received the same iron-fortified wheat products, but the products were also fortified with 30 mg Zn as zinc sulfate per kg flour. Children in group Zn-9 received the same iron-fortified wheat products, but the products were also fortified with 90 mg Zn as zinc sulfate per kg flour. The wheat products were prepared at the Institute of Agro-Industrial Development of the National Agrarian University, La Molina, under the supervision of the study personnel. Samples of each of the fortified foods were analyzed periodically by atomic absorption spectrophotometry to confirm appropriate levels of fortification.
Absorption studies
As many as 4 children were enrolled in the absorption studies at one time. Before breakfast on the morning of study day 1, the fasting children were transported from their homes to the study unit, which was located in a building near the study community. The children's height and weight were recorded, and a 7-mL venous blood sample was collected to measure baseline hemoglobin and plasma zinc and ferritin concentrations. On each of the 2 following days (study days 2 and 3), the children were taken again to the study unit, where they received the assigned study diets for breakfast, lunch, and dinner between 0700 and 0800, 1200 and 1300, and 1700 and 1800, respectively. The composition of each of the meals provided during the first 2 d of each absorption study is shown in Table 2. All foods were individually weighed before serving, and any leftovers were weighed to calculate actual intakes. On both days, small sips of aqueous solutions containing oral zinc tracer doses of 0.15 mg 70Zn/meal were given along with both breakfast and lunch, beginning mid-way through the meals; and 0.3 mg of the 68Zn tracer was provided with each of the dinners to measure zinc absorption separately from the latter meals, which did not contain any fortified foods. The rationale for labeling meals for 2 d was to limit the total amount of tracer added to the meals each day. A total of 0.6 mg 70Zn and 0.6 mg 68Zn was provided over the 2 d. A single 1 mg dose of the 67Zn tracer was administered intravenously on day 3, a few minutes after the children had finished lunch. This time was selected on the basis of simulation studies utilizing Miller's compartmental model of zinc metabolism (20).
For the remainder of the metabolic period (study days 4–10), a nursing aide stayed with each of the children at their homes for 12 h each day to record their dietary intakes by direct weighing, as described above, and to assist with specimen collection. During these days, the same study meals (without isotopic tracers) were delivered to the children's homes. Beginning on study day 5, 2 separate midstream urine samples were collected each day for the next 6 d (ie, on study days 5–10). A 24-h urine collection was also performed on study day 6. Urine samples were stored in zinc-free, acid-washed 50-mL plastic tubes at –20°C until analysis.
After completing the first set of metabolic studies, the children continued to receive the same study meals for breakfast and lunch, which were delivered to their homes each day during the following 6 wk. They consumed their usual home diets in the evenings, and a total of 7 24-h dietary recall histories were collected, approximately once weekly (generally on weekdays), to estimate their total daily iron and zinc intakes during this period between the 2 sets of absorption studies. Energy, zinc, and iron intakes were expressed both as the total amounts consumed per day and as the amounts consumed divided by the children's body weights.
A second round of absorption studies was performed during study days 51–60, using the same procedures as described for the first round of studies.
Isotope preparation
Accurately weighed quantities of zinc oxide enriched with 67Zn, 68Zn or 70Zn (Trace Sciences International, Richmond Hill, Ontario, Canada) were dissolved in 0.5 M H2SO4 to prepare stock solutions. For preparation of the orally administered zinc isotopes, 68Zn and 70Zn, the stock solutions were diluted with triply deionized water and titrated to pH 5.0 with metal-free ammonium hydroxide. For preparation of the intravenously administered zinc isotope, 67Zn, the pH of the stock solution was adjusted to 6.0 with ammonium hydroxide and diluted with sterile isotonic sodium chloride to a zinc concentration of 1.5 mmol/L. Oral and intravenous solutions were filtered through a 0.2 μm filter, and the zinc concentrations of these solutions were measured by atomic absorption spectrophotometry with mass correction factors applied (21). Accurately weighed quantities were stored in plastic tubes (oral doses) or sealed sterile vials (intravenous doses). The intravenous doses were tested for pyrogens immediately before being stored.
Isotope analyses and other biochemical assays
Urinary zinc isotopic ratios were analyzed by using a modification of the method of Veillon et al (22). Urine samples were transferred to Pyrex beakers (Corning Inc, Reston, VA), dried at 100°C, and ashed for 24 h at 450°C in a muffle furnace. A few drops of concentrated nitric acid (Fisher Scientific, Pittsburgh) were added to the samples, which were then redried on a hot plate and ashed in the muffle furnace for a second 24-h period. Each sample was then dissolved in 2.5 mL ammonium acetate buffer (pH 5.6), and zinc was extracted into hexane after the addition of pyridine (Fisher Scientific). The solvents were then evaporated, and any remaining organic material was digested by adding concentrated nitric acid and 30% (by vol) hydrogen peroxide and heating at 100°C on a heating block. Finally, zinc was dissolved in 4 mL 2% (by vol) nitric acid distilled at subboiling temperature (Optima; Fischer Scientific) for subsequent isotopic ratio analysis. Glass tubes used in the extraction were washed in aqua regia and rinsed in Milli-Q triply deionized water (Millipore, Billerica, MA) before use.
For the analyses of zinc isotopic ratios (67Zn/66Zn, 68Zn/66Zn, and 70Zn/66Zn), 8-mL aliquots of the samples were prepared in 2% (by vol) nitric acid to achieve a final concentration of 50 ppb. The resulting solutions were introduced into an inductively coupled plasma mass spectrometer (PlasmaQuad 3; VG Elemental, Cheshire, United Kingdom), using an auto-sampler (ASX-500, Model 510, CETAC, Omaha) and peristaltic pump (Perimax 12, CPETEC, Erding, Germany). Instrument parameters were: argon gas flow: 13L/m; intermediate gas flow: 1.4 L/m; nebulizer gas flow: 0.82 L/m; forward power: 1350W; temperature of pneumatic nebulizer: 4°C; sample flow rate: 1 mL/min. Data acquisition parameters were set as follows: dwell times were 3 milliseconds for 66Zn, 4 milliseconds for 67Zn, 3 milliseconds for 68Zn and 5 milliseconds for 70Zn, one point per peak, 1800 sweeps, 10 replicates and 50 ns dead time. A natural zinc standard was run every 6 samples and 2% HNO3 every 12 samples. Counts per second of nitric acid were subtracted from counts per second of all samples. The precision of this method is <0.3% RSD for 67Zn/66Zn and 68Zn/66Zn and <0.6% RSD for 70Zn/66Zn. All isotope analyses were done at the Pediatric Nutrition Laboratory at the University of Colorado Heath Sciences Center. A full description of the analytic method has been published (23).
Isotopic enrichments were calculated from the measured isotopic ratios. Calculations of isotopic enrichment data accounted for the presence of the other isotopes. For each particular isotopic label used, enrichment was defined as all zinc in the sample from an isotopically enriched source divided by the total amount of zinc in the sample. FAZ was determined from the ratio of urinary enrichment with the oral doses to urinary enrichment with the intravenous dose (24). Meal-specific estimates of the quantities of zinc absorbed were calculated by multiplying the FAZs by the amounts of zinc in the respective meal(s). FAZb,l and AZb,l refer to absorption of zinc from breakfast and lunch, whereas FAZd and AZd refer to absorption of zinc from dinner and AZb,l.d refers to AZ from all meals.
The blood samples drawn by finger prick during screening were analyzed for hemoglobin concentration with a portable hemoglobin photometer (Hemocue; AB Leo Diagnostics, Helsingborg, Sweden). The blood samples collected during the metabolic periods were analyzed for hemoglobin concentrations using a specific, colorimetric assay kit (Wiener Labs; Rosario, Argentina). Plasma ferritin concentrations were determined by immunoradiometry (Diagnostics Products Corporation; Los Angeles). Plasma zinc concentrations were determined by inductively coupled plasma mass spectrometer after wet ashing (25).
Sample size estimate and statistical analyses
A sample size of 20 children per group was estimated to be sufficient to detect a difference in FAZ of 0.10 between groups, assuming a baseline mean FAZ of 0.10 based on previous unpublished studies in Peruvian infants, and an SD of ±0.09 based on previous studies in Chinese women (26) and allowing for 20% attrition.
A two-factor analysis of variance (ANOVA) was used to compare the 3 groups at baseline (SPSS for WINDOWS, version 10.0; SPSS Inc, Chicago). Repeated-measures ANOVA, with zinc group as the between-subject variable, time (days 1–2 or days 51–52) as the within-subject variable, and a group-by-time interaction, was used to compare differences in dietary zinc intake, FAZ, and AZ by study groups and metabolic period. Analysis of covariance (ANCOVA) was used to compare differences in the mean change in plasma zinc concentration, with control for initial values. All comparisons were done at the 5% level of significance.
RESULTS
Characteristics of study subjects
Of the 58 children originally assigned to one of the treatment groups, results of the absorption studies were available from 41 children at the initial time point and from 31 children at both time points; these 2 sets of information form the basis of the current report. The reasons for the missing results were as follows: 1) we were unable to administer the tracers successfully to 17 children during the first round of absorption studies (because of the failure to place the intravenous line or spillage of the oral tracer), 2) we lost 8 children to follow-up between the first and second rounds of absorption tests, and 3) we failed to administer the tracers to 2 children in the second round. There were no significant differences in the rates of incomplete studies by treatment group. The children with complete absorption data at both time points differed from those without complete data only with regard to sex distribution (P = 0.019) and initial hemoglobin concentration (P = 0.011), but neither of these variables was associated with the main outcomes of the absorption studies. Thus, the results of the absorption studies completed in the subset of children with longitudinal measurements can be considered to be generally representative of the whole group of high-risk children in the study community.
The baseline characteristics of the 41 children included in at least the first set of absorption studies are presented in Table 3. The treatment groups did not differ with regard to age, percentage males, initial mean body weight and height, and initial mean hemoglobin and plasma zinc and ferritin concentrations. Children in group Zn-9 had significantly lower height-for-age z scores than did those in group Zn-0, but there were no group-related differences in weight-for-height or weight-for-age. As per the selection criteria, all children were anemic at baseline (hemoglobin concentration <110 g/L). Eight (20.5%) of the 39 children with acceptable specimens had low fasting plasma zinc concentrations (<65 μg/dL) at baseline.
Dietary zinc intake at baseline and between absorption studies
The mean baseline (preintervention) usual dietary intakes of energy, protein, iron, zinc, and phytate of the children included in the absorption studies are shown by study group in Table 4. The mean baseline energy intakes of the study subjects were 6% less than those of the children included in the background studies, presumably because the children enrolled in the absorption studies were selected on the basis of their low height-for-age. The study subjects' mean total zinc intakes of 5 mg/d were similar to the amounts observed during the background studies, as were the mean phytate:zinc molar ratios (13).
The mean dietary intakes and body weights during the period between the absorption studies are also presented in Table 4 for the 31 subjects who participated in both rounds of absorption studies. Their body weights increased significantly during the period of study, but there were no significant changes in their mean total daily energy intakes, expressed as either kcal/d or kcal · kg body wt–1 · d–1. Iron intakes increased significantly during the interim period because the children were provided with iron-fortified wheat products, and zinc intakes increased in accordance with the level of zinc fortification. There were no changes in phytate intake from baseline, so the phytate:zinc molar ratios varied in relation to the amount of zinc intake.
Zinc intake and zinc absorption during tracer studies
For both sets of absorption studies, the same oral isotopic tracer was used to measure zinc absorption from the breakfast and lunch meals, both of which contained iron- or iron- and zinc-fortified wheat products. Thus, combined results are presented for zinc intake and absorption from these 2 meals. Dinner meals did not contain any fortified foods, and a different oral isotopic tracer was used to measure zinc absorption, so results for the dinners are presented separately, as are the combined results for AZ for all meals.
Initial studies
During the first 2 d of the first round of absorption studies, the children consumed 80 ± 8% of the energy offered at breakfast and lunch; the mean amounts of energy consumed did not differ significantly by treatment group (Table 5). As expected, zinc intakes from breakfast and lunch (and total daily zinc intakes from all meals combined) were directly related to the level of zinc fortification. Although the mean FAZb,l (from breakfast and lunch combined) was inversely related to zinc intake from these meals (P < 0.001), the mean AZb,l was progressively greater in children who received the higher intakes (P < 0.001). The mean energy and zinc intakes from the dinner meal did not differ significantly by study group, but the mean FAZ and AZ values from dinner (FAZd and AZd) were significantly lower in the group that received the highest level of zinc fortification at breakfast and lunch than in the group that did not receive any zinc-fortified foods (ANCOVA: P = 0.015 for FAZd and P = 0.012 for AZd). Despite the lower FAZs in children who received the zinc-fortified foods, AZs from all meals combined were greater in those who received the zinc-fortified foods (P = 0.01).
Interestingly, a curvilinear model provided a better fit to the data relating the individual means for FAZb,l to mean zinc intake from breakfast and lunch than did a linear model (departure from linearity: P = 0.006; Figure 2). In other words, the decline in FAZb,l per unit intake was more pronounced at the lower than at the higher end of the range of zinc intake. Notably, inspection of the relation between zinc intake and AZb,l suggests that there were diminishing returns as the level of fortification increased (Figure 2), although the quadratic term in this model was not statistically significant (P = 0.42), perhaps because of one outlying value, as shown in the figure. If this value is eliminated from consideration, the quadratic term in this model is marginally significant as well (P = 0.078).
Longitudinal studies
Longitudinal data for dietary intake and zinc absorption from the subset of 31 children with complete data at both time points are presented in Table 6. Because the children gained weight between the initial and final absorption studies, the major variables for zinc intake and absorption are presented both before and after being divided by body weight. Final mean zinc intakes at breakfast and lunch, expressed as mg/d, did not change significantly from those observed during the initial studies; however, the mean zinc intakes decreased slightly in relation to body weight in all treatment groups (P = 0.029). The mean FAZs from the zinc-fortified (breakfast and lunch) meals did not change significantly during the period of observation, and the inverse relation between mean zinc intake and FAZb,l that was observed at baseline remained significant in the final studies. The mean AZb,l, expressed as mg/d, decreased slightly, but not significantly, from baseline; although these changes from the initial values were significant when expressed in relation to body weight (P = 0.007).
Total zinc intakes from dinner increased significantly from the initial to the final studies, although these differences disappeared when the intakes were expressed in relation to body weight. Mean FAZd and AZd values decreased significantly compared with the initial results, possibly because of the increased zinc intakes, although the group-wise differences that were observed at baseline were no longer statistically significant during the second round of studies.
The mean AZs from all meals combined tended to decrease slightly between the 2 rounds of studies (P = 0.054); these changes were significant when expressed in relation to body weight (P < 0.001). Nevertheless, the mean AZb,l,d was still positively related to the level of fortification and, hence, to the total amount of zinc consumed. Although the relative group-wise differences in AZb,l,d were still present when the data were expressed in relation to body weight, these latter results were not statistically significant, possibly because of the greater variability in these data.
Change in weight, height, and biochemical indicators of zinc and iron status
The children in all groups combined gained 1.28 ± 1.29 kg and 2.94 ± 1.43 cm during the course of the study and increased their height-for-age, weight-for-height, and weight-for-age z scores, although the magnitude of these changes did not differ by study group (Table 3).
The mean plasma zinc concentrations did not change significantly in any of the treatment groups during the course of the study, although the percentage of children with plasma zinc concentrations <65 μg/dL decreased significantly from 20.5% to 3.3% (P = 0.046; Table 3). All children were anemic initially (hemoglobin <110 g/L). Mean hemoglobin concentrations increased significantly during the study in all treatment groups, and there were no significant group-wise differences in hemoglobin concentration at the end of the study period, when only 8.8% still had low hemoglobin concentrations. There were no significant changes in geometric mean plasma ferritin concentrations from the initial to the final measurements.
DISCUSSION
The results of this study indicate that, despite the lower FAZ that occurs when young children's dietary zinc intakes are increased by fortifying a common staple food, total AZ increases as more zinc is consumed. The mechanism for the effect of increased zinc intake on FAZ is not certain; this may represent simple saturation kinetics of the zinc transporters that are responsible for the uptake of zinc by the enterocyte (15) or down-regulation of the number of transporters or specific mucosal receptors (27). In the case of the latter explanation, however, the specific transporters or receptors that were affected would have had to respond to the dietary changes within 24–48 h, because the differences in absorption were already present during the first set of tracer studies, which were completed within 1-2 d of introduction of the study diets. Notably, the relation between zinc intake and AZ suggests that with increasingly higher zinc intakes there is a diminishing return on total zinc absorption, as was described previously (13).
In general, the relations between zinc intake and zinc absorption remained unchanged during the 7-wk period of continued consumption of zinc-fortified foods, although there is some suggestion that the efficiency of zinc absorption may begin to decline after continued exposure to zinc-fortified foods. In particular, during the interval from the initial to the final absorption studies, there was a reduction of 9% in FAZ from the fortified breakfast and lunch meals (from 0.23 to 0.21, all groups combined), which was not statistically significant (P = 0.12), and there was a significant decline of 15% in FAZ from the nonfortified dinner meals (from 0.45 to 0.38; P < 0.001). It is uncertain, however, whether the small observed decreases in FAZ were due to the cumulative effect of increased AZb,l,d on the zinc status of the children who received the zinc-fortified foods or because the children's zinc intakes increased slightly as they increased in age and body size. Notably, there was no interaction between treatment group and time, which suggests that the time-related changes in FAZ may not have been due to zinc fortification per se, although the small sample size limits the statistical power to assess this interaction.
A novel finding of the current study was the fact that those children who consumed more zinc from the zinc-fortified breakfast and lunch meals absorbed less zinc from the nonfortified dinners during the initial absorption studies. These results are consistent with recent observations in one adult volunteer whose FAZ at breakfast was reduced on the morning after consumption of a single 20-mg supplement of zinc (KM Hambidge, unpublished observations, 2004). Because the lunch and dinner meals in the present study were separated by 4–5 h, it is unlikely that very much of the zinc that was consumed during the earlier meals would have remained in the small intestinal lumen at the time that the dinners were consumed. However, it is conceivable that the zinc transporters or receptors involved in zinc uptake may have remained saturated for a longer period of time by zinc consumed at earlier meals.
The fact that the intervention had only a small effect on serum zinc concentration was unexpected because earlier studies in young children have consistently found a moderately large response to zinc supplementation with doses ranging from 3 to 20 mg/d (1). This difference between the serum responses to zinc fortification and to zinc supplementation may be due to the lower efficiency of zinc absorption that occurs when zinc is delivered with food than when an aqueous supplement is provided between meals (15). Alternatively, it is possible that the duration of the study was too short for plasma zinc concentrations to be affected or that the sample size was inadequate to detect such changes. A sample size of 32 per group would have been necessary to detect a statistically significant change of 0.8 SD, which was the magnitude of change observed in the meta-analysis of earlier supplementation trials (1).
It is notable that the children's hemoglobin concentrations increased in response to the dietary treatments with iron-fortified foods, even though these foods delivered a net increase of only 1 mg Fe/d. Although the increase in mean hemoglobin concentration tended to be less in the groups that received higher levels of zinc fortification, these differences were not statistically significant, possibly because of the relatively small sample size. Nevertheless, only 3 children were still mildly anemic at the end of the study, despite the concomitant provision of zinc in the fortified foods. Thus, the levels of zinc fortification provided in this study would not be likely to affect hematologic status adversely if provided continuously over longer periods of time along with iron fortification.
It is worth reemphasizing that the current studies were conducted in children who were selected because of their presumptive high risk of current or prior zinc deficiency. Specifically, the children had a low height-for-age, which may be related to poor zinc status (1), anemia, and low plasma ferritin concentrations, which suggests that their dietary intake of bioavailable iron (and possibly, therefore, of bioavailable zinc) was less than adequate. The children's mean baseline dietary zinc intakes were 5 mg/d, and 5 of the 41 children (12%) had usual zinc intakes that were less than their age-specific estimated average requirements (3 mg/d for children aged 1–3 y and 4 mg/d for children aged 4–8 y; 13). Moreover, the moderately high phytate:zinc ratio of the diet might have limited zinc absorption to some extent (23). However, only 20% of the children had low plasma zinc concentrations at baseline, which suggests that the population from which they were drawn was only marginally at risk for zinc deficiency (9). Thus, it is not certain whether these children were truly zinc deficient at the time of the absorption studies or whether this might have affected the study outcomes.
One of the underlying objectives of this study was to determine for this population the appropriate level of zinc fortification, which depends on the children's usual dietary zinc intake, the amount of the fortified food vehicle that is consumed, and the total zinc absorption from the mixed diet and fortified foods in relation to the age-specific requirements for AZ. Several expert groups have developed estimates of the physiologic requirements for AZ. For example, the recent publication on dietary reference intakes (DRIs) for North America (13) suggests that children 1–3 y of age need to absorb 0.74 mg Zn/d and children 4–8 y of age need to absorb 1.20 mg Zn/d. By contrast, the International Zinc Nutrition Consultative Group (IZiNCG) concluded that children in these respective age ranges need to absorb 0.53 and 0.83 mg Zn/d (9). Notably, in the present study, 10–30% of the children who did not receive any zinc-fortified foods had total daily AZs less than the DRI estimates of adequate absorption (depending on whether round 1 or round 2 is considered) and 10% had AZs less than the IZiNCG estimates. By contrast, 11–22% of the children in group Zn-3 and none of those in group Zn-9 had total AZs less than the DRI estimates, and none of the children in either of these groups absorbed less than the IZiNCG estimates. Thus, for children in this population, fortification at a level slightly higher than 30 mg Zn/kg wheat flour would be desirable.
Among the current group of 3–4-y-old children included in the absorption studies, wheat intakes averaged 150 g/d when wheat products were provided by the study. However, other children in the study community consumed only 70 g wheat/d when no additional wheat products were provided by the project team. Thus, if the consumption of wheat products is assumed to remain at the baseline (preintervention) level, it might be necessary to fortify these products at even higher levels to achieve the same AZs observed among the study subjects. Alternatively, a zinc-fortification program might be linked with other efforts to promote increased consumption of the fortified food products. It is important to recognize that the highest level of zinc fortification (90 mg/kg wheat) used in the current study might be excessive for a universal fortification program because any adults consuming as much as 400 g wheat/d, which is within the range of conceivable intakes, would receive 36 mg Zn. This amount of fortificant zinc in addition to zinc provided by other dietary sources would be likely to exceed the currently proposed safe upper limit of 40 mg/d for adults (9, 13). Thus, the highest level of fortification used in the current project would be appropriate only for wheat products specifically targeted to young children.
In summary, we found that despite the reduction in FAZ that occurs when zinc intake is increased by fortifying a common staple food, fortification allows for greater total absorption of zinc from the diet. Moreover, the increased level of zinc absorption is sustained during 2 mo of continued consumption of the fortified foods. In this population, fortification at a level 30 mg Zn/kg wheat would be desirable to ensure that zinc absorption exceeds the physiologic requirements of almost all children in the age range that was studied. Because of the apparent diminishing returns for zinc absorption as the level of zinc fortification is increased further, it is unlikely that there would be any practical benefit of increasing levels of fortification beyond 70–80 mg/kg wheat flour. Importantly, almost all of the previously anemic children enrolled in the study recovered from anemia, so zinc fortification did not prevent utilization of the iron in the doubly fortified foods.
ACKNOWLEDGMENTS
We appreciate the contributions of Sian Lei, who completed the stable-isotope analyses, and Jamie Westcott and Leland Miller, who assisted with the quality control and analysis of the isotope data.
DLdR, KMH, NFK, and KHB were responsible for the conceptualization and the design of the study. DLdR and MEP implemented the clinical procedures and supervised the data collection. MS completed the laboratory analyses. JMP assisted with the data analysis and statistical modeling procedures. DLdR and KHB drafted the manuscript, which was reviewed by all coauthors. None of the authors had any financial conflicts of interest.
REFERENCES
Brown KH, Peerson JM, Rivera J, Allen LH. Effect of supplemental zinc on the growth and serum zinc concentrations of prepubertal children: a meta-analysis of randomized controlled trials. Am J Clin Nutr 2002;75:1062–71.
Fraker PJ, King LE. Reprogramming of the immune system during zinc deficiency. Annu Rev Nutr 2004;24:277–98.
Bhutta ZA, Black RE, Brown KH, et al. Prevention of diarrhea and pneumonia by zinc supplementation in children in developing countries: pooled analysis of randomized controlled trials. Zinc Investigators' Collaborative Group. J Pediatr 1999;135:689–97.
Bhutta ZA, Bird SM, Black RE, et al. Therapeutic effects of oral zinc in acute and persistent diarrhea in children in developing countries: pooled analysis of randomized controlled trials. Am J Clin Nutr 2000;72:1516–22.
Sazawal S, Black RE, Menon VP, et al. Zinc supplementation in infants born small for gestational age reduces mortality: a prospective, randomized, controlled trial. Pediatrics 2001;108:1280–6.
Caulfield LE, Zavaleta N, Shankar AH, Merialdi M. Potential contribution of maternal zinc supplementation during pregnancy to maternal and child survival. Am J Clin Nutr 1998;68(suppl):499S–508S.
Black MM. Zinc deficiency and child development. Am J Clin Nutr 1998;68(suppl):464S–9S.
Gibson RS. Zinc nutrition in developing countries. Nutr Res Rev 1994;7:151–73.
International Zinc Nutrition Consultative Group. Assessment of the risk of zinc deficiency in populations and options for its control. Food Nutr Bull 2004;25(suppl):S94–204. (IZiNCG technical document 1.)
Lotfi M, Venkatesh Mannar MG, Merx RJHM, Naber-van den Heuvel P. Micronutrient fortification of foods. Current practices, research, and opportunities. Ottawa, Canada: The Micronutrient Initiative, 1996.
Jackson MJ, Jones DA, Edwards RH, Swainbank IG, Coleman ML. Zinc homeostasis in man: studies using a new stable isotope-dilution technique. Br J Nutr 1984;51:199–208.
Lee DY, Prasad AS, Hydrick-Adair C, Brewer G, Johnson PE. Homeostasis of zinc in marginal human zinc deficiency: role of absorption and endogenous excretion of zinc. J Lab Clin Med 1993;122:549–56.
Food and Nutrition Board, Institute of Medicine. Dietary reference intakes of vitamin A, vitamin K, arsenic, boron, chromium, copper, iodine, iron, manganese, molybdenum, nickel, silicon, vanadium, and zinc. Washington, DC: National Academy Press, 2001.
Jalla S, Westcott J, Steirn M, Miller LV, Bell M, Krebs NF. Zinc absorption and exchangeable zinc pool sizes in breastfed infants fed meat or cereal as first complementary food. J Pediatr Gastroenterol Nutr 2002;34:35–41.
Lnnerdal B. Dietary factors influencing zinc absorption. J Nutr 2000;130(suppl):1378S–83S.
Instituto de Investigacion Nutricional. Tabla de composición de alimentos Peruanos. (Peruvian food-composition table.) Lima, Peru: Instituto de Investigacion Nutricional, 1996.
WorldFood Dietary Assessment System. Internet: http://www.fao.org/infoods/software_worldfood_en.stm (accessed 8 October 2004).
Beaton GH, Milner J, Corey P, et al. Sources of variance in 24 hour recall data: implications for nutrition study design and interpretation. Am J Clin Nutr 1979;32:2546–9.
Ogden CL, Kuczmarski RJ, Flegal KM, et al. Centers for Disease Control and Prevention 2000 growth charts for the United States: improvements to the 1977 National Center for Health Statistics version. Pediatrics 2002;109:45–60.
Miller LV, Krebs NF, Hambidge KM. Development of a compartmental model of human zinc metabolism: idetifiability and multiple studies analyses. Am J Physiol Regul Integr Comp Physiol 2000;279:R1671–84.
Peirce PL, Hambidge KM, Goss CH, Miller LV, Fennessey PV. Fast atom bombardment mass spectrometry for the determination of zinc stable isotopes in biological samples. Anal Chem 1987;59:2034–37.
Veillon C, Patterson KY, Moser-Veillon PB. Digestion and extraction of biological materials for zinc stable isotope determination by inductively coupled plasma mass spectrometry. J Anal Atomic Spectrom 1996;11:727–30.
Hambidge KM, Huffer JW, Raboy V, et al. Zinc absorption from low-phytate hybrids of maize and their wild-type isohybrids. Am J Clin Nutr 2004;79:1053–9.
Friel JK, Naake VL Jr, Miller LV, Fennessey PV, Hambidge KM. The analysis of stable isotopes in urine to determine the fractional absorption of zinc. Am J Clin Nutr 1992;55:473–7.
Clegg MS, Keen CL, Lonnerdal B, Hurley LS. Influence of ashing techniques on the analysis of trace elements in animal tissue. I. Wet ashing. Biol Trace Elem Res 1981;3:107–15.
Sian L, Mingyan X, Miller LV, Tong L, Krebs NF, Hambidge KM. Zinc absorption and intestinal losses of endogenous zinc in young Chinese women with marginal zinc intakes. Am J Clin Nutr 1996;63:348–53.
Cousins RJ, McMahon RJ. Integrative aspects of zinc transporters. J Nutr 2000;130(suppl):1384S–7S., http://www.100md.com(Daniel López de Romaa, Ma)