| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research Paper |
Mineral Bioavailability Laboratory, Jean Mayer USDA Human Nutrition Research Center on Aging, Tufts University, Boston, Massachusetts
Address correspondence to: Miguel Navarro, Ph.D., Department of Nutrition and Food Science, School of Pharmacy, University of Granada, 18071-Granada, SPAIN. E-Mail: nalarcon{at}platon.ugr.es
ABSTRACT
Objective: To estimate the micronutrient (riboflavin, folate, vitamin C, vitamin B12, iron, zinc and copper) bioavailability in healthy adults from a multi-micronutrient dietary supplement to assess the possible influence on it by the tablet disintegration properties and by the relative intestinal permeability of subject.
Methods: The bioavailability of seven micronutrients from a single brand of multi-micronutrient dietary supplement was measured on two separate occasions in the presence of a standardized test meal in 15 healthy adult subjects. Each subject visited the Metabolic Research Unit on four separate randomized occasions for an absorption test. One test measured the intestinal permeability. The other three tests measured the postprandial changes in plasma or serum concentrations after consuming a test meal alone (control:placebo effect), or the test meal with either whole or crushed and powdered dietary supplements. 15 healthy Caucasian adult volunteers, aged 42 ± 14 years.
Results: The 12 hour-post-dose AUC for riboflavin, folate and vitamin C (whole and crushed tablet), and that for vitamin B12 (only for the crushed tablet treatment) and iron (only for the whole tablet treatment) were all significantly (p < 0.001) higher than after a test meal alone. In contrast there was no significant increase in the AUC after supplement intake for zinc and copper. Neither the form of the supplement for all micronutrients tested nor intestinal permeability of the subject for riboflavin, folate, vitamin C, iron, zinc and copper influenced the postdose nutrient AUC. In contrast, for vitamin B12 the intestinal permeability of the subject influenced significantly the nutrient AUC (p = 0.003).
Conclusion: Tablet disintegration characteristics of this dietary supplement did not limit absorption of these seven micronutrients. The intestinal permeability of subject was only positively correlated with the B12 bioavailability. Results are suggestive of using multi-micronutrients dietary supplements as a vehicle to decrease the prevalence of multiple micronutrient deficiencies overall for vitamins in healthy adults.
Key words: micronutrient bioavailability, supplements, intestinal permeability, disintegration rate
INTRODUCTION
Approximately 40% of the American population regularly consumes micronutrient supplements, and a significant portion of these take a combination vitamin and mineral product [1,2]. In the U.S. many people use multivitamin and mineral supplements that can provide one or more times the Recommended Dietary Intakes (RDI) for many of the individual nutrients, despite doubts about the need for micronutrient supplementation for most of the population, especially in healthy subjects consuming a balanced diet [3,4]. On the other hand, in developing countries multiple micronutrient deficiencies may be common, and there is a growing sentiment that the delivery of multinutrient supplements to these populations would be an efficacious public health measure [5]. However, there is no labeling requirement to provide information about the bioavailability of the individual micronutrients from supplements. A number of dietary and physiological factors can influence the efficiency of intestinal absorption of essential nutrients. In addition, bioavailability of micronutrients from dietary supplements can be influenced by the form of the nutrient, e.g., mineral salt chosen for the supplement, the possibility of various negative nutrient-nutrient interactions. In addition, pharmaceutical-related factors, such as the degree or rate of disintegration of a tablet are known to be important in determining the bioavailability of various medication [6–8]. However, little information is available concerning the effect of tablet characteristics on the bioavailability of micronutrient from dietary supplements. Moreover, the relationship between disintegration rate and bioavailability of specific nutrients can be complex. For example, some reports have suggested that a more rapid tablet disintegration rate may lead to higher [6,9], lower [7] or no effect on micronutrient absorption [10]. Thus, in the present study we tested the hypothesis that crushing and grinding the supplement into powdered form, which might be done by some consumers to be able to consume the relatively large multimicronutrient supplements, would alter micronutrient bioavailability.
A common characteristic of intestinal nutrient transport is that nutrients are transported by both saturable and nonsaturable pathways, the latter possibly representing paracellular (intercellular) pathway [11–13]. The paracellular transport pathway is concentration dependent and not down-regulated by high nutrient intakes, presumably because it represents a transport pathway that reflects the general permeability characteristics of the intestine for the given micronutrients. For example, the paracellular pathway is responsible for increasing amounts of calcium being absorbed as the oral dose of calcium increases, despite saturation of the vitamin D-regulated calcium transport pathway [14]. Therapeutically, the paracellular transport pathway can be utilized to deliver life-saving nutrients in the face of defective cellular transport or when conditions of hormonal deficit occur (e.g., vitamin D deficiency of dietary or congenital nature leading to the development of rickets). This pathway is also of significant importance in the delivery of most pharmaceutical preparations [15,16]. Little is known about what factors may determine individual differences in the diffusion of nutrients through the paracellular pathway. Intestinal permeability can vary considerably between healthy adults. For example, a recent study in young and older adults [17] found that interindividual differences in permeability, measured by a simple non-invasive lactulose/mannitol absorption test, were as much as fivefold. One of the hallmarks of intestinal nutrient absorption is the great degree of heterogeneity of absorption efficiency between individuals. However, to our knowledge, no studies have been conducted in humans to determine if relative intestinal permeability characteristics are related to micronutrient absorption. Thus, we hypothesized that subjects with high intestinal permeability would have relatively high rates of micronutrient absorption when high oral doses of micronutrients are delivered as can occur in persons consuming dietary supplements.
MATERIALS AND METHODS
Subjects
Fifteen healthy Caucasian adult volunteers, eleven women and four men, aged 42 ± 14 years, were recruited for the study. None had a significant medical problem known to affect nutrient metabolism, such as gastrointestinal disease, renal disease, diabetes or other endocrine diseases or mild anemia. All subjects completed a medical questionnaire and underwent a medical examination. They were asked to refrain from taking any supplemental vitamins and minerals for 14 days prior to the start of the study. Premenopausal subjects had a pregnancy test performed within seven days prior to admission day. The protocol was approved by the Human Investigation Review Board at Tufts University. All subjects were fully informed of the aims of the study and signed an informed written consent.
Study Design
The bioavailability of seven micronutrients (vitamin C, vitamin B12, folate, riboflavin, iron, zinc and copper) from a single brand of multivitamin and mineral supplement (from vitamin A to zinc, Centrum®, Lederle Consumer Health Division, Whitehall-Robins, Madison NJ) was measured on two separate occasions in the presence of a standardized test meal in 15 healthy adult subjects who fasted overnight. The test meal consisted of two pieces of bread, two pats of butter, 100 g applesauce, 30 g jam, and 122 g orange juice. No coffee or tea was allowed with the test meal. All subjects were studied under the supervision of trained research nursing staff at the Metabolic Research Unit of the Jean Mayer USDA Human Nutrition Research Center on Aging at Tufts University. Each subject visited the metabolic research unit on four separate occasions for an absorption test. One test measured intestinal permeability and was performed without a test meal. The other three tests measured the postprandial changes in plasma or serum micronutrient concentrations after consuming a test meal alone (control), or the test meal with either whole or crushed and powdered dietary supplement. In each of the latter two tests, the subjects were given a single oral bolus dose of a multiple micronutrients as four multivitamin and mineral supplement tablets from a leading selling over-the-counter brand. The micronutrient supplements provided 240 mg vitamin C, 6.8 mg riboflavin, 1.6 mg folate, 96 µg vitamin B12, 72 mg iron, 60 mg zinc and 8 mg copper. Several other nutrients were also provided in the supplement tablets, but their concentration was not measured in blood (Table 1). Subjects took four times the USRDA for each micronutrient (Table 1) in order to result in a measurable rise in the plasma nutrient concentrations. In these conditions micronutrient bioavailability can be related with the intestinal permeability of individuals and pharmaceutical form of the formulation.
|
Assessment of Intestinal Permeability
Permeability of the small intestine was evaluated by a noninvasive mannitol-lactulose absorption test and was based on the urinary excretion of orally administered mannitol and lactulose load [17]. After a 12 hour overnight fast, 10 g of lactulose and 5 g of mannitol were given orally in 108 mL of water. Urine was collected for six hours. Urinary lactulose and mannitol were determined by high performance liquid chromatography analysis, as described elsewhere [18], and is based on a modification of the method of Delaunty and Hollander [19] for estimating urinary lactulose and mannitol. Total urinary excretion was calculated for each of the sugars and the results were expressed as percent recovery of the ingested sugar probe. Relative intestinal permeability was calculated as the ratio of excreted lactulose to mannitol. This test has been used as a measure of intestinal integrity in healthy and diseased individuals and is not affected by age or changes in renal creatinine clearance [17]. The test is based on the principle of differential absorption pathways of these two sugars in the small intestine. Mannitol is absorbed transcellularly through aqueous pores in the cell membrane, whereas lactulose is absorbed paracellularly via extrusion zones at the tip of the villous and tight junctions [20,21]. Urine was collected in opaque plastic urine jugs containing thimerosal as an additive over six hours following the ingestion of the lactulose-mannitol solution. The intestinal permeability procedure was scheduled as convenient any time during the study.
Blood Biochemical Analysis
For the determination of riboflavin, a slightly modified method of that described by Floridi et al. [22] was used. Plasma was deproteinized by addition of a 6% (w/v) trichloracetic acid -20% (w/v) acetonitrile solution. After centrifugation (12,000 x g for 10 minutes at 0 -2° C), the supernatant was heated for 15 minutes at 60° C. By this procedure, flavin adenine dinucleotide (FAD) was hydrolyzed to yield flavin mononucleotide (FMN), and then the flavocoenzymes FAD and FMN can be analyzed by reversed-phase HPLC as FMN. This hydrolysis step has the advantage of avoiding artificially reduced concentrations of FAD [23]. For the vitamin C assay, the plasma sample was deproteinized with perchloric acid/EDTA immediately after collection [24] the clear supernatant was frozen at -70°C until further analysis. For vitamin C measurements, a 2,4-dinitrophenylhydrazine based method was used [25]. Plasma folate and vitamin B12 levels were measured by the Quantaphase II radioassay according to manufacturer’s instructions.
Mineral Analysis
Blood was collected in trace element-free containers, allowed to clot, and serum was harvested for subsequent determination of iron, zinc and copper by flame atomic absorption spectrophotometry (Perkin-Elmer model 5000 AA with automatic burner control, AS 50). Serum samples were diluted 1:5 with deionized water, and the analysis was performed against standards prepared in 5% glycerol to approximate the viscosity characteristics of the diluted samples [26,27]. In order to avoid any significant interferences by contaminating hemoglobin iron, serum for iron determination was pretreated with 20% (w/v) trichloracetic acid plus heating at 90°C for 15 minutes [28].
Statistical Analysis
Data were analyzed by repeated measures ANOVA and GLM factorial model of period (phase 1, 2, 3), treatment (placebo, whole tablet, crushed tablet) and subject using SPSS Version 7.5 software. Assessment of significant differences between treatments was made using Tukey’s HSD multiple comparison procedure. The 12 hour plasma vitamin or serum mineral AUC was calculated by the trapezoidal rule as a summary measure of vitamin or mineral bioavailability. A log transformation was applied to riboflavin and iron data, and a square root transformation was applied to folate and zinc data to achieve normality and equality of variances in these measure. The data for the other nutrients was analyzed in the natural scale. A PC-based software program [29] was initially used in planning the experiment to calculate sample size for comparison of two treatments with each person as their own control. For the sake of estimating sample size, we assumed that intra-subject variability in the bioavailability measure was about 25%, and a 25% mean treatment effect was expected. An alpha level of 0.05 and a power of 0.90 were used. This calculation led to a sample size estimate of approximately 13 subjects.
RESULTS
Vitamin Bioavailability from the Micronutrient Supplement
For illustrative purposes, Figs. 1 and 2 shows the mean plasma folate and iron concentrations, respectively, for all 15 subjects over 12 hours for the control test meal alone and following the whole and crushed supplement tablet treatments. Inspection of the plasma folate profile illustrates a flat plasma folate baseline following the test meal alone compared to the rapid postprandial rise in plasma folate during the first four hours following supplement administration in either whole or crushed form. Generally similar time profiles were evident for riboflavin and vitamin C following supplement administration (data not shown). Table 2 lists the micronutrient bioavailability parameter AUC calculated from plasma micronutrient concentration over 12 hours and the percent difference in the AUC compared to each subject’s AUC for that nutrient obtained after the test meal alone (100%). Following consumption of the supplement, there was a statistically significant increase in the AUC for folate, riboflavin and vitamin C. Vitamin B12 did not demonstrate a significant overall treatment effect (p = 0.069), but crushed tablet treatment was found to yield a significantly higher average nutrient AUC than placebo test. Compared to the postprandial control AUC, supplement administration increased the AUC by 8% to 98%, depending upon the specific vitamin examined. When these subjects were tested with the crushed tablet, the AUC for riboflavin, folate, vitamin C and vitamin B12 was essentially identical to that observed after the consumption of whole tablets, clearly suggesting that the tablet disintegration characteristics of this multimicronutrient supplement did not limit absorption of these four vitamins. Anyway, taking into account that standard procedure in pharmacokinetics is to subtract control values from each treatment of the two treatment groups (to obtain a baseline correction) and then calculate and (compare) the AUCs for the two treatment groups in order to estimate bioavailability, neither the existence of significant differences for the four vitamins was found (p > 0.05).
|
|
|
Relationship of Intestinal Permeability Measurements to Micronutrient Bioavailability
The mean fractional excretion of lactulose dose in the 15 healthy subjects studied was 0.225 ± 0.083 of the administered dose; the mean fractional excretion of the mannitol dose was 0.137 ± 0.034. Relative intestinal permeability, defined as the ratio of % lactulose dose excreted in the urine divided by the % mannitol dose excreted in the urine over six hours, was 1.7 ± 0.7 (data not shown). There was a threefold range in relative intestinal permeability (0.9 to 2.9). However, we observed no significant correlations (data not shown) between relative intestinal permeability and the rise in AUC for riboflavin, folate, vitamin C or minerals. Thus, intestinal permeability did not appear to influence micronutrient absorption. Contrarily, for vitamin B12 a significant correlation between the relative intestinal permeability and the AUC for serum vitamin B12 was found (p < 0.05, r = 0.439).
DISCUSSION
Despite the widespread popularity of multiple micronutrient dietary supplements, the bioavailability of the nutrients provided in these supplements is largely unknown. We believe the observations made in this study concerning blood vitamin and mineral concentrations following supplement administration are of interest and relevant to the question of nutrient bioavailability from multimicronutrient supplements. This notion is all the more important given recent suggestions to promote multiple vitamin and mineral supplementation of healthy older adults or populations in developing countries where multiple deficiencies in micronutrients are common, especially among women of reproductive age where these deficiencies can have important effects on pregnancy outcome and the health of the newborn [5].
For measuring the present quantity of micronutrients absorbed from commercially available dietary supplements, we chose an indirect measure (increment in the postdose plasma AUC) in an attempt to simultaneously estimate bioavailability (absorption) of several micronutrients from a popular over-the-counter dietary supplement. It needs to be emphasized that estimates of AUC as an index of bioavailability can be affected by a number of factors. For example, AUC measures can be altered by physiologic factors other than the rate of absorption of the micronutrient, such as the volume of distribution of the micronutrient in the extracellular space, the rate of transport of the micronutrient out of the measured (plasma) compartment, the possible rate of conversion of the delivered substrate into other vitamers or breakdown products, as well as the storage in intestinal cells or liver and a very slow release of some nutrients from these tissues. However, because each subject in this study served as his or her own control, these potential confounding factors are unlikely to systematically influence the interpretation of the AUC, and it is likely that the AUC is a fair representation of the vitamin and mineral absorption from the dietary supplement. Intestinal absorption can be affected by pharmaceutical factors, such as the tablet disintegration rate, that affects the rate of physical release or possibly solubility of substances delivered in tablet form by determining positional availability of the micronutrients throughout the GI tract as a function of time and thereby the extent of intestinal absorption. For example, slow release of the micronutrients from a tablet could result in suboptimal concentrations of the nutrient at intestinal locations in which absorption efficiency for that micronutrient is maximal, such as the upper small intestine. However, with the possible exception of iron, we found no remarkable differences in the absorption between the whole tablet and crushed tablet modes of administration for most nutrients studied. When we employed the standard procedure in pharmacokinetics, we found the same result without statistically significant influence of formulation type in bioavailability of all nutrients (p > 0.05), included the iron (p = 0.439). In addition, with the possible exception of vitamin B12, we did not find that estimates of relative intestinal permeability were significantly correlated to micronutrient bioavailability. Consequently, it may be hypothesized that the observed AUC for micronutrient transport (folate, riboflavin, vitamin C, iron, zinc and copper) did not significantly increase in subjects with higher intestinal permeability, probably due to the physiological capacity of the human organism to adapt to different nutrient intakes through several mechanisms of absorption available, which has been also noticed in previous human studies [21,30]. In the case of vitamin B12 a significant correlation between the relative intestinal permeability and the AUC for serum vitamin B12 was found (p < 0.005). However the relationship is very weak due to that the coefficient correlation (r = 0.439) indicates that only about 19% (i.e., r squared) of the AUC variability is explained by intestinal permeability.
A notable finding from our study was that there can be clear differences in the apparent bioavailability of studied vitamins compared to minerals delivered from the same multi-micronutrient supplement. One practical implication of this observation is that it will not be possible to simply measure the bioavailability of any single nutrient as a marker of general bioavailability of all micronutrients delivered in a multinutrient supplement. Additional comparisons between various supplements will be necessary in the future to determine whether some individual nutrients may act as reasonable and convenient markers of bioavailability of a class of nutrients, such as water-soluble vitamins or minerals. In our current study, with the possible exception of vitamin B12, when the supplement was taken as a whole tablet, the area-under-the-curve (absorption) of the water-soluble vitamins significantly increased following supplement administration. The significant increase in plasma AUC for all four vitamins tested suggests that this multinutrient supplement was a likely source of bioavailable riboflavin, folate, vitamin C and probably vitamin B12. Significant overall treatment effects were observed for folate, riboflavin and vitamin C, with average AUC of these nutrients from both crushed and whole tablet being significantly higher than that for the placebo (p < 0.001, Table 2). However, vitamin B12 did not demonstrate a significant overall treatment effect, but crushed tablet treatment was found to yield a significantly higher average AUC than placebo (Table 2). Possibly, these findings could be related with the non-saturation of the transcellular transport pathway or with the content of vitamins studied (vitamin C, folate and riboflavin) in the tested meal given and individually considered in the placebo test.
In contrast, the absorption overall of zinc and copper was questionable given the apparently minimal postdose AUC response. The design of the study did not allow us to determine whether the apparent poor bioavailability of these minerals was a reflection of the study (high dose micronutrient delivery), a unique characteristic of the tested product, or whether it is a general characteristic of this class of products.
The explanation of the low bioavailability of minerals from the dietary supplement may be quite complex. There are several possible reasons to explain a low mineral AUC response to supplement administration. One possibility could be that the absolute dose of mineral was too low to result in a measurable rise in plasma mineral concentrations. However, given that we intentionally designed the study to provide a large oral mineral load in our attempt to assess the relationship between intestinal permeability and presumably paracellular absorption mechanisms, this is unlikely. The oral dose of iron in this study was 72 mg. Previous reports by other investigators assessing iron bioavailability from prenatal supplements have observed significant changes in AUC following similar or lower iron doses [31–34]. Likewise, the oral dose of zinc (60 mg) and copper (8 mg) was substantial.
It is interesting to note that vitamin B12 was the only nutrient to demonstrate a significant period effect with average absorption of it decreasing from phase 1 (the beginning of the study) to phase 3. A plausible explanation for this observation may be that the vitamin B12 status of the subjects had improved with the time in the study (because of the supplements) and that they had consequently reduced the efficiency of B12 absorption. Another study performed in healthy older adults taking a multivitamin/mineral supplement showed a decrease of the prevalence of supobtimal plasma levels of vitamin B12 [35].
The AUC for the minerals could be affected by the disintegration rate of the tablet if a relatively slow disintegrating tablet released these minerals into the intestinal lumen distal to efficient sites of mineral absorption. However, our observation that there was clearly no difference in the AUC between crushed and whole tablets for iron, zinc and copper suggests that tablet disintegration in this case could not explain the relatively low AUC.
Various dietary factors are known to specifically inhibit the absorption of mineral and would not influence vitamin bioavailability. In this study the micronutrient supplement was provided in the context of a test meal. However, we think that a food effect on mineral bioavailability is unlikely given that the composition of the test meal was such that it did not provide significant levels of mineral-inhibiting substances, such as polyphenols or phytate [36]. Moreover, from a practical viewpoint, it is not uncommon for people to take their supplements along with a meal.
The mixture of minerals in the supplement (Table 1) may have resulted in important mineral-mineral interactions that limited overall mineral bioavailability of the trace elements [37]. Earlier studies investigating the bioavailability of iron from prenatal supplements demonstrated that the calcium and magnesium content of the supplement influences iron bioavailability [31]. The total dose of calcium and magnesium supplied by the supplement used in this study was not trivial. Consumption of four supplement tablets provided approximately 640 mg of calcium and 400 mg of magnesium. This level of these minerals has been reported to have negative effects on both iron [38–40] and zinc [41,42] bioavailability. Negative interactions between calcium and zinc [42] and zinc and copper [43,44] have also been reported. In the present study, in supplementation tests, we also found negative interactions between the 12 hour-postdose AUC for zinc and that for copper (p < 0.01, r = -0.491). This finding shows the influence that zinc exerts in the copper storage in intestinal cells by inducing the synthesis of the Cu-binding ligand metallothioneins in the mucosal cells [45,46]. Consequently a very slow release to circulation is produced.
An additional possible explanation of the apparently low mineral bioavailability from the supplement relates to the mineral salt forms utilized in micronutrient supplements (namely zinc and cupric oxides, and ferric fumarate). In this sense, because of the high element-to-salt mass ratio it is not uncommon for zinc and copper oxides to be used in multimicronutrient supplements to help minimize the size of the final tablet. Various reports have indicated the potential low relative bioavailability of zinc and copper oxide mineral salts [47,48]. Additional study of these factors will be necessary to identify the primary determinants of low mineral bioavailability from the dietary supplement.
Regardless of the possible factors contributing to apparent low mineral bioavailability from this supplement, it needs to be remembered that no comparison was made with any other multi-mineral supplement. Thus, it is impossible to state unequivocally that the mineral bioavailability from this supplement is any less bioavailable than any other currently available dietary multivitamin and mineral supplement source until additional direct comparative studies are performed. Rather, we prefer to view these initial observations as a warning to remember that the label on the bottle of all dietary supplements attests only to the presence of a particular amount of a given substance in the supplement tablet and gives no guarantee of its bioavailability, and as an encouragement for more research in this area.
CONCLUSIONS
For the multi-micronutrient dietary supplement studied neither the disintegration tablet rate nor intestinal permeability of the subject influenced the 12 hour-postdose nutrient AUC for riboflavin, folate, vitamin C, iron, zinc and copper; nevertheless, for vitamin B12 the intestinal permeability of the subject influenced significantly (p < 0.005) the AUC. The 12 hour-postdose AUC for riboflavin, vitamin C and folate for both forms of the supplement and that for vitamin B12 only for crushed tablet treatment were all significantly (p < 0.001) higher than after a test meal alone, suggesting that these water-soluble vitamins were being absorbed from the supplement. In contrast, there was no significant increase in the AUC after supplement administration for zinc and copper for both forms of the supplement and for iron for crushed tablet treatment, suggesting questionable mineral bioavailability from the multi-micronutrient supplement, despite the delivery of substantial amounts of iron (72 mg), zinc (60 mg) and copper (8 mg). Obtained results are suggestive of using multi-micronutrients dietary supplements formulated at about 100% Daily Value as a vehicle to combat the prevalence overall of suboptimal vitamin status in healthy adults and improve their micronutrient status to levels associated with reduced risk for several chronic diseases [35].
ACKNOWLEDGMENTS
We wish to express our appreciation to the volunteers who participated in this study and to the entire staff of the Metabolic Research Unit at the Jean Mayer Human Nutrition Research Center on Aging at Tufts University for their facilitation of the work. In addition, we would like to thank the staff of the Nutrition Evaluation Laboratory and Drs. P. Bagley and J. Selhub of the Vitamin Bioavailability Laboratory for assistance and advice in vitamin biochemical analysis and to Ms. J. Bradley for assistance in data management. This material is based upon work supported wholly by the U.S. Department of Agriculture, under agreement No. 58-1950-9-001. Any opinions, findings, conclusions, or recommendations expressed in this publication are those of the authors and do not necessarily reflect the view of the US Department of Agriculture.
FOOTNOTES
Dr. Navarro was a on a NATO fellowship in the Mineral Bioavailability Laboratory of the Tufts University. His present address is Department of Nutrition and Food Science, School of Pharmacy, University of Granada, SPAIN.
Received June 25, 2002. Accepted November 5, 2002.
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  A. K Kant and B. I Graubard Ethnic and socioeconomic differences in variability in nutritional biomarkers Am. J. Clinical Nutrition, May 1, 2008; 87(5): 1464 - 1471. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
