Document reference: (August 2003)


N.F. Krebs
Section of Nutrition, Department of Pediatrics
University of Colorado Health Sciences Center
Box C225, Denver, CO 80262, USA


Zinc stable isotopes have been especially useful for studies of pediatric zinc bioavailability and homeostasis. Application of stable isotope methodology has confirmed that the gastrointestinal tract is the primary site of zinc homeostasis, involving both absorption of exogenous zinc and regulation of endogenous intestinal zinc. This report reviews findings from investigations in normal exclusively breastfed and formula fed infants, and factors which can affect the bioavailability of the diet. Also presented are findings from isotope studies in which the condition of the host impacts zinc homeostasis, including infants and children with cystic fibrosis and low birth weight infants.


Zinc deficiency in infants and children has now been well documented and appears to be very common, perhaps as common as iron deficiency. Results of zinc supplementation trials have convincingly demonstrated beneficial effects of zinc supplementation on linear growth and weight gain [1], on prevalence of such common infectious diseases as diarrhea (acute and persistent) and pneumonia [2], and on mortality in small for gestational age infants [3]. Key issues in infant and child zinc nutrition can be addressed by application of zinc stable isotopes, including zinc bioavailability, homeostasis and metabolism, requirements, and ultimately prevention & treatment of deficiency. The latter is an especially worthy goal, because of the substantial morbidity and mortality associated with zinc deficiency.

Considerations for stable isotope methodology in pediatric studies

The first and foremost aspect of zinc stable isotopes is their safety for use in pediatric studies. There are 3 zinc stable isotopes (67Zn, 68Zn, 70Zn) with sufficiently low natural abundance to allow them to be used as tracers. This availability of multiple isotopes allows essentially concurrent oral and intravenous administration of different isotopes in the same subject under different experimental conditions. Because the stable isotopes are more abundant than the zinc radioisotopes, careful attention must be given to minimizing doses to avoid perturbation of physiologic processes simply by the dose of "tracer." Unless one is interested only in determination of fractional absorption from single foods, there are considerable advantages to administering the isotopes (as an extrinsic label) with several or all meals over a 24-hour course. This allows representation of zinc absorption from a whole diet, and also allows division of the dose into smaller increments, thus minimizing potential impact on normal physiologic processes.

To obtain the most complete information from the stable isotope studies, meticulous collection of both urine and stool is necessary. Details of approaches to determine fractional absorption from fecal monitoring or dual isotope ratio in urine have been described elsewhere [4]. Because of the critical role of intestinal endogenous zinc in overall zinc homeostasis, it has become apparent that it is important to determine endogenous fecal zinc excretion, which necessitates collecting stool samples for several days. Endogenous fecal zinc can be determined by isotope dilution technique, which involves measurement of total zinc and isotopic enrichment in fecal samples and urine enrichment over the same time period [4]. For stool collections, portable collection containers have been developed which can be used in different positions and apparatus for the infant, such as in a stroller, swing, and infant seat. Complete 24 hr urine collections are not necessary, and thus specimens can be adequately obtained with an adhesive bag that can be emptied from the bottom, without disturbing the adhesion to the skin.

Normal Zn homeostasis and bioavailability

Studies in normal 2-5 month old infants who were either exclusively breastfed or formula fed reveal dramatic differences in variables of zinc homeostasis related to feeding type [5,6]. Zinc intake from human milk between 2 and 4 months of age is approximately 1/7 that of formula fed infants, while fractional absorption is more than 2-fold greater, with an average (SD) of 0.54 (.08) for human milk and 0.22 (0.04) for cow milk based formula. Thus the difference in absorbed zinc is only about 2.5-fold, averaging 0.6 and 1.6 mg/d for the breastfed and formula fed infants, respectively. Excretion of endogenous fecal zinc is also greater in the formula fed infants compared to the breastfed infants, resulting in further narrowing of the difference in net absorbed zinc between the two groups. These investigations have highlighted relationships between variables of zinc homeostasis: total absorbed zinc is positively correlated with endogenous fecal zinc. For breastfed infants, it is both the favorable absorption of the zinc from human milk plus the ability to minimize losses of endogenous zinc from the intestine that are critical to their ability to achieve appropriate net absorption of zinc with much lower daily intakes.5

Affectors of zinc bioavailability

There are several factors which may potentially have an adverse impact on the bioavailability of zinc in a given food or diet. There are still considerable gaps in our knowledge of the effects, interplay, and net impact of these factors. Stable isotopes offer powerful methodology to examine these relationships. Examples of factors that may affect bioavailability of zinc from foods and overall diets include especially phytic acid, a potent inhibitor of zinc absorption, and supplemental iron, which may interact with zinc at the site of absorption from the gastrointestinal tract and/or have a systemic influence on secretion or reabsorption of intestinal endogenous zinc.

For populations primarily reliant on plant-based diets, a high intake of inhibitors of zinc absorption, rather than a low zinc intake per se, is considered to be a major etiologic factor for zinc deficiency. Phytic acid is the primary example of such an inhibitor, and various phytate reduction strategies have thus been proposed and investigated. These will be discussed in the manuscript by Hambidge et al in this series.

The possible interaction between supplemental iron and zinc is an important area of interest because of the highly prevalent co-existence of deficiency of these two minerals. Treatment approaches will need to be tailored to minimize interference with bioavailability of each micronutrient. We have explored the effect of the level of iron fortification in infant formula on zinc absorption from primarily breastfed infants who are receiving small amounts of formula. Sixteen primarily breastfed infants, 3-4 months of age, who also were receiving modest amounts of formula (6 oz/day maximum intake) were randomized to either low iron formula (4.5 mg Fe/L) or to high iron formula (12 mg Fe/L). Human milk feeds were extrinsically labeled with 70Zn and formula was labeled with 67Zn. The labelled human milk was fed over 4 or 5 feeds and the formula over 2 feeds in a 24 hr period. Variables of zinc homeostasis were determined by previously described methods [4,5,6] with determination of zinc intake, fractional absorption, and absorbed zinc all determined separately for the formula and human milk. Briefly, preliminary results of these studies include a significantly lower fractional absorption of zinc from human milk (p < 0.05) and a higher (non-significant) daily excretion of endogenous fecal zinc, and lower net absorption of zinc (p = 0.06) in those infants receiving the high iron formula [7]. These results suggest a systemic rather than luminal effect of the higher iron intake on zinc bioavailability, and clearly support an interaction between level of iron intake and zinc homeostasis. The results also emphasize the importance of carefully designed studies that examine all aspects of zinc homeostasis.

Host effects on zinc homeostasis

Zinc homeostasis has been examined in several different clinical conditions. We have reported that in infants and children with cystic fibrosis (CF), there are a number of perturbations. In young infants with CF who were approximately 6 weeks old and who had been identified by newborn screen rather than by clinical symptoms, nearly one third of the infants had plasma zinc levels in the range of zinc deficiency [8]. In detailed metabolic studies using both intravenous and oral stable isotope tracers, we also found that fractional absorption was lower in both breastfed and formula fed infants, averaging 0.40 for the breastfed infants, and 0.13 for the formula fed infants. Equally striking were the relatively large losses of endogenous fecal zinc, which were correlated with the amount of fat in the stools. Furthermore, in contrast to findings in normal populations, there was no correlation between the amount of zinc absorbed and the amount of endogenous fecal zinc. The combination of the lower fractional absorption and relatively excessive endogenous fecal zinc losses resulted in mean negative net absorption [9]. Older children and adolescents with CF and pancreatic insufficiency were also found to have significantly improved absorption of exogenous zinc while receiving exocrine pancreatic enzyme replacement therapy compared to baseline absorption [10]. These studies suggest that in the setting of pancreatic insufficiency (and other abnormalities of the gastrointestinal tract), there are alterations in both absorption of dietary zinc and in handling of intestinal endogenous zinc. The combination of these perturbations likely contributes to the suboptimal zinc status that has been observed in young infants and children with CF. The correlation between fecal fat and endogenous fecal zinc suggests that fat malabsorption may interfere with reabsorption of endogenous zinc and may be important in other settings with gastrointestinal tract dysfunction.

Infants who are born with low birth weight account for a large percentage of births worldwide and are at risk for reduced survival and overall poorer outcomes. Zinc supplementation trials of small-for-gestational age infants, who make up the majority of low birth weight infants in developing countries, have demonstrated substantial benefit, including a significant reduction in mortality [3]. This is strongly suggestive of high risk for zinc deficiency in this population. There are a number of factors which may contribute to high risk of zinc deficiency in infants born with low birth weight. These include relatively high physiologic demands associated with catch-up growth; risk of inadequate intake, especially if premature and not receiving zinc fortified human milk or formula; risk of some gastrointestinal tract immaturity and/or dysfunction including relative pancreatic insufficiency and "physiologic" fat malabsorption; and lower levels of hepatic metallothionein-bound zinc, which may normally be available as a source of zinc to the young infant.

We have studied hospitalized preterm infants at approximately 2 wk postnatal age, who were on full enteral feeds - either infant formula designed for preterm infants or human milk with added commercial human milk fortifier. Both intravenous and oral isotopes were administered and full metabolic studies were completed. Net absorption of zinc from the stable isotopes and net zinc retention (including losses of zinc in urine) were both within the range estimated for normal intrauterine accretion rates. Furthermore, a significant positive correlation between net absorbed zinc and daily weight gain was observed [11]. The results of these studies suggest that, at least with generous zinc intakes, preterm low birth weight infants demonstrate adequate zinc homeostatic processes.

We have also recently examined the size of the exchangeable zinc pool (EZP) [12] at birth in low birth weight infants, most of whom were born prematurely, and have found a positive correlation between EZP size and birth weight (unpublished data). Although additional data will be required, the initial observations support the hypothesis that low birth weight infants are born with lower body zinc "stores," which may contribute to their apparent increased risk of zinc deficiency, especially if they do not receive generous dietary zinc intake in the early weeks to months of post-natal life.


The careful application of zinc stable isotope methods to normal and pathologic conditions in infants offers potentially invaluable new insight into zinc homeostasis and underlying causes of zinc deficiency. From the sum of the findings described above, the data suggest the following relationships. First, the amount of dietary zinc intake is positively related to the amount of absorbed zinc, but some dietary components, such as phytic acid and iron, can alter the percent and amount that is absorbed. The amount of endogenous fecal zinc is variable and normally is strongly related to the amount of absorbed zinc. This relationship is critical to maintaining zinc homeostasis in conditions of low zinc intake, such as in normal breastfed infants. The relationship can be disturbed by malabsorption, such as occurs in pancreatic insufficiency and CF. It is also possible that endogenous losses may be adversely affected in other conditions, such as infectious diarrhea or malabsorption, but this will require additional studies in pathologic conditions. Recent findings suggest that the EZP size is related to birth weight, and low birth weight infants may have lower zinc "stores" to access in the early months of postnatal life.

Future research using stable isotopes will likely continue to enhance our understanding of zinc homeostasis under normal and abnormal conditions, and thus will enable better prediction of populations at risk of zinc deficiency. Further, stable isotope methods offer great potential to provide an important evaluation component of various intervention strategies to treat and prevent deficiencies of zinc and other micronutrients.

Acknowledgements:: The research work described herein was supported by the following grants: NIH grants RR00069, T32-DK07658, DK02240, DK48520, #2T73MC00011-04, P30-DK-48520, and CA49981; and by the Pew Nutrition Fellowship T86-00279-023, and the Dannon Institute Nutrition Fellowship. The author also acknowledges the critical input from K. Michael Hambidge, MD, Sanju Jalla, PhD, Leland V. Miller, BA, Lei Sian, MD, Jamie E. Westcott, MS and other members of our research team for the data presented in this manuscript.


1. K. H. BROWN, J. M. PEERSON, J. RIVERA, L.H. ALLEN. Am J Clin Nutr 75 (2002): 1062.
2. The Zinc Investigators' Collaborative Group. J Pediatr 135 (1999): 689.
3. S. SAZAWAL, R.E. BLACK, V.P. MENON, P. DINGHRA, L.E. CAULFIELD, U. DHINGRA, A. BAGATI. Pediatrics 108 (2001):1280.
4. N.F. KREBS, L.V. MILLER, V. NAAKE, L. SIAN, J.E. WESTCOTT, P.V. FENNESSEY, K.M. HAMBIDGE. J Nutr Bioch 6 (1995): 292.
5. N.F. KREBS, C.J. REIDINGER, L.V. MILLER, K.M. HAMBIDGE. Pediatr Res 39 (1996): 661.
6. N.F. KREBS, C.J. REIDINGER, L.V. MILLER, M. BORSCHEL. J Pediatr Gastroenterol Nutr 30 (2000):29.
7. N.F. KREBS, M. STEIRN, J.E. WESTCOTT. Pediatr Res 130 (2000): 204A.
8. N.F. KREBS, M. SONTAG, F. ACCURSO, K.M. HAMBIDGE. J Pediatr 133 (1998):761.
9. N.F. KREBS, J.E. WESTCOTT, T.D. ARNOLD, B.M. KLUGER, F.J. ACCURSO, L.V. MILLER, K.M. HAMBIDGE. Pediatr Res 48 (2000): 256.
10. D. EASLEY, N.F. KREBS, M. JEFFERSON, L.V. MILLER, J. ERSKINE, F.J. ACCURSO, K.M. HAMBIDGE. J Pediatr Gastro Nutr 26 (1998):136.
11. S. JALLA, N.F. KREBS, D.J. RODDEN, L.V. MILLER. Pediatric Res 41(1997):233A.
12. L. V. MILLER, K. M. HAMBIDGE, V.L. NAAKE, Z. HONG, J.E. WESTCOTT, FENNESSEY P.V. J Nutr 124 (1994): 268.

Return to NAMLS-7 Zinc Session (overview)

NAMLS Homepage