Contents - Previous - Next
This is the old United Nations University website. Visit the new site at http://unu.edu
Xu-Nian Liu, Jingnian Kang, Li Zhao, and Fernando E. Viteri
Iron deficiency in vulnerable groups is widespread in China, as in most developing countries, and iron supplementation has been accepted for many years as an effective targeted measure to control it. However, the current practice of daily dosing frequently causes unfavourable side effects that result in non-compliance. We randomly assigned 238 3-6-year-old children attending a kindergarten in Chang-ji City, Xinjiang, China, of whom 37% were anaemic, to receive 6 mg of iron per kilogram of body weight as FeSO4 daily, twice a week, or weekly under direct supervision for three months. The weekly dose was as effective as the daily dose in controlling anaemia, had insignificant side effects (4% of the children experienced side effects in contrast to 36% of those receiving the daily dose), and, on the evidence of serum ferritin distribution patterns, avoided temporary iron overload. The twice-weekly dose had no advantages over the weekly dose. On the basis of these results, long-term targeted preventive supplementation with weekly doses of iron should be considered among groups where iron deficiency and anaemia are prevalent.
Iron deficiency is a public health problem in China as well as in most developing countries [1-2]. Clinical experience shows that infants and young children in north-western China often have anaemia and iron deficiency, diagnosed as low serum ferritin levels, high free erythrocyte protoporphyrins, and low percentage saturation of transferrin. This would also be expected in view of the high prevalence of iron-deficiency anaemia among pregnant women in that region: over 50% between the third and eighth months of pregnancy have haemoglobin concentrations below the WHO cut-off point of 110 g/L, corrected for altitude above sea level .
Preventing and correcting iron-deficiency anaemia are particularly urgent among infants and young children because of the negative consequences it carries, some of which may be long-lasting if not permanent [4, 5]. For this reason, treating anaemia with 6 mg of iron per kilogram of body weight per day, supplementing with 1-3 mg of iron/kg/day after weaning or earlier in low-birthweight infants, and fortifying milk and weaning foods have been proposed [6, 7].
The daily administration of relatively high doses of iron results in a rapid decline in iron absorption . This produces a gastrointestinal epithelium permanently loaded with iron. This condition is unphysiological and possibly responsible for the side effects, which are directly proportional to the iron dose .
Taking these factors into account, Viteri et al.  conducted a series of experimental studies in iron-normal and iron-deficient rats given iron doses equivalent to those suggested by WHO for pregnant women. They demonstrated that administration of iron supplements every three days (coinciding with gut mucosal turnover time in the rat) improved true iron absorption by more than 2.5 times compared with a similar dose given daily. This supplementation avoided the well-known phenomenon of iron-absorption blockage after the ingestion of repeated iron doses  and the sustained iron loading of the gastrointestinal lumen and mucosa among previously iron-deficient rats .
On the basis of these very encouraging results in animals, we decided to test the hypothesis that intermittent iron supplementation is also more efficient than daily iron administration in humans. Since intestinal mucosa renewal time in humans is between five and six days , a weekly dose of iron should be almost as effective as a similar daily dose, and side effects, possibly caused by a temporary iron overload, should be minimized. We had an ideal setting to carry out this research in a large day-care centre (kindergarten) at Changji City that cares for children 3-6 years old.
TABLE 1. Adjustment of iron doses according to children's weight
|Child's weight (kg)||Tablets givena||Iron per kg (mg)|
a. Each whole tablet contains 60 mg of elemental iron.
Subjects and methods
The total population of 246 healthy children, 3-6 years of age, attending the kindergarten of Chang-ji Autonomous Prefecture, in Xinjian province, China, were enrolled in the study. The kindergarten operates six days a week and has nine large classrooms, each with adjoining bedrooms (for naps), bathrooms, and food-heating facilities.
Two centrally prepared meals and two snacks are offered daily, varying little from day to day, and are the same for all children, who eat ad libitum, including extra helpings. The main dish consists of noodles, vegetables, and chicken or fish soup. Snacks are seasonal fruit and candy.
There are three of the nine classrooms for each of three age groups, 3-4, 4-5, and 5-6 years; and within each age group the children are assigned randomly to one of the three rooms. For our study, the classrooms for each age group were randomly chosen for one of three different iron-supplementation regimens, lasting three months. Thus a similar number of children, with similar age and sex distributions, were assigned to each regimen, as follows:
» group 1 (89 children) received a daily dose of iron, six days per week;
» group 2 (74 children) received the same dose but only twice a week, once every three or four days;
» group 3 (83 children) received the same dose but only once a week, every seven days.
Thus, in each week, and over the course of the study, the children in group 2 received one-third and those in group 3 one-sixth the total dosage received by those in group 1.
The iron dose was aimed to be therapeutic, providing close to 6 mg of elemental iron per kilogram of body weight. This dosage was chosen to ensure enough daily iron intake for anaemic children and was 10 to 12 times the recommended daily iron intake. It also was proportionately similar to the 120-180 mg of iron recommended for pregnant women in the developing world .
The basic preparation consisted of 300-mg FeSO4·7H2O tablets containing 60 mg of elemental iron. To adjust the dose for the child's weight, tablets were divided into halves or quarters, providing 30 or 15 mg of elemental iron respectively, and given as shown in table 1. The tablets were administered by the teachers one hour after breakfast under direct supervision to make sure the child swallowed them.
The children's ages ranged from 36 to 71 months; 238 (107 boys, 131 girls) of the 246 children enrolled completed the study. Sex distribution by age group and by haematological status was homogeneous. Five children left the kindergarten during the study because of personal reasons (two each from the daily and weekly groups, one from the twice-weekly group). Three children from the daily group discontinued supplementation because of persistent nausea; two of these were non-anaemic. Exclusion criteria were chronic infectious diseases, cardiopathies, or respiratory diseases, and intake during the previous month of supplements or drugs containing iron or specially prescribed iron-rich and absorption-promoting foods for the month prior to entering the study.
The protocol was approved by the Human Subjects Committee of Chang-ji Hospital and Prefecture. All parents or guardians and the kindergarten authorities gave their informed consent for their children's participation.
Height and weight were measured by attending physicians using standardized anthropometric techniques  before the start of supplementation and at the end of the study. The scale and stadiometer used for all measurements were located in the health office of the kindergarten.
Before iron supplementation was started and one day after the last dose was administered, 5 ml of antecubital venous blood was collected, with minimal stasis, into tubes containing sodium EDTA. The blood was always collected before breakfast. Blood haemoglobin concentration was measured with cyan-methaemoglobin kits and serum ferritin with double-antibody radioimmunoassay kits prepared and standardized by the Atomic Energy Institute in Beijing. Radioactivity of 125I was counted in an FJ 2008 model automatic gamma counter (Xian). Haemoglobin and ferritin standards and control samples were obtained from the Tianjin Children's Health Institute, Tianjin, China, which is standardized against WHO standards. Between-duplicate errors were 2% and 4% for haemoglobin and serum ferritin respectively. The between-run variability was always below 4% for haemoglobin and 6% for serum ferritin. Samples and runs falling outside these ranges of error were repeated.
The teachers and helpers kept a record of any health alteration or complaint that each child manifested during the study. These were then tabulated, without knowledge of the children's supplementation regimen, by two nurses in charge of the clinic at the kindergarten with the assistance of a nonparticipating physician, to estimate the rate of undesirable side effects for each child. At the end of the study, the children with side effects were tabulated according to their respective regimens.
Data analyses included descriptive statistics, analysis of variance with the Newman-Keuls sequential test of significance between subclasses, covariance, and Pearson correlations performed by SAS-PC programmes (SAS Institute, Cary, N.C., USA) to compare changes between initial and final values by experimental group and by initial anaemic or non-endemic status, defined by initial haemoglobin concentration below or above the WHO-suggested cut-off value of 110 g/L. Simple and cumulative distributions were also estimated.
TABLE 2. Ages of the children in the study
|Supplementation regimen||No.||Age (months)|
TABLE 3. Haemoglobin levels in children before and after three months' iron supplementation
|Supplementation regimen||No.||Haemoglobin (g/L)|
|Non - anaemic|
Supplementation doses were approximately 6 mg
of iron per kilogram of body weight (table 1).
Classification as non-anaemic or anaemic based on child's haemoglobin level before supplementation.
a. All values after supplementation are higher than those before supplementation (p <.01).
b. Lower than the values after supplementation for the other two regimens in the non-anaemic group (p < .05).
The children were well nourished by anthropometric measurements, none being more than two Z scores below the WHO standards in height-for-age, weight-for-age, and weight-for-height (data not presented).
Table 2 shows the ages of the children, divided into non-anaemic and anaemic groups (Hb > 110 g/ L or < 110 g/L respectively), by supplementation regimen. The anaemic children were younger than the non-anaemic (p < .05).
Table 3 presents the haemoglobin concentrations before and after supplementation of the children in the three regimens classified by their non-anaemic or anaemic condition at the start of the study. The "before" values were similar within each group. As expected, the response of the anaemic children was larger in every regimen than that of the non-anaemics. The increments in haemoglobin observed with the three regimens in both groups were similar. At the final evaluation no children had haemoglobin levels below 110 g/L, and all fell within a normal probability plot for normally (Gaussian) distributed values (fig. 1). All the anaemic and 37% of the non - anaemic children responded to the three iron-supplementation regimens with increments greater than 10 g of haemoglobin per litre. That value would be considered iron deficient but not anaemic. The 94 non-endemic children whose haemoglobin concentration did not increase more than 10 g/L during the course of supplementation had levels of 129.5 ± 0.72 g/L (mean ± SE) and can be considered to have had adequate iron nutrition even before entering the study.
FIG. 1. Cumulative frequency distribution of haemoglobin values (g/dl) among 3-6-year-old children after three months of iron supplementation daily (black circles), twice a week (squares), or weekly (hollow circles)
TABLE 4. Serum ferritin levels before and after three months' iron supplementation
|Supplementation regimen||No.||Serum ferritin (µg/L)|
a. Ail mean values before
supplementation in the non-anaemic group are higher than those ID
the anaemic group (p < .001).
b. All mean values after supplementation are higher than those before supplementation (p < .01).
c. Lower than the values after supplementation for the other two regimens in the non-anaemic group (p < .05).
d. Mean values after supplementation for the different regimens in the anaemic group are significantly different from each other (p <.05).
FIG. 2. Distribution of children who were anaemic at the start of the study by serum ferritin levels after three months of iron supplementation daily (squares, solid line), twice a week (hollow circles, dashed line) or weekly (triangles, dotted line) compared with the distribution of non-anaemic children before supplementation (black circles, solid line)
Initial and final serum ferritin concentrations for both non-anaemic and anaemic children and the three regimens are shown in table 4. It is evident that the values were significantly lower in the anaemic children. As in the case of haemoglobin, all three regimens resulted in significant increases in serum ferritin. However, the final concentrations were significantly lower among the non-anaemic children receiving the supplements daily or twice weekly than in their anaemic counterparts. Non-anaemic as well as anaemic children receiving iron only once a week had lower values than those receiving more frequent doses. Among the anaemic children, the final serum ferritin concentrations were proportional to total weekly iron intake and significantly different from each other.
The mean initial serum ferritin concentration among the 94 non-responders was 31.3 + 0.74 µg/L. None of these children had concentrations below 12 µg/L. Figure 2 compares the distribution of the levels of the anaemic children after supplementation and the original levels of the non-responders (iron well-nourished). It is evident that the distribution of values of the children who received the weekly dose is superimposable on that of the non-anaemic non-responders. The children supplemented twice weekly had a distribution of values that was very close to being normal, with a mean 1.5 times greater than that of the non-responders. Those supplemented daily had a skewed distribution with a higher median and a longer tail to the right. The higher levels surpassed the highest levels of the non-responders by 24 µg/L or more.
The number of children who experienced undesirable side effects once or more in the course of the three supplementation regimens is shown in table 5. Thirty-two (37%) of the children receiving a daily supplement had undesirable side effects, with nausea and vomiting being most common. Three of these children were dropped from the supplementation because of this. In contrast, only five (7%) and three (4%) of the children supplemented twice weekly and weekly respectively experienced undesirable side effects. It is important to note that side effects occurred in both the anaemic and non-anaemic groups receiving daily and twice weekly supplements but in none of the anaemic children given weekly supplements.
TABLE 5. Undesirable side effects among non-anaemic and anaemic children, by supplementation regimen
a. Some children had more than one symptom.
The prevalence of anaemia among the apparently normal preschool-age children in Chang-ji City, as defined by the 110 g/L WHO cut-off point for haemoglobin concentration, was 36%. As expected, anaemic children were younger than non-anaemic ones. On the other hand, on the basis of a haemoglobin response to iron supplementation of 10 g/L or greater, 59% of the children could be classified as responsive to iron and therefore most probably were iron deficient. These included all those classified as anaemic as well as 37% of the non-anaemics. The lowest initial haemoglobin concentration was 84 g/L, so the severity of anaemia can be considered moderate in comparison with the severe anaemia in parts of South Asia and Africa. At the end of the trial no child had a haemoglobin concentration below 110 g/L, and only two of the originally non-anaemic children had a lower level after supplementation than before (regression to the mean).
In controlling anaemia, the daily, twice-weekly, and weekly supplementation regimens appeared similarly effective. On the other hand, the weekly dose appeared much more efficacious and better tolerated since it achieved the same haematological response with one-sixth of the daily dose and produced essentially no side effects. The twice-weekly regimen appeared of intermediate efficacy but was also well tolerated. The daily dose was the least efficacious and was less well tolerated.
The final cumulative frequency distribution of haemoglobin was identical for the three regimens and reached a normal Gaussian distribution. This distribution was also shared by both the non-anaemic group as a whole and the non-anaemic subgroup that did not respond to iron supplementation with haemoglobin increments of 10 or even 5 g/L, thus considered fully normal, non-iron-deficient. These results are interpreted as definitive evidence for the effectiveness of any of the three regimens in controlling ferropenic anaemia in this population.
A conservative serum ferritin level for defining depleted iron stores (probable iron deficiency) in children is below 12 µg/L [16,17]. In this study, 32% of the anaemic and none of the non-anaemic children fulfilled this criterion. On the other hand, all the anaemic children and 4 of the 56 non-anaemic children who responded to supplementation by an increase in haemoglobin concentration of 10g/L or more had serum ferritin levels below 20 µg/L (7%). This means that 64% of all the responders had an initial serum ferritin value below that level. None of the non-responders and only 2 children of 238 after supplementation had levels below 20 µg/L. Again, these results are interpreted as definitive evidence of the effectiveness of any of the three regimens in controlling iron deficiency in this population.
The response of serum ferritin to the three regimens must also be analysed in terms of the possibility of temporary iron overload among the anaemic children receiving therapeutic doses even 12 weeks after beginning therapy. This response was described in daily-supplemented, iron-deficient, anaemic rats that continued to accumulate intestinal and liver iron in contrast to their non-anaemic counterparts receiving the same amount of iron . We do not know the exact mechanisms of this phenomenon. Among the children who were non-anaemic before supplementation, including all the non-responders, none had serum ferritin values above 50 µg/L. Also, none of these children had serum ferritin levels below 20 or above 50 µg/L. after supplementation with any of the three regimens.
Among the anaemic weekly-supplemented children, none had serum ferritin levels above 44 µg/L. In contrast, the 20 (64%) who received daily supplementation had levels above 50 µg/L. as did 9 (33%) of those receiving supplements twice weekly. Levels were as high as 74 and 68 µg/L. respectively. We interpret these findings as an indication of a phenomenon in the daily and twice-weekly supplemented children similar to that observed in our experimental work with rats. In these animals, the administration of iron every three days prevented temporary iron overload and increased the efficacy of supplementation [10, 12].
The correlation coefficients of regression analysis indicate a very high correlation between initial haemoglobin and log-transformed serum ferritin values in all the children (r=.860). When the children are divided between anaemic and non-anaemic, the correlation is stronger in the former (r=.567), which strongly suggests that the anaemia was due to iron deficiency. This was confirmed by the response to supplementation.
The difference in the frequency of reported (and observed) undesirable side effects between the children receiving daily and intermittent supplementation is striking, favouring both intermittent supplementation regimens. We do not }now whether there was temporary iron overload to which those symptoms may have been related or whether they reflected only a local response in the gastro-intestinal tract.
These results, supported by experimental evidence in animals, should stimulate a revision of iron-supplementation practices proposed for populations with a high prevalence of iron deficiency or at high risk of developing it. Iron-supplementation programmes have been plagued by poor adherence, a high frequency of undesirable side effects, and overall ineffectiveness. To this should be added the possibility that the daily administration of relatively high iron doses may reduce the absorption of other trace elements ; a weekly dose should be much safer in this regard.
The public health implications of these results are the object of other documents and publications . We believe that at present an effective iron-supplementation programme would be a feasible strategy for controlling iron-deficiency anaemia in older infants and young children. Experience with this strategy in children is very limited in the developing world because of several factors: few programmes and even fewer evaluations, low coverage, cost, and difficulties in implementation. These are partly due to inadequate information on the need to provide supplements to children who appear healthy; to the reluctance of mothers to provide iron to children, given their personal experience with undesirable side effects; and to operational constraints associated with the daily regimen. Several of these constraints may be overcome by a better-tolerated weekly iron dose. The strategy of preventive supplementation [21, 22]-that is, providing long-term weekly iron supplements to prevent iron deficiency to population groups at risk of iron-deficiency anaemia- should be considered seriously in light of the results of this study. Taking a weekly pill for a long time is not a strange concept in areas where malaria is endemic. Building on this experience and possibly incorporating small vitamin A doses  in areas where it may also be deficient, may boost further the concept of preventive supplementation.
In China, as in most of the developing world, the effective control of iron deficiency and anaemia requires exploring and using all possible strategies together, including means to ensure adequate birth-weight and duration of gestation, late cord ligation, breast-feeding, consumption of high-quality diets and favourable meal patterns, food fortification (milk, weaning and staple foods), and preventive iron supplementation. In our opinion, the most feasible measures to implement in the short term are effective prenatal iron and folate supplementation for improving the outcome of pregnancy, delayed cord ligation, support of breast-feeding, and long-term weekly iron supplementation. However, the last is not contemplated yet in China or in most of the developing world, mainly because the target is the anaemia of pregnancy, and partly because, despite many efforts, prenatal iron supplementation has not proved very successful . This negative image extends to other iron supplementation programmes.
Several factors have been identified as responsible for this lack of success, including low coverage, improper information systems, inefficient programme management and supplement distribution, late initiation of prenatal care, and rejection of the programmes . The last is due partly to undesirable side effects produced by the daily intake of the recommended doses proposed for developing nations by WHO . About 30% of subjects receiving this dosage ( > 120 mg of iron as FeSO4) daily have been reported to experience such side effects , a proportion similar to that observed with the daily dose in this study. The development of iron tablets using the gastric delivery system is promising, mainly because, by improving and delaying iron absorption, they reduce the daily dose that is ingested and the undesirable gastro-intestinal side effects . However, production constraints and cost make this technology unavailable at present. The logistical advantage of giving a weekly dose rather than a small daily dose (i.e., a supplement of 1 mg/kg/day) should be taken into account.
The administration of a weekly dose of iron as proposed on the basis of the results of this study would obviate many of the obstacles to iron supplementation. It is more flexible in its application, so that communities could become directly responsible for its implementation under the guidance of public health authorities. It is certainly more economical, particularly if community-based, and should make it easier to increase coverage.
Acknowledgement is made to the Han Suyin Foundation and the University of California at Berkeley and San Francisco for making possible a scholar exchange for Dr. Liu Xunian at the Department of Nutritional Sciences at the University of California at Berkeley. Thanks are also given to the Chang-ji Hospital authorities for partial support of the study and for allowing Dr. Liu Xunian an extended leave of absence, and to the authorities, personnel and families of the kindergarten of Chang-ji Autonomous Prefecture in Xinjian, China, where the study was conducted.
1. Chang Y. Zhai F. Li W et al. Nutritional status of preschool children in poor rural areas of China. Bull WHO 1994;72:105-12.
2. Chen XC, Wang WG, Yan C, Yin TA, Xu QM. Studies in iron deficiency and anemia, rickets and zinc deficiency and their prevention among Chinese preschool children. Prog Food Nutr Sci 1992;16:263-77.
3. Guo ZX, Wang LL, Han L, Guo BZ. The study of prevention and cure, and investigation of iron status in reproductive women and adolescent children in Len Zhuo area. In: Zhung HM, ed. Obstetrics and gynecology textbook. 3rd ed. Beijing: People's Republic Hygiene Agency, 1990:110-13.
4. Scrimshaw NS. Functional significance of iron deficiency. In: Enwonwu CO, ed. Functional significance of iron deficiency. Nashville, Tenn, USA: Center for Nutrition, Meharry Medical College, 1989:1-13.
5. Lozoff B. Jimenez E, Wolf AW. Long term developmental outcome of infants with iron deficiency. N Engl J Med 1991;325:687-95.
6. International Nutritional Anemia Consultative Group. Combating iron deficiency in Chile: a case study. Washington, DC: ILSI/Nutrition Foundation, 1986.
7. Academy of Pediatrics, Committee on Nutrition. Iron supplementation for infants. Pediatrics 1976;58:765-8.
8. Norrby A. Iron absorption studies in iron deficiency. Stand J Haematol 1974;11(suppl.20):4-125.
9. Solvell L. Oral iron therapy: side effects. In: Hallberg L, Harwerth HG, Vannotti A, eds. Iron deficiency, pathogenesis, clinical aspects, therapy. New York: Academic Press. 1970:573-83.
10. Viteri FE, Liu X-N, Martin A, Tolomei K. Iron supplementation of normal and iron-deficient rats: 1. Retention of daily and intermittent iron supplements. J Nutr 1995;125 (in press).
11. Brown Jr EB, Dubach R. Moore CV. Studies in iron transportation and metabolism: XI. Critical analysis of mucosal block by large doses of inorganic iron in human subjects. J Lab Clin Med 1958;52:335-55.
12. Martin A, Tolomei K, Viteri FE. Iron metabolism in Fe supplemented rats. FASEB J 1990;4:A1076.
13. Trier, JS. Morphology of the epithelium of the small intestine. Handbook of physiology, sec. 6. Vol III: Alimentary canal. Ch. 63. Washington, DC: American Physiological Society, 1968:1125-75.
14. De Maeyer EM, ed. Preventing and controlling iron deficiency anaemia through primary health care. Geneva: WHO, 1989.
15. Habicht JP. Estandardización de métodos epidemiológicos cuantitativos sobre el terreno. Bol Of San Pan 1974;76:375-84.
16. Cook JD, Lipschitz DA, Miles LEM, Finch CA. Serum ferritin as a measure of iron stores in normal subjects. Am J Clin Nutr 1974;27:681-87.
17.. Hallberg L, Bengtsson C, Lapidus L. Screening for iron deficiency: an analysis based on bone-marrow examinations and serum ferritin determinations in a population sample of women. Br J Haematol 1993;85:787-98.
18. Solomons NW. Competititive interaction of iron and zinc in the diet: consequences for human nutrition. J Nutr 1986;116:927-35.
19. Viteri FE. Report to WHO on global strategy for the control of iron deficiency. Geneva: World Health Organization, Nutrition Unit, 1993.
20. WHO/UNICEF/UNU. Consultation on prevention of iron deficiency and anaemia. Geneva: World Health Organization, 1993.
21. Viteri FE, Hercberg S, Galan P, Guiro A, Preziosi P. Absorption of iron supplements administered daily or weekly: a collaborative study. In: Nestlé Foundation for the Study of the Problems of Nutrition in the World. Annual report 1993. Lausanne: Nestlé Foundation, 1994;82-96.
22. Viteri FE. The consequences of iron deficiency and anaemia in pregnancy on maternal health, the fetus and the infant. SCN News 1994;11:14-18.
23. Rhamathullah L, Underwood BA, Thulasiraj RD et al. Reduced mortality among children in southern India receiving a small weekly dose of vitamin A. N Engl J Med 1990;323:929-35.
24. WHO, Maternal Health and Sate Motherhood Programme, Division of Family Health. Iron supplementation during pregnancy: Why aren't women complying? A review of available information. Geneva: World Health Organization, 1990.
25. Gillespie S. Kevany J. Mason J. eds. Controlling iron deficiency: a report based on an ACC/SCN workshop. ACC/SCN State-of-the-An Series, Nutrition Policy Discussion Paper no. 9. Geneva: ACC/SCN, 1991.
26. Hallberg L, Sölvell L, Brise H. Search for substances promoting the absorption of iron: studies on absorption and side effects. Acta Med Stand 1966;459(suppl): 11-21.
27. Charoenlarp P. Dhanamitta S. Kaewvichit R et al. A WHO collaborative study on iron supplementation in Burma and in Thailand. Am J Clin Nutr 1988;47:28097.
28. Simmons WK, Cook JD, Bingham KC et al. Evaluation of a gastric delivery system for iron supplementation in pregnancy. Am J Clin Nutr 1993;58:622-6.
Contents - Previous - Next