The Journal of Nutrition Vol. 127 No. 8 August 1997, pp. 1451-1455
Copyright ©1997 by the American Society for Nutritional Sciences

Long-Term Oral Supplementation with Iron Is Not Harmful for Young Children in a Poor Community of Bangladesh^1,2

Amal K. Mitra³, Syed M. Akramuzzaman, George J. Fuchs⁴, Mohammad M. Rahman, and Dilip Mahalanabis

International Centre for Diarrhoeal Disease Research, Bangladesh, Mohakhali, Dhaka 1212, Bangladesh

ABSTRACT

INTRODUCTION

SUBJECTS AND METHODS

RESULTS

DISCUSSION

FOOTNOTES

LITERATURE CITED

ABSTRACT

The effect of long-term oral iron supplementation on morbidity due to diarrhea, dysentery and respiratory infections in 349 children, aged 2-48 mo, living in a poor community of Bangladesh, was evaluated in this double-blind study. The treatment group received 125 mg of ferrous gluconate (15 mg elemental iron) plus multivitamins and the controls received only multivitamins, daily for 15 mo. House-to-house visits were made on alternate days by trained community health workers for recording symptoms and duration of illnesses and for monitoring medicine intake. Seventy-six percent of the children continued the syrup for over 1 y. No untoward effects were noticed in either treatment group. The attack rates for diarrhea, dysentery and acute respiratory tract infections (ARI) were 3, 3 and 5 episodes per child per year, respectively. Each episode of diarrhea lasted a mean of 3 d, and those of dysentery and ARI, 5 d. The two treatment groups did not differ in the number of episodes, mean duration of each episode, or total days of illnesses due to diarrhea, dysentery and ARI. However, a 49% greater number of episodes of dysentery was observed with iron supplementation in a subset of the study children who were less than 12 mo old (P = 0.03). The results of this study suggest that long-term oral iron supplementation is not harmful for older children in a poor community. Further studies are needed to demonstrate the safety and efficacy of iron administration in young infants.

INTRODUCTION

Iron deficiency is the most common nutritional disorder among children in the developing world (Yip 1994). Young children are at great risk of developing iron deficiency, particularly during the rapid phases of growth (Viteri 1993). The prevalence of iron deficiency anemia among children aged 2 to 6 y in Karachi, Pakistan, was found to be 47% (Hamedani et al. 1987). More than half of these children had serum ferritin levels below normal. Iron deficiency is associated with reduced cell-mediated immunity (Chandra 1983, Pearson and Pitcock 1976, Stockman 1981) and reduced microbial-killing capacity of leucocytes (Chandra 1976). Programs of mass supplementation and food fortification have been advocated in many countries (Keusch 1990, Tomkins 1991, Yip 1994). Iron treatment of deficient individuals promotes growth (Chwang et al. 1988, Lawless et al. 1994), work performance, and mental development, and decreases morbidity and mortality for certain diseases (Angeles et al. 1993, Lawless et al. 1994, Yip 1994). However, the effect of iron administration on the risk of disease morbidity and mortality is controversial. Parenteral iron treatment is associated with exacerbation of certain infections, in particular malaria and respiratory diseases in infants in populations where malaria is endemic (Smith et al. 1989). Parenteral iron is also associated with increased risk of serious Escherichia coli sepsis in neonates (Oppenheimer 1989). In Santiago, Chile, daily feeding of iron-enriched milk for 6 mo was associated with an increased frequency of watery diarrhea and persistent diarrhea, but fewer episodes of Shigella infections (Brunser et al. 1993). In contrast, oral iron supplementation reduced infectious disease morbidity in certain countries (Angeles et al. 1993, Chwang et al. 1988). The unresolved debate over the interaction of iron and morbidity prompted us to evaluate the effect of iron supplementation on morbidity in children through a controlled prospective study.

---

SUBJECTS AND METHODS

Subjects. The study included infants and young children aged 2 to 48 mo of either sex and excluded children who were critically ill or who had evidence of congenital malformation or metabolic disorders. The study was conducted in a peri-urban slum village of Nandipara, about 10 km northeast of Dhaka, Bangladesh, from January 1990 through December 1991. This village is a low-lying area often flooded by river water. The inhabitants of Nandipara are of very poor socioeconomic status, and most families are with head of household being rickshaw-pullers, day-laborers or fishermen. The International Centre for Diarrhoeal Disease Research, Bangladesh (ICDDR,B) has maintained a day-care clinic in this community since 1981. The villagers seek care for all illnesses from the ICDDR,B clinic or occasionally from local practitioners. The most frequent ailments for which people attend the clinic include common cold, fever, pneumonia, dysentery, watery diarrhea, scabies, other skin infections, gingivitis and anemia. Informed consent was taken from a legal guardian before enrolling a child to this study. The study was approved by the Human Ethics Review Committee of the ICDDR,B.

Case definitions. Diarrhea was defined as three or more liquid stools in a 24-h period, except for breast-fed infants, for which the mothers' report of diarrhea was considered as an episode. Twenty-three percent (35 of 152) of the study children and 29% (46 of 157) controls were partially or fully breast-fed. The exact number of days that the child experienced watery, mucoid or bloody stools was recorded. If only watery type of stools occurred throughout the episode, it was recorded as watery diarrhea, whereas the presence of blood and/or mucus in any stool during the episode was considered to indicate dysentery. Acute respiratory tract infection (ARI) was defined as cough (or difficult breathing) and fast breathing (50 breaths or more per minute in a child less than 1 y old, 40 breaths or more per minute in a child aged 12 mo to 5 y), with or without chest indrawing (WHO 1981). Three or more symptom-free days differentiated episodes of diarrhea or respiratory tract infections.

Sample size. In a previous survey in Nandipara, the diarrheal attack rate (mean ± SD) was 3.7 ± 2 per child year (Henry, F., unpublished data). We considered a difference of 25% or more in diarrheal disease incidence due to the intervention as worthwhile. Considering a 5% alpha error, 90% confidence and 20% dropout or migration, the total sample size needed was 184 children. According to the clinic record, the incidence of respiratory infections was higher than that of diarrhea in this population.

Randomization. After a baseline census and household mapping, the population was divided into four homogenous clusters by geographic location. Block randomization was used to assign an approximately equal number of children to the intervention and control groups from each cluster. Each eligible child was individually randomized to one of the two groups and assigned an identity number. The test (iron plus vitamins) and the control (vitamins only) medications were of similar color and smell and dispensed in identical bottles; the composition is given in Table 1. If there was more than one eligible child in a family, each was assigned to the same group (iron or control). A master randomization code and list was kept sealed at the ICDDR,B. Group allocation was double-blinded, and the code was revealed to the investigators only after completion of data analysis.

Table 1. Composition of the iron and the control syrups [View Table]

Supplementation procedures. Mothers were advised to administer one teaspoon of syrup (5 mL) containing 125 mg of ferrous gluconate (15 mg elemental iron) plus vitamins or vitamins only to their children every day immediately before breakfast. The dose of iron was selected according to WHO/UNICEF/UNU recommendation (Anonymous 1994). Supplementation and morbidity surveillance were maintained for 15 mo. Compliance to the protocol was monitored by trained community health workers (one for each household cluster). Health workers visited assigned houses daily and recorded the amount of medicine taken by checking and returning the empty bottles to the clinic. On an average, the children consumed 80% (4 mL) of the dose.

Morbidity surveillance. House-to-house visits were made on alternate days to record symptoms and duration of diarrhea, dysentery and respiratory infections, any referrals to the study physicians or other health care providers, and any deaths. Two study physicians attended the weekly clinic and provided treatment for the patients. A record was maintained at the clinic of the type of illnesses and of any medicines prescribed. The investigators made unscheduled spot visits to monitor the activities of the workers. At these visits, problems encountered were discussed and advice made accordingly. Health assistants supervised the volunteers and summarized the daily records.

Statistical analysis. Data files were reviewed and cleaned by visual checks and by logical and range checks, and the files were analyzed using SPSS for Windows (version 6.1.2, SPSS Inc., Chicago, IL). Results regarding baseline characteristics are given as means ± SD, and statistical significance was assessed using Student's t test. Data that were not normally distributed were log-transformed to achieve normality. The morbidity data for watery diarrhea, dysentery and acute respiratory infections had non-normal distribution even after transformation; therefore, the morbidity data are presented as median and quartile values. Statistical comparisons were made by the Mann-Whitney U test (Daniel 1987). A standard technique for the analysis of quantitative outcomes as suggested by the World Health Organization (Cousens and Kirkwood 1990) was followed. To reduce any variability in morbidity rates due to age, sex or nutritional status, the data were further stratified. The results remained the same when the data for those who were noncompliers (six with noncompliance shortly after enrollment, and eight with intake for less than 30 d) were included, i.e., following the "intent to treat" design of a randomized trial. The data analysis was also performed based on the duration of supplementation (e.g., including only those with 6+ mo or 9+ mo of supplementation); it showed no significant differences in the disease morbidity between the two treatment groups.

---

RESULTS

A total of 349 children aged 2-48 mo were enrolled. Forty children (20 in each group) dropped out of the study, and 152 in the iron group and 157 controls remained in the study. Seventy-three percent (111 of 152) of children in iron group and 78% (123 of 157) controls continued supplement for 1 y or more (Table 2). The reasons for drop-out were migration (23), noncompliance to intervention shortly after enrollment (6), death (1 control due to drowning), too sick to continue (1), overage (1), and syrup intake for less than 30 d (8). However, there was no indication of differential attrition.

Table 2. Distribution of the study children by the duration of supplementation [View Table]

Characteristics of the study population. The total population of the village was 3712, with 550 children below 5 y of age. The majority of the study children (114 in iron and 108 controls) were between 1 and 4 y old. The average ages of the children in the iron and control groups were similar (31 and 28 mo), and the girls outnumbered the boys (59 vs. 51%). The majority of the children had mild to moderate grade of undernutrition. The mothers were young and malnourished, having an average body mass index of 18 ± 2 kg/m². Eighty percent of mothers and 60% of fathers had no formal education, and average family income was US $50 per month. The people lived in clusters of thatched houses having an average floor space of 12 m². Ninety percent of the people had access to tubewell water for drinking, whereas only 40% had access to a sanitary latrine. The two treatment groups did not differ significantly in their baseline characteristics (Table 3).

Table 3. Baseline characteristics of the study population1 [View Table]

Table 4. Effect of daily oral doses of iron on morbidity due to diarrhea and dysentery in children of different age categories1 [View Table]

Disease morbidity. After supplementation, the median number of episodes of watery diarrhea and dysenteric illnesses was three per child per year (Table 4). The average duration of each episode was 3 d for watery diarrhea and 4 d for dysentery. The total days of illness per child per year were 8 and 6 for watery diarrhea in the treatment and the control groups, respectively, and 11 for dysentery in each group. No significant difference was observed in diarrheal morbidity between the two groups for either sex. However, infants aged 2-11 mo who received iron had a higher number of episodes and total days of illness due to dysentery compared with those who received the control medication. The children did not differ according to treatment groups in diarrheal morbidity when categorized by nutritional status. The children suffered more often from respiratory illnesses than diarrhea (5 vs. 3 episodes per child per year). Each episode of ARI lasted for 5 d and each child suffered from a total of 24 d from ARI (Table 5). The two treatment groups did not differ in the number of episodes, average duration of each episode or total days of illnesses of ARI. There were no sex- or age-related differences (<12 mo vs. 12-48 mo) in respiratory illnesses between the two treatment groups. The children supplemented with iron or control medicine did not differ in their respiratory morbidity when categorized by nutritional status.

Table 5. Effect of daily oral doses of iron on morbidity due to acute respiratory tract infection in children of different age categories1 [View Table]

---

DISCUSSION

This study demonstrates that long-term iron supplementation did not increase morbidity of watery diarrhea or respiratory infections but did increase dysenteric illnesses in young infants. The reason could be that free iron is essential for the multiplication of bacteria including species of Candida, Escherichia, Mycobacterium, Pasteurella, Shigella and Staphylococcus (Weinberg 1974). Generally, sufficient exogenous iron to achieve at least 60-80% saturation to serum transferrin is necessary. We did not attempt to define the etiology of illness in this study; however, the commonly isolated pathogens are viruses in young children presenting with watery diarrhea and Shigella in children with dysentery in Bangladesh (Stoll et al. 1982b). A surveillance report, using a systematic sample of approximately 100,000 hospital patients in Bangladesh, showed that stool mucus was the chief symptom in 97 and 88%, and stool blood the chief symptom in 83 and 55%, of patients with S. dysenteriae type 1 (n = 66) and S. flexneri (n = 270) infection, respectively (Stoll et al. 1982a). In vitro studies showed that iron depletion of medium inhibits growth of Shigella and other bacteria, and iron excess promotes it (Weinberg 1974). Bacteria have the ability to produce "siderophores," such as phenolate or hydroxymate compounds, to solubilize and assimilate the metal. Therefore it is perhaps likely that iron enhanced growth for Shigella species in young infants who may not be able to metabolize iron as efficiently as do older children.

The debate over the interaction of iron and infection in the clinical setting is unresolved. Studies in Gambian children showed that fever associated with parasitemia occurred more frequently in the iron-treated group than in the placebo group (Smith et al. 1989). This difference was larger for the children having higher rate of parasitemia (10 or more positive fields per 100 high power fields) than for those having a lower rate of parasitemia. Morbidity due to malaria illness was not a part of our study, because malaria is not endemic in our study population. Another community-based study in children of low socio-economic stratum suggested that chronic iron intake through iron-enriched milk (12 mg/L) increased the incidence of diarrhea compared with consumption of control milk (Brunser et al. 1993). These studies suggest that iron deficiency states may be relatively protective and that conditions of iron excess may predispose to infection. These concepts most often seem to pertain to certain infections such as malaria and Yersinia infections and to situations in which parenteral iron has been used. There has been inadequate information to determine whether iron is equally harmful in other childhood diseases. Our study aimed to determine whether long-term iron supplementation increases the risk of childhood infection. Because any possible untoward effects of iron supplementation would probably be magnified with long-term use, we continued daily supplementation for 15 mo. We observed a greater number of episodes and days of illness of dysentery in young infants less than 12 mo old compared with older children aged 12-48 mo, but we observed no increase in watery diarrhea or ARI morbidity due to the supplementation with iron.

One limitation of our study is that we did not assess the efficacy of iron supplementation on laboratory indices of iron status (such as cell morphology, hemoglobin status and serum ferritin) or on physiological function tests. However, previous studies clearly demonstrated that oral iron supplementation substantially improves hematological and iron status (Chwang et al. 1988, Smith et al. 1989). Recent evidences indicated the efficacy of once or twice weekly compared with multiple daily dosing of iron (Liu et al. 1995, Ridwan et al. 1996). Whether less frequent dosing of iron would reduce the incidence of dysentery in young infants as observed in our study remains to be determined.

In conclusion, our data demonstrate that daily oral iron supplementation to young children is not associated with increased risk of watery diarrhea, dysentery or ARI; however, dysentery episodes were higher in a subset of young infants receiving iron administration. The effect of iron supplementation on young infants should be investigated further.

FOOTNOTES

Manuscript received 6 September 1996. Initial reviews completed 18 February 1997. Revision accepted 21 April 1997.

LITERATURE CITED

• Angeles I. T., Schultink W. J., Matulessi P., Gross R., Sastroamidjojo S. Decreased rate of stunting among anemic Indonesian preschool children through iron supplementation. Am. J. Clin. Nutr. 1993; 58:339-342 [Abstract/Free Full Text]
• Anonymous (1994) Report of the WHO/UNICEF/UNU Consultation on Indicators and Strategies for Iron Deficiency and Anaemia Programmes. Draft report, 6-10 December 1993, World Health Organization, Geneva, Switzerland.
• Brunser O., Espinoza J., Araya M., Pacheco I., Cruchet S. Chronic iron intake and diarrhoeal disease in infants. A field study in a less-developed country. Eur. J. Clin. Nutr. 1993; 47:317-326
• Chandra R. K. Iron and immunocompetence. Nutr. Rev. 1976; 34:129-132 [Medline]
• Chandra R. K. Nutrition, immunity and infection: present knowledge and future directions. Lancet 1983; 1:688-691 [Medline]
• Chwang L. C., Soemantri A. G., Pollitt E. Iron supplementation and physical growth of rural Indonesian children. Am. J. Clin. Nutr. 1988; 47:496-501 [Abstract/Free Full Text]
• Cousens, S. N. & Kirkwood, B. R., eds. (1990) Outcome Measures in Prospective Studies of Childhood Diarrhoea and Respiratory Infections: Choosing and Using Them. Diarrhoeal Disease Control Programme. Programme for Control of Acute Respiratory Infections. WHO/CDD/EDP/90.2, WHO/ARI/RES/90.4.
• Daniel, W. W. (1987) Biostatistics: A Foundation for Analysis in the Health Sciences, 5th ed. John Wiley & Sons, New York, NY.
• Hamedani P., Hashmi K. Z., Manji M. Iron depletion and anaemia: prevalence, consequences, diagnostic and therapeutic implications in a developing Pakistani population. Curr. Med. Res. Opin. 1987; 10:480-485 [Medline]
• Keusch G. T. Micronutrients and susceptibility to infection. Ann. N.Y. Acad. Sci. 1990; 587:181-188 [Medline]
• Lawless J. W., Latham M. C., Stephenson L. S., Kinoti S. N., Pertet A. M. Iron supplementation improves appetite and growth in anemic Kenyan primary school children. J. Nutr. 1994; 124:645-654
• Liu X.-N., Kang J., Zhao L., Viteri F. Intermittent iron supplementation is efficient and safe in controlling iron deficiency and anemia in preschool children. Food Nutr. Bull. 1995; 16:139-146
• Oppenheimer, S. J. (1989) Iron and infection: the clinical evidence. Acta Paediatr. Scand. 361 (suppl): 53-62.
• Pearson H. A., Pitcock J. A. The role of iron in host resistance. Adv. Pediatr. 1976; 23:1-33 [Medline]
• Ridwan E., Schultink W., Dillon D., Gross R. Effects of weekly iron supplementation on pregnant Indonesian women are similar to those of daily supplementation. Am. J. Clin. Nutr. 1996; 63:884-890 [Abstract/Free Full Text]
• Smith A. W., Hendrickse R. G., Harrison C., Hayes R. J., Greenwood B. M. The effects on malaria of treatment of iron-deficiency anaemia with oral iron in Gambian children. Ann. Trop. Paediatr. 1989; 9:17-23 [Medline]
• Stockman J. A. III Infections and iron: too much of a good thing? Am. J. Dis. Child. 1981; 135:18-20 [Abstract/Free Full Text]
• Stoll B. J., Glass R. I., Huq M. I., Khan M. U., Banu H., Holt J. E. Epidemiologic and clinical features of patients infected with Shigella who attended a diarrheal disease hospital in Bangladesh. J. Infect. Dis. 1982a; 146:177-183 [Medline]
• Stoll B. J., Glass R. I., Huq M. I., Khan M. U., Holt J. E., Banu H. Surveillance of patients attending a diarrhoeal disease hospital in Bangladesh. Br. Med. J. 1982b; 285:1185-1188
• Tomkins A. Recent developments in the nutritional management of diarrhoea. 1. Nutritional strategies to prevent diarrhoea among children in developing countries. Trans. R. Soc. Trop. Med. Hyg. 1991; 85:4-7
• Viteri, F. E. (1993) A Guide for the Global Control of Nutritional Anemias and Iron Deficiency. Nutrition Unit, World Health Organization, Geneva, Switzerland.
• Weinberg E. D. Iron and susceptibility to infectious disease. Science 1974; 184:952-956 [Free Full Text]
• World Health Organization Clinical management of acute respiratory infections in children: a WHO memorandum. Bull. WHO 1981; 59:707-716
• [Medline] Yip, R. (1994) Iron deficiency: contemporary scientific issues and international programmatic approaches. J. Nutr. 124 (suppl.): 1479S-1490S.

0022-3166/97 $3.00 ©1997 American Society for Nutritional Sciences