Iron and Zinc Supplementation Improved Iron and Zinc Status, but Not Physical Growth, of Apparently Healthy, Breast-Fed Infants in Rural ...
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
The Journal of Nutrition
Community and International Nutrition
Iron and Zinc Supplementation Improved Iron
and Zinc Status, but Not Physical Growth, of
Apparently Healthy, Breast-Fed Infants in Rural
Communities of Northeast Thailand1
Emorn Wasantwisut, Pattanee Winichagoon,* Chureeporn Chitchumroonchokchai, Uruwan Yamborisut,
Atitada Boonpraderm, Tippawan Pongcharoen, Kitti Sranacharoenpong, and Wanphen Russameesopaphorn
Downloaded from https://academic.oup.com/jn/article/136/9/2405/4664952 by guest on 28 February 2021
Institute of Nutrition, Mahidol University, Salaya, Nakhon Pathom 73170, Thailand
Abstract
Iron deficiency is prevalent in children and infants worldwide. Zinc deficiency may be prevalent, but data are lacking. Both
iron and zinc deficiency negatively affect growth and psychomotor development. Combined iron and zinc supplementation
might be beneficial, but the potential interactions need to be verified. In a randomized, placebo-controlled trial using 2 3 2
factorial design, 609 Thai infants aged 4–6 mo were supplemented daily with 10 mg of iron and/or 10 mg of zinc for 6 mo to
investigate effects and interactions on micronutrient status and growth. Iron supplementation alone increased hemoglobin
and ferritin concentrations more than iron and zinc combined. Anemia prevalence was significantly lower in infants
receiving only iron than in infants receiving iron and zinc combined. Baseline iron deficiency was very low, and iron
deficiency anemia was almost nil. After supplementation, prevalence of iron deficiency and iron deficiency anemia were
significantly higher in infants receiving placebo and zinc than in those receiving iron or iron and zinc. Serum zinc was higher
in infants receiving zinc (16.7 6 5.2 mmol/L), iron and zinc (12.1 6 3.8 mmol/L) or iron alone (11.5 6 2.5 mmol/L) than in the
placebo group (9.8 6 1.9 mmol/L). Iron and zinc interacted to affect iron and zinc status, but not hemoglobin. Iron
supplementation had a small but significant effect on ponderal growth, whereas zinc supplementation did not. To
conclude, in Thai infants, iron supplementation improved hemoglobin, iron status, and ponderal growth, whereas zinc
supplementation improved zinc status. Overall, for infants, combined iron and zinc supplementation is preferable to iron or
zinc supplementation alone. J. Nutr. 136: 2405–2411, 2006.
Introduction
use of iron and zinc together as a supplement at a prophylactic
Deficiencies of 2 important trace minerals, iron and zinc, are a dose has provoked controversy over their potential interaction
major public health problem in many developing countries (1). and resulting adverse impact on iron or zinc status (18) and
These problems exist even among populations where severe cognition (19). More data on the biological interaction between
protein-energy malnutrition has been alleviated. Iron deficiency these 2 nutrients are needed before further recommendations can
in young children can result in impaired cognitive and psycho- be made.
motor development (2–4). Iron supplementation was inconsis- Plant-based complementary foods in developing countries are
tently shown to reverse these impairments (2,4,5). Several poor sources for bioavailable iron, zinc, and other micronutri-
studies indicate that zinc deficiency results in poor growth in ents. Adequate nutrients from breast milk during the first 6 mo of
infants (6,7) and children (8–10) and a depressed appetite (8,11). life, complemented by high-quality complementary food during
Zinc deficiency may also lead to impaired motor development in the second half of the first year, are needed for the optimal growth
infants (12,13) and thus can interfere with cognitive perfor- and development of infants (20). The untimely introduction of
mance. Furthermore, zinc is essential for the integrity of the poor-quality complementary food can result in an inadequate
immune system, and a deficiency can result in reduced immu- uptake of iron and can deplete iron stores. In Thailand, rice gruel
nocompetence and decreased resistance to infection (14–17). or chewed glutinous rice is the main portion of complementary
Zinc supplementation administered to zinc-deficient children food introduced to young infants, even before the age of 4 mo.
enhances linear growth and reduces morbidity (17). Concurrent Although the major sources of iron and zinc in rural Thailand are
iron and zinc deficiency may be common among populations that animal foods, these are given to infants late or are consumed by
consume small amounts of animal-source foods. However, the infants only occasionally in small amounts (21). Growth faltering
starts at ;6 mo of age in NE Thailand (22). Furthermore, a 1995
1
Supported by UNICEF and the Thrasher Research Fund. national nutrition survey in Thailand indicated that stunting
* To whom correspondence should be addressed. E-mail: nupwn@mahidol.ac.th. remained at 15.6% among children ,5 y (23). There are no data
0022-3166/06 $8.00 ª 2006 American Society for Nutrition. 2405
Manuscript received 23 December 2005. Initial review completed 7 February 2006. Revision accepted 1 June 2006.on zinc deficiency in Thai infants, but 34% of NE Thai school- were given between meals to avoid interference from the absorption of
aged children are found to be marginally zinc deficient (serum food. The supplement was given daily for a duration of 6 mo. The village
zinc ,9.9 mmol/L) (24). Similarly, data on iron deficiency anemia health volunteers were trained and visited each family 2–3 times/wk to
in infants are scarce, but anemia among pregnant women and identify problems associated with the ingestion of the supplement, to
encourage mothers to administer the supplements, and to keep a record
school-aged children has been a public health problem (25).
of infants’ illness (symptoms). Every week, 4 field workers were assigned
We implemented a community-based, randomized controlled to visit each village to check on each mother and to assist village health
trial that provided 6 mo of supplementation to 4 to 6-mo–old volunteers in monitoring the compliance and morbidity data. Supple-
infants in rural NE Thailand. The daily supplementation of iron ment compliance was recorded by mothers or village health volunteers
and/or zinc among these rural Thai infants was conducted to test onto a monthly calendar provided with each supplement bottle. These
the hypothesis that supplementation of iron or zinc alone, or records were reviewed weekly by field workers. At the monthly field visit,
iron and zinc combined, can improve iron and zinc status and the research team measured the leftover syrup collected by the village
growth of infants 4–6 mo of age compared with infants receiving health volunteers and dispensed new bottles to be distributed to the
a placebo. The interactions between iron and zinc on iron and mothers.
zinc status indicators were also examined. This article reports Data collection
the efficacy of these supplements on biochemical and anthropo- Socio-demographics, health, feeding practices, and morbidity
metric outcomes. status. A culturally appropriate questionnaire, which included socio-
Downloaded from https://academic.oup.com/jn/article/136/9/2405/4664952 by guest on 28 February 2021
economic status, household income, maternal age, weight, height, and
medicinal supplements that mothers gave to their infants during the
Materials and Methods month prior to the intervention, was completed by the mothers at
enrollment. Once each month, the research data collection team visited
Study site and study design the field area. Data on the infant’s health and feeding practices (breast-
The study sites were 3 rural districts of Khon-Kaen Province, located feeding frequency and intake of complementary food) were also col-
;450 km northeast of Bangkok. These districts are similar to other rural lected. During the monthly field visit, morbidity records from the village
northeast districts, based on socio-demographic profiles for population, health volunteers were verified with the mother by one of the
village, and household size. A total of 106 villages were included in the investigators. When applicable, morbidity was also verified with the
study. Most of the inhabitants were engaged in subsistence agriculture. health card at the health centers or district hospital.
The study was a randomized 2 3 2 factorial, double-blind, placebo-
controlled trial, with infants receiving iron-only, zinc-only, iron plus zinc
Anthropometry. Anthropometric measurements were performed on
supplements, or a placebo. A sample size of 130 infants/group was cal-
each infant using standard procedures (26). The child’s body weight
culated to detect a difference in length-for-age (LAZ)2 of 0.15 (with a SD
(wearing light clothing) was measured to the nearest 5 g using a portable
of 0.9) with 95% CI and power of 0.90 among the groups, or to detect a
beam balance (Detecto Infant Scale). Recumbent length was measured to
difference in LAZ of 0.11 when using factorial analysis. Allowing for a
the nearest 0.1 cm using a length of wooden board with a sliding foot
dropout of 15%, a sample size of 150/group was included. The study was
piece. Anthropometry was assessed every 2 mo by 2 field researchers who
approved by the Human Ethics Committee of Mahidol University.
were standardized periodically during the data collection. The coefficient
of reliability, based on technical error of measurement (27), was 0.98.
Subject recruitment and randomization to treatment
Two readings were taken at each measurement. For all infants, LAZ,
Lists of infants aged ,6 mo were obtained from the local health personnel
weight-for-age Z-scores (WAZ), and weight-for-length Z-scores (WLZ)
and village health volunteers. When eligible infants reached the age of
were calculated (EpiInfo 6.02) and based on the National Center for
4–6 mo, a verbal informed consent was obtained from the mothers or
Health Statistics of 1977 (28). Birth weight data were retrieved from the
caretakers. The same information about the project was explained a second
baby’s health card. Most of the births in the study area occurred at the
time and the verbal informed consent verified. Eligibility criteria were that
district hospital where birth weight was measured at the time of birth
infants were predominantly breast fed and free from apparent congenital
and recorded on the baby’s health card. Gestational age was estimated by
abnormalities and that their parents agreed to participate in the study.
recalling the last menstrual period and the completed weeks of gestation
Infants having hemoglobin (Hb) ,80 g/L, chronic illnesses, or who were
were recorded. Once a woman was suspected of being pregnant by the
bottle fed were excluded. A total of 675 infants (339 boys and 336 girls)
village health volunteers, she was invited to meet with the research team
aged 4–6 mo were enrolled in the study. Infants were stratified by age and
at monthly field visits to obtain the earliest possible recall of the last
sex and then random numbers were used to assign individual infants to
menstrual period.
supplemental groups. The randomization was done by a statistician who
was not involved in the study. Blood collection
A blood sample was taken from a subset of infants (whose mothers
The supplements and monitoring of compliance agreed to blood collection) at baseline and the end of the 6-mo
Supplements were made by PT Kenrose as a syrup and supplied by supplementation period. Nonfasting blood samples were taken at 800
UNICEF of Indonesia. The bottles of syrup were coded at the production and 1400 h. Three mL of blood was obtained from the antecubital vein,
site. The code allocation was kept at UNICEF, Jakarta, until the end of using regular syringes. Random samples of syringes from each batch
data analysis. The syrup contained either 5 g/L (89.5 mmol/L) iron (as were tested for contamination by iron and zinc and were found to be
ferrous sulfate), 5 g/L (76.5 mmol/L) zinc (as zinc sulfate), or both iron uncontaminated. All glassware, plasticware, and microcontainer tubes
and zinc in the same concentrations. The 4 types of supplements were were acid-washed. One mL of blood was put in a tube containing the
indistinguishable in appearance, color, taste, or odor. All supplements anticoagulant, EDTA, and 2 mL were allowed to clot and then were
contained vitamin C (15 g/L). Vitamin A drops at a dose of 1500 mg centrifuged for serum.
retinol equivalent (RE) were given to each subject at the beginning of the Blood samples were stored in a cool box immediately after the blood
study. Each mother was provided with 2 bottles of syrup and a calendar was taken. Separation of serum was done at the hospital laboratory and
to record each time she administered the syrup. Mothers were carefully completed by 1500 of each field visit day. Samples were frozen at 220C
instructed, in the presence of village health volunteers from respective prior to analyses for zinc and ferritin. Hemoglobin was determined by
communities, to administer the supplement using an oral medication the cyanmethemoglobin method (29), and serum ferritin by enzyme-
syringe (2 mL/dose, equivalent to 10 mg of iron or zinc/d). Supplements linked immunoassay (30,31), using the ES-33 system (Boehringer
Mannheim). Serum zinc was measured by graphite furnace atomic
2
Abbreviations used: WAZ, weight-for-age Z-score; LAZ, length-for-age Z-score; absorption spectrophotometry (WL 213.9, current 15 mA and slit high
WLZ, weight-for-length Z-score; Hb, hemoglobin, ID, iron deficiency; IDA, iron 0.7) (32). For both ferritin and zinc, the initial and final samples were
deficiency anemia; RE, retinol equivalent. analyzed in the same batch. Certified reference materials were used for
2406 Wasantwisut et al.Hb, serum ferritin, and zinc. CBC-8 hematology controls (from R&D
Systems) were used for quality control of Hb, and had the following
values (means 6 SD): low, 61 6 3; normal, 134 6 4; and high, 178 6 5 g/
L. The mean (SD, %CV) for ferritin references were as follows: low, 12.8
(1.11, 3.1%); normal, 90.2 (5.5, 4.0%); and high, 550 (25.0, 5.2%) mg/L.
For serum zinc (using the Bovine Serum Reference Material 1598,
National Institute of Standards and Technology) the means 6 SD value
was 13.6 6 0.9 mmol/L vs. a pooled serum mean (SD, %CV) of 13.6
mmol/L (0.1, 0.89%).
Because there was no consensus on the cutoff for defining anemia,
iron deficiency, and iron deficiency anemia for young infants, the fol-
lowing were used: whereas WHO recommended Hb ,110 g/L to define
anemia in children ,5 (33), Domellöf et al. (34) proposed age-specific
Hb cutoffs for defining anemia with ,105 g/L for 4- to 6-mo–old infants
and ,100 g/L for 9-mo–old infants. However, there were no data for
10- to 12-mo olds. In the present study, infants were measured at 4–6 mo
at baseline, and at 10–12 mo at the end point. Thus, anemia was defined
as Hb either below the age-specific cutoffs from Domellöf et al., Hb Figure 1 Trial profile.
Downloaded from https://academic.oup.com/jn/article/136/9/2405/4664952 by guest on 28 February 2021
,105 g/L at 4–6 mo of age and Hb ,100 g/L at 10–12 mo, or at the
WHO recommended cutoff of ,110 g/L for infants. In addition, the age-
specific cutoffs for ferritin were also proposed (,20 mg/L for 4 mo, ,9 Background characteristics. The majority of the mothers had
mg/L for 6 mo, and ,5 mg/L for 9 mo). However, only a few infants a primary education, and 20–27% had a secondary education.
at these ages in the present study had ferritin ,12 mg/L. Therefore, Their incomes were derived from both agricultural and nonag-
2 cutoffs were used for defining iron deficiency (ID), namely, ferritin ,12 ricultural activities. Maternal age, height, parity, education, and
or ,20 mg/L. For iron deficiency anemia (IDA), the 2 cutoff systems for income did not differ among the groups. Baseline characteristics
Hb and ferritin, ,12 or ,20 mg/L, were used. For serum zinc, the cutoff did not differ among the 4 groups (Table 1). The mean age of
of ,9.9 mmol/L and ,10.7 mmol/L for the fasting and nonfasting states infants at baseline was 4.5 mo. The incidence of low birth weight
were used, respectively (35). and preterm delivery (gestational age ,37 wk) was 9 and 11%,
respectively. The prevalence of low birth weight or preterm
Statistical analysis delivery did not differ among groups. Overall, the study infants
Data were checked for normality, using the Kolmogorov-Smirnov test of were apparently healthy with only 3% stunted (LAZ ,22,
normality. Because of skewed data on serum concentrations of ferritin NCHS) and none were underweight. The LAZ of infants in the
and zinc, these data were transformed to natural logarithms. Differences
zinc group tended to be worse than other groups (P ¼ 0.055).
among groups were analyzed with ANOVA for continuous variables and
the chi-square test for prevalence. Where there was a significant effect,
multiple comparisons were performed using the Bonferroni post-hoc test. Biochemical indicators. The mean concentrations of hemo-
The estimated-effect sizes were analyzed using the postintervention globin and zinc, and the median concentration of ferritin in
values as the dependent variable in an ANCOVA model, in which the serum, did not differ among the groups at baseline (Table 2).
baseline values were included as a covariate. Gender and/or birth weight After 6 mo of supplementation, infants receiving iron supple-
were included in the model as potential confounders. Similar analysis ments (iron and iron plus zinc groups) had significantly higher
procedures were performed for both biochemical and anthropometric hemoglobin concentrations than those receiving a placebo or
parameters. Main effects were considered significant at P , 0.05, and in- zinc (Table 2). Infants not receiving iron had a decrease in
teractions were considered significant at P , 0.1. Differences were con- hemoglobin concentrations of 11.0 g/L between baseline and
sidered significant at P , 0.05. SPSS, version 11.5, was used for all
6 mo (P , 0.001). The declines in hemoglobin concentrations in
statistical analysis.
the placebo and zinc groups might have been due to the lack of
supplemental iron. The overall effect of iron supplementation on
end point hemoglobin concentrations was an increase of 10.8
Results (95% CI: 8.0, 13.6) g/L, whereas the effect of zinc supplemen-
Trial profile. In total, 609 infants completed the study and were tation with or without iron was a decrease in hemoglobin
included in the analysis. There were 9, 13, 24, and 19 infants concentrations by 3.4 g/L (95% CI: 26.3, 0.6). The interaction
from the placebo, iron, zinc, and iron plus zinc groups,
respectively, who did not complete the study (Fig. 1). Reasons TABLE 1 Baseline characteristics of infants who completed
for dropout during the study included health problems (n ¼ 2), the study1
moving (n ¼ 49), and side effects attributed to the supplements
(severe diarrhea, n ¼ 3). Infants whose mothers introduced Placebo Iron Zinc Iron and zinc
bottle feeding during the study (n ¼ 4) and those not cooperative Subjects, n 153 153 151 152
(n ¼ 5) were excluded from the analysis. Infants not completing Male:Female, %:% 52:48 50:50 51:49 50:50
the study did not differ from those who completed the study in Age, mo 4.5 6 0.5 4.5 6 0.6 4.5 6 0.6 4.4 6 0.5
any of the characteristics at baseline. Infants who had biochem- Birthweight, g 3097 6 414 3057 6 417 3045 6 469 3061 6 365
ical data did not differ from those without these data in their sex,
Anthropometry
age, and anthropometric measurements, both at study entry or
Length, cm 62.6 6 2.2 62.2 6 2.5 62.0 6 2.3 61.9 6 2.1
exit. The median compliance was .90% [median (range) of 96
Weight, kg 6.6 6 0.8 6.6 6 0.8 6.5 6 0.8 6.4 6 0.7
(57–100), 96 (70–100), 95 (70–100), and 95 (71–100) for
LAZ 20.66 6 0.75 20.66 6 0.72 20.73 6 0.74 20.65 6 0.65
placebo, iron, zinc, and iron plus zinc groups, respectively].
WAZ 20.14 6 0.86 20.08 6 0.82 20.22 6 0.82 20.17 6 0.74
More than 95% of infants received .80% of intended doses of
WLZ 0.43 6 0.79 0.51 6 0.77 0.41 6 0.79 0.39 6 0.79
each of the supplements. Supplement compliance among the
groups did not differ. 1
Values are means 6 SD, n ¼ 609.
Iron and zinc supplementation of breast-fed infants 2407TABLE 2 Biochemical indicators of infants at baseline and after 6-mo supplementation with iron, zinc, or both
Estimated effect size Interaction
Placebo Iron Zinc Iron and zinc Iron, 95% Cl Zinc, 95% Cl Iron and zinc, P-value
Subjects, n 66 65 58 67
Hemoglobin,1,*g/L
Baseline 116.7 6 11.6 118.1 6 11.2 113.4 6 11.5 114.2 6 11.8 10.8y (8.0, 13.6) 23.4y (26.3, 20.6) 0.990
End point 106.3 6 13.0b 118.0 6 12.2a 101.6 6 13.2b 112.9 6 11.4a
Serum ferritin,2,* mg/L
Baseline 65.2 (38.5, 101.2) 81.1 (42.5, 115.4) 70.4 (40.9, 99.2) 80.4 (51.9, 113.3) 31.8y (27.1,36.6) 28.5y (0.10, 0.14) 0.044
End point 24.2c (18.7, 37.3) 69.6a (47.1, 115.4) 24.4c (15.6, 45.5) 50.7b (30.9, 98.7)
Serum zinc,1,*mmol/L
Baseline 11.1 6 2.7 10.4 6 2.5 11.5 6 2.9 11.3 6 2.3 20.8y (1.2, 1.3) 3.0y (2.8, 3.2) ,0.001
End point 9.8 6 1.9c 11.5 6 2.5b 16.7 6 5.2a 12.1 6 3.8b
Downloaded from https://academic.oup.com/jn/article/136/9/2405/4664952 by guest on 28 February 2021
1
Values are means 6 SD,
2
Values are medians [IQR (inter-quartile range, calculated by 75th percentile 2 25th percentile)].
* For Hb and serum zinc: ANOVA, significant difference among groups, P , 0.001. For ferritin: Kruskal Wallis test, significant difference among groups, P , 0.001. For post-hoc,
Bonferroni tests, values in a row with superscripts without a common letter differ, P , 0.05.
y
Significant difference by ANCOVA controlled for gender, P , 0.05.
between the 2 nutrients on hemoglobin concentrations was not zinc groups. On the contrary, the prevalence of IDA increased
significant. markedly in the other 2 (no iron) groups, especially among infants
Baseline ferritin concentrations did not differ among the receiving only zinc (15.4 and 27.6%, respectively). Overall, the
groups (P ¼ 0.201) but, after 6-mo of supplementation, the 2 prevalence of ID and IDA decreased among infants receiving iron
groups receiving iron supplementation (iron or iron plus zinc) alone or iron combined with zinc more than among infants in the
had significantly higher ferritin concentrations than those placebo or zinc group. Similar patterns in the prevalence of IDA
receiving only zinc or placebo (P , 0.001). There also was an from baseline among groups were observed using the WHO
interaction between iron and zinc supplementation (P ¼ 0.044). cutoff for Hb (.110 g/L for infants of all ages).
With iron supplementation, ferritin concentration increased by The reduction in the prevalence of low serum zinc concen-
31.8 mg/L (95% CI: 27.1, 36.6), but a decrease of 8.5 mg/L (95% trations was substantial and clearest in the infants supplemented
CI: 0.10, 0.14) occurred with zinc supplementation. with zinc only (,2% had low serum zinc based on both cutoffs).
Similarly, serum zinc concentrations (P , 0.001) differed A decrease in prevalence of low serum zinc also occurred
among groups at the end of the study. Infants receiving zinc had in infants receiving iron only, defined either as ,10.7 or
the highest serum zinc concentration (P , 0.001), but the zinc ,9.9 mmol/L, whereas the prevalence in the iron plus zinc
concentrations of the 2 groups that received iron did not differ, group was almost the same as in the iron group. The prevalence
and both were higher than that of the placebo group. The of low serum zinc was higher in the placebo group at the end of
interaction between iron and zinc supplementation on serum supplementation than in all other groups, regardless of the
zinc concentration was significant (P , 0.001). The overall cutoffs used.
effect of iron supplementation on end point serum zinc concen-
trations was a decrease of 0.8 mmol/L, compared with an Growth. Infants gained 1.59 6 0.49 kg and increased 8.5 6 1.4 cm
increase of 3.0 mmol/L from zinc supplementation (P , 0.001). in length. Supplementation did not affect the linear growth of
infants; neither LAZ nor length differed among the groups at the
Prevalence of anemia, ID, IDA, and zinc deficiency. At end of supplementation. Even when only infants with LAZ
baseline, the overall prevalence of anemia was 18% using the below –1.5 at recruitment (n ¼ 67) were considered, there were
age-specific cutoff of Hb ,105 g/L and 28.8% using the WHO no significant effects of supplementation.
cutoff of Hb ,110 g/L; groups did not differ for either cutoff When data were analyzed by ANCOVA, controlled for gender
(P . 0.1) (Table 3). Reduced anemia prevalence was greater in and birth weight, iron supplementation significantly improved
infants receiving only iron than in infants receiving iron and zinc ponderal growth, with an estimated effect size of 80 g over the
but, in both groups, the reduction was more than the placebo 6-mo supplementation period (P ¼ 0.034). Similarly, the WLZ
group (P , 0.05). Infants receiving only zinc had a greater were improved by iron supplementation (P ¼ 0.003). Zinc
prevalence of anemia than the placebo group when both cutoffs, supplementation did not improve any growth indicators. Zinc
105/100 g/L and 110 g/L, were used. and iron tended to interact to affect the LAZ (P ¼ 0.103) but not
Despite the high prevalence of anemia in most groups, at the other anthropometric indicators (Table 4).
baseline, very few infants had ID and IDAwhen ferritin ,12 mg/L
was used; but the prevalence became more substantial when a
higher cutoff for ferritin (,20 mg/L) was used. Using age-specific
Discussion
cutoffs for Hb, there was no IDA (Hb ,105 1 ferritin ,12 mg/L)
in all groups at baseline. However, at postintervention, the Despite global concerns about micronutrient deficiencies and
prevalence of IDA (Hb ,100 1 ferritin ,12 mg/L) increased in their health and functional consequences among young children,
the placebo group and was especially increased in the group with data in Thailand are scarce due to limitations in blood
zinc only. Using a higher ferritin cutoff with age-specific Hb examination. Iron deficiency anemia is known to be prevalent
cutoffs, IDA at baseline was corrected in the iron and the iron plus in developing countries, but data on zinc deficiency in young
2408 Wasantwisut et al.TABLE 3 Prevalence of anemia, ID, IDA, and zinc deficiency mothers and village health volunteers on their compliance
in infants at baseline and after 6-mo supplementation calendar, infant morbidity was low. The use of a more definitive
with iron, zinc, or both1,2 indicator to rule out or confirm inflammation, such as C-reactive
protein, would provide a clearer explanation of the situation.
Placebo Iron Zinc Iron1Zinc The low morbidity and the reasonably good hygienic condition
Subjects, n 66 65 58 67
of the community, however, suggested that infection might not
%
be a problem. Finally, vitamin A is known to be a significant
determinant of Hb, especially among school-aged children in NE
Anemia
Thailand (37). Although Vitamin A supplementation was given
Age-specific cutoff 3
at baseline, the dose might not have been adequate to maintain
Baseline, Hb ,105 g/L 15.2 12.3 22.4 25.4
serum retinol level during the entire 6-mo period of this study.
End point, Hb ,100 g/L 33.3 4.6 41.4 13.4
Hence, vitamin A deficiency could underlie the anemia observed
WHO cutoff, Hb ,110 g/L
in infants at the end point of supplementation.
Baseline 27.3 23.1 31.0 38.8
Regardless of the ferritin cutoffs (12 or 20 mg/L) used, an
End point 60.6 16.9 69.0 41.8
improvement in the prevalence of ID and IDA after iron sup-
Iron deficiency plementation was clearly observed, compared with the placebo
Downloaded from https://academic.oup.com/jn/article/136/9/2405/4664952 by guest on 28 February 2021
Serum ferritin, ,12mg/L group, which suggests the benefit of iron supplementation in
Baseline 1.5 0 3.4 0 reducing anemia in late infancy. Infants receiving the placebo or
End point 6.1 0 15.5 3.0 zinc had a decrease in hemoglobin concentrations of .10g/L
Serum ferritin, ,20mg/L over the 6-mo supplementation period, whereas infants receiv-
Baseline 4.5 6.2 5.2 4.5 ing iron alone or iron combined with zinc had no decrease, or
End point 33.3 4.6 41.4 10.4 only a small decrease, in hemoglobin concentrations. Hemoglo-
Iron deficiency anemia bin concentrations substantially change during infancy, and the
Age-specific cutoff for Hb 1 ferritin ,12 mg/L current cutoffs for anemia have been recently questioned (34).
Baseline, Hb ,105g/L 1 ferritin ,12mg/L 0 0 0 0 The prevalence of anemia remains high without iron sup-
End point, Hb ,100g/L 1 ferritin ,12mg/L 3.1 0 13.8 0 plementation. The present study shows, however, that infants
Hb ,110 g/L 1 ferritin ,12 mg/L can sustain hemoglobin concentrations when they are provided
Baseline 1.5 0 0 0 with sufficient iron. Moreover, the marked decline in ferritin and
End point 6.2 0 15.5 1.5 zinc concentrations in the placebo group between recruitment
Age specific cutoff for Hb 1 ferritin ,20mg/L and end of the study clearly shows that iron or zinc status can be
Baseline, Hb ,105g/L 1 ferritin ,20mg/L 0 6.2 0 3.0 compromised in infants who do not receive iron or zinc during
End point Hb ,100g/L 1 ferritin ,20mg/L 15.4 0 27.6 0 the second half of infancy.
Hb ,110 g/L 1 ferritin ,20 mg/L Furthermore, although zinc supplementation was efficient in
Baseline 4.5 6.2 1.7 3.0 improving zinc status, the addition of iron substantially reduced
End point 25.8 1.5 36.2 7.5 this effect. On the contrary in our study, we found a beneficial
Zinc deficiency effect of iron supplementation on zinc status, with serum zinc
Serum zinc, ,9.9 mmol/L concentrations significantly higher in the infants receiving iron
Baseline 27.7 46.2 34.5 25.4 than infants receiving a placebo. Conflicting findings on the
End point 51.5 32.3 1.7 32.8 beneficial effect of iron supplementation on zinc status have
Serum zinc ,10.7 mmol/L been reported (38,39). These opposite findings may be related to
Baseline 47.4 63.1 44.8 46.3 the extent of the deficiency. Similar findings were reported where
End point 66.7 44.6 1.7 43.3 the level of deficiency was nearly the same as in the present study
(38) but not where the deficiency was more severe (39). Both of
1
Prevalence at baseline did not differ among groups. these studies had the same study protocol. The underlying
2
End point values among groups differed, P , 0.05.
3
Age-specific cutoffs: Hb ,105 g/L at 4–6 mo old; Hb ,100 at 9 mo old was adopted
mechanisms by which iron supplementation might improve zinc
for use also at 10–12 mo old (34). status are unclear, however.
When zinc was given in combination with iron, hemoglobin
concentrations were lower than when only iron was supple-
children are limited. To our knowledge, this study is the first to mented. The effect of iron supplementation in improving ferritin
show that zinc deficiency is a problem among infants. Using a was also reduced when zinc was given in combination with iron,
conservative cutoff of 9.9 mmol/L (35), more than half of the which is consistent with findings reported earlier from Indonesia
infants in the placebo group were zinc deficient. (38,39) that suggest zinc supplementation might interfere with
ID and IDA were surprisingly low, given the high prevalence the transport of iron into the intestinal cells. However, Kordas
of anemia (Hb ,110 g/L at .20%). In this region, hemoglo- and Stoltzfus (40) examined studies using Caco-2 cells and
binopathies, especially heterozygous E, were reported to be as concluded that zinc does not depend on intestinal divalent metal
high as 35–40% among pregnant women and school-aged transporter, and hence does not compete with iron through this
children (36,37). This could partially explain our finding. Serum mechanism.
ferritin concentrations did not differ among normal Hb-type The finding that infants receiving iron had higher serum
children or among those with heterozygotes E (HbAE) but were ferritin than those receiving placebo suggests the need for
significantly higher among children with homozygotes E (37). additional iron from iron-rich sources during the second half of
Because the predominant form of abnormal Hb in this region is infancy. The substantial increase in the prevalence of anemia and
HbAE (;85–90%), the relatively high ferritin found among iron deficiency anemia in the infants receiving only zinc,
infants in this study should not be attributed to this condition. In compared with the placebo group, is a concern. This was also
addition, according to the illness-symptoms records kept by consistent with a study conducted in Indonesia by Dijkhuizen
Iron and zinc supplementation of breast-fed infants 2409TABLE 4 Growth of the infants after 6-mo supplementation with iron, zinc, or both1
Estimated effect size Interaction
Placebo Iron Zinc Iron and zinc Iron, 95% Cl Zn, 95% CI Iron and zinc, P-value
Subjects, n 153 153 151 152
Age, mo 10.5 6 0.5 10.5 6 0.5 10.5 6 0.5 10.4 6 0.6
Length, cm 70.6 6 2.4 70.6 6 2.5 70.6 6 2.5 70.4 6 2.2 20.04 (20.24, 0.16) 0.04 (20.16, 0.25) 0.465
Weight, kg 8.1 6 1.0 8.2 6 0.9 8.0 6 0.9 8.1 6 0.9 0.08y (0.01, 0.16) 0.03 (20.05, 0.11) 0.636
LAZ 20.99 6 0.86 21.0 6 0.8 21.0 6 0.9 21.0 6 0.7 20.03 (20.10, 0.03) 20.01 (20.07, 0.06) 0.103
WAZ 21.3 6 0.9 21.2 6 0.9 21.3 6 0.9 21.3 6 0.8 0. 06 (20.01, 0.14) 0.01 (20.06, 0.08) 0.254
WLZ 20.7 6 0.7 20.5 6 0.7 20.7 6 0.8 20.6 6 0.8 0.121y (0.04, 0.20) 0.002 (20.08, 0.08) 0.709
1
Values are means 6 SD.
y
ANCOVA analysis, controlled for gender and birth weight, P , 0.05.
Downloaded from https://academic.oup.com/jn/article/136/9/2405/4664952 by guest on 28 February 2021
et al. (38). From a public health point of view, this means that may outweigh the risk of functional consequences associated
providing zinc without iron might cause a significant increase in with these deficiencies.
iron deficiency with possible detrimental effects on psychomotor
development. The apparent strong negative effect of zinc on
hemoglobin concentrations has not been reported in children of Acknowledgments
another age range. Two studies in Indonesian infants also We thank Dr. Frank Wieringa and Dr. Marjoleine Dijkhuizen
demonstrated an increase in the prevalence of iron deficiency for advice on statistical analysis and structure of the article.
anemia in infants receiving only zinc compared with iron-
supplemented infants. In contrast, zinc supplementation in older
Mexican children did not result in an increase in the prevalence
of iron deficiency (41). Literature Cited
Despite the low serum zinc concentrations observed in this 1. Tomkins A. Malnutrition, morbidity and mortality in children and their
study, no association of zinc deficiency and stunting was found mothers. Proc Nutr Soc. 2000;59:135–46.
(data not shown), and zinc supplementation did not improve 2. Oski FA, Honig AS, Helu B, Howanitz P. Effect of iron therapy on
linear or ponderal growth. The beneficial effect of iron on behavior performance in nonanemic, iron- deficient infants. Pediatrics.
1983;71:877–80.
ponderal growth shows that, although only 6% of the infants in
3. Walter T, De Andraca I, Chadud P, Perales CG. Iron deficiency anemia:
the placebo group had iron deficiency anemia according to adverse effects on infant psychomotor development. Pediatrics. 1989;
standard cutoffs, iron was beneficial for these infants. 84:7–17.
It is important to note that other micronutrients and possibly 4. Lozoff B, Brittenham G, Wolf AW, McClish DK, Kuhnert PM, Jimenez
macronutrient deficiencies may exist (38). Although the infants E, Jimenez R, Mora LA, Gomez I, Krauskoph D. Iron deficiency anemia
appeared to have adequate growth at baseline (i.e., #4–6 mo of and iron therapy effects on infant developmental test performance.
Pediatrics. 1987;79:981–95.
age), the rapid decline of anthropometric status at the end point
5. Idjradinata P, Pollitt E. Reversal of developmental delays in iron
is a concern. Whereas the prevalence of stunting increased from
deficient anemic infants treated with iron. Lancet. 1993;341:1–4.
3 to 11%, the prevalence of WAZ below 22 was as high as 20%
6. Walravens PA, Hambidge KM. Growth of infants fed a zinc
compared with only 0.2% at baseline. The lowest prevalence of supplemented formula. Am J Clin Nutr. 1976;29:1114–21.
stunting and underweight was observed in the iron group at 7. Schlesinger L, Arevalo M, Arredondo S, Diaz M, Lonnerdal B, Stekel A.
the end of the intervention period (9 vs. 11–12% for LAZ and Effect of a zinc-fortified formula on immunocompetence and growth of
15.7 vs. 18–23% for WAZ in other groups) (data not shown). malnourished infants. Am J Clin Nutr. 1992;56:491–8.
This pattern clearly indicates that supplementation with only 2 8. Chen XC, Yin TA, He JS, Ma QY, Han ZM, Li LX. Low levels of zinc in
micronutrients (as used in this study) does not prevent a faltering hair and blood, pica, anorexia and poor growth in Chinese preschool
children. Am J Clin Nutr. 1985;42:694–700.
in growth. This was also noted in another recent study (42).
9. Gibson RS, Smit Vanderkooy PD, MacDonald AC, Goldman A, Ryan
In conclusion, anemia and iron and zinc deficiencies are a BA, Berry M. A growth limiting, mild zinc deficiency syndrome in some
problem among apparently healthy infants in NE Thailand. southern Ontario boys with low height percentiles. Am J Clin Nutr.
Without iron and/or zinc supplementation, anemia, iron, and 1989;49:1266–73.
zinc deficiency can be aggravated, as found in the placebo group. 10. Brown KH, Peerson JM, Rivera J, Allen LH. Effect of supplemental zinc
Consequently, such deficits may affect crucial developmental on the growth and serum zinc concentrations of prepubertal children: a
meta-analysis of randomized controlled trials. Am J Clin Nutr. 2002;
functions. The low prevalence of ID and IDA warrants addi-
75:1062–71.
tional study to identify causes of anemia among young infants.
11. Umeta M, West CE, Haider J, Deurenberg P. Hautvast, JGAJ. Zinc
The roles of hemoglobinopathy and possible confounding effects supplementation and stunted infants in Ethiopia,: a randomized
of mild infection need to be verified before a definitive controlled trial. Lancet. 2000;355:2021–6.
conclusion can be made for a situation in which mild infection 12. Bentley ME, Caulfield LE, Ram M, Santizo MC, Hurtado E, Rivera JA,
and concurrent marginal micronutrient deficiencies may exist, Ruel MT, Brown KH. Zinc supplementation affects the activity patterns
such as in the present study population. In this study, iron sup- of rural Guatemalan infants. J Nutr. 1997;127:1333–8.
plementation increased ferritin, reduced iron deficiency, iron 13. Sazawal S, Bentley M, Black RE, Dhingra P, George S, Bhan MK. Effect
of zinc supplementation on observed activity in low socioeconomic
deficiency anemia, and overall anemia, improved serum zinc, Indian preschool children. Pediatrics. 1996;98:1132–7.
and benefited ponderal growth, whereas zinc supplementation 14. Black RE. Therapeutic and preventive effects of zinc on serious
improved zinc status. Despite the interaction between iron and childhood infectious diseases in developing countries. Am J Clin Nutr.
zinc, overall, the benefit from providing both micronutrients 1998;68:476s–9s.
2410 Wasantwisut et al.15. Bhutta ZA, Bird SM, Black RE, Brown KH, Meeks Gardner J, Hidayat 29. Cortius HCM, Roth M. Clinical biochemistry principle and methods.
A, Khatun F, Martorell R, Ninh NX, et al. Therapeutic effects of oral New York: Walter de Gruyter Berlin.1974. p.1306–12.
zinc in acute and persistent diarrhea in children in developing countries: 30. Alfrey CP. Serum ferritin assay. CRC Crit Rev Lab Sci. 1978;9:
pooled analysis of randomized controlled trials. Am J Clin Nutr. 2000; 179–203.
72:1516–22.
31. Jacobs A, Miller F, Worwood M, Beamish MR, Wardrop CA. Ferritin in
16. Walker CF, Black RE. Zinc and the risk for infectious disease. Annu Rev the serum of normal subjects and patients with iron deficiency and iron
Nutr. 2004;24:255–75. overload. BMJ. 1972;4:206–8.
17. Roy SK, Tomkins AM, Haider R, Behren RH, Akramuzzaman SM, 32. D’Haese PC, Lamberts LV, Vanheule AO, De Broe ME. Direct
Mahalanabis D, Fuchs GJ. Impact of zinc supplementation on subse- determination of zinc in serum by Zeeman atomic absorption spec-
quent growth and morbidity in Bangladeshi children with acute trometry with a graphite furnace. Clin Chem. 1992;38:2439–43.
diarrhoea. Eur J Clin Nutr. 1999;53:529–34.
33. World Health Organization (WHO). Iron deficiency anaemia assess-
18. Walker CF, Kordas K, Stoltzfus RJ, Black RE. Interactive effects of iron ment, prevention, and control. A guide for programme managers.
and zinc on biochemical and functional outcomes in supplementation Geneva: World Health Organization; 2001.
trials. Am J Clin Nutr. 2005;82:5–12.
34. Domellöf M, Dewey KG, Lönnerdal B, Cohen RJ, Hernell O. The
19. Black MM, Baqui AH, Zaman K, Persson AK, Arifeen SE, Le K, Mc diagnostic criteria for iron deficiency in infants should be reevaluated.
Nary SW, Parveen M, Hamadani JA, Black RE. Iron and zinc J Nutr. 2002;132:3680–6.
supplementation promote motor development and exploratory behavior
35. International Zinc Nutrition Consultative Group (IZiNCG). Technical
among Bangladeshi infants. Am J Clin Nutr. 2004;80:903–10.
document #1. Assessment of the risk of zinc deficiency in populations
Downloaded from https://academic.oup.com/jn/article/136/9/2405/4664952 by guest on 28 February 2021
20. Dewey KG, Brown KH. Update on technical issues concerning and options for its control. Food Nutr Bull. 2004;25:S130–62.
complementary feeding of young children in developing countries and
36. Charoenlarp P, Dhanamitta S, Keawvichit R, Silprasert A, Suwanaradd
implications for intervention programs. Food Nutr Bull. 2003;24:5–28.
C, Na-Nakorn S, Prawatmuang P, Vatanavicharn S, Nutcharas U,
21. Tontisirin K, Winichagoon P, Chavasit V, Yamborisut U. Supplementary Pootrakul PA. WHO collaborative study on iron supplementation in
feeding. In: Winichagoon et al., editors. Integrating food and nutrition Burma and in Thailand. Am J Clin Nutr. 1988;47:280–97.
into development. Thailand’s experiences and future visions. Mahidol
37. Thurlow RA, Winichagoon P, Green T, Wasantwisut E, Pongcharoen T,
University, Thailand: Institute of Nutriton; 1992. p. 133–45.
Baily KB, Gibson RS. Only a small proportion of anemia in northeast
22. Chusilp K, Somnasang P, Kridpon W, Wongkham S, Sribonlue P, Thai schoolchildren is associated with iron deficiency. Am J Clin Nutr.
Mahaverawat U, Yongvanit P, Sawakontha S, Waterlow J. Observation 2005;82:380–7.
on the development of stunting in children of the Khon Kaen region of
38. Dijkhuizen MA, Wieringa FT, West CE, Martuti S, Muhilal. Effects of
Thailand. Eur J Clin Nutr. 1992;46:475–87.
iron and zinc supplementation in Indonesian infants on micronutrient
23. Division of Nutrition, Department of Health, Ministry of Public Health. status and growth. J Nutr. 2001;131:2860–5.
The 4th National Nutrition Survey. 1995.
39. Lind T, Lönnerdal B, Stenlund H, Ismail D, Seswandhana R, Ekström
24. Udomkesmalee E, Dhanamitta S, Yhoung-aree J, Rojroongwasinkul N, EC, Persson LÅ. A community-based randomized controlled trial of
Smith JC. Biochemical evidence suggestive of suboptimal zinc and iron and zinc supplementation in Indonesian infants: interactions
vitamin A status in school children in Northeast Thailand. Am J Clin between iron and zinc. Am J Clin Nutr. 2003;77:883–90.
Nutr. 1990;52:564–7.
40. Kordas K, Stoltzfus RJ. New evidence of iron and zinc interplay at the
25. Winichagoon P. Prevention and control of anemia: Thailand experi- enterocyte and neural tissues. J Nutr. 2004;134:1295–8.
ences. J Nutr. 2002;132:862s–6s.
41. Rosado JL, Lopez P, Munoz E, Martinez H, Allen LH. Zinc
26. Gibson RS. Principles of nutritional assessment. New York: Oxford supplementation reduced morbidity, but neither zinc nor iron supple-
University Press. 1990. p.163–86. mentation affected growth or body composition of Mexican pre-
27. Ulijaszek SJ, Kerr DA. Anthropometric measurement error and the schoolers. Am J Clin Nutr. 1997;65:13–9.
assessment of nutritional status. Br J Nutr. 1999;82:165–77. 42. Ramakrishnan U, Aburto N, McCabe G, Martorell R. Multimicronu-
28. Hamill PV, Drizd TA, Johnson CL, Reed RB, Roche AF, Moore WM. trient interventions but not vitamin A or iron interventions alone
Physical growth: national center for health statistics percentiles. Am J improve child growth: results of 3 meta-analyses. J Nutr. 2004;134:
Clin Nutr. 1979;32:607–29. 2592–602.
Iron and zinc supplementation of breast-fed infants 2411You can also read
