This is the old United Nations University website. Visit the new site at http://unu.edu
References
Kornelia Buzina-Suboticanec, Ratko Buzina, Ana Stavljenic, Meri Tadinac-Babic, and Vesna Juhovic-Markus
Kornelia Buzina-Suboticanec and Ratko Buzina are affiliated with the Institute of Public Health in Zagreb, Croatia. Ana Stavljenic is affiliated with the School of Medicine in the University of Zagreb, and Meri Tadinac-Babic is affiliated with the Faculty of Philosophy in the University of Zagreb. Vesna Juhovic-Markus is affiliated with the Medical Center in Velika Gorica, Croatia.
Mention of the names of firms and commercial products does not imply endorsement by the United Nations University.
Editorial introduction
Moderate to severe anaemia in infancy has been shown to have a lasting impact on cognitive performance [1, 2], but if the anaemia in infancy has been mild, the effects are reversible. Iron deficiency at any age has been shown to have adverse effects on cognitive performance [3], with the response to iron supplementation depending on the circumstances. Studies of adolescent girls in Pennsylvania in the United States found an adverse effect of iron-deficiency anaemia [4, 5]. The present study is significant because it demonstrates that even in a relatively well-nourished population, at least with normal weight-for-age, iron deficiency still has a highly significant effect on cognitive performance that can be corrected by iron supplementation. Although not discussed in this paper, iron deficiency also has an adverse effect on immune competence and resistance to infection and on physical work capacity [6]. The important message is that iron deficiency should be prevented wherever it occurs, whether in developing or industrialized-country populations.
Nevin S. Scrimshaw
References
1. Pollitt E, Gorman KS, Engle PL, Martorell R, Rivera J. Early supplementary feeding and cognition. Monographs of the Society for Research in Child Development. Serial 235. 1993, 58(7).
2. Lozoff B, Jimenez E, Wolf AW. Long-term developmental outcome of infants with iron deficiency. N Engl J Med 1991;325:687-95.
3. Scrimshaw NS. Malnutrition, brain development, learning, and behavior. Nutr Res 1998;18:351-79.
4. Brunner AB, Joffe A, Duggan AK, Casella JF, Brandt J. Randomized study of cognitive effects of iron supplementation in non-anaemic iron-deficient adolescent girls. Lancet 1996;348:992-6.
5. Beard JL, Vernon-Feagans L, Piñero D, Whitfield K. Iron deficiency and cognitive performance in teen mothers. Appl Dev Sci (in press).
6. Scrimshaw NS. Functional significance of iron deficiency. In: Enwonwu CE, ed. Functional significance of iron deficiency. Annual nutrition workshop series. Vol III. Nashville, Tenn, USA: Center for Nutrition, Meharry Medical College, 1990:1-13.
Abstract
This study examined the effect of iron supplementation on cognitive function by a double-blind intervention trial in nine-year-old mildly anaemic schoolchildren. Their nutritional status was assessed by anthropometric measurements and the following biochemical values: haemoglobin, haematocrit, red blood cell count (RBC), mean corpuscular haemoglobin (MCH), mean corpuscular volume (MCV), mean corpuscular haemoglobin concentration (MCHC), serum iron, total iron-binding capacity (TIBC), and transferrin saturation. In addition, biochemical values of vitamin A, vitamin C, thiamine, riboflavin, pyridoxine, and zinc were measured. The cognitive assessment was performed using an abbreviated Wechsler Intelligence Scale for Children-Revised (WISC-R) containing six subtests: arithmetic, similarities, digit span, picture completion, block design, and digit symbol (coding), in order to obtain information on both verbal and non-verbal aspects of intelligence. There were highly significant correlations of the WISC-R scores with initial height-for-age, haemoglobin, haematocrit, and transferrin saturation, and a correlation with MCHC. After completion of the baseline examination, one group of children was given a supplement containing 100 mg of iron for 10 weeks while the other group received a placebo. Iron supplementation had a positive effect on the biochemical measures of iron status, with haemoglobin, haematocrit, transferrin saturation, RBC, MCH, and MCHC all showing statistically significant increases (p<.05). Iron supplementation also resulted in a statistically significant improvement in total WISC-R score (p<.01). This effect was primarily the result of improved performance on nonverbal subtests, of which improvements in block design and coding were statistically significant (p<.01). The small increase in the sum of scaled scores from the verbal subtest was not significant (p>.05), but within the verbal subtest there was a significant improvement on the similarities part of the test (p<.05). The effects of iron supplementation were more pronounced in children with initially lower haemoglobin values. It is concluded that iron supplementation in nine-year-old schoolchildren with haemoglobin levels between 110 and 119 g/L will result in an improvement of cognitive functions, even though they are not otherwise malnourished.
Introduction
Iron deficiency is the most common nutritional deficiency in both developing and industrialized countries [1]. The consequences of iron deficiency include impaired immunity, increased morbidity from infectious disease, and decreased physical capacity. In infants and children, studies have demonstrated impaired motor development and coordination, impaired language and scholastic achievement, psychological and behavioural effects, and decreased work capacity [2-6].
The possible effects of iron deficiency on behavioural development were reviewed by Pollitt and Leibel [7], as well as by an International Conference on Iron Deficiency and Behavioral Development [8]. It was concluded that iron deficiency in infants and children is associated with lower scores on tests of development and with impaired learning and school achievement. These findings were of public health concern because of the potential effect of iron deficiency on education and consequently on social and economic development.
Over the past decade and a half, the relationship between iron deficiency and cognitive performance has received increasing attention and confirmation. Studies have indicated that iron therapy may favourably affect developmental test scores in some, but not all, anaemic children [9]. The first controlled demonstration of an adverse effect of subclinical iron deficiency on learning and behaviour in young children was by Pollitt et al. in Cambridge, Massachusetts, USA [10, 11]. On a battery of cognitive and behavioural tests, moderately iron-deficient three- to six-year-old children had lower test scores than those with normal iron status. These values returned to normal after 12 weeks of supplementary oral iron. However, when Pollitt repeated these studies on pre-school children with iron-deficiency anaemia in Guatemala, these effects were not reversed by iron supplementation [10, 12].
In this paper we report results from a double-blind intervention trial in which the impact of iron supplementation on the cognitive function of nine-year-old children was assessed. In addition to haematological measurements, nutritional status was assessed by anthropometric and biochemical values in order to identify nutritional variables possibly associated with iron nutrition status and intellectual functions.
Methods
Sample selection
Since relatively few cases of iron-deficiency anaemia are reported in Croatian schoolchildren in the official national health statistics, a pilot study to screen blood haemoglobin levels was carried out in cooperation with the school health services in a rural area of central Croatia. Children from an elementary school near Zagreb with a relatively high prevalence of mild anaemia were selected. In order to minimize the effect of age on iron nutrition status as well as on cognitive functions, only children attending the third grade of elementary education and between 8.7 and 9.6 years of age were selected.
The children remained under the regular supervision of the local school health services. There were no endemic infectious diseases in the area, and intestinal parasites were not a problem. Following oral and written explanation of the purpose and procedures of the study, parents gave written consent for their children to participate.
Study design
All children received initial anthropometric, biochemical, and psychological examinations to collect baseline pretreatment data. Children from one class (supplemented group) were then given tablets for 10 weeks containing 100 mg of iron in the form of ferri-glycine sulfate (Orferon Retard-Pliva, Zagreb), while children from the other class (control group) received placebo tablets identical in appearance. The tablets were distributed daily, except Sunday, in a double-blind manner under the supervision of the teachers. The school health nurse visited classes and teachers twice a week in order to confirm compliance with the distribution.
It was logistically unacceptable to assign the children within each class randomly to different treatment conditions. However, because a child’s placement in a class is random, the two groups of children were comparable. This is supported by the lack of any significant difference between classrooms on the many baseline measures collected during this study.
All examinations were repeated after 10 weeks of supplementation. Thirty-one (18 boys and 13 girls) of the 34 children assigned to the supplemented group and 29 (17 boys and 12 girls) of the 32 children assigned to the control group completed all biochemical and cognitive examinations and were included in the analysis of treatment effects.
Assessment of nutritional status
Anthropometric examinations included measurements of body weight, body height, and bicipital, tricipital, subscapular, and midaxillary skinfolds. On the basis of weight and height measurements, the relative body weight (weight as percentage of standard weight-for-height) as a criterion of current nutrition status and the relative body height (percentage of standard height-for-age) as a criterion of growth performance were calculated. Relative body weight and relative body height were calculated using National Center for Health Statistics data as the reference standard [13]. The percentage of body fat was calculated from skinfold measurements according to the equation of Durnin and Rahaman [14].
Venous blood (20 ml) was drawn from each subject using disposable plastic needles and syringes. Approximately 3 ml of the whole blood was placed into an EDTA-containing tube for blood cell counts and other haematological examinations. An equal amount of the whole blood was mixed with acid citrate dextrose stabilizer containing 7.3 g/L citric acid, 22.0 g/L sodium anhydride, and 24.5 g/L D-glucose monohydrate for the enzymatic determination of thiamine, riboflavin, and pyridoxine in the red blood cells. The remaining blood was allowed to clot, and 0.5 ml of serum was separated and mixed with 4.5 ml of 5% metaphosphoric acid. This sample and the rest of the serum were frozen at -30°C for further analyses. All blood samples were drawn between 8 and 10 a.m. Iron status was assessed by measuring haemoglobin, haematocrit, and red blood cell count (RBC) and by calculating mean corpuscular haemoglobin (MCH), mean corpuscular volume (MCV), and mean corpuscular haemoglobin concentration (MCHC) using standard automated blood-cell counter procedures and the ABX-ARGOS counter (Roche). Serum iron and total iron-binding capacity (TIBC) were measured by the ferrozine method [15] and Spectrum autoanalyser (Abbot, Irving, Tex, USA), and transferrin saturation was calculated as the ratio of serum iron to TIBC × 100.
In addition, the following biochemical values were measured: vitamins A and E in serum by high-performance liquid chromatography [16] and vitamin C in serum by a modification of the fluorometric method of Brubacher and Vuilleumier [17]. Erythrocyte thiamine was determined on the basis of transketolase enzyme reactivation (ETK), erythrocyte riboflavin on the basis of the glutathione reductase reactivation test (EGR), and erythrocyte pyridoxine on the basis of glutamic oxalacetic transaminase (EGOT) reactivation [18, 19]. Zinc in serum was measured by atomic absorption spectrophotometry [20].
Assessment of cognitive function
Cognitive function was assessed between 8 a.m. and noon Monday through Saturday, with children from each class examined on alternate days. The testing was done in a separate, quiet room in the school library with which the children were familiar. The environment was comfortable and free of distraction.
The abbreviated Wechsler Intelligence Scale for Children-Revised (WISC-R) containing six subtests-arithmetic, similarities, digit span, picture completion, block design, and digit symbol (coding)-was used to obtain information on both verbal and non-verbal aspects of intelligence in individual subjects.
In the coding test, children have to substitute symbols for numbers as quickly as possible. The score represents the total number of correct symbols written during a fixed time. The coding test primarily assesses visual-motor coordination, visual encoding, and short-term memory, concentration, and sustained attention.
The arithmetic test involves primary mental arithmetic problems, each having a time limit. The scores represent the total number of correct answers. The arithmetic test primarily assesses numerical reasoning, concentration, verbal comprehension and integration, and knowledge of basic numerical operations. Other factors influencing performance are short-term memory, attention, and interest in and comfort with mathematics.
The similarities test is composed of 13 questions inquiring how the concepts are related to each other. It is considered to reflect logical abstract (categorical) thinking and verbal concept formation. Other factors influencing performance are long-term memory, concentration, and cultural experience.
The digit-span test involves the immediate recall of increasingly longer strings of digits that are read to the children. One set has to be recalled forward as they are given and a second set has to be recalled backwards. The forward and backward items were analysed separately, and the children’s scores represent the total number of correct responses. This test measures attention, short-term auditory memory, and auditory sequencing.
In picture completion, the subject is asked to find the missing parts in a picture. The test is said to reflect visual alertness as well as visual recognition and identification (long-term visual memory).
In block design, the subject is required to copy a sample design using cubes painted in different colours and patches. This test breaks down the analysis of the whole into its component parts-nonverbal concept formation and spatial visualization.
Data analysis
The paired t test was used for pre- and post-treatment comparisons. Pearson’s r correlation coefficients were determined from scattergrams, and regression analysis was used to evaluate the relationship between measures of cognitive function and anthropometric and biochemical measurements of nutritional status. All statistical analyses were done using the statistical software package SPSS for Windows 6.0 (SPSS, Chicago, Ill, USA).
Results
The results of anthropometric assessment of nutritional status (table 1) show that when relative body weight was used as an indicator of energy balance, most children had adequate dietary energy intake. Only one child had a weight below 80% of the reference standard, indicating undernutrition, whereas 7.6% of the children had weights greater than 125% of the reference standard. The percentage of body fat calculated from skinfold measurements showed that 7.4% of the boys and 10.1% of the girls were slightly obese.
Data on relative body height show that 98% of the children had attained satisfactory height-for-age, indicating that stunting, which usually reflects moderate to severe malnutrition during early childhood and is associated with poorer mental performance, was not a problem in the population studied.
Children with lower haemoglobin values (below 120 g/L) were slightly shorter than children with haemoglobin levels above 120 g/L, but these differences were not statistically significant. On the basis of the relative body weight and the percentage of body fat, two indices of soft-tissue development, the non-anaemic children appeared to have a better nutritional status than the mildly anaemic ones. However, only the difference in boys reached the level of statistical significance (p <.05).
The results of biochemical examinations (table 2) show that, despite the adequate energy intake and satisfactory mean values for biochemical indices of nutritional status, approximately 40% of the children had iron deficiency, a rather high prevalence. However, as judged by serum values, other micronutrient deficiencies were also present, including riboflavin (31%), pyridoxine (18%), and vitamin C (9%) deficiencies.
Correlation coefficients between baseline anthropometric and biochemical measurements of nutritional status and summary scores from the WISC-R administered prior to intervention are given in table 3. The only statistically significant association between anthropometric or biochemical measurements of nutritional status and total WISC-R score was with transferrin saturation (p<.0.1). However, the sum of scaled scores from verbal tests of the WISC-R was significantly correlated with haemoglobin (p <.01), haematocrit (p <.01), transferrin saturation (p<.01), MCHC (p <.05), and relative body height (p <.01). It was also positively correlated with red cell thiamine content (p <.05). None of the nutritional status indicators were significantly associated with the sum of scaled scores from the nonverbal tests of the WISC-R.
TABLE 1. Baseline anthropometric measures (mean ± SD) of nutritional status according to sex and haemoglobin concentration
|
Measurement |
Total sample |
Hb < 120 g/L |
Hb ³ 120 g/L |
|||
|
Boys (n = 38) |
Girls (n = 28) |
Boys (n = 15) |
Girls (n = 13) |
Boys (n = 23) |
Girls (n = 15) |
|
|
Age (yr) |
9.3 ± 0.4 |
9.1 ± 0.2 |
9.2 ± 0.4 |
9.1 ± 0.2 |
9.3 ± 0.3 |
9.1 ± 0.3 |
|
Weight (kg) |
30.3 ± 6.3 |
30.6 + 5.5 |
27.6 ± 4.4 |
30.2 ± 7.3 |
32.1 ± 6.8 |
31.3 ± 3.6 |
|
Weight-for-height |
101.6 ± 12.9 |
104.1 ± 14.3 |
95.6 ± 8.7b |
100.6 ± 12.4 |
105.5 ± 13.91b |
107.1 ± 15.5 |
|
Height (cm) |
135.0 ± 4.9 |
134.9 ± 4.5 |
133.4 ± 4.5 |
134.7 + 5.2 |
136.1 ± 4.9 |
135.2 + 3.8 |
|
Height-for-age |
101.1 ± 3.9 |
101.6 + 3.2 |
100.1 ± 4.5 |
101.4 ± 4.1 |
101.8 ± 3.5 |
101.8 ± 2.2 |
|
Body fat (%) |
16.5 ± 4.2 |
24.2 ± 4.7 |
15.2 ± 2.7 |
22.7 ± 4.9 |
17.3 ± 3.8 |
25.6 ± 4.4 |
TABLE 2. Baseline biochemical measures of nutritional status
|
Measure |
Baseline value (mean ± SD) |
Criterion of deficiency |
% Deficient |
|
|
Boys |
Girls |
|||
|
Haemoglobin (g/L) |
120.4 ± 5.2 |
<120a |
39.5 |
46.4 |
|
Haematocrit (%) |
38.8 ± 1.5 |
<34a |
0.0 |
0.0 |
|
Serum iron (mol/L) |
17.4 ± 7.2 |
<10.8a |
10.5 |
25.0 |
|
Transferrin saturation |
24.8 ± 7.5 |
<15.0a |
7.9 |
3.6 |
|
MCV (m3) |
83.8 ± 2.4 |
<76.0a |
0.0 |
0.0 |
|
MCH (pg) |
26.3 ± 1.2 |
<26.0a |
57.9 |
39.3 |
|
MCHC (%) |
310.1 ± 13.8 |
<320.0a |
75.6 |
75.0 |
|
Vitamin A (g/L) |
284.7 ± 42.2 |
<100.0b |
0.0 |
0.0 |
|
Vitamin C (mg/L) |
5.5 ± 2.4 |
<2.0b |
10.5 |
7.1 |
|
Vitamin E (mg/L) |
7.4 ± 1.6 |
<5.0b |
2.6 |
3.6 |
|
Thiamine (a-ETK)c |
10 ± 0.05 |
>1.25b |
0.0 |
0.0 |
|
Riboflavin (a-EGR)c |
1.29 + 0.10 |
>1.30b |
26.3 |
39.3 |
|
Pyridoxine (a-EGOT)c |
1.89 ± 0.12 |
>2.0b |
13.2 |
25.0 |
|
Zinc (mol/L) |
13.9 + 1.2 |
<10.7d |
2.6 |
0.0 |
Iron supplementation had the expected positive effect on the indicators of iron status, with haemoglobin, haematocrit, transferrin saturation, RBC, MCH, and MCHC showing statistically significant improvement (table 4). However, 10 weeks of iron supplementation did not significantly affect either serum iron concentrations or MCV. In the placebo group, there were no significant increases in the measures of iron status.
In both the iron-supplemented and the placebo groups, there was a statistically significant increase in serum concentrations of vitamin C (p <.001) and vitamin E (p <.05). The increase in the activity of the ETK and EGR tests, indicating a reduced dietary intake of thiamine and riboflavin in both groups, was not significant in either group. In both groups there was also a slight, but statistically significant, reduction in serum zinc values (p <.05).
Iron supplementation had a positive and statistically significant effect on several scaled scores from the WISC-R (table 5). Iron supplementation improved performance on all three nonverbal tests, of which improvements in block design and coding were statistically significant and the sum of the scaled scores from nonverbal subtests was statistically significant (p <.01). The slight increase in the sum of scaled scores from the verbal subtest was not statistically significant (p >.05). The only significant effect of iron supplementation on verbal scores was on the similarities subtest (p <.05).
The effects of iron supplementation were more pronounced in children with initial haemoglobin values below 120 g/L than in children with values of 120 g/L or more (table 6). The mean haemoglobin value of the former group of children increased from 115.2 to 120.8 g/L with supplementation (p <.001), whereas the mean haemoglobin value of the latter group only increased from 125.4 to 127.8 g/L (p >.05). However, despite the positive effect of iron supplementation on mean biochemical indicators of iron nutrition status, 6 out of 18 subjects (33.3%) with initial haemoglobin values below this limit continued to have values below 120 g/L after 10 weeks of iron supplementation.
TABLE 3. Correlations (Pearson r) between anthropometric and biochemical indicators of nutritional status and WISC-R scores
|
Indicator |
WISC-R score |
||
|
Verbal |
Non-verbal |
Total |
|
|
Weight-for-height (% of standard) |
0.0801 |
-0.1652 |
-0.0758 |
|
Height-for-age (% of standard) |
0.3844*** |
0.1521 |
0.1051 |
|
Body fat (%) |
0.0658 |
-0.0379 |
0.0093 |
|
Haemoglobin |
0.3950*** |
0.0055 |
0.2265 |
|
Haematocrit |
0.2943** |
0.0462 |
0.1319 |
|
Serum iron |
0.1343 |
0.1429 |
0.1802 |
|
Transferrin saturation |
0.3574** |
0.1851 |
0.3367** |
|
MCV |
0.0928 |
0.0024 |
0.054 |
|
MCH |
0.1843 |
0.0431 |
0.0722 |
|
MCHC |
0.2435* |
0.0388 |
0.1655 |
|
RBC |
0.202 |
0.0354 |
0.0878 |
|
Vitamin A |
0.1225 |
0.1049 |
0.0078 |
|
Vitamin C |
0.1812 |
0.0892 |
0.0367 |
|
Vitamin E |
0.1829 |
0.2425* |
0.0746 |
|
Thiamine |
0.2466* |
0.0069 |
0.1338 |
|
Riboflavin |
0.0307 |
0.0861 |
0.0458 |
|
Pyridoxine |
0.1106 |
0.0426 |
0.0934 |
|
Zinc |
0.0813 |
0.1238 |
0.1364 |
TABLE 4. Effects of iron supplementation on biochemical measures of nutritional status (mean ± SD) in comparison with the group receiving the placebo
|
Measure |
Iron supplementation (n = 31) |
Placebo (n = 29) |
||||
|
Pre |
Post |
p |
Pre |
Post |
p |
|
|
Haemoglobin (g/L) |
119 ± 6 |
123 ± 6 |
.05 |
121 ± 4 |
121 ± 6 |
NS |
|
Haematocrit (%) |
38.3 ± 1.7 |
39.3 ± 1.7 |
.05 |
38.3 ± 1.2 |
37.8 ± 1.6 |
NS |
|
Serum iron (mol/L) |
17.3 ± 5.6 |
17.4 ± 9.5 |
NS |
17.2 ± 4.4 |
16.4 ± 4.0 |
NS |
|
Transferrin saturation |
25.9 ± 8.5 |
30.2 ± 8.3 |
.05 |
24.2 ± 6.8 |
25.5 ± 7.0 |
NS |
|
MCV (µ3) |
83.7 ± 2.6 |
84.1 ± 2.6 |
NS |
84.2 ± 2.0 |
84.1 ± 2.4 |
NS |
|
MCH (pg) |
25.5 ± 0.9 |
27.2 ± 1.1 |
.001 |
26.8 ± 1.1 |
27.0 ± 1.4 |
NS |
|
MCHC (%) |
303 ± 8 |
323 ± 8 |
.001 |
316 ± 10 |
320 ± 9 |
NS |
|
RBC (× 106) |
4.5 ± 0.2 |
4.7 ± 0.2 |
.05 |
4.6 ± 0.2 |
4.5 ± 0.2 |
.05 |
|
Vitamin A (g/L) |
274 ± 53 |
296 ± 50 |
NS |
296 ± 28 |
302 ± 25 |
NS |
|
Vitamin C (mg/L) |
5.1 ± 2.5 |
11.1 ± 3.7 |
.001 |
5.8 ± 2.5 |
11.1 ± 2.0 |
.001 |
|
Vitamin E (mg/L) |
6.9 ± 1.4 |
8.0 ± 1.2 |
.01 |
7.9 ± 1.7 |
9.0 ± 1.1 |
.05 |
|
Thiamine (a-ETK) |
1.11 ± 0.05 |
1.13 ± 0.05 |
NS |
1.08 ± 0.05 |
1.12 ± 0.08 |
NS |
|
Riboflavin (a-EGR) |
1.28 ± 0.13 |
1.35 ± 0.13 |
NS |
1.30 ± 0.08 |
1.34 ± 0.07 |
NS |
|
Pyridoxine (a-EGOT) |
1.90 ± 0.12 |
1.92 ± 0.11 |
NS |
1.88 ± 0.12 |
1.87 ± 0.08 |
NS |
|
Zinc (mol/L) |
13.8 ± 1.3 |
11.7 ± 1.2 |
.01 |
14.0 ± 1.2 |
13.3 ± 1.6 |
.05 |
|
Scale |
Iron supplementation (n = 31) |
Placebo (n = 29) |
|||||
|
Pre |
Post |
p |
Pre |
Post |
p |
||
|
Verbal |
30.5 ± 3.8 |
32.2 ± 4.4 |
NS |
30.4 ± 4.2 |
30.9 ± 4.3 |
NS |
|
|
|
Arithmetic |
10.8 ± 2.7 |
10.6 ± 2.0 |
NS |
10.0 ± 2.1 |
10.8 ± 2.1 |
NS |
|
|
Similarities |
10.5 ± 2.2 |
12.0 ± 2.6 |
.05 |
11.9 ± 1.4 |
11.3 ± 1.9 |
NS |
|
|
Digit span |
9.2 ± 2.2 |
9.8 ± 1.7 |
NS |
8.6 ± 2.2 |
8.9 ± 2.1 |
NS |
|
Non-verbal |
28.4 ± 6.9 |
33.5 ± 7.4 |
.01 |
29.3 ± 3.7 |
30.5 ± 3.4 |
NS |
|
|
|
Picture completion |
8.8 ± 2.8 |
9.9 ± 3.1 |
NS |
8.9 ± 1.8 |
9.6 ± 1.8 |
NS |
|
|
Block design |
9.1 ± 3.4 |
11.2 ± 2.8 |
.01 |
10.2 ± 2.2 |
10.4 ± 1.7 |
NS |
|
|
Coding |
10.5 ± 2.6 |
12.4 ± 3.7 |
.01 |
10.2 ± 1.4 |
10.5 ± 1.7 |
NS |
|
Total |
58.8 ± 8.9 |
65.8 ± 9.1 |
.01 |
59.7 ± 5.7 |
61.3 ± 6.1 |
NS |
|
|
Measure |
Hb < 120 g/L (n = 18) |
Hb ³ 120 g/L (n = 13) |
||||||
|
Pre |
Post |
D |
p |
Pre |
Post |
D |
p |
|
|
Haemoglobin (g/L) |
115.2 ± 2.5 |
120.8 ± 4.9 |
5.6 |
.001 |
125.4 ± 4.5 |
127.8 ± 4.2 |
2.4 |
NS |
|
Haematocrit (%) |
37.1 ± 1.4 |
38.1 ± 0.8 |
0.1 |
NS |
39.3 ± 1.5 |
40.8 ± 1.3 |
1.5 |
.05 |
|
Serum iron (mmol/L) |
17.2 ± 11.4 |
18.3 ± 5.9 |
1.1 |
NS |
17.7 ± 6.2 |
16.4 ± 5.0 |
1.3 |
NS |
|
Transferrin saturation (%) |
23.9 ± 9.1 |
31.6 ± 8.0 |
7.7 |
.05 |
28.5 ± 7.0 |
28.3 ± 8.7 |
0.2 |
NS |
|
RBC (×106) |
4.5 ± 0.2 |
4.6 ± 0.2 |
0.1 |
NS |
4.6 ± 10.2 |
4.7 ± 0.3 |
0.1 |
NS |
|
MCV (m3) |
82.8 ± 2.0 |
83.6 ± 2.3 |
0.8 |
NS |
84.9 ± 3.1 |
84.8 ± 2.8 |
0.1 |
NS |
|
MCH (pg) |
25.1 ± 0.8 |
27.0 ± 1.2 |
1.9 |
.001 |
26.1 ± 0.8 |
27.6 ± 0.9 |
1.5 |
.001 |
|
MCHC (%) |
301 ± 8 |
321 ± 7 |
20.0 |
.001 |
307 ± 6.4 |
325 ± 9.1 |
18 |
.001 |
|
WISC-R verbal |
29.6 ± 4.1 |
31.1 ± 4.4 |
1.5 |
NS |
31.7 ± 3.2 |
34.3 ± 3.3 |
2.6 |
NS |
|
WISC-R non-verbal |
28.6 ± 6.6 |
34.7 ± 7.5 |
6.1 |
.05 |
28.9 ± 7.5 |
31.8 ± 7.3 |
2.9 |
NS |
|
WISC-R total |
58.2 ± 8.8 |
65.7 ± 9.3 |
7.5 |
.05 |
59.8 ± 9.4 |
66.2 ± 8.5 |
6.4 |
NS |
Discussion
Iron supplementation during a period of 10 weeks had a positive and statistically significant effect on the mean values of haemoglobin, haematocrit, red cell count, MCH, MCHC, and transferrin saturation. Iron supplementation did not significantly affect serum iron level and MCV. In the placebo group, the average blood haemoglobin values did not change during the period of the study; those of children with initial haemoglobin values below 120 g/L increased slightly but non-significantly (p >.05).
The fact that test scores also correlate with anthropometry and socio-economic status raises the question of the extent to which coexistent nutritional deficiencies contributed to the differences. During the period of the study, some of the other biochemical indicators of nutritional status changed as well. There was a highly significant increase in the vitamin C values in both groups. This increase could be explained by a seasonal increase in the dietary intake of ascorbic acid, which usually takes place at the end of the spring season. At this time, sources of ascorbic acid become available again after a shortage during the late winter and early spring. This phenomenon was repeatedly observed in our earlier studies with schoolchildren that were initiated in early spring and terminated before summer vacations in the middle of June [24]. The higher ascorbic acid content in the diet [25] might have improved iron absorption sufficiently to account for the increase in haemoglobin levels in children from the placebo group who had initially low haemoglobin values.
Serum vitamin E increased significantly in both the iron-supplemented and the placebo groups. The increase in vitamin E biochemical status was probably the result of the seasonal increase in the consumption of green leafy vegetables, which contain appreciable amounts of this nutrient [26].
In order to assess the possible influence of vitamin C and vitamin E on the WISC-R scores, we carried out an analysis of variance between the increase (D) in WISC-R scores in the supplemented and placebo groups, using the increase (D) in vitamins C and E as covariates. The increase in the biochemical values of these two vitamins did not have any significant effect on the WISC-R scores. Thus this study does not provide evidence for a significant effect of nutrients other than iron.
It is noteworthy that the serum zinc content was reduced at the end of the study in both groups, but more significantly in the iron-supplemented group. Although higher intakes of dietary iron have been reported to have a negative effect on the absorption of zinc [27-29], the significant decrease in serum zinc levels in the placebo group indicates that iron supplementation was not the only factor in serum zinc reduction in the iron-supplemented group.
The positive effect of iron supplementation on WISC-R scores was primarily on the non-verbal component of the test. Out of the three subtests of the non-verbal component, the differences between changes in block design and coding (DT2- T1) were statistically significant in comparison with changes in the placebo group (p <.05). The difference in picture completion scores did not reach statistical significance. In regard to the verbal component, however, only the increase in scores in the similarities subtest in the iron -supplemented group was statistically significantly different from the changes in the placebo group (DT2 - T1 =2.34; t = 3.529; p <.001).
In the placebo group, despite the absence of any significant increases in the mean values of iron nutrition status indicators, and a decrease in RBC, there was a non-significant increase in the total WISC-R scores at the end of the study. This increase in the test scores in the second testing in the non-supplemented group could be attributed to greater familiarity and consequently to a more comfortable feeling of the examinees under test situations, as described earlier [8].
In earlier studies, the effects of double-blinded oral supplementation with iron or a placebo were examined in rural Indonesian schoolchildren classified as anaemic or non-anaemic [30]. The haematological. status returned to normal after supplementation with 10 mg of ferrous sulfate per kilogram body weight per day for three months, and the anaemic group markedly improved their learning and achievement scores with iron supplementation. No change was observed in non-anaemic children or in anaemic children given placebo. Three other studies in Indonesia had similar results [12, 31, 32].
Improvement in cognitive test performance in preschool children and schoolchildren after haematological response to iron supplementation was also reported from Egypt [12]. However, the effect of supplementation was not significant when the entire supplemented group was compared with the group receiving the placebo. Studies in 8- to 15-year-old schoolboys in India compared the effect of 30- and 40-mg doses of elemental iron and a placebo on a number of tests of cognitive function, using the Indian adaptation of the WISC [33]. Both doses improved recent memory, attention, auditory memory, auditory sequencing, visual-motor coordination, and visual perception. Furthermore, they found that the prophylactic dose of 60 mg of elemental iron per day improved attention, memory, and concentration. The conclusion from these studies was that iron supplementation would improve scholastic performance. The same conclusion was reached by Soemantri in Indonesia and by Pollitt and co-workers in Indonesia and Egypt.
In a double-blinded study of children 9 to 11 years of age in Bangkok, a significant positive association was found between iron status and performance on the Raven Progressive Matrices used to measure IQ, on the Thai language test, and on a mathematics test [34]. Even children who were iron depleted without anaemia had significantly lower scores on the Thai language test than did iron-repleted children. However, no improvement was observed after the administration of 100 mg of iron per day as ferrous sulfate. These results differ from those obtained by the same author in Indonesia and Egypt. The explanation proposed is that reversibility depends on whether damage occurs during a critical period of brain growth during infancy. Iron-deficiency anaemia in schoolchildren and adults may or may not be a continuation of the same deficiency in infancy. Where it is not, the effects on cognitive function are reversible, as in the present study as well as in four separate studies in India [33] and on adolescent girls in the United States [35].
It is evident that key iron-containing compounds in the brain respond to iron deficiency at any age, even when the deficiency is not sufficiently severe to affect haematopoiesis. However, the mechanism by which iron-deficiency anaemia in infancy has a lasting effect, even when the individual is no longer iron deficient, is unknown. Work from Israel with iron-deficient rats suggests that iron deficiency at a critical developmental state reduces the number of central dopamine neurotransmitters [36, 37]. It is postulated that interference with iron metabolism at an early age in humans could cause similar irreversible damage to brain function. It has also been proposed that the underlying problem is a reduction in non-haem iron in the brain or other systemic changes in the organism associated with the reduction of the transport of oxygen [38]. It is of interest that in our study the six children with initially low haemoglobin values increased their WISC-R scores after supplementation, without an increase in haemoglobin values.
It is concluded that mild iron deficiency in otherwise adequately nourished nine-year-old schoolchildren with haemoglobin levels between 110 and 119 g/L may affect behaviour and cognitive functions, and that iron supplementation resulted in an improvement of haematological status and cognitive functions. The overall results of this and the other studies cited indicate that iron deficiency can be educationally disadvantageous at any age, independently of ethnicity and of physical and social environment.
1. The World Health Report 1997. Geneva: World Health Organization, 1997.
2. Dallman PR. Manifestations of iron deficiency. Semin Hematol 1982;19:19-30.
3. Lozoff B. Behavioral alterations in iron deficiency. Adv Pediatr 1988;35:331-59.
4. Soemantri AG, Pollitt E, Kim I. Iron deficiency anemia and educational achievement. Am J Clin Nutr 1985;42:1221-8.
5. Scrimshaw NS. Functional significance of iron deficiency. In: Enwonwu CO, ed. Functional significance of iron deficiency. Annual Nutrition Workshop Series. Vol. III. Nashville, Tenn, USA: Center for Nutrition, Meharry Medical College, 1990:1-13.
6. Prasad AN, Prasad C. Iron deficiency: non-hematological manifestations. Prog Food Nutr Sci 1991;15:255-83.
7. Pollitt E, Leibel RL. Iron deficiency and behavior. J Pediatr 1976;88:372-81.
8. Haas JD, Fairchild MW. Summary and conclusions of the International Conference on Iron Deficiency and Behavioral Development. Am J Clin Nutr 1989; 50(suppl):703-5.
9. Lozoff B. Methodologic issues in studying behavioral effects of infant iron-deficiency anemia. Am J Clin Nutr 1989;50(suppl):641-54.
10. Pollitt E, Lewis N, Leibel RL, Greenfield DB. Iron deficiency and Flay behaviour in preschool children. In: Garry PJ, ed. Human nutrition. Washington, DC: American Association for Clinical Chemistry, 1981:290-301.
11. Pollitt E, Viteri F, Saco-Pollitt C, Leibel RL. Behavioural effects of iron deficiency anemia in children. In: Pollitt E, Leibel RL, eds. Iron deficiency: brain biochemistry and behavior. New York: Raven Press, 1982:195-208.
12. Pollitt E, Soemantri AG, Yunis F, Scrimshaw NS. Cognitive effects of iron-deficiency anemia (letter). Lancet 1985;1(8421):158.
13. NCHS growth charts. (HRA 76 - 1120,25,3). Rockville, Md, USA: United States Public Health Service, Health Resources Administration, 1976.
14. Durnin JVGA, Rahaman MM. The assessment of the amount of far in the human body from measurements of skinfold thickness. Br J Nutr 1967;21:681-9.
15. Carter P. Spectrophotometric determination of serum iron at the submicrogram level with a new reagent (ferrozine). Anal Biochem 1971;40:450-8.
16. Vuilleumier JP, Keller ME, Gysel D, Hunziker F. Clinical chemical methods for the routine assessment of the vitamin status in human populations. Part I: The fat-soluble vitamins A and E, and beta-carotene. Int J Vit Nutr Res 1983;53:265-72.
17. Brubacher G, Vuilleumier JP. In: Curtius HCH, Roth M, eds. Clinical biochemistry: principles and methods. Vol I. New York, Berlin: de Gruyter, 1974:989-97.
18. Vuilleumier JP, Keller ME, Retenmaier R, Hunziker F. Clinical chemical methods for the routine assessment of the vitamin status in human populations. Part II: The water-soluble vitamins B1, B2 and B6. Int J Vit Nutr Res 1983;53:359-70.
19. Vuilleumier JP, Keller HE, Keck E. Clinical chemical methods for the routine assessment of the vitamin status in human populations. Part III: The apoenzyme stimulation test for vitamin B1, B2 and B6 adapted to the Cobas-Bio Analyzer. Int J Vit Nutr Res 1990;60:126-35.
20. Henry RJ. Clinical chemistry: principles and techniques. New York: Harper and Row, 1974.
21. World Health Organization. Indicators and strategies for iron deficiency programmes. Geneva: WHO, 1998.
22. Haller J. The vitamin status assessment. An international overview. Int J Vit Nutr Res (in press).
23. Schrijver J. Biochemical markets for micronutrient status and their interpretation. In: Pietrzik K, ed. Modern lifestyles, lower energy intake and micronutrient status. London: Springer-Verlag, 1991:55-85.
24. Buzina R, Brodarec A, Jusic M, Milanovic N, Kolombo V, Brubacher G. Epidemiology of angular stomatitis and bleeding gums. Int J Vit Nutr Res 1973;43:401-15.
25. Hallberg L, Brune M, Rossander-Hulthen L. Is there a physiological role of vitamin C in iron absorption? In: Burns JJ, Rivers JM, Machlin LJ, eds. Third conference on vitamin C. New York: New York Academy of Sciences, 1987; 498:324-32.
26. Bauernfeind J. Tocopherols in foods. In: Machlin LJ, ed. Vitamin E: a comprehensive treatise. New York: Marcel Dekker, 1980:99-167.
27. Solomons NW, Jacob RA. Studies on the bioavailability of zinc in humans: effects of heme and nonheme iron on the absorption of zinc. Am J Clin Nutr 1981;34: 475-82.
28. Meadows NJ, Grainger SL, Ruse W, Keeling PW, Thompson RP. Oral iron and the bioavailability of zinc. BMJ 1983;287:1013-4.
29. Dawson EB, Albers J, McGanity WJ. Serum zinc changes due to iron supplementation in teen-age pregnancy. Am J Clin Nutr 1989;50:848-52.
30. Soemantri AG. Preliminary findings on iron supplementation and learning achievement of rural Indonesian Children. Am J Clin Nutr 1989;50(suppl 3):698-702.
31. Soewondo S, Husaini M, Pollitt E. Effect of iron deficiency on attention and learning processes in preschool children: Bandung, Indonesia. Am J Clin Nutr 1989(suppl 3): 667-74.
32. Idjradinata P, Pollitt E. Reversal of developmental delays in iron-deficient anaemic infants treated with iron. Lancet 1993;341:1-4.
33. Seshadri A, Gopaldas T. Impact of iron supplementation on cognitive functions in preschool and school-aged children: the Indian experience. Am J Clin Nutr 1989;50(suppl 3):675-86.
34. Pollitt E, Hathirat P, Kotchabhakdi NJ, Missell L, Valyasevi A. Iron deficiency and educational achievement in Thailand. Am J Clin Nutr 1989;50(suppl 3):687-97.
35. Brunner AB, Joffe A, Duggan AK, Casella JF, Brandt J. Randomized study of cognitive effects of iron supplementation in non-anaemic iron-deficient adolescent girls. Lancet 1996;348:992-6.
36. Youdim MBH, Ben-Schachar D, Yehuda S. Putative biological mechanisms of the effect of iron deficiency on brain biochemistry and behavior. Am J Clin Nutr 1989; 50:607-17.
37. Yehuda S, Youdim MBH. Brain iron: a lesson from animal models. Am J Clin Nutr 1989;50:618-29.
38. Pollitt E. Iron deficiency and cognitive functions. Annu Rev Nutr 1993;13:521-37.
