点击显示 收起
1 From the Nutrition Program, School of Population Health, The University of Queensland, Queensland, Australia (FA, GCM, and GW), and the Institute of Nutrition and Food Science, University of Dhaka, Dhaka, Bangladesh (MRK, MA, RK, CPB, and BN)
2 Supported by a grant from the Nestlé Foundation, Vevey. Switzerland. 3 Address reprint requests to F Ahmed, Nutrition Program–Division of International Health, School of Population Health, The University of Queensland, Public Health Building, Level 4, Herston Road, Queensland 4006, Australia. E-mail: f.ahmed{at}sph.uq.edu.au.
ABSTRACT
Background: Although iron deficiency is a major cause of anemia, other micronutrient deficiencies may also play a role.
Objective: We examined whether multiple micronutrient supplementation is more efficacious than is supplementation with iron and folic acid alone for improving the hemoglobin and iron status of anemic adolescent girls in Bangladesh.
Design: Anemic (hemoglobin < 12.0 g/dL) girls (n = 197) aged 14–18 y from rural schools in Dhaka District were entered into a randomized double-blind trial and received twice-weekly supplements of iron and folic acid (IFA group) or multiple micronutrients (15 micronutrients, including iron and folic acid; MMN group) for 12 wk.
Results: At recruitment, the characteristics of the girls in the 2 groups were not significantly different, except for family size and body mass index. At the end of the study, although both groups benefited significantly from supplementation, mean changes in hemoglobin and serum ferritin concentrations were not significantly different between groups. Compared with the IFA group, girls in the MMN group had significantly greater increases in mean serum vitamin A, plasma vitamin C, red blood cell folic acid, and riboflavin concentrations (assessed as erythrocyte glutathione reductase activation coefficient). After 12 wk of supplementation, only the prevalence of vitamins A and C and riboflavin deficiencies decreased more significantly in the MMN group than in the IFA group.
Conclusions: Twice-weekly MMN supplementation for 12 wk significantly improved the status of the micronutrients assessed but was not more efficacious than was supplementation with iron and folic acid alone in improving the hematologic status of anemic adolescent girls. More frequent doses may be needed to achieve full benefit.
Key Words: Anemia • iron deficiency • multiple micronutrients • adolescent girls • Bangladesh
INTRODUCTION
Anemia affects more than a billion people globally and is recognized as a major public health problem in developing countries. Iron deficiency is considered to be the main cause of anemia (1). Other possibly important causes of anemia include malaria, hookworm, chronic infection, and deficiencies of micronutrients such as riboflavin, folic acid, and vitamins A, C, and B-12 (2-4). These micronutrients affect hemoglobin synthesis either directly, by impairing erythropoiesis, or indirectly, by affecting absorption or mobilization (3, 5-8). In a recent review by Fishman et al (8), the importance of various micronutrients in erythropoiesis is discussed in more detail.
In Bangladesh, although the prevalence of anemia in adolescent girls is very high, estimates vary widely: 43% in rural girls (9) and 20–40% in urban girls (10, 11). A large proportion of the anemia problem is related to iron deficiency (12, 13). Furthermore, studies have shown that a significant proportion of the adolescent girls in the country suffer from subclinical vitamin A deficiency (11, 14). Dietary surveys also indicate that a large proportion of adolescent girls do not meet the daily requirements for various micronutrients, including iron, vitamins A and C, and riboflavin (15, 16).
In populations, single micronutrient deficiencies are rare, and it is likely that iron deficiency in anemic subjects is compounded by multiple micronutrient (MMN) deficiencies (17, 18). Allen et al (19) suggested that the coexistence of MMN deficiencies, including iron deficiency, may increase the risk of anemia and limit the hematologic response to iron supplementation. In 1999, UNICEF/WHO/UNU recommended that MMN supplementation should not only be promoted for use in pregnant women but also for use in adolescent girls in developing countries to prevent anemia and other micronutrient deficiencies and to improve stores before the onset of pregnancy (20). Furthermore, it was suggested that the supplement should be given on a once or twice weekly basis, because this is thought to be a more practical option for implementing an intervention program (20).
Although 2 studies in pregnant women (21, 22) and 1 study in nonpregnant women (23) recently examined the relative efficacy of MMN supplementation compared with supplementation with iron alone on hematologic improvement, data on adolescent girls are not available. Furthermore, none of the abovementioned studies measured the effect of MMN supplements on micronutrient status other than iron status, which would help in the interpretation of the hematologic findings. Therefore, we conducted this study to examine whether twice-weekly MMN supplementation was more efficacious than was supplementation with iron and folic acid alone in improving the hemoglobin and iron status of anemic adolescent girls in Bangladesh. Furthermore, we sought to determine whether twice-weekly MMN supplementation was able to improve the status of other micronutrients.
SUBJECTS AND METHODS
Study setting and subjects
The subjects were postmenarcheal, nonpregnant girls aged 14–18 y who were in the schools of selected rural Thanas (subdistricts) of Dhaka District in Bangladesh. Schools were selected purposively and school permission was sought; girls in grades 8–10 were invited to participate. Both the subjects’ and their parents’ written consent to participate in the study was sought, and all eligible girls were screened for anemia. Initial screening for anemia was carried out with a B-Hemoglobin Analyzer (HemoCue, Angelholm, Sweden) on a finger-prick capillary blood sample. Those who were identified as anemic (hemoglobin < 12.0 g/dL) were then invited to provide a venous blood sample (baseline sample) on a prefixed date, and their hemoglobin concentration was measured again with a cyanmethemoglobin method. The girls who were found to be anemic on the basis of the baseline blood sample were recruited for the study. The girls with severe anemia (hemoglobin < 8.0 g/dL) were excluded from the study and referred to the hospital. The girls with any sign of chronic infection or metabolic disorder were also excluded from the study.
The sample size for the study was calculated on the basis of an expected difference of 0.5 g/dL for hemoglobin between 2 treatment groups. From our previous experience, we estimated the SD of the hemoglobin concentration for the population to be 1.1 g/dL (11). To detect this difference with a power of 80% at the 5% significance level and after correction for an allowance of 20% dropout during follow-up, each of the 2 groups required 95 subjects. The study protocol was approved by the ethical committee of Medical Research Council, Dhaka, Bangladesh, and the University of Queensland, Brisbane, Australia. The study was conducted from April to October 2003.
Study design and intervention
A randomized, double-blind, experimental trial was used, and the subjects entered into the study were randomly assigned to 1 of 2 treatment groups: to receive either iron and folic acid (IFA) or MMN tablets. Randomization was carried out by using 2 color-coded groups, and the subjects were allocated to 1 of the 2 color codes by lottery carried out by an independent researcher as they were added to the list at the time of recruitment. Subjects had equal probability of being allocated to either of the 2 groups. All the personnel and investigators were blinded to the group assignment. The composition of the MMN tablets, which contained 15 micronutrients, was similar to that proposed by UNICEF and WHO (Table 1) (20). The IFA tablets contained 30 mg Fe and 400 µg folic acid, which was comparable with the iron and folic acid contents of the MMN tablets. The biochemical analyses of the tablets to determine nutrient contents were carried out by the quality control laboratory of the manufacturer (UniMed and UniHealth Manufacturer Limited, Dhaka, Bangladesh). Both supplements were identical in appearance and were color coded at the factory; these details were kept at the University of Dhaka in sealed envelops that were opened only after the preliminary data analysis was completed.
View this table:
TABLE 1. Composition of the supplements1
All subjects received the tablets twice weekly for 12 wk, because this time period was previously found to be effective at producing a significant increase in hemoglobin with weekly iron and folate supplementation in adolescent female factory workers (12).
Compliance
The enrolled subjects were supplemented twice weekly for 12 wk. The tablets were given to the subjects during their lunch break, on school premises, under close supervision of field assistants. The field assistants ensured that the tablets were swallowed with a glass of water and maintained a written record of consumption of the supplement for each subject as a measure of compliance.
Data and sample collection
Socioeconomic data were collected from parents with the use of a self-administered questionnaire. At baseline, trained field staff obtained personal and health-related information about the subjects and recorded anthropometric measurements. Five milliliters of venous blood were drawn from each subject. An aliquot of blood was dispensed into a heparinized tube, and the remaining portion was transferred into a centrifuge tube for the collection of serum. Twenty microliters of whole blood were directly transferred into 5 mL Drabkin’s reagent for the measurement of hemoglobin. Hemolysates for erythrocyte (red blood cell; RBC) folate and erythrocyte glutathione reductase activation coefficient (EGRAC) assays was prepared by using aliquots of heparinized blood and packed cells, respectively, and transported to the laboratory on dry ice. The remaining portion of the heparinized blood was centrifuged (1800 x g, 10 min, room temperature), and 200 µL plasma was transferred into a tube containing 800 µL 5% trichloro-acidic acid. The supernatant fluid was collected after centrifugation and stored in dry ice for transport to the laboratory for plasma ascorbic acid (vitamin C) assay. Hemoglobin was measured spectrophotometrically within 8 h. Serum samples were separated from the clotted blood in the laboratory after centrifugation (1800 x g, 10 min, room temperature). Appropriate aliquots were taken in separate tubes for measurement of vitamins A and B-12, ferritin, and C-reactive protein (CRP). All of the specimens, except those for vitamin B-12 and RBC folic acid, were stored at –20 °C until further analysis. Specimens for folic acid and vitamin B-12 assay were stored at –80 °C. At the end of the 12-wk supplementation period, another 5 mL venous blood was drawn from each subject and processed in the same way as for the baseline sample.
Analytic procedure
Hemoglobin concentration was measured with the use of commercial kit (Roche Diagnostics, Mannheim, Germany). Serum ferritin was measured by enzyme-linked immunosorbent assay with commercial kits (BioCheck Inc, Burlingame, CA), and the interassay CV was 5.4%. RBC folic acid and serum vitamin B-12 concentrations were measured by chemiluminescence with a commercial kit (IMMULITE; Diagnostic Products Corporation, Los Angeles, CA), and the interassay CVs of 9 replicates were 10.4% and 10.1%, respectively. Serum retinol (vitamin A) concentration was measured as described elsewhere (11), and the interassay CV was 4.6%. Plasma vitamin C was measured with the dinitrophenylhydrazine method according to Lowry et al (24). Serum CRP was measured by nephelometry with a commercial kit (Turbox Plus; Orion Diagnostica, Espoo, Finland), and the interassay CV was 2.5%. EGRAC (an indicator of riboflavin) was measured according to the method of Vudhivai et al (25), and the CV was 3.5%.
Statistical analysis
Statistical analysis was carried out by using SPSS (version 12; SPSS Inc, Chicago, IL) and SAS (SAS for WINDOWS version 8; SAS institute Inc, Cary, NC). Univariate analyses of the selected variables were carried out. Normality of distribution of the biochemical variables was assessed with the Kolmogorov–Smirnov goodness-of-fit test. Hemoglobin and serum ferritin concentrations at baseline, and serum vitamin B-12 concentrations at both baseline and after supplementation, were not normally distributed. These data were transformed appropriately for statistical analysis; for presentation, the values were back transformed to original units. All data are presented as means ± SEMs.
The analysis was performed on an intent-to-treat basis. The baseline characteristics of the 2 treatment groups were compared by one-way analysis of variance or chi-square analysis as appropriate. To assess the difference in mean changes in concentrations of hemoglobin and other biochemical measures from baseline to the end of the 12-wk supplementation period between 2 treatment groups, analysis of covariance was used. The main explanatory variables were treatment group, total dose of supplement received and baseline value of the response variable. Any other differences between the 2 treatment groups were treated as potential confounding variables and were also included in the model. Furthermore, in populations in whom infection is common, it is difficult to interpret the data for serum ferritin and serum vitamin A, which are acute phase reactants (26). Therefore, the serum concentration of CRP was measured to identify those who might have inflammation (CRP > 5.0 mg/L) at the time of the study and were taken into account during the data analysis of serum vitamin A and ferritin concentrations.
The prevalence of deficiency for each biochemical index was expressed as the percentage of girls below the appropriate cutoff value. Anemia was defined as a hemoglobin concentration < 12.0 g/dL, based on WHO recommendations for nonpregnant women (27). Below-normal concentrations for other nutritional variables were defined as follows: <12 ng/mL for serum ferritin (28), <0.29 mg/dL for plasma vitamin C (29), <140 ng/mL for RBC folic acid (Diagnostic Products Corporation), <200 pg/mL for serum vitamin B-12 (30), <30.0 µg/dL for serum retinol (vitamin A) (31), and EGRAC 1.4 for riboflavin (5). The difference in prevalence for each of the variables between baseline and the end of the trial, by treatment group, was estimated by estimating a generalized estimating equation with the use of a binomial model with a logit link with time as a repeated-measures option for PROC GENMOD (SAS Institute). However, for the comparison of the prevalence of anemia, all subjects were anemic at baseline in both treatment groups, so the generalized estimating equation model could not provide an interaction result. Furthermore, all subjects in the MMN group had a normal vitamin C status at the end of the 12-wk treatment period. Thus, only a chi-square test was performed between the 2 treatment groups at the end of the study to compare the prevalence of anemia and vitamin C deficiency.
RESULTS
Of the 197 recruited subjects, 178 completed the 12-wk study protocol: 90% of the MMN group and 91% of the IFA group. A total of 19 subjects were lost to follow-up for various reasons: withdrawal from the study, refusal to give blood after supplementation, absence on the day of the postsupplementation blood collection (Figure 1). Socioeconomic characteristics and mean age, height, body weight, body mass index, hemoglobin concentrations, serum ferritin concentrations, and vitamin A concentrations of the subjects who entered the final analysis were not significantly different from those of the subjects who did not enter the final analysis (data not shown).
FIGURE 1.. Selection process for the study participants and reasons for loss to follow-up in a randomized double-blind trial with multiple micronutrient (MMN) and iron and folic acid (IFA) supplementation. Hb, hemoglobin. HemoCue (HemoCue Inc, Angelholm, Sweden).
On the basis of the compliance record, 42% of the subjects in the IFA group received 23–24 of the possible 24 doses, 35% received 21–22 doses, 17% received 18–20 doses, and 6% received 17 doses. In the MMN group, 51% of the subjects received 23–24 doses, 21% received 21–22 doses, 20% received 18–20 doses, and 8% received 17 doses. The mean (±SEM) number of tablets received by the MMN group (21.4 ± 0.3) was not significantly different (P = 0.36) from the number received by the IFA group (21.8 ± 0.2).
At baseline, there were no significant differences in mean age, monthly family income, body weight, or height between the 2 treatment groups (Table 2). However, the MMN group came from relatively smaller-sized families and had significantly lower body mass indexes than did the IFA group. Furthermore, there were no significant differences in parents’ education level or occupation between the 2 treatment groups (data not shown). At baseline, there were no significant differences in hemoglobin, serum ferritin, RBC folic acid, serum vitamin B-12, serum vitamin A, and plasma vitamin C concentrations and EGRAC between the 2 treatment groups (Table 2).
View this table:
TABLE 2. Baseline anthropometric, sociodemographic, and biochemical characteristics of the subjects who completed the study protocol, by treatment group1
After 12 wk of supplementation, mean changes in hemoglobin, serum ferritin, and serum vitamin B-12 were not significantly different between the 2 treatment groups (Table 3). Mean changes in RBC folic acid, serum vitamin A, and plasma vitamin C concentrations were significantly higher, whereas the mean change in EGRAC was significantly lower (p < 0.001) in the MMN group than in the IFA group.
View this table:
TABLE 3. Mean increase in hemoglobin and other biochemical measures of micronutrient status after 12 wk of supplementation in the subjects who completed the study protocol1
At baseline, the prevalence of anemia and other micronutrient deficiencies was not significantly different between the 2 treatment groups (Table 4). After 12 wk of supplementation, the prevalence of anemia and iron and folic acid deficiencies decreased significantly in both treatment groups. However, the differences between the 2 groups were not statistically significant. At the end of the study, the prevalence of vitamins A and C and riboflavin deficiencies decreased significantly only in the MMN group, whereas the prevalence of vitamin deficiencies remained unchanged in the IFA group, except for vitamin B-12 deficiency, which increased significantly in this group. After supplementation, the prevalences of vitamins A and C and riboflavin deficiencies were significantly lower in the MMN group than in the IFA group. Compared with the IFA group, the odds ratio for being deficient in riboflavin or vitamin A after the supplementation period decreased by 73% and 53%, respectively, in the MMN group. However, the decrease was statistically significant only for riboflavin (P = 0.005).
View this table:
TABLE 4. Prevalence of anemia and biochemical deficiencies of selected micronutrients at baseline and at the end of supplementation in the subjects who completed the study protocol1
DISCUSSION
In the present study, we found that after 12 wk of intermittent (twice weekly) supplementation, both the MMN and IFA groups had significantly higher hemoglobin and serum ferritin concentrations. However, the mean increases in hemoglobin and serum ferritin were not statistically significantly different between the 2 treatment groups. The response to treatment had 2 important features: 1) despite a sizeable reduction in the prevalence of anemia and iron deficiency, determined on the basis of serum ferritin concentrations after the supplementation protocol, there were no significant differences between the 2 groups, and 2) although there was an overall reduction in the prevalence of anemia and iron deficiency, nearly two-thirds of the girls still remained anemic, and a substantial proportion remained iron deficient in both treatment groups after 12 wk of supplementation. This suggests that twice-weekly doses of MMN supplements as proposed by UNICEF/WHO/UNU (20) may not be an effective strategy for rectifying iron deficiency and anemia in Bangladeshi adolescent girls.
Two studies in pregnant women (21, 22) and 1 study in nonpregnant women (24) also found no added benefit of MMN supplements in improving hemoglobin and iron status. One group argues that this lack of benefit could be due to the addition of zinc and iron in a ratio of 1:2 in the MMN group (21), because there is a potential competitive interaction between zinc and iron (32). In our study, although the MMN supplements contained zinc and iron at a ratio of 1:2—because the study was not designed to test the effect of zinc on iron uptake—we were unable to make any judgement on the above explanation. On the other hand, the lack of additional benefit of MMN supplementation to hemoglobin formation and iron status may be explained by some other reasons.
One of the reasons could be the use of twice-weekly supplementation in both treatment groups with a dose of 30 mg Fe, which was equivalent to the recommended dietary allowance (RDA) for Indian adolescent girls (33). Given the fact that iron intake through diet was grossly inadequate in these girls, twice-weekly supplementation with an RDA-equivalent dose of iron meant that they met requirements only for 2 d/wk, and, thus, iron was still limiting and failed to show any additional effect of MMN supplementation on hematopoiesis. This is supported by the fact that a substantial proportion of the girls remained iron deficient at the end of the study.
The second possible explanation for the lack of additional effects may have been the low prevalence of other micronutrient deficiencies in the study participants. However, this possibility was unlikely because the baseline data clearly indicated that these girls had a range of micronutrient deficiencies, including iron deficiency. A related possibility is that the MMN supplements fail to improve the micronutrient status other than iron. Until now, little has been known about the effect of MMN supplementation on the status of various micronutrients, except for iron and vitamin A (21, 22, 34). This is important for understanding the underlying factors involved in the hematologic response to MMN supplements, especially in populations in whom concurrent MMN deficiencies are common. Therefore, we also examined whether twice-weekly MMN supplements improved the status of other micronutrients. At the end of the trial, there was a significantly higher mean increase in RBC folic acid in the MMN group than in the IFA group, although both groups were given an equal dose of folic acid, which was 4 times the RDA for adolescent girls (33). The mechanism of this effect is not clear. The other micronutrients contained in the supplement were reasonably well absorbed, as indicated by the significantly higher mean increase in serum vitamin A and plasma vitamin C concentrations and lower mean change in EGRAC in the MMN group than in the IFA group. The decrease in EGRAC in the MMN group after supplementation suggests an improvement in riboflavin status. Angeles-Agdeppa et al (34) supplemented female Indonesian adolescents with MMNs containing iron, folic acid, and vitamins C and A for 12 wk, given daily or weekly, and observed a significant improvement in vitamin A status compared with placebo. Previously, we also found a significant increase in serum vitamin A concentrations in anemic girls who received a weekly dose of iron, folic acid, and vitamin A for 12 wk compared with iron and folic acid only or placebo (12).
After 12 wk of supplementation, the prevalences of vitamin A, riboflavin, and vitamin C deficiencies decreased by 43%, 32%, and 100%, respectively, in the MMN group. Similarly, the odds ratios indicated that the girls in the MMN group were less likely to be deficient in vitamin A and riboflavin than were the girls in the IFA group. These results indicate that MMN supplements can substantially reduce vitamins A and C and riboflavin deficiencies in this population. Furthermore, the prevalence of vitamin B-12 deficiency was found to increase significantly in the IFA group, whereas there was only a slight but nonsignificant increase in the MMN group. This indicates that twice-weekly MMN supplements for 12 wk may not be able to reduce vitamin B-12 deficiency but can protect from a worsening of the deficiency. Thus, our study showed that MMN supplements can be beneficial in improving the status of some important micronutrients.
Because folic acid, riboflavin, and vitamins A, C, and B-12 are known to have a direct or indirect role in hematopoiesis (3, 5-8), and because we found a significant reduction in the prevalence of a low status for various micronutrients after MMN supplementation for 12 wk, it is reasonable to expect to see some improvement in the status of hemoglobin and iron in the girls who received MMN supplements. Instead, we observed no additional effect of MMN supplementation on the status of hemoglobin and iron. It is important to note that a sizeable proportion of the girls still remained deficient in several micronutrients after 12 wk of MMN supplementation (Table 4). It may be possible that the additional effect of MMN supplements would only be seen after micronutrient status, including iron, had reached optimal levels.
Anemia may be caused by many nonnutritional factors, such as malaria, hookworms, chronic infections, and hemoglobinopathies (4, 35). The risk of malaria is low in Bangladesh, being restricted to hilly areas. In the present study we excluded any subject with overt chronic infections, and, at baseline, we examined the worm load of the subjects and observed only a few cases of mild hookworm infection. However, we do not have any information on the trends of hemoglobinopathies in the study population. Thus, there is no evidence to suggest that other nonnutritional causes of anemia provide an alternative explanation for the results observed here.
In conclusion, twice-weekly supplementation with MMNs for 12 wk is not more efficacious than is supplementation with iron and folic acid alone in improving the hematologic status of anemic adolescent girls. However, even though the effect on anemia and iron status was not as expected, the other benefit of MMN supplementation in combination with iron and folic acid is the enhancement of the status of other micronutrients. Furthermore, our findings suggest that more frequent doses, possibly even a daily dose, are required to replete micronutrient status and thereby permit optimum hematopoiesis.
ACKNOWLEDGMENTS
We thank UniMed and UniHealth Manufacturers Limited, Dhaka, for providing the MMN and iron folate preparations and the Biomedical Research Group of BIRDEM, Dhaka, for helping with the RBC folate and vitamin B-12 assays.
FA took the lead in study planning and design, provided guidance on the data collection and analysis of the data, and wrote the manuscript. MRK supervised the blood collection and laboratory analysis and contributed to the study planning and writing of the manuscript. MA coordinated the field work and contributed to the study planning and writing of the manuscript. RK supervised the fieldwork and contributed to the study planning and the writing of the manuscript. GCM contributed to the study design and the writing of the manuscript. CPB was responsible for the blood collection and laboratory analysis and contributed to the writing of the manuscript. BN was responsible for the field data collection and contributed to the writing of the manuscript. GW contributed to the data analysis and interpretation. None of the authors had any financial or personal conflict of interest.
REFERENCES