| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research Communication |
1 From the Department of International Health, Rollins School of Public Health, Emory University, Atlanta (UR and RM), and the Centro de Investigación en Nutrición y Salud, Instituto Nacional de Salud Pública, Cuernavaca, Mexico (TG-C, LMN, and JR).
2 Presented at the International Union of Nutritional Sciences Congress, August 27–September 1, 2001, Vienna.
3 Supported by the Thrasher Research Fund; UNICEF, New York; Conacyt; and the Instituto Nacional de Salud Pública, Mexico.
4 Address reprint requests to U Ramakrishnan, Emory University, Rollins School of Public Health, 1518 Clifton Road, NE, Atlanta, GA 30322. E-mail: uramakr{at}sph.emory.edu.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Objective: We conducted a randomized, double-blind clinical trial in semirural Mexico to compare the effects of multiple micronutrient (MM) supplements with those of iron supplements during pregnancy on birth size.
Design: Pregnant women (n = 873) were recruited before 13 wk of gestation and received supplements 6 d/wk at home, as well as routine antenatal care, until delivery. Both supplements contained 60 mg Fe, but the MM group also received 1–1.5 times the recommended dietary allowances of several micronutrients.
Results: At recruitment, the women in the 2 groups were not significantly different in age, parity, economic status, height, or hemoglobin concentration but differed significantly in marital status (4.6% and 2.0% of women in the MM and iron-only groups, respectively, were single mothers) and mean (± SD) body mass index (in kg/m2; 24.6 ± 4.3 and 23.8 ± 3.9 in the iron-only and MM groups, respectively). Losses to follow-up (25%) and compliance (95%) did not differ significantly between the groups. In intent-to-treat analyses (MM group: n = 323; iron-only group: n = 322), mean (± SD) birth weight (2.981 ± 0.391 and 2.977 ± 0.393 kg in the MM and iron-only groups, respectively) and birth length (48.61 ± 1.82 and 48.66 ± 1.83 cm in the MM and iron-only groups, respectively) did not differ significantly between the groups.
Conclusion: These findings suggest that MM supplementation during pregnancy does not lead to greater infant birth size than does iron-only supplementation.
Key Words: Iron • multivitamins • minerals • supplements • pregnancy • intrauterine growth retardation • Mexico • birth size
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The possibility that improvements in dietary quality rather than in the quantity of food consumed during pregnancy may be effective in reducing fetal growth retardation has generated considerable excitement among international agencies. A few years ago, the results of a prospective, observational cohort study among teenage mothers in the United States suggested that the use of multiple vitamin and mineral supplements reduces the risk of preterm delivery and LBW (9). A study using a double-blind, placebo-controlled design among HIV-infected but asymptomatic women (n = 1067) in Tanzania found that multivitamin supplements dramatically reduce preterm delivery (39%) and LBW (44%) (10). The prevalence of LBW in the 2 groups that received multivitamin supplements with or without vitamin A was only 
9%, whereas the prevalence of LBW in the groups that received vitamin A or placebo was 14.5% and 17.2%, respectively. There was no effect of vitamin A, and the mean birth weight of the infants whose mothers received multivitamins was 120 g higher than that of those whose mothers received placebo. Note that all the women received routine iron folate supplements but that none of the supplements contained zinc, which was shown to improve birth weight in some studies (7). Motivated by these results, UNICEF and the World Bank introduced multiple vitamin and mineral supplements in programs for pregnant women and efforts to evaluate the effect of these supplements on birth outcomes, but no results from these efforts are yet available (11). We conducted a double-blind, randomized controlled trial to compare the effects of multiple micronutrient (MM) supplements with those of iron-only supplements during pregnancy on birth outcomes in a semirural community in Mexico.
|
SUBJECTS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
for the actual composition of the supplement) and was similar to supplements that are commercially available, whereas the supplement received by the control group contained only iron, which was consistent with the standard practice of the Ministry of Health in Mexico at the time the study was conducted (13). Both supplements contained 60 mg Fe in the form of ferrous sulfate.
|
25% of the subjects assigned to each color; pregnant women were allocated to the preassigned color code as they were added to this list at the time of recruitment. All study personnel and investigators were blinded to the group assignment, the details of which were kept at Emory University and the INSP in sealed envelopes that were opened only after preliminary data analysis was completed.
Data collection
At recruitment, a prenatal examination that included a detailed obstetric history, physical examination, anthropometric assessment, and blood work was carried out by the study physician and a team of trained nurses at the study headquarters. The first supplement was consumed at the study headquarters, after which trained workers visited the women’s homes 6 d/wk until delivery to administer supplements and record their consumption. Socioeconomic status was determined by using a questionnaire that included questions regarding education, ethnicity, water and sanitation, quality of housing, household size, occupation, and possessions such as a television set, radio, or bicycle; an index of economic status was derived from these data by using factor analysis similar to that used in previous studies in this setting (14). All subjects were asked to come to the study headquarters for routine prenatal care visits at 26, 32, and 37 wk of pregnancy and at 1 mo after delivery, during which anthropometric measurements and blood samples were also obtained. Anthropometric measurements of height, weight, triceps and thigh skinfold thicknesses, and midupper arm circumference were obtained by highly trained anthropometrists using standard procedures (15). Venous blood samples were collected at baseline and at 32 wk of gestation, and biochemical measurements of micronutrient concentrations (serum ferritin, retinol, folate, and zinc) are currently under way. Hemoglobin concentrations in capillary blood samples obtained by finger prick were measured at the field headquarters by using a HemoCue (HemoCue Inc, Mission Viejo, CA). Appropriate referral and treatment for high-risk pregnancies were provided by the study physician, who worked closely with the local health authorities. All births were followed up by a trained field worker who obtained anthropometric measurements (weight, length, and head circumference) at birth (< 72 h) either at home or at the hospital. Cord blood samples were also obtained in a subsample. All the infants were also followed up until 3 mo of age, and infant growth and feeding practices were monitored at 1, 2, and 3 mo of age. Data entry and checking for errors were carried out on an ongoing basis with supervision by the INSP staff at the INSP headquarters in Cuernavaca. Additional checking of data for errors was carried out at Emory University.
Data analysis
The main outcome variables were measures of birth size (birth weight, length, and ponderal index) and gestational age based on recall of the last menstrual period. In addition, the incidences of preterm delivery (< 37 wk of gestation), LBW (< 2.5 kg), intrauterine growth retardation (IUGR), and intrauterine stunting (IUS) were also compared between the 2 treatment groups. IUGR and IUS were defined as values below the 10th percentile for the birth weight and birth length, respectively, appropriate for the infant’s gestational age and sex from reference data from Miller and Hassanein (16). Anemia and overweight at recruitment were defined by using the World Health Organization definitions of hemoglobin concentration < 110 g/L and BMI > 25, respectively (17, 18). Compliance was calculated by expressing the total number of tablets consumed as a percentage of the total number of study days in which supplements could have been consumed. We used an intent-to-treat design in which all pregnancies assigned to treatment between July 1997 and 31 December 1999 were included, and the effectiveness of randomization was tested by comparing the 2 groups for selected sociodemographic, health, and nutrition characteristics at recruitment. Comparisons between the final sample with information about birth outcomes and the sample lost to follow-up were also performed for selected baseline characteristics and measures of compliance. The analysis for birth size outcomes was restricted to singleton live births. All comparisons were made by using Student’s t tests of means for normally distributed variables and chi-square tests of proportions for categorical variables. Adjusted analyses with multivariate techniques (general linear models and logistic regression) were done to control for factors that differed between groups. In addition, effect modification by characteristics selected before data analysis (maternal BMI at recruitment, infant sex, and socioeconomic status) was tested for birth weight. All statistical analyses were conducted by using SAS 6.12 (SAS Institute Inc, Cary, NC). Because some of the women had > 1 pregnancy, comparisons of the main outcomes of interest were made by using repeated-measures analysis (SAS PROC MIXED). A P value < 0.05 was used for all group comparisons.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
indicates that the 2 treatment groups did not differ significantly in most of the characteristics, including age, parity, number of weeks pregnant at study entry, hemoglobin concentration, and maternal height. Almost one-third of the women were overweight, with a significantly higher proportion in the iron-only group (38.8%) than in the MM group (32.3%). Approximately one-third of the pregnancies were primiparous, and the proportion of women who had > 1 pregnancy included in the study was 12.6%. The prevalence of anemia at recruitment did not differ significantly between the groups. Because socioeconomic status was measured during the course of the study, data were not available in all cases, especially among women lost to follow-up. However, the groups did not differ significantly in maternal schooling, ethnicity, and economic status. Approximately three-fourths of the women had completed primary school, and 10% had more than a secondary school education. Approximately 30% of the women lived in homes that were made of good-quality construction material. Although most of the women had partners, the proportion of single mothers was significantly higher in the MM group than in the iron-only group.
|
did not indicate any significant differences in baseline characteristics. As expected, compliance and the total number of supplements consumed during the study were significantly lower among the pregnancies lost to follow-up than among the final sample. Weight and length measurements at birth were available for 98.1% (n = 633) and 94.1% (n = 607) of the pregnancies, respectively; these measurements were taken within 24 h of birth, and > 75% of the infants were measured in the first hour in both the MM and iron-only groups. Although 55% of births occurred in area hospitals, 89% of all birth measurements were obtained by the project-trained anthropometrist, and the percentage of birth measurements made by the project-trained anthropometrist did not differ significantly between the 2 groups. In the final sample of pregnancies with known birth outcomes, the groups did not differ significantly in several characteristics, including age, hemoglobin concentration, and compliance, but significant differences in prepregnant weight, BMI, and marital status remained.
|
. There were no significant differences between the groups in any of these outcomes. The overall incidences of LBW, preterm delivery, IUGR, and IUS were 8.7%, 7%, 10.9%, and 13.0%, respectively. The overall incidence of postterm deliveries (> 41 wk of gestation) was 12.3%, and the incidence of postterm deliveries did not differ significantly between the groups. Because of the observed differences in maternal BMI and marital status, the comparisons were also adjusted for these 2 variables, but the results remained unchanged, indicating that MM supplementation in this population had no overall beneficial effect compared with that of iron-only supplementation. Because some women had > 1 pregnancy, repeated-measures analysis was used, and the adjusted mean (± SE) differences between the MM and iron-only groups in birth weight, birth length, and ponderal index were 16.1 ± 31 g (P = 0.60), -0.002 ± 0.151 cm (P = 0.99), and 0.014 ± 0.018 kg/cm3 (P = 0.42), respectively. The adjusted odds ratios for the MM group compared with the iron-only group for LBW, IUGR, IUS, and preterm deliveries were 0.98 (95% CI: 0.55, 1.74), 0.82 (95% CI: 0.49, 1.39), 0.96 (95% CI: 0.59, 1.57), and 1.16 (95% CI: 0.60, 2.23), respectively. Testing for interactions that were specified a priori showed no selective effects of treatment by maternal overweight (BMI at recruitment > 25), infant sex, or tertiles of economic status.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
320 per group, exceeding the stipulated sample sizes. A sample size of 250 per group has more than 80% power to detect meaningful effect sizes (0.23 SD) on birth size (100 g and 2.5 cm for birth weight and birth length, respectively) and gestational age (5 d). This sample size also has > 80% power to detect a 50% reduction (effect size = 0.28) in the incidence of LBW, IUGR, and IUS, with the assumption of a baseline value of 20% in the control group. However, calculations made after the study showed that the final sample sizes, despite being larger than the stipulated sizes, had inadequate power (< 80%) to detect a reduction of 50% in LBW, IUGR, IUS, and prematurity, because the incidence of these outcomes was lower than expected (23).
One possible explanation for the lack of effects is that the extent of micronutrient deficiencies and poor birth outcomes may not have been severe enough to see an effect. Indeed, preliminary analyses suggest that the supplements may not have been efficacious in improving the biochemical status of zinc, iron, and vitamin A (24–26). The explanation for the lack of effects is not that the study population was free of deficiencies. Although the prevalence of anemia in our sample was lower than expected (27), preliminary results showed that iron, zinc, and folate deficiencies each affected 30–40% of the women and that more than one-half of the subjects had 
2 micronutrient deficiencies (28). The prevalence of LBW in our study (8.7%) represents one-half that reported for the control group in the Tanzanian study (17.2%; 10), in which improvements were observed; mean birth weights however, were similar in both the Tanzanian study and our study. As noted earlier, we had expected slightly higher rates of LBW on the basis of earlier estimates (29). Perhaps the provision of high-quality and regular antenatal care, including referrals and treatment of infections, combined with high compliance may have contributed to lower incidences of LBW and anemia. However, antenatal care was provided in Tanzania as well. Other differences between the 2 studies were as follows: 1) the subjects in the Tanzanian study were HIV-positive, which may have affected their response, and 2) the inclusion of zinc in our supplement [zinc supplementation was recently shown in Bangladesh (30) and Peru (31) not to improve birth size] may have interfered with the absorption of other nutrients in the supplement (32).
In summary, the reasons for the lack of improvement in birth size in our study are not clear, and much remains unknown about the effects of MM supplements during pregnancy. The results of other similar trials that are currently underway in different study settings are needed before reaching a conclusion on the role of MM supplements in improving birth size. Even though there were no improvements in birth size, there may have been other benefits such as improvements in maternal nutritional status, maternal and child micronutrient status, and child growth and development. These are being investigated in our ongoing studies of postnatal development. For example, although zinc supplementation during pregnancy failed to reduce the incidence of LBW in Bangladesh, the intervention was associated with reduced morbidity during the first 6 mo of infancy (33). In conclusion, the findings of the present study suggest that among relatively healthy women living in a semirural community in a developing country that is in transition, multivitamin and mineral supplementation during pregnancy does not improve birth size in comparison with iron-only supplementation.
|
ACKNOWLEDGMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  F. Tofail, L. A. Persson, S. El Arifeen, J. D Hamadani, F. Mehrin, D. Ridout, E.-C. Ekstrom, S. N Huda, and S. M Grantham-McGregor Effects of prenatal food and micronutrient supplementation on infant development: a randomized trial from the Maternal and Infant Nutrition Interventions, Matlab (MINIMat) study Am. J. Clinical Nutrition, March 1, 2008; 87(3): 704 - 711. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S Palma, R Perez-Iglesias, D Prieto, R Pardo, J Llorca, and M Delgado-Rodriguez Iron but not folic acid supplementation reduces the risk of low birthweight in pregnant women without anaemia: a case control study J. Epidemiol. Community Health, February 1, 2008; 62(2): 120 - 124. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. A Haws, A. L Thomas, Z. A Bhutta, and G. L Darmstadt Impact of packaged interventions on neonatal health: a review of the evidence Health Policy Plan., July 1, 2007; 22(4): 193 - 215. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. W. Fawzi, G. I. Msamanga, W. Urassa, E. Hertzmark, P. Petraro, W. C. Willett, and D. Spiegelman Vitamins and Perinatal Outcomes among HIV-Negative Women in Tanzania N. Engl. J. Med., April 5, 2007; 356(14): 1423 - 1431. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Gupta, M. Ray, T. Dua, G. Radhakrishnan, R. Kumar, and H. P. S. Sachdev Multimicronutrient Supplementation for Undernourished Pregnant Women and the Birth Size of Their Offspring: A Double-blind, Randomized, Placebo-Controlled Trial Arch Pediatr Adolesc Med, January 1, 2007; 161(1): 58 - 64. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. A. Haws and G. L. Darmstadt Integrated, Evidence-Based Approaches to Save Newborn Lives in Developing Countries Arch Pediatr Adolesc Med, January 1, 2007; 161(1): 99 - 102. [Full Text] [PDF]
|
|
|
|
 |
|
 M. A. Orjuela, L. Titievsky, X. Liu, M. Ramirez-Ortiz, V. Ponce-Castaneda, E. Lecona, E. Molina, K. Beaverson, D. H. Abramson, and N. E. Mueller Fruit and Vegetable Intake during Pregnancy and Risk for Development of Sporadic Retinoblastoma Cancer Epidemiol. Biomarkers Prev., June 1, 2005; 14(6): 1433 - 1440. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. H Allen Multiple micronutrients in pregnancy and lactation: an overview Am. J. Clinical Nutrition, May 1, 2005; 81(5): 1206S - 1212S. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Z. A. Bhutta, G. L. Darmstadt, B. S. Hasan, and R. A. Haws Community-Based Interventions for Improving Perinatal and Neonatal Health Outcomes in Developing Countries: A Review of the Evidence Pediatrics, February 1, 2005; 115(2/S1): 519 - 617. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. E Moriarty-Craige, U. Ramakrishnan, L. Neufeld, J. Rivera, and R. Martorell Multivitamin-mineral supplementation is not as efficacious as is iron supplementation in improving hemoglobin concentrations in nonpregnant anemic women living in Mexico Am. J. Clinical Nutrition, November 1, 2004; 80(5): 1308 - 1311. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Friis, E. Gomo, N. Nyazema, P. Ndhlovu, H. Krarup, P. Kaestel, and K. F Michaelsen Effect of multimicronutrient supplementation on gestational length and birth size: a randomized, placebo-controlled, double-blind effectiveness trial in Zimbabwe Am. J. Clinical Nutrition, July 1, 2004; 80(1): 178 - 184. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 U. Ramakrishnan, L. M. Neufeld, T. Gonzalez-Cossio, S. Villalpando, A. Garcia-Guerra, J. Rivera, and R. Martorell Multiple Micronutrient Supplements during Pregnancy Do Not Reduce Anemia or Improve Iron Status Compared to Iron-Only Supplements in Semirural Mexico J. Nutr., April 1, 2004; 134(4): 898 - 903. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
U. Ramakrishnan Nutrition and low birth weight: from research to practice Am. J. Clinical Nutrition, January 1, 2004; 79(1): 17 - 21. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Christian Micronutrients and Reproductive Health Issues: An International Perspective J. Nutr., June 1, 2003; 133(6): 1969S - 1973. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
