| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research Communication |
1 From the Division of Nutrition and Physical Activity, National Center for Chronic Disease Prevention and Health Promotion, Centers for Disease Control and Prevention, Atlanta (MEC, IP, and RY); the MetroHealth Medical Center, Supplemental Nutrition Program for Women, Infants, and Children, Cleveland (LI); UNICEF, Beijing (RY); and the Department of Pediatrics, Columbia University College of Physicians and Surgeons, New York (GMB).
2 Supported by the US Department of Health and Human Services, Centers for Disease Control and Prevention (Cooperative Agreement no. U50/CCU 50855) and the National Institutes of Health (grant no. RO1 HL52447).
3 Address reprint requests to ME Cogswell, Division of Nutrition and Physical Activity, MS K-25, CDC, 4770 Buford Highway NE, Atlanta, GA 30341-3717. E-mail: mec0{at}cdc.gov.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Objective: We tested the hypothesis that administration of a daily iron supplement from enrollment to 28 wk of gestation to initially iron-replete, nonanemic pregnant women would reduce the prevalence of anemia at 28 wk and increase birth weight.
Design: Between June 1995 and September 1998, 513 low-income pregnant women in Cleveland were enrolled in the study before 20 wk of gestation. Of these, 275 had a hemoglobin concentration 
110 g/L and a ferritin concentration 20 µg/L and were randomly assigned to receive a monthly supply of capsules containing either 30 mg Fe as ferrous sulfate or placebo until 28 wk of gestation. At 28 and 38 wk of gestation, women with a ferritin concentration of 12 to < 20 µg/L or < 12 µg/L received 30 and 60 mg Fe/d, respectively, regardless of initial assignment. Almost all the women received some supplemental iron during pregnancy. We obtained infant birth weight and gestational age at delivery for 117 and 96 of the 146 and 129 women randomly assigned to receive iron and placebo, respectively.
Results: Compared with placebo, iron supplementation from enrollment to 28 wk of gestation did not significantly affect the overall prevalence of anemia or the incidence of preterm births but led to a significantly higher mean (± SD) birth weight (206 ± 565 g; P = 0.010), a significantly lower incidence of low-birth-weight infants (4% compared with 17%; P = 0.003), and a significantly lower incidence of preterm low-birth-weight infants (3% compared with 10%; P = 0.017).
Conclusion: Prenatal prophylactic iron supplementation deserves further examination as a measure to improve birth weight and potentially reduce health care costs.
Key Words: Iron deficiency • anemia • iron supplementation • pregnancy • low birth weight • small-for-gestational age infants • preterm delivery
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
100 g/L (1, 2). Specifically, evidence about beneficial effects of iron supplementation during pregnancy on functional outcomes is inconclusive. Moreover, the theoretical possibility of adverse effects, such as oxidative damage, with administration of iron supplements during pregnancy has been raised (3).
In 1993, the Institute of Medicine (IOM) proposed a complex program of selective iron supplementation during pregnancy, which was based on screening pregnant women in their first and second trimesters by using hemoglobin and ferritin concentrations (Table 1
) (4). In particular, the IOM recommended that iron supplements not be given to women who are nonanemic (hemoglobin concentration 110 g/L) and iron replete (ferritin concentration > 20 µg/L). Women with anemia or decreased iron stores would receive different doses of iron depending on their hemoglobin or ferritin concentrations. In the third trimester, all women would receive iron supplements, but the dose would differ by hemoglobin concentration. Recognizing the lack of available evidence, the IOM recommended that the guidelines be evaluated before implementation (4). The Centers for Disease Control and Prevention and the American College of Obstetricians and Gynecologists recommend a daily iron supplement (30 mg) as prophylaxis for iron deficiency during pregnancy (5, 6). In the present study, we tested the hypothesis that administration of a daily iron supplement from enrollment to 28 wk of gestation to initially iron-replete, nonanemic pregnant women would reduce the prevalence of anemia at 28 wk of gestation. During the study, we added birth weight and gestational age as outcomes because of a need for more information on the functional consequences of iron supplementation during pregnancy (1, 2).
|
|
SUBJECTS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Enrollment
The purpose and design of the study was explained to each eligible participant in writing (fact sheet written by GMB) and verbally by a WIC dietitian (LI) trained in the informed consent procedure, with emphasis on the requirement for random allocation to receive no supplemental iron or standard iron supplementation. Among the WIC enrollees, 513 agreed to participate in the study. No compensation was given for participation. Initial venous blood samples (5 mL) were drawn, and each volunteer was given an appointment for her first study visit within 10 d of enrollment. The protocol required participants to visit the WIC clinic monthly and to have blood drawn at 28 and 38 wk of gestation. At their monthly visits to the WIC clinic, the participants returned their supplement bottles with any unused capsules to the WIC dietitian (LI), who asked the participants about any side effects that they had experienced.
Iron and hematologic measurements
Iron and hematologic studies, including complete blood count, ferritin, and erythrocyte protoporphyrin, were conducted on the blood samples at enrollment, 28 wk of gestation, and 38 wk of gestation. Measurement of erythrocyte protoporphyrin, however, was added after the start of the study, and thus data are not available for all the subjects. After collection in the WIC clinic, the blood samples were sent immediately by a pneumatic tube carrier transport system to the Hematology Research Laboratory, which is housed in an adjacent section of the building. The samples were immediately removed from the transport system and refrigerated at 4 °C until analysis (whole blood or red cell studies) or separation of the blood components (serum, plasma) for storage at -70 °C. Complete blood counts and zinc erythrocyte protoporphyrin measurements were carried out within 4 h of receipt. Serum iron studies and ferritin measurements were carried out in the order of receipt in batches of 
100 samples per assay after the first thawing of previously frozen specimens.
Hemoglobin concentrations and mean corpuscular volumes were measured with an electronic counter (Coulter Electronics, Hialeah, FL). Plasma ferritin concentrations were measured by using enzyme-linked immunosorbent assay (Ramco Laboratories, Inc, Houston). Zinc erythrocyte protoporphyrin was measured fluorimetrically (ZnP Model 4000 Hematofluorimeter; Environmental Sciences Associates, Inc, Bedford, MA). An internal quality control system ensured that appropriate controls were run with each assay and that the calibration of spectrophotometers, the hematofluorimeter, balances, pipettes, and other equipment was periodicly verified.
Randomization
The women were randomly allocated to treatment groups on the basis of random numbers generated by the study data manager using a computerized algorithm (STATVIEW II; Abacus Concepts, Inc, Berkeley, California). Of the 513 women who gave initial blood samples, 275 had a hemoglobin concentration 110 g/L and a ferritin concentration 20 µg/L (Figure 1). These women were randomly allocated to receive by mouth each day soft gelatin capsules containing either 30 mg Fe as ferrous sulfate (n = 146) or placebo (n = 129) until 28 wk of gestation. Of the 238 women not included in this report, 95 had a ferritin concentration of 12 to < 20 µg/L and were randomly assigned to receive daily capsules containing 30 or 60 mg Fe, 126 had a ferritin concentration < 12 µg/L and were prescribed 60 mg Fe/d, 15 were referred for medical evaluation (because they were anemic with a serum ferritin concentration > 20 µg/L), and 2 were missing data for ferritin or hemoglobin concentration.
|
Reassessments at 28 and 38 wk of gestation
At 28 wk of gestation, venous blood samples were drawn, and iron and hematologic studies were conducted. The women were assigned to the prescriptions listed in Table 2 according to their hemoglobin and serum ferritin concentrations, and capsules were dispensed in a bottle containing a 31-d supply at the monthly clinic visit. This procedure was repeated at 38 wk of gestation.
|
Main outcomes
The primary outcome was the prevalence of anemia (hemoglobin concentration < 110 g/L) at the 28-wk visit. We also examined the distributions of birth weight, birth length, and gestational age and the proportion of infants who had a low birth weight (< 2500 g), were born preterm (< 37 wk of gestation), were born at term but had a low birth weight ( 37 wk of gestation and < 2500 g), and were small-for gestational age (7). We compared means and SDs between term births because this approach provides another way to compare fetal growth between groups (8). We also examined the proportion of infants who were small and preterm based on 1) a birth weight < 2500 g and a gestation of < 37 wk, and 2) their "residual birth weight" (8).
Subsidiary outcomes
Subsidiary outcomes included average maternal weight gain, iron status at 28 wk of gestation, adherence to iron supplementation, and side effects. Iron-status measurements included mean cell volume and hemoglobin, ferritin, and erythrocyte protopophyrin concentrations. For this study, "iron deficiency anemia" was defined as a hemoglobin concentration < 110 g/L and a ferritin concentration < 12 µg/L, and "absent iron stores" was defined as a ferritin concentration < 12 µg/L.
Adherence to the supplementation regimen was calculated on the basis of the number of capsules remaining in the bottle at each return visit:
 | (1) |
We also calculated the proportion of women who took > 50% of the capsules received. Total iron taken was defined as the sum of the number of capsules taken times the amount of iron per capsule.
Side effects were determined at each visit by the WIC dietitian (LI), who asked the subjects, "Did you have any complaints or difficulty with taking the capsules?" The women who responded positively to this question at 1 visit were recorded as having experienced side effects. Some women who returned for monthly visits did not bring their bottles but answered questions about side effects, and thus the proportion of women with data on side effects was larger than that with data on adherence.
Sample size and power
Sample size calculations for the study were originally based on a relative difference of 50% between the 2 groups in the prevalence of anemia during the third trimester, ie, 30% compared with 15%. Although the prevalence of anemia at 28 wk of gestation among women in the study population who were not exposed to prenatal iron supplements was unknown, the prevalence of anemia during the third trimester among low-income women enrolled in WIC was 30% before this study. On the basis of 2 independent samples with a two-sided significance level of 0.05 and a ß-error specification of 0.20, 120 subjects per group were needed (9). On the assumption of an overall rate of loss to follow-up of 10%, 133 subjects per group were required (266 nonanemic, iron-replete women). Actual loss to follow up (Table 2
) affected the power of the study. With 86 women in the placebo group who had hemoglobin data at 28 wk of gestation, the statistical power to find a 50% difference between the 2 groups in anemia risk at 28 wk of gestation at a two-tailed probability of < 0.05 was 40% (10).
We also hypothesized that the 2 study groups would differ in their distributions of birth weight and gestational age. With 96 women in the placebo group (n = 192 in the study population), and if the true difference between the 2 treatments was 0.407 times the SD, the probability of detecting a difference between the 2 groups at a two-sided significance level of 5% was 80% (10). Given that the SDs for mean birth weight and gestational age at delivery in the study sample were 573 g and 2.2 wk, respectively, the study had 80% power at a two-sided significance level of 5% to detect differences in mean birth weight and mean gestational age of 233 g and 0.9 wk, respectively. The minimum detectable difference may actually have been a bit smaller because the group that was provided with iron supplements from enrollment to 28 wk of gestation actually included 117 women. Because infants born to mothers who smoke during pregnancy weigh, on average, 150–300 g less than do those born to mothers who do not smoke (11), we thought that the difference in birth weight described above (ie, 233 g) was clinically significant.
Statistical analyses
We analyzed the outcomes on the basis of the randomization of the study participants. Because of losses to follow-up, we did not have data on third-trimester iron status, infant birth weight, gestational age at delivery, or adherence for some of the women. We compared the distributions of sociodemographic, health, and behavioral characteristics and of initial iron-status measures between the women with missing data and those with complete data. We also compared the reasons for missing data by randomization group. Among the women with third-trimester iron-status measures, we compared average iron-status measures between the 2 study groups and calculated the absolute difference between the 2 groups in the proportion of women with third-trimester anemia, low iron stores, and iron deficiency anemia. Among women with data on infant birth weight, we compared distributions of birth weight, gestational age, birth length, and maternal weight gain during pregnancy between the 2 study groups. We also calculated the absolute difference between the 2 groups in the risks of low birth weight, preterm delivery, small-for-gestational age, preterm delivery with low birth weight, and term delivery with low birth weight.
We used t tests to compare continuous outcomes between the iron and placebo groups. Because the distributions of ferritin concentrations were skewed, we log-transformed ferritin concentrations before comparing the means with a t test. The results were then retransformed into antilogarithms to recover the original units and were expressed as geometric means and percentiles in the tables. We used chi-square tests and Fisher exact tests, where appropriate, to compare differences in categorical outcomes (12). To adjust for factors that differed by randomization group, we calculated mean differences in continuous outcomes by using multiple linear regression, and we adjusted odds ratios for dichotomous outcomes by using multiple logistic regression. We used SAS (version 8; SAS Institute Inc, Cary, NC) for all analyses (13).
To determine whether missing data biased our results, we compared our complete case analysis (n = 213) with that obtained by using a 3-step multiple imputation process (n = 275) (14, 15). We created 40 imputed datasets, 10 for each of 4 theoretical scenarios: 1) the women who were missing birth-weight data took the prescribed supplement or placebo, 2) regardless of initial treatment group assignment, the women who were missing birth-weight data did not take iron supplements, 3) regardless of initial treatment group assignment, the women who were missing birth-weight data took iron supplements, and 4) the women who were missing birth-weight data switched treatments (ie, the women assigned to receive iron supplements did not take them, and the women assigned to receive placebos took iron supplements instead). With multiple imputation, data are assumed to be missing at random (ie, the lack of data depends on the observed variables only, including supplemental iron on the basis of 1 of the 4 theoretical scenarios), and we felt that our data did not violate this assumption. Variables used to impute missing data included the following: theoretical supplementation after assignment (on the basis of 1 of the 4 scenarios presented above), gestational age at study entry, natural log of initial ferritin concentration, initial hemoglobin concentration, maternal age at last menstrual period, ethnic group, parity, smoking status at entry into WIC, prepregnancy weight (in kg), and adherence to the supplementation regimen. We used the PROC MI and PROC MIANALYZE procedures in the SAS software with a Markov chain Monte Carlo Approach to account for the complicated nature of the missing data (13–15).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
). There were no other significant differences between the groups in demographic characteristics, behavioral characteristics, or initial iron status (Table 3).
|
and Table 2). Compared with the women with complete data, the women without data on either of these 2 outcomes entered the study, on average, 1.5 wk earlier in pregnancy (P = 0.001) and were more likely to be black and non-Hispanic (36.2% compared with 19.4%; P = 0.0025). Initial iron status and other maternal characteristics did not differ significantly between these groups. The proportion of women with missing data and the reasons for missing data did not differ significantly between the 2 treatment groups (P = 0.198, chi-square test).
At 28 wk of gestation, 196 women returned for iron measures (Table 2). For most of these women, the dose of supplemental iron was increased at 28 wk of gestation. Among 110 women who were initially assigned to receive 30 mg Fe/d and who returned for iron measures at 28 wk of gestation, 62 (56%) were without iron stores and were prescribed 60 mg Fe/d, 25 (23%) had depleted iron stores and continued to receive 30 mg Fe/d, 18 (16%) were iron replete and continued to receive 30 mg Fe/d (Table 2), and 5 (5%) were referred for medical evaluation. Among 86 women who were initially assigned to receive placebo and who returned for iron measures at 28 wk of gestation, 56 (65%) were without iron stores and were prescribed 60 mg Fe/d, 11 (13%) were iron depleted and were prescribed 30 mg Fe/d, 15 (17%) were iron replete and continued to receive placebo, and 5 (6%) were referred for medical evaluation. None of the women had a hemoglobin concentration < 90 g/L at 28 wk of gestation. At 38 wk of gestation, 144 women returned for iron measures. Only one woman treated with placebo throughout her pregnancy remained nonanemic and iron replete (data not shown).
Seventy-three women were missing adherence data. Compared with the women with adherence data, those without adherence data entered the study, on average, 1.2 wk earlier in pregnancy (P = 0.019). Otherwise, maternal characteristics did not differ significantly between these 2 groups. Adherence to the supplementation regimen between the initial visit and 28 wk of gestation did not differ significantly between the 2 treatment groups (Table 4). The women who were randomly assigned to receive iron took a mean of 2 g Fe between the initial study visit and 28 wk of gestation. By delivery, the women who were initially assigned to receive iron had taken a mean (± SD) of 2.9 ± 1.8 g Fe, and the women who were initially assigned to receive placebo had taken a mean of 0.8 ± 1.1 g Fe (P < 0.0001).
|
). The associations of iron supplementation with absent iron stores and iron deficiency anemia were stronger after adjustment for prepregnancy weight and log of initial ferritin concentration: the prevalence of absent iron stores and of iron deficiency anemia was 14.3 percentage points (P = 0.031) and 10 percentage points (P = 0.062) lower, respectively, among the women assigned to receive iron supplements than among those assigned to receive placebo.
|
). The distribution of gestational age at birth was also shifted to the right among the women assigned to receive iron from enrollment to 28 wk of gestation, with the curves for the 2 treatment groups meeting at 36 wk of gestation (Figure 3). Ten infants born to women in the iron supplement group were delivered at 36 wk of gestation, whereas only 2 infants were born to women in the placebo group at 36 wk of gestation. Compared with the infants born to the women in the placebo group, the infants born to the women in the iron supplement group were significantly heavier (by a mean of 206 g) and significantly less likely to have a low birth weight (Table 6). Most of the difference between the 2 treatment groups in the incidence of low birth weight was accounted for by the difference in the incidence of small preterm infants. When the incidence of small preterm infants was estimated by analysis of residual birth-weight distributions (data not shown), small preterm infants were born to 1.6% of the women in the iron supplement group and to 12.7% of the women in the placebo group (P = 0.004). Although the difference between the 2 treatment groups in the proportion of small-for-gestational age infants was significant, the difference in the proportion of term infants with low birth weight was not. Among the women who gave birth at term ( 37 wk of gestation), the mean (± SD) birth weight of the infants born to the women who received iron supplements from enrollment to 28 wk of gestation was 118 ± 460 g higher than that of the infants born to the women who received placebo (P = 0.0836). These data also suggest that most of the difference between the 2 treatment groups in the proportion of small-for-gestational age infants occurred preterm. The overall difference in preterm births between the 2 groups was not significant, primarily because most of the births at 36 wk of gestation occurred in the iron supplement group (Figure 3). Adjustment for self-reported prepregnancy weight and the log of initial ferritin concentration did not significantly change the associations of iron supplementation with infant birth outcomes. After adjustment for the log of initial ferritin concentration and prepregnancy weight, the odds ratio for low birth weight among the infants born to the women in the iron supplement group was 0.24 (95% CI: 0.08, 0.68) relative to that of the infants born to the women in the placebo group. The corresponding odds ratios adjusted for log of initial ferritin concentration and prepregnancy weight from the multiple imputation analyses of the 275 women and infants were 0.25–0.34.
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
A recent comprehensive review concluded that virtually all published intervention studies had major flaws that preclude definitive conclusions about the effect of iron supplementation on birth outcomes (16). In our study, the randomization, the use of the placebo control until 28 wk of gestation, and double blinding eliminated many factors (eg, internal biases and confounding) other than iron that could have potentially explained the higher observed birth weight in the iron supplement group. Nonetheless, our study also has limitations.
Despite randomization, the women who were assigned to the placebo group had significantly higher prepregnancy weight and initial iron stores than did the women who were assigned to the iron supplement group, and these differences could contribute to an underestimate of the effect of iron supplementation. Although adjustment for these characteristics made no difference with respect to the main outcomes, we found larger reductions among the iron supplement group in both the prevalence of absent iron stores and iron deficiency anemia in the third trimester after adjustment for prepregnancy weight and initial iron stores.
We did not find any differences between the 2 treatment groups in the proportion of women lost to follow-up or in the reasons for the loss to follow-up, but the women who were lost to follow-up entered the study earlier in gestation than did those who were not lost to follow-up and were more likely to be black than were those who were not lost to follow-up. We used these factors as well as others to conduct a sensitivity analysis based on low-birth-weight data multiply imputed for women with missing data by using 4 assumptions for actual iron supplementation (14, 15) (n = 275) and found that the reductions in low birth weight in the iron supplement group were substantial regardless of the assumption.
Although self-reported birth-weight data have previously been shown to be accurate (17), the reporting of the last menstrual period has not been evaluated among women enrolled in WIC. It is possible that errors in reporting may have affected our estimates of gestational age and attenuated the differences in the incidence of preterm delivery. We found that most of the births at 36 wk (n = 12) occurred among the women who received iron supplements (n = 10), and all of the infants resulting from these births weighed 2500 g. When we defined small preterm infants by using residual birth weight (8), we found larger reductions in the incidence of small preterm infants in the iron supplement group than when we defined small preterm infants as infants who were born at < 37 wk of gestation and who had a birth weight < 2500 g. Because the analysis of residual birth weight does not include gestational age, it may be more accurate.
In our study, the rates of low birth weight (9.9%) and preterm delivery (12.7%) were higher than those in the US population during the study years of 1995–1998 (ie, 7.3–7.6% and 11.0–11.6%, respectively) (18). Although the women were enrolled in our study before 20 wk of gestation, MetroHealth Medical Center acts as the referral center for women with high-risk pregnancies, and a substantial proportion of the women in our population were smokers (36%), thus increasing the risk of adverse outcomes. We did not have data on pregnancy complications (eg, hypertension) in the study population. Even with randomization, it is possible that the rates of pregnancy complications were higher in the placebo group than in the iron supplement group and that these differences accounted for the difference in the proportion of small preterm infants. However, the rates of smoking and gestational weight gain were not significantly different between the 2 treatment groups, suggesting that these 2 factors at least did not account for the difference in the proportion of small preterm infants.
Little is known about the pathophysiologic means whereby iron supplementation may affect the regulation of gestation and fetal growth. Iron deficiency and iron deficiency anemia may lead to changes in factors such as norepinepherine concentration (19, 20), cortisol and corticotropin-releasing hormone concentrations (21), and indexes of oxidative stress that may adversely affect gestation, fetal growth, or both (21). In the future, researchers should consider measuring these factors.
In our study, iron supplementation had a strong effect on birth weight, but the effect on anemia during the third trimester was not significant. Most of the cases of anemia appeared to be due to iron deficiency in the iron supplement group (12.7/19.8 x 100 = 64%) and in the placebo group (20.9/26.7 x 100 = 78%), but the incidence of iron deficiency anemia in the iron supplement group was also not significantly lower than that in the placebo group. Adjustment of the results for initial ferritin concentration and prepregnancy weight suggested that, compared with placebo, iron supplementation from enrollment to 28 wk of gestation led to a significantly lower proportion of women with low iron stores at 28 wk; however, even with iron supplementation, most of the women developed low iron stores. How can iron supplementation have little to no effect on iron status at 28 wk of gestation but increase infant birth weight?
It is possible that iron supplements may be preferentially transferred to the placenta and fetus, thus contributing to higher birth weight rather than to higher maternal iron stores. However, the mechanisms for the preferential transfer of iron and for higher birth weight as a result of increased iron to the fetus are unknown. Maternal iron deficiency in rats results in compensatory changes in the iron transport mechanisms of the placenta, which in turn minimizes the level of iron deficiency in the fetus (22), but how this applies to humans is uncertain. In the present study, 30 mg Fe in the form of supplements may have been enough to meet fetal requirements but may not have been enough to maintain maternal stores.
Other mechanisms may have little to do with iron deficiency. Iron supplementation may improve the appetite of the mother (23), thus increasing energy consumption and resulting in increased intrauterine growth. Our study did not include measures of energy consumption. Another explanation is that iron supplementation may lead to increased plasma volume expansion, and thus in the placebo group in our study, plasma volume may have failed to expand adequately. Although this seems like a plausible explanation, one study indicated that iron supplements resulted in increased red blood cell volume, but not plasma volume, during pregnancy (24).
The 1993 IOM recommendations to prevent iron deficiency anemia (4) are complex and based on the assumption that measurements of hemoglobin and of ferritin (which is costly to measure) provide reliable means of identifying women who would benefit from iron supplementation during the first and second trimester of pregnancy. In our study, most of the initially iron-replete, nonanemic women had decreased or absent iron stores by 28 wk of gestation. Indeed, only one woman treated with placebo throughout her pregnancy remained nonanemic and without iron depletion. Most importantly, in contrast to the US Preventive Services Task Force (1, 2) and IOM recommendations (4), our study provides evidence that infants may benefit substantially from maternal iron supplementation beginning early in gestation. Iron supplementation of initially nonanemic pregnant women without iron depletion from the first trimester to 28 wk of gestation may have important benefits by preventing small preterm births. Because small preterm births are a major determinant of perinatal morbidity and mortality (8), iron supplementation could reduce health care costs. Furthermore, accumulating evidence suggests that in adult life, infants with low birth weight may be at greater risk of various chronic disorders, including type 2 diabetes, hypertension, and coronary artery disease (25, 26). Iron supplementation during pregnancy deserves further examination as a measure to improve birth outcomes and reduce health care costs.
|
ACKNOWLEDGMENTS |
|---|
GMB, LI, IP, and RY designed the randomized controlled trial. GMB, MEC, and IP designed the birth-weight study. LI enrolled the participants, dispensed the study drugs, performed pill counts, and asked subjects about their side effects. MEC led the data analysis. All investigators contributed to data interpretation and the writing of the manuscript. None of the authors had any conflicts of interest.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONSUBJECTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  E. R. Kabyemela, M. Fried, J. D. Kurtis, T. K. Mutabingwa, and P. E. Duffy Fetal Responses during Placental Malaria Modify the Risk of Low Birth Weight Infect. Immun., April 1, 2008; 76(4): 1527 - 1534. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. B Belfort, S. L Rifas-Shiman, J. W Rich-Edwards, K. P Kleinman, E. Oken, and M. W Gillman Maternal iron intake and iron status during pregnancy and child blood pressure at age 3 years Int. J. Epidemiol., April 1, 2008; 37(2): 301 - 308. [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]
|
|
|
|
 |
|
 K. Tolentino and J. F. Friedman An Update on Anemia in Less Developed Countries Am J Trop Med Hyg, July 1, 2007; 77(1): 44 - 51. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. S. Jarrah, J. O. Halabi, A. E. Bond, and J. Abegglen Iron Deficiency Anemia (IDA) Perceptions and Dietary Iron Intake Among Young Women and Pregnant Women in Jordan J Transcult Nurs, January 1, 2007; 18(1): 19 - 27. [Abstract] [PDF]
|
|
|
|
 |
|
 G. R. Lubach and C. L. Coe Preconception Maternal Iron Status Is a Risk Factor for Iron Deficiency in Infant Rhesus Monkeys (Macaca mulatta) J. Nutr., September 1, 2006; 136(9): 2345 - 2349. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. S. Andersen, L. Gambling, G. Holtrop, and H. J. McArdle Maternal Iron Deficiency Identifies Critical Windows for Growth and Cardiovascular Development in the Rat Postimplantation Embryo J. Nutr., May 1, 2006; 136(5): 1171 - 1177. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. S Golub, C. E Hogrefe, A. F Tarantal, S. L Germann, J. L Beard, M. K Georgieff, A. Calatroni, and B. Lozoff Diet-induced iron deficiency anemia and pregnancy outcome in rhesus monkeys Am. J. Clinical Nutrition, March 1, 2006; 83(3): 647 - 656. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Alderman and J. R. Behrman Reducing the Incidence of Low Birth Weight in Low-Income Countries Has Substantial Economic Benefits World Bank Res. Obs., March 1, 2006; 21(1): 25 - 48. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Z. Mei, M. E. Cogswell, I. Parvanta, S. Lynch, J. L. Beard, R. J. Stoltzfus, and L. M. Grummer-Strawn Hemoglobin and Ferritin Are Currently the Most Efficient Indicators of Population Response to Iron Interventions: an Analysis of Nine Randomized Controlled Trials J. Nutr., August 1, 2005; 135(8): 1974 - 1980. [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]
|
|
|
|
|
|
T. O Scholl Iron status during pregnancy: setting the stage for mother and infant Am. J. Clinical Nutrition, May 1, 2005; 81(5): 1218S - 1222S. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Racasan, B. Braam, H. A. Koomans, and J. A. Joles Programming blood pressure in adult SHR by shifting perinatal balance of NO and reactive oxygen species toward NO: the inverted Barker phenomenon Am J Physiol Renal Physiol, April 1, 2005; 288(4): F626 - F636. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L Gambling, H. S Andersen, A Czopek, R Wojciak, Z Krejpcio, and H. J McArdle Effect of timing of iron supplementation on maternal and neonatal growth and iron status of iron-deficient pregnant rats J. Physiol., November 15, 2004; 561(1): 195 - 203. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Vahratian, A. M. Siega-Riz, D. A. Savitz, and J. M. Thorp Jr. Multivitamin Use and the Risk of Preterm Birth Am. J. Epidemiol., November 1, 2004; 160(9): 886 - 892. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. G. Ronnenberg, R. J. Wood, X. Wang, H. Xing, C. Chen, D. Chen, W. Guang, A. Huang, L. Wang, and X. Xu Preconception Hemoglobin and Ferritin Concentrations Are Associated with Pregnancy Outcome in a Prospective Cohort of Chinese Women J. Nutr., October 1, 2004; 134(10): 2586 - 2591. [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]
|
|
|
|
|
|
K. M Rasmussen and R. J Stoltzfus New evidence that iron supplementation during pregnancy improves birth weight: new scientific questions Am. J. Clinical Nutrition, October 1, 2003; 78(4): 673 - 674. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||
