Daily Multivitamins With Iron to Prevent Anemia in High-Risk Infants: A Randomized Clinical Trial Paul L. Geltman, Alan F.

Meyers, Supriya D. Mehta, Carlo Brugnara, Ivan Villon, Yen A. Wu and Howard Bauchner Pediatrics 2004;114;86-93 DOI: 10.1542/peds.114.1.86

The online version of this article, along with updated information and services, is located on the World Wide Web at: http://www.pediatrics.org/cgi/content/full/114/1/86

PEDIATRICS is the official journal of the American Academy of Pediatrics. A monthly publication, it has been published continuously since 1948. PEDIATRICS is owned, published, and trademarked by the American Academy of Pediatrics, 141 Northwest Point Boulevard, Elk Grove Village, Illinois, 60007. Copyright 2004 by the American Academy of Pediatrics. All rights reserved. Print ISSN: 0031-4005. Online ISSN: 1098-4275.

Downloaded from www.pediatrics.org. Provided by Indonesia:AAP Sponsored on April 15, 2011

Daily Multivitamins With Iron to Prevent Anemia in High-Risk Infants: A Randomized Clinical Trial
Paul L. Geltman, MD, MPH*; Alan F. Meyers, MD, MPH*; Supriya D. Mehta, PhD, MHS; Carlo Brugnara, MD; Ivan Villon, MD*; Yen A. Wu, MPH*; and Howard Bauchner, MD*
ABSTRACT. Objective. The goal of this study was to assess the effectiveness of multivitamins with iron as prophylaxis against iron deficiency and anemia in infancy. Methods. The study was a double-blinded, randomized, pragmatic, clinical trial conducted at 3 urban primary care clinics. Subjects included healthy, full-term infants who were enrolled at their 6-month well-child visit. Infants were randomly assigned to receive standard-dose multivitamins with or without iron (10 mg/ day). Parents administered multivitamins by mouth daily for 3 months. Laboratory results at 9 months of age were analyzed for the presence of anemia and/or iron deficiency. Anemia was defined as hemoglobin level <11.0 g/dL. Iron deficiency was initially defined as any abnormal laboratory value of the following: mean corpuscular volume combined with red cell distribution width or zinc protoporphyrin (with blood lead level <10 g/dL) for most subjects and ferritin, transferrin saturation, or reticulocyte hemoglobin content for a subset. Subsequent analyses defined iron deficiency as any 2 abnormalities of the above laboratory outcomes, except hemoglobin. Results. The control (n 138) and intervention (n 146) groups were equivalent with respect to all important sociodemographic and nutritional variables. At 9 months of age, anemia was found in 21% of infants (n 58). A total of 229 (81%) had iron deficiency on the basis of 1 abnormal laboratory indicator and 139 (49%) on the basis of 2 abnormal laboratory indicators. No difference existed in the occurrence of anemia and iron deficiency between the intervention and control groups. In the intervention group, 22% and 78% of 138, respectively, were anemic or had 1 abnormal laboratory outcome indicative of iron deficiency. In the control group, 19% and 84% of 144 were anemic or iron deficient. When stratified by adherence, no differences in hematologic outcomes between groups were noted. However, in multivariate logistic regression, infants whose mothers were anemic during pregnancy were 2.15 times more likely than others to have any laboratory abnormality (95% confidence interval: 1.14 4.07). Increasing adherence, regardless of group assignment, was associated with a 0.56 times reduced risk of any abnormality (95% confidence interval: 0.41 0.76). Conclusion. On the basis of intention-to-treat analysis, multivitamins with iron was not effective in preventing iron deficiency or anemia in 9-month-old infants. However, effective prevention and treatment of maternal anemia during pregnancy and giving multivitamins with or without additional iron during infancy may prove to be important approaches to the prevention of iron deficiency among high-risk children. Because of the consequences of iron deficiency and its high prevalence among low-income infants, additional investigation in these areas is warranted. Pediatrics 2004;114:86 93; nutrition, anemia, iron deficiency, vitamins, pediatrics.
ABBREVIATIONS. WIC, Special Supplemental Nutrition Program for Women, Infants, and Children; Hb, hemoglobin; MCV, mean corpuscular volume; RDW, red cell distribution width; Fer, ferritin; TS, transferrin saturation; CHr, reticulocyte hemoglobin content; CDC, Centers for Disease Control and Prevention; OR, odds ratio; CI, confidence interval.

From the *Department of Pediatrics, Boston University School of Medicine and Boston Medical Center, Boston, Massachusetts; Department of Emergency Medicine, Boston University School of Medicine and Boston Medical Center, Boston, Massachusetts; and Departments of Pathology and Laboratory Medicine, Harvard Medical School and Childrens Hospital, Boston, Massachusetts. Received for publication Jan 2, 2003; accepted Jan 9, 2004. Reprint requests to (P.G.) Whittier Street Health Center, 1125 Tremont St, Roxbury, MA 02120. E-mail: pgeltman@bu.edu PEDIATRICS (ISSN 0031 4005). Copyright 2004 by the American Academy of Pediatrics.

ron deficiency is the most common nutrient deficiency and cause of anemia in childhood.1 Iron deficiency is associated with numerous adverse health effects, particularly cognitive impairment, that may be irreversible2,3 and has been shown to be a precursor of lead poisoning.4 In the United States, an estimated 720 000 (9%) children aged 1 to 2 years are iron deficient, 240 000 (3%) of whom are anemic.5,6 These conditions disproportionately affect low-income children. Despite reports of declining prevalence of iron deficiency anemia among young children,68 data from the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC)9 and local studies10 suggest that rates of iron deficiency and anemia of 15% to 19% persist among low-income infants and toddlers. Few recent US studies have assessed clinic-based approaches to primary prevention using supplementation with medicinal iron. Instead, studies have focused on primary prevention using dietary interventions, with varying results.1114 The beneficial effects of ferrous sulfate supplementation have been supported by other recent research conducted overseas. Domellof et al15 found a reduction in the prevalence of anemia in healthy, full-term, breastfed Honduran infants who were given supplemental iron from 4 or 6 months to 9 months of age but not among Swedish infants, who had a much lower baseline prevalence of anemia. At 9 months of age, 9% of the supplemented Honduran infants were anemic, compared

86

PEDIATRICS Vol. 114 No. 1 July 2004 Downloaded from www.pediatrics.org. Provided by Indonesia:AAP Sponsored on April 15, 2011

with 29% of the infants who received placebo. Similarly, Dijkhuizen et al16 found a 38% reduction of the prevalence of anemia in healthy, full-term Indonesian infants who received iron supplementation compared with infants who received placebo. Because vitamin supplements are widely used among preschool children,17 we believed that parents would be receptive to administering a daily vitamin supplement to their infants. In addition, a recent pediatric commentary in Pediatric News18 called for universal iron prophylaxis for infants and toddlers in lieu of current recommendations of the American Academy of Pediatrics to screen and treat anemic children.19 From 1996 to 1998, we conducted a randomized clinical trial of the use of multivitamins with iron as prophylaxis against iron deficiency anemia in infants from 6 to 9 months of age.20 We chose this age range because full-term infants typically exhaust maternally derived iron stores by then.21 In this study, at 9 months of age, anemia was documented in 11% of the infants in the intervention group who received multivitamins with iron and in 19% in the control group who received multivitamins without iron. After stratifying by adherence, infants in the intervention group were 50% less likely to be anemic. This previous study, however, was limited by poor adherence, loss to follow-up, and limited laboratory data with which to assess iron deficiency. Because of these limitations, we conducted this randomized clinical trial.
METHODS
The study design was a randomized, double-blinded, clinical trial of multivitamins with iron supplementation of infants. Infants between 5 and 7 months of age were identified through clinic registration records at 3 primary care sites, the Pediatric Primary Care Clinic at Boston Medical Center (Boston, MA), the Uphams Corner Health Center (Dorchester, MA), and the Whittier Street Health Center (Roxbury, MA). The study was approved and monitored by the Institutional Review Board of Boston Medical Center/Boston University Medical Center. Written, informed consent was obtained for all subjects. This study was designed to be consistent with current clinical practice at each site so as to simulate how the intervention would be implemented in actual clinical practice. No laboratory studies were done at enrollment because this is not routinely done in clinical practice and our pilot studies indicated that enrollment would be greatly reduced by parental refusal to permit this additional phlebotomy. No additional patient visits were added to the standard protocols for well-child visits. Parents were reminded about visits and called with reminders after missed appointments as is normally done at these clinics. Sample size calculations were based on rates of anemia documented in our first study, with an expected reduction of the rate of anemia by 50% among the infants who received multivitamins with iron. Using a prevalence of 22% and assuming an of .05 and power of 80%, 196 subjects were needed in each group. At their 6-month well-child visits, all infants who received care at the study sites were approached and screened for enrollment in the study. Infants with premature birth (37 weeks); low birth weight (2500 g); major medical, hematologic, or gastrointestinal conditions; and previous supplement use were excluded from the study. At enrollment, parents were interviewed regarding demographics and dietary history, including age in days, gender, race, birth weight in pounds, duration of breastfeeding, use of whole cow milk, introduction of meats, use of iron-fortified cereal, use of iron-fortified formula, receipt of WIC benefits, and education by their pediatrician regarding iron-rich foods. We assessed household food security and hunger with the short form of the US Department of Agriculture Household Food Security Scale.22 The infants were randomized to 2 study groups. Vitamins were

repackaged and relabeled to maintain blinding with randomization coordinated by the Boston Medical Center Pharmacy. Research assistants enrolled subjects at each site into the 2 study groups. Randomization was implemented through previous computer-generated randomized assignment of subject numbers to each group by the pharmacy. Blinding of the researchers, research assistants, and parents was maintained throughout the study. The group assignment listing was kept in a sealed envelope until study completion. The intervention group received standard, nonprescription infant multivitamin drops with iron. Each milliliter of the vitamins contained the following: elemental iron, 10 mg; vitamin A, 1500 IU; vitamin C, 35 mg; vitamin D, 400 IU; vitamin E, 5 IU; thiamine, 0.5 mg; riboflavin, 0.6 mg; niacin, 8 mg; vitamin B6, 0.4 mg; and vitamin B12, 2 g. The control group received the same multivitamins without iron. Parents were instructed to administer 1 mL of the multivitamins by mouth to their infants each day. Combined with other dietary sources of iron, it was estimated that infants who received the multivitamins with iron would receive 2 mg/kg/day elemental iron. Parents were given a 3-month supply of vitamins. At the 9-month well-child visit, hematologic outcomes were measured. Given high rates of missed and delayed appointments, all infants who had specimens drawn between 9 and 12 months of age were included in analyses. These outcomes included hemoglobin (Hb), zinc protoporphyrin, blood lead, mean corpuscular volume (MCV), and red cell distribution width (RDW). For a subset of the infants from whom sufficient blood could be obtained, additional hematologic outcomes included ferritin (Fer), transferring saturation (TS), and reticulocyte hemoglobin content (CHr). Because the infants who had the additional tests conducted were evenly distributed between both study groups, we used the full number of subjects with a Hb level in each group for reporting the prevalences of summary outcomes, such as any lab abnormality. Cutoff values for Hb, RDW, MCV, Fer, and TS followed recommendations of the US Centers for Disease Control and Prevention (CDC).1 The CDC guidelines are the standard of care for primary care practice in the United States and suggest the following cutoff values for children younger than 2 years: Hb 11 g/dL, RDW 14.5%, MCV 77 fL, Fer 15 g/L, and TS 16%. Although some recent research literature5,15 cites lower cutoff values for Fer, TS, and MCV, we chose to use the current CDC guidelines that are intended for clinical practice, particularly primary care. In defining anemia for black infants, no adjustments of Hb values were made; the CDC does not recommend adjustment of Hb cutoff because the basis of variation in Hb distribution by race has not been determined.1 The zinc protoporphyrin cutoff (35 g/dL with a blood lead 10 g/dL) followed widely used guidelines from the Childhood Lead Poisoning Prevention Laboratory, Massachusetts Department of Public Health. CDC guidelines do not address CHr; therefore, the CHr cutoff (27 pg) was chosen on the basis of published reference data from the laboratory at which the testing was performed.23 The definition of iron deficiency followed commonly accepted criteria that are used in clinical practice at our study sites. The CDC guidelines, although defining cutoff points for abnormal tests of iron sufficiency, do not offer definitions of iron deficiency based on these tests, alone or in combination, with the exception of low serum Fer concentration.1 MCV and RDW were used as a combined measure to minimize the effect of thalassemia trait in interpreting results. In applying a more rigorous definition of iron deficiency anemia, we also determined the prevalence of iron deficiency on the basis of any 2 abnormal laboratory values. We analyzed TS and CHr both as separate indicators of iron deficiency and as a combined measure. In addition, we analyzed available prenatal maternal Hb values nearest the delivery date using CDC trimester-specific cutoff values to define anemia.1 Maternal Hb levels can fluctuate as a result of hemodilution near term; however, we chose these near-term values to have greater consistency in the timing of the testing during the pregnancy. During monthly reminder telephone calls, research assistants assessed adherence by parental report. Adherence was measured by verbal report with the following categories: refused to answer, vitamins given every day (6 7 days per week), vitamins given usually (4 5 days per week), vitamins given sometimes (13 days per week), or vitamins never given. Full adherence was defined as

ARTICLES Downloaded from www.pediatrics.org. Provided by Indonesia:AAP Sponsored on April 15, 2011

87

vitamins given every day, and partial adherence was defined as vitamins given usually or vitamins given sometimes. Although we attempted to assess adherence via bottle return, nearly all parents failed to return bottles. Primary analyses were intentto-treat; subsequent analyses included only those with full adherence and then any known adherence. Adherence level was analyzed as none, sometimes, usually, and every day, with each category representing a unit increase over the next. Analysis included comparisons of means using t tests and 2 tests from bivariate 2 2 table analyses. Results were analyzed for all children who completed the study (and had complete blood counts tested, 282 of 284 with laboratory test results) and then stratified on the basis of adherence. Using multivariate logistic regression and stratified analyses, we examined the associations between baseline data and adherence with abnormal hematologic outcomes in the infants. Analyses were conducted using STATA 7.0 for Windows (Stata Corp, College Station, TX).

RESULTS

Research assistants approached 585 infants for enrollment between February 2000 and July 2001. Of these, 126 did not meet eligibility criteria: 57 were already taking vitamins; 35 were born prematurely; 23 had low birth weight; 5 had major medical problems, notably sickle cell anemia; 5 had acute medical problems (2 with severe constipation. 1 with active tuberculosis. and 2 seen when sick); and 1 exceeded the enrollment age (57 months). Of 459 eligible infants, 76 declined to participate, 5 moved before their 6-month visit, 1 reported that their physician advised against participation, and 1 did not speak English or Spanish. Therefore, 376 infants were enrolled in the study. (Fig 1) The 376 enrolled infants and the 76 whose families declined to participate did not differ by gender, maternal educational level, percentage born in the

United States, or percentage of mothers born in the United States; however, white and Asian families were less likely to participate than black or Latino families (P .04). Of 22 eligible white infants and 9 Asian infants, only 14 (64%) and 6 (67%), respectively, were enrolled. In comparison, 83% of black and 87% of Latino infants were enrolled. A total of 179 infants were in the intervention group, and 197 were in the control group. A total of 284 (76%) infants, 138 (76%) in the intervention group and 146 (74%) in the control group, had laboratory results obtained between 9 and 12 months of age and were included in analyses. Those with laboratory test results that were analyzed did not differ from those without results except that those with results were more likely to receive WIC, a finding that is expected, given WIC requirements for anemia testing on recertification at 1 year of age. Of the families in the intervention group, 16 moved, were lost to follow-up, or refused blood sampling, and 25 did not have laboratory values obtained until after 12 months of age; in the control group, 21 and 30, respectively, were not included in analyses. Two of 284 children with laboratory tests done between 9 and 12 months did not have a complete blood count, leaving 146 infants in the control group and 138 in the intervention group. The control and intervention groups were similar with respect to all sociodemographic characteristics, including pregnancy history and household food security, and infant characteristics, including age, gender, race, breastfeeding, WIC enrollment, and birth

Fig 1. Patient flow diagram.

88

TRIAL OF MULTIVITAMINS WITH IRON TO PREVENT ANEMIA IN INFANTS Downloaded from www.pediatrics.org. Provided by Indonesia:AAP Sponsored on April 15, 2011

weight. In addition to being evenly distributed between study group, gender did not affect any hematologic outcome. The only difference between groups was that infants in the intervention group were more likely than control subjects to be accompanied by their mothers (95% vs 85%, P .03; Table 1).
Hematologic Outcomes

ence level, no statistically significant differences in any hematologic outcomes between study groups were noted (Table 4).
Hematologic Outcomes Based on Maternal Hb Level and Adherence Level

Of the 282 infants with laboratory tests obtained between 9 and 12 months, only 44 (16%) had completely normal hematologic values: 24 (17%) in the intervention group and 20 (14%) in the control group. Thus, more than four fifths (238) of these 282 infants were either anemic or possibly iron deficient. A total of 229 (81%) infants had iron deficiency defined as 1 laboratory outcome abnormality except Hb, and 139 (49%) infants had iron deficiency defined as 2 abnormalities. Anemia was noted in 58 (21%). Anemia with 2 other laboratory outcome abnormalities was noted in 22 (8%). The occurrence of anemia and iron deficiency was statistically similar in the 2 groups. No statistically significant differences in any of the other iron deficiency measures, either alone or in combination, emerged between intervention and control groups (Tables 2 and 3). When assessing adherence to the daily dosing regimen by parental report, parents of 111 (41%) reported giving the vitamins every day, 59 (22%) usually, and 60 (22%) sometimes. Rates of full adherence did not differ between intervention and control groups (full adherence of 40.6% and 41.0%, respectively). When results were stratified by adherTABLE 1.

We found high rates of maternal anemia during pregnancy, noted in the mothers of 99 (43%) of 232 infants for whom maternal Hb values were obtained. Low maternal Hb was significantly associated with an increased risk of any abnormal laboratory test value (odds ratio [OR]: 2.24; 95% confidence interval [CI]: 1.23 4.06) and a tendency toward iron deficiency (OR: 1.65; 95% CI: 0.98 2.79). In addition, preliminary analyses revealed an association between increasing categorical level of adherence and improved hematologic outcomes in the infants in both study groups (Table 5). A categorical increase in adherence level had a corresponding decrease in risk of any abnormal laboratory test value (OR: 0.55; 95% CI: 0.42 0.72) and Hb 11.0 g/dL (OR: 0.67; 95% CI: 0.51 0.87). In a multivariate logistic regression model controlling for study group assignment, the associations between low maternal Hb level and risk of any abnormal laboratory test value remained statistically significant, as did increasing adherence level (Table 6).
DISCUSSION

In our intent-to-treat analysis, prescribing prophylactic multivitamins with iron at 6 months of age did not reduce the prevalence of either iron deficiency or

Descriptive Data of 284 Infants Included in Analyses by Group Assignment Variable Intervention Group (Multivitamins With Iron, n 138) 185 65 76 3.40 39.6 95% (131) 20% (28) 51% (70) 35% (48) 15.2 93% (128) 90% (124) 77% (106) 38% (52) 90% (124) 0% (0) 92% (127) 23% (32) 95% (131) 28% (38) 16% (22) Control Group (Multivitamins With No Iron, n 146)* 183 73 70 3.39 39.8 85% (124) 24% (35) 44% (64) 29% (42) 16.2 90% (131) 88% (128) 77% (113) 32% (47) 93% (136) 0% (0) 96% (140) 18% (27) 90% (132) 22% (32) 12% (18)

Demographics Age (d; mean) Gender (n; female) Race (n; black) Birth weight (kg; mean) Gestational age (wk; mean) Mother accompanying infant Mother with 12 y of school Mother with 12 y of school Mother married Pregnancy history Weight gain (kg; mean) Self-reported use of vitamins and/or iron Received WIC during pregnancy Infant nutrition history Ever breastfed Breastfed at enrollment Formula fed at enrollment Cow milk in diet at enrollment Rice cereal with iron in diet at enrollment Meat in diet at enrollment Enrolled in WIC program Household receiving Food Stamps Household with food insecurity or hunger

* P .05 by 2 analyses except for mother accompanying infant (P 03).

ARTICLES Downloaded from www.pediatrics.org. Provided by Indonesia:AAP Sponsored on April 15, 2011

89

TABLE 2.

Distribution of Laboratory Values* Intervention Group (Multivitamins With Iron) Control Group (Multivitamins With No Iron) 11.7 (11.511.8) 78.0 (77.078.7) 14.2 (14.014.6) 32 (3034) 32 (26.638.5) 16.5 (14.818.8) 28.4 (27.829.1) 11.8 (11.612.0) Total 11.7 (11.611.8) 77.9 (77.078.1) 14.2 (14.014.3) 31 (3033) 31 (27.636.8) 18.1 (16.019.5) 28.2 (27.828.7) 11.9 (11.812.1)

Hb MCV RDW ZP Fer Transferrin CHr Maternal Hb

11.7 (11.511.8) 77.7 (76.078.8) 14.1 (13.714.3) 30.5 (2933) 29.2 (25.537.7) 19.0 (16.621.0) 28.0 (27.528.6) 12.0 (11.812.3)

ZP indicates zinc protoporphyrin. * Mean and 95% CI are presented for normally distributed data (Hb and maternal Hb); median and 95% CI obtained by binomial method are presented for nonnormally distributed data (MCV, RDW, ZP, Fer, Transferrin, and CHr). Differences in laboratory values between groups were tested by t test for normally distributed data and by nonparametric K-sample test for equality of medians for nonnormally distributed data. All P .05. TABLE 3. Counts* Hematologic Outcomes by Group Enrollment for 282 Infants With Complete Blood Intervention Group (Multivitamins With Iron) Hematologic outcomes Any abnormal laboratory results Iron deficiency only (1 abnormality) Iron deficiency only (2 abnormalities) Hb 11.0 g/dL Hb 11.0 g/dL and 2 other abnormal laboratory values Iron deficiency measures MCV 77 fL RDW 14.5% ZP 35 g/dL Fer 15 g/L TS 16% CHr 27 pg Combined measures Abnormal MCV RDW Abnormal CHr TS * All P .05 by 2 analyses. 83% (114/138) 78% (108/138) 48% (66/138) 22% (31/138) 8% (11/138) 44% (61/138) 38% (52/138) 35% (46/132) 17% (16/96) 37% (36/97) 29% (28/97) 25% (34/138) 14% (14/97) Control Group (Multivitamins With No Iron) Total

86% (124/144) 84% (121/144) 51% (73/144) 19% (27/144) 8% (11/144) 39% (56/144) 42% (60/144) 39% (54/137) 12% (11/95) 48% (45/93) 27% (26/95) 19% (28/144) 16% (15/93)

84% (238/282) 81% (229/282) 49% (139/282) 21% (58/282) 8% (22/282) 41% (117/282) 40% (112/282) 37% (100/269) 14% (27/191) 43% (81/190) 28% (54/192) 22% (62/282) 15% (29/190)

anemia in 9-month-old infants. Additional analyses indicated that infants whose mother was anemic during pregnancy were twice as likely as other infants to be anemic at 9 months of age and that adherence with multivitamin supplementation, regardless of iron content, was associated with a reduction in abnormal hematologic measures. Our finding of an association between maternal prenatal anemia and iron deficiency in the infant has not been previously reported in the United States and deserves additional study. Limited data from developing countries suggest that a causal link exists,2428 and a biologically plausible mechanism has been documented in one study.29 The high prevalence of prenatal anemia among low-income women has persisted, with the CDCs Pregnancy Nutrition Surveillance System reporting a third-trimester pregnancy rate of 29% among women enrolled in WIC, a level that has not changed during the previous 7 years of monitoring.30 If maternal anemia is a causal factor, then efforts at primary prevention of iron deficiency in infants could be aimed in an entirely new direction. Similarly, it is biologically plausible that multivi90

tamins with or without additional iron could be associated with improved hematologic outcomes. For example, the role of vitamin C in improved absorption of dietary iron is well established.31,32 Alternatively, adherence to the use of vitamins may be a proxy for overall better infant care practices or other unidentified maternal characteristics. It is possible that parents may have inflated or overestimated the amount of vitamins actually given to their infants; however, this would not explain the association between adherence and improved hematologic outcomes. If reported use of the vitamins was a proxy for other maternal characteristics, then it seems equally plausible that these characteristics would predispose parents toward truly better adherence and not just a false reporting of adherence estimates. This study has a number of limitations. Measures of iron deficiency and iron deficiency anemia are imprecise; thus, laboratory diagnosis of this common disorder is not simple.1 The diagnosis must rely on a combination of measures. We chose to use a variety of biochemical and hematologic measures either alone or in combination. Also, we could not assess or control for the prevalence of -thalassemia trait in

TRIAL OF MULTIVITAMINS WITH IRON TO PREVENT ANEMIA IN INFANTS Downloaded from www.pediatrics.org. Provided by Indonesia:AAP Sponsored on April 15, 2011

TABLE 4.

Main Hematologic Outcomes by Group Enrollment and Adherence Level Intervention Group (Multivitamins With Iron) Control Group (Multivitamins With No Iron) Total OR (95% CI)

Full adherers only Any abnormal laboratory results Hb 11.0 g/dL Iron deficiency only Hb 11.0 g/dL and 2 abnormal laboratory values Full or partial adherers only Any abnormal laboratory results Hb 11.0 g/dL Iron deficiency only Hb 11.0 g/dL and 2 abnormal laboratory values Nonadherers only Any abnormal laboratory results Hb 11.0 g/dL Iron deficiency only Hb 11.0 g/dL and 2 abnormal laboratory values

51% (28/55) 15% (8/55) 36% (20/55) 9% (5/55)

57% (32/56) 13% (7/56) 45% (25/56) 5% (3/56)

54% (60/111) 14% (15/111) 41% (45/111) 7% (8/111)

0.78 (0.371.64) 1.19 (0.403.55) 0.71 (0.331.52) 1.77 (0.407.78)

63% (70/111) 20% (22/111) 43% (48/111) 13% (14/111)

68% (81/119) 17% (20/119) 51% (61/119) 8% (9/119)

66% (151/230) 18% (42/230) 47% (109/230) 10% (23/230)

0.80 (0.461.38) 1.22 (0.632.39) 0.72 (0.431.22) 1.76 (0.734.26)

87% (20/23) 39% (9/23) 48% (11/23) 13% (3/23)

84% (16/19) 32% (6/19) 53% (10/19) 21% (4/19)

86% (36/42) 36% (15/42) 50% (21/42) 17% (7/42)

1.25 (0.227.05) 1.39 (0.385.01) 0.83 (0.242.79) 0.56 (0.112.90)

TABLE 5.

Main Hematologic Outcomes Stratified by Adherence Status Full Most 71% (42/59) 20% (12/59) 51% (30/59) 14% (8/59) Some 82% (49/60) 25% (15/60) 57% (34/60) 12% (7/60) None 86% (36/42) 36% (15/42) 50% (21/42) 17% (7/42) P Value* .001 .020 .209 .331

Any abnormal laboratory results Hb 11.0 g/dL Iron deficiency only Anemia (Hb 11.0) with any 2 abnormal values * All P by 2 analyses. TABLE 6.

54% (60/111) 14% (15/111) 41% (45/111) 7% (8/111)

Results of Multivariate Logistic Regression: Factors Associated With Hematologic Outcomes Any Abnormal Laboratory Value (OR, 95% CI, P Value) Hb 11 (OR, 95% CI, P Value) 1.24, 0.642.38, .528 1.23, 0.642.24, .543 0.65, 0.490.87, .003 Iron Deficiency Only (OR, 95% CI, P Value) 0.77, 0.451.33, .352 1.55, 0.902.67, .115 0.88, 0.691.12, .308

Intervention group* Low maternal Hb Increasing categorical level of adherence

0.88, 0.481.61, .673 2.15, 1.144.07, .018 0.56, 0.410.76, .001

* Reference is control group. Reference is normal maternal Hb level. Reference level of adherence is none.

our study population. Also, while laboratory diagnosis may be imprecise, the results were unchanged when the data were reanalyzed on the basis of 2 abnormal laboratory findings. Our finding of 49% of infants with evidence of iron deficiency is consistent with other studies. It is possible that some of the children with anemia did not have iron deficiency as has been suggested in recent literature33; however, these children should have been randomly distributed between the 2 groups. We did not assess baseline iron status because this is not standard practice and may have discouraged many parents from enrolling their infants in the study. The study was conducted as a pragmatic randomized clinical trial and followed

methods similar to those of other peer-reviewed research.34 As such, we did not alter clinic practice except by adding the additional laboratory measures. Children with preexisting anemia or iron deficiency should have been equally distributed between control and intervention groups as a result of the randomization. Furthermore, Irigoyen et al35 demonstrated that iron deficiency anemia is uncommon among 6-month-old infants who are enrolled in the WIC program. The additional loss to follow-up among enrolled subjects also reflects the difficulty in both providing clinical care and conducting research with low-income populations. We did not quantify other sources of dietary iron, such as meat consumption. The range of meat consumption by the infants
91

ARTICLES Downloaded from www.pediatrics.org. Provided by Indonesia:AAP Sponsored on April 15, 2011

should have been evenly distributed between study groups, and we found no differences between groups in terms of qualitative reporting of any meat consumption. In addition, analyses indicated that children who declined to participate or had no blood specimens drawn and those who had blood specimens were similar to the entire cohort. We also carefully followed CONSORT requirements for analyzing and reporting randomized clinical trials.36 As demonstrated in this study and others,10 including the CDCs Pediatric Nutrition Surveillance System,9 the high prevalence of iron deficiency and iron deficiency anemia among low-income infants continues. This is of substantial concern, particularly in light of the recent findings that a period of iron deficiency is associated with subsequent elevated blood lead levels4 and that the impact of elevated lead levels 10 g/dl on neurocognitive function may be greater than the impact of higher levels.37 The elevated risk of iron deficiency among low-income children has persisted despite numerous recommendations for primary intervention strategies. To date, no evidence satisfactorily explains this increased risk. It is likely that iron deficiency in lowincome infants results from a combination of factors, including insufficient dietary iron content, food insecurity, the early introduction of whole milk, and delayed introduction of meat into the infant diet. However, if, as suggested by our data and that of others,2428 maternal iron deficiency is a causative factor in infantile iron deficiency, then the persistence of this problem among low-income infants would be expected, as the high prevalence of iron deficiency among pregnant low-income women has not changed in recent years.30 This hypothesis deserves additional exploration in the high-risk populations of the United States. For the primary prevention of iron deficiency in full-term infants, the CDC currently recommends exclusive breastfeeding for the first 4 to 6 months, encouraging the consumption of iron-rich foods and supplemented formula and cereals, avoiding whole cow milk until after 12 months of age and limiting its consumption to 24 oz thereafter, and considering iron supplementation for breastfed infants whose consumption of iron from supplementary foods is judged to be inadequate by 6 months of age.1 Although overseas studies of the use of iron supplementation as a primary prevention strategy have been encouraging, the results of our study suggest that multivitamins with iron may not prevent iron deficiency anemia among the population of urban, low-income infants in the United States. We are hopeful that the use of multivitamins, with or without iron, may yet prove to be a successful primary prevention approach. However, more studies will be necessary to test this strategy. We believe that because of the consequences of iron deficiency and its high prevalence among low-income infants, additional investigation of the prevention of iron deficiency among infants is needed. Until an effective primary prevention strategy can be identified, universal screening for secondary prevention of iron deficiency among high-risk populations of infants
92

and preschool children, as recommended by the CDC,1 is warranted.

ACKNOWLEDGMENTS
This work was funded by a grant from the Gerber Foundation and by National Institute of Child Health and Human Development grant 1K24HD42489-O1A2 (Dr Bauchner). Presented in part at the Ambulatory Pediatrics Association at the Pediatric Academic Societies Annual Meeting; Baltimore, MD; May 7, 2002, and May 5, 2003.

REFERENCES
1. Centers for Disease Control and Prevention. Recommendations to prevent and control iron deficiency in the United States. MMWR Recomm Rep. 1998;47(RR-3):125 2. Pollitt E. Iron deficiency and cognitive function. Annu Rev Nutr. 1993; 13:521537 3. Pollitt E. Iron deficiency and educational deficiency. Nutr Rev. 1997;55: 133140 4. Wright RO, Tsaih SW, Schwartz J, Wright RJ, Hu H. Association between iron deficiency and blood lead level in a longitudinal analysis of children followed in an urban primary care clinic. J Pediatr. 2003;142: 9 14 5. Looker AC, Dallman PR, Carroll MD, et al. Prevalence of iron deficiency in the United States. JAMA. 1997;277:973976 6. Centers for Disease Control and Prevention. Iron deficiencyUnited States, 1999 2000. MMWR Morb Mortal Wkly Rep. 2002;51:897 899 7. Yip R, Binkin NJ, Fleshood L, Trowbridge FL. Declining prevalence of anemia among low-income children in the United States. JAMA. 1987; 258:1619 1623 8. Sherry B, Mei Z, Yip R. Continuation of the decline in prevalence of anemia in low-income infants and children in five states. Pediatrics. 2001;107:677 682 9. Polhamus B, Dalenius K, Thompson D, et al. Pediatric Nutrition Surveillance 2001 Report. Atlanta, GA: US Department of Health and Human Services, Centers for Disease Control and Prevention; 2003 10. Eden AN, Mir MA. Iron deficiency in 1- to 3-year-old children: a pediatric failure? Arch Pediatr Adolesc Med. 1997;151:986 988 11. Walter T, Dallman PR, Pizarro F, et al. Effectiveness of iron-fortified infant cereal in prevention of iron deficiency anemia. Pediatrics. 1993; 91:976 982 12. Pizarro F, Yip R, Dallman PR, Olivares M, Hertrampf E, Walter T. Iron status with different infant feeding regimens: relevance to screening and prevention of iron deficiency. J Pediatr. 1991;118:687 692 13. Engelmann MDM, Sandstrom B, Michaelsen KF. Meat intake and iron status in late infancy: an intervention study. J Pediatr Gastroenterol Nutr. 1998;26:26 33 14. Childs F, Aukett A, Darbyshire P, et al. Dietary education and iron deficiency anaemia in the inner city. Arch Dis Child. 1997;76:144 147 15. Domellof M, Cohen FJ, Dewey KG, Hernell O, Rivera LL, Lonnerdal B. Iron supplementation of breast-fed Honduran and Swedish infants from 4 to 9 months of age. J Pediatr. 2001;138:679 687 16. Dijkhuizen MA, Wieringa FT, West CE, Martuti S, Muhilal. Effects of iron and zinc supplementation in Indonesian infants on micronutrient status and growth. J Nutr. 2001;131:2860 2865 17. Yu SM, Kogan MD, Gergen P. Vitamin-mineral supplement use among preschool children in the United States. Pediatrics. 1997;100(5). Available at: pediatrics.org/cgi/content/full/100/5/e4 18. Eden AN. Iron supplement policy is inadequate. Pediatr News. 2001; March:40 19. American Academy of Pediatrics, Committee on Nutrition. Pediatric Nutrition Handbook. 4th ed. Elk Grove, IL: American Academy of Pediatrics; 1998 20. Geltman PL, Meyers AF, Bauchner H. Daily multivitamins with iron to prevent anemia in infancy: a randomized clinical trial. Clin Pediatr. 2001;40:549 554 21. Dallman PR, Siimes MA, Stekel A. Iron deficiency in infancy and childhood. Am J Clin Nutr. 1980;33:537541 22. Blumberg SJ, Bialostowsky D, Hamilton WL, Briefel RR. The effectiveness of a short form of the Household Food Security Scale. Am J Public Health. 1999;89:12311234 23. Brugnara c, Zurakowski D, DiCanzio J, Boyd T, Platt O. Reticulocyte hemoglobin content to diagnose iron deficiency in children. JAMA. 1999;281:22252230 24. Colomer J, Colomer C, Gutierrez D, et al. Anaemia during pregnancy as a risk factor for infant iron deficiency: report from the Valencia Infant Anaemia Chort (VIAC) study. Paediatr Perinat Epidemiol. 1990;4:196 204

TRIAL OF MULTIVITAMINS WITH IRON TO PREVENT ANEMIA IN INFANTS Downloaded from www.pediatrics.org. Provided by Indonesia:AAP Sponsored on April 15, 2011

25. Preziosi P, Prual A, Galan P, Daouda H, Boureima H, Hercberg S. Effect of iron supplementation on the iron status of pregnant women: consequences for newborns. Am J Clin Nutr. 1997;66:1178 1182 26. Kilbride J, Baker TG, Parapia LA, Khoury SA, Shuqaidef SW, Jerwood D. Anaemia during pregnancy as a risk factor for iron-deficiency anaemia in infancy: a case-control study in Jordan. Int J Epidemiol. 1999;28:461 468 27. Blot I, Diallo D, Tchernia G. Iron deficiency in pregnancy: effects of the newborn. Curr Opin Hematol. 1999;6:6570 28. de Pee, S, Bloem MW, Sari M, Kiess L, Yip R, Kosen S. The high prevalence of low hemoglobin concentration among Indonesian infants aged 35 months is related to maternal anemia. J Nutr. 2002;132: 22152221 29. Choi JW, Kim CS, Lee J, Pai S. Erythropoietic activity and soluble transferrin receptor level in neonates and maternal blood. Acta Paediatr. 2000;89:675 679 30. Centers for Disease Control and Prevention. Pregnancy Nutrition Surveillance 1996 Full Report. Atlanta, GA: US Department of Health and Human Services, Centers for Disease Control and Prevention; 1998: 1516

31. Bothwell TH. Overview and mechanisms or iron regulation. Nutr Rev. 1995;53:237245 32. Siegenberg D, Baynes RD, Bothwell TH, et al. Ascorbic acid prevents the dose-dependent inhibitory effects of polyphenols and phytates on nonheme-iron absorption. Am J Clin Nutr. 1994;53:537541 33. Bogen DL, Krause JP, Serwint JR. Outcome of children identified as anemic by routine screening in an inner-city clinic. Arch Pediatr Adolesc Med. 2001;155:366 371 34. Little P, Gould C, Williamson I, Moore M, Warner G, Dunleavey J. Pragmatic randomized controlled trial of two prescribing strategies of acute otitis media. BMJ. 2001;322:336 342 35. Irigoyen M, Davidson LL, Carriero D, Seaman C. Randomized, placebocontrolled trial of iron supplementation in infants with low hemoglobin levels fed iron-fortified formula. Pediatrics. 1991;88:320 326 36. Moher D, Schulz DF, Altman D, for the CONSORT Group. The CONSORT statement: revised recommendations for improving the quality of reports of parallel-group randomized trials. JAMA. 2001;285:19871991 37. Canfield RL, Henderson CR Jr, Cory-Slechta DA, Cox C, Jusko TA, Lanphear BP. Intellectual impairment in children with blood lead concentrations below 10 g per deciliter. N Engl J Med. 2003;348:15171526

COW PARTS EVERYWHERE

Parts of a slaughtered cow go everywhere. The official British BSE Inquiry Report put it this way: It has been said, and not altogether facetiously, that the only industry in which some parts of the cow are not used is concrete production. The problem isnt just global meat. Its global cow. Heres the scale of the question. In 2002 commercial slaughterhouses in the United States killed 36 780 000 cattle and calves. What isnt meat leaves the slaughterhouse for the rendering plant. There it is converted into basic raw materials that are processed all around the world into a thousand different forms, most broken down all the way to their molecular components, into proteins and fats and fatty acids. Just how widely these are dispersed industrially is nearly endless. Vaccines are often prepared in media that may contain byproducts from slaughtered cattle. Until recently, heparin, a widely prescribed anticoagulant, was made from bovine mucosa and lungs, and steroids come from adrenal glands. Chemicals derived from bovine tissue appear in plastics, paper coatings, rubber, and asphalt. Glycerin appears in countless products. Collagen is a bovine by-product.

Klinkenborg V. New York Times. January 20, 2004

Submitted by Student

ARTICLES Downloaded from www.pediatrics.org. Provided by Indonesia:AAP Sponsored on April 15, 2011

93

Daily Multivitamins With Iron to Prevent Anemia in High-Risk Infants: A Randomized Clinical Trial Paul L. Geltman, Alan F. Meyers, Supriya D. Mehta, Carlo Brugnara, Ivan Villon, Yen A. Wu and Howard Bauchner Pediatrics 2004;114;86-93 DOI: 10.1542/peds.114.1.86
Updated Information & Services References including high-resolution figures, can be found at: http://www.pediatrics.org/cgi/content/full/114/1/86 This article cites 31 articles, 17 of which you can access for free at: http://www.pediatrics.org/cgi/content/full/114/1/86#BIBL This article has been cited by 2 HighWire-hosted articles: http://www.pediatrics.org/cgi/content/full/114/1/86#otherarticles This article, along with others on similar topics, appears in the following collection(s): Therapeutics & Toxicology http://www.pediatrics.org/cgi/collection/therapeutics_and_toxico logy Information about reproducing this article in parts (figures, tables) or in its entirety can be found online at: http://www.pediatrics.org/misc/Permissions.shtml Information about ordering reprints can be found online: http://www.pediatrics.org/misc/reprints.shtml

Citations

Subspecialty Collections

Permissions & Licensing

Reprints

Downloaded from www.pediatrics.org. Provided by Indonesia:AAP Sponsored on April 15, 2011