Association of Infant Feeding Practices with Iron Status and Hematologic Parametersin Thai Infants at 6 Months of Age

Background: An infant's iron intake in the first 6 months of life comes solely from milk 23 intake. However, infants' feeding practices vary, and their association with infants' iron status 24 and hematologic parameters has not been well studied. We aimed to evaluate how different 25 infant feeding practices associate with iron status and hematologic parameters among 6- 26 month-old Thai infants. Methods : In a retrospective chart review, we identified 403 infants who attended a well-baby 28 clinic and received laboratory screening for anemia (complete blood count and serum ferritin) 29 at 6-month visits. Infants were categorized into four groups according to feeding practices. 30 Hematologic parameters and incidence of anemia (hemoglobin [Hb]<11 g/dL), iron 31 deficiency (ID; ferritin<12 ng/mL), and iron deficiency anemia (IDA; Hb<11 g/dL and 32 ferritin<12 ng/mL) were compared between groups. Univariate and multiple logistic 33 regression models were used to identify IDA associated factors among 6-month-old infants. Results : In total, 105 infants were breastfed (BF), 78 were breastfed with iron 35 supplementation (BI), 109 infants were mixed-fed (breast milk and formula) with or without 36 iron supplementation starting at age 4 months (MF), and 111 infants were formula-fed (FF). 37 The BF group had the highest incidence of anemia, ID, and IDA. Anemia was found in 38 38.1% of BF infants compared with 21.8% of BI, 19.3% of MF, and 16.2% of FF infants 39 (p<0.001). ID was found in 28.6% of BF infants compared with 3.8% of BI, 3.7% of MF, and 40 0.9% of FF infants (p<0.001). IDA was found in 17.1% of BF infants compared with 2.6% of 41 BI, 0.9% of MF, and 0.9% of FF infants (p<0.001). In multivariate logistic regression, higher 42 weight gain during 0–6 months slightly increased the risk of IDA and higher birth weight slightly Conclusion: The incidence of anemia, ID, and IDA at age 6 months was higher in BF than 46 FF or partially BF infants. However, iron supplements in BF infants starting at 4 months 47 significantly reduced their ID and IDA incidence. 48

significantly reduced their ID and IDA incidence. scores than children without this condition [5]. In their follow-up study, many children who 79 had severe, chronic ID in infancy continued to experience behavioral and developmental 80 problems 10 years later, although their anemia had been corrected [6].

81
A full-term infant has iron stores at birth that are sufficient until 4-6 months of age. 82 During the first 6 months of life, infant iron status depends more on the iron store at birth 83 than the iron intake from breast milk. The iron content in breast milk is low and decreases 84 over the lactation period. Nevertheless, because breast milk is considered the best nutrition 85 for infants during the first year of life, especially in the first 6 months, the World Health infants under 6 months of age be exclusively breastfed, with no need for complementary 90 feeding or vitamin and mineral supplementation [8,9]. Previous studies on iron status in 91 infants have shown that anemia and IDA occur more frequently in breastfed infants than 92 those who are formula-fed [10,11]. A cross-sectional study of iron status in breastfed infants 93 age 3-5 months showed a higher prevalence of ID and IDA among 5-month-old compared 94 with 3-month-old infants [12]. A longitudinal study following the iron status of healthy 95 breastfed infants at 4 and 6 months of age showed an increase in the prevalence of ID ( ng/mL, which is defined as within the range of insufficiency to deficiency [16,17]. Data in 110 this patient age group also suggest that IDA is associated with low birth weight, low dietary 111 iron intake, low household income, and long duration of breastfeeding [17][18][19]. It is possible commonly performed in many institutions, may be too late; anemia may occur earlier because 114 the neonatal iron store is exhausted before 6 months.

115
Because feeding is the primary source of iron intake for infants, feeding practices 116 considerably impact their iron status. Infant feeding practices during the first 6 months of life 117 vary widely. Some infants are exclusively breastfed whereas others are given infant formula 118 or a combination of both breast milk and formula. Although iron supplementation for 119 breastfed infants, as recommended in the AAP guideline, is not endorsed by Thai health 120 organizations, some clinical practitioners in Thailand adhere to this practice because iron 121 supplementation may have a preventive effect on ID in exclusively or partially breastfed 122 infants. Some infants may also be given complementary foods early, before they have 123 reached 6 months of age.

124
In this study, we investigated the association of infant feeding practices with iron 125 status and hematologic parameters among infants at 6 months of age using historical data of 126 patients examined in a well-baby clinic. Our primary objective was to compare the 127 prevalence of anemia, ID, and IDA at 6 months of age between infants who received different

134
We performed a retrospective cohort study and reviewed the electronic medical  Only the medical records of infants with available information from 4-month and 6-month 151 well-baby visits in the electronic database were included in the study. Premature infants with 152 gestational age at birth less than 35 weeks and infants who received iron supplements before 153 4 months of age were excluded from the study.

154
A self-administered questionnaire is routinely provided to the parents of infants 155 visiting a well-baby clinic at CNMI, to be completed before each physician encounter. These consisted of infants' baseline characteristics, including sex, gestational age, weight, length, 166 and head circumference at birth, and the data from continuous infant monitoring at the well-167 baby clinic at 4 and 6 months of age including weight, length, head circumference, and infant 168 feeding practice. We collected laboratory results, including CBC and serum ferritin levels at 169 6 months of age. Infants were categorized into four groups according to feeding practices: 1) 170 breastfed infants (BF) were those whose parents reported in the 4-month and 6-month 171 questionnaires that the infant was fed breast milk (without infant formula or iron 172 supplements); 2) breastfed with iron supplements (BI) referred to infants whose parents 173 documented in the 4-month and 6-month questionnaires that the infant was fed breast milk 174 and also received iron supplements prescribed by a physician starting at the 4-month visit; 3) 175 mixed-fed with or without iron supplements (MF) comprised infants whose parents reported 176 in either the 4-or 6-month questionnaire that the infant was fed breast milk and infant 177 formula together or was fed breast milk at the 4-month visit and fed infant formula at the 6-178 month visit (infants in this group may or may not have received iron supplementation); and 4) 179 formula-fed infants (FF) were those whose parents documented in the 4-month and 6-month   (Table 1).

230
The mean age of infants at the time of laboratory evaluation was 189±11 days in BF, 190±11  At 6 months of age, 28.6% of infants in the BF group had ID, 38.1% had anemia, and 238 17.1% had IDA, which were all significantly higher than these proportions in the other 239 groups. In contrast, there was no significant difference in the rate of infants who had anemia 240 without ID between the four groups (Table 2).

241
Univariate logistic regression analysis was performed to determine the predictive 242 factors for IDA in 6-month-old infants. Male sex and greater weight gain during 0-6 months 243 were associated with increased risk of IDA whereas older gestational age and higher birth 244 weight were associated with decreased risk of IDA. The BI, MF, and FF groups were 245 associated with decreased risk of IDA, with BF infants as the reference group. In contrast, 246 small for gestational age and complementary feeding initiated before 6 months were not 247 associated with the risk of IDA. In the multivariate logistic regression model, which included 248 factors with p<0.1 in univariate logistic regression analysis, greater weight gain during 0-6 249 months slightly increased the risk of IDA and higher birth weight slightly decreased this risk.

250
Infant feeding patterns strongly predicted IDA in the multivariate logistic regression model.

251
With BF infants as the reference group, the BI, MF, and FF groups had 90.4%, 97.5%, and 252 96.9% decreased risk of IDA, respectively (Table 3). Our study found that 28.6% and 17.1% of Thai infants who were breastfed but did not 256 receive iron supplementation had ID and IDA, respectively, at 6 months of age. These 257 incidences were significantly higher than those among formula-fed or partially breastfed 258 infants. We found that breastfed infants who received 1-2 mg/kg/day of iron supplementation 259 starting at 4 months of age had iron status and hematologic parameters comparable to those When infants in all groups were combined, the overall prevalence of anemia in 6-265 month-old infants was 23.9%. which is comparable to the prevalence previously reported in 266 multiple studies among infants and children in similar age groups in Thailand. A study by 267 Suwannakeeree et al. found a 29.1% prevalence of anemia in 9-month-old infants [19], and 268 Tantracheewathorn et al. found 26.4% prevalence in 9-12 month-old infants [18]. infants in our study was 5.5%, which was lower than previously reported prevalence among 272 9-12 month-old infants in Thailand (14.3% by Tantracheewathorn et al. and 17.9% by 273 Suwannakeeree et al.) [18,19]. This difference in IDA prevalence was likely owing to 274 differences in the ages of the study populations. The iron store at birth is the source for iron 275 utilization in infants until approximately 6 months of age. The risk of IDA among infants 276 over 6 months old is increased unless appropriate complementary feeding is given.

277
Additionally, the feeding patterns differed between studies. Our study included a large 278 number of formula-fed infants, many of who were breastfed or partially breastfed and 279 receiving iron supplements beginning at 4 months old, which may help to prevent IDA.

280
In our study, 15%-20% of infants in each group were anemic despite having normal 281 serum ferritin levels at 6 months. Anemia among these infants was unlikely to have been infants had ID at 6 months of age, and nearly one in every five breastfed infants had IDA.

299
Given the potential negative effects of even pre-anemic ID on early development, waiting 300 until infants reach 6 months of age before starting supplementation may be too late. A meta-301 analysis of iron supplementation among breastfed infants at an early age showed limited 302 evidence regarding the effect of iron supplements in the prevention of IDA [26]. However, 303 the impact of supplements depends on the initial infant iron status, as shown in a randomized 304 control trial that compared the prevalence of IDA among exclusively breastfed infants with 305 iron supplementation during ages 4-9 and 6-9 months. That study was conducted in two areas 306 (Honduras and Sweden) with relatively different amounts of dietary iron consumption. Iron 307 supplementation in breastfed infants from 4 months effectively reduced the prevalence of ID 308 at 6 months of age only in Honduras, where the baseline prevalence of ID is high and dietary 309 iron intake is low [27]. Neonatal iron storage has a more significant effect on serum ferritin 310 levels in early infancy than iron consumption from breast milk [28]. Our study findings imply stores at birth, which are affected by maternal nutrition and dietary intake during pregnancy.

313
Iron supplements starting at 4 months of age in breastfed infants, as recommended by the 314 AAP, can offer a reasonable approach to reduce the risk of ID and IDA among Thai infants.

315
Our results showed a significantly lower rate of ID and IDA among breastfed infants 316 with iron supplementation, as well as those who were partially breast-fed and formula-fed, 317 compared with infants who were breastfed and did not receive iron supplements. Our 318 multivariate logistic regression analysis confirmed these findings and showed that iron 319 supplementation among breastfed infants might reduce the risk of IDA by 90.4%. In addition 320 to iron supplements, evidence from a systematic review suggests that the introduction of 321 solids at 4 months may have a beneficial effect on the rate of IDA in breastfed infants [29]. In 322 our study, complementary foods were introduced to infants in each group between age 4 and 323 6 months at a different rate. However, complementary feeding did not seem to affect iron 324 status, which showed no association with IDA in univariate analysis (Table 3). The absence 325 of an association between complementary feeding and IDA may be owing to feeding 326 practices in Thailand, where typical foods given to infants at this age are usually plant-based 327 foods with low iron content [30]. As a retrospective study, we did not have information 328 regarding the types of foods given to infants.

329
Universal iron supplementation of breastfed infants at an early age is controversial.

330
Iron homeostasis in infants younger than age 6 months is limited compared with older infants 331 [31]. Iron absorption is not effectively downregulated in infants at an early age, who have regarding infants' growth with daily iron supplementation, we found no significant differences in growth parameters among infants with different feeding patterns. However, our 338 study included a relatively small sample size, and follow-up assessment was performed over 339 a short period. Thus, we cannot support nor refute the above concerns. Adequately powered 340 trials are needed to investigate the non-hematological benefits and risks of iron 341 supplementation in this patient population. As an alternative to iron supplementation, 342 measures to increase iron stores at birth among infants so as to provide them with adequate 343 iron until late infancy, such as delayed cord clamping, have been reported [33].

344
This study has some limitations. Our study was performed at a single center with a 345 limited sample size, so it was not sufficiently powered to be able to adequately assess the Anemia is not uncommon among Thai infants at 6 months of age. In our study, 364 significantly higher incidence of ID and IDA was found in Thai breastfed than formula-fed or 365 partially breastfed infants. However, iron supplementation of breastfed infants with 1-2 366 mg/kg/day starting at 4 months of age significantly decreased the incidence of ID and IDA. We thank all the parents for bringing their infants to our hospital and allowing us to provide 380 the best service possible for them. We especially thank the interns, pediatric hospitalists, and 381 attending staff at CNMI for their hard work in performing patient follow-up in the well-baby 382 clinic at CNMI. We would like to thank Assoc. Prof. Sasivimol Rattanasiri and Mrs.

383
Umaporn Udomsubpayakul for their assistance in the statistical analysis. We are grateful to