Early Nutrition and Its Effect on Growth, Body Composition, and Later Obesity

Introduction

A balanced and nutritionally adequate diet in early life is essential for optimal growth and healthy life both in the short and long term. Many studies are published within this research area. In this review, early nutrition covers infant feeding (mainly breastfeeding),complementary feeding, and nutrition in early childhood with special focus on intake of cow's milk.

We performed a nonsystematic literature search in PubMed using the terms ``breastmilk[or] human milk [or] complementary feeding [and] growth [or] body composition.” The search resulted in 648 papers of which we selected 12 recent publications published between July 1, 2021 and June 30, 2022. We find these publications of special interest based on their contribution to the research within this field, novelty, and quality. We have divided the papers into 5 topics: human milk appetite-regulating hormones and infant growth(2 papers); early feeding and body composition (3 papers); complementary feeding, growth and adiposity (4 papers); cow’s milk, fat and adiposity (2 papers); historical overview of 25 years of research on human milk and lactation (1 paper).

Human Milk Appetite-Regulating Hormones and Infant Growth

Appetite-regulating hormones in human milk: a plausible biological factor for Obesity risk reduction?

Comments:  In recent years, hormones in human milk (HM) have been suggested to affect appetite regulation in infants through similar mechanisms as the endogenously produced hormones [1], and thereby affect growth. Leptin is the most investigated appetite-regulating hormone (ARH) and is secreted from the adipose tissue [2, 3]. Endogenously Produced leptin affects appetite through receptors in the hypothalamus of the brain by reducing energy intake (EI) and increasing energy expenditure (EE) [4]. Animal Studies have found that orally ingested HM leptin can enter the circulation [5], which makes the hypothesis biologically plausible. However, evidence on the influence of HM ARH on infant growth is sparse and the high risk of confounding in observational studies limits the ability to infer causality and draw conclusions. We found 2 studies addressing these issues.

Castell et al. reviewed the current literature within the area of HM leptin and infant growth outcomes with the posed hypothesis that HM leptin affects appetite regulation and thereby growth. They studied 18 papers, including both original research and other reviews, published from January 1, 2015 to December 31, 2019. The studies were mainly observational cohort studies with sample sizes ranging from n = 20 up to n = 350. Two of the original research papers presented relationships between HMleptin concentrations and infant weight [6, 7] and/or body composition, although results were slightly conflicting. Brunner et al. measured HM leptin at 6 weeks and 4 months and infant anthropometric measurements were collected at birth and at 6 weeks, 4 months, and at 1 and 2 years. They found inverse associations between HMleptin and infant weight and lean body mass at 4 months, but not at later time points. This might suggest that appetite regulation is more likely to affect fat-free mass (FFM)accretion rather than fat deposition during breastfeeding. Leptin in HM declined across lactation; thus increased concentrations in early lactation might be driving the association. The other study by Nuss et al., found that HM leptin inversely correlated with infant growth measures such as weight and fat mass (FM) percentage at 4 to 8 weeks postpartum, but only in infants of mothers with normal weight compared to overweight [7]. This suggests an effect on deposition of fat compared to lean mass which is contrary to the findings by Brunner et al.

The review also presents studies investigating the association between infant plasma leptin and growth depending on feeding practice. One of the studies found higher plasma leptin in breastfed infants compared to formula-fed infants [8], which could reflect either orally ingested HM leptin or increased endogenous production of leptin. Conversely, Breij et al. found higher serum leptin in formula-fed infants compared to breastfed infants [9] and serum leptin correlated furthermore positively with infant FM percentage. One of the main limitations is the lack of evidence regarding absorption of the hormones in the infant gut. As such, we cannot determine that HM leptin is the main contributor to circulating levels of the hormones and/or for associations seen with infant growth.

The results from this review overall emphasize the conflicting evidence within this area. The studies diverge in several important aspects including method of milk sample collection, sample size, and study design, which increases the risk of bias and complicates comparison. Other factors could be of importance when investigating the associations between HM hormones and infant outcomes such as infant milk intake(MI) or bacterial colonization [10].

The study by Larson-Meyer et al. investigates the hormones leptin, peptide YY (PYY),glucagon-like peptide 1 (GLP-1), and ghrelin in HM. They further pose 3 aims of their study, namely, to investigate (1) the hormone concentrations across lactation from 1 to 6 months, (2) maternal predictors of the hormones in HM, and (3) the associations between HM hormones and infant growth. They use a combination of fore- and hindmilk and have a sample size of n = 22 mothers at 1 month and n = 15 at 6 months. They Found declining concentrations of milk fat as well as leptin, ghrelin, and PYY from 1 to6 months of lactation. Furthermore, milk fat and leptin were positively associated with maternal body mass index (BMI), which is also supported in the literature [11]. Lastly,the authors found that milk fat in foremilk at 1 month was positively associated with weight-for-age z-score, whereas GLP-1 and leptin at 1 month were negatively associated with weight-for-age z-score at 6 and 12 months, respectively. The authors mention that the associations seen between milk fat and maternal adiposity might be driven by a few mothers with obesity whose milk could contain higher fat concentrations. As leptin is partly secreted by the adipose tissue, HM leptin might correlate with milk fat and could mask associations between HM fat and infant outcomes. These results illustrate the complexity of this research area, as we cannot elucidate whether genetic disposition for obesity is the true predictor for infant weight. Furthermore, the authors chose a foremilk and a hindmilk sample, which may not represent a complete feed, and concentrations of milk fat and/or leptin might be either over- or underestimated compared to the infant’s actual intake.

In conclusion, the area of HM hormones is still controversial. This, however, does not make the field of research less important, only challenging for the researchers. Statistical analyses and the resulting conclusions have to be made carefully and with respect to relevant limitations.

Early Feeding and Body Composition

The “drive to eat” hypothesis: energy expenditure and fat-free mass but not adiposity are associated with milk intake and energy intake in 12-week infants

Comments: In this reanalysis of previously collected data, Wells et al. explored the drive-to-eat hypothesis among 48 infants at 12 weeks of age. The hypothesis posits that rather than FM and associated adipokines working to maintain energy balance through appetite alterations, appetite instead adapts as a function of EE and accordingly FFM.Participants were predominantly breastfed (n = 24) or formula-fed (n = 24), healthy,full-term infants who had participated in a British study to investigate energy metabolism. Intake of supplementary foods was minimal though was permitted after 11 weeks. MIs were estimated by test-weighing of the infant or bottle over two 24-h periods and EI was estimated.

Body composition and EE were measured with deuterium dilution and dilution spaces calculated via back-extrapolation. The sleeping metabolic rate (SMR) was measured as a proxy for, and in place of, the basal metabolic rate using a Deltatrac Metabolic monitor measuring respiratory gas exchange.

Pearson’s correlations showed that MI and EI were positively correlated with FFM but not FM, while also being positively correlated with SMR and EE. As a tissue with greater metabolic activity, FFM but not FM was correlated with SMR and EE. Spearman’s correlations showed that MI and EE were positively correlated with weight gain over the 1-week data collection period, but EI was not. In multiple regression models adjusted for mid-parental height, MI was associated with SMR independent of FFM, but FFM was not significantly associated with MI in the same model. In a further model EE and FFM were both independently associated with MI. When investigating EI instead of MI, FFM was independently associated in models including either SMR or EE which were each independently associated with EI.

These results support that MI and EI appear to match FFM and EE in infants, but not FM. This finding is new in infants and suggests that adipose-derived hormones, afront-running theory in infant satiety and appetite regulation, may not necessarily function as in adults. This study provides a glimpse at the “other side of the coin” of infant growth and reminds us that reverse causality should always be considered, i.e.,perhaps the infants with greater MI grow faster, or perhaps the faster growth stimulates greater MI. It is also worth highlighting the significance of mid-parental height when included in the models to account for a heritability of growth-drive. That this appears to influence infant MI, EI, and weight-gain suggests a stimulatory effect on appetite that can be incorporated into future studies investigating appetite regulation.

This investigation had strong methodology throughout, with precise measures of body composition and EE using a reference technique. The assessment of SMR in infancy is also something seldom seen in the field of infant feeding so brings a novel aspect to the discussion of infant appetite drive.

Early infant feeding effect on growth and body composition during the first 6 years and neurodevelopment at age 72 months

Comments: This study assessed differences in growth and body composition longitudinally over the first 6 years between infants fed with breast milk (BF), soy-based formula (SF), or cow’s milk-based formula (MF) from 3 to 12 months.
A large sample of 600 healthy, term-born infants were recruited between ages 1 and2 months. Parents had chosen which diet their infant was to follow before enrolling, and were excluded if this choice changed between 2 and 12 months or if they introduced complementary foods before 4 months. The formula groups were provided with appropriate formula to strengthen adherence and standardize formulations. Within the breastfeeding group, half breastfed until 12 months, a quarter weaned to formula between 9 and 12 months, and the remaining quarter weaned to formula before 9 months. Growth assessment took place at 3, 6, 9, 12, 24, 36, 48, 60, and 72 months, which included body composition measurements via dual-energy X-ray absorptiometry (DXA). Complementary food was recorded by completion of 3-day food records.
The final analysis contained 178 BF, 179 MF, and 169 SF infants. There were some feeding-group differences at baseline, which we would expect as the groups were not randomized; gestational age, birthweight, maternal IQ, and parental education were all higher in the BF group.

At each time point from 24 to 72 months, infant BMI was lower in the BF group compared to the SF group. At 9 and 12 months, BF infants were around 0.5 kg lighter than both MF and SF infants. At measurements from 36 to 72 months, BF infants were 3–4 kg lighter than SF infants. At 6, 9, and 12 months, BF infants were shorter than both groups of formula-fed infants. Fat mass index (FMI) was higher in BF infants compared to SF infants at 3 and 6 months, which reversed at 36 and 48 months, when FMI was higher in the SF children. At 60 and 72 months, BF children had lower BMI compared to both SF and MF children. BF infants tended to have lower fat-free mass index (FFMI)than formula-fed infants up to 6 months, after which we expect complementary feeding (CF) to begin contributing to growth variation. Interestingly, FFMI was higher in SF children at 36 and 72 months by 0.3–0.4 kg/m2 compared to MF, a result which could arise from the differences in length. Possibly contributing to some of these differences was the observed lower EI in BF infants in the first year versus formula-fed infants. This could be seen as lower EI in the BF group contributing to reduced FFMI or, following from the paper by Wells et al. commented on above, as a greater drive-to-eat due to the higher FFMI in formula groups.

The longitudinal approach and reference body composition technique in DXA represent great strengths of this study. Furthermore, the impressive sample size is larger than many studies we see in this area. The lower BMI observed in BF infants is a familiar result and supports previous research indicating a protective effect of breastfeeding against obesity. This characteristic of higher FMI and lower FFMI of breastfed versus formula-fed infants in infancy supports the hypothesis that higher-protein formula feeding can indeed influence body composition at this age and can switch direction of association into later childhood. This study also makes the important observation that SF feeding does not result in unfavorable changes to infant body composition when compared to MF, although breastfeeding remains optimal.

Infant feeding practices associated with adiposity peak and rebound in the EDEN mother-child cohort

Comments: Later age and higher magnitude of infant BMI peak, also called adiposity peak (AP),and younger age when BMI starts to increase again, called adiposity rebound (AR),have been associated with the risk of obesity in many studies. This study is based on the French Eden cohort and the study included 1,225 children. Data on infant feeding practice were collected at birth and 4, 8, and 12 months and growth pattern was modeled on growth data from the child’s health booklet up to 12 months. Mean age forAP was 9.9 months and for AR 5.5 years. Interestingly, this study found that child sex had a moderating effect on the association between infant feeding and AP and AR.For boys, longer breastfeeding duration was related to a reduced BMI at AP, which is associated with a lower risk of adiposity later in childhood. Later age at AR was associated with duration of breastfeeding in girls, but in boys it was associated with delayed introduction of complementary foods. This study highlights how infant sex plays an important role in the association between early nutrition, growth, and later risk of obesity and thereby underlines that infant sex should be included in future analysis of how early nutrition is affecting growth.

Complementary Feeding, Growth and Adiposity

Complementary feeding methods – a review of the benefits and risks

Comments: While there are several aspects and types of CF methods, this paper focuses only on baby-led weaning (BLW). Twenty-nine studies of BLW were reviewed and the authors conclude that there is suggestive evidence that BLW can reduce food fussiness and improve satiety responsiveness, but the results are far from conclusive. The potential negative effects of BLW are choking and underweight, but there is no evidence to support such effects. BLW can in theory reduce the risk of overweight, but only a few of the included studies have examined this. These studies lean toward BLW reducing the risk of being overweight. The author concludes that there is a need for more high-quality studies examining the effect of BLW on growth.

Complementary feeding caregivers’ practices and growth, risk of overweight/obesity, and other non-communicable diseases: a systematic review and meta analysis

Comments: During the year covered by this review, a large group of Italian researchers have published 3 papers on CF in the journal Nutrients: 2 systematic reviews and meta-analyses and 1 concept paper. The 3 papers focused on different aspects of the effects of CF on growth, risk of overweight, and other noncommunicable diseases (NCDs). 

The focus of the systematic review and meta-analysis by Verga et al. is how the timing of CF affects growth and risk of NCDs. They compared the start of CF between 4 and 6 months with start at 6 months and the outcomes were growth at 12 months, overweight/obesity at 3–6 years, and iron status. They could only identify 7 studies addressing these issues, and none of them found significant differences on these outcomes. They therefore conclude that the review supports recommendations from theWHO and the EFSA (European Food Safety Authority), namely, that there is no advantage of introducing CF before the age of 6 months.

The systematic review and meta-analysis by Bergamini et al. investigate how caregiver CF practices affect infant growth, overweight/obesity, risk of choking, dental caries, and risk of NCDs. They included several feeding practices: studies on BLW andresponsive CF where the active behavior of the child is prioritized; studies on a modified special version of BLW called Baby-Led Introduction to SolidS (BLISS), where infants at each meal are offered 3 different foods, rich in iron, energy, or fiber; studies examining the effects of nonresponsive CF, where the caregivers are overly active(forcing) or overly passive in relation to feeding of the infant. The authors’ overall assessment of the evidence from the few randomized controlled trials included was that it was low. They concluded that it was not possible to state that either BLW or BLISShad a preventive effect on later overweight. However, they concluded that responsive feeding can result in lower incidence of overweight/obesity and that non responsive feeding can lead to either excess weight or lower weight.

The concept paper by Caroli et al. contains recommendations on CF from a working group with members from several Italian scientific societies with expertise in pediatric nutrition. The paper presents 38 specific recommendations on CF. For each of these recommendations, it is stated if it is an expert opinion, or if the evidence is weak or strong. The panel consensus is also mentioned, and it differs between 100% for many of the recommendations, down to 75% consensus. Among the 38 recommendations are the following: protein intake should not exceed 14% of total EI for children between 6 and 24 months old; CF should not be introduced before 6 months in breastfed and formula-fed infants, if the infants are growing well; unmodified cow’s milk should not be given before 12 months and from 12 to 24 months it is suggested touse formula, as an alternative to cow’s milk to limit protein intake. Furthermore, the amount of cow’s milk, if given, should be less than 500 mL (panel consensus only 75%)and it is suggested not to use BLW. The concept paper also mentions important research areas where more evidence is needed, e.g., age and time window when a specific nutrient may act as trigger for a programming effect, and the impact of CF feeding styles like BLW and responsive and nonresponsive feeding.

Last year, Nutrients published a paper authored by some of the same Italian authors discussing if breastfed and formula-fed infants need different CF [12]. One of the main focuses of the paper is protein intake and type of milk offered to the child. The authors suggest that the advice for CF should differ between breastfed and formula-fed infants, mainly to prevent a too high protein intake in formula-fed infants. Thereby they go against the advice from ESPGHAN, i.e., that the recommendations should be the same independently of feeding practice to avoid confusion for the parents [13]. They also suggest that during the second year (12–24 months), what they call young child formula can be used to meet the age-related nutrient requirements. In a recent paper by Lutter et al. [14], it is stated that what they call follow-up formula and growing-up milks is deemed unnecessary and not recommended by the WHO and many pediatric societies.

Cow's Milk, Fat, and Adiposity

Association of cow’s milk intake in early childhood with adiposity and cardiometabolic risk in early adolescence

Comments: Overweight and obesity in childhood have been of major concern as they often track into adulthood and are difficult to reverse. Therefore, it is essential to have valid and updated strategies and recommendations to prevent overweight and obesity already in childhood. Cow’s milk is a main food offered in early childhood and widely consumed in western countries. According to the “early protein hypothesis,” high amounts of early protein would stimulate growth, especially fat tissue, which could lead to later increased risk of obesity [15]. Therefore, whole cow’s milk is generally first recommended to be introduced at the age of 12 months [16, 17]. However, at the age of 2 years it is recommended to switch to lower-fat cow’s milk to reduce the EI and consequently minimize the risk of excess weight gain [18, 19].

Until lately, dairy fat was often considered to have negative effects on young children. However, more recent studies have found that milk and dairy product consumption was neutral or inversely associated with adiposity in children and adolescents [19].Previous studies investigating the influence of fat content of cow’s milk consumed in early childhood and later risk of overweight and obesity have shown conflicting results and often have important limitations such as lacking adjustment for potential confounders [19–21]. We have selected 2 publications examining the association between intake of cow’s milk fat in early childhood and risk of later adiposity. They give an important contribution toward understanding the relationship between the fat content in cow’s milk consumed in early childhood and later risk of overweight and obesity. They both included subjects from prospective cohort studies and adjusted for salient potential confounders.

In a recent study by Vanderhout et al., the associations between intake of cow’s milk fat and child adiposity measured as BMI z-score (z-BMI) among healthy children were examined. They included 7,467 children in the age range from 9 months to 8 years from a Canadian cohort who reported intake of cow’s milk. Of these, 4,699 had repeated measurements. The intake of cow’s milk fat given as skim (0.1%), 1%, 2%, andwhole (3.25%) milk was collected by dietary questionnaires completed by the parents, and milk consumption was also categorized as either whole milk or reduced-fat milk (0.1–2%). The outcome was z-BMI, and overweight and obesity were defined according to the WHO criteria for children older than 5 years, as z-BMI scores >+1 and>+2 SD, respectively, and applied for all children across ages for consistency. Relevant Covariates included inter alia volume of cow’s milk and sugary drinks, duration of breastfeeding, parental characteristics, and birth weight. Mean age at baseline was2.6 ± 1.5 years and mean intake of cow’s milk was 475 ± 300 mL/day. At baseline, most children (56%) consumed whole milk whereas 34, 8, and 3% of the children consumed 2, 1, and 0.1% milk, respectively. Mean time for follow-up was 2.7 ± 1.7 years. Information regarding breastfeeding, CF, and parental characteristics was obtained from the parents by standardized questionnaires. They found that an increase of 1% in fat content of the milk consumed corresponded to 0.05 lower z-BMI also when adjusting for potential confounders. This result was supported by comparing children consuming whole cow’s milk to children consuming reduced-fat milk. The children who consumed whole cow’s milk had 16% lower odds of being overweight and 18% lower odds of obesity at follow-up compared to children consuming reduced-fat milk.

These findings, which challenge the recommendation of consumption of reduced-fat milk from age 2 years, are supported by the study by McGovern et al. They investigated the associations between consumption of cow’s milk (fat content and frequency) in early childhood and adiposity and cardiometabolic risk in adolescence. In this study, 796 children from a prospective cohort established in Boston consuming cow's milk were included. The intake of cow’s milk fat given as whole milk, 2%, 1%, and skim milk was collected at baseline. In addition, the frequency of MI was assessed. Outcomes were measures of body composition in early adolescence and included i.a. zBMI and overweight or obesity defined as z-BMI ≥ 85th percentile using CDC growth references. Furthermore, body FM was assessed by bioelectrical impedance (BIA)analysis and lean mass, total fat, and trunk FM by DXA. Covariates included, among others, parental characteristics, sex, birthweight, breastfeeding duration, EI and sugary drinks, and z-BMI in early childhood. The mean age at baseline and follow-up was 3.2 and 13.2 years, respectively. Most children consumed whole or 2% milk (30.8 and 32.4%, respectively) and 26.5 and 10.3% of the children consumed 1% or skim milk,respectively. MI in early childhood was estimated as frequency (mean 2.3 ± 1.2 times/day). They compared the intake of milk with higher fat content (whole milk and 2%)versus intake of milk with reduced fat (1% and skim milk) for the outcomes measuring early adiposity in different models with adjustment for increasing number of covariates. They found that intake of higher-fat milk compared to lower-fat milk was associated with lower adiposity for all body composition techniques in models adjusted for most covariates. However, when adjusting for z-BMI at baseline and change in z-BMIfrom 2 to 3 years, only overweight or obesity remained significant, corresponding to40% lower odds of overweight or obesity in early adolescence for children who consumed high-fat milk compared to children consuming low-fat milk in early childhood. Frequency of MI in early childhood was not associated with adiposity and neither frequency nor fat content of intake of cow’s milk was associated with cardiometabolic risk in adolescence.

The 2 studies both showed that intake of low-fat milk in early childhood was not associated with reduced risk of overweight or obesity later in life, but suggested that an inverse relationship might exist. The follow-up periods in the studies were different. The study by Vanderhout et al. had a relatively short mean follow-up time, which is insufficient to evaluate long-term effects. However, they applied a longitudinal design as many of the children had repeated measures. BMI was used as an outcome in both studies but is not a direct measure of body composition. To manage this, McGovern et al. also used DXA and BIA to assess body composition showing the same trends and directions as for z-BMI. Dietary registrations had limitations in both studies, where MI was not assessed in the study by McGovern et al., and total EI was missing in the study by Vanderhout et al. A strength of both studies was the inclusion of important covariates in the models reducing the risk of confounding which is very important, as families following the dietary guidelines generally may be more prone to follow other health advice. In addition, the risk of reverse causality is relevant to consider, i.e., leaner children may be offered higher-fat milk by the parents and vice versa. Both studies agree that the findings are not sufficient to alter the recommendation for fat content in cow’s milk consumed in early childhood. Future studies should include randomized trials to establish any causal relationship between cow’s milk fat consumed in early infancy and later adiposity.

Historical Overview

25 years of research in human lactation: from discovery to translation

Comments: Geddes et al. have published an impressive review, which builds on the research output from the group investigating research on HM and lactation at the University ofWestern Australia. Twenty-five years ago, Peter Hartmann established this research unit, now called the Geddes Hartmann Human Lactation Research Group. He is the last author of the review. He died in 2021, 80 years old. The review conceptualized a biological framework on how maternal and infant factors influence HM composition, and how it is related to infant growth, development,and health. It is a very comprehensive review with 46 pages and 337 references. It covers a broad range of topics from breast anatomy, milk secretion, physiology of milk removal, and milk composition to infant intake, growth, body composition, andhealth. Among the many details included in the review are 5 interesting figures showing the possible pathways of lactocrine programming of the infant. The figures show how maternal body composition is influencing milk components and how these components influence appetite control and body composition. The 5 groups of milk components are proteins, immune factors, appetite hormones, glucocorticoids, and carbohydrates. This review underlines the importance of a broad range of findings emerging from this research group through the past 25 years.

References

1. Fields DA, Schneider CR, Pavela G. A narrative reviewof the associations between six bioactive components inbreast milk and infant adiposity. Obesity (Silver Spring).2016;24:1213–21
2. Kratzsch J, Bae YJ, Kiess W. Adipokines in humanbreast milk. Best Pract Res Clin Endocrinol Metab.2018;32:27–38.
3. Obradovic M, Sudar-Milovanovic E, Soskic S, Essack M,Arya S, Stewart AJ, et al. Leptin and obesity: role andclinical implication. Front Endocrinol (Lausanne).2021;12:585887.
4. Pandit R, Beerens S, Adan RAH. Role of leptin in energyexpenditure: the hypothalamic perspective. Am J Physiol Regul Integr Comp Physiol. 2017;312:R938–47
5. Casabiell X, Piñeiro V, Tomé MA, Peinó R, Diéguez C,Casanueva FF. Presence of leptin in colostrum and/orbreast milk from lactating mothers: a potential role inthe regulation of neonatal food intake. J Clin EndocrinolMetab. 1997;82:4270–73.
6. Brunner S, Schmid D, Zang K, Much D, Knoeferl B,Kratzsch J, et al. Breast milk leptin and adiponectin inrelation to infant body composition up to 2 years. Pediatr Obes. 2015;10:67–73
7. Nuss H, Altazan A, Zabaleta J, Sothern M, Redman L.Maternal pre-pregnancy weight status modifies the influence of PUFAs and inflammatory biomarkers inbreastmilk on infant growth. PLoS One.2019;14:e0217085.
8. Socha P, Hellmuth C, Gruszfeld D, Demmelmair H, Rzehak P, Grote V, et al. Endocrine and metabolic biomarkers predicting early childhood obesity risk. Nestle NutrInst Workshop Ser. 2016;85:81–8.
9. Breij LM, Mulder MT, van Vark-van der Zee LC, Hokken-Koelega ACS. Appetite-regulating hormones in early life and relationships with type of feeding and bodycomposition in healthy term infants. Eur J Nutr.2017;56:1725–32.
10. Sanchez M, Panahi S, Tremblay A. Childhood obesity: arole for gut microbiota? Int J Environ Res Public Health.2014;12:162–75.
11. Savino F, Sardo A, Rossi L, Benetti S, Savino A, Silvestro L. Mother and infant body mass index, breast milkleptin and their serum leptin values. Nutrients.2016;8:E383.
12. Caroli M, Vania A, Tomaselli MA, Scotese I, Tezza G,Verga MC, et al. Breastfed and formula-fed infants: needof a different complementary feeding model? Nutrients.2021;13:3756
13. Fewtrell M, Bronsky J, Campoy C, Domellöf M, Embleton N, Fidler Mis N, et al. Complementary feeding: aposition paper by the European Society for PaediatricGastroenterology, Hepatology, and Nutrition (ESPGHAN) Committee on Nutrition. J Pediatr Gastroenterol Nutr. 2017;64:119–32.
14. Lutter CK, Grummer-Strawn L, Rogers L. Complementary feeding of infants and young children 6 to 23months of age. Nutr Rev. 2021;79:825–46.
15. Koletzko B, Beyer J, Brands B, Demmelmair H, Grote V,Haile G, et al. Early influences of nutrition on postnatalgrowth. Nestle Nutr Inst Workshop Ser. 2013;71:11–27
16. Muth ND. Recommended drinks for children age 5 &younger. Available from: https://www.healthychildren.org/English/healthy-living/nutrition/Pages/Recommended-Drinks-for-Young-Children-Ages-0–5.aspx(accessed August 22, 2022).
17. ESPGHAN Committee on Nutrition, Agostoni C, Braegger C, Decsi T, Kolacek S, Koletzko B, et al. Breast-feeding: a commentary by the ESPGHAN Committee onNutrition. J Pediatr Gastroenterol Nutr. 2009;49:112–25.
18. Gidding SS, Dennison BA, Birch LL, Daniels SR, Gillman MW, Gilman MW, et al. Dietary recommendationsfor children and adolescents: a guide for practitioners.Pediatrics. 2006;117:544–59.
19. Vanderhout SM, Aglipay M, Torabi N, Jüni P, da CostaBR, Birken CS, et al. Whole milk compared with reduced-fat milk and childhood overweight: a systematicreview and meta-analysis. Am J Clin Nutr.2020;111:266–79.
20. Louie JCY, Flood VM, Hector DJ, Rangan AM, Gill TP.Dairy consumption and overweight and obesity: a systematic review of prospective cohort studies. Obes Rev.2011;12:e582–92.21 O’Sullivan TA, Schmidt KA, Kratz M. Whole-fat or reduced-fat dairy product intake, adiposity, and cardiometabolic health in children: a systematic review. AdvNutr. 2020;11:928–50.