Background
Breastfeeding is considered by the World Health Organization (WHO) as the best choice for infant feeding. Human milk composition is used as a guideline for establishing minimum and maximum levels of nutrients in formulas in order to meet infant’s needs. Lipids are major components in human milk, providing 45–55% of total energy intake. They fulfill various metabolic and physiological functions critical for the development, growth and health of the newborn [
1]. Human milk contains a wide variety of lipid species, present as milk fat globules with a core containing triacylglycerols (more than 98% of total lipids), surrounded by a milk fat globule membrane with complex lipids (phospholipids and sphingolipids), esterified cholesterol, proteins and glycoproteins. Human milk provides alpha-linolenic (ALA) and linoleic acids (LA) which are essential fatty acids (EFA). They can be endogenously converted by the newborn into long-chain derivatives of Omega 3 and Omega 6 families, such as docosahexaenoic acid (DHA) and arachidonic acid (ARA), respectively [
2]. However, this conversion is considered as low in humans. Human milk also contains preformed DHA and ARA, at levels strongly influenced by the mother’s diet. These fatty acids (FA) constitute the main components of the brain and they have an important impact on neuronal and visual functions [
3].
Fat sources used in most infant formulas currently marketed are a mixture of plant oils. Indeed, since the middle of twentieth century, infant formulas have been enriched in EFA-rich plant oils and bovine milk fat has been progressively removed. Infant formulas can also be optionally supplemented with fish/algae/single cell oils providing preformed DHA and ARA. However, lipids of plant oils are not comparable to lipids of human milk in terms of FA diversity, triacylglycerol structure, fat globule composition, complex lipids and cholesterol contents. Consequently, infant formulas provide milk-specific FA and cholesterol only when dairy lipids are used as a fat source in combination with plant oils [
1]. Also, the nature of fat sources used in formulas strongly influences their FA composition. In particular, short and medium chain FA, lauric acid, myristic acid and palmitic acid contents can vary considerably according to the use of palm oil, coco oil or dairy lipids [
1]. Moreover, plant oils do not possess the specific triglyceride structure found in breast milk or cow’s milk with palmitic acid in sn2 position. The triglyceride structure is of particular importance because long-chain saturated FA in the sn2 position are more efficiently digested and absorbed [
4].
Dairy lipids specifically contain about 10% of myristic acid and 10% of short and medium chain FA (C4-C10) while plant oils provide much less of these FA. Short chain FA represent a rapid source of energy for the infant, because they enter directly the portal vein and are known to be completely absorbed and oxidized. Myristic and short/medium chain FA could also increase bioavailability and conversion of n-3 PUFA [
5]. Also, it was shown in adults that a moderate intake of dairy lipids and rapeseed oil improved DHA levels in erythrocytes and blood fluidity [
6,
7]. Furthermore, recent studies in omega 3-deficient rodents have demonstrated that for a similar ALA content, a blend of dairy lipids and rapeseed oil induced a higher level of brain DHA than a blend of plant oils even if supplemented with preformed DHA [
8,
9].
Consequently, dairy lipids could represent an interesting strategy to improve lipid composition of infant formula in order to better mimic human milk composition and structure and to optimize growth and health of the infant. In this study, we evaluated the impact on growth as well as the tolerance to a formula containing a blend of dairy lipids and plant oils during the first 4 months of life.
Discussion
This study demonstrated that 4-month consumption of a formula containing dairy lipids was associated with normal growth in healthy term infants and was well tolerated. There were no statistically significant differences in weight, length, or head circumference growth rates among groups fed with formula containing either a mix of dairy and plant lipids, or a blend of plant oils (with or without ARA and DHA). Long Chain PUFA addition had no effect on growth. In this study, growth rates of formula-fed were close to those of breastfed babies. Furthermore, weight, length and head circumference measurements during the 4 first months of life in all groups were similar to the WHO growth standards.
One limitation of this study is that sample size was not initially calculated for the specific outcomes presented here, in particular weight gain. According to the guidance provided by the American Academy of Pediatrics [
13], the number of subjects of a specified sex needed in each group to detect a 3 g/day difference in weight gain over a 3.5 month-period with a power of 0.8 in a one- tailed test is 28. Consequently, this study might be underpowered to detect a statistical difference in weight gain. Body composition was also similar in the 3 formula groups. It has been previously described that formula-fed had different body composition compared to breastfed babies, with higher fat-free mass (in grams) at 4 months and higher fat-free mass changes between enrollment and 4 months [
14]. However, in this previous study, fat mass and fat-free mass at 4 months (both expressed in % of body weight) were similar between breastfed and formula-fed babies [
14]. In our study, fat mass and fat-free mass values in 4 month-old formula-fed infants were close to those described by Gianni et al., 2014 (around 27% and 73% of body weight, respectively). Fat-free mass change after 4 months was also lower in breastfed infants (+ 62%) than in formula-fed babies (+ 74–83%). These results suggest that formula-fed infants show different body composition development during the first 4 months of life compared to breastfed babies. However, the nature of fat used in the tested formulas did not significantly affect body composition.
No significant differences were noted among the 3 formula groups in gastrointestinal parameters (stool frequency, consistency or color), occurrence of gastrointestinal symptoms (abdominal pain, flatulence, regurgitation or vomiting) or infant’s behavior. Stool frequency tended to be higher in breastfed than formula-fed babies (with 42% and 20% of breastfed infants at 1 month at 3 months, respectively, having more than 4 stools/day versus 4 and 0% of formula-fed ones). This result is in agreement with previous data showing that breastfed infants pass stools almost 50% more often than formula-fed ones [
15]. Moreover, solid stools were observed more often in formula-fed than in breastfed babies (19–24% at 1 month and 9–24% at 3 months in formula-fed versus 4 and 0% in breastfed). It has been previously reported that hard stools are associated with formula feeding and are related to the presence of excreted fatty acid soaps and calcium [
16]. Indeed, palmitic acid from plant oils is differentially arranged in the triglyceride (mainly in sn1 and sn3 position, only 5–20% in sn2 position also called beta-palmitate) from palmitic acid from breast milk (60% of beta-palmitate). Palmitic acid from plant oils is poorly absorbed and can react with calcium to form insoluble soaps leading to decreased fat and calcium absorption and increased stool hardness [
17]. In dairy lipids, around 45% of palmitic acid is esterified in sn2 position [
18]. In infant formula, this results in a lower proportion of beta-palmitate in formulas containing only vegetable oils compared with formulas containing dairy lipids or structured lipids [
19]. In our study, the formula containing dairy lipids contained more beta-palmitate than formulas with blends of plant oils. However, no difference was observed among these formula on the occurrence of solid stools. This could be due to methodological considerations with a low number of babies in each group and the inaccuracy of stool consistency report (only 2 time points with evaluation on the last 2 days). Further investigation would be required especially to evaluate if the nature of fat in formula has an impact on fat content in stools.
A high frequency of gastrointestinal symptoms was present throughout the study, especially regurgitations which affected around 50% of babies at 1 month and around 30% of babies at 3 months, without any difference among groups. However, regurgitation can be considered as a physiological phenomenon in the first year of life and these frequencies are in accordance with previous data reporting that more than 50% of infants aged 0–5 months regurgitate at least once a day [
20]. Intermittent episodes of abdominal pain were also observed in 27–44% of formula-fed infants at 1 month and 9–13% at 3 months, which is slightly higher than in the usual range of incidence for colic [
21]. However, in our study, the incidence of colic, as defined by Wessel [
22], was not evaluated. Therefore, the tolerability and safety of a formula containing a mix of dairy and plant lipids did not appear to be different from formulas containing blend of plant oils with or without ARA and DHA.
Acknowledgements
We would like to thank participating families, Elisa Garavaglia and Domenica Mallardi for their contribution to this study, Soladis for statistical analysis and Aurélie Lemaitre for study product management.