Prebiotic oligosaccharides (OS) are the third most prevalent component in breast milk, but they are virtually absent in cow's milk( Reference Kunz, Rudloff and Baier 1 ). Human milk OS are comprised of a complex mixture of glycan compounds. Prebiotics are non-digestible food ingredients that stimulate the growth and/or activity of bacteria in the digestive system in ways that are claimed to be beneficial to human health. It is well known that the gastrointestinal microbiota develops in different ways in breast-fed and standard formula-fed infants( Reference Harmsen, Wildeboer-Veloo and Raangs 2 ). There are many aspects about the composition of breast milk and that of infant formula that contribute to these differences, including their carbohydrate, protein, Fe and phosphor content. The gastrointestinal microbiota is considered to be an organ within an organ that contributes to host nutrition, the developmental regulation of intestinal angiogenesis, protection from pathogens and the development of immune response( Reference Kunz, Rudloff and Baier 1 , Reference Johnson and Versalovic 3 ).
The concept of dietary modulation of the human colonic microbiota with prebiotics was developed about 20 years ago( Reference Gibson and Roberfroid 4 ). In infant formula, the main prebiotics that are used include galacto-oligosaccharide (GOS), fructo-oligosaccharide (FOS) and/or polydextrose (PDX). There is strong clinical support that GOS, which have some similarities to human milk OS, are beneficial for both digestive and immune health( Reference Roberfroid, Gibson and Hoyles 5 ). This was also shown for inulin-type fructans( Reference Kunz, Rudloff and Baier 1 ). The prebiotic effect is defined as ‘the selective stimulation of growth and/or activity(ies) of one or a limited number of microbial genus(era)/species in the gut microbiota that confer(s) health benefits to the host’( Reference Roberfroid, Gibson and Hoyles 5 ).
Prebiotics and bifidogenic effect, gut metabolic activity and stool consistency
Prebiotics are added to infant formula with the aim of achieving a bifidogenic effect on the gastrointestinal microbiota of the host. This effect is by definition global. Protein, in terms of both source (such as whey or α-lactalbumin) and amount, lactose, low phosphor and Fe, lipids (β-palmitate) and nucleotides are all known to have a bifidogenic effect( Reference Yaron, Shachar and Abramas 6 ).
Prebiotics and the bacterial composition of the gastrointestinal microbiota
The type of prebiotic supplement may have a different bifidogenic effect, seeing as short-chain prebiotics are mainly fermented in the caecum and ascending colon. Long-chain OS are fermented in the entire colon( Reference Stewart, Timm and Slavin 7 ). The transcriptional effect of a semi-synthetic medium containing GOS, which, like human milk, contains a large amount of lactose and galactose, on the Bifidodobacterium longum transcriptome has revealed substantial similarity to that of human milk( Reference González, Klaassens and Malinen 8 ). GOS and lactulose support the most favourable growth characteristics, whereas relatively poor Lactobacillus and Bifidodobacterium growth was observed with inulin, maltodextrin and PDX( Reference Watson, O'Connell Motherway and Schoterman 9 ). When comparing different formulas, including oligofructose/FOS (50:50, 4 or 8 g/l), GOS/FOS (90:10, 8 g/l) and standard formula, the bacterial composition in the first two formulas was closer to that of the breast-fed pattern( Reference Veereman-Wauters, Staelens and Van de Broek 10 ). The addition of a GOS/FOS mixture or of GOS only to infant formula has a stimulating effect on the growth of Bifidodobacterium ( Reference Knol, Scholtens and Kafka 11 – Reference Costalos, Kapiki and Apostolou 18 ). GOS alone was shown to increase Lactobacillus ( Reference Ben, Li and Feng 14 , Reference Ben, Zhou and Zhao 15 ). The proportion of Bifidodobacterium was higher in the prebiotic group, but not significantly (P= 0·262), whereas faecal clostridia was significantly lower (P= 0·042)( Reference Scholtens, Alliet and Raes 12 ). Similar results were reported with a GOS/FOS mixture: there was a significant increase in Bifidodobacterium and a decrease in clostridia( Reference Ben, Zhou and Zhao 15 ). Bifidodobacterium breve growth was higher and Clostridium difficile was lower in 365 infants that were fed GOS at 4 g/l in infant formula and 5 g/l in follow-up formula as compared to control formula( Reference Sierra, Bernal and Blasco 13 ). Supplementation with other prebiotics, such as PDX and GOS, led to comparable results( Reference Scalabrin, Mitmesser and Welling 19 ).
In preterm infants, FOS-supplemented formula has been shown to have positive effects on gut microbiota development: the number of Bifidodobacterium and proportion of infants colonised were significantly higher in the study group (P= 0·032 and P= 0·03, respectively), which resulted in a significant reduction of Escherichia coli and Enterococci (P= 0·029 and P= 0·025, respectively)( Reference Kapiki, Costalos and Oikonomidou 20 ).
The magnitude of the prebiotic impact may be age dependent, seeing as the bifidogenic effect has been reportedly larger when administration was begun in early infancy( Reference Nakamura, Gaskins and Collier 21 ). However, GOS supplementation also significantly increased Bifidodobacterium numbers in vivo and in vitro in women and men who were more than 50 years old( Reference Walton, van den Heuvel and Kosters 22 ). Prebiotics have also been shown to have the potential to reduce disturbances in the gut microbiota that were induced by antibiotics( Reference Ladirat, Schuren and Schoterman 23 ). Changes in the relative abundance of bacteria after antibiotic treatment and the growth of Bifidodobacterium and Lactobacillus after the addition of GOS were dependant on the type of antibiotic and dose amount( Reference Ladirat, Schuren and Schoterman 23 ).
No local or systemic side effects were recorded( Reference Fedorak and Madsen 17 ).
It can be concluded that the majority of prebiotics are able to influence the intestinal microbiota of young infants.
Prebiotics and faecal pH
Faecal pH was lower (P= 0·01) in breast-fed and prebiotic formula-fed infants than it was in standard infant formula-fed infants( Reference Scholtens, Alliet and Raes 12 ). Infant formula containing 0·24 g/100 ml( Reference Ben, Li and Feng 14 ) and 0·44 g/100 ml of GOS( Reference Sierra, Bernal and Blasco 13 ) has been shown to bring faecal pH to a similar level as that of infants fed human milk. Intestinal pH modulates the intestinal environment by inhibiting or favouring the growth of different bacterial populations; a low intestinal pH results in a decrease of pathogenic micro-organisms( Reference Costalos, Kapiki and Apostolou 18 ).
Prebiotics and metabolic activity
Feeding infants a GOS/FOS or a GOS-only formula has resulted in a similar effect on the metabolic activity of the microbiota as that found in breast-fed infants( Reference Knol, Scholtens and Kafka 11 , Reference Sierra, Bernal and Blasco 13 , Reference Bakker-Zierikzee, Alles and Knol 24 ). The changes in SCFA and lactate (and pH) in one study's prebiotic group represented a fermentation profile that was closer to that observed in breast-fed infants as compared to infants fed control formula( Reference Knol, Scholtens and Kafka 11 ). Because SCFA represent the metabolic activity of intestinal microbiota( Reference Bakker-Zierikzee, Alles and Knol 24 ), these findings suggest that the gut microbiota of infants fed a prebiotic-supplemented formula develops similar to that in breast-fed infants.
Acetic acid, whose formation is consistent with Bifidodobacterium metabolism, was the major SCFA synthesised( Reference Rodriguez-Colinas, Kolida and Baran 25 ). Purified GOS can be considered bifidogenic( Reference Rodriguez-Colinas, Kolida and Baran 25 ). The effect of GOS on intestinal metabolic activity has been observed in a recent study: GOS supplementation resulted in a significant increase in total SCFA, higher acetic acid and lower butyric and propionic acids as compared to breast-fed infants( Reference Sierra, Bernal and Blasco 13 ).
Prebiotics and stool patterns
Different levels of GOS (0·24 g/100 ml and 0·4 g/100 ml) in infant formula increase stool frequency, as does breast milk( Reference Ben, Li and Feng 14 – Reference Rinne, Gueimonde and Kalliomäki 16 ). Prebiotics increase the bacterial mass and the osmotic water-binding capacity in the gut lumen. These actions increase stool weight and frequency, and they also soften the stool, which indirectly contributes to decreased transit time.
Compared to infants who received standard formula, infants who received prebiotic supplementation experienced softer stools that were similar to the consistency reported in breast-fed infants( Reference Schmelzle, Wirth and Skopnik 26 ). Standard infant formula with 4 g/l GOS had beneficial effects on both stool consistency and additional measures of gastrointestinal tolerance by newborn infants during the first 4 months of life( Reference Williams, Choe and Price 27 ). Infants fed a prebiotic GOS/FOS mixture had softer stools than those fed standard infant formula (P= 0·026)( Reference Scholtens, Alliet and Raes 12 ). The infants that consumed PDX/GOS (1:1, 4 g/l) for 60 d had softer stools than the control group at all times (P< 0·001)( Reference Scalabrin, Mitmesser and Welling 19 ). No differences in formula intake or infant fussiness or gassiness were observed. During study weeks 1 and 2 and at 60 d of age, stool consistency ratings were higher (i.e. softer stools) for infants in the PDX/GOS and GOS groups v. the control group and remained higher at 120 d for the PDX/GOS group (all P< 0·05)( Reference Scalabrin, Mitmesser and Welling 19 ). In a systematic review, prebiotics were reported to have increased stool frequency but had no impact on stool consistency, the incidence of colic, spitting up/regurgitation, crying, restlessness or vomiting( Reference Mugambi, Musekiwa and Lombard 28 ). When comparing a mixture of oligofructose/FOS (50:50; 4 or 8 g/l), GOS/FOS (90:10, 8 g/l) and standard formula, the stool consistency of infants who were fed 8 g/l of GOS/FOS-supplemented formula was closer to the breast-fed pattern( Reference Veereman-Wauters, Staelens and Van de Broek 10 ).
Follow-up formulas with only 5 g/l GOS that were provided to infants aged 6 months to 1 year resulted in a higher frequency and softer consistency of stools( Reference Sierra, Bernal and Blasco 13 ). Supplementation with 5 g/l GOS did not alter stool frequency, but it did decrease stool consistency( Reference Fedorak and Madsen 17 ).
Several prebiotic mixtures have been evaluated in different studies. In a systematic review about the effect of prebiotics on stool frequency and consistency( Reference Rao, Srinivasjois and Patole 29 ), four randomised controlled trials (RCT) demonstrated that the stool frequency of the prebiotic group was similar to that of breast-fed infants and was significantly higher than that of infants fed standard formula. Similar data were obtained regarding stool consistency: six RCT showed significantly softer stools in the prebiotic group( Reference Rao, Srinivasjois and Patole 29 ). In a large study in 1130 infants, the stool consistency in the prebiotic group was softer than in the control group at 8, 16 and 24 weeks (P< 0·001) and closer to that of the breast-fed group( Reference Piemontese, Giannì and Braegger 30 ).
Prebiotics and general health
GOS has been shown to reduce Ca excretion in faeces in normal and gastrectomised rats, and thus to prevent osteopenia( Reference dos Santos, Tsuboi and Araújo 31 ). The addition of a specific mixture of 0·6 g/100 ml of short-chain galacto-oligosaccharide (scGOS) and long-chain fructo-oligosaccharides (lcFOS) in a 9:1 ratio to infant formula did not decrease cholesterol concentrations, although the levels in formula-fed infants were significantly lower than those in breast-fed infants( Reference Alliet, Scholtens and Raes 32 ). One open trial of 342 healthy infants compared standard infant formula with a GOS/FOS-enriched formula. The formulas were given for 6 months, and at the age of 12 months, the incidence of gastroenteritis was lower (0·12 ± 0·04 v. 0·29 ± 0·05 episodes/child per 12 months; P= 0·015), the number of children with more than three episodes tended to be lower (17/60 v. 29/65; P= 0·06) and the number of children with multiple antibiotic courses per year was lower (24/60 v. 43/65; P= 0·004) in the enriched-formula group( Reference Bruzzese, Volpicelli and Squeglia 33 ). Infants that were fed an scGOS/lcFOS mixture were reported in another study to have fewer episodes of physician-diagnosed overall and upper respiratory tract infections (P< 0·01), fewer fever episodes (P< 0·00001) and fewer antibiotic prescriptions (P< 0·05) during the period of intervention and up to the age of 2 years( Reference Arslanoglu, Moro and Schmitt 34 , Reference Arslanoglu, Moro and Boehm 35 ). According to these results, early dietary intervention with an OS prebiotic mixture had a protective effect against infections( Reference Arslanoglu, Moro and Schmitt 34 ). However, in other randomised clinical trials, these results have not been confirmed. In one large, randomised, double-blind, placebo-controlled trial performed in 830 healthy term infants in seven centres in five Western European countries, adding specific prebiotics to standard formula-feeding had no effect on reducing the number of fever episodes( Reference van Stuijvenberg, Eisses and Grüber 36 ). These results are in alignment with those from another recent randomised, double-blind, placebo-controlled clinical trial that used only GOS as a prebiotic source in infant formula; in that study, no differences were observed in allergic manifestations up to 12 months of age, and there were no differences in the numbers of diarrhoea episodes or prescriptions of antibiotics( Reference Sierra, Bernal and Blasco 13 ). This difference in results may be explained by differences in the study populations: in the last study, all of the infants were healthy term infants who lived in modern environments with a low risk of infectious disease, but this was not the case in some of studies described earlier. Further studies in populations with a higher incidence of infections, perhaps in developing countries, and further studies on the effects of prebiotics on allergy manifestations in infants with a high risk of atopy are needed( Reference Sierra, Bernal and Blasco 13 ).
Prebiotics and growth
A transient increase in body weight was observed in children on prebiotics as compared to the control group during the first 6 months of follow-up( Reference Bruzzese, Volpicelli and Squeglia 33 ). Prebiotics in formula increased weight gain but had no impact on length or head circumference gain( Reference Scholtens, Alliet and Raes 12 , Reference Mugambi, Musekiwa and Lombard 28 ). However, the quality of this evidence is compromised by imprecision, an inconsistency of results, the use of different study preparations and publication bias( Reference Mugambi, Musekiwa and Lombard 28 ). As a consequence, evidence for the benefit of symbiotics added to infant formula is equally limited.
A multi-centre, randomised, double-blind, placebo-controlled trial including 1130 healthy term infants showed similar mean weight, length, head circumference, skinfold thickness and arm circumference gains in all infants( Reference Arslanoglu, Moro and Boehm 35 ). The skinfold thicknesses assessed in the breast-fed group at 8 weeks were strikingly larger than those in the formula-fed infants, whereas at 52 weeks, they were strikingly smaller( Reference Arslanoglu, Moro and Boehm 35 ).
Other studies have not observed these differences in growth parameters. One study that compared prebiotic-enriched to standard infant formula showed no group differences in growth rates from 14 to 120 d of age( Reference van Stuijvenberg, Eisses and Grüber 36 ). Discontinuation rates were not significantly different among the study groups. The tolerance of a formula enriched with FOS/GOS was excellent, and no differences in growth parameters were observed after 6 and 12 weeks( Reference Scholtens, Alliet and Raes 12 , Reference Arslanoglu, Moro and Schmitt 34 , Reference van Stuijvenberg, Eisses and Grüber 36 ). According to a systematic review from 2009, ten publications evaluated the effect of prebiotics on growth during the 1st year of life( Reference Braegger, Chmielewska and Decsi 37 ). None of these trials showed a significant difference in growth parameters; infants fed the supplemented formula had only slightly better weight gain (weighted mean difference 1·07 g, 95 % CI 0·14, 1·99; four RCT)( Reference Braegger, Chmielewska and Decsi 37 ). This is in agreement with the results obtained recently by Sierra et al. ( Reference Sierra, Bernal and Blasco 13 ), which observed no significant differences in growth.
Prebiotics and immune parameters
A greater number of Bifidodobacterium may promote postnatal maturation of the immune system and thus have a protective effect against infections and allergies. The correlation between Bifidodobacterium and the amount of intestinal secretory IgA (sIgA) has been known for many years( Reference Scholtens, Alliet and Raes 12 , Reference Fukushima, Kawata and Hara 38 , Reference Yasui, Mike and Ohwaki 39 ). Prebiotic inulin and oligofructose clearly modulate immunological processes at the level of the gut-associated lymphoid tissue, and they may be associated with significant health benefits in infants and patients who have intestinal inflammatory diseases( Reference Seifert and Watzl 40 ). The prebiotic effect that was observed after the intervention period suggests that an immune modulating effect through modification of the intestinal microbiota may be the principal mechanism of action( Reference van Stuijvenberg, Eisses and Grüber 36 ). One area of research on modern infant formula focuses on the development of the immune system. The addition of long-chain PUFA is outside the topic of the present review, but they are shown to positively influence the development of the immune system( Reference Field, Van Aerde and Robinson 41 ).
In practice, these different trial parameters are often combined, which makes it more difficult to know ‘which parameter did what’. This has been observed in the RCT where 0·44 g/100 ml and 0·5 g/100 ml of GOS were added to infant and follow-up formulas, respectively; long-chain PUFA and nucleotides were also included in the standard formula( Reference Sierra, Bernal and Blasco 13 ). The addition of a specific mixture of 0·6 g/100 ml of scGOS and lcFOS in a 9:1 ratio to infant formula showed that immune parameters at weeks 8 and 26 did not differ between the two groups( Reference Raes, Scholtens and Alliet 42 ). The mixture of prebiotic OS did not change the basal level of the measured parameters of the developing immune system in healthy infants with a balanced immune system during the first 6 months of life as compared to that in standard formula-fed infants and that in exclusively breast-fed infants( Reference Raes, Scholtens and Alliet 42 ). No significant differences were detected between PDX/GOS supplementation and the control group in changes from baseline to 30 or 60 d for sIgA( Reference Scalabrin, Mitmesser and Welling 19 ). After 26 weeks of intervention with the same mixture, the concentration of sIgA was higher (P< 0·001) in the scGOS/lcFOS group (719 μg/g) than it was in the control group (263 μg/g), whereas the concentration in the breast-fed group was similar to that in the intervention group( Reference Ben, Zhou and Zhao 15 ). In another study, GOS reduced colitis by modulating the function and trafficking of natural killer cells in mice( Reference Gopalakrishnan, Clinthorne and Rondini 43 ).
There is only one study which provided evidence that GOS/FOS supplementation induced a beneficial antibody profile( Reference van Hoffen, Ruiter and Faber 44 ). A GOS/FOS mixture reduced the total Ig response and modulated the immune response to cow milk protein but left the response to vaccination intact( Reference van Hoffen, Ruiter and Faber 44 ). In a recent multi-centre, randomised, double-blind, placebo-controlled trial carried out in 365 infants, sIgA concentration decreased in both groups, and there was a trend towards a lower decrease in the GOS group( Reference Sierra, Bernal and Blasco 13 ). This suggests a protective effect on the development of a gut mucosa immune response, given that the sIgA decrease was more stable over time in the infants who consumed the GOS-containing formula; it also reflects a similar behaviour to that observed in breast-fed infants( Reference Scalabrin, Mitmesser and Welling 19 , Reference Bakker-Zierikzee, Tol and Kroes 45 ).
Prebiotics and allergies and atopic dermatitis
Prebiotics were shown to reduce Ig free light chain concentrations in infants who were at risk for allergies( Reference Schouten, Van Esch and Kormelink 46 ). GOS has been shown to block atopic dermatitis, such as skin lesions, in mice by at least partly inducing the production of IL-10 and suppressing the production of cytokines, such as IL-17, which are involved in skin inflammation( Reference Tanabe and Hochi 47 ). In a large study in 259 high-risk infants, prebiotic supplements to formula-feeding reduced the development of atopic dermatitis at the age of 6 months( Reference Moro, Arslanoglu and Stahl 48 ). However, systematic reviews have failed to prove the efficiency of prebiotics in atopic dermatitis prevention( Reference Williams and Grindlay 49 ). Early dietary intervention with OS has been shown to have a protective effect against allergy manifestations. When they were started early in life, OS (scGOS/lcFOS) had a protective effect against allergy manifestations in high-risk infants that lasted beyond infancy until the age of 5 years for atopic dermatitis and allergic rhinoconjunctivitis( Reference Williams and Grindlay 49 , Reference Vos, van Esch and M'Rabet 50 ). Although the intervention group had a 75 % reduction in the prevalence of persistent wheezing (4·8 v. 14 %), no significance was shown( Reference Williams and Grindlay 49 , Reference Vos, van Esch and M'Rabet 50 ). The specific scGOS/lcFOS mixture did, however, reduce the allergic reaction to ovalbumin in mice( Reference Vos, van Esch and M'Rabet 50 ). The use of GOS/FOS in dietary products has also been shown to possibly stimulate the adaptive immune response in a T helper 1 direction and subsequently inhibit infections and T helper 2-related immune disorders, such as allergies, in human subjects( Reference Vos, Haarman and Buco 51 ).
The preventive effects of prebiotic feeding during perinatal and post-weaning periods in mouse models of allergies were analysed by studying biomarkers related to tolerance (IgG2a, IgA, interferon-γ, transforming growth factor-β and IL-10), to allergies (IgE, IgG1, IL-4, IL-17 and symptoms) and to microbiota (propionate and MyD88)( Reference Gourbeyre, Desbuards and Grémy 52 ). Regardless of diet, sensitised mice exhibited similar levels of IgE, IgG1, CD-23, IL-4, IL-17 and symptoms( Reference Gourbeyre, Desbuards and Grémy 52 ). Prebiotic exposure during the perinatal and post-weaning periods induced the highest expression of biomarkers related to tolerance without affecting the biomarkers related to allergies( Reference Gourbeyre, Desbuards and Grémy 52 ). A Cochrane review concluded that there is some evidence that a prebiotic supplement added to infant formula may prevent eczema( Reference Osborn and Sinn 53 ). A reduced occurrence of early atopic dermatitis in infants that had a low risk for atopy has also been reported( Reference Grüber, van Stuijvenberg and Mosca 54 ). In another recent study in 365 infants, no differences were observed in the incidence of allergies or infections( Reference Sierra, Bernal and Blasco 13 ). In fact, a systematic review has concluded that the data available in relation to the use of prebiotics in infant formulas for the prevention of infections remain inconclusive( Reference Braegger, Chmielewska and Decsi 37 ).
It is unclear whether the use of prebiotics should be restricted to infants who are at a high risk for allergies and whether they may also have an effect on other allergic diseases, including asthma( Reference Braegger, Chmielewska and Decsi 37 ). Further research is needed before the routine use of prebiotics can be recommended for the prevention of allergies in formula-fed infants( Reference Braegger, Chmielewska and Decsi 37 ). Data on symbiotic combinations are still limited.
Conclusion
Even if there is not enough evidence to state that supplementation of infant formula with prebiotics does result in a relevant clinical benefit( Reference Braegger, Chmielewska and Decsi 37 ), there is evidence that prebiotics added to infant formula have beneficial effects on gut microbiota, metabolic activity, stool consistency and frequency and the development of some immune markers. Because prebiotics are present in breast milk as human milk OS, and because their addition to infant formula is safe, there seems to be no reason not to add them to infant formula, even though the evidence for the clinical benefit of doing so is limited. Future research should focus on specificity, safety, dosage and combinations.
