Effect of breast-feeding

Although the advantages of breast milk as the optimal form of feeding for children in the first months of life cannot be overemphasised, the role of exclusive breast-feeding in the prevention of allergic diseases is not clear⁽Reference Nwaru, Erkkola and Ahonen^9, Reference Dattner¹⁰⁾. Since the 1930s, the effect of breast-feeding on the risk of developing allergic diseases has been frequently studied and debated⁽Reference Pohlabeln, Mühlenbruch and Jacobs^11–Reference Greer, Sicherer and Burks¹³⁾. Conflicting findings have been reported from several studies investigating whether prolonged and exclusive breast-feeding increases, decreases or has no effect on the risks of asthma and allergy⁽Reference Nwaru, Erkkola and Ahonen^9, Reference Kramer, Matush and Vanilovich^12, Reference van Odijk, Kull and Borres¹⁴⁾. Some studies have reported greater degrees of protection with more exclusive and prolonged breast-feeding and several have noted a larger protective effect in children prone to atopy⁽Reference Kramer, Matush and Vanilovich^12, Reference Greer, Sicherer and Burks¹³⁾. However, recent studies have raised concern that breast-feeding may not protect children and may even increase the risk⁽Reference Kramer, Matush and Vanilovich^12, Reference Giwercman, Halkjaer and Jensen¹⁵⁾. In general, many of these studies have been non-randomised, retrospective or observational in design and, thus, inconclusive. Of course, it is not possible to truly randomise breast-feeding, which is always a confounding variable in these studies⁽Reference Greer, Sicherer and Burks¹³⁾.

It has been reported that among other environmental factors (early infections, early day-care attendance, presence of two or more siblings at birth and rural environment) breast-feeding for more than 3 months was protective for the development of asthma⁽Reference Midodzi, Rowe and Majaesic^16–Reference Kusunoki, Morimoto and Nishikomori¹⁸⁾. In 2001, two systematic reviews with meta-analysis on the risk of eczema and asthma reported lower incidence rates of these atopic diseases in children who were exclusively breastfed, with the stronger effect shown for infants with a family history of allergy⁽Reference Gdalevich, Mimouni and David^19, Reference Gdalevich, Mimouni and Mimouni²⁰⁾. A series of reports from the German Infant Nutritional Intervention Programme also found that breast-feeding reduces the incidence of atopic dermatitis, supporting the results of the meta-analysis⁽Reference Greer, Sicherer and Burks^13, Reference Gdalevich, Mimouni and David^19, Reference von Berg, Koletzko and Grubl²¹⁾. The Prevention and Incidence of Asthma and Mite Allergy birth cohort study that included 3115 Dutch children showed that breast-feeding (>16 weeks v. no breast-feeding) was significantly associated with a lower asthma prevalence from 3 to 8 years of age, in children of both non-allergic and allergic mothers⁽Reference Scholtens, Wijga and Brunekreef²²⁾. Furthermore, a study from New Zealand that enrolled a total of 1105 children, showed a significant protective effect of breast-feeding on infant wheezing and other adverse respiratory outcomes that may be early indicators of asthma. Even after adjustment for confounders, each month of exclusive breast-feeding reduced the risk of doctor-diagnosed asthma by 20%, wheezing by 12% and inhaler use by 14%⁽Reference Silvers, Frampton and Wickens²³⁾. Codispoti et al. showed that African–American infants receiving prolonged breast-feeding had significantly decreased risk of allergic rhinitis at the age of 3 years, although this breast-feeding effect was not seen in non-African–American children⁽Reference Codispoti, Levin and LeMasters²⁴⁾.

When compared with breast-feeding with supplemental cow's milk formula, exclusive breast-feeding for 4 months showed a protective effect on developing allergy in infants at high risk⁽Reference Greer, Sicherer and Burks¹³⁾. Breast-feeding with supplemental hydrolysed formula (both partially and extensively hydrolysed) also showed a positive effect compared with breast-feeding with supplemental cow's milk formula. However, breast-feeding with supplements of hydrolysed formulas showed no advantage compared with exclusive breast-feeding. Both groups showed a one-third decrease in the risk of atopic dermatitis compared with the risk of breast-feeding with supplements of cow's milk formula⁽Reference von Berg, Koletzko and Grubl²¹⁾.

Concerning asthma, in the study of Fonseca et al., the authors concluded that prolonged breast-feeding (6 months or longer) could reduce the risk of asthma in boys who live in the city⁽Reference Fonseca, Moreira and Moreira²⁵⁾. Furthermore, breast-feeding seems to decrease the wheezing episodes seen in younger children (<4 years of age) that are often associated with respiratory infections⁽Reference Greer, Sicherer and Burks¹³⁾.

In contrast to the above, a 2005 study published from Sweden found no effect of exclusive breast-feeding for less than 4 months on the incidence of atopic dermatitis in the first year of life with or without a family history of atopic disease⁽Reference Ludvigsson, Mostrom and Ludvigsson²⁶⁾. On the other hand, another 2005 study from Sweden found that exclusive breast-feeding for more than 4 months reduced the risk of atopic dermatitis at 4 years of age with or without a family history of allergy⁽Reference Kull, Bohme and Wahlgren²⁷⁾. In their review, Kramer and Kakuma also found no benefit of exclusive breast-feeding beyond 3 months of age on the incidence of atopic dermatitis in studies in which parents were not selected for risk of allergy⁽Reference Kramer and Kakuma²⁸⁾.

Furthermore, Kramer et al. conducted a cluster-randomised trial on 13 889 infants and found that prolonged and exclusive breast-feeding had no protective effect on allergic symptoms and diagnoses or on positive skin-prick tests⁽Reference Kramer, Matush and Vanilovich¹²⁾. Other studies restricted to populations with increased risk have even reported the increased risk from breast-feeding of eczema, wheezy disorder, asthma and sensitisation. Some authors reported a small protective effect on eczema and a dual effect, protecting in high-risk infants but increasing risk in infants without such heredity⁽Reference Giwercman, Halkjaer and Jensen¹⁵⁾.

Pohlabeln et al. observed a significantly increased risk of developing an atopic disease in children without a genetic allergic disposition who were exclusively breastfed for more than 4 months (OR 1·62, 95% CI 1·02, 2·56). In contrast, the appearance of allergic diseases in 2-year-old children with a family history of atopy was less common in breastfed children than in never-breastfed children. Nonetheless, when analysed in greater detail, children with a maternal-only predisposition had a considerably higher risk of allergic disease symptoms at the age of 2 years if they had been exclusively breastfed for more than 4 months than if they had never been breastfed. Children with a paternal-only predisposition had a significantly decreased risk if they had been breastfed for more than 4 months than if they had never been breastfed (OR 0·39, 95% CI 0·18, 0·83). Breast-feeding for ⩽4 months seemed to have no influence on increased or decreased risk of allergic disease in children with either a paternal or maternal allergic predisposition⁽Reference Pohlabeln, Mühlenbruch and Jacobs¹¹⁾.

An often-discussed explanation for the higher rates of allergies detected in children who have been breastfed for more than 4 months is the reversal of cause and effect. In other words, on detecting early signs of atopic disease in their infant, mothers who knew that their infants were at risk of developing allergy were more likely not only to breastfeed but also to breastfeed for a longer period of time (which would explain the higher number of children with allergic symptoms within the subgroup of children breastfed for more than 4 months)⁽Reference Pohlabeln, Mühlenbruch and Jacobs^11, Reference Kusunoki, Morimoto and Nishikomori^18, Reference von Berg, Koletzko and Grubl²¹⁾. To control for such an effect, Pohlabeln conducted an additional analysis excluding children with reported symptoms of allergic disease in the first 12 months of life and the results remained almost unchanged⁽Reference Pohlabeln, Mühlenbruch and Jacobs¹¹⁾. Kusunoki et al.⁽Reference Kusunoki, Morimoto and Nishikomori¹⁸⁾ showed that children with mixed and complete breast-feeding showed a significantly lower prevalence of bronchial asthma (P=0·04 and P=0·003 respectively). On the other hand, the prevalence of atopic dermatitis and food allergy were significantly higher in those with complete breast-feeding (P=0·04 and P=0·01 respectively). However, there was a higher proportion of complete breast-feeding among those with greater risk of allergic diseases (presence of family history, either eczema or wheeze within 6 months after birth or food allergy in infancy) and when analysis included these risks as confounding factors, the promoting effects of breast-feeding on atopic dermatitis and food allergy disappeared suggesting that this effect was most likely because of reverse causation⁽Reference Kusunoki, Morimoto and Nishikomori¹⁸⁾. Giwercman et al. has shown, in a birth cohort born of mothers with asthma, that duration of exclusive breast-feeding increased the infant's risk of eczema during the first 2 years of life. In contrast, the same study showed that breast-feeding protected infants from wheezy disorders and severe wheezy exacerbations⁽Reference Giwercman, Halkjaer and Jensen¹⁵⁾.

Overall, at the present time, firm conclusions about the role of breast-feeding in either preventing or delaying the onset of specific food allergies are not possible⁽Reference Greer, Sicherer and Burks¹³⁾. However, exclusive breast-feeding should always be encouraged.

Hydrolysed formula

When infants are not exclusively breastfed, partially hydrolysed formula (infant formula where cow's milk proteins were modified by hydrolysation processes) may be considered an effective measure to potentially reduce the risk of developing atopic dermatitis⁽Reference Alexander, Schmitt and Tran^3, Reference Alexander, Schmitt and Tran^6, Reference Alexander, Schmitt and Tran^29–Reference von Berg, Krämer and Link³¹⁾. Specifically, it was observed that feeding with partially-hydrolysed whey–protein formula instead of intact protein cow's milk formula reduces the risk of atopic dermatitis in infants, particularly in infants with a family history of allergy⁽Reference Alexander, Schmitt and Tran^3, Reference von Berg^6, Reference Greer, Sicherer and Burks^13, Reference van Odijk, Kull and Borres^14, Reference Alexander and Cabana²⁹⁾. Depending on the degree of modification, hydrolysed cow's milk formulas are differentiated into extensively and partially hydrolysed whey or casein hydrolysates⁽Reference von Berg⁶⁾. However, comparative studies of the various hydrolysed formulas indicate that not all formulas have the same protective benefit⁽Reference von Berg^6, Reference Greer, Sicherer and Burks^13, Reference von Berg, Koletzko and Grubl²¹⁾. Extensively and partially hydrolysed formulas can be used for primary prevention of allergy in infants at high atopic risk, while only extensively hydrolysed formulas are indicated for secondary prevention in patients with manifest cow's milk allergy. If atopic disease associated with cow's milk allergy has occurred, partially-hydrolysed formula is not recommended, because it contains potentially allergenic cow milk peptides⁽Reference von Berg, Koletzko and Grubl²¹⁾.

The role of partially hydrolysed and extensively hydrolysed formulas for the prevention of atopic disease has been the subject of many studies in both formula-fed and breastfed infants in the last 15 years⁽Reference Greer, Sicherer and Burks¹³⁾. The German Infant Nutritional Intervention Programme study showed that the incidence of atopic dermatitis was substantially reduced in those using the extensively hydrolysed casein-based formula and the partially-hydrolysed whey-based formula but not the extensively hydrolysed whey-based formula, compared with the incidence in those in the cow's milk formula group⁽Reference von Berg, Koletzko and Grubl²¹⁾. The prospective German birth cohort study GINIplus demonstrated that predisposed children without nutritional intervention had a 2·1 times higher risk for eczema than children without a familial predisposition. In other words, the data demonstrate that early intervention with hydrolysed infant formulas can substantially compensate up until the age of 6 years for an enhanced risk of childhood eczema due to familial predisposition to allergy⁽Reference von Berg, Krämer and Link³¹⁾.

Meta-analysis of clinical trial and intervention studies was conducted and included a total of eighteen articles on hydrolysed infant formula and the risk of atopic dermatitis. A statistically significant 44% reduced risk of atopic manifestations was found among infants receiving whey protein partially hydrolysed formula compared with infants receiving intact protein cow's milk formula (summary relative risk estimate 0·56, 95% CI 0·40, 0·77). In a subanalysis of four studies that reported results specifically for atopic dermatitis, the incidence of atopic dermatitis was reduced by 55%⁽Reference Alexander and Cabana²⁹⁾.

The use of amino acid-based formulas for prevention of atopic disease has not been studied. Soya formulas, on the other hand, have a long history of use for atopic disease in infants. In a recent meta-analysis of five randomised or quasi-randomised studies, the authors concluded that feeding with soya formula should not be recommended for the prevention of atopy in infants at high risk of developing allergy⁽Reference Greer, Sicherer and Burks¹³⁾.

In conclusion, exclusive breast-feeding should be encouraged as the standard for infant nutrition in the first months of life⁽Reference Alexander and Cabana^29, Reference Grimshaw, Allen and Edwards³⁰⁾. However, when infants are not exclusively breastfed, partially hydrolysed formula may be considered an effective measure to potentially reduce the risk of developing atopic dermatitis⁽Reference Alexander, Schmitt and Tran^3, Reference von Berg^6, Reference Grimshaw, Allen and Edwards^30, Reference von Berg, Krämer and Link³¹⁾.

Effect of introducing solids

Despite the lack of evidence, there have been a lot of changes in the timing of first exposure to solid foods over the last 40 years. In the 1960s, most infants had been exposed to solids by 4 months of age, while in the 1970s guidelines recommended delayed introduction of solids until after 4 months. By the late 1990s, expert bodies began to recommend delaying solids until after 6 months of age⁽Reference Koplin, Osborne and Wake⁸⁾. Until very recently, for infants with a family history of allergy expert guidelines recommended delaying introduction of allergenic foods (including avoiding eggs until 2 years and nuts until 3 years of age), as well as delaying solid foods until after 6 months and continuation of breast-feeding for at least 12 months⁽Reference Koplin, Osborne and Wake^8, Reference Greer, Sicherer and Burks^13, Reference McLean and Sheikh³²⁾. These recommendations have been challenged by recent population studies, creating the concern that the current practice of delaying complementary foods until 6 months of age may increase, rather than decrease, the risk of immune disorders⁽Reference Cochrane, Beyer and Clausen^1, Reference Nwaru, Erkkola and Ahonen^9, Reference Greer, Sicherer and Burks^13, Reference Anderson, Malley and Snell^33–Reference Zutavern, Brockow and Schaaf³⁵⁾.

In a systematic review of thirteen studies, five studies found a positive association between early introduction of solid foods (before 4 months of age) and eczema, one study found an association with pollen allergy, and others reported no association between early introduction of solid foods and the development of allergic diseases⁽Reference Tarini, Carroll and Sox³⁶⁾. In contrast, Sariachvili et al. showed that early introduction (within the first 4 months) of solid foods was associated with a reduced risk for eczema among children with allergic parents, whereas no significant effect was found among children with non-allergic parents⁽Reference Sariachvili, Droste and Dom³⁷⁾. Furthermore, a review by Prescott et al. showed an increased risk of allergy if solids were introduced before 3–4 months⁽Reference Prescott, Smith and Tang³⁴⁾. Some other prospective studies supported neither prolonged breast-feeding nor delayed introduction of solid foods for the prevention of allergic diseases in children⁽Reference Nwaru, Erkkola and Ahonen^9, Reference Zutavern, Brockow and Schaaf³⁵⁾. The results from the population-based, prospective, cohort study by Nwaru et al. provided evidence for increased risk of allergic sensitisation to food and inhalant allergens with delayed introduction of solid foods⁽Reference Nwaru, Erkkola and Ahonen^9, Reference Poole, Barriga and Leung³⁸⁾. However, the concept of reverse causality has to be considered in such studies because families with a history of allergic diseases or with infants with early signs of allergy may delay the introduction of solids. Therefore, separate analyses of data for the subgroup of children with parental history of asthma or allergic rhinitis were performed in the study by Nwaru et al. and results did not change, demonstrating no evidence of reverse causality⁽Reference Nwaru, Erkkola and Ahonen⁹⁾. These findings are consistent with those of Koplin et al. who showed that infants introduced to egg at 4–6 months had a lower risk of egg allergy than those introduced to egg after that time, particularly those introduced to egg at 10–12 months of age and after 12 months of age, even after adjusting for family history of allergy and infant-allergy symptoms⁽Reference Koplin, Osborne and Wake⁸⁾. Furthermore, a large, prospective, German population-based birth cohort study on asthma and allergic disease did not find evidence to support a delayed introduction of solids beyond 6 months of life for the prevention of atopic dermatitis and atopic sensitisation. However, in the analysis of data of that cohort at the age of 6 years, late introduction of solid foods was associated only with increased risk of sensitisation to food allergens and not with sensitisation to inhalant allergens, eczema, asthma or allergic rhinitis⁽Reference Zutavern, Brockow and Schaaf³⁵⁾. Moreover, data from wheat allergic patients obtained from a longitudinal birth cohort study showed that delaying exposure to wheat until after 6 months was associated with an increased risk of wheat allergy, not a protective effect⁽Reference Poole, Barriga and Leung³⁸⁾. It has also been shown that early introduction of fish into the child's diet was associated with less eczema development and a tendency to less asthma, while sensitisation was not associated with the timing of fish introduction⁽Reference Hesselmar, Saalman and Rudin³⁹⁾. Moreover, countries where peanuts are commonly used as weaning foods, like Israel for example, have low incidence of peanut allergy⁽Reference Du Toit, Katz and Sasieni⁴⁰⁾. A UK birth cohort study investigated the effect of age at the introduction of solid foods on wheezing, eczema, and atopy among children 5·5 years of age and showed that late introduction of eggs and milk was associated with increased risk of eczema. However, the retrospective collection of data on the introduction of solid foods might have introduced recall bias in that study⁽Reference Nwaru, Erkkola and Ahonen⁹⁾.

Based on the current available data, it was proposed that early complementary food introduction may hasten and/or maintain oral mucosal tolerance rather than increase the risk of food allergy⁽Reference Grimshaw, Allen and Edwards³⁰⁾. Tolerance to food allergens appears to be driven by regular, early exposure to these proteins during a ‘critical early window’ of development⁽Reference Prescott, Smith and Tang³⁴⁾. Although the timing of this window is not clear, current evidence suggests that this is most likely to be between 4 and 6 months of life and that delayed exposure beyond this period may increase the risk of food allergy, coeliac disease and islet-cell autoimmunity⁽Reference Koplin, Osborne and Wake^8, Reference Anderson, Malley and Snell^33, Reference Prescott, Smith and Tang³⁴⁾. There is also evidence that other factors such as favourable colonisation and continued breast-feeding promote tolerance and have protective effects during this period when complementary feeding is initiated⁽Reference Prescott, Smith and Tang³⁴⁾.

In summary, the evidence supporting the benefit of delaying the introduction of allergenic foods (cow's milk, fish, eggs and peanuts) beyond 6 months of age is contradictory⁽Reference Greer, Sicherer and Burks^13, Reference Grimshaw, Allen and Edwards³⁰⁾. This issue must be explored further with carefully controlled interventional trials⁽Reference Greer, Sicherer and Burks^13, Reference Grimshaw, Allen and Edwards^30, Reference McLean and Sheikh³²⁾. Also, at present, there are insufficient data to document a protective effect of any dietary intervention beyond 4–6 months of age for the development of atopic disease⁽Reference Greer, Sicherer and Burks¹³⁾. In 2008, based on the available data, the Nutritional Committee of the American Academy of Pediatrics updated the recommendations, and now it states that although solid foods should not be introduced before 4–6 months of age, there is currently no convincing evidence that delaying their introduction beyond this period has a significant protective effect on the development of atopic disease regardless of whether infants are fed cow's milk protein formula or human milk, and this includes delaying the introduction of foods that are considered to be highly allergenic, such as fish, eggs and foods containing peanut protein⁽Reference Cochrane, Beyer and Clausen^1, Reference Greer, Sicherer and Burks¹³⁾. The Committee on Nutrition of the European Society for Paediatric Gastroenterology and Nutrition (ESPGHAN) recently concluded that complementary foods may be introduced safely between 4 and 6 months, and 6 months of exclusive breast-feeding may not always provide sufficient nutrition for optimal growth and development⁽Reference Agostoni, Decsi and Fewtrell⁴¹⁾.

The role of probiotics and prebiotics

There is accumulating evidence that early colonisation of the intestinal tract by an appropriate intestinal microbiota is important for the healthy maturation of the immune system, including appropriate programming of oral tolerance to dietary antigens⁽Reference Cochrane, Beyer and Clausen¹⁾. As it is generally agreed that the intestinal microbiota plays an important physiological role in the postnatal development of the immune system, many attempts have been made to influence the intestinal microbiota and thereby the occurrence of atopic manifestations⁽Reference Gruber, van Stuijvenberg and Mosca⁴²⁾. Therefore, as a potential means of preventing allergies, dietary interventions to modulate the intestinal microbiota of infants using probiotics and prebiotics have been explored⁽Reference Cochrane, Beyer and Clausen¹⁾.

Probiotics are defined as microbial cell preparations or components of microbial cells that have a beneficial effect on the health and well being of the host, while prebiotics are food ingredients that stimulate selectively the growth and activity of bifidobacteria and lactobacilli in the gut and thereby benefit health⁽Reference Szajewska and Mrukowicz^43, Reference Cummings and Macfarlane⁴⁴⁾. In human subjects, a number of randomised controlled trials have shown a preventative effect of probiotic or prebiotic feeding on the development and severity of atopic dermatitis in infants. However, other recent clinical studies of probiotic supplementation to infants at risk have failed to demonstrate clear benefits, with one study even reporting that probiotic intervention increased sensitisation to allergen. Explanations for varied results among studies include differences in types and doses of the used bacterial strains, host factors (i.e. genetic predisposition to allergic disease, polymorphism in microbial recognition pathways etc.) and other environmental factors (diet, treatment with antibiotics and general microbial burden). Despite uncertainties regarding their efficacy, infant formulae are increasingly supplemented with probiotics, prebiotics or synbiotics⁽Reference Cochrane, Beyer and Clausen¹⁾.

Some trials show favourable results for probiotics with regard to atopic dermatitis and it was reported that prenatal and postnatal probiotic supplementation is an effective approach in preventing the development of eczema during 1 year of life in infants at high risk of allergy⁽Reference Grimshaw, Allen and Edwards^30, Reference Kim, Kwon and Ahn⁴⁵⁾. A pioneering placebo-controlled study in high-risk families with perinatal supplementation of the mothers' and infants' diet with Lactobacillus GG demonstrated a reduced prevalence of early atopic dermatitis in children⁽Reference Kalliomaki, Salminen and Arvilommi⁴⁶⁾. On the contrary, some controlled trials in infants at high risk of atopy failed to demonstrate a preventive effect of probiotics on the emergence of atopic dermatitis⁽Reference Gruber, van Stuijvenberg and Mosca⁴²⁾. Furthermore, probiotics have no proven preventive effect on the development of asthma⁽Reference Vael and Desager⁴⁷⁾.

Considering the above, the Committee of Nutrition of the ESPGHAN considers that there is still too much uncertainty to draw reliable conclusions from the available data and does not recommend the routine use of probiotic-supplemented formula in infants⁽Reference Braegger, Chmielewska and Decsi⁴⁸⁾.

Concerning prebiotics, previous data in infants at higher risk of development of atopy demonstrate that a specific mixture of prebiotic oligosaccharides has a protective effect against allergic manifestations⁽Reference Arslanoglu, Moro and Schmitt⁴⁹⁾. Moreover, Gruber et al. ⁽Reference Gruber, van Stuijvenberg and Mosca⁴²⁾ showed that formula supplementation with a specific mixture of acidic and neutral oligosaccharides is effective as primary prevention of atopic dermatitis also in low atopy risk infants. The authors speculated that the effect persists beyond the first birthday and may even result in a reduced incidence of respiratory allergy later in life⁽Reference Gruber, van Stuijvenberg and Mosca⁴²⁾. However, the ESPGHAN Committee of Nutrition considers that there is too much uncertainty to draw reliable conclusions from the available data. Considering that, the Committee does not recommend the routine use of formula supplemented with prebiotics in infants⁽Reference Braegger, Chmielewska and Decsi⁴⁸⁾.

Although probiotics and prebiotics are theoretically promising candidates for prevention of atopic diseases, at this stage, the data that confirm their immunologic or therapeutic effects are still lacking and there is currently not enough evidence to support the use of probiotics and/or prebiotics for the prevention of allergic disease in clinical practice⁽Reference Cochrane, Beyer and Clausen^1, Reference Grimshaw, Allen and Edwards^30, Reference Vael and Desager^47, Reference Braegger, Chmielewska and Decsi⁴⁸⁾. Further clinical results are required before definite recommendations on the use and effectiveness of prebiotics and probiotics in allergy prevention can be made⁽Reference Cochrane, Beyer and Clausen¹⁾.