Вы находитесь на странице: 1из 6

Pediatr Allergy Immunol 2008: 19: 375–380 Ó 2008 The Authors

DOI: 10.1111/j.1399-3038.2008.00718.x Journal compilation Ó 2008 Blackwell Munksgaard

PEDIATRIC ALLERGY AND


IMMUNOLOGY
Discussion Paper

The importance of early complementary


feeding in the development of oral tolerance:
Concerns and controversies
Prescott SL, Smith P, Tang M, Palmer DJ, Sinn J, Huntley SJ, Susan L. Prescott1, Peter Smith2, Mimi
Cormack B, Heine RG, Gibson RA, Makrides M. The importance of Tang3, Debra J. Palmer4, John Sinn5,
early complementary feeding in the development of oral tolerance: Sophie J. Huntley6, Barbara
Concerns and controversies. Cormack7, Ralf G. Heine8, Robert
Pediatr Allergy Immunol 2008: 19: 375–380. A. Gibson9 and Maria Makrides4
Ó 2008 The Authors 1
School of Paediatrics and Child Health Research,
Journal compilation Ó 2008 Blackwell Munksgaard University of Western Australia, Perth, WA, 2Bond
University Medical School, Gold Coast, Qld,
3
Department of Allergy and Immunology, Royal
Rising rates of food allergies in early childhood reflect increasing failure
ChildrenÕs Hospital, Parkville, Vic., 4WomenÕs and
of early immune tolerance mechanisms. There is mounting concern that ChildrenÕs Health Research Institute, Children, Youth
the current recommended practice of delaying complementary foods and WomenÕs Health Service, Adelaide, SA,
until 6 months of age may increase, rather than decrease, the risk of 5
Department of Neonatology, Royal North Shore
immune disorders. Tolerance to food allergens appears to be driven by Hospital, St Leonards, NSW, 6Department of
regular, early exposure to these proteins during a Ôcritical early windowÕ Nutrition and Dietetics, Flinders University, Adelaide,
of development. Although the timing of this window is not clear in SA, Australia, 7Nutrition Services, Auckland City
Hospital, Auckland, New Zealand, 8Department of
humans, current evidence suggests that this is most likely to be between Gastroenterology and Clinical Nutrition, Royal
4 and 6 months of life and that delayed exposure beyond this period ChildrenÕs Hospital, Parkville, Vic., 9School of
may increase the risk of food allergy, coeliac disease and islet cell Agriculture, Food and Wine, The University of
autoimmunity. There is also evidence that other factors such as Adelaide, Adelaide, SA, Australia
favourable colonization and continued breastfeeding promote tolerance
and have protective effects during this period when complementary Key words: infant feeding; weaning practices; allergy
feeding is initiated. This discussion paper explores the basis for concern prevention; growth and development
over the current recommendation to delay complementary foods as an
Prof. Susan L. Prescott, School of Paediatrics and
approach to preventing allergic disease. It will also examine the growing Child Health Research, University of Western
case for introducing complementary foods from around 4 months of Australia, Princess Margaret Hospital, PO Box D184,
age and maintaining breastfeeding during this early feeding period, for Perth, WA 6001, Australia
at least 6 months if possible. Tel.: +61 8 9340 8171
Fax: +61 8 9388 2097
E-mail: sprescott@meddent.uwa.edu.au

Accepted 9 January 2008

There has been a clear and worrying increase in a study (2), and had major implications for disease
diverse range of allergic and autoimmune diseases, burden in developing and semi-industrialized
which are all associated with an underlying failure countries. At that time, there was also a paucity
of immune tolerance (to allergens and self-anti- of evidence regarding the other risks and benefits
gens). This emphasizes the need to understand the of introducing complementary foods from 4 vs.
developmental origins of these conditions and, in 6 months of age (1). These recommendations
particular, the role of early feeding practices in were aimed at reducing morbidity in developing
both pathogenesis and prevention. countries, but may not be appropriate in the
In 2001, the World Health Organisation growing world population experiencing progres-
(WHO) revised its recommendation for exclusive sive industrialization and escalating risk of
breastfeeding from 4 to 6 months (1). This was immune dysregulation.
based (at least in part) on reduced gastrointesti- Similar recommendations have been adopted
nal infectious disease noted in a Belarussian for the prevention of allergic disease in the USA,
375
Prescott et al.

the UK, Australia and other industralized early when the gut colonization and local
countries, based on the theoretical concern for immune networks are less established may
increased gut permeability and immaturity of increase the risk of allergic or autoimmune
mucosal immunity in infants. However, there is disease (possibly through increased gut perme-
now mounting concern and some new evidence ability) (9). Some studies also suggest that
that this recommendation for delayed introduc- continued breastfeeding during introduction of
tion of complementary foods may have detri- complementary foods is important for promoting
mental consequences. In Western countries, tolerance (10). The timing of this Ôcritical win-
where these recommendations have been adopted dowÕ for oral allergen exposure is not clear in
into practice, rates of food allergy have escalated humans, but current evidence suggests that this
rather than declined in the last 10 yr (3–5). The may be between 4 and 6 months of life, as
purpose of this discussion paper is to highlight discussed below. This raises further concerns
these concerns and controversies (in term in- over recommendations for delaying introduc-
fants), together with the rationale for revising tions of complementary foods until after
guidelines to more accurately reflect current 6 months of age (1, 11), especially in industrial-
evidence, at least until more definitive studies ized countries where the incidence of allergy is
can be performed. high. It may be further argued that infants with a
genetic predisposition to allergic disease (failure
of oral tolerance) may require regular, larger
Implication for immune tolerance
ÔdosesÕ of food allergen to promote tolerance
Development of immune tolerance is a critical induction (as with immunotherapy). If this is the
process in early life. The rising rates of allergic case, allergen avoidance is likely to be detrimen-
and autoimmune diseases highlight the suscepti- tal in these children. However, at this stage this is
bility of these tolerance pathways to environ- not known.
mental changes. Although the mechanisms are
not clear, many of these conditions (including
The basis and evolution of current recommendations for
food allergies, coeliac disease and type 1 diabe-
infant feeding in allergy prevention
tes) manifest early in life, indicating that immune
dysregulation is a very early event. Animal In the last 25 yr, observational and primary
models suggest that tolerance is an antigen allergy prevention studies in allergy with respect
(allergen)-driven process and that exposure to to the early introduction of complementary foods
these proteins during a Ôcritical early windowÕ of have been inconclusive, and at best, several
development may be essential to this process studies demonstrate a transient increase in IgE,
(Fig. 1). This also appears to coincide with the particularly to milk if foods were introduced
establishment of healthy gut colonization, which before 3–4 months of age. Here, we explore the
has been shown to be essential in promoting historical basis for current recommendations:
tolerance to both allergens and self antigens (6). In the 1980s and 1990s, several studies
Delays in either colonization (6) or antigen/ reported an association between early introduc-
allergen exposure (7, 8) can lead to failure of oral tion of solids (<3–4 months) and eczema
tolerance. Conversely, allergen exposure too (12)14). This association did not persist beyond
12 months of age in two of these three cohorts
(18, 19). A number of subsequent prospective
studies have failed to demonstrate an association
risk window risk resolution between early introduction of complementary
Tolerance foods and either eczema or food allergy, and a
induction recent systematic review concluded that there
was Ôno strong evidence to support the associa-
Birth 3–4? 6–7? >12 months
tion between early solid feeding and the devel-
? ?
opment of persistent asthma, persistent food
Factors that influence the capacity for tolerance: allergy, allergic rhinitis, or animal dander allergyÕ
• optimal colonisation (17).
• genetic pre-disposition Despite the paucity of clear evidence, conser-
• allergen properties (dose, interval, timing, preparation)
• gut permeability/maturity/pH vative avoidance recommendations remain in
• continued breast feeding? place in many countries. These appear to be
• other immunomodulatory factors (fatty acids? stress? antioxidants?)
based on an early US study by Zeiger and Heller
Fig. 1. Possible Ôwindow of toleranceÕ for introduction of (18), which used combined dietary avoidance
complementary foods. strategies in infants at risk of allergic disease.
376
Early complementary feeding in the development of oral tolerance

Strategies included maternal allergen avoidance weak evidence and paradoxical effects in some
(in pregnancy and lactation), extensively hydro- studies (11). We also raised concerns about the
lysed formula if complementary feeding was lack of good evidence to support many food
required in the first year, and staged delay in allergen avoidance practices (some of which
introduction of solid foods: non-legume vegeta- avoid ÔallergenicÕ foods such as egg and peanuts
bles, rice cereal, meats and non-citrus fruits for between 2 and 4 yr of age). As indicated
between 6 and 12 months; cowÕs milk, wheat, above, any benefit of early food allergen avoid-
soy, corn and citrus fruits between 12 and ance appears to be largely in the first
18 months; eggs at 24 months; and peanut and 3–4 months, with very little clear or consistent
fish at 36 months. The main findings of these evidence that avoidance beyond this period
studies were as follows. reduces allergy risk. Therefore, in the context
of allergy prevention, we elected to maintain
(i) Reduction in food-associated atopic derma- the previous recommended duration of exclu-
titis, urticaria and/or gastrointestinal disease sive breastfeeding at Ô4–6 monthsÕ [consistent
by 12 months (5.1% vs. 16.4%; p = 0.007). with the ESPGHAN recommendations of
This was only significant with a grouped 5 months (21)], despite the WHO recommen-
symptom analysis. dations (1) for exclusive breastfeeding of all
(ii) In the prophylaxis (treated) group, there was infants for Ôat least 6 monthsÕ. Moreover, based
a lower rate of sensitization to foods at on the current evidence (presented below), there
24 months (16.5% vs. 29.4%; p = 0.019), may now be a case for revising this further to
due mainly to fewer positive milk skin tests Ô4 monthsÕ. Although the specific role in allergy
(1% vs. 12.4%; p = 0.001). prevention remains unclear, there is some
(iii) Serum IgE levels in the prophylaxis group evidence that continued breastfeeding while
were marginally lower only at 4 months. new foods are introduced is beneficial (10).
When breastfeeding is not possible, the use of
These findings could also be attributed to the hydrolysed formulas appears to confer some
use of an extensively hydrolysed formula. protective effect compared with normal cow
Despite this, the entire infant weaning structure milk-based formulas and these are still recom-
of this study has been incorporated into the mended, although it is generally recognized that
recommendations of the Committee on Nutri- more studies are needed and the protective
tion of the American Academy of Pediatrics in effects are not great. Thus, until more data are
2000 (19) and recently reinforced by a position available, it is difficult to justify the restriction
paper from the American College of Asthma of complementary foods until 6 months of age,
Allergy and Immunology (20). These avoidance particularly in industrialized countries.
recommendations have been presented at many
international meetings and propagated by pro-
Rationale for reconsidering the role of early food
fessional societies and government agencies to
allergen exposure
the wider community. The European Society
for Paediatric Allergology and Clinical Immu- The normal development of oral tolerance is an
nology Committee on Hypoallergenic Formulas antigen-driven process and may logically depend
and the European Society for Paediatric on regular exposure to foods and other antigens
Gastroenterology, Hepatology and Nutrition during a critical early window. These processes
(ESPGHAN) have been more circumspect, are also likely to depend on other conducive
suggesting the introduction of solids after exposures [such as favourable gut colonization
5 months (21). (6), breast milk (10) and/or other nutritional
In summary, the evidence for restricting immunomodulatory factors]. It is now increas-
infant diet (complementary foods) for up to ingly evident that allergen exposure is not the
6 months is weak, comprising inconsistent find- primary cause of the allergy epidemic, and that
ings of increased milk sensitization and eczema allergen avoidance may be unsuccessful, or even
with complementary foods given in the first detrimental in allergy prevention. Indeed,
3–4 months. although rising rates of immune disease are likely
to reflect a combination of many environmental
changes that compromise tolerance, the intro-
Summary of current guidelines for allergy prevention
duction of complementary foods must be con-
In 2005, we (Prescott and Tang) revised the sidered in this complex modern context.
Australian allergy prevention guidelines, remov- Understanding other early host–environment
ing the inhalant avoidance strategies in light of interactions is essential to this.
377
Prescott et al.

the USA (3), the UK (4) and Australia (5), rates


Evidence to support a role of earlier exposure to food
of food allergy have escalated rather than
proteins in immune tolerance
declined.
A number of recent studies suggest that exposure
to specific foods in the 4–6 months age range
New studies aimed at using early food allergen
may reduce the risk of food allergies (7) and
exposure to prevent allergy
autoimmunity (8, 22) compared with children
first exposed either before or after this ÔwindowÕ. Challenging many long-held concepts, there are
Specifically, children initially exposed to cere- now studies (in progress and in design) that will
als between ages 0 and 3 months [hazard ratio examine the hypothesis that earlier introduction
(HR), 4.32; 95% confidence interval (CI), 2.0– (rather than avoidance) and regular exposure to
9.35] and those who were exposed after 6 months ÔallergenicÕ foods (such as peanuts and egg) may
(HR, 5.36; 95% CI, 2.08–13.8) had increased risk reduce the risk of specific allergies to these
of islet cell autoantibodies than those who were foods.
exposed between 4 and 6 months (22). Similarly, There is currently a UK-based randomized
delayed exposure to gluten (after 6 months) was controlled trial investigating the regular con-
associated with increased risk of coeliac disease sumption vs. avoidance of peanut protein during
autoimmunity (HR, 1.87; 95% CI, 0.97–3.60) infancy. This trial involves infants (4–11 months
and biopsy-diagnosed coeliac disease (HR, 3.98; of age) with egg allergy, severe eczema or both.
95% CI, 1.18–13.46; p = 0.04) compared with The intervention group is fed at least 6 g peanut
those exposed at 4–6 months (8). Exposure prior protein weekly, distributed over at least three
to 4 months was also associated with increased meals each week. The primary outcome of this
risk of coeliac disease autoimmunity (HR, 5.17; study assesses the effects of this intervention on
95% CI, 1.44–18.57) and biopsy-diagnosed coe- the proportion of children with peanut allergy at
liac disease (HR, 22.97; 95% CI, 4.55–115.93; 5 yr of age. There is also a proposed Australian
p = 0.001). study that will examine the effects of early egg
Although early allergic disease outcomes are introduction in high-risk infants.
reduced in children who have not been exposed
to complementary foods prior to 4 months (12–
Other implications: growth, development and nutritional
14, 23), there is little evidence that avoidance
status
beyond 4 months is beneficial (24). Moreover,
avoidance beyond 6 months has been associated The age when constituent breast milk consump-
with increased risk of allergic disease (food tion will no longer meet infant nutrition require-
allergy, eczema, asthma) (7, 24). One recent ments is ill-defined. Current nutrient reference
Australian study (in which reverse causation was values (NRV) and energy requirements for the
excluded) showed that even delaying comple- first 6 months are largely based on average
mentary foods until after 3 months was associ- intakes of healthy breastfed infants. Despite the
ated with an increased risk of atopy at 5 yr of age rapid growth and development of infants during
(25). this period, many references ranges do not
Intervention studies that have used infant increase until 7 months when solid foods are
dietary avoidance strategies for allergy preven- currently recommended. Thus, NRV tend to be
tion are difficult to interpret because most have based more in feeding practice rather than
used these in combination with other strategies physiological needs. A systematic review by
such as hydrolysed formulas (26, 27), maternal Lanigan et al. (29) showed a paucity of random-
restrictions (28) and/or other environmental ized trials and little evidence to support a change
interventions (26, 28). Furthermore, dietary in recommended commencement of solids to
interventions also varied widely between studies 6 months.
(restricting ÔanyÕ complementary feeding for There are concerns that delayed introduction
between 3 and 6 months and restricting specific of complementary feeding could compromise
solids such as egg, dairy, wheat, nuts, fish and supply of nutrients essential for growth and
soy up from between 4–12 months). Although neuro-development. Exclusive breastfeeding for
some of these studies reported benefits on some 6 months has been associated with a higher risk
allergic outcomes, the findings are not consistent of anaemia (10%) compared with breastfed
and difficult to attribute specifically to infant infants receiving complementary feeding (2.3%)
dietary restrictions. Finally, in countries where (30). The WHO systematic review (31) also
early dietary restrictions (of complementary reported that infants exclusively breastfed to
foods) have been adopted into practice such as 6 months had a lower haemoglobin (p = 0.005)
378
Early complementary feeding in the development of oral tolerance

and ferritin level (p = 0.04) compared with United Kingdom children at school entry. J Allergy Clin
those infants fed mixed breastfeeding at Immunol 2007: 119: 1197–202.
5. Polous LM, Waters AM, Correll PK, Loblay RH,
4–6 months. Delayed complementary feeding Marks G. Trends in hospitalizations for anaphylaxis,
may in some populations increase the risk of angioedema, and urticaria in Australia, 1993-1994
deficiencies of iodine (32), zinc (33) and other to 2004-2005. J Allergy Clin Immunol 2007: 120: 878–
specific micronutrients (34). This raises particu- 84.
lar concerns that some infants may become 6. Sudo N, Sawamura S, Tanaka K, Aiba Y, Kubo C,
deficient in essential micronutrients if solid foods Koga Y. The requirement of intestinal bacterial flora
for the development of an IgE production system fully
are delayed. In addition to implications for susceptible to oral tolerance induction. J Immunol 1997:
growth, there are numerous other developmental 159: 1739–45.
issues surrounding the timing of dietary diver- 7. Poole JA, Barriga K, Leung DYM, et al. Timing of
sity, including the development of taste and food initial exposure to cereal grains and the risk of wheat
preferences, which also need to be addressed in allergy. Pediatrics 2006: 117: 2175–82.
this process. 8. Norris JM, Barriga K, Hoffenberg EJ, et al. Risk of
celiac disease autoimmunity and timing of gluten
introduction in the diet of infants at increased risk of
Conclusions disease. JAMA 2005: 293: 2343–51.
9. Halken S, Host A. Prevention of allergic disease.
New studies that address the role of early Exposure to food allergens and dietetic intervention.
exposure to allergenic foods (rather than avoid- Pediatr Allergy Immunol 1996: 7: 102–7.
10. Ivarsson A, Hernell O, Stenlund H, Persson LA.
ance) are greatly needed and awaited with utmost Breast-feeding protects against celiac disease. Am J Clin
interest. However, until these studies have been Nutr 2002: 75: 914–21.
completed, this issue is likely to remain unre- 11. Prescott SL, Tang ML. The Australasian Society of
solved. In the meantime, there is conflict between Clinical Immunology and Allergy position statement:
some allergy prevention guidelines that recom- summary of allergy prevention in children. Med J Aust
mend introductions of complementary foods 2005: 182: 464–7.
12. Fergusson DM, Horwood LJ, Shannon FT. Risk
from Ô4 to 6 monthsÕ [e.g. as currently recom- factors in childhood eczema. J Epidemiol Community
mend in Australia (11)] and the recommendation Health 1982: 36: 118–22.
for exclusive breastfeeding until Ô6 monthsÕ [by 13. Forsyth JS, Ogston SA, Clark A, Florey CD,
the WHO and many other international govern- Howie PW. Relation between early introduction of
ment and professional bodies]. The evidence for solid food to infants and their weight and illnesses
delayed introduction of complementary foods to during the first two years of life. BMJ 1993: 306: 1572–6.
14. Kajosaari M, Saarinen UM. Prophylaxis of atopic
6 months is very limited and difficult to justify in disease by six monthsÕ total solid food elimination.
the face of emerging evidence that this may be Evaluation of 135 exclusively breast-fed infants of
detrimental. Rather, there is a growing case for atopic families. Acta Paediatr Scand 1983: 72: 411–4.
further revising the recommended age for the 15. Fergusson DM, Horwood LJ, Shannon FT. Early
introduction of complementary foods to solid feeding and recurrent childhood eczema: a 10-year
Ô4 monthsÕ while breastfeeding is maintained for longitudinal study. Pediatrics 1990: 86: 541–6.
16. Kajosaari M. Atopy prophylaxis in high-risk infants.
at least 6 months where possible. Adv Exp Med Biol 1991: 310: 453–8.
17. Tarini BA, Carroll AE, Sox CM, Christakis DA.
References Systematic review of the relationship between early
introduction of solid foods to infants and the develop-
1. Department of Nutrition for Health and Development, ment of allergic disease. Arch Pediatr Adolesc Med
WHO. The optimal duration of exclusive breastfeeding: 2006: 160: 502–7.
report of an expert consultation. 2001. WHO/NHD/ 18. Zeiger R, Heller S. The development and prediction
01.09.2002. Available at: http://www.who.int/child- of atopy in high-risk children: follow-up at age seven
adolescent-health/New_Publications/NUTRITION/ years in a prospective randomized study of combined
WHO_CAH_01_24.pdf. maternal and infant food allergen avoidance. J Allergy
2. Kramer MS, Chalmers B, Hodnett ED, et al. Pro- Clin Immunol 1995: 96: 1179–90.
motion of breastfeeding intervention trial (PROBIT): 19. American Academy of Pediatrics, Committee on
a cluster-randomized trial in the Republic of Belarus. Nutrition. Hypoallergenic infant formulas. Pediatrics
Design, follow-up, and data validation. Adv Exp Med 2000: 106: 346–9.
Biol 2000: 478: 327–45. 20. Fiocchi A, AssaÕad A, Bahna S. Food allergy and the
3. Sicherer SH, Munoz-Furlong A, Sampson HA. introduction of solid foods to infants: a consensus
Prevalence of peanut and tree nut allergy in the United document. Ann Allergy Asthma Immunol 2006: 97:
States determined by means of a random digit dial 10–21.
telephone survey: a 5-year follow-up study. J Allergy 21. Host A, Koletzko B, Muraro A, et al. Joint State-
Clin Immunol 2003: 112: 1203–7. ment of the European Society for Paediatric Allergology
4. Hourihane JO, Aiken R, Briggs R, et al. The impact and Clinical Immunology (ESPACI) Committee on
of government advice to pregnant mothers regarding Hypoallergenic Formulas and the European Society for
peanut avoidance on the prevalence of peanut allergy in

379
Prescott et al.

Paediatric Gastroenterology, Hepatology and Nutrition STUDY-PROGRAMME. Part II: infant growth and
(ESPGHAN) Committee on nutrition used in infants health status to age 6 months. Eur J Nutr 2000: 39: 145–
for treatment and prevention of food allergy. Arch Dis 56.
Child 1999: 81: 80–4. 28. Arshad SH, Bateman B, Sadeghnejad A, Gant C,
22. Norris JM, Barriga K, Klingensmith G, et al. Tim- Matthews SM. Prevention of allergic disease during
ing of initial cereal exposure in infancy and risk of islet childhood by allergen avoidance: the Isle of Wight
autoimmunity. JAMA 2003: 290: 1713–20. prevention study. J Allergy Clin Immunol 2007: 119:
23. Zutavern A, Brockow I, Schaaf B, et al. Timing of 307–13.
solid food introduction in relation to atopic dermatitis 29. Lanigan J, Bishop J, Kimber A, Morgan J. Systematic
and atopic sensitization: results from a prospective birth review concerning the age of introduction of comple-
cohort study. Pediatrics 2006: 117: 401–11. mentary foods to the healthy full term infants. Eur J
24. Filipiak B, Zutavern A, Koletzko S, et al. Solid food Clin Nutr 2001: 55: 309–20.
introduction in relation to eczema: results from a four- 30. Chantry C, Howard C, Auinger P. Full breastfeeding
year prospective birth cohort study. J Pediatr 2007: 151: duration and risk for iron deficiency in US infants.
352–8. Breastfeed Med 2007: 2: 63–73.
25. Mihrshahi S, Ampon R, Webb K, et al. The association 31. Kramer MS, Kakuma R. The optimal duration of
between infant feeding practices and subsequent atopy exclusive breastfeeding a systematic review. Adv Exp
among children with a family history of asthma. Clin Med Biol 2004: 554: 63–77.
Exp Allergy 2007: 37: 671–9. 32. Skeaff SA. Are breast fed infants and toddlers in New
26. Chan-Yeung M, Ferguson A, Watson W, et al. The Zealand at risk of iodine deficiency. Nutrition 2005: 21:
Canadian Childhood Asthma Primary Prevention 325–31.
Study: outcomes at 7 years of age. J Allergy Clin 33. Walravens PA, Chakar A, Mokni R, Denise J,
Immunol 2005: 116: 49–55. Lemonnier D. Zinc supplements in breastfed infants.
27. Exl BM, Deland U, Secretin MC, Preysch U, Wall Lancet 1992: 340: 683–5.
M, Shmerling DH. Improved general health status in 34. Munns C. Prevention and treatment of infant and
an unselected infant population following an allergen- childhood Vit D deficiency in Aust and New Zealand:
reduced dietary intervention programme: the ZUFF- a consensus statement. Med J Aust 2006: 185: 268–72.

380

Вам также может понравиться