Miles, Lisa M. MMedSci, RNutr
The Scientific Advisory Committee on Nutrition (SACN) and the Committee on Toxicity (COT) are both advisory committees of independent experts that provide advice to the Department of Health and Food Standards Agency in the United Kingdom. In March 2011, the committees jointly completed an in-depth review examining the most appropriate time to introduce gluten into an infant’s diet. They considered the available evidence to see whether the time that gluten is introduced into an infant’s diet affects the chance of him/her developing celiac disease and type 1 diabetes mellitus (T1DM). The SACN/COT statement was published in response to a scientific opinion given by the European Food Safety Authority (EFSA, the keystone of European Union [EU] risk assessment regarding food and feed safety), which concluded that the introduction of gluten into an infant’s diet before 6 months of age while still breast-feeding might reduce the risk of developing celiac disease and T1DM. The recommendations by EFSA are inconsistent with current infant feeding advice from UK health departments. This article aims to provide a critique of the evidence base used to develop recent statements from SACN/COT and EFSA and offer a commentary on how different interpretations of the evidence have been reached.
INTRODUCING SOLID FOODS
UK health departments currently recommend exclusive breast-feeding for around the first 6 months of an infant’s life. It is recommended that solid foods are introduced at about 6 months of age and that breast-feeding continues beyond this time, along with appropriate types and amounts of solid foods.1 Similarly, the American Academy of Pediatrics recommends that complementary foods are introduced at around 6 months of age.2 In the United Kingdom, parents who choose to introduce solids before 6 months of age are advised to avoid giving foods containing gluten (as well as other commonly allergenic foods and foods thought to increase the risk of food poisoning) until the infant is 6 months of age.3
These recommendations have become a subject of debate, as a result of conflicting conclusions and advice from other organizations, including conclusions published by EFSA. A scientific opinion on the appropriate age for the introduction of complementary food into infant diets in the EU4 was produced by the EFSA Panel on Dietetic Products, Nutrition and Allergies and published in December 2009. The panel concluded that introducing complementary food into the diets of healthy term infants between the age of 4 and 6 months is safe and does not pose a risk for adverse health effects. This is inconsistent with the current UK recommendations to breast-feed exclusively for around the first 6 months of life and to avoid introduction of foods containing gluten before 6 months of age. It is also inconsistent with US and World Health Organization advice to breast-feed exclusively for 6 months.2,5
European Food Safety Authority’s conclusions on the appropriate age for introducing solids into infants’ diets are heavily influenced by its conclusions on the relationship between timing of introducing gluten and risk of the autoimmune conditions celiac disease and T1DM.
Celiac disease is characterized by a heightened immunological response to gluten in the diet of genetically predisposed individuals. Studies indicate that the prevalence of celiac disease in the United Kingdom lies around 1%.6,7 The prevalence of celiac disease in the United States is estimated to be 0.75% in those not at risk and between 2% and 5% in those at risk (eg, with a family history),8 and the prevalence of undiagnosed celiac disease is increasing.9 Celiac disease has a genetic association with certain types of type II human leukocyte antigens (HLAs); approximately 95% of people with celiac disease express the HLA-DQ2 haplotype, and the remainder express HLA-DQ8.10 The presence of one or other of these alleles is necessary for the development of celiac disease. However, the presence of either 1 or both of these alleles is widespread in the general population (estimates range from 25% to 40%11,12), yet only around 4% of these genetically susceptible individuals will develop celiac disease.13,14 The fact that this applies despite the widespread consumption of gluten-containing foods suggests that additional factors such as infant feeding, in particular, the timing of introduction of gluten into the infant diet, and/or infections in infancy may play a role in determining who develops the disease.15
TYPE 1 DIABETES MELLITUS
Type 1 diabetes mellitus is another autoimmune condition, and it is associated with celiac disease. It is thought that both genetic and environmental factors play a role in its development.16 There is a well-known predisposition to the disease associated with the HLA-DR3 and HLA-DR4 alleles, but also additional susceptibility associated with the HLA-DQ alleles, which also increase the risk of celiac disease.17 Individuals with T1DM and their first-degree relatives have an increased risk of celiac disease.11,18
European Food Safety Authority’s scientific opinion on the appropriate age for the introduction of complementary food into infant diets included conclusions about the appropriate age for introducing gluten into infants’ diets. The panel concluded that the early (<4 months) introduction of gluten might increase the risk of celiac disease and T1DM, whereas the introduction of gluten between 4 and 6 months while still breast-feeding might decrease the risk of celiac disease and T1DM. Overall, they supported the timing of the introduction of gluten not later than 6 months of age (preferably while still breast-feeding).4
Overall, SACN and COT considered the available evidence insufficient to support specific recommendations about the most appropriate time to introduce gluten into the infant diet. They did, however, indicate that introducing gluten into infants’ diets before 3 months of age might be linked to an increased risk of celiac disease, but they did not support EFSA’s conclusion to introduce gluten into the infant diet before 6 months. However, the conclusion on the role of breast-feeding was similar to that of EFSA; SACN/COT indicated there might be an increased chance of infants developing celiac disease if they are not being breast-fed when gluten is introduced into the diet.19
WHY DIFFERENT CONCLUSIONS?
At first glance, the inconsistency between the conclusions of EFSA and SACN/COT in relation to timing of introduction of gluten may seem unhelpful. However, a closer look at the scientific evidence reviewed by both committees highlights a number of study limitations, which can explain how different interpretations of the evidence have been reached.
First, the amount and quality of evidence were limited: few studies were identified that relate the timing of introduction of gluten into the infant diet to subsequent risk of celiac disease or T1DM. Those available were observational studies, and the lack of published intervention studies on gluten introduction and subsequent risk of celiac disease and T1DM is a major limitation of the evidence base.
It is difficult to investigate the epidemiology of celiac disease prospectively because it is a condition that is often underdiagnosed owing to the disparate nature of its signs and symptoms, which may be attributed to other conditions11; therefore, diagnoses can be relatively rare. Research studies therefore often identify at-risk individuals through HLA genotyping. Most of the studies identified by EFSA and SACN/COT were in high-risk populations identified in this way. Consequently, the quantitative risk estimates from these studies do not apply to the general population, although they do give a sense of the existence and direction of relationships.
The studies available were also limited in terms of their measurement of the timing of introduction of gluten into infant diets. The thresholds applied to measure exposure to gluten around 3 and 6 months of age applied in these studies were often unclear because the questionnaires were not always administered at exact time points. Ideally, studies examining the timing of introduction of gluten into the infant diet should have used a resolution of timing in weeks rather than months in order to prevent confusion. Different procedures for diagnosing celiac disease have also presented a challenge to the interpretation of studies investigating the timing of introduction of gluten into infants’ diets and celiac disease risk. Antibodies to the highly specific enzyme tissue transglutaminase have been identified as the major autoantibody involved in the celiac disease process20 and are often used as a diagnostic tool, along with endomysial antibodies.21 Alternatively, or in addition, gut tissue biopsies can be used. The predictive value of autoantibodies is less than 100%, so biopsy-confirmed celiac disease cases are thought to be more definitive, although more invasive. The studies reviewed by EFSA and SACN/COT used a range of outcome measures to identify cases of celiac disease, but many relied primarily on autoantibodies.
BREAST-FEEDING WHEN GLUTEN IS INTRODUCED INTO INFANTS’ DIETS
The conclusion, arrived at by both EFSA and SACN/COT, regarding the importance of breast-feeding when gluten is introduced into infants’ diets was based on a systematic review of 6 observational studies.22 A meta-analysis of 4 case-control studies included in this systematic review found that infants who were breast-fed when gluten was introduced had a 52% lower risk of celiac disease compared with those who were not breast-fed at that time. A strength of this systematic review is that it included only studies based on biopsy-confirmed celiac disease cases. However, it was not clear from the primary study reports whether investigators had assessed partial or exclusive breast-feeding, and duration of breast-feeding had been measured by variable methods. In addition, the exact timing and amount of gluten consumed were not stated in the original study reports, so this did not allow any conclusions to be made about these factors and risk of celiac disease. The studies included were all of case-control design so they may have been subject to recall bias. Furthermore, factors associated with longer duration of breast-feeding may have confounded the findings.
TIMING OF INTRODUCTION OF GLUTEN INTO INFANTS’ DIET AND RISK OF CELIAC DISEASE
Two studies of time trends in celiac disease among Swedish children23,24 found significant changes in the incidence of celiac disease from 1985 and suggested that these were temporally related to changes in national recommendations about the timing of introduction of gluten into the infant diet. They reported a sharp rise in incidence following recommendations to postpone introduction of gluten from 4 to 6 months of age and then a sharp decline in incidence after recommendations reverted to advise introduction of gluten from 4 months of age alongside breast-feeding. However, the quantity and form of gluten consumed over the periods studied were not measured directly, and so there is some uncertainty about the extent to which gluten was given during the process of introducing solids in Sweden. A type of gruel comprising cows’ milk and cereals is sometimes given during early weaning in Sweden. Conclusions were instead based merely on changes in recommendations, and the possibility of confounding factors cannot be excluded.
The most informative studies on timing of gluten introduction and risk of celiac disease were those by Norris et al15 and Ziegler et al,25 and both of these focused on high-risk populations. Norris et al15 reported a positive association between introduction of gluten after 6 months of age and risk of celiac disease defined by positive biopsy, which was statistically significant, and this is likely to be the major finding on which EFSA based its conclusions supporting the introduction of gluten into infants’ diets before 6 months of age. However, the dietary assessment method in the study of Norris et al15 has some limitations, which led to some uncertainty about the precise timing of introduction of gluten into the infants’ diets, and the number of subjects who had biopsy-confirmed celiac disease and had introduced gluten after 6 months was relatively small. The study by Ziegler et al,25 which was based on a cohort established primarily to investigate T1DM, was weaker overall than the study carried out by Norris et al15 because it did not use biopsy-confirmed celiac disease cases, but it found no relationship between gluten first introduced into the diet at post–6 months of age and risk of celiac disease The inconsistency between the findings of these 2 studies relating to first exposure to gluten at post–6 months limits the conclusions that can be drawn.
TIMING OF INTRODUCTION OF GLUTEN INTO INFANTS’ DIET AND RISK OF T1DM
Studies by Norris et al26 and Ziegler et al25 investigated the risk of T1DM by using high-risk populations and autoantibodies (insulin autoantibodies, glutamic acid decarboxylase autoantibodies, and/or tyrosine phosphatase–related antigen 2 autoantibodies) as outcome measures for T1DM rather than firm demonstration of the disease. Neither found statistically significant relationships between timing of introduction of gluten and risk of T1DM. Another study by Wahlberg et al27 showed mixed findings, depending on which autoantibodies were used as outcome measures. Overall, evidence relating age at introduction of gluten to risk of T1DM is weaker than that for celiac disease and is inconsistent.
The SACN/COT’s conclusions on the timing of introduction of gluten into infants’ diets have not changed the current advice of UK health departments. They each continue to advise that all mothers exclusively feed their babies breast milk for around the first 6 months of life and that solid foods should be introduced when an infant is around 6 months old, alongside continued breast-feeding. If, after speaking to a health professional, parents choose to introduce solid foods before 6 months, they should avoid gluten-containing foods and other foods that commonly cause allergies.
Introducing gluten before 6 months is not recommended, and there are no known benefits of delaying the introduction of gluten in the diet for any longer than 6 months. Once an infant is established on solid foods, gluten-containing foods such as bread or wheat-based cereals should be eaten regularly as celiac disease can be diagnosed only once gluten is in the diet. It is important that foods that contain gluten and others that commonly cause allergies are introduced one at a time so that an infant’s adverse reaction can be spotted. A diagnosis can be made if symptoms do occur.
There is evidence that continued breast-feeding while introducing gluten into the diet can protect against developing celiac disease so it is important to continue breast-feeding for at least 6 months. However, although breast-feeding initiation rates have increased in the United Kingdom, very few mothers continue to breast-feed beyond 6 months.28 Unfortunately, the available evidence does not show whether the relationship between breast-feeding while gluten is introduced into the diet and risk of celiac disease varies according to the age at which gluten is introduced.
Since EFSA and SACN/COT identified studies on timing of introduction of gluten and celiac disease/T1DM risk, a further relevant study29 has been published. This is a Swedish population-based study that measured timing of introduction of gluten, breast-feeding duration, and episodes of infectious disease and their relationship with biopsy-confirmed celiac disease. The age at which gluten was introduced was not associated with risk of future development of celiac disease. These findings are therefore consistent with the UK health departments’ advice on introducing gluten.
In terms of further research on timing of introduction of gluten and risk of celiac disease, a major research initiative is ongoing in the form of the EU-funded research project PreventCD. PreventCD is a European multicenter study, which studies the influence of infant nutrition and that of genetic, immunologic, and environmental factors on the risk of developing celiac disease. The project is investigating the hypothesis that it is possible to induce tolerance to gluten by introducing small quantities of gluten to infants, preferably while they are still being breast-fed, and that this might also reduce the risk for related autoimmune disorders. PreventCD encompasses 2 study designs and 2 study populations: (i) a European multicenter study: a prospective, double-blind, randomized dietary-intervention study among infants from families with high risk of celiac disease, and (ii) a Swedish population-based celiac disease screening study among 12-year-olds from the general population, divided into 2 birth cohorts that differ with respect to infant-feeding practices.30 The results of this study are expected at the end of 2011.
Regarding the wider debate about the appropriate age for introducing solids into the infant diet, the UK health departments have asked SACN/COT to conduct a review of infant and young child feeding policies. This will be informed by the SACN/COT statement on timing of introduction of gluten into infant diets, but the relationship between timing of introduction of gluten and risk of celiac disease and T1DM is clearly only 1 part of the jigsaw. The relationship between timing of introduction of solid foods and other health outcomes such as childhood infections and allergies, as well as normal growth and development, will also need to be considered. The SACN started work on this wider review in early 2011.
2. Gartner LM, Morton J, Lawrence RA, et al. Breastfeeding and the use of human milk. Pediatrics. 2005; 115:(2): 496–506.
3. Department of Health. Weaning and the Weaning Diet. London: The Stationery Office; 1994; .
4. European Food Safety Authority (EFSA) Panel on Dietetic Products, Nutrition and Allergies (NDA). Scientific opinion on the appropriate age for introduction of complementary feeding of infants. EFSA J. 2009;7(12):1423.
5. World Health Organization (WHO). The Optimal Duration of Exclusive Breastfeeding: Report on an Expert Consultation. Geneva, Switzerland: WHO; 2001.
6. Bingley PJ, Williams AJK, Norcross A, et al. Undiagnosed coeliac disease at age seven: population-based prospective birth cohort study. BMJ. 2004; 328: 322–323.
7. West J, Logan RF, Hill PG, et al. Seroprevalence, correlates, and characteristics of undetected coeliac disease in England. Gut. 2003; 52:(7): 960–965.
8. Fasano A, Berti I, Gerarduzzi T, et al. Prevalence of celiac disease in at-risk and not-at-risk groups in the United States: a large multicenter study. Arch Intern Med. 2003; 163: 286–292.
9. Rubio-Tapia A, Kyle RA, Kaplan EL, et al. Increased prevalence and mortality in undiagnosed celiac disease. Gastroenterology. 2009; 137: 88–93.
10. Myleus A, Ivarsson A, Webb C, et al. Celiac disease revealed in 3% of Swedish 12year old boys born during an epidemic. J Pediatr Gastroenterol Nutr. 2009; 49: 170–176.
11. National Institute for Health and Clinical Excellence (NICE). Coeliac Disease: Recognition And Assessment of Coeliac Disease. NICE Clinical Guideline 86. Manchester, UK: National Institute for Health and Clinical Excellence; 2009.
12. Wolters VM, Wijmenga C. Genetic background of celiac disease and its clinical implications. Am J Gastroenterol. 2008; 103: 190–195.
13. Schuppan D, Junker Y, Barisani D. Celiac disease: from pathogenesis to novel therapies. Gastroenterology. 2009; 137: 1912–1933.
14. Silano M, Agostoni C, Guandalini S, et al. Effect of the timing of gluten introduction on the development of celiac disease. World J Gastroenterol. 2010; 16:(16): 1939–1942.
15. 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 disease. JAMA. 2005; 293: 2343–2351.
16. Meloni T, Marinaro AM, Mannazzu M, et al. IDDM and early infant feeding. Sardinian case-control study. Diabetes Care. 1997; 20:(3): 340–342.
17. Atkinson MA, Eisenbarth GS. Type 1 diabetes: new perspectives on disease pathogenesis and treatment. Lancet. 2001; 358: 221–229.
18. Collin P, Kaukinen K, Valimaki M, et al. Endocrinological disorders and celiac disease. Endocr Rev. 2002; 23: 464–483.
20. Dieterich W, Ehnis T, Bauer M, et al. Identification of tissue transglutaminase as the autoantigen of celiac disease. Nat Med. 1997; 3: 797–801.
21. Rostrom A, Dube C, Cranney A, et al. The diagnostic accuracy of serologic tests for celiac disease: a systematic review. Gastroenterology. 2005; 128: S38–S46.
22. Akobeng AK, Ramanan AV, Buchan I, et al. Effect of breast feeding on the risk of coeliac disease: a systematic review and meta-analysis of observational studies. Arch Dis Child. 2006; 91: 39–43.
23. Carlsson A, Agardh D, Borulf S, et al. Prevalence of celiac disease: before and after a national change in feeding recommendations. Scand J Gastroenterol. 2006; 41: 553–558.
24. Ivarsson A, Persson LA, Nystrom L, et al. Epidemic of celiac disease in Swedish children. Acta Paediatr. 2000; 89: 165–171.
25. Ziegler A-G, Schmidt S, Huber D, et al. Early infant feeding and risk of developing type 1 diabetes-associated autoantibodies. JAMA. 2003; 290: 1721–1728.
26. Norris JM, Barriga K, Klingensmith G, et al. Timing of initial cereal exposure in infancy and risk of islet autoimmunity. JAMA. 2003; 290: 1713–1720.
27. Wahlberg J, Vaarala O, Ludvigsson J. ABIS-Study Group. Dietary risk factors for the emergence of type 1 diabetes-related autoantibodies in 2 1/2 year-old Swedish children. Br J Nutr. 2006; 95:(3): 603–608.
29. Welander A, Tjernberg AR, Montgomery SM, et al. Infectious disease and risk of later celiac disease in childhood. Pediatrics. 2010; 12: e530–e536.
30. Hogen Escha CE, Rosen A, Auricchio R, et al. The PreventCD Study design: towards new strategies for the prevention of coeliac disease. Eur J Gastroenterol Hepatol. 2010; 22: 1424–1430.