Portal de Periódicos da CAPES

Early-life antibiotic exposure and childhood food allergy: A systematic review

2019; Elsevier BV; Volume: 144; Issue: 5 Linguagem: Inglês

10.1016/j.jaci.2019.08.001

1097-6825

Stejara A. Netea, Nicole L. Messina, Nigel Curtis,

Allergic Rhinitis and Sensitization

Use of antibiotics in children has increased over the last few decades. Concurrent with this, there has also been a dramatic increase in the prevalence of allergic diseases, including food allergy. A large study in Australia reported that nearly one in five 12-month-old infants is sensitized to at least 1 food, and more than 10% have challenge-proved food allergy.1Osborne N.J. Koplin J.J. Martin P.E. Gurrin L.C. Lowe A.J. Matheson M.C. et al.Prevalence of challenge-proven IgE-mediated food allergy using population-based sampling and predetermined challenge criteria in infants.J Allergy Clin Immunol. 2011; 127: 668-676, e1-2Abstract Full Text Full Text PDF PubMed Scopus (755) Google Scholar Several studies suggest a relationship between antibiotic exposure and allergic disease, particularly asthma and eczema.2Obiakor C.V. Tun H.M. Bridgman S.L. Arrieta M.C. Kozyrskyj A.L. The association between early life antibiotic use and allergic disease in young children: recent insights and their implications.Expert Rev Clin Immunol. 2018; 14: 841-855Crossref Scopus (20) Google Scholar Antibiotics might promote the development of allergy by interfering with the microbiome-dependent normal maturation of the immune system through an imbalance in T-cell maturation favoring a TH2 cell response. In this systematic review we aimed to identify studies investigating the association between antibiotic exposure and food allergy and the factors affecting this relationship. A systematic search of Medline, Embase, and PubMed was done in February 2019 to identify studies that investigated the association between early-life antibiotic exposure and childhood food allergy (Fig 1). Articles were included if they evaluated an association between antibiotics and food allergy in children (0-18 years of age). Food allergy was defined as either (1) reported symptoms directly related to food or drink intake together with a positive IgE blood test or skin prick test result and/or positive oral food challenge result or (2) diagnosis of food allergy by a physician. Articles were excluded if they used alternative definitions for food allergy (eg, skin prick test results only or outcome based on parental questionnaire only; Fig 1). Of 1058 unique articles identified in the original systematic search (February 2019), 6 relevant articles met the inclusion criteria (2 prospective3Eggesbo M. Botten G. Stigum H. Nafstad P. Magnus P. Is delivery by cesarean section a risk factor for food allergy?.J Allergy Clin Immunol. 2003; 112: 420-426Abstract Full Text Full Text PDF PubMed Scopus (189) Google Scholar, 4Risnes K.R. Belanger K. Murk W. Bracken M.B. Antibiotic exposure by 6 months and asthma and allergy at 6 years: findings in a cohort of 1,401 US children.Am J Epidemiol. 2011; 173: 310-318Crossref PubMed Scopus (197) Google Scholar and 4 retrospective5Dowhower Karpa K. Paul I.M. Leckie J.A. Shung S. Carkaci-Salli N. Vrana K.E. et al.A retrospective chart review to identify perinatal factors associated with food allergies.Nutr J. 2012; 11: 87Crossref PubMed Scopus (20) Google Scholar, 6Love B.L. Mann J.R. Hardin J.W. Lu Z.K. Cox C. Amrol D.J. Antibiotic prescription and food allergy in young children.Allergy Asthma Clin Immunol. 2016; 12: 41Crossref PubMed Scopus (33) Google Scholar, 7Hirsch A.G. Pollak J. Glass T.A. Poulsen M.N. Bailey-Davis L. Mowery J. et al.Early-life antibiotic use and subsequent diagnosis of food allergy and allergic diseases.Clin Exp Allergy. 2017; 47: 236-244Crossref PubMed Scopus (58) Google Scholar, 8Li M. Lu Z.K. Amrol D.J. Mann J.R. Hardin J.W. Yuan J. et al.Antibiotic exposure and the risk of food allergy: evidence in the US Medicaid pediatric population.J Allergy Clin Immunol Pract. 2019; 7: 492-499Abstract Full Text Full Text PDF Scopus (9) Google Scholar). Of these, 4 articles (1 prospective4Risnes K.R. Belanger K. Murk W. Bracken M.B. Antibiotic exposure by 6 months and asthma and allergy at 6 years: findings in a cohort of 1,401 US children.Am J Epidemiol. 2011; 173: 310-318Crossref PubMed Scopus (197) Google Scholar and 3 retrospective6Love B.L. Mann J.R. Hardin J.W. Lu Z.K. Cox C. Amrol D.J. Antibiotic prescription and food allergy in young children.Allergy Asthma Clin Immunol. 2016; 12: 41Crossref PubMed Scopus (33) Google Scholar, 7Hirsch A.G. Pollak J. Glass T.A. Poulsen M.N. Bailey-Davis L. Mowery J. et al.Early-life antibiotic use and subsequent diagnosis of food allergy and allergic diseases.Clin Exp Allergy. 2017; 47: 236-244Crossref PubMed Scopus (58) Google Scholar, 8Li M. Lu Z.K. Amrol D.J. Mann J.R. Hardin J.W. Yuan J. et al.Antibiotic exposure and the risk of food allergy: evidence in the US Medicaid pediatric population.J Allergy Clin Immunol Pract. 2019; 7: 492-499Abstract Full Text Full Text PDF Scopus (9) Google Scholar) found a significant association between early-life antibiotic exposure and food allergy in childhood. All 4 attempted to adjust for protopathic bias but did not correct for all potential factors. Two of the 4 studies found the association correlated with antibiotic exposure (number of courses). Two studies did not find an association between antibiotic exposure and food allergy. The results are outlined in Table I.3Eggesbo M. Botten G. Stigum H. Nafstad P. Magnus P. Is delivery by cesarean section a risk factor for food allergy?.J Allergy Clin Immunol. 2003; 112: 420-426Abstract Full Text Full Text PDF PubMed Scopus (189) Google Scholar, 4Risnes K.R. Belanger K. Murk W. Bracken M.B. Antibiotic exposure by 6 months and asthma and allergy at 6 years: findings in a cohort of 1,401 US children.Am J Epidemiol. 2011; 173: 310-318Crossref PubMed Scopus (197) Google Scholar, 5Dowhower Karpa K. Paul I.M. Leckie J.A. Shung S. Carkaci-Salli N. Vrana K.E. et al.A retrospective chart review to identify perinatal factors associated with food allergies.Nutr J. 2012; 11: 87Crossref PubMed Scopus (20) Google Scholar, 6Love B.L. Mann J.R. Hardin J.W. Lu Z.K. Cox C. Amrol D.J. Antibiotic prescription and food allergy in young children.Allergy Asthma Clin Immunol. 2016; 12: 41Crossref PubMed Scopus (33) Google Scholar, 7Hirsch A.G. Pollak J. Glass T.A. Poulsen M.N. Bailey-Davis L. Mowery J. et al.Early-life antibiotic use and subsequent diagnosis of food allergy and allergic diseases.Clin Exp Allergy. 2017; 47: 236-244Crossref PubMed Scopus (58) Google Scholar, 8Li M. Lu Z.K. Amrol D.J. Mann J.R. Hardin J.W. Yuan J. et al.Antibiotic exposure and the risk of food allergy: evidence in the US Medicaid pediatric population.J Allergy Clin Immunol Pract. 2019; 7: 492-499Abstract Full Text Full Text PDF Scopus (9) Google ScholarTable ISummary of studies that have investigated the association between early-life exposure to antibiotics and childhood food allergyFirst author, year of publicationStudy design/countryNo. of participants/age groupAntibiotic exposureOutcome definitionResultsRelevant allergic factors included in adjusted resultsLinear association observedAntibiotic class specifiedEggesbo et al,3Eggesbo M. Botten G. Stigum H. Nafstad P. Magnus P. Is delivery by cesarean section a risk factor for food allergy?.J Allergy Clin Immunol. 2003; 112: 420-426Abstract Full Text Full Text PDF PubMed Scopus (189) Google Scholar 2003Prospective cohort/Norway2803/0-2.5 yFirst 6 mo of lifeParent-reported immediate reaction to egg plus positive IgE test result or positive OFC result∗Second outcome measure (questionnaire-based symptoms) did not fit the inclusion criteria.OR, 1.4 (95% CI, 0.6-3.3); aOR, 1.5 (95% CI, 0.6-3.7)NoneNoNoRisnes et al,4Risnes K.R. Belanger K. Murk W. Bracken M.B. Antibiotic exposure by 6 months and asthma and allergy at 6 years: findings in a cohort of 1,401 US children.Am J Epidemiol. 2011; 173: 310-318Crossref PubMed Scopus (197) Google Scholar 2011Prospective cohort/United States1401/0-6 yFirst 6 mo of lifeParent-reported allergic reaction (interview) plus report of a positive blood IgE test or positive SPT resultOR, 1.93 (95% CI, 1.38-2.71); aOR, 1.59 (95% CI, 1.10-2.28)LRI,parental asthmaNoNoDowhower Karpa et al,5Dowhower Karpa K. Paul I.M. Leckie J.A. Shung S. Carkaci-Salli N. Vrana K.E. et al.A retrospective chart review to identify perinatal factors associated with food allergies.Nutr J. 2012; 11: 87Crossref PubMed Scopus (20) Google Scholar 2012Retrospectivecase-control/United States99 cases and 192 control subjects/0-18 yFirst 18 y of ageICD-9-CM codes plus positive IgE test or positive SPT result or pediatrician-diagnosed food allergyOR, 1.35 (95% CI, 0.680-2.68)NoneNoNoLove et al,6Love B.L. Mann J.R. Hardin J.W. Lu Z.K. Cox C. Amrol D.J. Antibiotic prescription and food allergy in young children.Allergy Asthma Clin Immunol. 2016; 12: 41Crossref PubMed Scopus (33) Google Scholar 2016Retrospectivecase-control/United States1504 cases and5995 controlsubjects/0-3 yFirst year of lifePhysician's diagnosis with matching ICD-9-CM code≥1 AB course vs none: OR, 1.22 (95% CI, 1.08-1.37); ≥1 AB course vs none: aOR, 1.21 (95% CI, 1.06-1.39); >5 AB courses vs none: aOR, 1.64 (95% CI, 1.31-2.05)Asthma,eczema,wheezeYes; increase in aORYesHirsch et al,7Hirsch A.G. Pollak J. Glass T.A. Poulsen M.N. Bailey-Davis L. Mowery J. et al.Early-life antibiotic use and subsequent diagnosis of food allergy and allergic diseases.Clin Exp Allergy. 2017; 47: 236-244Crossref PubMed Scopus (58) Google Scholar 2017Retrospective case-control/United States484 cases and 598 control subjects/2 mo-7 yMilk allergy: up to 30 d before diagnosis†Multiple other results were presented in the original article.Non-milk food allergy: 0-12 mo†Multiple other results were presented in the original article.Outpatient, inpatient, or emergency department encounter or medication order plus ICD-9-CM codesMilk allergy diagnosed up to 300 d†Multiple other results were presented in the original article.1-2 AB courses vs none: OR, 1.43 (95% CI, 0.97-2.09); ≥3 AB courses vs none: OR, 2.39 (95% CI, 1.25-4.59); 1-2 AB courses vs none: aOR, 1.35 (95% CI, 0.84-2.17); ≥3 AB courses vs none: aOR, 3.65 (95% CI, 1.75-7.60)Non-milk food allergy diagnosed up to 700 d†Multiple other results were presented in the original article.1-2 AB courses vs none: OR, 1.27 (95% CI, 0.96-1.69); ≥3 AB courses vs none: OR, 1.95 (95% CI, 1.38-2.75); 1-2 AB courses vs none: aOR, 1.01 (95% CI, 0.66-1.55); ≥3 AB courses vs none: aOR, 1.40 (95% CI, 0.85-2.33)Infants with otherallergic diseasesYes;increase in ORYesLi et al,8Li M. Lu Z.K. Amrol D.J. Mann J.R. Hardin J.W. Yuan J. et al.Antibiotic exposure and the risk of food allergy: evidence in the US Medicaid pediatric population.J Allergy Clin Immunol Pract. 2019; 7: 492-499Abstract Full Text Full Text PDF Scopus (9) Google Scholar 2019Retrospective case-control/United States500,647 antibiotic users and 500,647 antibiotic non-users/0-4 yFirst year of lifeICD-9-CM code≥1 vs none: HR, 1.40 (95% CI, 1.34-1.45)1 vs none: HR, 1.49 (95% CI, 1.41-1.58); ≥2 vs none: HR, 1.31 (95% CI, 1.24-1.39)≥1 vs none: aHR, 1.46 (95% CI, 1.38-1.55)Rhinitis,dermatitis,toxic effect,and other adversefood reactionsYes;no increase in HRNoAB, Antibiotic; aHR, adjusted hazard ratio; aOR, adjusted odds ratio; HR, hazard ratio; LRI, lower respiratory tract infection; OFC, oral food challenge; OR, odds ratio; SPT, skin prick test.∗ Second outcome measure (questionnaire-based symptoms) did not fit the inclusion criteria.† Multiple other results were presented in the original article. Open table in a new tab AB, Antibiotic; aHR, adjusted hazard ratio; aOR, adjusted odds ratio; HR, hazard ratio; LRI, lower respiratory tract infection; OFC, oral food challenge; OR, odds ratio; SPT, skin prick test. Overall, our systematic review found evidence to suggest an association between early-life antibiotic exposure and childhood food allergy. However, this association was found in only 4 of 6 studies. There are several potential explanations for the discrepancies in the results of the 6 studies. First, study design and methods differed between the studies. In particular, the number of participants varied, and one study (with only 291 participants) that did not find a significant association between antibiotic exposure and allergy might have been insufficiently powered to find a difference.5Dowhower Karpa K. Paul I.M. Leckie J.A. Shung S. Carkaci-Salli N. Vrana K.E. et al.A retrospective chart review to identify perinatal factors associated with food allergies.Nutr J. 2012; 11: 87Crossref PubMed Scopus (20) Google Scholar Second, the definition of food allergy outcomes differed between studies. Of the 4 studies that found a significant association, in 3 studies food allergy was defined by using International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) codes, which do not distinguish between food intolerance and food allergy.6Love B.L. Mann J.R. Hardin J.W. Lu Z.K. Cox C. Amrol D.J. Antibiotic prescription and food allergy in young children.Allergy Asthma Clin Immunol. 2016; 12: 41Crossref PubMed Scopus (33) Google Scholar, 7Hirsch A.G. Pollak J. Glass T.A. Poulsen M.N. Bailey-Davis L. Mowery J. et al.Early-life antibiotic use and subsequent diagnosis of food allergy and allergic diseases.Clin Exp Allergy. 2017; 47: 236-244Crossref PubMed Scopus (58) Google Scholar, 8Li M. Lu Z.K. Amrol D.J. Mann J.R. Hardin J.W. Yuan J. et al.Antibiotic exposure and the risk of food allergy: evidence in the US Medicaid pediatric population.J Allergy Clin Immunol Pract. 2019; 7: 492-499Abstract Full Text Full Text PDF Scopus (9) Google Scholar The fourth study that found a significant association defined food allergy by using a more robust definition, namely self-reported allergic reaction (using a standardized interview) plus a positive IgE test or skin prick test result.4Risnes K.R. Belanger K. Murk W. Bracken M.B. Antibiotic exposure by 6 months and asthma and allergy at 6 years: findings in a cohort of 1,401 US children.Am J Epidemiol. 2011; 173: 310-318Crossref PubMed Scopus (197) Google Scholar One of the 2 studies that did not find a significant association also used ICD-9-CM codes but included the results of IgE or skin prick tests to define food allergy.5Dowhower Karpa K. Paul I.M. Leckie J.A. Shung S. Carkaci-Salli N. Vrana K.E. et al.A retrospective chart review to identify perinatal factors associated with food allergies.Nutr J. 2012; 11: 87Crossref PubMed Scopus (20) Google Scholar The other study that did not find an association used parent-reported allergic reaction (interview) plus a report of a positive blood IgE test or positive skin prick test result.3Eggesbo M. Botten G. Stigum H. Nafstad P. Magnus P. Is delivery by cesarean section a risk factor for food allergy?.J Allergy Clin Immunol. 2003; 112: 420-426Abstract Full Text Full Text PDF PubMed Scopus (189) Google Scholar Among the 4 studies that found an association, the extent to which protopathic bias was considered, and the adjustment models used differed, which affects the interpretation of their results. Because IgE-mediated food allergies have been associated with concomitant IgE-mediated allergic diseases,9Alduraywish S.A. Lodge C.J. Campbell B. Allen K.J. Erbas B. Lowe A.J. et al.The march from early life food sensitization to allergic disease: a systematic review and meta-analyses of birth cohort studies.Allergy. 2016; 71: 77-89Crossref PubMed Scopus (104) Google Scholar 3 studies6Love B.L. Mann J.R. Hardin J.W. Lu Z.K. Cox C. Amrol D.J. Antibiotic prescription and food allergy in young children.Allergy Asthma Clin Immunol. 2016; 12: 41Crossref PubMed Scopus (33) Google Scholar, 7Hirsch A.G. Pollak J. Glass T.A. Poulsen M.N. Bailey-Davis L. Mowery J. et al.Early-life antibiotic use and subsequent diagnosis of food allergy and allergic diseases.Clin Exp Allergy. 2017; 47: 236-244Crossref PubMed Scopus (58) Google Scholar, 8Li M. Lu Z.K. Amrol D.J. Mann J.R. Hardin J.W. Yuan J. et al.Antibiotic exposure and the risk of food allergy: evidence in the US Medicaid pediatric population.J Allergy Clin Immunol Pract. 2019; 7: 492-499Abstract Full Text Full Text PDF Scopus (9) Google Scholar adjusted for other allergic diseases. In 2 studies the association between antibiotic exposure and food allergy remained unchanged,6Love B.L. Mann J.R. Hardin J.W. Lu Z.K. Cox C. Amrol D.J. Antibiotic prescription and food allergy in young children.Allergy Asthma Clin Immunol. 2016; 12: 41Crossref PubMed Scopus (33) Google Scholar, 8Li M. Lu Z.K. Amrol D.J. Mann J.R. Hardin J.W. Yuan J. et al.Antibiotic exposure and the risk of food allergy: evidence in the US Medicaid pediatric population.J Allergy Clin Immunol Pract. 2019; 7: 492-499Abstract Full Text Full Text PDF Scopus (9) Google Scholar whereas in the other study the association between antibiotic exposure and nonmilk food allergy was no longer significant.7Hirsch A.G. Pollak J. Glass T.A. Poulsen M.N. Bailey-Davis L. Mowery J. et al.Early-life antibiotic use and subsequent diagnosis of food allergy and allergic diseases.Clin Exp Allergy. 2017; 47: 236-244Crossref PubMed Scopus (58) Google Scholar This suggests that the other allergy diagnoses accounted for the observed association between antibiotics and food allergy in one of the studies. However, exclusion of cases that have other allergic diseases could also lead to an underestimation of a clinically relevant effect of antibiotic exposure on IgE-mediated food allergy, even if the latter is a secondary or concomitant effect of another IgE-mediated allergic disease. One of the studies accounted for the age of antibiotic exposure and found the association to be similar across different age groups.7Hirsch A.G. Pollak J. Glass T.A. Poulsen M.N. Bailey-Davis L. Mowery J. et al.Early-life antibiotic use and subsequent diagnosis of food allergy and allergic diseases.Clin Exp Allergy. 2017; 47: 236-244Crossref PubMed Scopus (58) Google Scholar However, although remaining statistically significant, the association between antibiotic use and food allergy weakened as the time between antibiotic exposure and diagnosis increased. The association between antibiotics and food allergies is likely to be multifactorial. Because the microbiome plays a critical role in development of the immune system, antibiotic-induced changes in the microbiome are likely to be the predominant mechanism by which antibiotic exposure influences the risk of food allergy.1Osborne N.J. Koplin J.J. Martin P.E. Gurrin L.C. Lowe A.J. Matheson M.C. et al.Prevalence of challenge-proven IgE-mediated food allergy using population-based sampling and predetermined challenge criteria in infants.J Allergy Clin Immunol. 2011; 127: 668-676, e1-2Abstract Full Text Full Text PDF PubMed Scopus (755) Google Scholar As a result of their stronger influence on the microbiome, broad-spectrum (in contrast to narrow-spectrum) antibiotics are likely to have a greater influence on the development of the immune system and allergic disease. In support of this, the 2 studies that differentiated between the effect of antibiotic type both found the strongest association with broad-spectrum antibiotics.6Love B.L. Mann J.R. Hardin J.W. Lu Z.K. Cox C. Amrol D.J. Antibiotic prescription and food allergy in young children.Allergy Asthma Clin Immunol. 2016; 12: 41Crossref PubMed Scopus (33) Google Scholar, 7Hirsch A.G. Pollak J. Glass T.A. Poulsen M.N. Bailey-Davis L. Mowery J. et al.Early-life antibiotic use and subsequent diagnosis of food allergy and allergic diseases.Clin Exp Allergy. 2017; 47: 236-244Crossref PubMed Scopus (58) Google Scholar In conclusion, there is some evidence to support the hypothesis that early-life antibiotic exposure is associated with the development of childhood food allergy. However, an association was not consistently found in all studies. Whether this association is causal or correlational remains uncertain because of inconsistencies between studies in design and the inability to completely exclude protopathic bias. It also remains uncertain whether any association is only applicable in certain children or is dependent on antibiotic exposure (number of courses, dose, duration, and type), age of antibiotic exposure, or spectrum of antibiotic activity. Future prospective studies should minimize protopathic bias (by excluding antibiotic exposure close to the time of allergy diagnosis and accurately recording the indication for antibiotic use), which is correct for other allergic diseases and assess the role of number of courses, dose, duration, and type (narrow vs broad spectrum) of antibiotic exposure. We thank Ms Poh Chua for assistance with the search strategy.