Advertisement

Relationship between birth weight or fetal growth rate and postnatal allergy: A systematic review

  • Amy L. Wooldridge
    Affiliations
    Robinson Research Institute, University of Adelaide, Adelaide, Australia

    Adelaide Medical School, University of Adelaide, Adelaide, Australia

    School of Human Sciences, University of Western Australia, Perth, Australia
    Search for articles by this author
  • Mark McMillan
    Affiliations
    Robinson Research Institute, University of Adelaide, Adelaide, Australia

    Adelaide Medical School, University of Adelaide, Adelaide, Australia

    Vaccinology and Immunology Research Trials Unit (VIRTU), Women's and Children's Hospital, North Adelaide, Australia
    Search for articles by this author
  • Manpreet Kaur
    Affiliations
    Robinson Research Institute, University of Adelaide, Adelaide, Australia

    School of Medicine, Deakin University, Waurn Ponds, Australia
    Search for articles by this author
  • Lynne C. Giles
    Affiliations
    Robinson Research Institute, University of Adelaide, Adelaide, Australia

    School of Public Health, University of Adelaide, Adelaide, Australia
    Search for articles by this author
  • Helen S. Marshall
    Affiliations
    Robinson Research Institute, University of Adelaide, Adelaide, Australia

    Adelaide Medical School, University of Adelaide, Adelaide, Australia

    Vaccinology and Immunology Research Trials Unit (VIRTU), Women's and Children's Hospital, North Adelaide, Australia
    Search for articles by this author
  • Kathryn L. Gatford
    Correspondence
    Corresponding author: Kathryn L. Gatford, PhD, Robinson Research Institute, Level 6, Adelaide Health & Medical Sciences Building, University of Adelaide, Adelaide SA 5005, Australia.
    Affiliations
    Robinson Research Institute, University of Adelaide, Adelaide, Australia

    Adelaide Medical School, University of Adelaide, Adelaide, Australia
    Search for articles by this author
Published:October 12, 2019DOI:https://doi.org/10.1016/j.jaci.2019.08.032

      Background

      Individual susceptibility to allergic diseases is developmentally programmed by early-life exposures. Evidence from preclinical studies suggests that intrauterine growth restriction is protective against later inflammatory responses to allergens.

      Objective

      We sought to evaluate whether prenatal growth affects susceptibility to allergy in human subjects.

      Methods

      We systematically searched for relevant studies in 11 databases, including Web of Science, ProQuest, EMBASE, and PubMed. We included only studies that corrected for gestational age or were restricted to full-term infants to separate effects of fetal growth from those of prematurity.

      Results

      The 42 eligible studies included prospective and retrospective cohort, cross-sectional, and case-control studies. Only 2 studies reported allergic asthma. A birth weight increase of 1 kg was associated with a 44% greater risk of food allergy in children (odds ratio [OR], 1.44; 95% CI, 1.04-1.99; P = .001), a 17% greater risk of ever allergic dermatitis in children (OR, 1.17; 95% CI, 1.04-1.32; P = .008), and a 34% greater risk of ever or current allergic dermatitis in infants up to 2 years of age (OR, 1.34; 95% CI, 1.08-1.68; P = .009). Risks of allergic rhinitis were not associated with birth weight.

      Conclusions

      The results of these meta-analyses suggest that intrauterine growth restriction protects against allergic diseases in human subjects consistent with preclinical evidence but that effects might differ between allergic diseases. The strongest evidence is available for infancy and early childhood, and additional studies in older children and adults are needed to determine whether the effects of prenatal growth on each allergic disease persist or differ between those with severe and mild phenotypes.

      Graphical abstract

      Key words

      Abbreviations used:

      AA (Allergic asthma), AD (Atopic dermatitis), AR (Allergic rhinitis), BW (Birth weight), COPSAC (Copenhagen Prospective Studies on Asthma in Childhood), FA (Food allergy), ISAAC (International Study of Asthma and Allergies in Childhood), IUGR (Intrauterine growth restriction), OR (Odds ratio), PROBIT (Promotion of Breastfeeding Intervention Trial), Treg (Regulatory T)
      Allergic diseases, including atopic dermatitis (AD)/eczema, allergic rhinitis (AR)/hay fever, food allergy (FA)/anaphylaxis, and asthma, are estimated to affect 30% to 40% of the world's population.
      • World Allergy Organization
      World Allergy Organization (WAO) white book on allergy: update 2013.
      Although genetic factors are important determinants of susceptibility to allergic diseases,
      • Ferreira M.A.
      • Vonk J.M.
      • Baurecht H.
      • Marenholz I.
      • Tian C.
      • Hoffman J.D.
      • et al.
      Shared genetic origin of asthma, hay fever and eczema elucidates allergic disease biology.
      environmental factors must underlie recent rapid increases in prevalence.
      • Mullins R.J.
      Paediatric food allergy trends in a community-based specialist allergy practice, 1995-2006.
      • Galassi C.
      • De Sario M.
      • Biggeri A.
      • Bisanti L.
      • Chellini E.
      • Ciccone G.
      • et al.
      Changes in prevalence of asthma and allergies among children and adolescents in Italy: 1994-2002.
      • Poulos L.M.
      • Waters A.M.
      • Correll P.K.
      • Loblay R.H.
      • Marks G.B.
      Trends in hospitalizations for anaphylaxis, angioedema, and urticaria in Australia, 1993-1994 to 2004-2005.
      The fact that these increases have been largest in younger age groups, particularly for FA and food-induced anaphylaxis,
      • Mullins R.J.
      Paediatric food allergy trends in a community-based specialist allergy practice, 1995-2006.
      • Galassi C.
      • De Sario M.
      • Biggeri A.
      • Bisanti L.
      • Chellini E.
      • Ciccone G.
      • et al.
      Changes in prevalence of asthma and allergies among children and adolescents in Italy: 1994-2002.
      • Poulos L.M.
      • Waters A.M.
      • Correll P.K.
      • Loblay R.H.
      • Marks G.B.
      Trends in hospitalizations for anaphylaxis, angioedema, and urticaria in Australia, 1993-1994 to 2004-2005.
      further implies that exposures before and during pregnancy and in early life are important determinants of a subject's susceptibility to allergy. Exposures, including high maternal body mass index, cigarette smoke exposure during pregnancy, and delivery by means of cesarean section rather than vaginal birth are each associated with increased risk of allergic diseases in human subjects.
      • Grieger J.A.
      • Clifton V.L.
      • Tuck A.R.
      • Wooldridge A.L.
      • Robertson S.A.
      • Gatford K.L.
      In utero programming of allergic susceptibility.
      Conversely, a healthy maternal diet throughout pregnancy and larger family size appear to protect infants from later development of allergy, and there is preclinical and epidemiologic evidence that intrauterine growth restriction (IUGR) might be protective against allergic diseases other than asthma.
      • Grieger J.A.
      • Clifton V.L.
      • Tuck A.R.
      • Wooldridge A.L.
      • Robertson S.A.
      • Gatford K.L.
      In utero programming of allergic susceptibility.
      Results of preclinical studies suggest that restricted growth before birth is protective against later inflammatory responses to allergens, although sensitization, measured as the circulating antigen-specific IgE response, might not be reduced.
      • Landgraf M.A.
      • Landgraf R.G.
      • Jancar S.
      • Fortes Z.B.
      Influence of age on the development of immunological lung response in intrauterine undernourishment.
      • Landgraf M.A.
      • Landgraf R.G.
      • Silva R.C.
      • Semedo P.
      • Camara N.O.
      • Fortes Z.B.
      Intrauterine undernourishment alters TH1/TH2 cytokine balance and attenuates lung allergic inflammation in wistar rats.
      • Wooldridge A.L.
      • Bischof R.J.
      • Meeusen E.N.
      • Liu H.
      • Heinemann G.K.
      • Hunter D.S.
      • et al.
      Placental restriction of fetal growth reduces cutaneous responses to antigen after sensitization in sheep.
      In human subjects associations between the risks of many diseases and birth weight (BW) are observed across the BW range, with risks of cardiometabolic diseases generally negatively associated with BW. For example, in a cohort of 59- to 70-year-old men in east Hertfordshire, rates of impaired glucose tolerance decreased progressively with increasing BW independent of current weight.
      • Hales C.N.
      • Barker D.J.P.
      • Clark P.M.S.
      • Cox L.J.
      • Fall C.
      • Osmond C.
      • et al.
      Fetal and infant growth and impaired glucose tolerance at age 64.
      The risks of disease can also be affected by prenatal exposures without changes in BW.
      • Roseboom T.J.
      • van der Meulen J.H.P.
      • Ravelli A.C.J.
      • Osmond C.
      • Barker D.J.P.
      • Bleker O.P.
      Effects of prenatal exposure to the Dutch famine on adult disease in later life: an overview.
      Interestingly, directions of association between cancer risk and BW differ according to cancer type.
      • Spracklen C.N.
      • Wallace R.B.
      • Sealy-Jefferson S.
      • Robinson J.G.
      • Freudenheim J.L.
      • Wellons M.F.
      • et al.
      Birth weight and subsequent risk of cancer.
      Whether growth before birth affects rates of later allergic diseases in human subjects has not been systematically evaluated.
      In the large multicenter International Study of Asthma and Allergies in Childhood (ISAAC) studies, low BW was associated with decreased risk of AD but increased risk of asthma in the child's first year of life.
      • Mitchell E.A.
      • Clayton T.
      • Garcia-Marcos L.
      • Pearce N.
      • Foliaki S.
      • Wong G.
      Birthweight and the risk of atopic diseases: the ISAAC Phase III study.
      Consistent with these findings, the only systematic review of size at birth and AD published to date reported decreased risk of AD in low-BW (<2.5 kg; odds ratio [OR], 0.68; 95% CI, 0.63–0.75; P < .0001) and increased risk of AD in high-BW (>4.0 kg; OR, 1.1; 95% CI, 1.02–1.17; P = .01) infants compared with normal-BW infants (BW, 2.5–4.0 kg).
      • Panduru M.
      • Salavastru C.M.
      • Panduru N.M.
      • Tiplica G.S.
      Birth weight and atopic dermatitis: systematic review and meta-analyis.
      However, these studies each used absolute BW categories to assess exposure and did not account for major determinants of size at birth needed to determine whether low BW is due to restricted fetal growth, in particular gestational age at birth
      • Bleker O.P.
      • Buimer M.
      • van der Post J.A.M.
      • van der Veen F.
      Ted (G. J.) Kloosterman: on intrauterine growth. The significance of prenatal care. Studies on birth weight, placental weight and placental index.
      and ethnicity.
      • Ergaz Z.
      • Avgil M.
      • Ornoy A.
      Intrauterine growth restriction—etiology and consequences: what do we know about the human situation and experimental animal models?.
      Protection against allergic diseases by low BW might reflect the effects of gestational age rather than IUGR because in a study of Finnish adults, gestational age was positively associated with risk of adult allergic sensitization (atopy) independent of fetal growth.
      • Pekkanen J.
      • Xu B.
      • Järvelin M.R.
      Gestational age and occurrence of atopy at age 31—a prospective birth cohort study in Finland.
      Twin studies provide stronger evidence that IUGR is protective against allergic diseases in human subjects because siblings share common postnatal environments and gestational age, as well as similar genetics. In analyses of twins discordant for AD/allergic eczema, a 500-g greater BW was associated with nearly a 4-fold greater risk of AD in both monozygotic and dizygotic twins.
      • Lundholm C.
      • Ortqvist A.K.
      • Lichtenstein P.
      • Cnattingius S.
      • Almqvist C.
      Impaired fetal growth decreases the risk of childhood atopic eczema: a Swedish twin study.
      No systematic review has assessed whether fetal growth and consequent size at birth corrected for gestational age are determinants of allergic disease risk in human subjects.
      Therefore the objective of this systematic review was to synthesize the best available evidence on the relationship between size at birth or fetal growth and postnatal allergic diseases, specifically, AD, AR, allergic asthma (AA), and FA. To remove potential confounding caused by the effects of gestational age on BW, we only accepted absolute BW without correction for gestational age as an exposure measure in term cohorts. We also included studies of allergic diseases and BW or fetal size corrected for gestational age, which are usually expressed relative to population or customized growth data.

      Methods

       Protocol and registration

      Before commencement of the review, we developed a detailed protocol, which was published
      • Wooldridge A.L.
      • McMillan M.
      • Marshall H.S.
      • Gatford K.L.
      Relationship between birth weight or fetal growth rate and postnatal allergy: a systematic review protocol.
      and registered with the International Prospective Register of Systematic Reviews (PROSPERO; www.crd.york.ac.uk/prospero/, reference CRD42017054966).

       Search strategy

      The systematic literature review was performed using a 3-step strategy according to the published protocol. Initial limited searching of PubMed and EMBASE was used to identify key words and index terms, and these were used to produce the final search terms for PubMed (see Table E1 in this article's Online Repository at www.jacionline.org). This search syntax was adapted to conduct similar searches in EMBASE, OVID Medline, Web of Science, Informit Health, the Cochrane library, CINAHL, MedNar, ProQuest, Scopus, and Trove (Fig 1). Additional articles were found by searching lists of references of all included publications and relevant reviews.
      Figure thumbnail gr1
      Fig 1Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram for database search of studies investigating the association between size at birth or fetal growth corrected for gestational age and risk of allergic disease (AA, FA/anaphylaxis, AR/hay fever, or allergic dermatitis/eczema). GA, Gestational age.

       Screening

      Two of the authors (ALW, MK, and/or KLG) independently reviewed the titles and abstracts of all publications retrieved by the literature search, and any disagreements were decided by consensus. Publications were included for full-text screening if they were of human subjects, primary sources (eg, not reviews or commentaries), published in English, and with a title and/or abstract including an allergic disease or synonym (eczema, dermatitis, rhinitis, hay fever, asthma, eosinophilic disease, anaphylaxis, FA, skin prick test, IgE, urticaria, allergy, atopy, sensitization, and airway/respiratory hyperresponsiveness) and if the title and/or abstract mentioned an exposure of interest (BW/size/phenotype/characteristics, fetal growth measures, size for gestational age, IUGR, fetal growth restriction, or maternal/fetal pregnancy exposures, including smoking).
      Full texts of relevant studies were obtained, when available, for further assessment, and each was reviewed independently by 2 of the authors (ALW, MK, and/or KLG). Publications were included for data extraction if assessment of the full text confirmed they were not duplicates; met the abstract criteria; were full text (not conference abstracts); were not articles reporting trial protocols without results; had outcomes in the correct population (progeny); had outcomes analyzed by exposures of interest; had outcomes that were physician diagnosed or diagnosed by using clinical criteria, such as ISAAC questionnaires; had asthma that was specifically diagnosed as allergic; and had outcomes that were either in progeny who were born at full term (≥37 weeks completed gestational age) or were corrected for gestational age at birth. We accepted all articles with an actual or online publication date up to December 31, 2017. Disagreement at data screening stages was resolved by means of discussion between the authors. The main characteristics and key results of the included studies are described in Table E2 in this article's Online Repository at www.jacionline.org.

       Quality assessment

      Studies included after full-text screening were assessed independently by 2 reviewers (ALW, MM, MK, and/or KLG) for methodological quality by using standardized critical appraisal instruments from the Joanna Briggs Institute.
      • Moola S.
      • Munn Z.
      • Tufanaru C.
      • Aromataris E.
      • Sears K.
      • Sfetcu R.
      • et al.
      Disagreements that arose were resolved through discussion or with a third reviewer. Quality assessments of the included studies are provided in Table E3 (cohort studies), Table E4 (case-control studies), and Table E5 (cross-sectional studies) in this article's Online Repository at www.jacionline.org.

       Data extraction

      Independent double data extraction was performed by ALW, MM, MK, and/or KLG by using author-designed fields in REDcap.
      • Harris P.A.
      • Taylor R.
      • Thielke R.
      • Payne J.
      • Gonzalez N.
      • Conde J.G.
      Research electronic data capture (REDCap)—a metadata-driven methodology and workflow process for providing translational research informatics support.
      Discrepancies were resolved through consultation. During data extraction, we assessed studies using the same cohort for duplication of analyses. The studies published by Goksor et al
      • Goksor E.
      • Alm B.
      • Pettersson R.
      • Mollborg P.
      • Erdes L.
      • Aberg N.
      • et al.
      Early fish introduction and neonatal antibiotics affect the risk of asthma into school age.
      and Loid et al
      • Loid P.
      • Goksor E.
      • Alm B.
      • Pettersson R.
      • Mollborg P.
      • Erdes L.
      • et al.
      A persistently high body mass index increases the risk of atopic asthma at school age.
      each reported AA outcomes at the same age in children of the Western Sweden cohort. Therefore we excluded the study by Goksor et al
      • Goksor E.
      • Alm B.
      • Pettersson R.
      • Mollborg P.
      • Erdes L.
      • Aberg N.
      • et al.
      Early fish introduction and neonatal antibiotics affect the risk of asthma into school age.
      from the narrative review and retained only the report from Loid et al,
      • Loid P.
      • Goksor E.
      • Alm B.
      • Pettersson R.
      • Mollborg P.
      • Erdes L.
      • et al.
      A persistently high body mass index increases the risk of atopic asthma at school age.
      which described allergic disease outcomes for a greater number of BW categories.
      Another pair of studies based on the same cohort and using offspring born in the same birth years reported different subsets of subjects at the same outcome age (Promotion of Breastfeeding Intervention Trial [PROBIT] trial),
      • Anderson E.L.
      • Fraser A.
      • Martin R.M.
      • Kramer M.S.
      • Oken E.
      • Patel R.
      • et al.
      Associations of postnatal growth with asthma and atopy: the PROBIT study.
      • Kramer M.S.
      • Guo T.
      • Platt R.W.
      • Sevkovskaya Z.
      • Dzikovich I.
      • Collet J.P.
      • et al.
      Does previous infection protect against atopic eczema and recurrent wheeze in infancy?.
      whereas 2 articles reporting outcomes in children from the Copenhagen Prospective Studies on Asthma in Childhood (COPSAC) cohort reported outcomes in the same cohort at different ages.
      • Bisgaard H.
      • Halkjaer L.B.
      • Hinge R.
      • Giwercman C.
      • Palmer C.
      • Silveira L.
      • et al.
      Risk analysis of early childhood eczema.
      • Sevelsted A.
      • Bisgaard H.
      Neonatal size in term children is associated with asthma at age 7, but not with atopic dermatitis or allergic sensitization.
      These articles were each retained in the summaries and descriptive text, but only 1 study from each cohort was included in meta-analyses.
      For the PROBIT studies, we chose to include data from Anderson et al
      • Anderson E.L.
      • Fraser A.
      • Martin R.M.
      • Kramer M.S.
      • Oken E.
      • Patel R.
      • et al.
      Associations of postnatal growth with asthma and atopy: the PROBIT study.
      in meta-analyses of AD risk because this included the entire cohort, whereas Kramer et al
      • Kramer M.S.
      • Guo T.
      • Platt R.W.
      • Sevkovskaya Z.
      • Dzikovich I.
      • Collet J.P.
      • et al.
      Does previous infection protect against atopic eczema and recurrent wheeze in infancy?.
      conducted a comparison of AD cases and control subjects without AD. For the COPSAC studies, only the article by Sevelsted and Bisgaard
      • Sevelsted A.
      • Bisgaard H.
      Neonatal size in term children is associated with asthma at age 7, but not with atopic dermatitis or allergic sensitization.
      reported the effects of (estimated) BW and was able to be included in meta-analyses. Studies within the Danish National Birth Cohort
      • Bager P.
      • Melbye M.
      • Rostgaard K.
      • Benn C.S.
      • Westergaard T.
      Mode of delivery and risk of allergic rhinitis and asthma.
      • Linneberg A.
      • Simonsen J.B.
      • Petersen J.
      • Stensballe L.G.
      • Benn C.S.
      Differential effects of risk factors on infant wheeze and atopic dermatitis emphasize a different etiology.
      and from well-baby health checks in Fukuoka City, Japan,
      • Hikino S.
      • Nakayama H.
      • Yamamoto J.
      • Kinukawa N.
      • Sakamoto M.
      • Hara T.
      Food allergy and atopic dermatitis in low birthweight infants during early childhood.
      • Miyake Y.
      • Tanaka K.
      Lack of relationship between birth conditions and allergic disorders in Japanese children aged 3 years.
      each reported outcomes in cohorts from nonoverlapping ranges of birth years and were therefore all retained in analyses.

       Data synthesis

      We used the generalized least-squares method
      • Orsini N.
      • Li R.
      • Wolk A.
      • Khudyakov P.
      • Spiegelman D.
      Meta-analysis for linear and nonlinear dose-response relations: examples, an evaluation of approximations, and software.
      • Greenland S.
      • Longnecker M.P.
      Methods for trend estimation from summarized dose-response data, with applications to meta-analysis.
      to estimate the linear association of BW with each of the outcomes in each study. We applied random-effects meta-analysis models to account for the correlation between multiple categories of BW within a given study.
      In studies in which ORs were reported for categories of BW, we used the midpoint of the category in estimation of the linear relationship. Where an upper or lower limit for a BW category was not given, we assumed BW was normally distributed. The proportion of BW distribution that fell into a given category was then used along with the inverse normal probability distribution to estimate the value that corresponded to the midpoint of that category.
      Where studies reported the mean and SD, we used these values in back calculation of the BW category midpoint. When these summary statistics were not reported, we used values from similar contemporaneous studies for the estimation of the mean and SD in the back calculation. Where not stated, the number of cases for each outcome for each BW category were derived from the reported adjusted OR. The ORs and corresponding 95% CIs for each of the individual studies were transformed by using natural logarithms so as to normalize distributions and to allow calculation of SEs. The pooled log ORs and SEs for each of the outcomes for which data on effects of BW were available from 3 or more studies (FA, AR, and allergic eczema) were estimated by using a random-effects model weighted by the inverse variance estimate.
      • Orsini N.
      • Li R.
      • Wolk A.
      • Khudyakov P.
      • Spiegelman D.
      Meta-analysis for linear and nonlinear dose-response relations: examples, an evaluation of approximations, and software.
      ORs and 95% CIs per 1000-g increase in BW were then calculated by exponentiating 1000 times the coefficient and CI limits derived from the random-effects model. Heterogeneity for studies included in the meta-analyses was assessed by calculation of the I2 and τ2 statistics. All analyses were performed with Stata statistical software (version 15.1; StataCorp, College Station, Tex).

      Results

      Of the 15,093 studies identified from our search, 42 were included in the narrative review of the association between size at birth or fetal growth corrected for gestational age and risk of AA, FA, AR, or AD (Fig 1). BW was the most commonly reported exposure, and we were able to include 30 studies in meta-analyses of the association between BW corrected for gestational age and risk of FA, AR, or AD (Fig 1).

       AA

      Only 2 independent studies investigating the relationship between size at birth corrected for gestational age and AA were identified, both of which were conducted in Scandinavian populations.

       FA/anaphylaxis

      Four studies investigated FA outcomes, most of which investigated the incidence of FA in early childhood (before 2 years of age),
      • Hikino S.
      • Nakayama H.
      • Yamamoto J.
      • Kinukawa N.
      • Sakamoto M.
      • Hara T.
      Food allergy and atopic dermatitis in low birthweight infants during early childhood.
      • Gabet S.
      • Just J.
      • Couderc R.
      • Seta N.
      • Momas I.
      Allergic sensitisation in early childhood: Patterns and related factors in PARIS birth cohort.
      • Metsala J.
      • Lundqvist A.
      • Kaila M.
      • Gissler M.
      • Klaukka T.
      • Virtanen S.M.
      Maternal and perinatal characteristics and the risk of cow's milk allergy in infants up to 2 years of age: a case-control study nested in the Finnish population.
      although one study in an American population reported ever FA up to 3 to 10 years of age.
      • Renz-Polster H.
      • David M.R.
      • Buist A.S.
      • Vollmer W.M.
      • O'Connor E.A.
      • Frazier E.A.
      • et al.
      Caesarean section delivery and the risk of allergic disorders in childhood.
      Three of these studies considered an allergy-positive outcome as any history of FA (ever FA) until the age of assessment,
      • Hikino S.
      • Nakayama H.
      • Yamamoto J.
      • Kinukawa N.
      • Sakamoto M.
      • Hara T.
      Food allergy and atopic dermatitis in low birthweight infants during early childhood.
      • Metsala J.
      • Lundqvist A.
      • Kaila M.
      • Gissler M.
      • Klaukka T.
      • Virtanen S.M.
      Maternal and perinatal characteristics and the risk of cow's milk allergy in infants up to 2 years of age: a case-control study nested in the Finnish population.
      • Renz-Polster H.
      • David M.R.
      • Buist A.S.
      • Vollmer W.M.
      • O'Connor E.A.
      • Frazier E.A.
      • et al.
      Caesarean section delivery and the risk of allergic disorders in childhood.
      whereas one study assessed current FA as the outcome.
      • Gabet S.
      • Just J.
      • Couderc R.
      • Seta N.
      • Momas I.
      Allergic sensitisation in early childhood: Patterns and related factors in PARIS birth cohort.
      Because of the limited number of eligible studies of FA, we were not able to assess ever and current FA outcomes separately.
      Data on allergy ever to 3 years in Japanese children from Hikino et al
      • Hikino S.
      • Nakayama H.
      • Yamamoto J.
      • Kinukawa N.
      • Sakamoto M.
      • Hara T.
      Food allergy and atopic dermatitis in low birthweight infants during early childhood.
      were excluded from the meta-analysis because these were a small subset of the children that had already been assessed at 18 months in the same study. Data from these 3-year-olds were not included in our final meta-analysis to avoid resampling, but when we performed sensitivity analyses, their inclusion did not change our conclusion that the risk of FA increased with increasing BW. In a pooled analysis an increase of 1 kg in BW was associated with a 1.44-fold increase in the risk of ever FA in children (OR, 1.44; 95% CI, 1.04-1.99; P = .001; I2 = 70%; τ2 < 0.001; Fig 2).
      Figure thumbnail gr2
      Fig 2Meta-analysis of the relationship between BW corrected for gestational age and risk of FA in childhood. Data for each study are shown for each BW category reported. n/N, Number of cases and total population within each BW category; OR (95% CI), ORs and 95% CIs relative to the lowest BW group.

       AR/hay fever

      Sixteen studies investigated AR outcomes, with 14 of these able to be included in meta-analyses. Because AR was assessed at a wide range of ages, with some studies only considering current AR (within the past 12 months) and others considering any history of (ever) AR, we grouped these studies into AR ever in childhood, current AR in childhood, AR ever in adulthood, and current AR in adulthood. There was no evidence that the risks of AR ever in childhood (OR, 1.02; 95% CI, 0.91-1.14; P = .767; I2 = 50%; τ2 < 0.001), current AR in childhood (OR, 0.92; 95% CI, 0.69-1.23; P = .579; I2 < 10%; τ2 < 0.001), or AR ever in adulthood (OR, 0.97; 95% CI, 0.87-1.09; P = .645; I2 < 10%, τ2 < 0.001) were altered by BW (Fig 3).
      Figure thumbnail gr3
      Fig 3Meta-analysis of the relationship between BW corrected for gestational age and the risk of AR history (ever) in childhood (A), current AR in childhood (B), and AR history (ever) assessed in adulthood (C). Data for each study are shown for each BW category reported. n/N, Number of cases and total population within each BW category; OR (95% CI), ORs and 95% CIs relative to the lowest BW group.
      For AR ever in childhood, Lin et al
      • Lin M.H.
      • Hsieh C.J.
      • Caffrey J.L.
      • Lin Y.S.
      • Wang I.J.
      • Ho W.C.
      • et al.
      Fetal growth, obesity, and atopic disorders in adolescence: a retrospective birth cohort study.
       contained approximately 50 times greater numbers than the next largest study. Therefore we also analyzed this outcome excluding data from this study. Excluding Lin et al
      • Lin M.H.
      • Hsieh C.J.
      • Caffrey J.L.
      • Lin Y.S.
      • Wang I.J.
      • Ho W.C.
      • et al.
      Fetal growth, obesity, and atopic disorders in adolescence: a retrospective birth cohort study.
      decreased the precision of the estimate but had little effect on the estimated effect size and did not alter the conclusion of no effect of BW on AR ever in childhood (OR, 1.06; 95% CI, 0.89-1.26; P = .546; I2 = 56%; τ2 < 0.001).
      Only 2 independent studies investigating the relationship between size at birth corrected for gestational age and current AR in adulthood were identified, which did not permit a meta-analysis of this outcome. Although the larger of these 2 studies reported lower rates of current AR in small for gestational age–born adults than the remainder of the population,
      • Braback L.
      • Hedberg A.
      Perinatal risk factors for atopic disease in conscripts.
      a smaller twin study did not find differences in BW between the twin with AR and the twin without AR.
      • Ericsson C.H.
      • Svartengren M.
      • Mossberg B.
      • Pedersen N.
      • Camner P.
      Bronchial reactivity and allergy-promoting factors in monozygotic twins discordant for allergic rhinitis.

       Allergic dermatitis/eczema

      Similar to AR, ages and timeframes of AD diagnosis varied between studies for those able to be included in meta-analyses. Therefore we separated our analyses according to the age at AD assessment (children or adults) and whether only current AD (within 12 months) or ever AD (any history of AD) was reported. Sufficient studies were also available for this outcome to allow meta-analysis of the association between BW and AR in infancy up to 2 years of age. Overall, a BW increase of 1 kg was associated with a 17% greater risk of ever AD in children (OR, 1.17; 95% CI, 1.04-1.32; P = .008; I2 = 86%; τ2 < 0.001; Fig 4). For this analysis, we included the population-wide cohort from the Olesen et al
      • Olesen A.B.
      • Ellingsen A.R.
      • Olesen H.
      • Juul S.
      • Thestrup-Pedersen K.
      Atopic dermatitis and birth factors: historical follow up by record linkage.
      study, which had a much lower AD rate (5.6%) than many other studies, reflecting their use of specialist AD diagnosis as the outcome. In a sensitivity analysis we re-ran the meta-analysis using the subset of children in this study who were selected for increased incidence of preterm birth and using less stringent AD criteria.
      • Olesen A.B.
      • Ellingsen A.R.
      • Olesen H.
      • Juul S.
      • Thestrup-Pedersen K.
      Atopic dermatitis and birth factors: historical follow up by record linkage.
      Including this subset instead of the population-wide cohort made very little difference to the estimated OR for effects of a 1-kg increase in BW on rates of ever AD in childhood (OR, 1.20; 95% CI, 1.06-1.36; P = .003; I2 = 87%, τ2 < 0.001).
      Figure thumbnail gr4
      Fig 4Meta-analysis of the relationship between BW corrected for gestational age and the risk of ever having allergic dermatitis assessed in childhood. Data for each study are shown for each BW category reported. n/N, Number of cases and total population within each BW category; OR (95% CI), ORs and 95% CIs relative to the lowest BW group.
      Two studies contributed the majority of data for this outcome. Egeberg et al
      • Egeberg A.
      • Andersen Y.M.
      • Gislason G.
      • Skov L.
      • Thyssen J.P.
      Neonatal risk factors of atopic dermatitis in Denmark - Results from a nationwide register-based study.
      (total n = 609,548) contained approximately 8 times greater numbers than the next largest study, Lin et al
      • Lin M.H.
      • Hsieh C.J.
      • Caffrey J.L.
      • Lin Y.S.
      • Wang I.J.
      • Ho W.C.
      • et al.
      Fetal growth, obesity, and atopic disorders in adolescence: a retrospective birth cohort study.
      (total n = 74,688), and also substantially larger than the next largest study, Anderson et al
      • Anderson E.L.
      • Fraser A.
      • Martin R.M.
      • Kramer M.S.
      • Oken E.
      • Patel R.
      • et al.
      Associations of postnatal growth with asthma and atopy: the PROBIT study.
      (total n = 12,171). Excluding Egeberg et al
      • Egeberg A.
      • Andersen Y.M.
      • Gislason G.
      • Skov L.
      • Thyssen J.P.
      Neonatal risk factors of atopic dermatitis in Denmark - Results from a nationwide register-based study.
      and Lin et al
      • Lin M.H.
      • Hsieh C.J.
      • Caffrey J.L.
      • Lin Y.S.
      • Wang I.J.
      • Ho W.C.
      • et al.
      Fetal growth, obesity, and atopic disorders in adolescence: a retrospective birth cohort study.
      from the meta-analysis decreased the precision of the estimate for the effect of a 1-kg increase in BW on AD ever in childhood but had little effect on the estimated effect size and did not change the overall result (OR, 1.22; 95% CI, 1.04-1.44; P = .016; I2 = 80%; τ2 < 0.001).
      A 1-kg increase in BW was not associated with the risk of current AD in children overall (OR, 1.03; 95% CI, 0.87-1.22; P = .734; I2 = 66%; τ2 < 0.001; Fig 5). Excluding Egeberg et al
      • Egeberg A.
      • Andersen Y.M.
      • Gislason G.
      • Skov L.
      • Thyssen J.P.
      Neonatal risk factors of atopic dermatitis in Denmark - Results from a nationwide register-based study.
      from this meta-analysis did not substantially change the result (OR, 0.99; 95% CI, 0.81-1.22; P = .930; I2 = 58%, τ2 < 0.001). However, when we pooled data from studies of infants up to 2 years of age, an increase of 1 kg in BW was associated with a 34% greater risk of ever or current AD (OR, 1.34; 95% CI, 1.08-1.68; P = .009; I2 = 89%; τ2 < 0.001). This effect remained significant when we conducted a sensitivity analysis excluding data for the 2-year-old outcomes from Egeberg et al
      • Egeberg A.
      • Andersen Y.M.
      • Gislason G.
      • Skov L.
      • Thyssen J.P.
      Neonatal risk factors of atopic dermatitis in Denmark - Results from a nationwide register-based study.
      (OR, 1.41; 95% CI, 1.11-1.78; P = .005; I2 = 61%; τ2 < 0.001). Six studies, detailed in Table E2, were unable to be included in meta-analyses of the relationship between BW and AD because there was insufficient information available to calculate the BW exposure,
      • Agosti M.
      • Vegni C.
      • Gangi S.
      • Benedetti V.
      • Marini A.
      Allergic manifestations in very low-birthweight infants: a 6-year follow-up.
      • Sebok B.
      • Schneider I.
      • Harangi F.
      Familiar and environmental factors influencing atopic dermatitis in the childhood.
      because birth length rather than BW was assessed as the exposure,
      • Bisgaard H.
      • Halkjaer L.B.
      • Hinge R.
      • Giwercman C.
      • Palmer C.
      • Silveira L.
      • et al.
      Risk analysis of early childhood eczema.
      and because ORs for BW categories could not be calculated for 3 studies that reported only BW values or z scores for AD-positive and AD-negative groups.
      • Kramer M.S.
      • Guo T.
      • Platt R.W.
      • Sevkovskaya Z.
      • Dzikovich I.
      • Collet J.P.
      • et al.
      Does previous infection protect against atopic eczema and recurrent wheeze in infancy?.
      • AlMakoshi A.
      • Ellahi A.
      • Sallout B.
      • Devereux G.
      • Turner S.
      Fetal growth trajectory and risk for eczema in a Saudi population.
      • Belderbos M.E.
      • Knol E.F.
      • Houben M.L.
      • van Bleek G.M.
      • Wilbrink B.
      • Kimpen J.L.
      • et al.
      Low neonatal Toll-like receptor 4-mediated interleukin-10 production is associated with subsequent atopic dermatitis.
      Only one study investigating the relationship between BW corrected for gestational age and ever AD up to adulthood was identified,
      • Hesselmar B.
      • Dahlgren J.
      • Wennergren G.
      • Aberg N.
      • Albertsson-Wiklan K.
      Born small for gestational age: relation to future allergy and asthma.
      which did not permit meta-analysis of this outcome. We were also unable to perform a meta-analysis of the relationship between BW corrected for gestational age and current AD in adulthood because only 2 independent studies reported this exposure and outcome.
      • Hesselmar B.
      • Dahlgren J.
      • Wennergren G.
      • Aberg N.
      • Albertsson-Wiklan K.
      Born small for gestational age: relation to future allergy and asthma.
      • Steffensen F.H.
      • Sorensen H.T.
      • Gillman M.W.
      • Rothman K.J.
      • Sabroe S.
      • Fischer P.
      • et al.
      Low birth weight and preterm delivery as risk factors for asthma and atopic dermatitis in young adult males.
      Figure thumbnail gr5
      Fig 5Meta-analysis of the relationship between BW corrected for gestational age and the risk of current allergic dermatitis in childhood. Data for each study are shown for each BW category reported. n/N, Number of cases and total population within each BW category; OR (95% CI), ORs and 95% CIs relative to the lowest BW group.

       Risk of bias in included studies

      Most of the included studies were assessed as methodologically sound and of high quality across most domains, although few satisfied all of the quality assessment criteria. Common areas in which cohort studies were assessed as low quality included incomplete follow-up (<80%) and a lack of description of any strategies to address incompleteness (see Table E3). For almost all included studies, BW groups were all derived from the same populations and were assessed consistently. Most studies used recorded or directly assessed fetal or birth size as their exposure measure. One used an estimated BW obtained by using linear regression of neonatal weight z scores rather than direct measures of BW taken by midwives based on the lower variability of the former measure,
      • Sevelsted A.
      • Bisgaard H.
      Neonatal size in term children is associated with asthma at age 7, but not with atopic dermatitis or allergic sensitization.
      but this approach will introduce errors if neonatal growth is the variable. Others were limited by use of parent-recalled BW as the exposure measure, although in a recent meta-analysis recalled BW and recorded BW were strongly correlated (R = 0.90; 95% CI, 0.87-0.93), regardless of time since birth.
      • Shenkin S.D.
      • Zhang M.G.
      • Der G.
      • Mathur S.
      • Mina T.H.
      • Reynolds R.M.
      Validity of recalled v. recorded birth weight: a systematic review and meta-analysis.
      Recalled BW was estimated as accurate to within approximately 100 g, with the best accuracy for high-income countries,
      • Shenkin S.D.
      • Zhang M.G.
      • Der G.
      • Mathur S.
      • Mina T.H.
      • Reynolds R.M.
      Validity of recalled v. recorded birth weight: a systematic review and meta-analysis.
      where most of the studies of BW and allergic disease in our meta-analyses were conducted. Most of the studies assessed allergic diseases at ages that should capture the majority of cases of each disease.

      Discussion

      Our meta-analyses suggest that restricted fetal growth might protect against some allergic diseases in human subjects but that fetal growth has differential effects on different allergic diseases and that these might vary with age. Increasing BW, corrected for confounders, including gestational age at delivery, was associated with greater ORs for FA and AD, but not AR, in children. Protection against at least some allergic diseases in subjects whose growth was restricted before birth is broadly consistent with outcomes from animal studies showing protection against inflammatory responses after allergic sensitization in rats and sheep with IUGR at ages equivalent to adolescence in human subjects.
      • Landgraf M.A.
      • Landgraf R.G.
      • Silva R.C.
      • Semedo P.
      • Camara N.O.
      • Fortes Z.B.
      Intrauterine undernourishment alters TH1/TH2 cytokine balance and attenuates lung allergic inflammation in wistar rats.
      • Wooldridge A.L.
      • Bischof R.J.
      • Meeusen E.N.
      • Liu H.
      • Heinemann G.K.
      • Hunter D.S.
      • et al.
      Placental restriction of fetal growth reduces cutaneous responses to antigen after sensitization in sheep.
      The limited evidence available from the literature suggests that these associations between fetal growth and allergic diseases are not present in human adults but requires further investigation.
      The risk of FA increased with increasing BW, with a greater than 40% (95% CI, 4% to 99%) greater OR for each 1-kg increase in BW. The data used in this meta-analysis are mostly from studies of infants aged 2 years or less,
      • Hikino S.
      • Nakayama H.
      • Yamamoto J.
      • Kinukawa N.
      • Sakamoto M.
      • Hara T.
      Food allergy and atopic dermatitis in low birthweight infants during early childhood.
      • Gabet S.
      • Just J.
      • Couderc R.
      • Seta N.
      • Momas I.
      Allergic sensitisation in early childhood: Patterns and related factors in PARIS birth cohort.
      • Metsala J.
      • Lundqvist A.
      • Kaila M.
      • Gissler M.
      • Klaukka T.
      • Virtanen S.M.
      Maternal and perinatal characteristics and the risk of cow's milk allergy in infants up to 2 years of age: a case-control study nested in the Finnish population.
      with older children aged up to 10 years
      • Metsala J.
      • Lundqvist A.
      • Kaila M.
      • Gissler M.
      • Klaukka T.
      • Virtanen S.M.
      Maternal and perinatal characteristics and the risk of cow's milk allergy in infants up to 2 years of age: a case-control study nested in the Finnish population.
      making up less than 12% of the total number included in meta-analysis. Interestingly, BW was not a significant predictor of FA in children aged 3 to 10 years
      • Metsala J.
      • Lundqvist A.
      • Kaila M.
      • Gissler M.
      • Klaukka T.
      • Virtanen S.M.
      Maternal and perinatal characteristics and the risk of cow's milk allergy in infants up to 2 years of age: a case-control study nested in the Finnish population.
      or within the 3-year-old cohort reported by Hikino et al.
      • Hikino S.
      • Nakayama H.
      • Yamamoto J.
      • Kinukawa N.
      • Sakamoto M.
      • Hara T.
      Food allergy and atopic dermatitis in low birthweight infants during early childhood.
      The majority of the included studies recorded allergy to multiple food allergens,
      • Hikino S.
      • Nakayama H.
      • Yamamoto J.
      • Kinukawa N.
      • Sakamoto M.
      • Hara T.
      Food allergy and atopic dermatitis in low birthweight infants during early childhood.
      • Gabet S.
      • Just J.
      • Couderc R.
      • Seta N.
      • Momas I.
      Allergic sensitisation in early childhood: Patterns and related factors in PARIS birth cohort.
      • Renz-Polster H.
      • David M.R.
      • Buist A.S.
      • Vollmer W.M.
      • O'Connor E.A.
      • Frazier E.A.
      • et al.
      Caesarean section delivery and the risk of allergic disorders in childhood.
      although for 2 articles, the outcomes were parental report or medical record of physician's diagnosis of FA, which did not specify the allergen or allergens.
      • Hikino S.
      • Nakayama H.
      • Yamamoto J.
      • Kinukawa N.
      • Sakamoto M.
      • Hara T.
      Food allergy and atopic dermatitis in low birthweight infants during early childhood.
      • Renz-Polster H.
      • David M.R.
      • Buist A.S.
      • Vollmer W.M.
      • O'Connor E.A.
      • Frazier E.A.
      • et al.
      Caesarean section delivery and the risk of allergic disorders in childhood.
      Gabet et al
      • Gabet S.
      • Just J.
      • Couderc R.
      • Seta N.
      • Momas I.
      Allergic sensitisation in early childhood: Patterns and related factors in PARIS birth cohort.
      defined FA based on at least 1 increased IgE level specific to a food allergen, most commonly cow's milk and egg white, which was not confirmed by food challenge in this cohort of infants. The article by Metsala et al,
      • Metsala J.
      • Lundqvist A.
      • Kaila M.
      • Gissler M.
      • Klaukka T.
      • Virtanen S.M.
      Maternal and perinatal characteristics and the risk of cow's milk allergy in infants up to 2 years of age: a case-control study nested in the Finnish population.
      which also reported increasing ORs for FA with increasing BW, used a reimbursement code specific for a physician's diagnosis of cow's milk allergy. Allergies against cow's milk and egg white are usually transient, resolving with age in 70% to 90% of children with cow's milk allergy and more than half of children with egg allergy in comparison with peanut allergy, which only resolves with age in 10% to 20% of children.
      • Sanchez-Garcia S.
      • Cipriani F.
      • Ricci G.
      Food Allergy in childhood: phenotypes, prevention and treatment.
      The available evidence for effects of BW on FA risk is thus strongest for transient FA and in infancy. It is possible that effects of BW on FA differ between transient and persistent FA or between specific allergens, and additional studies with robust diagnosis of FA, ideally confirmed by food challenge, are needed in older children to test these hypotheses.
      Similar to relationships between BW and FA, the risk of AD increased with increasing BW, particularly in infants. The OR for a child ever having received a diagnosis of AD increased by 17% (95% CI, 4% to 32%) for each 1-kg increase in BW, and the effect was even stronger in infancy because the OR for diagnosis of AD within the first 2 years of life increased by 35% (95% CI, 7% to 68%) for each 1-kg increase in BW. Variable rates of AD across the included studies, ranging from 1.0% to 43.5%, might reflect differences in environmental exposures or genetic susceptibility.
      • Vaughn A.R.
      • Sivamani R.K.
      • Lio P.A.
      • Shi V.Y.
      Paternal vs. maternal factors in childhood atopic dermatitis.
      Despite this variability in AD rates across the 30 studies included in these meta-analyses, the relationships between BW and the adjusted OR for AD were remarkably consistent within each meta-analysis of this outcome in children and infants. This is particularly interesting given the report from the ISAAC phase III studies that low BW, defined as BW of less than 2.5 kg and unadjusted for gestational age, was associated with lower risks of AD within affluent countries but not in nonaffluent countries.
      • Mitchell E.A.
      • Clayton T.
      • Garcia-Marcos L.
      • Pearce N.
      • Foliaki S.
      • Wong G.
      Birthweight and the risk of atopic diseases: the ISAAC Phase III study.
      Despite data from 2 large studies contributing heavily to the overall analysis, the results remained consistent in sensitivity analyses excluding these studies. Unfortunately, insufficient studies were available to permit a meta-analysis of associations between BW and AD in adulthood.
      In contrast to associations observed between BW and both FA and AD in childhood, a meta-analysis of the human literature did not support a relationship between BW and AR in children or adults. AR or hay fever generally emerges in children aged at least 3 years in contrast to FA, which usually emerges in infancy and resolves in most children.
      • Sanchez-Garcia S.
      • Cipriani F.
      • Ricci G.
      Food Allergy in childhood: phenotypes, prevention and treatment.
      • Spergel J.M.
      Epidemiology of atopic dermatitis and atopic march in children.
      • Lau S.
      • Matricardi P.M.
      • Wahn U.
      • Lee Y.A.
      • Keil T.
      Allergy and atopy from infancy to adulthood: Messages from the German birth cohort MAS.
      We suggest 3 possible explanations for differential relationships between BW and allergic diseases at different ages and between different allergic diseases, if indeed these are not simply a reflection of few studies having been performed in older children and adults and therefore limited power to detect effects at these ages.
      First, it is possible that low BW protects against allergic diseases with onset in early childhood because of global downregulation of the immune system in infants with IUGR, which normalizes with aging and therefore does not reduce rates of sensitization and allergic disease onset in later childhood or in adults. Consistent with the hypothesis that IUGR downregulates immune function in infants and early childhood, thymus size is negatively correlated with fetal weight at a given gestational age,
      • Diemert A.
      • Hartwig I.
      • Pagenkemper M.
      • Mehnert R.
      • Hansen G.
      • Tolosa E.
      • et al.
      Fetal thymus size in human pregnancies reveals inverse association with regulatory T cell frequencies in cord blood.
      lower in fetuses with IUGR than control fetuses throughout the second half of gestation, and lower in fetuses with IUGR with abnormal Doppler flow compared with fetuses IUGR with normal flow.
      • Ekin A.
      • Gezer C.
      • Taner C.E.
      • Solmaz U.
      • Gezer N.S.
      • Ozeren M.
      Prognostic value of fetal thymus size in intrauterine growth restriction.
      Data from cohorts in Gambia, where fetal (seasonal) undernutrition decreases BW by 200 to 300 g, suggests immune responses to infection remain impaired because these subjects have greater rates of death from infection as young adults.
      • Moore S.E.
      • Cole T.J.
      • Collinson A.C.
      • Poskitt E.M.
      • McGregor I.A.
      • Prentice A.M.
      Prenatal or early postnatal events predict infectious deaths in young adulthood in rural Africa.
      Second, IUGR might result in a transient shift in the immune balance to favor a TH1 rather than TH2 response. Consistent with this hypothesis, relationships between immune balance at birth and later allergy have been demonstrated, as have relationships between BW and immune cell types at birth. Challenges of immune cells from cord blood induced greater expression of the transcription factor GATA-3, which promotes TH2 cytokine gene expression, in children with allergic diseases or IgE sensitization by 2 years compared with those who did not have allergy or sensitization.
      • Marschan E.
      • Honkanen J.
      • Kukkonen K.
      • Kuitunen M.
      • Savilahti E.
      • Vaarala O.
      Increased activation of GATA-3, IL-2 and IL-5 of cord blood mononuclear cells in infants with IgE sensitization.
      Cord blood regulatory T (Treg) cell concentrations are negatively correlated with fetal weight, and therefore greater Treg cell numbers might downregulate inflammatory responses in low-BW infants.
      • Diemert A.
      • Hartwig I.
      • Pagenkemper M.
      • Mehnert R.
      • Hansen G.
      • Tolosa E.
      • et al.
      Fetal thymus size in human pregnancies reveals inverse association with regulatory T cell frequencies in cord blood.
      • Ekin A.
      • Gezer C.
      • Taner C.E.
      • Solmaz U.
      • Gezer N.S.
      • Ozeren M.
      Prognostic value of fetal thymus size in intrauterine growth restriction.
      Consistent with this hypothesis, Treg cell numbers in cord blood negatively predicted the risks for AD and increased IgE levels to food allergens in children in the first year of life.
      • Hinz D.
      • Bauer M.
      • Roder S.
      • Olek S.
      • Huehn J.
      • Sack U.
      • et al.
      Cord blood Tregs with stable FOXP3 expression are influenced by prenatal environment and associated with atopic dermatitis at the age of one year.
      However, there is some evidence for protection against allergic disease in adults, such as the report of protection against AR in Swedish conscripts born small for gestational age,
      • Braback L.
      • Hedberg A.
      Perinatal risk factors for atopic disease in conscripts.
      suggesting that protective effects of IUGR against allergy might not be transient.
      Finally, IUGR might protect against less severe phenotypes of FA and AD, which usually resolve with aging, but not protect against the more severe and persistent phenotypes of these diseases.
      • Sanchez-Garcia S.
      • Cipriani F.
      • Ricci G.
      Food Allergy in childhood: phenotypes, prevention and treatment.
      • Spergel J.M.
      Epidemiology of atopic dermatitis and atopic march in children.
      • Lau S.
      • Matricardi P.M.
      • Wahn U.
      • Lee Y.A.
      • Keil T.
      Allergy and atopy from infancy to adulthood: Messages from the German birth cohort MAS.
      The latter hypothesis could explain the lack of relationship between BW and AD in young adult men at conscription, although these results come from a single study.
      • Steffensen F.H.
      • Sorensen H.T.
      • Gillman M.W.
      • Rothman K.J.
      • Sabroe S.
      • Fischer P.
      • et al.
      Low birth weight and preterm delivery as risk factors for asthma and atopic dermatitis in young adult males.
      If low BW is protective only against less persistent allergic phenotypes, we would also expect that FA in older children, which reflects nonresolving phenotypes, would not be associated with BW. Given the limited data available in older children, adolescents, and adults, additional studies in cohorts with known size and gestational age at birth are needed to resolve how and when IUGR alters immune development and susceptibility to allergic disease.
      The strengths of this systematic review include extensive searching of multiple databases, as well as literature cited in included studies and relevant reviews. We only included studies in which birth or fetal size measures were corrected for gestational age or in which only term-born children were included to enable the effects of fetal growth to be differentiated from those of preterm birth. Gestational age could otherwise confound assessments of relationships between fetal growth and allergic diseases because there is some evidence that premature birth is protective for allergic diseases.
      • Pekkanen J.
      • Xu B.
      • Järvelin M.R.
      Gestational age and occurrence of atopy at age 31—a prospective birth cohort study in Finland.
      Somewhat contradicting this theory, the rates of allergic diseases in the preterm-born studies that met our inclusion criteria were within (AD, 7.2% to 19.1%)
      • Agosti M.
      • Vegni C.
      • Gangi S.
      • Benedetti V.
      • Marini A.
      Allergic manifestations in very low-birthweight infants: a 6-year follow-up.
      • Barbarot S.
      • Gras-Leguen C.
      • Colas H.
      • Garrot E.
      • Darmaun D.
      • Larroque B.
      • et al.
      Lower risk of atopic dermatitis among infants born extremely preterm compared with higher gestational age.
      or greater than (AR, 30.1%)
      • Siltanen M.
      • Wehkalampi K.
      • Hovi P.
      • Eriksson J.G.
      • Strang-Karlsson S.
      • Jarvenpaa A.L.
      • et al.
      Preterm birth reduces the incidence of atopy in adulthood.
      the ranges reported in term-born groups and whole populations (AD, 1.0% to 43.5%; AR, 4.5% to 23.7%).
      Our outcome measures were restricted to a direct or reported physician's diagnosis of allergic disease or based on positive responses to questions about specific symptoms that have been clinically validated as indicating AR or AD. The majority of the latter studies used or adapted the ISAAC questionnaires.
      The International Study of Asthma and Allergies in Childhood (ISAAC) Steering Committee. Worldwide variation in prevalence of symptoms of asthma, allergic rhinoconjunctivitis, and atopic eczema: ISAAC.
      • Williams H.
      • Robertson C.
      • Stewart A.
      • Ait-Khaled N.
      • Anabwani G.
      • Anderson R.
      • et al.
      Worldwide variations in the prevalence of symptoms of atopic eczema in the International Study of Asthma and Allergies in Childhood.
      The International Study of Asthma and Allergies in Childhood (ISAAC) Steering Committee. Worldwide variations in the prevalence of asthma symptoms: the International Study of Asthma and Allergies in Childhood (ISAAC).
      One study in Danish children
      • Linneberg A.
      • Simonsen J.B.
      • Petersen J.
      • Stensballe L.G.
      • Benn C.S.
      Differential effects of risk factors on infant wheeze and atopic dermatitis emphasize a different etiology.
      used criteria developed for telephone interview–based diagnosis of AD in young children that were validated by a dermatologist's examination.
      • Benn C.S.
      • Benfeldt E.
      • Andersen P.K.
      • Olesen A.B.
      • Melbye M.
      • Bjorksten B.
      Atopic dermatitis in young children: diagnostic criteria for use in epidemiological studies based on telephone interviews.
      A reported physician's diagnosis has also been validated as consistent with direct diagnosis of allergic diseases. When validated against dermatology practice records of physicians' diagnoses, caregiver report (childhood) or self-report (adult) of a previous physician's diagnosis of recent (within 12 months) AD and caregiver report of childhood AD history had high specificity (≥0.89) and fair sensitivity (0.70-0.83).
      • Silverberg J.I.
      • Patel N.
      • Immaneni S.
      • Rusniak B.
      • Silverberg N.B.
      • Debashis R.
      • et al.
      Assessment of atopic dermatitis using self-report and caregiver report: a multicentre validation study.
      Adult report of AD history also had high specificity (0.97), although sensitivity was low (0.43).
      • Silverberg J.I.
      • Patel N.
      • Immaneni S.
      • Rusniak B.
      • Silverberg N.B.
      • Debashis R.
      • et al.
      Assessment of atopic dermatitis using self-report and caregiver report: a multicentre validation study.
      Although this means that studies requiring recall of a history of AD diagnosis in adults likely underestimate overall AD rates, we do not expect that recall of AD history would differ with BW. We did not accept self-report of allergy without clinical diagnosis as a valid outcome. Self-report is likely to be particularly unreliable for studies of FA, in which self-reported reactions to cow's milk, wheat, shrimp, and egg have little relationship with positive skin prick test responses to the same allergens in adults.
      • Woods R.K.
      • Stoney R.M.
      • Raven J.
      • Walters E.H.
      • Abramson M.
      • Thien F.C.
      Reported adverse food reactions overestimate true food allergy in the community.
      Insufficient studies were available to permit meta-analyses of associations between BW and AA at any age or FA and AD in adults. Therefore our findings of age- and disease-specific associations between BW and allergic disease require additional studies to explore the reasons for the observed differences. To enable further investigation of effects of fetal growth on allergic diseases, we recommend that all such studies should adjust for gestational age and child sex and ideally consider BW relative to comparable population data or parental height.
      Another limitation in our study arose because actual BW exposure data were not available for all studies, and therefore the exposure BW values used in analyses were estimates in many cases. We used the midpoint BW as the estimated BW exposure for categories in which upper and lower boundaries were provided and did not take the variability associated with these estimates into account in the analyses.
      There is also measurement variability in estimated BW for the lower and upper categories of BW (because only 1 end point is stated). For these categories, we estimated the exposure BW from the quantiles of the normal distribution corresponding to the proportion of BW distribution that fell into the respective category. These were based on estimated means and SDs in many cases, and these also introduce another source of variation. These assumptions are not testable, and therefore the uncertainty that these approximations introduce into the results is not captured in the CIs reported with pooled point estimates. Therefore we recommend that authors provide information regarding actual BW within each BW category, which would improve accuracy of meta-analyses of outcomes across studies. There was considerable heterogeneity evident across studies on the basis of I2 values, particularly for AD, although the τ2 values in each analysis suggested minimal spread in the study-specific effects. Taken together, this suggests that investigation of the shape of the relationship between BW and the outcomes considered here is warranted in planning of future studies.
      In conclusion, we have conducted the first systematic review and meta-analysis of associations between gestational age-adjusted or term BW and allergic disease in human subjects. Increases in BW were associated with greater ORs for FA in children and AD in children and infants but not AR in older children or adults.
      Key messages
      • Greater BW corrected for gestational age is associated with increased risk of FA and allergic dermatitis, but not AR, in childhood.
      • Additional studies are needed to determine underlying mechanisms and whether the effects of prenatal growth on allergic disease persist.
      We thank Mick Draper of the University of Adelaide library for expert assistance with scripting database searches. We also thank those authors who provided additional information to enable meta-analysis of outcomes.

      Supplementary data

      References

        • World Allergy Organization
        World Allergy Organization (WAO) white book on allergy: update 2013.
        World Allergy Organization, Milwaukee (WI)2013
        • Ferreira M.A.
        • Vonk J.M.
        • Baurecht H.
        • Marenholz I.
        • Tian C.
        • Hoffman J.D.
        • et al.
        Shared genetic origin of asthma, hay fever and eczema elucidates allergic disease biology.
        Nat Genet. 2017; 49: 1752-1757
        • Mullins R.J.
        Paediatric food allergy trends in a community-based specialist allergy practice, 1995-2006.
        Med J Aust. 2007; 186: 618-621
        • Galassi C.
        • De Sario M.
        • Biggeri A.
        • Bisanti L.
        • Chellini E.
        • Ciccone G.
        • et al.
        Changes in prevalence of asthma and allergies among children and adolescents in Italy: 1994-2002.
        Pediatrics. 2006; 117: 34-42
        • Poulos L.M.
        • Waters A.M.
        • Correll P.K.
        • Loblay R.H.
        • Marks G.B.
        Trends in hospitalizations for anaphylaxis, angioedema, and urticaria in Australia, 1993-1994 to 2004-2005.
        J Allergy Clin Immunol. 2007; 120: 878-884
        • Grieger J.A.
        • Clifton V.L.
        • Tuck A.R.
        • Wooldridge A.L.
        • Robertson S.A.
        • Gatford K.L.
        In utero programming of allergic susceptibility.
        Int Arch Allergy Immunol. 2016; 169: 80-92
        • Landgraf M.A.
        • Landgraf R.G.
        • Jancar S.
        • Fortes Z.B.
        Influence of age on the development of immunological lung response in intrauterine undernourishment.
        Nutrition. 2008; 24: 262-269
        • Landgraf M.A.
        • Landgraf R.G.
        • Silva R.C.
        • Semedo P.
        • Camara N.O.
        • Fortes Z.B.
        Intrauterine undernourishment alters TH1/TH2 cytokine balance and attenuates lung allergic inflammation in wistar rats.
        Cell Physiol Biochem. 2012; 30: 552-562
        • Wooldridge A.L.
        • Bischof R.J.
        • Meeusen E.N.
        • Liu H.
        • Heinemann G.K.
        • Hunter D.S.
        • et al.
        Placental restriction of fetal growth reduces cutaneous responses to antigen after sensitization in sheep.
        Am J Physiol Regul Integr Comp Physiol. 2014; 306: R441-R446
        • Hales C.N.
        • Barker D.J.P.
        • Clark P.M.S.
        • Cox L.J.
        • Fall C.
        • Osmond C.
        • et al.
        Fetal and infant growth and impaired glucose tolerance at age 64.
        BMJ. 1991; 303: 1019-1022
        • Roseboom T.J.
        • van der Meulen J.H.P.
        • Ravelli A.C.J.
        • Osmond C.
        • Barker D.J.P.
        • Bleker O.P.
        Effects of prenatal exposure to the Dutch famine on adult disease in later life: an overview.
        Mol Cell Endocrinol. 2001; 185: 93-98
        • Spracklen C.N.
        • Wallace R.B.
        • Sealy-Jefferson S.
        • Robinson J.G.
        • Freudenheim J.L.
        • Wellons M.F.
        • et al.
        Birth weight and subsequent risk of cancer.
        Cancer Epidemiol. 2014; 38: 538-543
        • Mitchell E.A.
        • Clayton T.
        • Garcia-Marcos L.
        • Pearce N.
        • Foliaki S.
        • Wong G.
        Birthweight and the risk of atopic diseases: the ISAAC Phase III study.
        Pediatr Allergy Immunol. 2014; 25: 264-270
        • Panduru M.
        • Salavastru C.M.
        • Panduru N.M.
        • Tiplica G.S.
        Birth weight and atopic dermatitis: systematic review and meta-analyis.
        Acta Dermatovenerol Croat. 2014; 22: 91-96
        • Bleker O.P.
        • Buimer M.
        • van der Post J.A.M.
        • van der Veen F.
        Ted (G. J.) Kloosterman: on intrauterine growth. The significance of prenatal care. Studies on birth weight, placental weight and placental index.
        Placenta. 2006; 27: 1052-1054
        • Ergaz Z.
        • Avgil M.
        • Ornoy A.
        Intrauterine growth restriction—etiology and consequences: what do we know about the human situation and experimental animal models?.
        Reprod Toxicol. 2005; 20: 301-322
        • Pekkanen J.
        • Xu B.
        • Järvelin M.R.
        Gestational age and occurrence of atopy at age 31—a prospective birth cohort study in Finland.
        Clin Exp Allergy. 2001; 31: 95-102
        • Lundholm C.
        • Ortqvist A.K.
        • Lichtenstein P.
        • Cnattingius S.
        • Almqvist C.
        Impaired fetal growth decreases the risk of childhood atopic eczema: a Swedish twin study.
        Clin Exp Allergy. 2010; 40: 1044-1053
        • Wooldridge A.L.
        • McMillan M.
        • Marshall H.S.
        • Gatford K.L.
        Relationship between birth weight or fetal growth rate and postnatal allergy: a systematic review protocol.
        JBI Database Syst Rev Implement Rep. 2016; 14: 11-20
        • Moola S.
        • Munn Z.
        • Tufanaru C.
        • Aromataris E.
        • Sears K.
        • Sfetcu R.
        • et al.
        Aromataris E. Munn Z. Chapter 7: systematic reviews of etiology and risk. The Joanna Briggs Institute, 2017 (Available at) (Accessed January 15, 2018)
        • Harris P.A.
        • Taylor R.
        • Thielke R.
        • Payne J.
        • Gonzalez N.
        • Conde J.G.
        Research electronic data capture (REDCap)—a metadata-driven methodology and workflow process for providing translational research informatics support.
        J Biomed Inform. 2009; 42: 377-381
        • Goksor E.
        • Alm B.
        • Pettersson R.
        • Mollborg P.
        • Erdes L.
        • Aberg N.
        • et al.
        Early fish introduction and neonatal antibiotics affect the risk of asthma into school age.
        Pediatr Allergy Immunol. 2013; 24: 339-344
        • Loid P.
        • Goksor E.
        • Alm B.
        • Pettersson R.
        • Mollborg P.
        • Erdes L.
        • et al.
        A persistently high body mass index increases the risk of atopic asthma at school age.
        Acta Paediatr. 2015; 104: 707-712
        • Anderson E.L.
        • Fraser A.
        • Martin R.M.
        • Kramer M.S.
        • Oken E.
        • Patel R.
        • et al.
        Associations of postnatal growth with asthma and atopy: the PROBIT study.
        Pediatr Allergy Immunol. 2013; 24: 122-130
        • Kramer M.S.
        • Guo T.
        • Platt R.W.
        • Sevkovskaya Z.
        • Dzikovich I.
        • Collet J.P.
        • et al.
        Does previous infection protect against atopic eczema and recurrent wheeze in infancy?.
        Clin Exp Allergy. 2004; 34: 753-756
        • Bisgaard H.
        • Halkjaer L.B.
        • Hinge R.
        • Giwercman C.
        • Palmer C.
        • Silveira L.
        • et al.
        Risk analysis of early childhood eczema.
        J Allergy Clin Immunol. 2009; 123: 1355-1360.e5
        • Sevelsted A.
        • Bisgaard H.
        Neonatal size in term children is associated with asthma at age 7, but not with atopic dermatitis or allergic sensitization.
        Allergy. 2012; 67: 670-675
        • Bager P.
        • Melbye M.
        • Rostgaard K.
        • Benn C.S.
        • Westergaard T.
        Mode of delivery and risk of allergic rhinitis and asthma.
        J Allergy Clin Immunol. 2003; 111: 51-56
        • Linneberg A.
        • Simonsen J.B.
        • Petersen J.
        • Stensballe L.G.
        • Benn C.S.
        Differential effects of risk factors on infant wheeze and atopic dermatitis emphasize a different etiology.
        J Allergy Clin Immunol. 2006; 117: 184-189
        • Hikino S.
        • Nakayama H.
        • Yamamoto J.
        • Kinukawa N.
        • Sakamoto M.
        • Hara T.
        Food allergy and atopic dermatitis in low birthweight infants during early childhood.
        Acta Paediatr. 2001; 90: 850-855
        • Miyake Y.
        • Tanaka K.
        Lack of relationship between birth conditions and allergic disorders in Japanese children aged 3 years.
        J Asthma. 2013; 50: 555-559
        • Orsini N.
        • Li R.
        • Wolk A.
        • Khudyakov P.
        • Spiegelman D.
        Meta-analysis for linear and nonlinear dose-response relations: examples, an evaluation of approximations, and software.
        Am J Epidemiol. 2012; 175: 66-73
        • Greenland S.
        • Longnecker M.P.
        Methods for trend estimation from summarized dose-response data, with applications to meta-analysis.
        Am J Epidemiol. 1992; 135: 1301-1309
        • Gabet S.
        • Just J.
        • Couderc R.
        • Seta N.
        • Momas I.
        Allergic sensitisation in early childhood: Patterns and related factors in PARIS birth cohort.
        Int J Hyg Environ Health. 2016; 219: 792-800
        • Metsala J.
        • Lundqvist A.
        • Kaila M.
        • Gissler M.
        • Klaukka T.
        • Virtanen S.M.
        Maternal and perinatal characteristics and the risk of cow's milk allergy in infants up to 2 years of age: a case-control study nested in the Finnish population.
        Am J Epidemiol. 2010; 171: 1310-1316
        • Renz-Polster H.
        • David M.R.
        • Buist A.S.
        • Vollmer W.M.
        • O'Connor E.A.
        • Frazier E.A.
        • et al.
        Caesarean section delivery and the risk of allergic disorders in childhood.
        Clin Exp Allergy. 2005; 35: 1466-1472
        • Lin M.H.
        • Hsieh C.J.
        • Caffrey J.L.
        • Lin Y.S.
        • Wang I.J.
        • Ho W.C.
        • et al.
        Fetal growth, obesity, and atopic disorders in adolescence: a retrospective birth cohort study.
        Paediatr Perinat Epidemiol. 2015; 29: 472-479
        • Braback L.
        • Hedberg A.
        Perinatal risk factors for atopic disease in conscripts.
        Clin Exp Allergy. 1998; 28: 936-942
        • Ericsson C.H.
        • Svartengren M.
        • Mossberg B.
        • Pedersen N.
        • Camner P.
        Bronchial reactivity and allergy-promoting factors in monozygotic twins discordant for allergic rhinitis.
        Ann Allergy. 1991; 67: 53-59
        • Olesen A.B.
        • Ellingsen A.R.
        • Olesen H.
        • Juul S.
        • Thestrup-Pedersen K.
        Atopic dermatitis and birth factors: historical follow up by record linkage.
        BMJ. 1997; 314: 1003-1008
        • Egeberg A.
        • Andersen Y.M.
        • Gislason G.
        • Skov L.
        • Thyssen J.P.
        Neonatal risk factors of atopic dermatitis in Denmark - Results from a nationwide register-based study.
        Pediatr Allergy Immunol. 2016; 27: 368-374
        • Agosti M.
        • Vegni C.
        • Gangi S.
        • Benedetti V.
        • Marini A.
        Allergic manifestations in very low-birthweight infants: a 6-year follow-up.
        Acta Paediatr Suppl. 2003; 91: 44-47
        • Sebok B.
        • Schneider I.
        • Harangi F.
        Familiar and environmental factors influencing atopic dermatitis in the childhood.
        J Eur Acad Dermatol Venereol. 2006; 20: 418-422
        • AlMakoshi A.
        • Ellahi A.
        • Sallout B.
        • Devereux G.
        • Turner S.
        Fetal growth trajectory and risk for eczema in a Saudi population.
        Pediatr Allergy Immunol. 2015; 26: 811-816
        • Belderbos M.E.
        • Knol E.F.
        • Houben M.L.
        • van Bleek G.M.
        • Wilbrink B.
        • Kimpen J.L.
        • et al.
        Low neonatal Toll-like receptor 4-mediated interleukin-10 production is associated with subsequent atopic dermatitis.
        Clin Exp Allergy. 2012; 42: 66-75
        • Hesselmar B.
        • Dahlgren J.
        • Wennergren G.
        • Aberg N.
        • Albertsson-Wiklan K.
        Born small for gestational age: relation to future allergy and asthma.
        Acta Paediatr. 2002; 91: 992-994
        • Steffensen F.H.
        • Sorensen H.T.
        • Gillman M.W.
        • Rothman K.J.
        • Sabroe S.
        • Fischer P.
        • et al.
        Low birth weight and preterm delivery as risk factors for asthma and atopic dermatitis in young adult males.
        Epidemiology. 2000; 11: 185-188
        • Shenkin S.D.
        • Zhang M.G.
        • Der G.
        • Mathur S.
        • Mina T.H.
        • Reynolds R.M.
        Validity of recalled v. recorded birth weight: a systematic review and meta-analysis.
        J Dev Orig Health Dis. 2017; 8: 137-148
        • Sanchez-Garcia S.
        • Cipriani F.
        • Ricci G.
        Food Allergy in childhood: phenotypes, prevention and treatment.
        Pediatr Allergy Immunol. 2015; 26: 711-720
        • Vaughn A.R.
        • Sivamani R.K.
        • Lio P.A.
        • Shi V.Y.
        Paternal vs. maternal factors in childhood atopic dermatitis.
        Dermatitis. 2017; 28: 241-245
        • Spergel J.M.
        Epidemiology of atopic dermatitis and atopic march in children.
        Immunol Allergy Clin North Am. 2010; 30: 269-280
        • Lau S.
        • Matricardi P.M.
        • Wahn U.
        • Lee Y.A.
        • Keil T.
        Allergy and atopy from infancy to adulthood: Messages from the German birth cohort MAS.
        Ann Allergy Asthma Immunol. 2019; 122: 25-32
        • Diemert A.
        • Hartwig I.
        • Pagenkemper M.
        • Mehnert R.
        • Hansen G.
        • Tolosa E.
        • et al.
        Fetal thymus size in human pregnancies reveals inverse association with regulatory T cell frequencies in cord blood.
        J Reprod Immunol. 2016; 113: 76-82
        • Ekin A.
        • Gezer C.
        • Taner C.E.
        • Solmaz U.
        • Gezer N.S.
        • Ozeren M.
        Prognostic value of fetal thymus size in intrauterine growth restriction.
        J Ultrasound Med. 2016; 35: 511-517
        • Moore S.E.
        • Cole T.J.
        • Collinson A.C.
        • Poskitt E.M.
        • McGregor I.A.
        • Prentice A.M.
        Prenatal or early postnatal events predict infectious deaths in young adulthood in rural Africa.
        Int J Epidemiol. 1999; 28: 1088-1095
        • Marschan E.
        • Honkanen J.
        • Kukkonen K.
        • Kuitunen M.
        • Savilahti E.
        • Vaarala O.
        Increased activation of GATA-3, IL-2 and IL-5 of cord blood mononuclear cells in infants with IgE sensitization.
        Pediatr Allergy Immunol. 2008; 19: 132-139
        • Hinz D.
        • Bauer M.
        • Roder S.
        • Olek S.
        • Huehn J.
        • Sack U.
        • et al.
        Cord blood Tregs with stable FOXP3 expression are influenced by prenatal environment and associated with atopic dermatitis at the age of one year.
        Allergy. 2012; 67: 380-389
        • Barbarot S.
        • Gras-Leguen C.
        • Colas H.
        • Garrot E.
        • Darmaun D.
        • Larroque B.
        • et al.
        Lower risk of atopic dermatitis among infants born extremely preterm compared with higher gestational age.
        Br J Dermatol. 2013; 169: 1257-1264
        • Siltanen M.
        • Wehkalampi K.
        • Hovi P.
        • Eriksson J.G.
        • Strang-Karlsson S.
        • Jarvenpaa A.L.
        • et al.
        Preterm birth reduces the incidence of atopy in adulthood.
        J Allergy Clin Immunol. 2011; 127: 935-942
      1. The International Study of Asthma and Allergies in Childhood (ISAAC) Steering Committee. Worldwide variation in prevalence of symptoms of asthma, allergic rhinoconjunctivitis, and atopic eczema: ISAAC.
        Lancet. 1998; 351: 1225-1232
        • Williams H.
        • Robertson C.
        • Stewart A.
        • Ait-Khaled N.
        • Anabwani G.
        • Anderson R.
        • et al.
        Worldwide variations in the prevalence of symptoms of atopic eczema in the International Study of Asthma and Allergies in Childhood.
        J Allergy Clin Immunol. 1999; 103: 125-138
      2. The International Study of Asthma and Allergies in Childhood (ISAAC) Steering Committee. Worldwide variations in the prevalence of asthma symptoms: the International Study of Asthma and Allergies in Childhood (ISAAC).
        Eur Respir J. 1998; 12: 315-335
        • Benn C.S.
        • Benfeldt E.
        • Andersen P.K.
        • Olesen A.B.
        • Melbye M.
        • Bjorksten B.
        Atopic dermatitis in young children: diagnostic criteria for use in epidemiological studies based on telephone interviews.
        Acta Derm Venereol. 2003; 83: 347-350
        • Silverberg J.I.
        • Patel N.
        • Immaneni S.
        • Rusniak B.
        • Silverberg N.B.
        • Debashis R.
        • et al.
        Assessment of atopic dermatitis using self-report and caregiver report: a multicentre validation study.
        Br J Dermatol. 2015; 173: 1400-1404
        • Woods R.K.
        • Stoney R.M.
        • Raven J.
        • Walters E.H.
        • Abramson M.
        • Thien F.C.
        Reported adverse food reactions overestimate true food allergy in the community.
        Eur J Clin Nutr. 2002; 56: 31-36