Caesarean delivery and the risk of atopic dermatitis in children

Abstract

BackgroundCaesarean delivery (C‐section) may disrupt maternal‐infant microbial transfer and alter immune system development and subsequent risk for atopic dermatitis.ObjectiveInvestigate the association between C‐section and atopic dermatitis by age four and examine potential sources of bias in the relationship in a large cohort study.MethodsMaternal and child information was collected through Kaiser Permanente Northern California's (KPNC) integrated healthcare system. Data sources included electronic medical records, pharmacy databases, state birth records, and prospectively collected breastfeeding surveys. Children were eligible if they were born in a KPNC or contracting hospital between 2005 and 2014 and had continuous enrolment in the KPNC system for at least four years (n = 173 105). Modified Poisson regression with robust variance estimation was used to estimate the association between C‐section and atopic dermatitis overall and when stratified by demographic and labour and delivery characteristics.ResultsAlthough unadjusted analyses showed a positive association between C‐section and atopic dermatitis [RR(95%CI): 1.06(1.03, 1.10)], this effect was attenuated towards the null after adjustment [aRR(95%CI): 1.02(0.99, 1.05)]. In stratified analyses, there was evidence that C‐section increased atopic dermatitis risk among certain subgroups (eg firstborns, overweight/obese pre‐pregnancy BMI), but associations were weak. C‐section delivery conditions indicative of the least exposure to maternal microbiome (ie no labour, short interval between membrane rupture and delivery) showed no evidence of association with atopic dermatitis. Estimated associations were not strongly influenced by intrapartum antibiotics, breastfeeding, missing data, or familial factors.ConclusionCaesarean delivery was not associated with atopic dermatitis by age four in this large US cohort. This association did not appear to be biased by intrapartum antibiotics, breastfeeding behaviour, C‐section indication, missing covariates, or familial factors.

Prepregnancy BMI, gestational weight gain and risk of childhood atopic dermatitis: A systematic review and meta‐analysis

“…As shown in Figure 1, a total of 1597 records were identified: 263 from PubMed, 568 from Embase, 95 from Cochrane, 296 from Web of Science, and 375 from Scopus. According to the inclusion and exclusion criteria, 13 studies 14‐18,21‐28 were included in the research and 12 studies 15‐18,21‐28 were included in the meta‐analysis.…”

“…The main characteristics of the included studies are shown in Table 1 (the specific data are provided in the supporting information: Supplementary specific study characteristics). A total of 13 studies 14‐18,21‐28 with a total of 114 485 participants were included in this systematic review and meta‐analysis. Sample sizes ranged from 339 to 38 874 participants, and age at outcome ranged from 6 months to 17 years.…”

“…Sample sizes ranged from 339 to 38 874 participants, and age at outcome ranged from 6 months to 17 years. Among the included studies, 11 were cohort studies 14‐16,18,21‐28 and 2 were cross‐sectional studies 17,21 . The studies were performed in Chile (1 study 27 ), China (1 study 17 ), Denmark (1 study 15 ), Finland (1 study 14 ), Japan (2 studies 16,21 ), Sweden (2 studies 24,25 ), the UK (1 study 26 ), and the USA (4 studies 18,22,23,28 ).…”

“…Among the included studies, 11 were cohort studies 14‐16,18,21‐28 and 2 were cross‐sectional studies 17,21 . The studies were performed in Chile (1 study 27 ), China (1 study 17 ), Denmark (1 study 15 ), Finland (1 study 14 ), Japan (2 studies 16,21 ), Sweden (2 studies 24,25 ), the UK (1 study 26 ), and the USA (4 studies 18,22,23,28 ). For maternal BMI and GWG, 5 studies 17,24‐28 used medical records, and 8 studies 14‐18,21‐23 used self‐reports, questionnaires, or questions.…”

Aetiology and prevention of eczema

“…Epidemiological studies have not yet successfully explained the marked variations seen between and within countries in eczema prevalence. Hygiene hypothesis type exposures such as antibiotics or mode of birth do not always associate with eczema in the same way as for other atopic disorders 5 . Perhaps the leading area of investigation for eczema prevention at the moment is work on skin barrier integrity, informed by our understanding of the role of defects in epidermal barrier function in disease onset.…”