Prenatal and early-life antibiotic exposure and the risk of atopic dermatitis in children: A nationwide population-based cohort study.

Abstract

Atopic dermatitis (AD) contributes to substantial social and financial costs in public health care systems. Antibiotic exposure during pregnancy has been proposed as a risk factor, but findings remain inconsistent. The aim of this study was to investigate the association between prenatal antibiotic use and childhood AD. We performed a population-based cohort study using data collected from the Taiwan Maternal and Child Health Database from 2009 to 2016. Associations were determined using Cox proportional hazards model and were adjusted for several potential covariates, including maternal atopic disorders and gestational infections. Children with and without maternal predispositions of atopic diseases and postnatal antibiotic/acetaminophen exposures within 1 year were stratified to identify the subgroups at risk. A total of 1,288,343 mother-child pairs were identified and 39.5% received antibiotics prenatally. Maternal antibiotic use during pregnancy was slightly positively associated with childhood AD (aHR 1.04, 95% CI 1.03-1.05), especially in the first and second trimesters. An apparent dose-response pattern was observed with an 8% increased risk when the exposure was ≥5 courses prenatally (aHR 1.08, 95% CI 1.06-1.11). Subgroup analysis showed the positive association remained significant regardless of postnatal infant antibiotic use, but the risk attenuated to null in infants who were not exposed to acetaminophen (aHR 1.01, 95% CI 0.96-1.05). The associations were higher in children whose mothers were without AD compared to those whose mothers were with AD. In addition, postnatal antibiotic or acetaminophen exposure of infants was associated with an increased risk of developing AD after 1 year of age. Maternal antibiotic use during pregnancy was associated with an increased risk of childhood AD in a dose-related manner. Further research may be warranted to investigate this variable using a prospectively designed study, and also to examine whether or not this association is specifically related to pregnancy.