Impact of maternal obesity on inhaled corticosteroid use in childhood: a registry based analysis of first born children and a sibling pair analysis.

Background: It has been proposed that maternal obesity during pregnancy may increase the risk that the child develops allergic disease and asthma, although the mechanisms underpinning this relationship are currently unclear. We sought to assess if this association may be due to confounding by genetic or environmental risk factors that are common to maternal obesity and childhood asthma, using a sibling pair analysis.

Methods: The study population comprised a Swedish national cohort of term children born between 1992 and 2008 to native Swedish parents. Maternal body mass index (BMI) was measured at 8-10 weeks gestation. Unconditional logistic regression models were used to determine if maternal obesity was associated with increased risk of inhaled corticosteroid (ICS) in 431,718 first-born children, while adjusting for potential confounders. An age-matched discordant sib-pair analysis was performed, taking into account shared genetic and environmental risk factors.

Results: Maternal over-weight and obesity were associated with increased risk that the child would require ICS (for BMI≥35 kg/m(2), aOR = 1.30, 95%CI = 1.10-1.52 compared with normal weight mothers) in children aged 6-12 years. Similar effects were seen in younger children, but in children aged 13-16 years, maternal obesity (BMI≥30) was related to increased risk of ICS use in girls (aOR = 1.28, 95%CI = 1.07-1.53) but not boys (OR = 1.05, 95%CI = 0.87-1.26). The sib-pair analysis, which included 2,034 sib-pairs older than six years who were discordant for both ICS use and maternal BMI category, failed to find any evidence that increasing maternal weight was related to increased risk of ICS use.

Conclusion: Maternal obesity is associated with increased risk of childhood ICS use up to approximately 12 years of age, but only in girls after this age. These effects could not be confirmed in a sib pair analysis, suggesting either limited statistical power, or the effects of maternal BMI may be due to shared genetic or environmental risk factors.