Mode of delivery and asthma in childhood and adolescence: Findings from the Millennium Cohort Study

Abstract

Previous research has shown a link between caesarean section (CS) and asthma,1 citing a disruption of infant intestinal flora leading to development of allergens.2 However, other studies have contradicted these findings reporting no association between the mode of delivery and asthma, especially adolescent asthma.1 The primary aim of this study was to explore the association between the mode of delivery and asthma development at age 7, 11 or 14 years, persistent asthma and adolescent onset asthma using data from the UK Millennium Cohort Study. We hypothesised that if a link between CS and asthma existed due to disruption in the infant intestinal microflora, then an association between planned CS and asthma would be observed in the current study. This research used data collected at 9 months and 7, 11 and 14 years from the Millennium Cohort Study (MCS), a nationally representative longitudinal study of children born in the UK between 2000 and 2002. A full description of our study population and ethics can be found here: https://doi.org/10.5281/zenodo.7438158. Data on the mode of delivery were obtained when children were aged 9 months. The mode of delivery was categorized as ‘spontaneous vaginal delivery’, ‘assisted vaginal delivery’, ‘induced vaginal delivery’, ‘emergency CS’, ‘planned CS’ and ‘CS after induction of labour’. The variable was defined as such to assess the effect of any intervention on labour compared to none (i.e., spontaneous, non-assisted vaginal delivery). A detailed description can be found here: https://doi.org/10.5281/zenodo.7438158. At ages 7, 11 and 14 years, respondents were asked a series of questions on asthma and wheezing in the children. Questions were taken from the International Study of Asthma and Allergies in Childhood (ISAAC) core questionnaire. Primary outcomes of interest included ever asthma (asthma at age 7, 11 or 14 years), persistent asthma (asthma at ages 7, 11 and 14 years) and adolescent onset asthma (asthma at age 14 years, but not prior). Secondary outcome included parent-reported ‘wheezing in the last 12 months’ at age 7 and 11 years. We controlled for the following potential confounders at pre-pregnancy or during pregnancy: maternal age, maternal education, ethnicity, income quintile, maternal body mass index (BMI), maternal asthma, maternal smoking during pregnancy and maternal diabetes. Additionally, we adjusted for hypertensive disorders of pregnancy (HDP) and small for gestational age (SGA) in a separate model. A directed acyclic graph and description of potential confounders can be found here: https://doi.org/10.5281/zenodo.7438158. Data were analysed using SPSS 28.0. Univariable and multivariable logistic regression was used to examine the association between mode of delivery and each outcome. For each outcome, four logistic regression models were performed. First, crude analysis (model 1). Second, we adjusted for socioeconomic factors including maternal age, maternal education, ethnicity and income quintile (model 2). Third, we adjusted for maternal BMI, maternal asthma, maternal smoking during pregnancy and maternal diabetes (model 3). Finally, we controlled for HDP and SGA (model 4). Bonferroni correction was used when a result was statistically significant to adjust for multiple testing. A description of all sensitivity analyses conducted can be found here: https://doi.org/10.5281/zenodo.7438158. There were 18,213 singleton mother–child pairs with data on the mode of delivery included at baseline; 13,088 at age 7; 12,530 at age 11 and 11,032 at age 14. Mother and child characteristics are outlined in Table 1. In the crude logistic regression model, emergency CS was significantly associated with ever asthma at age 7 years (OR: 1.27; 95% CI: 1.06–1.52). The association did not materially change when adjusted for potential confounders. No other consistent significant associations were observed at age 7 for planned CS or other modes or delivery (Table 2). At age 11 years, emergency CS was associated with a 42% increase in odds of ever having asthma compared to spontaneous vaginal delivery in the crude model (OR: 1.42; 95% CI:1.19–1.68). After adjusting for potential confounders, the associations remained broadly similar. Induced vaginal delivery was associated with asthma at age 11 across all models (OR: 1.14, 95% CI:1.01–1.29); however, it did not remain significant after Bonferroni correction (i.e., result was not significant at (0.05/3) p < .01) (Table 2). At age 14 years, emergency CS was associated with a 27% increase in odds of having asthma (OR: 1.27, 95% CI:1.03–1.58). This remained unchanged in adjusted models. Similarly, CS after induction was associated with a 36% increase in the odds of having asthma (OR 1.36; 95% CI:1.09–1.70), with similar results observed in the adjusted models (Table 2). It is worth noting that while point estimates were greater than 1 in most instances suggesting an association cannot be ruled out: the ORs for the association between planned CS and asthma at 7, 11 and 14 years were relatively close to one in almost all models. An association was observed for assisted vaginal delivery and persistent asthma in model 2 only (OR 1.39, 95% CI:1.07–1.80). Caesarean section after induction was also significantly associated with having persistent asthma (OR 1.46; 95% CI:1.09–1.94) (Table 2). After adjusting for all potential confounders, the association remained significant. No significant associations were observed between mode of delivery and adolescent-onset asthma for both crude and adjusted models. At age 7 years, induced vaginal delivery and emergency CS were significantly associated with wheezing in the last 12 months across all models, with results attenuating at age 11 years (Table 2). Results of secondary outcome analysis and all sensitivity analyses can be found here: https://doi.org/10.5281/zenodo.7438158. We have yielded two principal findings. First, associations were observed between emergency CS and asthma at ages 7, 11 and 14, with few associations observed between assisted and induced vaginal delivery and asthma at age 11 years and induced vaginal delivery and wheezing at age 7 years. Second, no associations were observed between planned CS and asthma in crude and adjusted models. Potential explanations for positive associations observed may include a disruption in the infant intestinal microflora following CS, although an association with planned CS would be expected in this instance. Furthermore, foetal stress during delivery, leading to changes in the development of the new-born immune system may increase the risk of developing atopic disorders.3 However, results would not likely lead to, or warrant a change in practice, and moreover, associations are unlikely to be causal and may be due to confounding by indication. Two large Swedish studies based on national health registers assessed the association between delivery by CS and asthma discharge diagnosis or use of asthma medications.4, 5 The study by Almqvist and colleagues included 87,555 sibling-pairs, allowing adjustment for familial environmental confounding and genetic factors, and found that emergency CS, but not elective CS, was associated with increased risk of asthma, which is consistent in our study.5 The second study found that planned CS contributed to a modestly increased risk in children aged between 2 and 5 years and not in older children.4 Finally, a recent study in New York, using the same ISAAC questionnaire as our study, reported a higher risk of wheeze in children born by emergency CS.2 The mode of delivery and data on potential confounders were self-reported and subject to recall bias. However, maternal reporting of mode of delivery in the MCS has previously been compared to hospital records and was found to be highly reliable, with an agreement of 94%–98%.6 Furthermore, as these data were collected prior to the knowledge of the asthma status of the child, any misclassification would likely be non-differential, potentially biasing results towards the null. Second, confounding by indication may have influenced our results as we do not know why individuals had an emergency or planned CS. Third, loss-to-follow up is another potential limitation in the current study. While the distributions of most variables were broadly similar between those who participated at baseline and at age 14 years, there were some minor differences for socio-economic position indicators. Additionally, those who reported asthma at age 7 years were more likely to be lost to follow-up at age 14 years which may have biased our results. Finally, while we controlled for ethnicity in the current study, this was categorized as a binary variable due to small numbers. As there is evidence that certain ethnic groups in the UK have distinct experiences of asthma, future studies could explore ethnic variations in asthma across mode of delivery.7 First, the study used a large, population-based cohort that was nationally representative of children born in the UK between 2000 and 2002. Second, a validated questionnaire was used to determine the outcome measure. This questionnaire has been widely used to measure childhood asthma and has been shown to have a specificity of 93%.8 Third, this study classified CS into emergency or planned, a limitation of some previous research. Additionally, we controlled for a wide range of potential confounders including several maternal and socio-economic factors, complications of pregnancy and maternal asthma. In conclusion, children born by emergency CS, but not planned CS had an increased likelihood of asthma diagnosis. These results suggest that the association may be due to confounding by indication or other residual confounding factors, and unlikely to be causal based on the microbiome theory. All authors were involved in the conception of the study. AOC and GMM performed the statistical analysis. All authors contributed to the interpretation of results. AOC wrote the first draft of the manuscript. All authors revised the manuscript for important intellectual content and approved the final version of the manuscript. All authors agree to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. We are grateful to the UK Data Service for making the data available to researchers and to all the children and families who took part in the study. Open access funding provided by IReL. This work was funded by University College Cork (UCC), Ireland (Summer Undergraduate Research Experience Award 2022). The authors report no conflict of interest. Ethical approval for the Millennium Cohort Study was obtained from an NHS Research Ethics Committee and has therefore been performed in accordance with the ethical standards laid down in the 1964 Declaration of Helsinki and its later amendments.

Topics & Concepts