The association between caesarean section and asthma or allergic disease continues to challenge
Article first published online: 13 MAR 2014
©2014 Foundation Acta Pædiatrica. Published by John Wiley & Sons Ltd
Volume 103, Issue 4, pages 349–351, April 2014
How to Cite
Almqvist, C. and Öberg, A. S. (2014), The association between caesarean section and asthma or allergic disease continues to challenge. Acta Paediatrica, 103: 349–351. doi: 10.1111/apa.12562
- Issue published online: 13 MAR 2014
- Article first published online: 13 MAR 2014
- Stockholm County Council
- Karolinska Institutet
- Swedish Heart-Lung foundation
- Swedish Research Council. Grant Number: 2011-3060
- SIMSAM. Grant Number: 340-2013-5867
- Karolinska Institutet
The rate of caesarean section (CS) has increased in developed countries and is now the most common surgical procedure in women of reproductive age. According to the World Health Organization, CS is indicated in up to 15% of deliveries, yet the majority of developed countries currently exceed this recommendation . This suggests that the increase is due to a rise in elective, prelabour CS, without a clear medical indication, rather than the need for emergency CS.
Mode of delivery may affect outcomes such as asthma and allergic disease in children many years after birth. Previous meta-analyses have reported a moderate 20% increased risk of allergic rhinitis and asthma in children delivered by CS , but no association with atopic dermatitis/eczema . More recently, studies linking the risk of asthma to mode of delivery have examined the roles of emergency CS after the onset of labour, versus elective prelabour CS [4-6], and instrumental use during vaginal delivery, involving forceps or vacuum extraction . Some studies have also performed sophisticated sibling analyses [5, 6]. Most of these aspects were reviewed by Pyrhönen et al.  in a recent issue of Acta Paediatrica, in a paper that further extends the existing literature.
The authors aimed to provide further clarification of the association between CS and the occurrence of allergic manifestations – food and pollen allergies, hay fever, atopic eczema and asthma – in a group of children aged between 1 and 4 years old . The target population (N = 4,779) was identified from the Finnish nationwide population register, and questionnaire data on 3181 participants were merged with allergy test results from skin prick tests, IgE antibodies and open food challenges. While consistent with a null hypothesis, the authors’ negative, nonsignificant findings on the association between CS and any allergic manifestation or positive test results for food, pollen or animal allergens also need to be considered in the context of a number of factors. These are limited power from not enough participants, and methodological issues pertaining to study design and the definitions of exposures and outcomes. However, the strengths of the study included objective markers for several allergic outcomes. While the authors also emphasised the representativeness of the sample, there should be some concern about participation bias and selection for blood testing, as well as potential recall bias for perinatal data. It would also have been very interesting to see mode of delivery; vaginal delivery, emergency and elective CS and instrumental delivery, explored in greater detail. Potential shortcomings aside, it is very important that negative findings such as these are reported and published, in order to add to the growing body of literature on this far from resolved topic.
The choice of birth mode of delivery is made based on maternal characteristics and anticipated paediatric outcomes. It is also related to the choice and preference of the pregnant mother or couple, as well as local clinical practice. Obstetricians assess the mode of delivery based on timing, progress and the degree of foetal distress. Normally a spontaneous vaginal delivery is considered the safest mode of delivery for both mother and child. The choice of emergency procedures, namely emergency CS or instrumental vaginal delivery, is based on many factors, such as maternal compliance, degree of foetal stress and progress of labour. If the baby needs to be delivered quickly, instrumental vaginal delivery is the best choice if the cervix is fully dilated and the foetal head is in a good position. Otherwise an emergency CS is normally the fastest and safest way of delivering the baby, when there are signs of foetal distress or when elective CS had already been agreed, but the woman has gone into labour before the planned procedure and instrumental or spontaneous vaginal delivery is not an option. An elective CS is chosen for maternal reasons, including extreme fear of labour, or a combination of maternal and foetal reasons, such as malpresentation, two or more previous CS procedures or multiple gestation.
A few recent studies have distinguished between elective CS and emergency CS. Of the 37 171 children included in the Norwegian Mother and Child Cohort Study, those delivered by CS had an increased likelihood of current asthma at 36 months of age [relative risk 1.17, 95% confidence interval (CI) 1.03–1.32], with similar findings among children delivered by emergency and elective CS. Two studies based on the Swedish national health registers showed an increased risk of asthma in children born with CS, but provided different and conflicting results for emergency and elective CS [5, 6]. Mechanisms to explain the association between birth mode of delivery and subsequent asthma or atopy could involve confounders such as maternal smoking, socio-economic status or family history of asthma. Sibling studies therefore provide an excellent opportunity to study the association between CS and asthma independent of shared environmental and genetic factors. Full siblings share approximately half of their segregating genes, some intrauterine exposures, maternal factors and early environment. In addition, siblings may be delivered using different methods. Traditionally, if associations seen in a cohort of siblings remain in sibling control analyses, this is taken to indicate that factors specific to each individual, such as exposure to vaginal flora or the indication for mode of delivery, are involved in the underlying causal pathways. Conversely, if the relationships change in the sibling control analyses, this would indicate an influence from factors common to the siblings, such as maternal factors. Such conclusions, however, also require careful consideration of the potential influence of measurement error and individual, nonshared confounders . Pyrhönen et al. had too few siblings to perform this type of control analyses. However, both of the studies based on the Swedish national health registers reported negative findings in older children after sibling control [5, 6], consistent with a lack of causal association between CS and childhood asthma.
Potential mechanisms for the presumed associations between CS and subsequent asthma or allergic disease were recently reviewed by Cho et al. . In line with the ‘hygiene hypothesis’, the intestinal microflora may be modified in those children who are not exposed to the vaginal flora. It has also been suggested that DNA methylation is higher in infants delivered by CS than in those vaginally delivered, and differences in immune biomarkers, gene expression and stress following CS compared with vaginal birth have also been proposed . However, all or parts of the association between CS and risk of asthma could also be explained by the underlying indications for CS. These could be related to the choice of elective CS due to an anxious mother, anthropometric measures or obstetric history, emergency situations such as prematurity, prolonged parturition or foetal asphyxia or subsequent diagnoses, including early respiratory stress.
If the vaginal microflora and/or epigenetics play a role in the association between CS and asthma, some guidance should be obtained from the difference in effect after vaginal delivery, instrumental vaginal delivery, elective or emergency CS on this association. In noninstrumental vaginal delivery, the foetus is exposed to microflora, there is no excessive stress for the mother, and there is normally no abnormal stress on the foetus, although there are cases of unexplained asphyxia. In instrumental vaginal delivery, indicated by stress in the mother, foetus or both, the foetus is also exposed to microflora. In emergency CS, there may be exposure to vaginal microflora if a failed forceps/vacuum extraction delivery requires a CS, if there is sign of intra-amniotic infection and, also theoretically, if the amniotic membranes are ruptured. Both maternal and foetal stress may be involved. For elective CS, there is normally no microflora exposure and no maternal or foetal stress. Epigenetic changes, such as those related to CS, may also be relevant in instrumental deliveries, although an instrumental delivery may be more similar to a vaginal delivery than a CS. Thus, future studies could help shed light on the potential differences between vaginal delivery, CS and instrumental deliveries on mechanisms such as microflora exposure, epigenetic changes, stress mediators and subsequent childhood outcomes.
In conclusion, the negative findings on the association between CS and allergic manifestations displayed in the Pyrhönen paper appear consistent with recent reports from larger cohorts, including sibling controls. Further investigations on the long-term effects of emergency and elective CS may include a systematic survey of all types of mode of delivery. This will help researchers to characterise the nature of the immune response and gene expression over time and to provide appropriate measurements of potentially confounding genetic and environmental factors.
Financial support was provided through the regional agreement on medical training and clinical research (ALF) between Stockholm County Council and Karolinska Institutet, the Swedish Heart-Lung foundation, the Swedish Research Council (grant no 2011-3060) and through the Swedish Initiative for Research on Microdata in the Social and Medical Sciences (SIMSAM) framework grant no 340-2013-5867, and the Strategic Research Program in Epidemiology at Karolinska Institutet.