Clinical & Experimental Allergy
  • Open Access

The impact of birth mode of delivery on childhood asthma and allergic diseases–a sibling study

Authors


Correspondence:

Associate Professor Catarina Almqvist, Department of Medical Epidemiology and Biostatistics and Lung and Allergy Unit, Astrid Lindgren Children's Hospital, PO Box 281, Karolinska Institutet, SE-171 77 Stockholm, Sweden.

E-mail: catarina.almqvist@ki.se

Abstract

Background

Caesarean section (CS) has been reported to increase the risk of asthma in offspring. This may be due to that infants delivered by CS are unexposed to vaginal flora, according to the ‘hygiene hypothesis’.

Objective

Our aim was to investigate if CS increases risk of childhood asthma, and if the risk increase remains after adjustment for familial confounding using sibling design.

Methods

A register-based cohort study with 87 500 Swedish sibling pairs was undertaken. Asthma outcome variables were collected from national health registers as diagnosis or asthma medication (ICD-10 J45-J46; ATC code R03) during the 10th or 13th year of life (year of follow-up). Mode of delivery and confounders were retrieved from the Medical Birth Register. The data were analysed both as a cohort and with sibling control analysis which adjusts for unmeasured familial confounding.

Results

In the cohort analyses, there was an increased risk of asthma medication and asthma diagnosis during year of follow-up in children born with CS (adjusted ORs, 95% CI 1.13, 1.04–1.24 and 1.10, 1.03–1.18 respectively). When separating between emergency and elective CS the effect on asthma medication remained for emergency CS, but not for elective CS, while both groups had significant effects on asthma diagnosis compared with vaginal delivery. In sibling control analyses, the effect of elective CS on asthma disappeared, while similar but non-significant ORs of medication were obtained for emergency CS.

Conclusions and Clinical Relevance

An increased risk of asthma medication in the group born by emergency CS, but not elective, suggests that there is no causal effect due to vaginal microflora. A more probable explanation should be sought in the indications for emergency CS.

Ancillary