Background and rationale
In the UK, a caesarean section (CS) is a common surgical procedure, with around one-third of babies being delivered this way.1–4 The risk of maternal post-partum infections is considerably higher following CS than following vaginal delivery (VD).5 National Institute for Health and Care Excellence (NICE) Guideline 192 on caesarean birth6 recommends offering women undergoing CS prophylactic antibiotics that work against the organisms causing wound infections, endometritis and urinary tract infections to reduce the risk of post-partum infections.
Before 2011, this NICE clinical guideline recommended giving intravenous prophylactic antibiotics after cord clamping to prevent the baby from being exposed to the maternal antibiotics. In 2011, the recommendation was changed to offering antibiotics to women before skin incision for CS.7 This change in recommendation was based on evidence that earlier administration reduces the risk of maternal infections, although most of these infections are mild and tend to respond well to treatment.8 A Cochrane review9 on the timing of intravenous prophylactic antibiotics for preventing post-partum infectious morbidity in women undergoing cs delivery included evidence from 10 randomised controlled trials (including a total of 5041 women). The review showed a significant reduction in the risk of combined post-partum infections [43%, 95% confidence interval (CI) 28% to 55%], wound infection (41%, 95% CI 19% to 56%) and endometritis (46%, 95% CI 21% to 64%) in women receiving antibiotics prior to cord clamping, compared with administering antibiotics after cord clamping.9
It is known, however, that these high-dose broad-spectrum pre-operative antibiotics cross the placenta, with concentrations detectable within 20–30 minutes, resulting in babies being exposed to the antibiotics at the time of birth.9,10 Although the Cochrane review found no evidence of neonatal adverse effects,9 intrapartum antibiotics have been shown to significantly disrupt the intestinal microbiota of babies.11 During birth, the gut is colonised by microbes12 and the mode of delivery is known to significantly affect the composition of intestinal microbiota both in the neonatal period and later in infancy.13 These microbiota appear to be key for intestinal immune tissue to differentiate resident commensal bacteria from potential pathogens.14 In addition, there is a growing body of evidence that intestinal microbe composition in infants plays a key role in the development of their immune system, including regulation of response to different antigens and inflammation, and disruptions to intestinal microbiota are associated with susceptibility to asthma, allergies and other immune-related diseases later in childhood.15–22
A recent systematic review14 explored the association between intestinal microbiota and the development of asthma, eczema and allergic sensitisation to food allergens in newborns and children. The review found evidence that reduced bacterial diversity and a greater or reduced abundance of particular bacteria was associated with the development of these health conditions. There is also an association between reduced intestinal bacterial diversity in the first months of life and development of allergic rhinitis.23
Some evidence also suggests that gut microbiota may have an influence on the development of autoimmune diseases. For example, gut microbiota have been linked to the development of type 1 diabetes,24 coeliac disease25 and systemic connective tissue disorders.26,27 The evidence in this area is acknowledged to be limited. It is, however, hypothesised that the mechanisms driving autoimmune disease development are similar to those involved in the development of allergic diseases.28
There is limited evidence that explores the link between the gut microbiome and infectious and inflammatory disease and other immune system-related conditions. Intestinal microbiota are likely to be involved in the development of necrotising enterocolitis and inflammatory bowel disease.29 It has also been hypothesised that childhood acute lymphoblastic leukaemia development could have an immune component driven by microbial exposures in early life.30
One study has shown an increased risk of cerebral palsy in children born to mothers who were given antibiotics in spontaneous pre-term labour.31 There is also a small body of research evidence that suggests associations between differences in gut microbiota and other neurodevelopmental conditions, but there is a consensus that further longitudinal assessment of the role of the gut microbiome in neurodevelopmental disorders is required.32,33
In the broader area of child health, there is a small amount of evidence looking at gut microbiota and symptoms, such as colic, and general health measures, such as expected weight gain. Infants affected by colic have been reported to have different levels of intestinal bacterial diversity from unaffected infants.34 Some types of gut bacteria appear to be associated with different growth patterns in babies.35
As antibiotic exposure around the time of birth leads to different gut microbiota patterns in newborns and different microbiota patterns are associated with longer-term health conditions, the change in the timing of when mothers receive prophylactic antibiotics presents a natural experiment. Babies born to mothers who were provided antibiotics after cord clamping are exposed to no antibiotics. Those born to mothers given pre-operative antibiotics are exposed to high-dose broad-spectrum antibiotics that are likely to affect their gut microbiota.
In this study, we aimed to compare the outcomes for babies born by CS who were and who were not exposed to antibiotics given to the mother at the time of birth. The primary outcomes of interest in this study were asthma and eczema. These are the two most frequent allergy-related, doctor-diagnosed conditions in childhood and are associated with changes in gut microbiota composition in infants.18,20 They are commonly seen in primary care and are well recorded.36,37 With a prevalence of 3% in children aged 0–5 years, asthma is a relatively common health condition in childhood, with the largest numbers of emergency hospital admissions for any long-term condition in young children in the UK.38,39 Although deaths due to asthma in children are rare, they do occur, accounting for 1–2% of all asthma deaths.38,40
Eczema is a very common condition in childhood, with a prevalence of 18% of ‘ever diagnosed eczema’ in children aged 0–5 years reported in the UK.41 Eczema usually develops in very early childhood and can affect a child’s sleep, their social and emotional development, and quality of life. It incurs significant societal costs because of lost school days, lost working days for parents and costs to the NHS, including prescription of topical corticosteroids.42
As the evidence regarding the role of microbiota in the development of diseases is evolving, we also assessed whether or not in-utero exposure to broad-spectrum antibiotics immediately prior to birth increases the risk of a range of other immune system-related health conditions in the first 5 years of life in children born by CS.
Aims and objectives
Overall aim
The overall aim of the study was to examine the impact, if any, that the change in the national recommendation from advising administering prophylactic antibiotics post-cord clamping to administering antibiotics before skin incision has had on the risk of allergic and other related health outcomes in children delivered by CS.
Primary objectives
The main objective was to investigate if prophylactic antibiotic exposure of the fetus immediately before birth (intervention) compared with no antibiotic exposure (comparator) increases the risk of (1) asthma and (2) eczema (outcomes) in children delivered by CS (population).
Secondary objectives
- To investigate the impact of exposure to pre-incision antibiotics in children delivered by CS on (1) other allergic and allergy-related health conditions, (2) autoimmune conditions, (3) infections and inflammation, (4) other immune system-related health conditions, (5) neurodevelopmental health conditions and (6) less specific child health measures (i.e. colic and failure to thrive).
- To investigate the impact of exposure to pre-incision antibiotics in children delivered by CS on health service use (defined as overall primary care consultation frequency and hospital admissions).
- To investigate if the effects of reducing post-partum infections in mothers, as reported in randomised controlled trials conducted outside the UK,9 can be replicated in the UK using routine health-care records.