Antenatal antibiotics: too much, too little, or just right?


  • MA Turrentine

    Corresponding author
    1. Obstetrics & Gynecology, Kelsey-Seybold Clinic, Houston, TX, USA
    2. Department of Obstetrics & Gynecology, Baylor College of Medicine and UTHealth Medical School, Houston, TX, USA
    • Correspondence: Dr MA Turrentine, Department of Obstetrics & Gynecology, Kelsey-Seybold Clinic, 1111 Augusta Drive, Houston, TX 77057, USA. Email

    Search for more papers by this author

As obstetricians, when faced with a potential antepartum infectious process, we must decide whether to treat the mother, the fetus, or both? Since the mid-20th century, the widespread availability and use of antibiotics has led to a dramatic reduction in illness and death from infectious diseases, with obstetric care not excepted. Current estimates have suggested that >40% of pregnant women will be given some type of antibiotic therapy immediately prior to delivery. However, despite the advent of modern antimicrobial treatments, maternal sepsis is still the third leading cause of maternal mortality in developed countries.[1] It has been estimated that for every maternal death, approximately 50 women experience severe morbidity.[2] The long-term effects of the current strategies to reduce short-term maternal and neonatal morbidities are less well recognised. So it begs the question, are we subjecting the mother–infant dyad to antibiotic exposure unnecessarily?

In this issue of BJOG, Drs Ledger and Blaser raise trenchant points on the use of antibiotics during pregnancy. With a 10–30% incidence of maternal colonisation with group B streptococcus (GBS), and with 33% of women undergoing caesarean delivery (two of the most common indications for maternal intrapartum antibiotic prophylaxis), between 1.5 and 2.0 million women (and their infants) will be exposed to antibacterial agents annually.[3, 4] What have been the risks and the benefits?

The deliberation over risk-factor versus universal screening, and then treating for maternal colonisation by GBS, was an arduous debate in the USA. This question is still contested in other parts of the world. Prior to the advent of maternal prophylaxis for GBS, the incidence of neonatal sepsis was 1.7 per 1000 births.[3] With the recommendation of universal screening and antibiotic prophylaxis in 2002, the rates of GBS sepsis have diminished by 85% (0.26 per 1000 births).[3, 5] This approach results in almost 700 women (and their unborn infants) being exposed to antibiotics to prevent one case of early-onset GBS disease; however, if the decision is made to treat women who are colonised with intrapartum GBS, it has been demonstrated that the most beneficial strategy is to screen and treat.[3] Most women (87%) who deliver at term and are colonised with GBS will receive antibiotic chemoprophylaxis.[3] Currently, the majority of cases (61.4%) of infant GBS disease occur in term newborns born to women who test negative for GBS prior to delivery.[3] A potential future solution might be rapid screening assays to detect maternal intrapartum GBS colonisation, which may optimise the real-time identification of GBS colonisation while simultaneously helping to minimise unnecessary antibiotic exposure. Unfortunately the clinical feasibility of these technologies still needs to be assessed. With the most common reason for missed intrapartum chemoprophylaxis in term pregnancies being delivery at <4 hours after admission to hospital (which is the optimal time period for GBS antibiotic prophylaxis), delaying the initiation of antibiotic therapy by even 1 hour will further increase the exposure to the at-risk infant.[3]

With over 1.2 million women having a caesarean delivery, even small decreases in infectious morbidity will have a substantial impact on maternal health and health care use.[4] Surgical scoring systems, such as the National Nosocomial Infections Surveillance (NNIS) risk index, were developed to enable comparisons of wound infection rates between surgeons, among institutions, or across time.[6] However, clinical studies using this risk index as a determinate to administer antibiotic prophylaxis are lacking. Limitations of the NNIS risk index have been noted. For caesarean deliveries the NNIS risk index did not use a multivariate model, and not all of the claimed risk factors are found to be so in multivariate analyses. Furthermore, the NNIS risk index does not stratify adequately the risk of surgical site infection developing in patients who have clean or clean-contaminated procedures (the majority of caesarean deliveries). Several randomised controlled trials and large observational studies have demonstrated that the preoperative administration of antibiotics at the time of caesarean delivery (compared with after umbilical cord clamping) will reduce serious maternal infectious complications (both endometritis and wound infection) by greater than 40%.[7, 8] Comparable reductions in infectious morbidities have also been noted in nonlaboring women undergoing caesarean delivery.[8, 9] Current evidence suggests that there has been no increase in neonatal infectious illness or the selection of antimicrobial-resilient bacteria causing neonatal sepsis, but contemporary studies have been underpowered to analyse these outcomes.[7, 9] We will have to await the results of future studies of sufficient size to evaluate potential neonatal detriments.

Although advantages have been shown for intrapartum antibiotic prophylaxis in these clinical scenarios, what are the long-term impacts on the newborn? Concerns have included the development of antibiotic-resistance bacterial strains in the neonate or an alteration of neonatal flora that might lead to serious long-term sequelae. Despite the increase in intrapartum antibiotic exposure, over the past decade decreases in all etiologies of neonatal early-onset sepsis rates have been shown for term and very low-birthweight infants.[10] Furthermore, no statistically significant changes in the incidence of ampicillin-resistant infections in these same groups of infants have been observed.[10] The human body is colonised by a vast number of microbes, collectively referred to as the human microbiota. It has been demonstrated that the microbiota of infants is impacted not only by the use of perinatal antibiotics, but by several multifactorial effects such as mode of infant delivery, the occurrence of breastfeeding, type of diet, and level of community sanitisation.[11] How these changes in the microbiota of the infant will impact them in childhood or as an adult is unclear. Whether this increases ones susceptibility to childhood or adult diseases (such as autism, asthma, diabetes, and obesity) is ambiguous because of confounding factors. Until large population studies controlling for these confounding factors are available, understanding these associations will be limited.

In our role as doctors we must balance the current best evidence in making decisions about the care of individual patients (both mother and infant), while at the same time optimising the delivery of health services. With guidelines from prominent organisations advocating intrapartum antibiotic prophylaxis for various clinical indications, a large number of women and their unborn infants will be exposed to antibiotics.[3, 9] Trying to reduce current intrapartum antibiotic use would seem to be a bit like ‘closing the stable door after the horse has bolted'. Until additional evidence is available that shows harmful outcomes for the neonate, we will have to balance the perils and advantages of the current preventive infectious practices. As obstetricians we need to ask ourselves: have we been overzealous to reduce infectious morbidity in the peripartum period? Or is the clarity of the indication for antibiotic use balancing the theoretical detrimental effect on the neonatal microbiome? In the words of Robert Southey:

At the table in the kitchen, there were three bowls of porridge. Goldilocks was hungry. She tasted the porridge from the first bowl. ‘This porridge is too hot!’ she exclaimed. So, she tasted the porridge from the second bowl. ‘This porridge is too cold’, she said. So, she tasted the last bowl of porridge. ‘Ahhh, this porridge is just right’, she said happily and she ate it all up.[12]

Disclosure of interests

The author does not have any potential conflicts of interest.

Contribution to authorship

I am the only author of this article.

Details of ethics approval

Not applicable.