Antibiotics for early-onset neonatal infection: a summary of the NICE guideline 2012

Authors

  • Moira A Mugglestone BSc MSc PhD FCMI,

    Director, Corresponding author
    1. National Collaborating Centre for Women's and Children's Health, Royal College of Obstetricians and Gynaecologists, Regent's Park, London, UK
    Search for more papers by this author
  • M Stephen Murphy BSc DCH MD FRCPI FRCPCH,

    Clinical Director for Children's Health, Senior Lecturer in Paediatrics and Child Health, Consultant Paediatric Gastroenterologist
    1. National Collaborating Centre for Women's and Children's Health, Royal College of Obstetricians and Gynaecologists, Regent's Park, London, UK
    2. College of Medical and Dental Sciences, University of Birmingham, Edgbaston, Birmingham, UK
    3. Birmingham Children's Hospital, NHS Foundation Trust, Birmingham, UK
    Search for more papers by this author
  • Cristina Visintin PhD,

    Project Manager
    1. National Collaborating Centre for Women's and Children's Health, Royal College of Obstetricians and Gynaecologists, Regent's Park, London, UK
    Search for more papers by this author
  • David T Howe DM FRCOG FRCS(Ed),

    Consultant and Honorary Senior Lecturer in Feto-Maternal Medicine
    1. Wessex Fetal Medicine Unit, Princess Anne Hospital, Southampton, UK
    Search for more papers by this author
  • Mark A Turner BSc MBChB PhD MRCP(UK) MRCPCH DRCOG

    Senior Lecturer in Neonatalogy and Consultant in Neonatology
    1. Department of Women's and Children's Health, University of Liverpool and Liverpool Women's Hospital, Liverpool, UK
    Search for more papers by this author

Abstract

Key content

  • Early-onset neonatal infection (infection arising within 72 hours of birth) is an important cause of morbidity and mortality and is often caused by Streptococcus agalactiae (group B streptococcus [GBS]).
  • Identifying and assessing risk factors for early-onset neonatal infection before and during labour and birth is integral to clinical management.
  • Intrapartum antibiotic prophylaxis (IAP) to prevent early-onset neonatal infection is effective when given to women with particular risk factors, including maternal GBS colonisation.
  • When IAP is given specifically to prevent early-onset neonatal infection with GBS the National Institute for Health and Care Excellence (NICE) recommends using benzylpenicillin.
  • Alternative antibiotic regimens are appropriate for women who are allergic to penicillin or where local microbiological surveillance data indicate antibiotic resistance.

Learning objectives

  • To be aware of the risk factors for early-onset neonatal infection.
  • To understand how IAP can reduce the risk of an early-onset neonatal infection.
  • To know which antibiotic to use for IAP of GBS.

Ethical issues

  • How should decisions about whether to administer IAP trade off the potential benefit to the baby of preventing an early-onset neonatal infection and potential harms to the woman, such as, allergic reactions and increased medicalisation of pregnancy?
  • How should decisions about the choice of antibiotics used in intrapartum prophylaxis take account of the risk of promoting antibiotic resistance?

Introduction

This review summarises evidence and guidance in the guideline on antibiotics for the prevention and treatment of early-onset neonatal infection published by the National Institute for Heath and Care Excellence (NICE).[1] The authors of this article highlight the aspects that are most relevant to obstetricians, namely the identification of risk factors for early-onset neonatal infection before and during birth, and indications for intrapartum antibiotic prophylaxis (IAP) for the prevention of early-onset neonatal infection with Streptococcus agalactiae (group B streptococcus [GBS]). Several other guidelines published by NICE and the Royal College of Obstetricians and Gynaecologists (RCOG) are relevant to this topic and are cited in the article where relevant.

Early-onset neonatal infection

The NICE guideline,[1] defines early-onset neonatal infection as infection occurring within 72 hours of birth. This definition was agreed during the guideline scoping process in the absence of a validated definition, although the literature uses definitions ranging from infection within 48 hours to within 1 week of birth. When reviewing the literature, the guideline development group (GDG) considered all relevant information irrespective of study authors’ definitions of early-onset infection.

A key factor in the management of early-onset neonatal infection is an awareness of causative micro-organisms. Two UK population-based neonatal infection surveillance studies, both of which defined early-onset neonatal infection as infection with onset within 48 hours of birth, were published during the development of the guideline.[2, 3] These studies showed that after exclusion of coagulase negative staphylocci (CONS), which are usually considered to be blood sample contaminants soon after birth, the common organisms were Gram-positive in about 75% of cases. These included GBS (50% of cases), non-pyogenic streptococci and Staphylococcus aureus. The most common Gram-negative organism was Escherichia coli. Both GBS and E. coli were susceptible in nearly all cases to antibiotic regimens combining benzylpenicillin and gentamicin, benzylpenicillin and amoxicillin, cefotaxime, or cefotaxime and amoxicillin. S.aureus isolates were resistant to benzylpenicillin and gentamicin in combination.

Risk factors for early-onset neonatal infection

The GDG conducted a systematic review of studies evaluating putative maternal and fetal risk factors (individually or in combination) as predictors of early-onset neonatal infection and their relevance in guiding clinical management before and after the birth. Eleven studies (all observational in design) were included in the systematic review, investigating risk assessment based on either single or a combination of factors.

None of the studies included in the systematic review demonstrated that single or combined features in the maternal or fetal history would be useful for predicting which babies would develop an early-onset neonatal infection. The GDG concluded that healthcare professionals should recognise the criteria listed in Box 1 as risk factors for early-onset neonatal infection, although not all of these were sufficiently important individually to warrant intervention in the form of IAP (see below). The GDG further recommended that, for women in labour, healthcare professionals should:

  • identify and assess any risk factors for early-onset neonatal infection,
  • monitor throughout labour for emergence of new risk factors, such as intrapartum fever higher than 38°C, or development of chorioamnionitis.

Box 1. Risk factors for early-onset neonatal infection

  • Invasive group B streptococcal infection in a previous baby
  • Maternal group B streptococcal colonisation, bacteriuria or infection in the current pregnancy
  • Prelabour rupture of membranes
  • Preterm birth following spontaneous labour (before 37 weeks of gestation)
  • Suspected or confirmed rupture of membranes for more than 18 hours in a preterm birth
  • Intrapartum fever higher than 38°C, or confirmed or suspected chorioamnionitis
  • Parenteral antibiotic treatment given to the woman for confirmed or suspected invasive bacterial infection (such as septicaemia) at any time during labour, or in the 24-hour periods before or after the birth. (This does not refer to intrapartum antibiotic prophylaxis.)
  • Suspected or confirmed infection in another baby in the case of a multiple pregnancy

Intrapartum antibiotic prophylaxis

The GDG conducted a systematic review of evidence from randomised controlled trials (RCTs) evaluating the effectiveness of antibiotic prophylaxis offered to pregnant women at, or shortly before, the expected time of labour and birth for the prevention of early-onset neonatal infection. The term IAP was used to refer to such practice. Since the NICE guideline on routine intrapartum care[4] covers IAP for prelabour rupture of membranes (PROM) at term, evidence relating to women with PROM was considered in the systematic review only where the study population included women with preterm PROM.

Thirty-seven articles reporting 34 RCTs and one non-comparative observational study were included in the review. The GDG agreed that they would recommend IAP only where the purpose of the intervention was to prevent early-onset neonatal infection with GBS. Population screening (universal testing) for GBS, or testing in particular groups of women, was not the focus of the GDG's review because routine screening for GBS is not recommended in the NICE guideline for routine antenatal care.[5]

Women with group B streptococcal colonisation

Five RCTs considered intrapartum antibiotics in women with GBS colonisation (Table 1). Having considered the evidence, the GDG's view was that women with GBS colonisation (including women in preterm labour) should be offered IAP with benzylpenicillin because:

Table 1. Key evidence for the effectiveness of intrapartum antibiotic prophylaxis for the prevention of early-onset neonatal infection in women with group B streptococcal infectiona
InterventionComparatorRelative risk95% confidence intervalStudy identifier
  1. a

    Only those outcomes for which statistically significant differences between treatment groups were reported are presented numerically. Numerical results for all outcomes considered by the GDG are presented systematically in the ‘full guideline’[15]

Incidence of neonatal infection
Incidence of early-onset group B streptococcal disease
BenzylpenicillinNo antibiotic prophylaxis0.130.02 to 0.97Tuppurainen and Hallman, 1989[11]
AmpicillinNo antibiotic prophylaxis0.040.003 to 0.69Morales et al., 1986[12]
Incidence of group B streptococcal sepsis or group B streptococcal bacteraemic disease
AmpicillinNo antibiotic prophylaxis0.110.01 to 0.89Boyer and Gotoff, 1986,[13] Mattoras et al., 1991[14]
Maternal outcomes (incidence of postpartum febrile morbidity)
AmpicillinNo antibiotic prophylaxis0.410.18 to 0.93Mattoras et al., 1991[14]
  • IAP with benzylpenicillin or ampicillin reduces the incidence of early-onset neonatal GBS disease,
  • there is no difference between benzylpenicillin and ampicillin in the incidence of suspected infection, neonatal mortality or neonatal duration of hospital stay, and,
  • benzylpenicillin has a narrow-spectrum of activity (whereas ampicillin is a broad-spectrum antibiotic) and so benzylpenicillin is less likely to promote antibiotic resistance.

The review did not identify any RCTs evaluating the effectiveness of IAP in pregnant women with GBS bacteriuria or maternal infection, nor in women with a previous baby affected by an invasive GBS infection. However, it is current practice to offer IAP to these groups of women (as exemplified by the RCOG green-top guideline for the prevention of early-onset neonatal GBS disease).[6] In the absence of evidence directing a change in practice, the GDG concluded that these groups of women should be offered IAP with benzylpenicillin.

Recognising that benzylpenicillin would not be suitable for all women (for example, those allergic to penicillin), the GDG recommended clindamycin as an alternative to benzylpenicillin unless microbiological surveillance data reveal local bacterial resistance patterns indicating a different antibiotic. Since resistance of GBS to clindamycin and other antibiotics varies markedly across England and Wales,[7] and local variations in antibiotic resistance might be important in determining second-line treatment options for individual women, the GDG emphasised that for women colonised with GBS, local laboratory data should guide the choice of antibiotics.

Women with preterm labour who are not colonised by group B streptococcus

Eleven articles reporting 10 RCTs considered intrapartum antibiotics in women with preterm labour (see Table 2). The GDG agreed that antibiotics given with the intention of preventing early-onset neonatal infection should have a narrow spectrum of activity (as with benzylpenicillin) rather than a broad spectrum (as with ampicillin and amoxicillin). Adding clavulanic acid or sulbactam to ampicillin or amoxicillin would give broader cover for micro-organisms, but the GDG noted from the evidence that there was an increased risk of necrotising enterocolitis with the use of co-amoxiclav (amoxicillin plus clavulanic acid).

Table 2. Key evidence for the effectiveness of intrapartum antibiotic prophylaxis for the prevention of early-onset neonatal infection in women with preterm laboura
InterventionComparatorRelative risk95% confidence intervalStudy identifier
  1. a

    Only those outcomes for which statistically significant differences between treatment groups were reported are presented numerically. Numerical results for all outcomes considered by the GDG are presented systematically in the ‘full guideline’[15]

Incidence of neonatal infection
Intravenous ampicillinPlacebo0.140.05 to 0.36Nadisauskiene and Bergstrom,1996[16]
Oral ampicillinPlacebo0.130.05 to 0.34Nadisauskiene et al., 1996[17]
Long-term paediatric follow-up
Incidence of any functional impairment in vision, hearing, speaking, ambulation, dexterity, emotion, cognition or pain at age 7 years
ErythromycinPlacebo1.131.002 to 1.28Kenyon et al., 2008[18]
Erythromycin plus co-amoxiclavPlacebo1.130.997 to 1.28Kenyon et al., 2008[18]
Erythromycin (with or without co-amoxiclav)No erythromycin (co-amoxiclav only or placebo)1.111.01 to 1.21Kenyon et al., 2008[18]
Incidence of cerebral palsy at age 7 years
Erythromycin (with or without co-amoxiclav)No erythromycin (co-amoxiclav only or placebo)1.901.20 to 3.01Kenyon et al., 2008[18]
Co-amoxiclav (with or without erythromycin)No co-amoxiclav (erythromycin only or placebo)1.671.06 to 2.61Kenyon et al., 2008[18]
Maternal outcomes (incidence of puerperal uterine infection)
Intravenous ampicillinPlacebo0.410.20 to 0.81Nadisauskiene and Bergstrom,1996[16]
Oral ampicillinPlacebo0.030.005 to 0.23Nadisauskiene et al., 1996[17]

The review identified no RCTs evaluating potential short-term benefits of IAP using benzylpenicillin among women who did not have GBS colonisation. However, the GDG emphasised the need for further research to evaluate the effectiveness of benzylpenicillin in women with preterm labour who do not have GBS colonisation.

Women with preterm prelabour rupture of membranes

Eighteen articles reporting 16 RCTs considered intrapartum antibiotics in women with preterm PROM (see Table 3). The GDG concluded that the increased risk of early-onset neonatal infection associated with preterm labour and ruptured membranes was effectively an increased risk of early-onset neonatal GBS infection. The GDG was aware of non-randomised clinical studies demonstrating the effectiveness of a risk-factor based approach to reducing the incidence of early-onset neonatal GBS infection.[8, 9] Since prematurity, rupture of membranes for more than 18 hours and PROM are independent risk factors for early-onset neonatal infection (Box 2), the GDG concluded that, in the absence of compelling evidence to support a stronger recommendation, IAP should be ‘considered’ for women in preterm labour with:

  • PROM of any duration, or
  • suspected or confirmed intrapartum rupture of membranes lasting more than 18 hours.
Table 3. Key evidence for the effectiveness of intrapartum antibiotic prophylaxis for the prevention of early-onset neonatal infection in women with preterm prelabour rupture of membranesa
InterventionComparatorRelative risk95% confidence intervalStudy identifier
  1. a

    Only those outcomes for which statistically significant differences between treatment groups were reported are presented numerically. Numerical results for all outcomes considered by the GDG are presented systematically in the ‘full guideline’[15]

Incidence of neonatal infection (incidence of blood culture-proven infection in babies born within 14 days of randomisation)
Erythromycin plus co-amoxiclavPlacebo0.700.51 to 0.97Kenyon et al., 2001[19]
Incidence of neonatal mortality
Ampicillin (with or without betamethasone)Placebo, betamethasone or no medication0.470.23 to 0.96

Amon et al., 1988,[20] Grable et al., 1996,[21] Morales et al., 1989,[22]

Owen et al., 1993[23]

Incidence of proven necrotising enterocolitis
Co-amoxiclavPlacebo4.071.67 to 9.91Kenyon et al., 2001[19]
Erythromycin plus co-amoxiclavPlacebo3.431.38 to 8.52Kenyon et al., 2001[19]

The GDG recommended benzylpenicillin as the first choice for IAP in these groups of women to target GBS as the predominant cause of early-onset neonatal infection. Unlike some other antibiotics (including erythromycin), benzylpenicillin crosses the placenta and is, therefore, a rational choice for prevention or treatment of GBS infection in the baby.

Women with chorioamnionitis

Three RCTs considered intrapartum antibiotics in women with chorioamnionitis (see Table 4). Having considered the evidence, the GDG noted that intrapartum antibiotic treatment for chorioamnionitis is usual obstetric practice, the intention being to benefit the woman (although the baby might benefit too). The GDG concluded that antibiotic treatment for women with chorioamnionitis (or intrapartum fever more than 38°C) should be determined by obstetric practice as defined in other guidelines, rather than treatment being designed to benefit the baby. Thus, no recommendation was made by the GDG regarding this group of women. However, the RCOG guideline on the prevention of early-onset neonatal GBS disease[6] recommends that in women with GBS colonisation and suspected chorioamnionitis, broad-spectrum antibiotics including an antibiotic active against GBS should replace GBS-specific antibiotics. The RCOG guideline similarly recommends that women with intrapartum fever (more than 38°C) should be offered broad-spectrum antibiotics.

Table 4. Key evidence for the effectiveness of intrapartum antibiotic prophylaxis for the prevention of early-onset neonatal infection in women with chorioamnionitisa
InterventionComparatorMean difference95% confidence intervalStudy identifier
  1. a

    Only those outcomes for which statistically significant differences between treatment groups were reported are presented numerically. Numerical results for all outcomes considered by the GDG are presented systematically in the ‘full guideline’[15]

Neonatal duration of hospital stay (days)
Intrapartum ampicillin plus gentamicinPostpartum ampicillin plus gentamicin1.9 fewer3.31 fewer to 0.49 fewerGibbs et al, 1988[24]

Box 2. Intrapartum antibiotic prophylaxis to prevent early-onset neonatal infection

  • Offer intrapartum antibiotic prophylaxis using intravenous benzylpenicillin to prevent early-onset neonatal infection for women who have had:
    • - a previous baby with an invasive group B streptococcal infection
    • - group B streptococcal colonisation, bacteriuria or infection in the current pregnancy
  • If the woman decides to take intrapartum antibiotic prophylaxis, give the first dose as soon as possible and continue prophylaxis until the birth of the baby
  • Consider intrapartum antibiotic prophylaxis using intravenous benzylpenicillin to prevent early-onset neonatal infection for women in preterm labour if there is prelabour rupture of membranes of any duration
  • Consider intrapartum antibiotic prophylaxis using intravenous benzylpenicillin to prevent early-onset neonatal infection for women in preterm labour if there is suspected or confirmed intrapartum rupture of membranes lasting more than 18 hours
  • Offer benzylpenicillin as the first choice for intrapartum antibiotic prophylaxis. If the woman is allergic to penicillin, offer clindamycin unless individual group B streptococcus sensitivity results or local microbiological surveillance data indicate a different antibiotic
  • If there are no signs of infection in a woman with prelabour rupture of the membranes at term, antibiotics should not be given to either the woman or the baby, even if the membranes have been ruptured for over 24 hours

Women with meconium-stained amniotic fluid

One RCT considered intrapartum antibiotics in women with meconium-stained amniotic fluid. There were no statistically significant differences in the incidence of neonatal sepsis, neonatal duration of hospital stay, or incidence of endometritis. The GDG concluded that the evidence did not support intrapartum antibiotics for the benefit of the baby in this group of women. It was also noted that usual obstetric practice is not to offer antibiotic treatment to women with meconium-stained amniotic fluid, and so the GDG made no recommendation for this group of women.

Conclusion

The GDG's recommendations relating to IAP to prevent early-onset neonatal infection are summarised in Box 2. The box also presents a recommendation from the NICE guideline on routine intrapartum care[4] specifying that in the absence of signs of infection and the absence of GBS colonisation in a woman with term PROM, antibiotics should not be given to the woman or the baby, even if the membranes have been ruptured for more than 24 hours.

The recommendations summarised in this article reflect the following key principles identified by the GDG as being important in determining clinical practice relating to the prevention and treatment of early-onset neonatal infection:

  • Unless it is dangerous, families should be offered choice. Informed choice should be supported through provision of information and, where appropriate, reassurance.
  • IAP should be administered in a timely manner to all eligible women who choose it.
  • Babies with suspected early-onset neonatal infection should be treated as quickly as possible.
  • Antibiotic exposure should be minimised in babies who do not have an early-onset neonatal infection. This is important because of the need to minimise antimicrobial resistance and increasing concerns about the impact of antibiotics on the development of the immune system.[10]
  • An integrated system of clinical care is needed to allow full implementation of the guideline recommendations.

Specific recommendations relating to information and support for women whose babies are at increased risk of developing an early-onset neonatal infection include maintaining communication with these women during labour and involving them in any handover of care (either when additional expertise is brought in because of the risk of infection or during planned changes in staff; the handover should include an update about the presence of infection). Another recommendation of relevance to obstetric practice is that pregnant women who have had GBS colonisation in a previous pregnancy but without infection in the baby should be reassured that this will not affect management of the birth in the current pregnancy. At discharge, women who have had GBS colonisation in the pregnancy but without infection in the baby should be informed that if they become pregnant again, this will not affect the management of the birth in the next pregnancy.[1]

Further reading

The NICE website publishes clinical guidelines in various formats aimed at healthcare professionals, patients and the public. It also publishes the NICE guidelines manual which describes the process and methods used in developing clinical guidelines. See http://www.nice.org.uk/guidance/cg/index.jsp

The RCOG website publishes clinical guidelines, including several that are relevant to prevention and treatment of infection during or after pregnancy. See http://www.rcog.org.uk/womens-health/guidelines

Acknowledgements

The National Collaborating Centre for Women's and Children's Health is funded by NICE to develop clinical guidelines. The authors were members of the GDG for the guideline on antibiotics for early-onset neonatal infection. The views expressed in this publication are those of the authors and not necessarily those of the Institute. The authors thank the referees for their comments which led to substantial improvements in the manuscript.

Contribution of authorship

All authors contributed to the initial draft and revisions and approved the final version for publication. MAM is the guarantor.

Disclosure of interests

All authors have completed the ICMJE form for disclosure of potential conflicts of interest at http://www.icmje.org/coi_disclosure.pdf (available on request from the corresponding author) and declare that they have: no support from any organisation for the submitted work; no financial relationships with any organisations that might have an interest in the submitted work in the previous three years; no other relationships or activities that could appear to have influenced the submitted work.

Ancillary