Article Text

Download PDFPDF

Missed opportunities for preventing group B streptococcus infection
  1. S Vergnano1,
  2. N Embleton2,
  3. A Collinson3,
  4. E Menson4,
  5. A Bedford Russell5,
  6. P Heath1
  1. 1
    Division of Child Health, St George’s, University of London, London, UK
  2. 2
    Neonatal Service, Royal Victoria Infirmary, Newcastle upon Tyne, UK
  3. 3
    Neonatal Unit, Royal Devon and Exeter NHS Foundation Trust, Exeter, UK
  4. 4
    Evelina Children’s Hospital, London, UK
  5. 5
    Heart of England NHS Trust, Birmingham, UK
  1. Correspondence to Stefania Vergnano, Division of Child Health, St George’s, University of London, Cranmer Terrace, London SW17 0RE, UK; stev{at}


Background: Group B streptococcus (GBS) is the most common cause of early onset (EO) neonatal infection in the UK. National guidelines for its prevention were introduced in 2003. We assessed the opportunities for prevention amongst cases of EO GBS using the electronic Neonatal Infection Surveillance Network (NeonIN).

Methods: Culture proven EO GBS cases occurring between 2004 and 2007 were identified prospectively in eight neonatal units participating in NeonIN. Data concerning risk factors, intrapartum antibiotic (IAP) use and infant outcome were collected retrospectively.

Results: There were 48 cases of GBS over the 4 years (0.52/1000 live-births); 22 male, median gestation 38 weeks. The most common clinical presentation was sepsis and the GBS-attributable mortality was 6%. Risk factors were present in 67% (32) and adequate IAP was given to six of these mothers (19%). If all women with risk factors received prophylaxis, 23 cases (48%) may have been prevented.

Conclusions: Better GBS prevention strategies are required in the UK.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Group B streptococcus (GBS) is the most common cause of early onset (EO) sepsis in the UK. The incidence of culture proven infection is 0.5 cases per 1000 infants, with a mortality of up to 10%.1 The inclusion of culture negative (“probable”) GBS infection may increase the total disease burden by two- to three-fold.2

A number of countries including the USA and Australia introduced EO GBS prevention guidelines in the 1990s and have successfully reduced infection rates in their populations. In the UK a national policy was introduced in 2003 by the Royal College of Obstetrics and Gynaecology (RCOG).3

The RCOG guidelines do not advocate a universal swab-based screening approach on the grounds of the relatively low incidence of GBS infection in the UK and the perceived risks associated with widespread use of antibiotics. They recommend discussing intrapartum antibiotic prophylaxis (IAP) with women with the following risk factors: prematurity <37 weeks, premature rupture of membranes (PROM) >18 h, a previous baby with GBS, maternal pyrexia and known carriage of GBS. Different weight is assigned to risk factors according to the likelihood of disease occurrence. The argument for prophylaxis is stronger in the presence of two or more risk factors.3 Recommended IAP is intravenous benzylpenicillin, or clindamycin if allergy to penicillins is reported, the prescribed antibiotic to be administered at least 2 h before delivery.3

A recent audit of GBS protocols from 171 UK obstetric units indicates that local recommendations are broadly consistent with the RCOG guidelines. However, significant variation was noted, the proportion recommending antibiotics for fever during labour, preterm labour or prolonged rupture of membranes being 61%, 49% and 52%, respectively.4 Reported practice is also noted to be variable. Surveillance data from the Health Protection Agency suggest that rates of EO GBS bacteraemia increased between 2005 and 2006 in England and Wales (from 0.31 to 0.37/1000 live births) (

We undertook a multi-centre review of cases of EO GBS with a particular focus on possible opportunities for prevention.


Cases of EO GBS were identified from the Neonatal Infection Surveillance Network (NeonIN) database to which eight neonatal units were contributing between 2004 and 2007. NeonIN received multi-centre research ethics committee approval from the London-Surrey Borders Research Ethic Committee 05/Q0806/34 in 2005.

EO GBS cases were defined as blood or cerebrospinal fluid (CSF) positive cultures in the first 6 days of life. The study-specific proforma collected data on the maternal risk factors listed in the RCOG guidelines: maternal pyrexia (>38°C), PROM >18 h, previous infant with EO GBS, antenatal detection of GBS and prematurity (<37 weeks). It also included information on type, dose and timing of any IAP, clinical course and outcome of the infection. The antibiotic used for intrapartum prophylaxis was deemed to be appropriate if GBS was known to be susceptible to it and it was administered intravenously.


During the study period 48 cases of EO GBS were identified from 96 217 live births in the contributing hospitals. The overall incidence was 0.5 per 1000 live births and did not vary significantly by year over the study period (data not shown). Fifty four per cent of affected infants were female (26); the median birth weight was 2869 g (range 576–4915 g) and median gestational age was 37 weeks (24–42 weeks).

Of the 48 babies with GBS, 39 (81%) were clinically symptomatic when cultures were obtained; GBS was isolated from blood in 43 and from CSF in five (10%). Five infants died; however, only three deaths were attributable to invasive GBS disease, thus the case fatality rate attributable to invasive GBS disease was 6% (3/48).

Thirty two mothers (67%) had at least one risk factor identified before delivery and 21 had two or more (44%) (fig 1). Only six of the 48 women received appropriate antibiotics in labour (12.5%). Of these, one mother had one risk factor (maternal fever), two had two risk factors (PROM and prematurity; maternal fever and prematurity) and three had three risk factors (PROM, prematurity and previous GBS; PROM, antenatal GBS and maternal fever; PROM, prematurity and antenatal GBS) (table 1). The antibiotics used for IAP were benzylpenicillin (two women), clindamycin (two) and co-amoxyclav (one). One woman received a cephalosporin and metronidazole for possible chorioamnionitis. All dosages were adequate.

Figure 1

Flow chart of intrapartum management of infants who subsequently developed early onset (EO) group B streptococcus (GBS) infection. IAP, intrapartum antibiotics.

Table 1

List of risk factors present

Of the 42 mothers who did not receive IAP, 26 (62%) had at least one risk factor. Women with more than one risk factor were more likely to have received IAP (5/21) than women with one risk factors (1/11): OR 3.13, 95% CI 0.27 to 82. In three cases, mothers presented to hospital in advanced labour with insufficient time for antibiotic administration, two of whom had one or more risk factors. Four women received oral erythromycin: one had three risk factors (PROM, prematurity and known GBS carriage) and three had two risk factors (PROM and prematurity, or prematurity and GBS carriage). None refused prophylaxis (fig 1).

Excluding the three mothers who presented late, had all 29 women with at least one risk factor received appropriate IAP, 23 cases might have been prevented (assuming an IAP effectiveness of 80%5). Had all 18 women with more than one risk factor received IAP, 14 cases may have been prevented (fig 1).


Cases of EO GBS from eight neonatal units in England between 2004 and 2007 were reviewed with regard to opportunities for prevention. The proportion of cases with risk factors was similar to that defined in other UK studies from the pre-guideline era.1 Of 48 EO GBS cases identified, 29 women had at least one risk factor and presented to hospital in time to receive prophylaxis, but only six received appropriate IAP (21%). Four women received oral erythromycin, suggesting that some clinicians may consider this to be adequate for GBS prophylaxis despite current macrolide resistance rates of 11%5 and lack of evidence on the efficacy of oral administration.

It is clear that no IAP policy will prevent all cases of EO GBS. IAP efficacy is estimated at around 80% and late presentation will preclude timely IAP administration. Furthermore, a risk-based prevention strategy does not prevent the approximately 33% of cases with no risk factors. In this audit an effective risk-based strategy which targets women with one or more risk factors has the potential to prevent approximately 50% of all EO GBS cases. IAP administration to women with two or more risk factors (as favoured in the RCOG guidelines) might prevent approximately 30% of EO cases.

We believe that the results of this audit are likely to be representative of practice in other UK units, indeed as NeonIN is a network with a particular interest in neonatal infection, the use of IAP may be greater than in other units. If this is the case, then it is clear that the existence of national guidelines on GBS prevention has not yet been translated into significant IAP use for mothers with risk factors for EO GBS. This may reflect lack of knowledge of the guidelines, which seems unlikely given a recent audit of practice,4 or uncertainty regarding their specific recommendations. The forthcoming update of these guidelines provides an opportunity to review this. Ultimately, an effective GBS vaccine provides the best prospect for prevention.


We thank Dr Nigel Kennea, Dr Hannah Baynes, Dr John Klein, Dr Timothy Watts, Dr Zoe Smith, Dr Lee Geraint, Dr Peter Chow, Dr John Chang, Dr Paul Clarke, Julia Hill, Dr Michael Robinson, Louise Stickland, Dr Imogen Storey, Dr Lorna Gillespie, Dr Bernd Reichert, Dr Mithilesh Lal, Dr Santosh Pattnayak and the NeonIN Neonatal Infection Surveillance Network.



  • Competing interests None.

  • Provenance and Peer review Not commissioned; externally peer reviewed.

  • Ethics approval NeonIN received multi-centre research ethics committee approval from the London-Surrey Borders Research Ethic Committee 05/Q0806/34 in 2005.