- Email: [email protected]
Antenatal prevention of neonatal group B streptococcal infection
Reviews in Gynaecological and Perinatal Practice 6 (2006) 218–225 www.elsevier.com/locate/rigapp
Antenatal prevention of neonatal group B streptococcal infection Sophie Beal a,*, Stephanie Dancer b b
a St James’s University Hospital, Leeds LS9 7TJ, United Kingdom Southern General Hospital, 1345 Govan Road, Glasgow G51 4TF, United Kingdom
Received 16 December 2005; accepted 15 May 2006 Available online 27 June 2006
Abstract Group B streptococci can be isolated from the vagina of 15–40% of pregnant women. Vertical transmission to the infant occurs in 50% of deliveries involving colonised women. Most infants remain asymptomatic, but 1–2% develop clinical infection, which is associated with significant morbidity and mortality. Vertical transmission can be successfully prevented by intrapartum administration of antibiotics. Other proposed methods include vaccines and intrapartum vaginal or neonatal washing with antiseptics. Selection of women for prophylactic antibiotics can be based on risk factors, screening or a combination of both. Benefits of prophylaxis should be balanced against cost, medicalisation of labour and the risks of anaphylaxis and bacterial resistance. We present an overview of vaginal group B streptococcal isolation methods and antenatal strategies for prevention of neonatal infection. # 2006 Elsevier B.V. All rights reserved. Keywords: Group B streptococcus; Streptococcus agalactiae; Screening; Antibiotic prophylaxis; Pregnancy complications (infectious); Vaccines
1. Introduction Maternal genital tract colonisation with Streptococcus agalactiae or Group B streptococci (GBS) is known to be associated with neonatal colonisation, and neonatal and maternal puerperal infections. GBS may contribute to intrauterine infection leading to stillbirth, neonatal pneumonia, meningitis and septicaemia. It has also been implicated in mid trimester miscarriage and premature rupture of membranes and is frequently associated with maternal postnatal endometritis, wound infection, asymptomatic bacteriuria and urinary tract infections [1]. Perinatal GBS infections were first described in the 1960s. By the 1970s GBS were the leading cause of neonatal infection and one of the most important causes of maternal endometritis and septicaemia. Initially neonatal GBS infection was associated with 20–50% mortality. Morbidity and mortality rates have both improved with advances in detection, prophylaxis and treatment [2]. * Corresponding author. E-mail addresses: [email protected] (S. Beal), [email protected] (S. Dancer). 1871-2320/$ – see front matter # 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.rigapp.2006.05.006
In the early 1980s clinical trials demonstrated vertical GBS transmission could be interrupted by the use of prophylactic intrapartum antibiotics (IAP) [3]. There is considerable controversy as to the best method of selection of parturient women for prophylaxis. The Royal College of Obstetricians and Gynaecologists (RCOG) has decided to opt for risk-based management rather than the universal screening currently advocated by the US Centre for Disease Control (CDC) [4,5]. This review examines the epidemiology of this organism, methods of identification and antenatal strategies for the prevention of neonatal infection.
2. Data selection We performed Medline, Embase and Cochrane database searches using the following keywords Group B streptococcus; S. agalactiae; prophylaxis; treatment; diagnosis; isolation; rapid testing; risk factors; screening; molecular biology; vaccines; polymerase chain reaction; latex agglutination; antigen detection; serotyping. We also looked at GBS guidelines produced by the RCOG, CDC and the Canadian Taskforce on Preventative Healthcare (CTFPHC) [4–6].
S. Beal, S. Dancer / Reviews in Gynaecological and Perinatal Practice 6 (2006) 218–225
3. Definition of used terms Colonisation is defined as the presence of a potential pathogen, e.g. GBS, that can be cultured without causing symptoms or disease. Infection is defined as the presence of clinical symptoms or illness due to a microorganism. Neonatal early-onset GBS infection is defined as GBS infection occurring within the first 7 days of life. In this review, transmission is describes the transmission of a potential pathogenic organism from one individual to another. Vertical transmission describes the transmission of such an organism from mother to fetus. Sepsis is a syndrome characterised by signs of infection and accompanied by septicaemia.
4. Colonisation and transmission It has been estimated that GBS can be cultured from the vagina of 15–40% of pregnant and postpartum women [1,7]. The gastro-intestinal tract acts as a reservoir for GBS in both men and women [6]. There is vertical transmission to the neonate in approximately 50% of pregnancies where the mother’s vagina is colonised [8]. Transmission is more likely if there is heavy maternal colonisation, maternal GBS bacteriuria, ruptured membranes, pre-term delivery or intrapartum pyrexia [9]. Nosocomial transmission between neonates may occasionally occur [2].
5. Infection 5.1. Neonatal infection Most neonates carrying GBS remain asymptomatic but in 1–2% of cases the neonate develops life-threatening infection with a risk of major long-term morbidity in survivors [1]. The most recent research estimates this to occur in 0.7/1000 live births within the UK [10]. The incidence varies geographically. GBS early-onset infection (EOI) occurs within the first 7 days of life, but most cases are evident within 72 h [2]. It is most commonly manifest as pneumonia or meningitis or has an unidentifiable focus of infection [1]. In the UK it is usually associated with serotypes 3 (38%), 1a (32%), and 5 (13%) [11]. Overall, the UK and US have similar incidences of 0.5/1000 live births [4,10]. The incidence within the UK varies from 0.21/1000 live births in Scotland to 0.73/1000 live births in Northern Ireland [10]. Late onset infection, associated predominantly with serotype 3, may occur at any time after the first week, even months later [10]. Cases are evenly distributed throughout the remaining first 90 days of life and generally present with either bacteraemia or meningitis. Less frequently, it may cause focal infections such as osteomyelitis, septic arthritis or cellulitis [1,2]. It is not always associated with maternal
219
vaginal colonisation and may represent horizontal transmission from hospital or community sources [1,12]. The risk of nosocomial transmission is increased with cramped nursery conditions, poor hand hygiene, prolonged hospital stay and high maternal population colonisation rates [2]. Mortality continues to decline due to neonatal care advances. It is greater in EOI (6% in term and 18% in preterm babies in the UK) [5]. 5.2. Maternal infection GBS is identified in 15% chorioamnionitis, 16% endometritis, 2–15% wound infections, 15% bacteraemia, 9–15% stillbirth and a 2–4% of maternal urinary tract infections [2,4]. GBS often co-exists with other bacteria in chorioamnionitis and endometritis [2]. Most maternal GBS infections, including bacteraemia, respond quickly to antibiotics, but bacteraemia is occasionally associated with abdominal abscess, necrotising fasciitis and meningitis [1].
6. Microbiological techniques GBS is a Gram-positive facultative coccus. Approximately 99% of GBS show b-haemolysis on blood agar plates [2]. 6.1. Isolation Isolation rates depend on the clinical and laboratory techniques used. They are improved when more than one appropriate site is swabbed (lower vagina, peri-urethral or ano-rectal areas). Combined antenatal vaginal and rectal swabs have a greater positive predictive value for intrapartum vaginal colonisation. Boyer et al. found a 17% vertical transmission rate when the rectal culture was positive but the vaginal culture negative [13]. The use of broth and antibiotic-containing media produce better isolation rates than non-selective media. One example of such is Todd-Hewitt broth developed for Gram-positive organisms, supplemented with nalidixic acid and either gentamicin or colistin (Lim Broth). In Ferreri and Blair’s study, GBS were isolated in 37% of women using antibioticcontaining broth but not agar [14]. However, all traditional culture techniques take at least 36 h, which means that the results of an intrapartum swab will usually be delayed until after delivery. The CDC defines heavy GBS colonisation as GBS isolated using standard agar techniques alone rather than antibiotic media. It is associated with a higher risk of neonatal transmission and infection [4]. 6.2. Identification 6.2.1. Traditional culture techniques Definitive GBS identification requires serologic detection of the group B carbohydrate antigen, however clinical
220
S. Beal, S. Dancer / Reviews in Gynaecological and Perinatal Practice 6 (2006) 218–225
laboratories generally use presumptive tests [1,2]. Identification, based on colonial morphology and Gram stain can often be made within 24 h of agar plating with final confirmation available within 48 h. GBS is differentiated from other streptococci using a combination of tests: the Christie, Atkins, and Munch-Peterson (CAMP) test, bacitracin sensitivity and bile esculin reaction [2]. 6.2.2. Rapid testing An ideal test would be rapid, sensitive and easy to use. Rapid immunological tests studied include the use of antigen detection (including co-agglutination, enzyme immunoassay and latex agglutination) and DNA hybridisation techniques, including polymerase chain reaction (PCR) [15]. PCR is emerging as the preferred method. Although latex agglutination and PCR both appear to have an excellent sensitivity and specificity, PCR provides the most rapid results. One hundred and twelve women were enrolled in a 2000 study assessing a real-time (fluorogenic) PCR assay [16]. Standard GBS culture, conventional PCR assay and rapid PCR assay were performed on rectal, vaginal and recto-vaginal swabs from women in labour. Compared with culture results, both PCR assays had a sensitivity of 97% and specificity of 100%. Real-time PCR results were available within 45 min. Conventional PCR took 100 min. The US Federal Drug Agency (FDA) has recently approved the intrapartum use of this test. If it proves to be cost effective on a large scale, it may supersede both the UK riskbased and the current US universal screening approach to antibiotic prophylaxis [2]. 6.2.3. Serotyping Until recently, most reference laboratories performed serotyping primarily by use of the Lancefield capillary precipitation test. This time-consuming method requires experienced staff. Latex agglutination is beginning to replace Lancefield typing. In one study, 203 out of 232 isolates were serotyped using latex agglutination, while the capillary precipitation test serotyped 184 isolates. Latex agglutination appeared to reduce the number of non-typable isolates by almost 50% [17].
7. Prevention 7.1. Intrapartum antibiotic administration Vertical transmission during labour can be successfully interrupted by intrapartum intravenous ampicillin [3], whereas oral antibiotic therapy during pregnancy fails to prevent maternal GBS colonisation at delivery in 30% of cases [18]. Benefits of antibiotic prophylaxis should be balanced against the cost, medicalisation of labour, risks of anaphylaxis, development of antibiotic resistance and the possible increase in non-GBS neonatal infections [5,19].
7.2. Specific management issues 7.2.1. Management of pre-term prolonged rupture of membranes Prematurity and prolonged rupture of membranes (PROM) both increase the risk of neonatal infection. Following the ORACLE trial, it is recommended that oral erythromycin be administered for 10 days or until delivery to all women in cases of PROM as this delays delivery [20]. This trial compared the use of prophylactic co-amoxiclav and erythromycin to no antibiotics in pre-term pre-labour ruptured membranes, regardless of GBS status. Data published recently from the ORACLE trial confirms a significant reduction in GBS neonatal colonisation, infection and mortality [21]. These findings are similar to those found in another large double-blind randomised controlled trial in Tennessee, US [22]. Specific antenatal prophylaxis for GBS colonisation is not considered necessary by the RCOG in the presence of PROM [5]. The term PROM study, a large non-blinded randomised controlled trial, reported a drastic reduction in neonatal transmission rates in cases of maternal colonisation with PROM at term when labour was induced with oxytocinon, rather than priming with prostaglandin or managed expectantly (2.5% versus >8%) [23]. 7.2.2. Caesarean section It is generally agreed that there is no need to give specific GBS prophylaxis during elective caesarean section, as the risk of GBS infection across intact membranes is very low. In a single retrospective study, no babies were identified with GBS infection following term caesarean section [24]. 7.3. Disadvantages of widespread intrapartum antibiotic use 7.3.1. Antibiotic resistance of GBS Although there has been no record of penicillin-resistant GBS organisms to date, some strains have developed resistance to macrolide and lincosamide antibiotics (erythromycin and clindamycin) [25]. Morales et al. found a significant increase in the proportion of GBS isolates resistant to erythromycin between those retrieved in 1997– 1998 and those retrieved in 1980–1983, and new resistance to clindamycin, associated with increased IAP use [26]. Clindamycin and erythromycin should be used only when penicillin is contraindicated. 7.3.2. Proliferation of non-GBS pathogens Even a broad-spectrum antibiotic such as ampicillin may result in the proliferation of Clostridium difficile, yeasts and ampicillin-resistant Gram-negative rods [19]. Stable or declining rates of non-GBS sepsis have been reported in the largest multi-centre studies, which have been based on case note review [4]. In a retrospective singlecentre study 30 infants were identified with early-onset E.
S. Beal, S. Dancer / Reviews in Gynaecological and Perinatal Practice 6 (2006) 218–225
coli infection from 61,500 deliveries over a 10-year period [27]. A rise in E. coli infections associated with increased IAP use was confined to infants weighing 1000–1500 g. However, Towers prospectively examined neonatal sepsis developing within the first 7 days of life and its relationship to IAP over a 6-year period [28]. Although this was a singlecentre study, a cohort of 30,000 deliveries was represented. Mothers had received IAP in 15 of 27 cases (56%) of nonGBS neonatal infection. During this time the overall administration of IAP had only increased to a maximum of 17%. Ampicillin-resistant organisms were responsible for 87% of infections in infants born to mothers who had received IAP. They caused only 17% of infections where IAP had not been given. 7.3.3. Anaphylaxis The estimated incidence of anaphylaxis due to penicillin is 1 in 10,000 people treated. This can be fatal in as many as a tenth of these. The RCOG estimates that universal screening where 30% of women were given intrapartum penicillin would result in an average of two UK deaths a year. Screening based on a single risk factor, where IAP was administered to 15% would result in one [5]. 7.4. Choice of antibiotics Ampicillin can be used for intrapartum prophylaxis, but many authorities prefer benzylpenicillin (penicillin G), because of its favourable pharmacokinetics and narrow spectrum of activity. Theoretically, these features should reduce the proliferation of non-GBS and antibiotic resistant microorganisms. However, ampicillin has been used in most clinical studies of intrapartum prophylaxis. Edwards et al. carried out a randomised controlled trial comparing ampicillin with benzylpenicillin intrapartum prophylaxis [29]. Although postpartum maternal culture rates of antibiotic resistant E. coli and Enterobacteriaceae spp. were higher than entry culture rates in both groups, there was no significant difference between the groups. Neonatal cultures were not performed. The sample size had been calculated to detect a 50% increase in the proportion of
221
antibiotic resistant organisms within the ampicillin group compared to the penicillin group. A smaller difference would not have reached statistical significance. In a recent small case-control study, infants with lateonset serious bacterial infections (defined as the presence of pathogenic bacteria cultured from the blood, cerebro-spinal fluid or urine) were much more likely to have received broad-spectrum intrapartum antibiotics than healthy infants. There was no significant difference in the numbers in each group that had received benzylpenicillin [30]. 7.5. Antenatal prophylaxis guidelines (Table 1) 7.5.1. Current RCOG guidelines 7.5.1.1. Background. Until recent RCOG guidelines were produced, there had been no consensus within the UK on GBS prophylaxis. CDC recommendations for universal screening were not necessarily appropriate to the UK. The UK incidence of EOI is similar to that seen in the US with widespread use of intrapartum antibiotics (0.5/ 1000 live births), despite similar vaginal colonisation rates [5]. The CDC guidelines were mainly based on data from a single large retrospective case note review study [4]. No randomised controlled trials directly comparing universal screening and a risk-based approach for selecting women for IAP have yet been carried out. Little UK data exist to support the use of either universal screening or risk factors to select women for administration of IAP [5]. Major organisational and funding changes would be required in the UK if universal screening were introduced to ensure quality and equality of service. Most UK laboratories, for instance, do not use the selective media required for optimal GBS culture because of cost. RCOG guidelines were systematically developed using standardised RCOG methodology [5]. Critical appraisal of the literature focused on the evidence provided by the CDC to support their universal screening strategy. The CTFPHC guidelines were also examined. These and the CDC guidelines are discussed below.
Table 1 Strategies for prevention of neonatal GBS infection 2003 RCOG guidelines [5]
2002 CDC guidelines [4]
2001 CTFPHC guidelines [6]
Criteria for IAP
<37 weeks, membranes ruptured >18 h, pyrexia >38 8C, previous affected infant GBS bacteriuria
Swabs at 35–37 weeks gestation, delivery prior to 37 weeks gestation
Pre-labour swabs performed on women with risk factors at 35–37 weeks (risk factors as for RCOG)
Culture technique
Does not rely on swabbing
Selective broth media
Selective broth media
Swabs used
None
Combined ano-vaginal swabs (can be performed by patient)
Combined recto-vaginal swabs
IAP
Intravenous penicillin (or clindamycin if allergic)
Intravenous penicillin (or clindamycin if allergic)
Intravenous penicillin (or clindamycin if allergic)
Neonatal antibiotics
Prompt antibiotic treatment if unwell; observation for 12 h if well
Prompt antibiotic treatment if unwell; observation for 24 h if well
No guidelines
222
S. Beal, S. Dancer / Reviews in Gynaecological and Perinatal Practice 6 (2006) 218–225
The RCOG guideline group concluded that it was not clear the benefits of universal screening outweighed the risks. Risk-based guidelines were largely formed by expert consensus following mathematical modelling to estimate the effects of proposed strategies within the UK [5]. 7.5.1.2. Recommendations. The RCOG guidelines recommend that IAP should be considered for mothers who have: had a baby previously with GBS infection; PROM; or been found incidentally to have GBS colonisation of the vagina. Antibiotics should also be offered to women with GBS bacteriuria. The argument for IAP strengthens in the presence of two or more risk factors. Antibiotics are not recommended to treat GBS colonisation prior to labour or for women with undergoing a planned caesarean section if the membranes are intact. Usually benzylpenicillin should be given, but intravenous clindamycin should be used in penicillin-allergic women [5]. 7.5.2. CDC guidelines 7.5.2.1. Background. The 1996 CDC guidelines recommended the use of one of two strategies: In the first, all pregnant women were to be screened using an ano-vaginal swab at 35–37 weeks gestation [31]. Swabs were to be cultured using selective broth media. Antibiotics were to be administered to all women identified as GBS colonised or to those where colonisation status was unknown at delivery. Alternatively, women were to receive IAP if they had one or more risk factors; delivery at less than 37 weeks gestation, intrapartum temperature greater than 38 8C or rupture of membranes for more than 18 h. Regardless of the approach decided upon, prophylaxis was also to be administered to women with bacteriuria or who had previously delivered an infant with EOI. The introduction of these guidelines was associated with a dramatic decline in the incidence of earlyonset but not late-onset GBS infection [4]. New 2002 CDC guidelines were produced following critical appraisal of new evidence. The CDC had sponsored a multi-state case note review of a randomised stratified sample of 629,912 live births (5144 live births), managed according to the 1996 guidelines. They found the universal screening approach to be 50% more effective than the risk factor based approach in prevention of neonatal GBS infection. Eighteen percent of women in the universal screening group who were identified as colonised with GBS did not present with risk factors and would not have been selected for intrapartum antibiotics unless they had been screened [32]. This study provided the main basis for the 2002 guideline changes and was the first large scale comparison of the 1996 CDC risk-based and universal screening strategies [32]. This was retrospective observational data and not part of a prospective controlled trial. It was assumed that all women without a documented culture had been managed according to the risk-based approach. This is likely to have produced bias, as others not managed according to the guidelines
would have been included in this group. To reduce this bias, all women who did not undergo antenatal screening and had risk factors, but did not receive IAP were excluded from the study. Confirmation of GBS infection may be more difficult in neonates where intrapartum antibiotics have been transferred across the placenta. Whilst universal screening results in a decreased incidence of EOI, there has been no agreement on a corresponding reduction in neonatal infections as a whole [2]. The implication is that there may be a significant proportion of neonates infected with GBS from whom cultures taken are negative as a result of IAP. Culture negative clinical infection should probably be used as an outcome in future research. 7.5.2.2. 2002 CDC guidelines. Current CDC recommendations for prevention of vertical transmission involve universal screening at 35–37 weeks gestation using a combined vaginal and ano-rectal swab that can be performed by women themselves. All colonised women should receive IAP. Prophylaxis is also advised for women with bacteriuria or who have previously delivered an infant that subsequently developed EOI. Where culture results are not available women should be managed according to risk factors [4]. A specific algorithm is provided for managing women with threatened pre-term labour. Unless a GBS urinary tract infection is diagnosed, antenatal antibiotics should not be prescribed to treat GBS or given to women having planned caesarean sections. Specimens should be placed in a non-nutritive transport medium (e.g. Amies or Stuart’s without charcoal). They should be inoculated into a selective broth media such as LIM broth, incubated overnight and then plated onto solid blood agar medium. In the penicillin-allergic woman susceptibility testing for clindamycin and erythromycin should be performed. Intravenous benzylpenicillin is recommended as IAP for those without a penicillin allergy as in the RCOG guidelines. Ampicillin is a less favourable option. In women with confirmed penicillin allergy, clindamycin or erythromycin is advised. 7.5.3. CTFPHC guidelines The CTFPHC recommends universal screening and administration of IAP to colonised women with risk factors (pre-term labour less than 37 weeks gestation, prolonged rupture of membranes for more than 18 h, maternal intrapartum temperature of greater than 38 8C, antenatal GBS bacteriuria) and those who have previously delivered a newborn with GBS infection regardless of their GBS colonisation status [6]. The CTFPHC recommend the same sample collection and culture techniques for universal screening as the CDC. They also suggest the same antibiotic regimes for IAP [4,6]. The guidelines were produced by expert consensus following systematic review of the literature. They were
S. Beal, S. Dancer / Reviews in Gynaecological and Perinatal Practice 6 (2006) 218–225
developed prior to the production of the 2002 CDC guidelines and the publication of observational US multi-state data. Like the RCOG and CDC, the CTFPHC failed to find any randomised controlled trials demonstrating a significant reduction in GBS infection associated with universal screening or risk factor approaches. However, examination of cumulative evidence from cohort studies showed reductions in early-onset infection when selection for IAP was based on universal screening, alone or combined with a single risk factor [6]. The strongest evidence to support the use of a combined universal screening and risk-based approach was provided by a single randomised controlled trial from the US. Colonised women in labour before 37 weeks gestation or with ruptured membranes for longer than 12 h, were randomised to receive either IAP or no antibiotics. Neonates whose mothers had received IAP showed a 51% reduction in GBS colonisation rates. Bacteraemia occurred in 6% of neonates within the control group, but in none of the treatment group [3]. Excluding women with intrapartum pyrexia or antenatal GBS bacteriuria is likely to have reduced the effect of IAP on neonatal colonisation and bacteraemia. These are both significant risk factors for neonatal infection. Weaknesses of the study include the relatively small numbers (80 women) in each group and the lack of blinding. As previously discussed, confirmation of GBS infection may be more difficult in neonates where antibiotics have been transferred across the placenta and the treatment effect may have been exaggerated. 7.6. Relative cost of universal screening and risk-based strategies The ideal IAP strategy would result in the minimum number of women necessary receiving IAP for the greatest possible reduction in EOI. Adverse antibiotic effects and costs would both be minimised. Case note review has suggested that perfect implementation of both risk-based and universal screening strategies in the US will result in a similar proportion receiving antibiotics [4,32]. However, the RCOG estimate that around 30% of women would receive IAP if universal GBS screening were introduced into the UK. Approximately 15% of women within the UK would receive antibiotics using the previous CDC risk-based strategy. These estimates assume a UK prevalence of 25% maternal GBS colonisation, based on a single 1986 UK study [5]. This may not reflect the current situation. A risk-based approach costs less initially as it avoids sample collection and laboratory costs. Screening costs need to be balanced against the cost of the intensive care for neonates with GBS infection. The cost of adverse effects associated with IAP should also be considered. US analyses of cost initially appear to calculate the overall cost of universal screening in the US to be similar to that of single risk factor based prophylaxis [4]. These studies included the additional cost of administering 48 h of
223
antibiotic prophylaxis to all neonates of mothers given IAP and prolonging hospital stay. The CDC, RCOG or CTFPHC do not currently recommend routine neonatal antibiotic prophylaxis [4–6]. The risk-based approach was found to be significantly less expensive than universal screening after excluding these costs. However, intensive care costs for neonates with GBS infection were not compared between the two strategies [33,34]. A randomised control trial comparing the cost-effectiveness of risk factor and universal screening approaches has been recommended by the RCOG. This would involve introducing optimum sample collection and laboratory techniques to participating hospitals. The RCOG estimate that to prevent one neonatal death by universal screening at least 24,000 UK women would have to be swabbed and 7000 would have to be given IAP. Using the previous CDC riskbased strategy, 625 UK women would have to receive IAP to prevent one case of EOI and 5882 women to prevent one neonatal death. It may be impossible to recruit enough UK women to give the suggested trial sufficient power [5].
8. Proposed alternative prevention strategies 8.1. Antiseptics In vitro studies have shown some microbicides to have strong activity against GBS. Chlorhexidine is cheap and may have a role in reducing GBS transmission in developing countries where antibiotics are not readily available. It has been found to have little or no impact on antibiotic resistance [2]. If proved effective, its widespread use would reduce the risk of antibiotic resistance (particularly ampicillin-resistant E. coli) by reducing antibiotic use. In several clinical trials, both neonatal GBS transmission and infection have been reduced by intrapartum vaginal application of chlorhexidine, or by neonatal washing at birth with chlorhexidine. Other studies have not supported these findings [2,35]. This may be due to heterogeneity of methodology between studies [35]. The large randomised controlled trial with the highest methodological quality and complete follow-up reported a significant reduction in the rates of respiratory distress syndrome, diagnosed or probable infection and admission to the neonatal unit [35,36]. However, only one case of confirmed EOI in each of the study and control groups was reported. Neonatal colonisation rates were not reported. Women with a high risk of vertical transmission (pre-term delivery or previous child with GBS infection) were excluded possibly diminishing the effect seen in the chlorhexidine group. A Cochrane systematic review selecting randomised and quasi-randomised trials comparing vaginal disinfection with chlorhexidine to placebo, or no treatment, found evidence of reduction in neonatal colonisation, but no definite effect on EOI [35].
224
S. Beal, S. Dancer / Reviews in Gynaecological and Perinatal Practice 6 (2006) 218–225
More definite effects have been seen in large prospective, but non-randomised trials. A large Malawi hospital trial of vaginal wiping and neonatal bathing with 0.25% chlorhexidine showed reduced neonatal morbidity and mortality, and maternal infection. There was no blinding, and assignment to treatment or control group was based on month of delivery [37]. Findings were similar in another large prospective study where labouring women received vaginal douching with either saline or a similar strength (0.2%) chlorhexidine solution [38]. 8.2. Vaccines The risk of neonatal infection is indirectly proportional to the levels of specific capsular polysaccharide antibodies in the maternal serum [39]. Protective IgG GBS antibodies appear to be transferred across the placenta. Unconjugated vaccines based on capsular polysaccharides have been shown to be only variably immunogenic. The covalent coupling of a protein antigen to a capsular polysaccharide stimulates T cells and significantly increases immunogenicity [40]. Vaccine composition is likely to differ depending on serotype distribution in each country. GBS types 1a, 1b, 2, 3 and 5 cause almost all maternal and neonatal GBS infection in the USA and UK. Phase 1 and 2 trials in healthy nonpregnant, and phase 1 trials in healthy pregnant women have already shown conjugate polysaccharide vaccines against these serotypes to be safe and highly immunogenic. Funding has not yet been secured for phase 3 trials, however, due to the litigation risk of such a trial in pregnant women [41]. Transplacental transport is less before 32–24 weeks gestation, which may limit the benefit of antenatal vaccines. Ideally such a vaccine could be given to teenage girls prior to pregnancy [2]. Systemic reactions to the vaccines have occurred in less than 2% of almost 500 healthy young adults [41]. These included low-grade fever, chills, headache or myalgia and resolved within 36 h. Immunization would hopefully prevent both early and lateonset neonatal infection by reducing the risk of vertical transmission. Theoretically, it may reduce or even eliminate GBS by mucosal immunological responses and may therefore reduce adverse pregnancy outcomes such as pre-term delivery, stillbirth and maternal infection. Vaccine use would also reduce the risk of further antibiotic resistance [2]. Recently the genomes of different GBS strains have been described, with implications for analysis and vaccine development [42].
9. Conclusions GBS is a major cause of infectious morbidity and mortality in the neonate. Vertical transmission can be successfully prevented in the majority of pregnancies by intrapartum administration of antibiotics. This benefit needs
to be balanced against the disadvantages of widespread intrapartum antibiotics, which include the potential for increased antibiotic resistance, proliferation of non-GBS organisms and anaphylaxis. At present, the RCOG recommend a risk-based strategy to select women for IAP. This contrasts with the US system of universal antenatal screening at 35–37 weeks gestation. Neither strategy has been proven to be superior. Major organisational and funding changes would be required in the UK if universal screening were introduced to ensure quality and equality of service. Further research is needed to ascertain the most appropriate and cost effective strategy for the UK.
References [1] Baker CJ, Edwards MS. Group B streptococcal infections. In: Remington J, Klein JO, editors. Infectious diseases of the newborn and infant. 5th ed. WB Saunders: Philadelphia, PA; 2000 . p. 1091– 156. [2] Gibbs RS, Schrag S, Schuchat A. Perinatal infections due to Group B streptococcus. Obstet Gynecol 2004;104(5 Pt 1):1062–76. [3] Boyer KM, Gotoff SP. Prevention of early-onset neonatal group B streptococcal disease with selective intrapartum chemoprophylaxis. New Engl J Med 1986;314:1665–9. [4] Centers for Disease Control Prevention. Prevention of perinatal group B streptococcal disease: revised guidelines from CDC. Morb Mortal Wkly Rep 2002;51:1–22. [5] Royal College of Obstetricians and Gynaecologists. Prevention of early onset neonatal group B streptococcal disease. Guideline No 36. London: RCOG; 2003. [6] Canadian Task Force on Preventative Health Care. Prevention of earlyonset group B streptococcal disease in the newborn: systematic review and recommendations. CTFPHC Technical Report 01-6. London ON: CTF; 2001. [7] Anthony BF, Eisenstadt R, Carter J, Kim KS, Hobel CJ. Genital and intestinal carriage of group B streptococci during pregnancy. J Infect Dis 1981;143:761–6. [8] Pass MA, Gray BM, Khare S, Dillon Jr HC. Prospective studies of group B streptococcal infections in infants. J Pediatr 1979;95:437–43. [9] Schuchat A, Deaver-Robinson K, Plikaytis BD, Zangwill KM, MohleBoetani J, Wenger JD. Multistate case-control study of maternal risk factors for neonatal group B streptococcal disease. Pediatr Infect Dis J 1994;13:623–9. [10] Heath PT, Balfour G, Weisner AM, Efstratiou A, Lamagni TL, Tighe H, et al. Group B streptococcal disease in UK and Irish infants younger than 90 days. Lancet 2004;363:292–4. [11] Weisner AM, Johnson AP, Lamagni TL, Arnold E, Warner M, Heath PT, et al. Characterisation of group B streptococci recovered from infants with invasive disease in England and Wales. Clin Infect Dis 2004;38:1203–8. [12] Noya FJD, Rench MA, Metzger TG, Colman G, Naidoo J, Baker CJ. Unusual occurrence of an epidemic of type Ib/c group B streptococcal sepsis in a neonatal intensive care unit. J Infect Dis 1987;155(6):1135– 44. [13] Boyer KM, Gadzala CA, Kelly PD, Burd LI, Gotoff SP. Selective intrapartum chemoprophylaxis of neonatal group B streptococcal early-onset disease. II. Predictive value of prenatal cultures. J Infect Dis 1983;148:802–9. [14] Ferreri P, Blair LL. Pharyngeal carriage of group B streptococci: detection by three methods. J Clin Microbiol 1977;6:136–9. [15] Yancey MK, Armer T, Clark P, Duff P. Assessment of rapid identification tests for genital carriage of group B streptococci. Obstet Gynecol 1992;80:1038–47.
S. Beal, S. Dancer / Reviews in Gynaecological and Perinatal Practice 6 (2006) 218–225 [16] Bergeron MG, Danbing K, Me´nard C, Francois FJ, Gagnon M, Bernier M, et al. Rapid detection of group B streptococci in pregnant women at delivery. New Engl J Med 2000;343:175–9. [17] Slotved HC, Elliott J, Thompson T, Konradsen HB. Latex assay for serotyping group B Streptococcus isolates. J Clin Microbiol 2003;41: 4445–7. [18] Hall RT, Barnes W, Krishnan L, Harris DJ, Rhodes PG, Fayez J, et al. Antibiotic treatment of parturient women colonized with group B streptococci. Am J Obstet Gynecol 1976;124:630–4. [19] Dancer SJ. How antibiotics can make us sick: the less obvious adverse effects of antimicrobial chemotherapy. Lancet Infect Dis 2004;4(10): 611–9. [20] Kenyon SL, Taylor DJ, Tarnow-Mordi W, ORACLE Collaborative Group. Broad-spectrum antibiotics for pre-term, pre-labour rupture of fetal membranes: the ORACLE I randomised trial. Lancet 2001; 357(9261):979–88. [21] Gilbert RE, Pike K, Kenyon SL, Tarnow-Mordi W, Taylor DJ. The effect of pre-partum antibiotics on the type of neonatal bacteraemia: insights from the MRC ORACLE trials. BJOG 2005;112(6): 830–2. [22] Mercer BM, Miodovnik M, Thurnau GR, Goldenberg RL, Das AF, Ramsey RD, et al. Antibiotic therapy for reduction of infant morbidity after pre-term premature rupture of the membranes. A randomized controlled trial. JAMA 1997;278(12):989–95. [23] Hannah ME, Ohlsson A, Wang EE, Matlow A, Foster GA, Willan AR, et al. Maternal colonisation with group B streptococcus and pre-labour rupture of membranes at term: the role of induction of labour. Term PROM Study Group. Am J Obstet Gynecol 1997;177(4):780–5. [24] Ramus RM, McIntyre DD, Wendel Jr GD. Antibiotic prophylaxis for group B Streptococcus is not necessary with elective caesarean section at term (abstract). Am J Obstet Gynecol 1999;180:S85. [25] Berkowitz K, Regan JA, Greenberg E. Antibiotic resistance patterns of group B streptococci in pregnant women. J Clin Microbiol 1990;28(1): 5–7. [26] Morales WJ, Dickey SS, Bornick P, Lim DV. Change in antibiotic resistance of group B streptococcus: impact on intrapartum management. Am J Obstet Gynecol 1999;181(2):310–4. [27] Joseph TA, Pyati SP, Jacobs N. Neonatal early-onset Escherichia coli disease. The effect of intrapartum ampicillin. Arch Pediatr Adolesc Med 1998;152(1):35–40. [28] Towers CV, Carr MH, Padilla G, Asrat T. Potential consequences of widespread antepartal use of ampicillin. Am J Obstet Gynecol 1998; 179(4):879–83. [29] Edwards RK, Clark P, Sistrom CL, Duff P. Intrapartum antibiotic prophylaxis 1: relative effects of recommended antibiotics on gramnegative pathogens. Obstet Gynecol 2002;100(3):534–9.
225
[30] Glasgow TS, Young PC, Wallin J, Kwok C, Stoddard G, Firth S, et al. Association of intrapartum antibiotic exposure and late-onset serious bacterial infections in infants. Pediatrics 2005;116(3):696–702. [31] Centers for Disease Control Prevention. Prevention of perinatal group B streptococcal disease: a public health perspective. Morb Mortal Wkly Rep 1996;45(RR7):1–24. [32] Schrag SJ, Zell ER, Lynfield R, Roome A, Arnold KE, Craig AS, et al. A population-based comparison of strategies to prevent early-onset group B streptococcal disease in neonates. New Engl J Med 2002; 347(4):233–9. [33] Mohle-Boetani JC, Schuchat A, Plikaytis BD, Smith JD, Broome CV. Comparison of prevention strategies for neonatal group B streptococcal infection. A population-based economic analysis. JAMA 1993;270(12):1442–8. [34] Fargason Jr CA, Peralta-Carcelen M, Rouse DJ, Cutter GR, Goldenberg RL. The pediatric costs of strategies for minimizing the risk of early-onset group B streptococcal disease. Obstet Gynecol 1997; 90(3):347–52. [35] Stade B, Shah V, Ohlsson A. Vaginal chlorhexidine during labour to prevent early-onset neonatal group B streptococcal infection [Systematic Review]. Cochrane Pregnancy and Childbirth Group. Cochrane Database Syst Rev 2005;4. [36] Burman LG, Christensen P, Christensen K, Fryklund B, Helgesson AM, Svenningsen NW, et al. Prevention of excess neonatal morbidity associated with group B streptococci by vaginal chlorhexidine disinfection during labour. The Swedish Chlorhexidine Study Group. Lancet 1992;340(8811):65–9. [37] Taha TE, Biggar RJ, Broadhead RL, Mtimavalye LA, Miotti PG, Justesen AB, et al. Effect of cleansing the birth canal with antiseptic solution on maternal and newborn morbidity in Malawi: clinical trial. Br Med J 1997;315:216–20. [38] Stray-Pedersen B, Bergan T, Hafstad A, Normann E, Grogaard J, Vangdal M. Vaginal disinfection with chlorhexidine during childbirth. Int J Antimicrob Agents 1999;12:245–51. [39] Baker CJ, Kasper DL. Correlation of maternal antibody deficiency with susceptibility to neonatal group B streptococcal infection. New Engl J Med 1976;294:753–6. [40] Kasper DL, Paoletti LC, Wessels MR, Guttormsen HK, Carey VJ, Jennings HL, et al. Immune response to type III group B streptococcal polysaccharide-tetanus toxoid conjugate vaccine. J Clin Invest 1996;98: 2308–14. [41] Baker CJ, Edwards MS. Group B streptococcal conjugate vaccines. Arch Dis Child 2003;88(5):375–8. [42] Maione D, Margarit I, Rinaudo CD, Masignani V, Mora M, Scarselli M, et al. Identification of a universal group B streptococcus vaccine by multiple genome screen. Science 2005;309(5731):148–50.
Antenatal prevention of neonatal group B streptococcal infection Sophie Beal a,*, Stephanie Dancer b b
a St James’s University Hospital, Leeds LS9 7TJ, United Kingdom Southern General Hospital, 1345 Govan Road, Glasgow G51 4TF, United Kingdom
Received 16 December 2005; accepted 15 May 2006 Available online 27 June 2006
Abstract Group B streptococci can be isolated from the vagina of 15–40% of pregnant women. Vertical transmission to the infant occurs in 50% of deliveries involving colonised women. Most infants remain asymptomatic, but 1–2% develop clinical infection, which is associated with significant morbidity and mortality. Vertical transmission can be successfully prevented by intrapartum administration of antibiotics. Other proposed methods include vaccines and intrapartum vaginal or neonatal washing with antiseptics. Selection of women for prophylactic antibiotics can be based on risk factors, screening or a combination of both. Benefits of prophylaxis should be balanced against cost, medicalisation of labour and the risks of anaphylaxis and bacterial resistance. We present an overview of vaginal group B streptococcal isolation methods and antenatal strategies for prevention of neonatal infection. # 2006 Elsevier B.V. All rights reserved. Keywords: Group B streptococcus; Streptococcus agalactiae; Screening; Antibiotic prophylaxis; Pregnancy complications (infectious); Vaccines
1. Introduction Maternal genital tract colonisation with Streptococcus agalactiae or Group B streptococci (GBS) is known to be associated with neonatal colonisation, and neonatal and maternal puerperal infections. GBS may contribute to intrauterine infection leading to stillbirth, neonatal pneumonia, meningitis and septicaemia. It has also been implicated in mid trimester miscarriage and premature rupture of membranes and is frequently associated with maternal postnatal endometritis, wound infection, asymptomatic bacteriuria and urinary tract infections [1]. Perinatal GBS infections were first described in the 1960s. By the 1970s GBS were the leading cause of neonatal infection and one of the most important causes of maternal endometritis and septicaemia. Initially neonatal GBS infection was associated with 20–50% mortality. Morbidity and mortality rates have both improved with advances in detection, prophylaxis and treatment [2]. * Corresponding author. E-mail addresses: [email protected] (S. Beal), [email protected] (S. Dancer). 1871-2320/$ – see front matter # 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.rigapp.2006.05.006
In the early 1980s clinical trials demonstrated vertical GBS transmission could be interrupted by the use of prophylactic intrapartum antibiotics (IAP) [3]. There is considerable controversy as to the best method of selection of parturient women for prophylaxis. The Royal College of Obstetricians and Gynaecologists (RCOG) has decided to opt for risk-based management rather than the universal screening currently advocated by the US Centre for Disease Control (CDC) [4,5]. This review examines the epidemiology of this organism, methods of identification and antenatal strategies for the prevention of neonatal infection.
2. Data selection We performed Medline, Embase and Cochrane database searches using the following keywords Group B streptococcus; S. agalactiae; prophylaxis; treatment; diagnosis; isolation; rapid testing; risk factors; screening; molecular biology; vaccines; polymerase chain reaction; latex agglutination; antigen detection; serotyping. We also looked at GBS guidelines produced by the RCOG, CDC and the Canadian Taskforce on Preventative Healthcare (CTFPHC) [4–6].
S. Beal, S. Dancer / Reviews in Gynaecological and Perinatal Practice 6 (2006) 218–225
3. Definition of used terms Colonisation is defined as the presence of a potential pathogen, e.g. GBS, that can be cultured without causing symptoms or disease. Infection is defined as the presence of clinical symptoms or illness due to a microorganism. Neonatal early-onset GBS infection is defined as GBS infection occurring within the first 7 days of life. In this review, transmission is describes the transmission of a potential pathogenic organism from one individual to another. Vertical transmission describes the transmission of such an organism from mother to fetus. Sepsis is a syndrome characterised by signs of infection and accompanied by septicaemia.
4. Colonisation and transmission It has been estimated that GBS can be cultured from the vagina of 15–40% of pregnant and postpartum women [1,7]. The gastro-intestinal tract acts as a reservoir for GBS in both men and women [6]. There is vertical transmission to the neonate in approximately 50% of pregnancies where the mother’s vagina is colonised [8]. Transmission is more likely if there is heavy maternal colonisation, maternal GBS bacteriuria, ruptured membranes, pre-term delivery or intrapartum pyrexia [9]. Nosocomial transmission between neonates may occasionally occur [2].
5. Infection 5.1. Neonatal infection Most neonates carrying GBS remain asymptomatic but in 1–2% of cases the neonate develops life-threatening infection with a risk of major long-term morbidity in survivors [1]. The most recent research estimates this to occur in 0.7/1000 live births within the UK [10]. The incidence varies geographically. GBS early-onset infection (EOI) occurs within the first 7 days of life, but most cases are evident within 72 h [2]. It is most commonly manifest as pneumonia or meningitis or has an unidentifiable focus of infection [1]. In the UK it is usually associated with serotypes 3 (38%), 1a (32%), and 5 (13%) [11]. Overall, the UK and US have similar incidences of 0.5/1000 live births [4,10]. The incidence within the UK varies from 0.21/1000 live births in Scotland to 0.73/1000 live births in Northern Ireland [10]. Late onset infection, associated predominantly with serotype 3, may occur at any time after the first week, even months later [10]. Cases are evenly distributed throughout the remaining first 90 days of life and generally present with either bacteraemia or meningitis. Less frequently, it may cause focal infections such as osteomyelitis, septic arthritis or cellulitis [1,2]. It is not always associated with maternal
219
vaginal colonisation and may represent horizontal transmission from hospital or community sources [1,12]. The risk of nosocomial transmission is increased with cramped nursery conditions, poor hand hygiene, prolonged hospital stay and high maternal population colonisation rates [2]. Mortality continues to decline due to neonatal care advances. It is greater in EOI (6% in term and 18% in preterm babies in the UK) [5]. 5.2. Maternal infection GBS is identified in 15% chorioamnionitis, 16% endometritis, 2–15% wound infections, 15% bacteraemia, 9–15% stillbirth and a 2–4% of maternal urinary tract infections [2,4]. GBS often co-exists with other bacteria in chorioamnionitis and endometritis [2]. Most maternal GBS infections, including bacteraemia, respond quickly to antibiotics, but bacteraemia is occasionally associated with abdominal abscess, necrotising fasciitis and meningitis [1].
6. Microbiological techniques GBS is a Gram-positive facultative coccus. Approximately 99% of GBS show b-haemolysis on blood agar plates [2]. 6.1. Isolation Isolation rates depend on the clinical and laboratory techniques used. They are improved when more than one appropriate site is swabbed (lower vagina, peri-urethral or ano-rectal areas). Combined antenatal vaginal and rectal swabs have a greater positive predictive value for intrapartum vaginal colonisation. Boyer et al. found a 17% vertical transmission rate when the rectal culture was positive but the vaginal culture negative [13]. The use of broth and antibiotic-containing media produce better isolation rates than non-selective media. One example of such is Todd-Hewitt broth developed for Gram-positive organisms, supplemented with nalidixic acid and either gentamicin or colistin (Lim Broth). In Ferreri and Blair’s study, GBS were isolated in 37% of women using antibioticcontaining broth but not agar [14]. However, all traditional culture techniques take at least 36 h, which means that the results of an intrapartum swab will usually be delayed until after delivery. The CDC defines heavy GBS colonisation as GBS isolated using standard agar techniques alone rather than antibiotic media. It is associated with a higher risk of neonatal transmission and infection [4]. 6.2. Identification 6.2.1. Traditional culture techniques Definitive GBS identification requires serologic detection of the group B carbohydrate antigen, however clinical
220
S. Beal, S. Dancer / Reviews in Gynaecological and Perinatal Practice 6 (2006) 218–225
laboratories generally use presumptive tests [1,2]. Identification, based on colonial morphology and Gram stain can often be made within 24 h of agar plating with final confirmation available within 48 h. GBS is differentiated from other streptococci using a combination of tests: the Christie, Atkins, and Munch-Peterson (CAMP) test, bacitracin sensitivity and bile esculin reaction [2]. 6.2.2. Rapid testing An ideal test would be rapid, sensitive and easy to use. Rapid immunological tests studied include the use of antigen detection (including co-agglutination, enzyme immunoassay and latex agglutination) and DNA hybridisation techniques, including polymerase chain reaction (PCR) [15]. PCR is emerging as the preferred method. Although latex agglutination and PCR both appear to have an excellent sensitivity and specificity, PCR provides the most rapid results. One hundred and twelve women were enrolled in a 2000 study assessing a real-time (fluorogenic) PCR assay [16]. Standard GBS culture, conventional PCR assay and rapid PCR assay were performed on rectal, vaginal and recto-vaginal swabs from women in labour. Compared with culture results, both PCR assays had a sensitivity of 97% and specificity of 100%. Real-time PCR results were available within 45 min. Conventional PCR took 100 min. The US Federal Drug Agency (FDA) has recently approved the intrapartum use of this test. If it proves to be cost effective on a large scale, it may supersede both the UK riskbased and the current US universal screening approach to antibiotic prophylaxis [2]. 6.2.3. Serotyping Until recently, most reference laboratories performed serotyping primarily by use of the Lancefield capillary precipitation test. This time-consuming method requires experienced staff. Latex agglutination is beginning to replace Lancefield typing. In one study, 203 out of 232 isolates were serotyped using latex agglutination, while the capillary precipitation test serotyped 184 isolates. Latex agglutination appeared to reduce the number of non-typable isolates by almost 50% [17].
7. Prevention 7.1. Intrapartum antibiotic administration Vertical transmission during labour can be successfully interrupted by intrapartum intravenous ampicillin [3], whereas oral antibiotic therapy during pregnancy fails to prevent maternal GBS colonisation at delivery in 30% of cases [18]. Benefits of antibiotic prophylaxis should be balanced against the cost, medicalisation of labour, risks of anaphylaxis, development of antibiotic resistance and the possible increase in non-GBS neonatal infections [5,19].
7.2. Specific management issues 7.2.1. Management of pre-term prolonged rupture of membranes Prematurity and prolonged rupture of membranes (PROM) both increase the risk of neonatal infection. Following the ORACLE trial, it is recommended that oral erythromycin be administered for 10 days or until delivery to all women in cases of PROM as this delays delivery [20]. This trial compared the use of prophylactic co-amoxiclav and erythromycin to no antibiotics in pre-term pre-labour ruptured membranes, regardless of GBS status. Data published recently from the ORACLE trial confirms a significant reduction in GBS neonatal colonisation, infection and mortality [21]. These findings are similar to those found in another large double-blind randomised controlled trial in Tennessee, US [22]. Specific antenatal prophylaxis for GBS colonisation is not considered necessary by the RCOG in the presence of PROM [5]. The term PROM study, a large non-blinded randomised controlled trial, reported a drastic reduction in neonatal transmission rates in cases of maternal colonisation with PROM at term when labour was induced with oxytocinon, rather than priming with prostaglandin or managed expectantly (2.5% versus >8%) [23]. 7.2.2. Caesarean section It is generally agreed that there is no need to give specific GBS prophylaxis during elective caesarean section, as the risk of GBS infection across intact membranes is very low. In a single retrospective study, no babies were identified with GBS infection following term caesarean section [24]. 7.3. Disadvantages of widespread intrapartum antibiotic use 7.3.1. Antibiotic resistance of GBS Although there has been no record of penicillin-resistant GBS organisms to date, some strains have developed resistance to macrolide and lincosamide antibiotics (erythromycin and clindamycin) [25]. Morales et al. found a significant increase in the proportion of GBS isolates resistant to erythromycin between those retrieved in 1997– 1998 and those retrieved in 1980–1983, and new resistance to clindamycin, associated with increased IAP use [26]. Clindamycin and erythromycin should be used only when penicillin is contraindicated. 7.3.2. Proliferation of non-GBS pathogens Even a broad-spectrum antibiotic such as ampicillin may result in the proliferation of Clostridium difficile, yeasts and ampicillin-resistant Gram-negative rods [19]. Stable or declining rates of non-GBS sepsis have been reported in the largest multi-centre studies, which have been based on case note review [4]. In a retrospective singlecentre study 30 infants were identified with early-onset E.
S. Beal, S. Dancer / Reviews in Gynaecological and Perinatal Practice 6 (2006) 218–225
coli infection from 61,500 deliveries over a 10-year period [27]. A rise in E. coli infections associated with increased IAP use was confined to infants weighing 1000–1500 g. However, Towers prospectively examined neonatal sepsis developing within the first 7 days of life and its relationship to IAP over a 6-year period [28]. Although this was a singlecentre study, a cohort of 30,000 deliveries was represented. Mothers had received IAP in 15 of 27 cases (56%) of nonGBS neonatal infection. During this time the overall administration of IAP had only increased to a maximum of 17%. Ampicillin-resistant organisms were responsible for 87% of infections in infants born to mothers who had received IAP. They caused only 17% of infections where IAP had not been given. 7.3.3. Anaphylaxis The estimated incidence of anaphylaxis due to penicillin is 1 in 10,000 people treated. This can be fatal in as many as a tenth of these. The RCOG estimates that universal screening where 30% of women were given intrapartum penicillin would result in an average of two UK deaths a year. Screening based on a single risk factor, where IAP was administered to 15% would result in one [5]. 7.4. Choice of antibiotics Ampicillin can be used for intrapartum prophylaxis, but many authorities prefer benzylpenicillin (penicillin G), because of its favourable pharmacokinetics and narrow spectrum of activity. Theoretically, these features should reduce the proliferation of non-GBS and antibiotic resistant microorganisms. However, ampicillin has been used in most clinical studies of intrapartum prophylaxis. Edwards et al. carried out a randomised controlled trial comparing ampicillin with benzylpenicillin intrapartum prophylaxis [29]. Although postpartum maternal culture rates of antibiotic resistant E. coli and Enterobacteriaceae spp. were higher than entry culture rates in both groups, there was no significant difference between the groups. Neonatal cultures were not performed. The sample size had been calculated to detect a 50% increase in the proportion of
221
antibiotic resistant organisms within the ampicillin group compared to the penicillin group. A smaller difference would not have reached statistical significance. In a recent small case-control study, infants with lateonset serious bacterial infections (defined as the presence of pathogenic bacteria cultured from the blood, cerebro-spinal fluid or urine) were much more likely to have received broad-spectrum intrapartum antibiotics than healthy infants. There was no significant difference in the numbers in each group that had received benzylpenicillin [30]. 7.5. Antenatal prophylaxis guidelines (Table 1) 7.5.1. Current RCOG guidelines 7.5.1.1. Background. Until recent RCOG guidelines were produced, there had been no consensus within the UK on GBS prophylaxis. CDC recommendations for universal screening were not necessarily appropriate to the UK. The UK incidence of EOI is similar to that seen in the US with widespread use of intrapartum antibiotics (0.5/ 1000 live births), despite similar vaginal colonisation rates [5]. The CDC guidelines were mainly based on data from a single large retrospective case note review study [4]. No randomised controlled trials directly comparing universal screening and a risk-based approach for selecting women for IAP have yet been carried out. Little UK data exist to support the use of either universal screening or risk factors to select women for administration of IAP [5]. Major organisational and funding changes would be required in the UK if universal screening were introduced to ensure quality and equality of service. Most UK laboratories, for instance, do not use the selective media required for optimal GBS culture because of cost. RCOG guidelines were systematically developed using standardised RCOG methodology [5]. Critical appraisal of the literature focused on the evidence provided by the CDC to support their universal screening strategy. The CTFPHC guidelines were also examined. These and the CDC guidelines are discussed below.
Table 1 Strategies for prevention of neonatal GBS infection 2003 RCOG guidelines [5]
2002 CDC guidelines [4]
2001 CTFPHC guidelines [6]
Criteria for IAP
<37 weeks, membranes ruptured >18 h, pyrexia >38 8C, previous affected infant GBS bacteriuria
Swabs at 35–37 weeks gestation, delivery prior to 37 weeks gestation
Pre-labour swabs performed on women with risk factors at 35–37 weeks (risk factors as for RCOG)
Culture technique
Does not rely on swabbing
Selective broth media
Selective broth media
Swabs used
None
Combined ano-vaginal swabs (can be performed by patient)
Combined recto-vaginal swabs
IAP
Intravenous penicillin (or clindamycin if allergic)
Intravenous penicillin (or clindamycin if allergic)
Intravenous penicillin (or clindamycin if allergic)
Neonatal antibiotics
Prompt antibiotic treatment if unwell; observation for 12 h if well
Prompt antibiotic treatment if unwell; observation for 24 h if well
No guidelines
222
S. Beal, S. Dancer / Reviews in Gynaecological and Perinatal Practice 6 (2006) 218–225
The RCOG guideline group concluded that it was not clear the benefits of universal screening outweighed the risks. Risk-based guidelines were largely formed by expert consensus following mathematical modelling to estimate the effects of proposed strategies within the UK [5]. 7.5.1.2. Recommendations. The RCOG guidelines recommend that IAP should be considered for mothers who have: had a baby previously with GBS infection; PROM; or been found incidentally to have GBS colonisation of the vagina. Antibiotics should also be offered to women with GBS bacteriuria. The argument for IAP strengthens in the presence of two or more risk factors. Antibiotics are not recommended to treat GBS colonisation prior to labour or for women with undergoing a planned caesarean section if the membranes are intact. Usually benzylpenicillin should be given, but intravenous clindamycin should be used in penicillin-allergic women [5]. 7.5.2. CDC guidelines 7.5.2.1. Background. The 1996 CDC guidelines recommended the use of one of two strategies: In the first, all pregnant women were to be screened using an ano-vaginal swab at 35–37 weeks gestation [31]. Swabs were to be cultured using selective broth media. Antibiotics were to be administered to all women identified as GBS colonised or to those where colonisation status was unknown at delivery. Alternatively, women were to receive IAP if they had one or more risk factors; delivery at less than 37 weeks gestation, intrapartum temperature greater than 38 8C or rupture of membranes for more than 18 h. Regardless of the approach decided upon, prophylaxis was also to be administered to women with bacteriuria or who had previously delivered an infant with EOI. The introduction of these guidelines was associated with a dramatic decline in the incidence of earlyonset but not late-onset GBS infection [4]. New 2002 CDC guidelines were produced following critical appraisal of new evidence. The CDC had sponsored a multi-state case note review of a randomised stratified sample of 629,912 live births (5144 live births), managed according to the 1996 guidelines. They found the universal screening approach to be 50% more effective than the risk factor based approach in prevention of neonatal GBS infection. Eighteen percent of women in the universal screening group who were identified as colonised with GBS did not present with risk factors and would not have been selected for intrapartum antibiotics unless they had been screened [32]. This study provided the main basis for the 2002 guideline changes and was the first large scale comparison of the 1996 CDC risk-based and universal screening strategies [32]. This was retrospective observational data and not part of a prospective controlled trial. It was assumed that all women without a documented culture had been managed according to the risk-based approach. This is likely to have produced bias, as others not managed according to the guidelines
would have been included in this group. To reduce this bias, all women who did not undergo antenatal screening and had risk factors, but did not receive IAP were excluded from the study. Confirmation of GBS infection may be more difficult in neonates where intrapartum antibiotics have been transferred across the placenta. Whilst universal screening results in a decreased incidence of EOI, there has been no agreement on a corresponding reduction in neonatal infections as a whole [2]. The implication is that there may be a significant proportion of neonates infected with GBS from whom cultures taken are negative as a result of IAP. Culture negative clinical infection should probably be used as an outcome in future research. 7.5.2.2. 2002 CDC guidelines. Current CDC recommendations for prevention of vertical transmission involve universal screening at 35–37 weeks gestation using a combined vaginal and ano-rectal swab that can be performed by women themselves. All colonised women should receive IAP. Prophylaxis is also advised for women with bacteriuria or who have previously delivered an infant that subsequently developed EOI. Where culture results are not available women should be managed according to risk factors [4]. A specific algorithm is provided for managing women with threatened pre-term labour. Unless a GBS urinary tract infection is diagnosed, antenatal antibiotics should not be prescribed to treat GBS or given to women having planned caesarean sections. Specimens should be placed in a non-nutritive transport medium (e.g. Amies or Stuart’s without charcoal). They should be inoculated into a selective broth media such as LIM broth, incubated overnight and then plated onto solid blood agar medium. In the penicillin-allergic woman susceptibility testing for clindamycin and erythromycin should be performed. Intravenous benzylpenicillin is recommended as IAP for those without a penicillin allergy as in the RCOG guidelines. Ampicillin is a less favourable option. In women with confirmed penicillin allergy, clindamycin or erythromycin is advised. 7.5.3. CTFPHC guidelines The CTFPHC recommends universal screening and administration of IAP to colonised women with risk factors (pre-term labour less than 37 weeks gestation, prolonged rupture of membranes for more than 18 h, maternal intrapartum temperature of greater than 38 8C, antenatal GBS bacteriuria) and those who have previously delivered a newborn with GBS infection regardless of their GBS colonisation status [6]. The CTFPHC recommend the same sample collection and culture techniques for universal screening as the CDC. They also suggest the same antibiotic regimes for IAP [4,6]. The guidelines were produced by expert consensus following systematic review of the literature. They were
S. Beal, S. Dancer / Reviews in Gynaecological and Perinatal Practice 6 (2006) 218–225
developed prior to the production of the 2002 CDC guidelines and the publication of observational US multi-state data. Like the RCOG and CDC, the CTFPHC failed to find any randomised controlled trials demonstrating a significant reduction in GBS infection associated with universal screening or risk factor approaches. However, examination of cumulative evidence from cohort studies showed reductions in early-onset infection when selection for IAP was based on universal screening, alone or combined with a single risk factor [6]. The strongest evidence to support the use of a combined universal screening and risk-based approach was provided by a single randomised controlled trial from the US. Colonised women in labour before 37 weeks gestation or with ruptured membranes for longer than 12 h, were randomised to receive either IAP or no antibiotics. Neonates whose mothers had received IAP showed a 51% reduction in GBS colonisation rates. Bacteraemia occurred in 6% of neonates within the control group, but in none of the treatment group [3]. Excluding women with intrapartum pyrexia or antenatal GBS bacteriuria is likely to have reduced the effect of IAP on neonatal colonisation and bacteraemia. These are both significant risk factors for neonatal infection. Weaknesses of the study include the relatively small numbers (80 women) in each group and the lack of blinding. As previously discussed, confirmation of GBS infection may be more difficult in neonates where antibiotics have been transferred across the placenta and the treatment effect may have been exaggerated. 7.6. Relative cost of universal screening and risk-based strategies The ideal IAP strategy would result in the minimum number of women necessary receiving IAP for the greatest possible reduction in EOI. Adverse antibiotic effects and costs would both be minimised. Case note review has suggested that perfect implementation of both risk-based and universal screening strategies in the US will result in a similar proportion receiving antibiotics [4,32]. However, the RCOG estimate that around 30% of women would receive IAP if universal GBS screening were introduced into the UK. Approximately 15% of women within the UK would receive antibiotics using the previous CDC risk-based strategy. These estimates assume a UK prevalence of 25% maternal GBS colonisation, based on a single 1986 UK study [5]. This may not reflect the current situation. A risk-based approach costs less initially as it avoids sample collection and laboratory costs. Screening costs need to be balanced against the cost of the intensive care for neonates with GBS infection. The cost of adverse effects associated with IAP should also be considered. US analyses of cost initially appear to calculate the overall cost of universal screening in the US to be similar to that of single risk factor based prophylaxis [4]. These studies included the additional cost of administering 48 h of
223
antibiotic prophylaxis to all neonates of mothers given IAP and prolonging hospital stay. The CDC, RCOG or CTFPHC do not currently recommend routine neonatal antibiotic prophylaxis [4–6]. The risk-based approach was found to be significantly less expensive than universal screening after excluding these costs. However, intensive care costs for neonates with GBS infection were not compared between the two strategies [33,34]. A randomised control trial comparing the cost-effectiveness of risk factor and universal screening approaches has been recommended by the RCOG. This would involve introducing optimum sample collection and laboratory techniques to participating hospitals. The RCOG estimate that to prevent one neonatal death by universal screening at least 24,000 UK women would have to be swabbed and 7000 would have to be given IAP. Using the previous CDC riskbased strategy, 625 UK women would have to receive IAP to prevent one case of EOI and 5882 women to prevent one neonatal death. It may be impossible to recruit enough UK women to give the suggested trial sufficient power [5].
8. Proposed alternative prevention strategies 8.1. Antiseptics In vitro studies have shown some microbicides to have strong activity against GBS. Chlorhexidine is cheap and may have a role in reducing GBS transmission in developing countries where antibiotics are not readily available. It has been found to have little or no impact on antibiotic resistance [2]. If proved effective, its widespread use would reduce the risk of antibiotic resistance (particularly ampicillin-resistant E. coli) by reducing antibiotic use. In several clinical trials, both neonatal GBS transmission and infection have been reduced by intrapartum vaginal application of chlorhexidine, or by neonatal washing at birth with chlorhexidine. Other studies have not supported these findings [2,35]. This may be due to heterogeneity of methodology between studies [35]. The large randomised controlled trial with the highest methodological quality and complete follow-up reported a significant reduction in the rates of respiratory distress syndrome, diagnosed or probable infection and admission to the neonatal unit [35,36]. However, only one case of confirmed EOI in each of the study and control groups was reported. Neonatal colonisation rates were not reported. Women with a high risk of vertical transmission (pre-term delivery or previous child with GBS infection) were excluded possibly diminishing the effect seen in the chlorhexidine group. A Cochrane systematic review selecting randomised and quasi-randomised trials comparing vaginal disinfection with chlorhexidine to placebo, or no treatment, found evidence of reduction in neonatal colonisation, but no definite effect on EOI [35].
224
S. Beal, S. Dancer / Reviews in Gynaecological and Perinatal Practice 6 (2006) 218–225
More definite effects have been seen in large prospective, but non-randomised trials. A large Malawi hospital trial of vaginal wiping and neonatal bathing with 0.25% chlorhexidine showed reduced neonatal morbidity and mortality, and maternal infection. There was no blinding, and assignment to treatment or control group was based on month of delivery [37]. Findings were similar in another large prospective study where labouring women received vaginal douching with either saline or a similar strength (0.2%) chlorhexidine solution [38]. 8.2. Vaccines The risk of neonatal infection is indirectly proportional to the levels of specific capsular polysaccharide antibodies in the maternal serum [39]. Protective IgG GBS antibodies appear to be transferred across the placenta. Unconjugated vaccines based on capsular polysaccharides have been shown to be only variably immunogenic. The covalent coupling of a protein antigen to a capsular polysaccharide stimulates T cells and significantly increases immunogenicity [40]. Vaccine composition is likely to differ depending on serotype distribution in each country. GBS types 1a, 1b, 2, 3 and 5 cause almost all maternal and neonatal GBS infection in the USA and UK. Phase 1 and 2 trials in healthy nonpregnant, and phase 1 trials in healthy pregnant women have already shown conjugate polysaccharide vaccines against these serotypes to be safe and highly immunogenic. Funding has not yet been secured for phase 3 trials, however, due to the litigation risk of such a trial in pregnant women [41]. Transplacental transport is less before 32–24 weeks gestation, which may limit the benefit of antenatal vaccines. Ideally such a vaccine could be given to teenage girls prior to pregnancy [2]. Systemic reactions to the vaccines have occurred in less than 2% of almost 500 healthy young adults [41]. These included low-grade fever, chills, headache or myalgia and resolved within 36 h. Immunization would hopefully prevent both early and lateonset neonatal infection by reducing the risk of vertical transmission. Theoretically, it may reduce or even eliminate GBS by mucosal immunological responses and may therefore reduce adverse pregnancy outcomes such as pre-term delivery, stillbirth and maternal infection. Vaccine use would also reduce the risk of further antibiotic resistance [2]. Recently the genomes of different GBS strains have been described, with implications for analysis and vaccine development [42].
9. Conclusions GBS is a major cause of infectious morbidity and mortality in the neonate. Vertical transmission can be successfully prevented in the majority of pregnancies by intrapartum administration of antibiotics. This benefit needs
to be balanced against the disadvantages of widespread intrapartum antibiotics, which include the potential for increased antibiotic resistance, proliferation of non-GBS organisms and anaphylaxis. At present, the RCOG recommend a risk-based strategy to select women for IAP. This contrasts with the US system of universal antenatal screening at 35–37 weeks gestation. Neither strategy has been proven to be superior. Major organisational and funding changes would be required in the UK if universal screening were introduced to ensure quality and equality of service. Further research is needed to ascertain the most appropriate and cost effective strategy for the UK.
References [1] Baker CJ, Edwards MS. Group B streptococcal infections. In: Remington J, Klein JO, editors. Infectious diseases of the newborn and infant. 5th ed. WB Saunders: Philadelphia, PA; 2000 . p. 1091– 156. [2] Gibbs RS, Schrag S, Schuchat A. Perinatal infections due to Group B streptococcus. Obstet Gynecol 2004;104(5 Pt 1):1062–76. [3] Boyer KM, Gotoff SP. Prevention of early-onset neonatal group B streptococcal disease with selective intrapartum chemoprophylaxis. New Engl J Med 1986;314:1665–9. [4] Centers for Disease Control Prevention. Prevention of perinatal group B streptococcal disease: revised guidelines from CDC. Morb Mortal Wkly Rep 2002;51:1–22. [5] Royal College of Obstetricians and Gynaecologists. Prevention of early onset neonatal group B streptococcal disease. Guideline No 36. London: RCOG; 2003. [6] Canadian Task Force on Preventative Health Care. Prevention of earlyonset group B streptococcal disease in the newborn: systematic review and recommendations. CTFPHC Technical Report 01-6. London ON: CTF; 2001. [7] Anthony BF, Eisenstadt R, Carter J, Kim KS, Hobel CJ. Genital and intestinal carriage of group B streptococci during pregnancy. J Infect Dis 1981;143:761–6. [8] Pass MA, Gray BM, Khare S, Dillon Jr HC. Prospective studies of group B streptococcal infections in infants. J Pediatr 1979;95:437–43. [9] Schuchat A, Deaver-Robinson K, Plikaytis BD, Zangwill KM, MohleBoetani J, Wenger JD. Multistate case-control study of maternal risk factors for neonatal group B streptococcal disease. Pediatr Infect Dis J 1994;13:623–9. [10] Heath PT, Balfour G, Weisner AM, Efstratiou A, Lamagni TL, Tighe H, et al. Group B streptococcal disease in UK and Irish infants younger than 90 days. Lancet 2004;363:292–4. [11] Weisner AM, Johnson AP, Lamagni TL, Arnold E, Warner M, Heath PT, et al. Characterisation of group B streptococci recovered from infants with invasive disease in England and Wales. Clin Infect Dis 2004;38:1203–8. [12] Noya FJD, Rench MA, Metzger TG, Colman G, Naidoo J, Baker CJ. Unusual occurrence of an epidemic of type Ib/c group B streptococcal sepsis in a neonatal intensive care unit. J Infect Dis 1987;155(6):1135– 44. [13] Boyer KM, Gadzala CA, Kelly PD, Burd LI, Gotoff SP. Selective intrapartum chemoprophylaxis of neonatal group B streptococcal early-onset disease. II. Predictive value of prenatal cultures. J Infect Dis 1983;148:802–9. [14] Ferreri P, Blair LL. Pharyngeal carriage of group B streptococci: detection by three methods. J Clin Microbiol 1977;6:136–9. [15] Yancey MK, Armer T, Clark P, Duff P. Assessment of rapid identification tests for genital carriage of group B streptococci. Obstet Gynecol 1992;80:1038–47.
S. Beal, S. Dancer / Reviews in Gynaecological and Perinatal Practice 6 (2006) 218–225 [16] Bergeron MG, Danbing K, Me´nard C, Francois FJ, Gagnon M, Bernier M, et al. Rapid detection of group B streptococci in pregnant women at delivery. New Engl J Med 2000;343:175–9. [17] Slotved HC, Elliott J, Thompson T, Konradsen HB. Latex assay for serotyping group B Streptococcus isolates. J Clin Microbiol 2003;41: 4445–7. [18] Hall RT, Barnes W, Krishnan L, Harris DJ, Rhodes PG, Fayez J, et al. Antibiotic treatment of parturient women colonized with group B streptococci. Am J Obstet Gynecol 1976;124:630–4. [19] Dancer SJ. How antibiotics can make us sick: the less obvious adverse effects of antimicrobial chemotherapy. Lancet Infect Dis 2004;4(10): 611–9. [20] Kenyon SL, Taylor DJ, Tarnow-Mordi W, ORACLE Collaborative Group. Broad-spectrum antibiotics for pre-term, pre-labour rupture of fetal membranes: the ORACLE I randomised trial. Lancet 2001; 357(9261):979–88. [21] Gilbert RE, Pike K, Kenyon SL, Tarnow-Mordi W, Taylor DJ. The effect of pre-partum antibiotics on the type of neonatal bacteraemia: insights from the MRC ORACLE trials. BJOG 2005;112(6): 830–2. [22] Mercer BM, Miodovnik M, Thurnau GR, Goldenberg RL, Das AF, Ramsey RD, et al. Antibiotic therapy for reduction of infant morbidity after pre-term premature rupture of the membranes. A randomized controlled trial. JAMA 1997;278(12):989–95. [23] Hannah ME, Ohlsson A, Wang EE, Matlow A, Foster GA, Willan AR, et al. Maternal colonisation with group B streptococcus and pre-labour rupture of membranes at term: the role of induction of labour. Term PROM Study Group. Am J Obstet Gynecol 1997;177(4):780–5. [24] Ramus RM, McIntyre DD, Wendel Jr GD. Antibiotic prophylaxis for group B Streptococcus is not necessary with elective caesarean section at term (abstract). Am J Obstet Gynecol 1999;180:S85. [25] Berkowitz K, Regan JA, Greenberg E. Antibiotic resistance patterns of group B streptococci in pregnant women. J Clin Microbiol 1990;28(1): 5–7. [26] Morales WJ, Dickey SS, Bornick P, Lim DV. Change in antibiotic resistance of group B streptococcus: impact on intrapartum management. Am J Obstet Gynecol 1999;181(2):310–4. [27] Joseph TA, Pyati SP, Jacobs N. Neonatal early-onset Escherichia coli disease. The effect of intrapartum ampicillin. Arch Pediatr Adolesc Med 1998;152(1):35–40. [28] Towers CV, Carr MH, Padilla G, Asrat T. Potential consequences of widespread antepartal use of ampicillin. Am J Obstet Gynecol 1998; 179(4):879–83. [29] Edwards RK, Clark P, Sistrom CL, Duff P. Intrapartum antibiotic prophylaxis 1: relative effects of recommended antibiotics on gramnegative pathogens. Obstet Gynecol 2002;100(3):534–9.
225
[30] Glasgow TS, Young PC, Wallin J, Kwok C, Stoddard G, Firth S, et al. Association of intrapartum antibiotic exposure and late-onset serious bacterial infections in infants. Pediatrics 2005;116(3):696–702. [31] Centers for Disease Control Prevention. Prevention of perinatal group B streptococcal disease: a public health perspective. Morb Mortal Wkly Rep 1996;45(RR7):1–24. [32] Schrag SJ, Zell ER, Lynfield R, Roome A, Arnold KE, Craig AS, et al. A population-based comparison of strategies to prevent early-onset group B streptococcal disease in neonates. New Engl J Med 2002; 347(4):233–9. [33] Mohle-Boetani JC, Schuchat A, Plikaytis BD, Smith JD, Broome CV. Comparison of prevention strategies for neonatal group B streptococcal infection. A population-based economic analysis. JAMA 1993;270(12):1442–8. [34] Fargason Jr CA, Peralta-Carcelen M, Rouse DJ, Cutter GR, Goldenberg RL. The pediatric costs of strategies for minimizing the risk of early-onset group B streptococcal disease. Obstet Gynecol 1997; 90(3):347–52. [35] Stade B, Shah V, Ohlsson A. Vaginal chlorhexidine during labour to prevent early-onset neonatal group B streptococcal infection [Systematic Review]. Cochrane Pregnancy and Childbirth Group. Cochrane Database Syst Rev 2005;4. [36] Burman LG, Christensen P, Christensen K, Fryklund B, Helgesson AM, Svenningsen NW, et al. Prevention of excess neonatal morbidity associated with group B streptococci by vaginal chlorhexidine disinfection during labour. The Swedish Chlorhexidine Study Group. Lancet 1992;340(8811):65–9. [37] Taha TE, Biggar RJ, Broadhead RL, Mtimavalye LA, Miotti PG, Justesen AB, et al. Effect of cleansing the birth canal with antiseptic solution on maternal and newborn morbidity in Malawi: clinical trial. Br Med J 1997;315:216–20. [38] Stray-Pedersen B, Bergan T, Hafstad A, Normann E, Grogaard J, Vangdal M. Vaginal disinfection with chlorhexidine during childbirth. Int J Antimicrob Agents 1999;12:245–51. [39] Baker CJ, Kasper DL. Correlation of maternal antibody deficiency with susceptibility to neonatal group B streptococcal infection. New Engl J Med 1976;294:753–6. [40] Kasper DL, Paoletti LC, Wessels MR, Guttormsen HK, Carey VJ, Jennings HL, et al. Immune response to type III group B streptococcal polysaccharide-tetanus toxoid conjugate vaccine. J Clin Invest 1996;98: 2308–14. [41] Baker CJ, Edwards MS. Group B streptococcal conjugate vaccines. Arch Dis Child 2003;88(5):375–8. [42] Maione D, Margarit I, Rinaudo CD, Masignani V, Mora M, Scarselli M, et al. Identification of a universal group B streptococcus vaccine by multiple genome screen. Science 2005;309(5731):148–50.