confidential: for review only - bmj · 2019-02-27 · confidential: for review only 2 steady in...

Confidential: For Review Only

Universal antenatal screening for carriage of group B

Streptococcus (GBS): State of the evidence

Journal: BMJ

Manuscript ID BMJ.2018.045010

Article Type: Analysis

BMJ Journal: BMJ

Date Submitted by the Author: 10-May-2018

Complete List of Authors: Seedat, Farah; University of Warwick, Warwick Medical School Geppert, Julia; University of Warwick, Warwick Medical School Stinton, Chris; University of Warwick, Warwick Medical School Patterson, Jacoby; University of Warwick, Warwick Medical School Freeman, Karoline; University of Warwick, Warwick Medical School

Johnson, Samantha; University of Warwick, Warwick Medical School Fraser, Hannah; University of Warwick, Warwick Medical School Brown, Colin; Public Health England Colindale, Bacteria Reference Department, National Infection Service Uthman, Olalekan; University of Warwick, Warwick Medical School Tan, Bee; University of Warwick, Warwick Medical School Robinson, Esther; Public Health England, Field Service, National Infection Service McCarthy, Noel; University of Warwick, Warwick Medical School Clarke, Aileen; University of Warwick, Warwick Medical School Marshall, John; UK National Screening Committee Visintin, Cristina; UK National Screening Committee

Mackie, Anne; UK National Screening Committee Taylor-Phillips, Sian; University of Warwick, Warwick Medical School

Keywords: Screening, Pregnancy, Policymaking, Evidence review, Infectious diseases, Fetal, maternal, and child health, Group B Streptococcus, Streptococcus agalactiae

https://mc.manuscriptcentral.com/bmj

BMJ

Confidential: For Review Only1

Universal antenatal screening for carriage of group B Streptococcus (GBS): State of the evidence

Responses to editor and reviewer comments from previous version of manuscript

Comment Response

Editorial team

1 Lacks international perspective. What happens in other countries,

particularly those that have elected for screening.

Each section of the analysis piece was informed by referenced papers from

international literature with any focus on the UK contextualised by these

papers. In response to this comment, we have added the following (bold

text) in the introduction about different countries adopting screening and

risk-based prevention (pages 2 and 3) to increase international focus:

“In the Netherlands, the incidence of EOGBS has increased under

risk-based prevention from 0.11 to 0.19 per 1,000 live births

(p<0.001).13 By contrast, under risk-based prevention the incidence of

EOGBS in New Zealand has decreased from 0.5 to 0.26 per 1,000 live

births, from 0.4 to 0.3 per 1,000 live births in Sweden, and from 0.6 to

0.2 per 1,000 live births in Denmark (from 1995 to 2002).14-16

”

“Screening was first recommended in the US in 1996, when either risk-

based prevention or universal screening were recommended.17 This was

followed by a revision in 2002 recommending universal screening.18 19 The

incidence of EOGBS decreased in the US from 1.7 per 1,000 live births

to 0.6 in the 1990s, to 0.3 in 2004 after the introduction of universal

screening,20 21

and was most recently estimated at 0.22 in 2016.22 The

largest decline occurred when either risk-based or screening

prevention were recommended,21 making it unclear how much of the

decrease was a result of screening itself. After the US guidelines were

published, universal GBS screening guidelines were developed in many

countries,1 23

including Australia,24 Canada,

25-27 Belgium,

28 29 France,

30

Germany,31 32 Italy,

33 34 Spain,

35 and Switzerland.

36 After the introduction

of universal screening, the incidence of EOGBS was reported as

Page 1 of 49


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


steady in Emilia-Romagna in Italy at around 0.28 per 1,000 live births

from 2003 to 2008.37 The rate decreased from 1.1 to 0.7 per 1,000 live

births in Spain,38 but more than doubled in Alberta in Canada from

0.15 per 1,000 live births in 2003 to 0.34 in 2013.39”

2 Insufficient context about what happens in practice (e.g. what options are

available to people, women requesting private tests, the values and

priorities that inform decision making).

These issues were not the focus of our work although they are clearly of

vital importance in the interpretation of our work. We have referred to

them as follows on page 12 and stated that further work should be

undertaken in this area:

“There is also a lack of evidence about what currently happens in

practice. Little is known about the impact of private GBS carriage

testing or the values and priorities informing women’s decision-

making. Further work should be undertaken to understand this.”

3 We felt it did not spell out the relative weight of benefits and harms

clearly enough or in sufficient detail.

A major point of our analysis was to highlight that, internationally,

screening has been implemented without sufficient consideration of the

harms that may be a consequence of the overdiagnosis from screening for

GBS. The relative weight of benefits and harms of screening is

tremendously important but there is very little discussion of these issues in

the literature. Our analysis here is based on a systematic review of the

available literature, which shows that the harms are not yet fully

understood. However, to clarify the relative weight of the benefits and

harms, we have restructured the section on GBS screening into sub-

sections on the benefits of screening and the harms of screening (see pages

7 to 11) where we have expanded on the IAP harms, providing more

details on the findings of the systematic review.

Please also see the answer to the first reviewer 1 comment in the next line

of the table, which says: ‘in the section on the benefits and harms of

screening on page 7, we have stated that “in the absence of RCTs, it is

difficult to quantify the potential impact of adding universal screening to

risk-based practice.” while on page 11 we conclude that “The current

evidence on the benefits and harms of universal GBS screening is based

Page 2 of 49


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


mainly on observational studies that are small, subject to bias, and/or

have applicability concerns. Therefore, the balance of benefits and harms

from GBS screening and IAP cannot be quantified, especially as the long-

term impact of EOGBS and of IAP is poorly understood.”

Reviewer 1

General comments

1 Prenatal screening for GBS is an important topic affecting maternity and

neonatal practice and a substantial minority of pregnant women. It is

relevant to the BMJ readership as an analysis piece. However, to appeal to

and be understood by a general audience, the arguments need to

simplified, with less detailed quoting of figures and more clarity about the

strength of evidence.

We thank the reviewer for his/her positive comments. We agree that the

argument needs to be simplified and to do this we have restructured the

paper as mentioned above, whereby the section on universal screening is

split into the evidence on the benefits followed by the evidence of the

harms (pages 7 to 11).

We also agree that the strength of evidence is an important piece of this

analysis and we have discussed the strength of evidence in every section of

paper and the conclusion. For example, in the section on IAP on page 7,

we have mentioned that “Due to the high risk of bias identified in the

three included small randomised controlled trials (RCTs) which were

conducted more than 20 years ago, the authors concluded that there is

no valid information to inform clinical practice.76 Although IAP may

be effective in reducing EOGBS infections, the poor evidence base

makes it uncertain how effective it is.” Similarly, in the section on the

benefits and harms of screening on page 7, we have stated that “in the

absence of RCTs, it is difficult to quantify the potential impact of

adding universal screening to risk-based practice” while on page 11 we

conclude that “The current evidence on the benefits and harms of

universal GBS screening is based mainly on observational studies that

are small, subject to bias, and/or have applicability concerns.

Therefore, the balance of benefits and harms from GBS screening and

IAP cannot be quantified, especially as the long-term impact of

EOGBS and of IAP is poorly understood.” Finally, in the conclusion on

page 12 we state that “The harm from widespread IAP to thousands of

Page 3 of 49


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


pregnant women and their babies is unknown while the evidence on

the benefit from a screening programme is uncertain due to lower

quality studies with serious limitations. The long-term morbidity of

EOGBS is also largely unknown as is the impact of screening upon it.”

We have removed quoting of figures where possible. We feel that the

remaining figures (along with their limitations/strength of evidence) are

important to gauge the flow of people in the population in a screening

programme. These figures show the numbers/proportions of people who

suffer morbidity and mortality from EOGBS, the number of pregnant

women that need to be treated, and the number of babies who would be

saved by screening. These figures could help readers to understand the

magnitude of the potential overdiagnosis more readily and so have been

retained.

2 The conclusion about introduction of prenatal screening for GBS

colonisation, pending a better test for mothers who transmit to their baby,

is appropriately cautious but the reasoning, that we do not fully understand

the natural history of GBS to be able to screen and treat, is weak. The

volume of high quality research into GBS since the early 1980s, including

trials and before vs after studies, far exceeds the research base for many

infections in early childhood.

Our point here is that because we do not understand the natural history of

why some mothers transmit GBS and have a baby with EOGBS, while

others do not, we cannot better identify and treat only those mothers who

would be at highest risk of GBS. This lack of natural history information

means that with the current test there would be a substantial potential for

overdiagnosis and overtreatment; the harms of which are unknown. If we

had this information, overdiagnosis and overtreatment could be reduced.

We have clarified our statement on page 12 as follows (bold text):

“Selective maternal culture is not an accurate predictor of EOGBS disease

in neonates, and a lack of understanding about why some colonised

mothers have a neonate with EOGBS limits the ability to identify an

approach that reduces the rate of overdiagnosis. At present, if a GBS

screening programme is implemented, it would offer all term

pregnant women the antenatal GBS culture test, but over 99% of

screen-positive mothers (and their babies) would be overdiagnosed

and unnecessarily receive IAP, as they would not have a neonate with

Page 4 of 49


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


EOGBS even without intervention. The harm from widespread IAP to

thousands of pregnant women and their babies is unknown while the

evidence on the benefit from a screening programme is uncertain due

to lower quality studies with serious limitations.”

Specific comments

3 The introduction helpfully cites the criteria for assessing a screening

programme. The 4 concepts (1 prevalence and seriousness; 2 accurate test;

3 effective treatment; and 4 effective programme) are useful and could be

followed through in the assessment of evidence and conclusions. I suggest

deleting the more detailed sections in para 1 p 3, which has different

numbering and is unnecessary.

Thank you for this helpful comment. We have gone through and reworded

headings to make the links to the criteria more explicit and have

restructured the section on GBS screening into sub-sections on the benefits

of screening and the harms of screening to ensure that the readers can see

it is followed through to the conclusions.

We have also deleted the more detailed sections from paragraph 1 on page

3 and re-written the paragraph as follows:

“Here, we summarise the best available evidence on universal GBS

screening from two comprehensive evidence reviews undertaken for

the UK National Screening Committee (NSC) and have also included

key papers published since the conclusions of those reviews.43 44

The

reviews address UK NSC screening criteria, examining 1) the natural

history (asymptomatic stage to illness) and epidemiology (severity,

incidence) of EOGBS; 2) the accuracy of the GBS culture screening

test (is it able to separate women at high risk of having a neonate with

EOGBS from those at low risk?); 3) the effectiveness of IAP

treatment; and 4) whether a universal GBS screening programme

delivers more benefit than harm (at a reasonable cost).”

4 There should be more discussion of the context of EOGBS, which

accounts for around half of early onset bacteraemia. Risk-based treatment

addresses infection due to GBS and to some extent, depending on

spectrum of effect, other organisms. GBS screening might reduce needed

treatment in GBS negative women with risk factors at term. This argument

strengthens the authors’ focus on screening or not for low risk women at

term.

We have added more context about EOGBS in the introduction section as

well as more information on the current approach of risk-based prevention

in the UK on page 2 as follows (bold text):

“Group B Streptococcus (Streptococcus agalactiae, GBS) is a Gram-

positive bacterium that colonises the gastrointestinal and genitourinary

tract in approximately 30% of healthy adults.1-3 It usually causes no harm.1

However, GBS is the commonest cause of neonatal sepsis and

Page 5 of 49


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

meningitis in many developed countries.4 In the UK, GBS causes

invasive disease in the first six days of life (early onset GBS or

EOGBS) in around 1 of every 2,000 live births.55 Neonates with

EOGBS can develop sepsis, pneumonia, and meningitis, and around

5% to 10% of these babies can die as a result.5-7

To prevent EOGBS, a risk-based strategy has been recommended in

the UK since 2003.8-11

Pregnant women presenting with GBS risk

factors are offered intrapartum antibiotic prophylaxis (IAP), usually

intravenous penicillin, during labour.8-11

The incidence of EOGBS has

increased over this period in the UK by around 0.09 per 1,000 live

births,5 6 however, as the strategy has been imperfectly implemented,

6

12 it is difficult to identify its impact. In the Netherlands, the incidence

of EOGBS has increased under risk-based prevention from 0.11 to

0.19 per 1,000 live births (p<0.001).13 By contrast, under risk-based

prevention the incidence of EOGBS in New Zealand has decreased

from 0.5 to 0.26 per 1,000 live births, from 0.4 to 0.3 per 1,000 live

births in Sweden, and from 0.6 to 0.2 per 1,000 live births in Denmark

(from 1995 to 2002).14-16

A criticism of the risk-based strategy is that risk factors are absent in

30% to 60% of pregnancies that result in EOGBS who would be

excluded from prophylactic prevention, although of course we have no

way of telling in advance who these cases are.”

In the epidemiology section on page 4, we have also added the long-term

morbidity of EOGBS:

“The long-term outcomes in EOGBS survivors are not well

researched. Neurological impairment at hospital discharge or last

paediatric review has been reported in up to 15.8% of EOGBS

survivors.55-57

GBS meningitis survivors have a higher risk of long-

term neurodevelopmental disability.56 58 59

”

Page 6 of 49


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


After careful consideration we do not think that introducing GBS

screening would reduce needed treatment in GBS negative women with

risk factors as the introduction of GBS screening would be in addition to

risk-based prevention as opposed to replacing it. Indeed, the modelling

project conducted by the UK National Screening Committee found that a

strategy of screening and risk-based prevention would not reduce needed

treatment.

5 The figure is helpful but number dense: proportionate symbols (using data

visualisation techniques) could be helpful and could bring more clarity

about the proportion at term who might benefit from screening.

We thank you for your suggestion. We have changed the figure and added

data visualisation tools to improve it. We have focused only the outcomes

of most interest and re-calculated the proportion of women at term that

might benefit. We have added a few sentences in the natural history and

test accuracy section:

Natural history section, page 5: “Figure 1 summarises the natural

history and epidemiology of GBS for a hypothetical cohort of term

pregnant women in 2000 (no prevention guideline) and 2014 (risk-

based prevention guideline). It is clear that EOGBS is an important

health problem, yet there is a poor understanding about the natural

history of GBS from maternal GBS colonisation to the long-term

outcomes of EOGBS.”

Test accuracy section, page 6: “Based on the UK surveillance data

shown in Figure 1, we estimate that the positive predictive value

(PPV) of antenatal culture predicting EOGBS may be around 2/1,000

(0.2%). The 2000-01 UK surveillance study shows that there were

around 205 term neonates affected by EOGBS per year in the UK at a

time with no national prevention guideline.7 There were 679,029

babies born in the UK in 2000,64 of whom approximately 573,448

reached full term (37 weeks onwards) and were not born by elective

caesarean section.65 If around 22% of women were colonised with

GBS,47 that would equate to approximately 126,159 GBS carriers at 35

to 37 weeks gestation. Therefore, if there was 100% test uptake and no

Page 7 of 49


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


change in GBS colonisation status from screening until labour, the

PPV of a screening test performed at 35 to 37 weeks for detecting

EOGBS in the neonate would be 0.16% (205/126,159). Performing the

same calculations using 2014 surveillance figures gives a PPV value of

0.25% (350/138,933),6 64 66

which would deliver an unacceptably high

level of false positive results.”

6 The authors could draw more on a previous, detailed cost effective

modelling study, albeit 10 years ago, which paid close attention to the

timing of screening and processing of culture results. The change in the

estimated proportion of women treated (increasing from 11% to

27%; http://www.bmj.com/content/bmj/335/7621/655.full.pdf see table 4)

is proportionately similar to the figures quoted on p9 para 1, though the

authors do not report the over proportion treated. The authors could also

reference that the estimated NNT from this study is consistent with the

figure of more than 99% of women being over treated (see table 55 p76,

strategy

99 file:///S:/ICH/PPP_CENB_CEBCHAdmin/RUTH%20ADMIN%20FO

LDER/Ruth's%20reviews/3001483fullhta%20gbs.pdf)

Thank you for this suggestion. We have added in the study on page 9 as

follows (bold text):

“An examination of the potential harms is a standard part of the

assessment of any screening proposal. Based on the estimated PPVs of

culture testing indicated above (0.2%), approximately 138,933 term

pregnant women per year would be eligible for IAP. Approximately 99.8%

(138,655 pregnant women and babies) would be overdiagnosed and

overtreated as they would have never had a neonate with EOGBS even

without IAP. This makes assessing the harms of treatment particularly

important. A cost-effectiveness model published in the UK also

estimated that adding screening to risk-based prevention would result

in 99.8% overtreatment (increasing antibiotic use in the pregnant

population from 11% to 27%).77”

7 The call for RCTs with mortality due to EOGBS as an outcome (p6

section 4) is unrealistic. Such trials would have to involve several

countries. Bacteraemia or sepsis are sufficiently serious outcomes and

should remain the focus of feasible clinical studies.

We agree that the focus of an RCT should be on bacteraemia and sepsis as

they are more feasible. However, for screening programmes, EOGBS

mortality it is an outcome of major importance. We have edited the

sentence on page 7 as follows (bold text) to reflect the feasibility

problems:

“There have been no RCTs assessing the effects of antenatal screening on

the reduction of morbidity, clinical outcomes, and mortality from EOGBS.

RCTs on the effects of screening on the clinical outcomes and

mortality from EOGBS would be challenging due to immense sample

size requirements.”

8 A minor, but important point is that the Cochrane trial estimate of We have added the estimate from the NIHR study on page 7 as follows

Page 8 of 49


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


treatment effectiveness for neonatal colonisation (given a colonised

mother) is contentious and highly susceptible to variation in the statistical

methods for handling zero cell sizes. The aforementioned NIHR cost

effectiveness study used an alternative Bayesian method and produced a

more extreme treatment

effect file:///S:/ICH/PPP_CENB_CEBCHAdmin/RUTH%20ADMIN%20

FOLDER/Ruth's%20reviews/3001483fullhta%20gbs.pdf table 33.

(bold text):

“Although IAP may be effective in reducing EOGBS infections, the poor

evidence base makes it uncertain how effective it is. Echoing the

uncertainty, a meta-analysis in 2007 using different statistical methods

produced a more extreme treatment effect of IAP, estimating a risk

reduction in EOGBS of 97%.77”

9 Recommendation 1 in box p 11, for a RCT of risk based versus culture

screening, is contentious and not sufficiently justified. The argument

made in this review focusses on uncertainty about screening versus no

treatment for term pregnancies without risk factors. However,

recommendation 1, does not restrict the trial question to this group. Why

not? Culture-based screening instead of risk based screening for preterm

and high risk groups could be harmful by delaying treatment for GBS and

other early onset causes of infection.

Please see answers to questions 10 and 11.

10 The authors should refer to the earlier ‘value of information’ study based

on the NIHR HTA cost effectiveness modelling study, which addressed

the same question about a trial

(file:///S:/ICH/PPP_CENB_CEBCHAdmin/RUTH%20ADMIN%20FOL

DER/Ruth's%20reviews/3001483fullhta%20gbs.pdf ). The NIHR study

concluded that investment in research into prenatal vaccination for GBS

was worthwhile, and if effective, vaccination combined with risk based

screening would be expected to be more cost effective than culture based

screening (negating the need for a trial) and would limit the proportion of

women treated to around 11% (see also

BMJ https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1995477/).

We thank the reviewer for this suggestion and have included this in the

debate on page 11:

“In 2007, a cost-effectiveness model concluded that investments in

prenatal vaccination research should be prioritised over screening

research.77 However, interest has remained to evaluate the impact of

universal GBS screening. In 2017, the Health Technology Assessment

launched an expression of interest for a cluster randomised trial

assessing whether screening women for carriage of GBS in late

pregnancy reduces the occurrence of early-onset sepsis, whether the

benefits outweigh the harms, and whether it would be cost-effective.”

11 The authors have omitted to refer to the fact that the NIHR HTA

programme is currently commissioning a RCT on culture-based screening

versus risk based screening (https://www.nihr.ac.uk/funding-and-

support/funding-opportunities/1786-the-clinical-and-cost-effectiveness-of-

At the time that this paper was initially submitted to the BMJ, the call for

the RCT was not yet advertised. As suggested, we have removed the

recommendation from Box 1 and added a paragraph to discuss the NIHR

call and stimulate the discussion on this:

Page 9 of 49


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


screening-for-group-b-streptococcus-gbs-in-pregnancy/6627). This call is

surprising, given that the HTA commissioning board rejected such a

proposal based on the previous cost effectiveness analysis in 2007. What

has changed in the epidemiology of the condition or evidence for test

accuracy or treatment effectiveness to justify the very substantial costs of

such a large trial? Furthermore, the HTA recommends a cluster RCT,

which would not allow the detailed analysis of ‘who’ (i.e. risk groups) is

susceptible to mother to child transmission and disease, as raised by the

authors in this piece. One change since the HTA last decided on this

matter is progress in developing a vaccine. However, this would argue

against a RCT at this time. This analysis piece is not sufficiently up to

date and explicit about the scientific evidence, and other influences, on the

recommendation for a RCT. I suggest they delete this recommendation

and discuss the evidence to support (or not) the NIHR HTA call for a

RCT.

“In 2007, a cost-effectiveness model concluded that investments in

prenatal vaccination research should be prioritised over screening

research.77 However, interest has remained to evaluate the impact of

universal GBS screening. In 2017, the Health Technology Assessment

launched an expression of interest for a cluster randomised trial

assessing whether screening women for carriage of GBS in late

pregnancy reduces the occurrence of early-onset sepsis, whether the

benefits outweigh the harms, and whether it would be cost-effective.”

12 The authors could be clearer about whether there is a statistically

significant increase in the rate of EOGBS, and whether this is mirrored in

other countries, with and without screening. Point prevalence comparisons

are less relevant than comparisons of trends as maternal GBS colonisation

and the incidence of EOGBS is expected to vary between populations.

Thank you for this helpful comment. Yes, this was a statistically

significant increase, which we have added in on page 4 (bold text):

“This was statistically different to the earlier enhanced surveillance

conducted in 2000-01 (before national prevention guidelines), which found

a rate of 0.48 per 1,000 live births.7”

Minor comments

13 P3 line 26 – ‘thought to be passed’ – I think there is enough evidence to

indicate that this is how infection is transmitted to the neonate. Possibly

add ‘colonise vagina or rectum’.

Changed to “is passed” on page 3.

14 References 14 and 15 are not obtainable. Links should be given. Links added.

Reviewer 2

Summary

This article is based on a review commissioned by the UK National screening

committee (NSC), that evaluates the proposal that the UK should introduce a

universal screening programme to detect GBS carriage in late pregnancy. The

We thank the reviewer for his/her positive comments. We agree, although

screening may be effective (how effective we do not know), the

programme would generate substantial overtreatment from IAP, which

Page 10 of 49


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


review concludes that universal GBS screening does not meet the UK NSC

criteria for introducing a screening programme.

This is probably a reasonable argument but it should be acknowledged that it is

a difficult area, as even based on the authors’ conservative estimates,

introduction of a universal GBS screening programme and IAP (to replace ‘risk

based’ IAP) might potentially prevent ~212 additional EOGBS sepsis episodes

and 5-6 neonatal deaths per year. Given the morbidity and mortality potentially

averted by a universal screening approach, I think the authors need to be very

clear that the argument they are making is that the programme is not cost-

effective and/or that universal screening programmes and IAP have potential

unquantified harms (due to lack of RCT evidence) e.g. antibiotic

resistance/changes in neonatal gut flora etc. If the major emphasis is on

unintended harms of a screening programme, then more attention should be

given to describing recent studies quantifying known harms (at present, many of

the possible ‘harms’ are not referenced) e.g. disruption of neonatal flora (Azad

et al BJOG 2016). Most recent studies have not shown an impact of IAP on

antimicrobial resistance.

may have potential unquantified harms.

To make this argument clear, we have stated the following in the stand-

first on page 1:

“In this paper, Seedat and colleagues review the evidence

underpinning this decision, addressing key issues yet to be considered

before introducing a screening programme, including the possibility of

harm from the high levels of unnecessary treatment with antibiotic

prophylaxis due to a poor test and the uncertainty of screening

effectiveness.”

In the conclusion on page 12, we have acknowledged that this is a difficult

area and that screening should not be offered as there is a huge amount of

overtreatment and that IAP has potential unquantified harms, as follows

(bold text):

“Universal antenatal GBS culture screening is a complex area and a

poor evidence base means that, so far, it does not meet UK NSC criteria

needed to introduce screening programmes. Selective maternal culture is

not an accurate predictor of EOGBS disease in neonates, and a lack of

understanding about why some colonised mothers have a neonate with

EOGBS limits the ability to identify an approach that reduces the rate

of overdiagnosis. At present, if a GBS screening programme is

implemented, it would offer all term pregnant women the antenatal

GBS culture test, but over 99% of screen-positive mothers (and their

babies) would be overdiagnosed and unnecessarily receive IAP, as

they would not have a neonate with EOGBS even without

intervention. The harm from widespread IAP to thousands of

pregnant women and their babies is unknown while the evidence on

the benefit from a screening programme is uncertain due to lower

quality studies with serious limitations.”

We have also expanded on the review of the harms from IAP describing

the studies on pages 9 to 11, including the consistent evidence on gut

Page 11 of 49


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


microbiota and the inconsistent evidence on antibiotic resistance.

Given that this journal is intended for a general audience, the article makes

several assumptions about prior knowledge of neonatal EOGBS disease

prevention approaches and does not really explain why there have been repeated

calls ‘in the media’ to change from a risk-based to a universal screening

approach. I think it would be fair to acknowledge in the introduction that many

countries (e.g. US after 2002 CDC guidelines, and many European countries)

have adopted a universal GBS screening programme and have seen significant

reductions in EOGBS rates. Most studies have also shown lower rates of

EOGBS disease in neonates with universal screening rather than a risk based

approach (see ref 46), acknowledging that no RCTs have been done.

As mentioned above, we have added the following information in the

introduction on pages 2 to 3:

“A criticism of the risk-based strategy is that risk factors are absent in

30% to 60% of pregnancies that result in EOGBS who would be

excluded from prophylactic prevention, although of course we have no

way of telling in advance who these cases are. There are regular media

and political calls to introduce a universal antenatal GBS screening

programme in addition to the current risk-based strategy. Universal

screening would involve the collection of specimens using rectovaginal

swabs at 35 to 37 weeks gestation, which are processed using selective

culture media, to identify women colonised with GBS so that IAP can be

offered to all women testing positive. Screening was first recommended in

the US in 1996, when either risk-based prevention or universal screening

were recommended.17 This was followed by a revision in 2002

recommending universal screening.18 19 The incidence of EOGBS

decreased in the US from 1.7 per 1,000 live births to 0.6 in the 1990s,

to 0.3 in 2004 after the introduction of universal screening,20 21

and

was most recently estimated at 0.22 in 2016.22 The largest decline

occurred when either risk-based or screening prevention were

recommended,21 making it unclear how much of the decrease was a

result of screening itself. After the US guidelines were published,

universal GBS screening guidelines were developed in many countries,1 23

including Australia,24 Canada,25-27 Belgium,28 29 France,30 Germany,31 32

Italy,33 34

Spain,35 and Switzerland.

36 After the introduction of universal

screening, the incidence of EOGBS was reported as steady in Emilia-

Romagna in Italy at around 0.28 per 1,000 live births from 2003 to

2008.37 The rate decreased from 1.1 to 0.7 per 1,000 live births in

Spain,38 but more than doubled in Alberta in Canada from 0.15 per

1,000 live births in 2003 to 0.34 in 2013.39”

Page 12 of 49


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


We have, however, made a decision to keep the discussion of the studies

comparing screening to risk-based prevention in the section on the

effectiveness of screening where they can be described in more detail

along with the strength of evidence, rather than include them in the

introduction.

Finally, we have replaced ref 46 with a more recent and better conducted

systematic review of studies between 2000 and 2013 (Kurz and Davis,

2015).

I think there are some small issues not sufficiently covered:

-more attention to the state of the evidence for alternative screening approaches

eg GBS detection in labour? (given that the authors feel that screening at 35-37

weeks is not accurate)

Although we have focussed on culture based testing, as this is the

recommended test in the UK while rapid testing is not currently

recommended (as referenced in the test accuracy section), we have now

added information about PCR testing in the test section on pages 6 to 7 as

follows (bold text) in order to provide some discussion about this:

“A better test is required; however, a poor understanding of the natural

history of GBS limits the ability to identify such an approach. Rapid in-

labour tests have been developed. The most promising is real-time

polymerase chain reaction (PCR) testing. However, there are some

practical limitations that require further development before such

rapid tests can be recommended. These include cost, the length and

complexity of processing, inability to determine antibiotic sensitivity,

and/or they have lower sensitivity compared to culture.73-75

Furthermore, this approach to testing may not increase the PPV for

EOGBS compared to antenatal culture screening as only a small

proportion of colonised mothers in labour would have a neonate with

EOGBS.”

-Are there any established maternal harms/benefits from universal screening and

treatment? E.g. reduction in chorioamnionitis?

As far as we are aware, there is no information on the maternal benefits of

GBS screening. Most of the data on the maternal impact of IAP are for

Page 13 of 49


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


other purposes such as preterm labour and post-partum infection. We have

added the following information (bold text) on maternal harms into the

section on the harms of screening on pages 10 to 11:

“Maternal anaphylaxis is another important harm to consider as it

has potentially fatal consequences. However, as it is rare, it is difficult

to explore in well-designed studies other than very large RCTs. In the

US, four anaphylactic cases associated with GBS prophylaxis were

reported since the introduction of guidelines in 1996 up to 2010.19

Another study in Texas reported 19 cases of all-cause maternal

anaphylaxis (2.7 cases per 100,000 deliveries) in 2004-05.111 Eleven

cases were due to penicillin and cephalosporin and the authors

speculated that it was likely that these antibiotics were used for GBS

prevention. In the UK, the rate of all-cause maternal anaphylaxis has

been reported at 1.6 per 100,000 maternities (37 cases in three years,

11 due to penicillin) and one was a result of GBS prophylaxis. Two

mothers (5%) died and 14 (38%) mothers and 7 (41%) neonates

required intensive care admission.112 113 Other reported harms in small

observational studies at risk of bias, included neonatal respiratory

distress,114 maternal thrush,

115 and childhood atopic dermatitis.

116

- it should probably be acknowledged that ‘risk based’ strategies to identify

women for IAP are imperfectly applied, as demonstrated in the UK 2016 Audit

of current practice in preventing early-onset neonatal group B streptococcal

disease.

We agree that risk-based prevention may not have been perfectly applied

to assess its impact and we have added this in on page 2:

“The incidence of EOGBS has increased over this period in the UK by

around 0.09 per 1,000 live births,5 6 however, as the strategy has been

imperfectly implemented,6 12

it is difficult to identify its impact.”

Specific comments

1

Stand-first

I think this short summary could be re-phrased to provide more rationale as to

what questions are still to be decided in implementing a screening program e.g.

relatively few deaths averted versus unknown harms of universal screening and

We have changed the stand-first as follows (bold text) to provide questions

still to be decided in implementing a programme:

“Based on research evidence, the UK National Screening Committee

recommended that routine screening for group B Streptococcus carriage in

late pregnancy should not be introduced into the UK. In this paper, Seedat

Page 14 of 49


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


IAP. I am not sure about the claim that GBS is currently the most common

cause of neonatal infectious disease mortality, although it remains the most

common cause of neonatal sepsis (see below).

and colleagues review the evidence underpinning this decision,

addressing key issues yet to be considered before introducing a

screening programme, including the possibility of harm from the high

levels of unnecessary treatment with antibiotic prophylaxis due to a

poor test and the uncertainty of screening effectiveness.”

Introduction

1 P2, line 1 first sentence claim that GBS is most common infectious cause

of death. This should be referenced as recent studies have reported GBS is

most common cause of neonatal sepsis BUT E. coli may be the most

common cause of sepsis-related death (Simonsen et al, Clinical

Microbiology Reviews 2014). See also ‘Stand-first’ claim.

Thank you for pointing this out. We have changed the sentence on page 2

to:

“GBS is the commonest cause of neonatal sepsis and meningitis in

many developed countries.4”

2 P2, line 15 ‘pregnant women reporting with GBS risk factors’ might be

‘presenting with’?

Changed to “presenting”.

3 In the first paragraph it would be reasonable to mention that several other

high-resource countries have adopted the universal screening practice,

particularly after the US CDC guidelines were issued.

We have added the following sentence to the first paragraph on page 3 as

the introduction has been restructured according the comments above:

“After the US guidelines were published, universal GBS screening

guidelines were developed in many countries,1 23

including Australia,24

Canada,25-27

Belgium,28 29

France,30 Germany,

31 32 Italy,

33 34 Spain,

35 and

Switzerland.36”

4 It would be fair to quote reported rates of neonatal GBS sepsis in countries

that have adopted universal screening.

As mentioned above, we have added the following (bold text) on pages 2

to 3:

“Screening was first recommended in the US in 1996, when either risk-

based prevention or universal screening were recommended.17 This was

followed by a revision in 2002 recommending universal screening.18 19 The

incidence of EOGBS decreased in the US from 1.7 per 1,000 live births

to 0.6 in the 1990s, to 0.3 in 2004 after the introduction of universal

screening,20 21

and was most recently estimated at 0.22 in 2016.22 The

largest decline occurred when either risk-based or screening

prevention were recommended,21 making it unclear how much of the

decrease was a result of screening itself. After the US guidelines were

Page 15 of 49


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


published, universal GBS screening guidelines were developed in many

countries,1 23

including Australia,24 Canada,

25-27 Belgium,

28 29 France,

30

Germany,31 32 Italy,33 34 Spain,35 and Switzerland.36 After the introduction

of universal screening, the incidence of EOGBS was reported as

steady in Emilia-Romagna in Italy at around 0.28 per 1,000 live births


births in Spain,38 but more than doubled in Alberta in Canada from

0.15 per 1,000 live births in 2003 to 0.34 in 2013.39”

Epidemiology and natural history of a condition

1 Could the title be re-phrased to ‘Epidemiology and natural history of

GBS’?

Changed to “Natural history and epidemiology of EOGBS” on page 3

as natural history is presented first.

2 P3, line 57 reference for GBS as most common causes of neonatal death is

quite old (?2001 edition of Remington and Klein)- see above, could these

figures be updated? For example, the authors quote recent figures showing

a decline in GBS related mortality in the UK in recent years in the

subsequent paragraph.

Thank you for pointing this out. Changed to “EOGBS causes

considerable morbidity and mortality” on page 4.

3 P4, para 1: several figures from the UK enhanced surveillance programme

are quoted but the implications of the numbers are not really spelt out.

This would be helpful for the general reader. Are rates of EOGBS

incidence higher or lower than other countries with universal GBS

screening? If these are not comparable, then why not?

As mention in question 1 from the Editorial Team and question 4 from

Reviewer 2 three rows above, we have added the incidence of EOGBS in

other countries in the introduction: ‘The incidence of EOGBS decreased in

the US from 1.7 per 1,000 live births to 0.6 in the 1990s, to 0.3 in 2004

after the introduction of universal screening,20 21 and was most recently

estimated at 0.22 in 2016.22 The largest decline occurred when either risk-

based or screening prevention were recommended,21 making it unclear

how much of the decrease was a result of screening itself. After the

introduction of universal screening, the incidence of EOGBS was reported

as steady in Emilia-Romagna in Italy at around 0.28 per 1,000 live births


births in Spain,38 but more than doubled in Alberta in Canada from 0.15

per 1,000 live births in 2003 to 0.34 in 2013.39’

Page 16 of 49


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


We have also added the following sentence in this paragraph on page 4:

“The incidence of EOGBS in developed countries was recently

estimated at 0.37 per 1,000 live births.53”

4 p4, line 24-25: ‘Rates of GBS in term women’ should read ‘rate of

EOGBS in infants born to women at term’. This should be changes

throughout the paragraph to be accurate.

Changed accordingly on page 5.

5 P4, para 2: It should be clarified that women in pre-term labour would

usually receive antibiotics whether a risk-based approach or universal

screening based approach is used.

At the time that we submitted this review, this was not the case.

The RCOG only recommended GBS IAP for preterm births in Sept 2017.

We have added the following on page 5:

“Given this burden in preterm neonates, guidelines in the UK were

updated in September 2017 to recommend offering IAP to all women

in preterm labour.11

Most countries with risk-based or universal

screening prevention also offer IAP for preterm labour.”

Clinical effectiveness of IAP treatment

1 ?leave off the word ‘treatment’ in title, as IAP is by definition

‘prophylaxis’

Removed on page 7.

2 P6, line 47 ‘of whom approximately 99.8%’ (do you mean infants, rather

than pregnant women?).

It would be both pregnant women and their babies as pregnant women

would be treated but it is to prevent disease in the infant, therefore the

infant would also be affected. Changed as follows (bold text) on page 9:

“Approximately 99.8% (138,655 pregnant women and babies) would be

overdiagnosed and overtreated as they would have never had a neonate

with EOGBS even without IAP.”

3 P6 final para: I think the potential harms should be spelt out separately for

mothers and infants.

We have expanded this section and stated the harms to infants followed by

mothers. We have focused on the key findings of the report instead of

stating all of the investigated harms in infants and mothers as the evidence

on many of the harms was inconclusive (see pages 9 to 11).

4 P7, para 1: no references for studies that found an increase in AMR,

childhood atopic disease etc? It is difficult to refute the claims in the text

as the studies are not referenced.

We have now included the references for the studies in this section (pages

9 to 11).

Page 17 of 49


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

5 P7, line 22: should this read: ‘the UK Department of health’s

antimicrobial resistance strategy’?

Agree, rephrased on page 11.

Clinical effectiveness of GBS screening

1 The recent UK study observational study published in BMJ (reference 23)

did report a reduction in EOGBS disease with a screening programme

when only women who were actually screened were included in the

analysis; perhaps this should be acknowledged.

We have added this as suggested and stated the limitations of such an

analysis. We have also included the follow up study from this centre. On

page 8, we state the following:

“Additional analysis of mothers who were actually screened compared

to the pre-screening period was reported as statistically significant

(0.16 versus 0.99 per 1,000 live births, RR 0.16, p<0.05). However, the

mothers who accepted screening may have been systematically

different. The authors acknowledged that there were statistically

significant differences in several maternal characteristics between the

screening and risk-based periods. In addition, there was a low rate of

IAP in the carrier population and the authors suggested that factors

beyond the screening programme may have influenced the reduction

in EOGBS. The authors published a second study after the cessation

of their GBS programme finding that EOGBS incidence had increased

from 0.33 to 1.79 per 1,000 live births, which was statistically different

when adjusting for ethnicity (RR 5.67, p=0.009).82 Again, this study

may be biased due to the inclusion of differing populations across

differing time periods.”

Conclusion and future research

1 It is probably a bit dogmatic to sat that selective culture at 35-37 weeks is

not an accurate predictor of colonisation status in labour, although it is fair

to say that it may not be an accurate indicator of EOGBS disease in

neonates. Figures that the authors quote for PPV are 60-80% for

colonisation at time of labour.

We have changed the sentence as follows on page 12:

“Selective maternal culture is not an accurate predictor of EOGBS

disease in neonates, and a lack of understanding about why some

colonised mothers have a neonate with EOGBS limits the ability to

identify an approach that reduces the rate of overdiagnosis.”

Box 1

I think the authors should consider revising the order of their recommendations,

and making them more succinct. In particular, the authors clearly feel that an

As per the recommendations of reviewer 1, we have removed the

recommendation of an RCT and moved it into the main text on page 11.

Page 18 of 49


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


RCT of risk-based versus universal screening approach will still not yield a test

with a high positive predictive value and therefore an RCT might not alter their

assessment that a screening based approach is not indicated. Maybe they should

emphasise other approaches that should be addressed in research e.g. rapid GBS

detection in labour, predictive tests for neonatal EOGBS etc

We have also moved the other research recommendations into the text on

pages 11 to 12 (as follows) and removed the extra details in each

recommendation to make them succinct. These recommendations do

address research to find a better approach to identify high risk women,

including development of rapid tests and predicting neonatal EOGBS:

“In addition to trial evidence, research to understand the natural

history of GBS could help to improve the balance of benefits and

harms for future proposed screening programmes. Better

identification of the women at most risk of having a neonate with

EOGBS could reduce the amount of overtreatment to thousands of

women and their babies. To this end, research is needed to better

predict which mothers with GBS carriage will transmit GBS to the

neonate, which mothers will have a neonate that develops EOGBS,

and which neonates with GBS colonisation will progress to EOGBS

disease. Similarly, test accuracy research is needed to reliably detect

GBS colonisation and bacterial load during labour, including

development of the latest in-labour tests and accurate measurement of

who is colonised in labour and how heavily. Finally, research is

required on the long-term follow up of neonates who are diagnosed

with EOGBS to understand the clinical morbidity experienced.”

If this response does not meet the reviewer’s suggestion, please expand

and we will be happy to adjust.

Key messages

I don’t think it is fair to say the screening rectovaginal culture is not an accurate

predictor of colonisation status in labour, although it may not be an accurate

indicator of EOGBS disease in neonates. Figures the authors quote for PPV are

60-80% for colonisation at time of labour.

We have changed the sentence as follows (bold text) on page 13:

“The natural history, particularly, the development from maternal GBS

carriage to EOGBS disease is poorly understood and maternal carriage

is a poor predictor of EOGBS.”

Page 19 of 49


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


1

Universal antenatal screening for carriage of group B Streptococcus (GBS): State of the

evidence

Farah Seedat PhD,1

Julia Geppert PhD,1 Chris Stinton PhD,

1 Jacoby Patterson MD,

1 Karoline

Freeman MSc,1

Samantha Ann Johnson MA,1

Hannah Fraser BSc,1 Colin Stewart Brown

FRCPath,2

Olalekan A Uthman PhD,1

Bee Tan FRCS,1

Esther R Robinson BMBCh DPhil,3 Noel

Denis McCarthy MB Dphil,1

Aileen Clarke MD,1 John Marshall MA,

4 Cristina Visintin PhD,

4

Anne Mackie FFPH,4 Sian Taylor-Phillips PhD

1

1. Division of Health Sciences, University of Warwick Medical School, Gibbet Hill

Campus, Coventry, CV4 7AL, UK.

2. Bacteria Reference Department, National Infection Service, Public Health England,

61 Colindale Ave, London, NW9 5EQ, England.

3. Field Service, National Infection Service, Public Health England, Seaton House

Nottingham, NG2 4LA.

4. UK National Screening Committee, Floor 5, Wellington House, 133-155 Waterloo

Road, London, SE1 8UG.

Corresponding author: Dr Sian Taylor-Phillips, Division of Health Sciences, University of

Warwick Medical School, Gibbet Hill Campus, Coventry, CV4 7AL, UK, E) S.Taylor-

[email protected] , T) +44(0) 2476 575882

Stand-first:

Based on research evidence, the UK National Screening Committee recommended that

routine screening for group B Streptococcus carriage in late pregnancy should not be

introduced into the UK. In this paper, Seedat and colleagues review the evidence

underpinning this decision, addressing key issues yet to be considered before introducing a

screening programme, including the possibility of harm from the high levels of unnecessary

treatment with antibiotic prophylaxis due to a poor test and the uncertainty of screening

effectiveness.

Page 20 of 49


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

2

Introduction

Group B Streptococcus (Streptococcus agalactiae, GBS) is a Gram-positive bacterium that

colonises the gastrointestinal and genitourinary tract in approximately 30% of healthy

adults.1-3

It usually causes no harm.1 However, GBS is the commonest cause of neonatal

sepsis and meningitis in many developed countries.4 In the UK, GBS causes invasive disease

in the first six days of life (early onset GBS or EOGBS) in around 1 of every 2,000 live births.55

Neonates with EOGBS can develop sepsis, pneumonia, and meningitis, and around 5% to

10% of these babies can die as a result.5-7

To prevent EOGBS, a risk-based strategy has been recommended in the UK since 2003.8-11

Pregnant women presenting with GBS risk factors are offered intrapartum antibiotic

prophylaxis (IAP), usually intravenous penicillin, during labour.8-11

The incidence of EOGBS

has increased over this period in the UK by around 0.09 per 1,000 live births,5 6

however, as

the strategy has been imperfectly implemented,6 12

it is difficult to identify its impact. In the

Netherlands, the incidence of EOGBS has increased under risk-based prevention from 0.11

to 0.19 per 1,000 live births (p<0.001).13

By contrast, under risk-based prevention the

incidence of EOGBS in New Zealand has decreased from 0.5 to 0.26 per 1,000 live births,

from 0.4 to 0.3 per 1,000 live births in Sweden, and from 0.6 to 0.2 per 1,000 live births in

Denmark (from 1995 to 2002).14-16

A criticism of the risk-based strategy is that risk factors are absent in 30% to 60% of

pregnancies that result in EOGBS who would be excluded from prophylactic prevention,

although of course we have no way of telling in advance who these cases are. There are

regular media and political calls to introduce a universal antenatal GBS screening

programme in addition to the current risk-based strategy. Universal screening would involve

the collection of specimens using rectovaginal swabs at 35 to 37 weeks gestation, which are

processed using selective culture media, to identify women colonised with GBS so that IAP

can be offered to all women testing positive. Screening was first recommended in the US in

1996, when either risk-based prevention or universal screening were recommended.17

This

was followed by a revision in 2002 recommending universal screening.18 19

The incidence of

EOGBS decreased in the US from 1.7 per 1,000 live births to 0.6 in the 1990s, to 0.3 in 2004

after the introduction of universal screening,20 21

and was most recently estimated at 0.22 in

Page 21 of 49


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


3

2016.22

The largest decline occurred when either risk-based or screening prevention were

recommended,21

making it unclear how much of the decrease was a result of screening

itself. After the US guidelines were published, universal GBS screening guidelines were

developed in many countries,1 23

including Australia,24

Canada,25-27

Belgium,28 29

France,30

Germany,31 32

Italy,33 34

Spain,35

and Switzerland.36

After the introduction of universal

screening, the incidence of EOGBS was reported as steady in Emilia-Romagna in Italy at

around 0.28 per 1,000 live births from 2003 to 2008.37

The rate decreased from 1.1 to 0.7

per 1,000 live births in Spain,38

but more than doubled in Alberta in Canada from 0.15 per

1,000 live births in 2003 to 0.34 in 2013.39

Most screening recommendations have been made by clinical organisations.24 31 32 34-36 40 41

Crucially, however, screening recommendations should be assessed against screening

criteria,42

which require that the harms as well as benefits of a programme, are examined.

Here, we summarise the best available evidence on universal GBS screening from two

comprehensive evidence reviews undertaken for the UK National Screening Committee

(NSC) and have also included key papers published since the conclusions of those reviews.43

44 The reviews address UK NSC screening criteria, examining 1) the natural history

(asymptomatic stage to illness) and epidemiology (severity, incidence) of EOGBS; 2) the

accuracy of the GBS culture screening test (is it able to separate women at high risk of

having a neonate with EOGBS from those at low risk?); 3) the effectiveness of IAP

treatment; and 4) whether a universal GBS screening programme delivers more benefit than

harm (at a reasonable cost).

1) Natural history and epidemiology of EOGBS

GBS is passed from mother to neonate if the bacteria colonise the vagina around the time of

labour. The GBS carriage rate in pregnant women varies, and a recent meta-analysis

estimated that in developed countries GBS is found in around 19% of tested women.45

In

the UK, it is estimated to be around 21% to 22%,46 47

although a small recent study in a

London centre reported rates as high as 29%.48

Crucially, not all neonates born to GBS-

colonised mothers will be colonised, and not all colonised neonates will go on to develop

EOGBS. Of the women testing GBS-positive during pregnancy, a systematic review

estimated that around 31% are estimated to test culture negative in labour.49

In women

Page 22 of 49


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


4

testing positive during labour, there is a 36% (95% confidence interval [CI] 28.1% to 45.0%)

chance that they will transmit GBS to their neonates.50

The majority of neonates colonised

with GBS will remain asymptomatic; however, about 3% (95% CI 1.6% to 4.7%) will develop

EOGBS disease.50

Internationally, the risk of EOGBS in GBS colonised mothers (in settings

without an IAP policy) was recently estimated at around 1%.51

This means that in the

proposed screening programme a large number of women who carry GBS would be eligible

for antibiotics during labour when there is no risk of EOGBS to their neonates. The

mechanisms by which this common commensal infects and subsequently harms a small

proportion of neonates remain unclear.52

The incidence of EOGBS in developed countries was recently estimated at 0.37 per 1,000

live births.53

In the UK and the Republic of Ireland, enhanced national surveillance during

2014-15 recorded an EOGBS incidence rate of 0.57 per 1,000 live births (518 cases).5 6

This

was statistically different to the earlier enhanced surveillance conducted in 2000-01 (before

national prevention guidelines), which found a rate of 0.48 per 1,000 live births.7 While this

is the best available data in the UK, the studies have only been conducted at two points in

time (14 years apart) making it unclear how figures might fluctuate annually. The authors

also mention that the observed increase might be at least in part due to technical

improvements in bacterial culture practices, increased awareness of neonatal GBS, or

increased case ascertainment. Finally, only culture-proven EOGBS cases were included, and

although this is the standard outcome reported in GBS studies internationally, it is likely to

underestimate the true burden of the disease due to IAP given to women with risk factors.

EOGBS causes considerable morbidity and mortality. Neonates rapidly develop sepsis in 63%

of EOGBS cases, pneumonia in around 24%, and meningitis in around 13%.6 Around 5% of

EOGBS cases died in the UK in 2014-15; this was statistically different from 10.6% reported

in 2000-01.6 Of all early neonatal deaths (within seven days) in the UK, it was estimated that

1.7% were related to GBS in 2014.54

The long-term outcomes in EOGBS survivors are not

well researched. Neurological impairment at hospital discharge or last paediatric review has

been reported in up to 15.8% of EOGBS survivors.55-57

GBS meningitis survivors have a

higher risk of long-term neurodevelopmental disability.56 58 59

Page 23 of 49


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


5

Preterm birth is an important risk factor for EOGBS, and both incidence and case fatality

have been inversely associated with gestational age at birth.6 The rate and outcomes of

EOGBS in infants born to women at term is the important issue for screening because the

proposed screening test is at 35 to 37 weeks gestation and would only identify neonates

born after this time. However, the rate of EOGBS is rarely reported in this way. In the UK,

the rate of EOGBS in infants born to women at term has not changed much since earlier

estimates of 0.38 per 1,000 live births in 2007.60

Indeed, comparing the incidence rate of

EOGBS in infants born to women at term only (instead of all women) between the UK and

countries that have universal screening guidelines reduces the gap in incidence (US: 0.22 per

1,000 live births,61

Italy: 0.24 per 1,00037

). Given this burden in preterm neonates,

guidelines in the UK were updated in September 2017 to recommend offering IAP to all

women in preterm labour.11

Most countries with risk-based or universal screening

prevention also offer IAP for preterm labour.

Aside from preterm labour, a number of maternal factors have been associated with EOGBS,

including incidental GBS colonisation, a previous infant with GBS disease, GBS bacteriuria,

intrapartum fever, and suspected chorioamnionitis.11 62

If a woman presents with these risk

factors in labour, the UK guidelines recommend that IAP is offered.8-11

However, in 2014-15,

the percentage of EOGBS neonates born at term without these maternal risk factors

(therefore not eligible for IAP) was around 65%.6 The percentage of term EOGBS deaths

without maternal risk factors was around 56% to 67% (n=5-6/9) depending on the risk

factors included in the analysis.6 It is this group of term neonates with no preceding risk

factors which a screening programme could impact.

Figure 1 summarises the natural history and epidemiology of GBS for a hypothetical cohort

of term pregnant women in 2000 (no prevention guideline) and 2014 (risk-based prevention

guideline). It is clear that EOGBS is an important health problem, yet there is a poor

understanding about the natural history of GBS from maternal GBS colonisation to the long-

term outcomes of EOGBS.

2) Accuracy of selective rectovaginal culture at 35 to 37 weeks

Page 24 of 49


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


6

The suggested screening test for detecting GBS in the UK is selective enriched culture of

rectovaginal swabs taken at 35 to 37 weeks gestation.63

Studies tend to use detection of

GBS carriage at birth as the measure of test accuracy for culture of swabs at 35 to 37 weeks.

However, the outcome of interest should be EOGBS disease in the neonates. Based on the

UK surveillance data shown in Figure 1, we estimate that the positive predictive value (PPV)

of antenatal culture predicting EOGBS may be around 2/1,000 (0.2%). The 2000-01 UK

surveillance study shows that there were around 205 term neonates affected by EOGBS per

year in the UK at a time with no national prevention guideline.7 There were 679,029 babies

born in the UK in 2000,64

of whom approximately 573,448 reached full term (37 weeks

onwards) and were not born by elective caesarean section.65

If around 22% of women were

colonised with GBS,47

that would equate to approximately 126,159 GBS carriers at 35 to 37

weeks gestation. Therefore, if there was 100% test uptake and no change in GBS

colonisation status from screening until labour, the PPV of a screening test performed at 35

to 37 weeks for detecting EOGBS in the neonate would be 0.16% (205/126,159). Performing

the same calculations using 2014 surveillance figures gives a PPV value of 0.25%

(350/138,933),6 64 66

which would deliver an unacceptably high level of false positive results.

Two prospective studies67 68

included in a systematic review found that the PPV of selective

rectovaginal culture at 35 to 37 weeks gestation to predict GBS carriage at birth was 67.4%

and 78.6%.49 67 68

The reported negative predictive values were 96.7% and 93.5%.67 68

Small

recent studies identified in the 2016 UK NSC review showed similar values.69-72

Overall,

approximately 20% to 30% of pregnant women who test positive for GBS at 35 to 37 weeks

gestation, test negative in labour. It is unclear whether the reported changes in colonisation

status are a result of false test results at the point of screening or if they reflect natural

fluctuations in GBS colonisation during late pregnancy.

A better test is required; however, a poor understanding of the natural history of GBS limits

the ability to identify such an approach. Rapid in-labour tests have been developed. The

most promising is real-time polymerase chain reaction (PCR) testing. However, there are

some practical limitations that require further development before such rapid tests can be

recommended. These include cost, the length and complexity of processing, inability to

determine antibiotic sensitivity, and/or they have lower sensitivity compared to culture.73-75

Page 25 of 49


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


7

Furthermore, this approach to testing may not increase the PPV for EOGBS compared to

antenatal culture screening as only a small proportion of colonised mothers in labour would

have a neonate with EOGBS.

3) Effectiveness of intrapartum antibiotic prophylaxis

Intrapartum antibiotic prophylaxis is currently the recommended mainstay of EOGBS

prevention. A Cochrane meta-analysis published in 2014 estimated that the use of IAP

reduced the risk of culture-proven and probable EOGBS compared with no treatment by

83%.76

However, IAP did not reduce the incidence of all-cause mortality, mortality from GBS,

or mortality from other infections. Due to the high risk of bias identified in the three

included small randomised controlled trials (RCTs) which were conducted more than 20

years ago, the authors concluded that there is no valid information to inform clinical

practice.76

Although IAP may be effective in reducing EOGBS infections, the poor evidence

base makes it uncertain how effective it is. Echoing the uncertainty, a meta-analysis in 2007

using different statistical methods produced a more extreme treatment effect of IAP,

estimating a risk reduction in EOGBS of 97%.77

4) Benefits and harms of a universal GBS screening programme

Benefits

The evidence on the clinical effectiveness of universal GBS screening is limited and focusses

on incidence rather than clinical outcomes. There have been no RCTs assessing the effects

of antenatal screening on the reduction of morbidity, clinical outcomes, and mortality from

EOGBS. RCTs on the effects of screening on the clinical outcomes and mortality from EOGBS

would be challenging due to immense sample size requirements. However, in the absence

of RCTs, it is difficult to quantify the potential impact of adding universal screening to risk-

based practice. The risk of bias from observational study designs is well-documented, in

particular, due to the risk of confounding factors contributing to results. The majority of

studies on GBS screening use historical controls and compare the rates of EOGBS during

different periods of time in which different GBS prevention guidelines were introduced. The

control periods (no screening and/or risk-based prevention) precede the universal screening

Page 26 of 49


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

8

periods. Risk of bias is higher in these studies as participants in the study and control period

are not contemporaneous so other changes occurring between these periods may

contribute to the results. Even in the few observational studies that compare screening to

concurrent controls, these studies often retrospectively compare women with a culture

result to all other women; this may be biased as people who accept screening are

systematically different to those who do not.78

Another limitation, as mentioned above, is

that many studies do not report the findings in term pregnant women alone, which is the

eligible population that screening could impact. Finally, many studies only include culture-

proven EOGBS, thus changes may reflect a decreased likelihood of neonatal cultures being

positive due to IAP use, with culture negative cases of EOGBS being undetected. This may

distort the impact of screening.

A systematic review published in 2015 of nine observational studies from Turkey, Australia

and the US found that the odds of EOGBS under universal screening were 55% lower than

under risk-based prevention for all neonates and for term neonates only (three studies).79

There were limitations which may affect the internal and external validity of the studies.

Risk-based guidelines were not always clear and may have differed between studies, there

was a risk of misclassification bias, and the majority of the studies were based in the US.79

Findings from two more recent US studies also consistently report a decreased EOGBS

incidence with universal screening compared to the era without any screening, however,

reported benefits compared to the era with a risk-based approach are inconsistent.80 81

A

similar recent study in a UK maternity unit found that the rate of EOGBS fell from 0.99 per

1,000 live births in the risk-based period to 0.33 per 1,000 live births during the screening

period; however, this did not reach statistical significance (risk ratio [RR] 0.33, p=0.08).48

Additional analysis of mothers who were actually screened compared to the pre-screening

period was reported as statistically significant (0.16 versus 0.99 per 1,000 live births, RR

0.16, p<0.05). However, the mothers who accepted screening may have been systematically

different. The authors acknowledged that there were statistically significant differences in

several maternal characteristics between the screening and risk-based periods. In addition,

there was a low rate of IAP in the carrier population and the authors suggested that factors

beyond the screening programme may have influenced the reduction in EOGBS. The authors

Page 27 of 49


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


9

published a second study after the cessation of their GBS programme finding that EOGBS

incidence had increased from 0.33 to 1.79 per 1,000 live births, which was statistically

different when adjusting for ethnicity (RR 5.67, p=0.009).82

Again, this study may be biased

due to the inclusion of differing populations across differing time periods.

Internationally there has been a suggestion for studies to explore all-cause early-onset

sepsis, however, these findings have been contradictory.61 83 84

The impact of universal GBS screening on clinical outcomes of EOGBS is unclear. In relation

to mortality, results are inconsistent. Two studies conducted in the US did not find a change

in the mortality rate between periods with and without universal screening.80 81

On the

other hand, a Hungarian study reported decreased EOGBS mortality rates from 19.5%

(29/149) to 1.6% (1/63) after the introduction of screening compared with no prevention.85

Harms

An examination of the potential harms is a standard part of the assessment of any screening

proposal. Based on the estimated PPVs of culture testing indicated above (0.2%),

approximately 138,933 term pregnant women per year would be eligible for IAP.

Approximately 99.8% (138,655 pregnant women and babies) would be overdiagnosed and

overtreated as they would have never had a neonate with EOGBS even without IAP. This

makes assessing the harms of treatment particularly important. A cost-effectiveness model

published in the UK also estimated that adding screening to risk-based prevention would

result in 99.8% overtreatment (increasing antibiotic use in the pregnant population from

11% to 27%).77

Recently, an expert group convened by the UK NSC completed a modelling

exercise concluding that adding screening to a risk-based strategy in the UK would result in

an additional 1,675 to 1,854 women receiving IAP to prevent one EOGBS case and 24,065 to

32,087 to prevent one EOGBS death.47

This is similar to another study where the authors

found that 1,191 women would need to be treated with IAP to prevent one EOGBS case.86

Although the models contain a level of uncertainty because of the limitations and gaps in

the evidence, the estimates all support the high levels of overtreatment that would occur

when adding screening to risk-based prevention.

Page 28 of 49


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


10

A systematic review of 30 studies found little evidence to quantify the potential harms of

IAP to mothers and babies.87

Although a range of adverse effects were investigated, studies

specifically on GBS prophylaxis were observational and at high or unclear risk of bias while

13 RCTs at lower risk of bias investigated different drugs and regimens to GBS prophylaxis.

Key findings from the review were around changes in gut microbiota, long term outcomes,

and antibiotic resistance.

There was consistent evidence that IAP for GBS prophylaxis alters neonatal gut microbiota

up to 90 days of age.88-94

These findings have been mirrored in similar studies,95 96

however,

it is unknown whether microbiota alterations specifically from GBS prophylaxis are

associated with long-term clinical outcomes. Early infant exposure to antibiotics and gut

microbiota changes have been implicated in autism, metabolic problems such as obesity

and diabetes, and atopic, inflammatory, and autoimmune problems such as asthma,

allergies, necrotising enterocolitis, and Crohn’s disease.97-100

The impact of IAP on antibiotic

resistance was inconsistent in the literature, with some evidence of an increase in the

resistance of some antibiotics for some pathogens, with others showing no increase.93 101-105

Globally, most GBS isolates are susceptible to penicillin,106

however, in the US in 2005, 0.2%

of GBS isolates were reaching the upper level of susceptibility for one or more beta-

lactams.107

Finally, there was a particular lack of information on the long-term outcomes of

IAP. There was evidence from only one RCT using different IAP to GBS prophylaxis which

showed that IAP was moderately associated with serious consequences of functional

impairment and cerebral palsy at seven years of age.108

However, the biological plausibility

of IAP causing cerebral palsy is unknown and another trial on IAP for preterm rupture of

membranes found no evidence that IAP was associated with cerebral palsy.108-110

Maternal anaphylaxis is another important harm to consider as it has potentially fatal

consequences. However, as it is rare, it is difficult to explore in well-designed studies other

than very large RCTs. In the US, four anaphylactic cases associated with GBS prophylaxis

were reported since the introduction of guidelines in 1996 up to 2010.19

Another study in

Texas reported 19 cases of all-cause maternal anaphylaxis (2.7 cases per 100,000 deliveries)

in 2004-05.111

Eleven cases were due to penicillin and cephalosporin and the authors

speculated that it was likely that these antibiotics were used for GBS prevention. In the UK,

Page 29 of 49


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


11

the rate of all-cause maternal anaphylaxis has been reported at 1.6 per 100,000 maternities

(37 cases in three years, 11 due to penicillin) and one was a result of GBS prophylaxis. Two

mothers (5%) died and 14 (38%) mothers and 7 (41%) neonates required intensive care

admission.112 113

Other reported harms in small observational studies at risk of bias, included

neonatal respiratory distress,114

maternal thrush,115

and childhood atopic dermatitis.116

The current evidence on the benefits and harms of universal GBS screening is based mainly

on observational studies that are small, subject to bias, and/or have applicability concerns.

Therefore, the balance of benefits and harms from GBS screening and IAP cannot be

quantified, especially as the long-term impact of EOGBS and of IAP is poorly understood.

The scale of overtreatment and range of plausible harms require a better understanding

before we can be confident that IAP at this scale is a safe undertaking. In the era of

antimicrobial resistance, such a widespread IAP strategy may also be challenging in relation

to the UK Department of Health and Social Care’s antimicrobial resistance strategy to

reduce unnecessary use of antibiotics.117

Research agenda for EOGBS prevention

In 2007, a cost-effectiveness model concluded that investments in prenatal vaccination

research should be prioritised over screening research.77

However, interest has remained to

evaluate the impact of universal GBS screening. In 2017, the Health Technology Assessment

launched an expression of interest for a cluster randomised trial assessing whether

screening women for carriage of GBS in late pregnancy reduces the occurrence of early-

onset sepsis, whether the benefits outweigh the harms, and whether it would be cost-

effective.

In addition to trial evidence, research to understand the natural history of GBS could help to

improve the balance of benefits and harms for future proposed screening programmes.

Better identification of the women at most risk of having a neonate with EOGBS could

reduce the amount of overtreatment to thousands of women and their babies. To this end,

research is needed to better predict which mothers with GBS carriage will transmit GBS to

their neonate, which mothers with GBS carriage will have a neonate that develops EOGBS,

and which neonates with GBS colonisation will progress to EOGBS disease. Similarly, test

Page 30 of 49


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


12

accuracy research is needed to reliably detect GBS colonisation and bacterial load during

labour, including development of the latest in-labour tests and accurate measurement of

who is colonised in labour and how heavily. Finally, research is required on the long-term

follow up of neonates who are diagnosed with EOGBS to understand the clinical morbidity

experienced.

There is also a lack of evidence about what currently happens in practice. Little is known

about the impact of private GBS carriage testing or the values and priorities informing

women’s decision-making. Further work should be undertaken to understand this.

Conclusions

GBS infection is an important health problem and more work to understand and prevent

neonatal invasive disease is required. Universal antenatal GBS culture screening is a

complex area and a poor evidence base means that, so far, it does not meet UK NSC criteria

needed to introduce screening programmes. Selective maternal culture is not an accurate

predictor of EOGBS disease in neonates, and a lack of understanding about why some

colonised mothers have a neonate with EOGBS limits the ability to identify an approach that

reduces the rate of overdiagnosis. At present, if a GBS screening programme is

implemented, it would offer all term pregnant women the antenatal GBS culture test, but

over 99% of screen-positive mothers (and their babies) would be overdiagnosed and

unnecessarily receive IAP, as they would not have a neonate with EOGBS even without

intervention. The harm from widespread IAP to thousands of pregnant women and their

babies is unknown while the evidence on the benefit from a screening programme is

uncertain due to lower quality studies with serious limitations. The long-term morbidity of

EOGBS is also largely unknown as is the impact of screening upon it. Given the state of the

evidence base, the balance of the benefits and harms from introducing universal GBS

screening cannot be quantified.

Page 31 of 49


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

13

Boxes

Figure caption / legend:

Figure 1. Natural history of GBS in a hypothetical cohort of term pregnant women in year

2000 (no prevention guideline) and 2014 (risk-based guideline)

Abbreviations: GBS Group B Streptococcus, EOGBS early-onset group B Streptococcus, NPV Negative predictive value, PPV

Positive predictive value

Notes: Due to the uncertainties of the data, the numbers should be treated cautiously for a sense of scale but not as exact

estimates. Data estimates and sources:

a. Pregnant women available for screening in 2000 and 2014: All livebirths taken from the Office for National

Statistics,64

then elective caesarean sections and preterm births (<37 weeks) were removed from the cohort

using HES estimates,65 66

as elective caesarean sections are not at risk of EOGBS and preterm births are not

eligible for screening. Note: Rate for pre-term births in 2000 is taken from 2004-05.

b. Maternal GBS carriage: 22%.47

c. Number of EOGBS disease cases and mortality taken from British Paediatric Surveillance Unit study.5-7

d. Long-term disability: 8.7-15.8% of surviving EOGBS cases.55-57

e. Short-term EOGBS morbidity: Meningitis 13.2%; Sepsis 63.1%; Pneumonia 23.7%.6

f. EOGBS cases with maternal risk factors: 33-37% of EOGBS cases will have at least one risk factor for intrapartum

antibiotic prophylaxis.6

Key messages

1. Early-onset group B Streptococcus disease (< 7 days of life, EOGBS) is an

important health problem and efforts should continue to better understand

and prevent it. However, a screening programme is not currently

appropriate as internationally agreed screening criteria are not met.

2. The natural history, particularly, the development from maternal GBS

carriage to EOGBS disease is poorly understood and maternal carriage is a

poor predictor of EOGBS.

3. This could lead to over 99% of screen-positive women and their babies

receiving unnecessary intrapartum antibiotic prophylaxis (IAP).

4. The addition of universal screening to risk-based prevention appears to

reduce the incidence of EOGBS. However, a lack of high quality evidence on

the harms and benefits of GBS screening and widespread IAP means that

quantifying the impact of universal screening in addition to risk-based

prevention is not possible.

Page 32 of 49


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


14

References

1. Rodriguez-Granger J, Alvargonzalez JC, Berardi A, et al. Prevention of group B

streptococcal neonatal disease revisited. The DEVANI European project. Eur J Clin

Microbiol Infect Dis 2012;31(9):2097-104. doi: 10.1007/s10096-012-1559-0

[published Online First: 2012/02/09]

2. Edwards M, Baker C. Streptococcus agalactiae (group B streptococcus). In: Mandell G,

Bennett J, Dolin R, eds. Principles and practice of infectious diseases. 7 ed.

Philadelphia: Elsevier 2010.

3. Edwards M, Nizet V. Group B streptococcal infections. In: Remington J, Klein J, Wilson C,

et al., eds. Diseases of the fetus and newborn infant. 7 ed. Philadelphia: Elsevier

2011:419–69.

4. Le Doare K, Heath PT. An overview of global GBS epidemiology. Vaccine 2013;31 Suppl

4:D7-12. doi: 10.1016/j.vaccine.2013.01.009 [published Online First: 2013/08/30]

5. O'Sullivan C, Lamagni T, Efstratiou A, et al. P3 Group B Streptococcal (GBS) disease in UK

and Irish infants younger than 90 days, 2014–2015. Archives of disease in childhood

2016;101(Suppl 1):A2-A2. doi: 10.1136/archdischild-2016-310863.3

6. O’Sullivan C, Heath PT, on behalf of the British Pediatric Surveilance Unit. Group B

Streptococcal (GBS) disease in UK and Irish infants younger than 90 days, 2014-2015,

2016.

7. Heath PT, Balfour G, Weisner AM, et al. Group B streptococcal disease in UK and Irish

infants younger than 90 days. Lancet 2004;363(9405):292-4. doi: 10.1016/S0140-

6736(03)15389-5 [published Online First: 2004/01/31]

8. National Institute for Health and Clinical Excellence. Antibiotics for early-onset neonatal

infection: antibiotics for the prevention and treatment of early-onset neonatal

infection. United Kingdom: National Institute for Health and Clinical Excellence,

2012.

9. Royal College of Obstetricians and Gynaecologists. Prevention of Early Onset Neonatal

Group B Streptococcal Disease. Guideline No 36: RCOG, 2003.


Group B Streptococcal Disease. Green-top Guideline No. 36. 2 ed: Royal College of

Obstetricians and Gynaecologists, 2012.

Page 33 of 49


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


15


Group B Streptococcal Disease. Green-top Guideline No. 36. 3 ed: RCOG, 2017.

12. Royal College of Obstetricians & Gynaecologists (RCOG). Audit of current practice in

preventing early-onset neonatal group B streptococcal disease in the UK. First

report., 2015.

13. Bekker V, Bijlsma MW, van de Beek D, et al. Incidence of invasive group B streptococcal

disease and pathogen genotype distribution in newborn babies in the Netherlands

over 25 years: a nationwide surveillance study. The Lancet Infectious Diseases

2014;14(11):1083-89. doi: 10.1016/S1473-3099(14)70919-3

14. Darlow BA, Voss L, Lennon DR, et al. Early-onset neonatal group B streptococcus sepsis

following national risk-based prevention guidelines. The Australian & New Zealand

journal of obstetrics & gynaecology 2016;56(1):69-74. doi: 10.1111/ajo.12378


15. Hakansson S, Lilja M, Jacobsson B, et al. Reduced incidence of neonatal early onset

group B streptococcal infection after promulgation of guidelines for risk-based

intrapartum antibiotic prophylaxis in Sweden: analysis of a national population-

based cohort. Acta obstetricia et gynecologica Scandinavica 2017 doi:

10.1111/aogs.13211 [published Online First: 2017/08/24]

16. Ekelund K, Konradsen HB. Invasive group B streptococcal disease in infants: a 19-year

nationwide study. Serotype distribution, incidence and recurrent infection.

Epidemiology and infection 2004;132(6):1083-90. [published Online First:

2005/01/08]

17. Schuchat A, Whitney CG, Zangwill K. Prevention of perinatal group B streptococcal

disease: a public health perspective. 1996

18. Schrag S, Gorwitz R, Fultz-Butts K, et al. Prevention of perinatal group B streptococcal

disease. Revised guidelines from CDC. MMWR Recomm Rep 2002;51(RR-11):1-22.


19. Verani JR, McGee L, Schrag SJ. Prevention of perinatal group B streptococcal disease--

revised guidelines from CDC, 2010. MMWR Recomm Rep 2010;59(RR-10):1-36.


Page 34 of 49


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


16

20. Schrag SJ, Zywicki S, Farley MM, et al. Group B streptococcal disease in the era of

intrapartum antibiotic prophylaxis. N Engl J Med 2000;342(1):15-20. doi:

10.1056/NEJM200001063420103 [published Online First: 2000/01/06]

21. Centers for Disease Control and Prevention. Early-onset and late-onset neonatal group B

Streptococcal disease – United States, 1996–2004. MMWR CDC Surveill Summ

2005;54:1205–8.

22. Centers for Disease Control and Prevention. Active Bacterial Core Surveillance Report,

Emerging Infections Program Network, Group B Streptococcus, 2016, 2018.

23. Homer CS, Scarf V, Catling C, et al. Culture-based versus risk-based screening for the

prevention of group B streptococcal disease in newborns: a review of national

guidelines. Women and birth : journal of the Australian College of Midwives

2014;27(1):46-51. doi: 10.1016/j.wombi.2013.09.006 [published Online First:

2013/11/19]

24. Royal Australian and New Zealand College of Obstetricians and Gynaecologists. Maternal

Group B Streptococcus in pregnancy: screening and management, C-Obs 19.

Melbourne: RANZCOG, 2016.

25. Money D, Allen VM. The prevention of early-onset neonatal group B streptococcal

disease. Journal of obstetrics and gynaecology Canada : JOGC = Journal d'obstetrique

et gynecologie du Canada : JOGC 2013;35(10):939-48. doi: 10.1016/s1701-

2163(15)30818-5 [published Online First: 2013/10/30]

26. Money DM, Dobson S. The prevention of early-onset neonatal group B streptococcal

disease. Journal of obstetrics and gynaecology Canada: JOGC = Journal d'obstetrique

et gynecologie du Canada: JOGC 2004;26(9):826-40.

27. Canadian Task Force on Preventive Health Care. Prevention of group B streptococcal

infection in newborns: Recommendation statement from the Canadian Task Force

on Preventive Health Care. CMAJ 2002;166(7):928-30.

28. Melin P. Prevention of perinatal group B streptococcal diseases: Belgian guidelines.

Round Table Series - Royal Society of Medicine 2007(85):29-41.

29. Melin P, Verschraegen G, Mahieu L, et al. Towards a Belgian consensus for prevention of

perinatal group B streptococcal disease. Indian Journal of Medical Research,

Supplement 2004;119:197-200.

Page 35 of 49


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


17

30. Agence Nationale d’Accreditation et d’Evaluation en Sante. Antenatal prevention of

early neonatal bacterial infection - Clinical Practice Guidelines. France: ANAES, 2001.

31. Leitlinien der Gesellschaft für Neonatologie und Pädiatrische Intensivmedizin (GNPI),

Deutschen Gesellschaft für Gynäkologie und Geburtshilfe, Deutschen Gesellschaft

für Pädiatrische Infektiologie (DGPI), et al. Prophylaxe der Neugeborenensepsis

(frühe Form) durch Streptokokken der Gruppe B: AWMF, 2008.

32. Berufsverband der Frauenärzte e.V. (BVF) BDfKeVB, Deutsche Gesellschaft für

Gynäkologie und Geburtshilfe (DGGG), Deutsche Gesellschaft für Hygiene und

Mikrobiologie (DGHM) DGfPID, et al. Prophylaxe der Neugeborenensepsis - frühe

Form - durch Streptokokken der Gruppe B: AWMF, 2016.

33. Società Italiana di Medicina Perinatale. Proposta: lineeguida per la prevenzione delle

infezioni perinatali da streptococcoβ-emolitico di gruppo B. Boll Soc Ital Med

Perinatale 1996;1:21–24.

34. Berardi A, Lugli L, Baronciani D, et al. Group B streptococcal infections in a northern

region of Italy. Pediatrics 2007;120(3):e487-93. doi:

http://dx.doi.org/10.1542/peds.2006-3246

35. Sociedad Espanola de Ginecologıa y Obstetricia (SEGO), Sociedad Espanola de

Neonatologıa (SEN), Sociedad Espanola de Enfermedades Infecciosas y Microbiologıa

Clı´nica (SEIMC), et al. Prevention of perinatal group B streptococcal infections:

revised Spanish recommendations 2012. Prog Obstet Ginecol 2012;55(7):337—46.

36. Kind C. Betreuung der Neugeborenen von Müttern, die mit Streptokokken der Gruppe B

kolonisiert sind. Swtizerland: Swiss Society of Neonatology, Undated.

37. Berardi A, Lugli L, Baronciani D, et al. Group B Streptococcus early-onset disease in

Emilia-romagna: review after introduction of a screening-based approach. Pediatr

Infect Dis J 2010;29(2):115-21. doi: 10.1097/INF.0b013e3181b83cd9 [published

Online First: 2009/11/17]

38. Lopez Sastre JB, Fernandez Colomer B, Coto Cotallo GD, et al. Trends in the

epidemiology of neonatal sepsis of vertical transmission in the era of group B

streptococcal prevention. Acta paediatrica (Oslo, Norway : 1992) 2005;94(4):451-7.


Page 36 of 49


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


18

39. Alhhazmi A, Hurteau D, Tyrrell GJ. Epidemiology of Invasive Group B Streptococcal

Disease in Alberta, Canada, from 2003 to 2013. Journal of Clinical Microbiology

2016;54(7):1774-81. doi: 10.1128/jcm.00355-16

40. Hisanori M, Hiramatsu Y, Mitsuhiko K, et al. Guidelines for obstetrical practce in Japan:

Japan Society of Obstetrics and Gynecology, 2011.

41. Mechurova A. Group B Streptococcus infection, screening, treatment -

Recommendation. [Czech] Infekce streptokoky skupiny B, screening, lecba v

tehotenstvi - Doporucene postupy. Prakticky Lekar 2006;86(2):92-94.

42. Wilson JMG, Jungner G. Principles and practice of screening for disease. Geneva,: World

Health Organization 1968.

43. Seedat F, Taylor-Phillips S, Geppert J, et al. Universal antenatal culture-based screening

for maternal Group B Streptococcus (GBS) carriage to prevent early-onset GBS

disease United Kingdom: UK National Screening Committee, Public Health England;

2016 [Available from: https://legacyscreening.phe.org.uk/groupbstreptococcus.

44. Bazian Ltd. Screening for Group B Streptococcal infection in pregnancy: External review

against programme appraisal criteria for the UK National Screening Committee (UK

NSC): UK National Screening Committee; 2012 [Available from:

https://legacyscreening.phe.org.uk/groupbstreptococcus.

45. Russell NJ, Seale AC, O'Driscoll M, et al. Maternal Colonization With Group B

Streptococcus and Serotype Distribution Worldwide: Systematic Review and Meta-

analyses. Clinical infectious diseases : an official publication of the Infectious Diseases

Society of America 2017;65(suppl_2):S100-s11. doi: 10.1093/cid/cix658 [published


46. Jones N, Oliver K, Jones Y, et al. Carriage of group B streptococcus in pregnant women

from Oxford, UK. J Clin Pathol 2006;59(4):363–66.

47. UK National Screening Committee. Early onset streptococcal (EOGBS) disease: a report

of a modelling exercise: UK NSC, 2016.

48. Gopal Rao G, Nartey G, McAree T, et al. Outcome of a screening programme for the

prevention of neonatal invasive early-onset group B Streptococcus infection in a UK

maternity unit: an observational study. BMJ Open 2017;7(4):e014634. doi:

10.1136/bmjopen-2016-014634

Page 37 of 49


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


19

49. Valkenburg-van den Berg AW, Houtman-Roelofsen RL, Oostvogel PM, et al. Timing of

group B streptococcus screening in pregnancy: a systematic review. Gynecologic and

obstetric investigation 2010;69(3):174-83. doi: 10.1159/000265942 [published


50. Colbourn T, Gilbert R. An overview of the natural history of early onset group B

streptococcal disease in the UK Early Hum Dev 2007;83:149–56.

51. Russell NJ, Seale AC, O'Sullivan C, et al. Risk of Early-Onset Neonatal Group B

Streptococcal Disease With Maternal Colonization Worldwide: Systematic Review

and Meta-analyses. Clinical infectious diseases : an official publication of the

Infectious Diseases Society of America 2017;65(suppl_2):S152-s59. doi:

10.1093/cid/cix655 [published Online First: 2017/11/09]

52. Seedat F, Brown C, Stinton C, et al. Bacterial load and molecular markers associated with

early-onset Group B Streptococcus. A systematic review and meta-analysis. Pediatr

Infect Dis J In press,

53. Madrid L, Seale AC, Kohli-Lynch M, et al. Infant Group B Streptococcal Disease Incidence

and Serotypes Worldwide: Systematic Review and Meta-analyses. Clinical infectious

diseases : an official publication of the Infectious Diseases Society of America

2017;65(suppl_2):S160-s72. doi: 10.1093/cid/cix656 [published Online First:

2017/11/09]

54. Manktelow B, on behalf of the MBRRACE-UK collaboration. Preliminary data on deaths

with Group B Streptococcal as a reported cause of death as reported to MBRRACE-

UK for births in 2014.: MBRRACE-UK, 2016.

55. Eastwood KA, Craig S, Sidhu H, et al. Prevention of early-onset Group B Streptococcal

disease - The Northern Ireland experience. BJOG: An International Journal of

Obstetrics and Gynaecology 2015;122(3):361-67. doi:

http://dx.doi.org/10.1111/1471-0528.12841

56. Matsubara K, Hoshina K, Kondo M, et al. Group B streptococcal disease in infants in the

first year of life: a nationwide surveillance study in Japan, 2011-2015. Infection

2017;45(4):449-58. doi: 10.1007/s15010-017-0995-2 [published Online First:

2017/02/27]

57. Matsubara K, Hoshina K, Suzuki Y. Early-onset and late-onset group B streptococcal

disease in Japan: A nationwide surveillance study, 2004-2010. International Journal

Page 38 of 49


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


20

of Infectious Diseases 2013;17(6):e379-e84. doi:

http://dx.doi.org/10.1016/j.ijid.2012.11.027

58. Bedford H, de Louvois J, Halket S, et al. Meningitis in infancy in England and Wales:

follow up at age 5 years. Bmj 2001;323(7312):533. doi: 10.1136/bmj.323.7312.533

59. Kohli-Lynch M, Russell NJ, Seale AC, et al. Neurodevelopmental Impairment in Children

After Group B Streptococcal Disease Worldwide: Systematic Review and Meta-

analyses. Clinical infectious diseases : an official publication of the Infectious Diseases

Society of America 2017;65(suppl_2):S190-s99. doi: 10.1093/cid/cix663 [published


60. Colbourn TE, Asseburg C, Bojke L, et al. Preventive strategies for group B streptococcal

and other bacterial infections in early infancy: cost effectiveness and value of

information analyses. Bmj 2007;335(7621):655. doi: 10.1136/bmj.39325.681806.AD


61. Weston EJ, Pondo T, Lewis MM, et al. The burden of invasive early-onset neonatal sepsis

in the United States, 2005-2008. The Pediatric infectious disease journal

2011;30(11):937-41. doi: 10.1097/INF.0b013e318223bad2 [published Online First:

2011/06/10]

62. Heath PT, Balfour GF, Tighe H, et al. Group B streptococcal disease in infants: a case

control study. Archives of disease in childhood 2009;94(9):674-80. doi:

10.1136/adc.2008.148874 [published Online First: 2009/05/22]

63. Public Health England. UK Standards for Microbiology Investigations: Detection of

Carriage of Group B Streptococci. London: PHE, 2015.

64. Office for National Statistics. Dataset: Vital Statistics: Population and Health Reference

Tables United kingdom: ONS; 2016 [Available from:

http://www.ons.gov.uk/peoplepopulationandcommunity/populationandmigration/p

opulationestimates/datasets/vitalstatisticspopulationandhealthreferencetables

accessed 29 April 2018.

65. Hospital Episodes Statistics. NHS Maternity Statistics - England, 2004-2005 England: NHS

Digital; 2006 [Available from: https://digital.nhs.uk/catalogue/PUB01674 accessed

23 March 2018.

Page 39 of 49


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


21

66. Hospital Episodes Statistics. NHS Maternity Statistics - England, 2014-2015 England: NHS

Digital; 2015 [Available from: https://digital.nhs.uk/catalogue/PUB19127 accessed

23 March 2018.

67. Easmon CS, Hastings MJ, Neill J, et al. Is group B streptococcal screening during

pregnancy justified? Br J Obstet Gynaecol 1985;92(3):197-201.

68. Valkenburg-van den Berg AW, Sprij AJ, Oostvogel PM, et al. Prevalence of colonisation

with group B Streptococci in pregnant women of a multi-ethnic population in The

Netherlands. Eur J Obstet Gynecol Reprod Biol 2006;124(2):178-83. doi:

10.1016/j.ejogrb.2005.06.007

69. Scasso S, Laufer J, Rodriguez G, et al. Vaginal group B streptococcus status during

intrapartum antibiotic prophylaxis. International Journal of Gynecology and

Obstetrics 2015;129(1):9-12. doi: http://dx.doi.org/10.1016/j.ijgo.2014.10.018

70. Szymusik I, Kosinska-Kaczynska K, Krolik A, et al. The usefulness of the universal culture-

based screening and the efficacy of intrapartum prophylaxis of group B

Streptococcus infection. Journal of Maternal-Fetal and Neonatal Medicine

2014;27(9):968-70. doi: http://dx.doi.org/10.3109/14767058.2013.845659

71. MacKay G, House MD, Bloch E, et al. A GBS culture collected shortly after GBS

prophylaxis may be inaccurate. Journal of Maternal-Fetal and Neonatal Medicine

2012;25(6):736-38. doi: http://dx.doi.org/10.3109/14767058.2011.596961

72. Kunze M, Zumstein K, Markfeld-Erol F, et al. Comparison of pre- and intrapartum

screening of group B streptococci and adherence to screening guidelines: a cohort

study. European Journal of Pediatrics 2015;174(6):827-35. doi:

http://dx.doi.org/10.1007/s00431-015-2548-y

73. Picard FJ, Bergeron MG. Laboratory detection of group B Streptococcus for prevention

of perinatal disease. Eur J Clin Microbiol Infect Dis 2004;23(9):665-71. doi:

10.1007/s10096-004-1183-8 [published Online First: 2004/07/20]

74. Daniels J, Gray J, Pattison H, et al. Rapid testing for group B streptococcus during labour:

a test accuracy study with evaluation of acceptability and cost-effectiveness. Health

technology assessment (Winchester, England), 2009.

75. Honest H, Sharma S, Khan KS. Rapid tests for group B Streptococcus colonization in

laboring women: a systematic review. Pediatrics 2006;117(4):1055-66. doi:

10.1542/peds.2005-1114 [published Online First: 2006/04/06]

Page 40 of 49


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

22

76. Ohlsson A, Shah Vibhuti S. Intrapartum antibiotics for known maternal Group B

streptococcal colonization. Cochrane Database of Systematic Reviews 2014(6) doi:

10.1002/14651858.CD007467.pub4

77. Colbourn T, Asseburg C, Bojke L, et al. Prenatal screening and treatment strategies to

prevent group B streptococcal and other bacterial infections in early infancy: cost-

effectiveness and expected value of information analyses. Health technology

assessment (Winchester, England) 2007;11(29):1-226, iii. [published Online First:

2007/07/27]

78. Raffle A, Gray M. Screening: Evidence and Practice. Oxford: Oxford University Press 2007

79. Kurz E, Davis D. Routine culture-based screening versus risk-based management for the

prevention of early-onset group B streptococcus disease in the neonate: a systematic

review. JBI database of systematic reviews and implementation reports

2015;13(3):206-46. doi: 10.11124/jbisrir-2015-1876 [published Online First:

2015/10/09]

80. Bauserman MS, Laughon MM, Hornik CP, et al. Group B Streptococcus and Escherichia

coli infections in the intensive care nursery in the era of intrapartum antibiotic

prophylaxis. Pediatr Infect Dis J 2013;32(3):208-12. doi:

http://dx.doi.org/10.1097/INF.0b013e318275058a

81. Ecker KL, Donohue PK, Kim KS, et al. The impact of group B Streptococcus prophylaxis on

early onset neonatal infections. Journal of Neonatal-Perinatal Medicine

2013;6(1):37-44. doi: http://dx.doi.org/10.3233/NPM-1363312

82. Gopal Rao G, Townsend J, Stevenson D, et al. Early-onset group B

Streptococcus (EOGBS) infection subsequent to cessation of screening-

based intrapartum prophylaxis: findings of an observational study in West London,

UK. BMJ Open 2017;7(11) doi: 10.1136/bmjopen-2017-018795

83. Lukacs SL, Schrag SJ. Clinical sepsis in neonates and young infants, United States, 1988-

2006. J Pediatr 2012;160(6):960-5 e1. doi: 10.1016/j.jpeds.2011.12.023 [published


84. Schrag SJ, Farley MM, Petit S, et al. Epidemiology of Invasive Early-Onset Neonatal

Sepsis, 2005 to 2014. Pediatrics 2016;138(6) doi: 10.1542/peds.2016-2013


Page 41 of 49


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


23

85. Horvath B, Grasselly M, Bodecs T, et al. Screening pregnant women for group B

streptococcus infection between 30 and 32 weeks of pregnancy in a population at

high risk for premature birth. International Journal of Gynecology and Obstetrics

2013;122(1):9-12. doi: http://dx.doi.org/10.1016/j.ijgo.2013.01.027

86. Angstetra D, Ferguson J, Giles WB. Institution of universal screening for Group B

streptococcus (GBS) from a risk management protocol results in reduction of early-

onset GBS disease in a tertiary obstetric unit. The Australian & New Zealand journal

of obstetrics & gynaecology 2007;47(5):378-82. doi: 10.1111/j.1479-

828X.2007.00760.x [published Online First: 2007/09/20]

87. Seedat F, Stinton C, Patterson J, et al. Adverse events in women and children who have

received intrapartum antibiotic prophylaxis treatment: a systematic review. BMC

Pregnancy and Childbirth 2017;17 doi: 10.1186/s12884-017-1432-3

88. Aloisio I, Mazzola G, Corvaglia LT, et al. Influence of intrapartum antibiotic prophylaxis

against group B Streptococcus on the early newborn gut composition and evaluation

of the anti-Streptococcus activity of Bifidobacterium strains. Applied Microbiology

and Biotechnology 2014;98(13):6051-60.

89. Aloisio I, Quagliariello A, De Fanti S, et al. Evaluation of the effects of intrapartum

antibiotic prophylaxis on newborn intestinal microbiota using a sequencing approach

targeted to multi hypervariable 16S rDNA regions. Applied Microbiology and

Biotechnology 2016;100(12):5537-46. doi: http://dx.doi.org/10.1007/s00253-016-

7410-2

90. Arboleya S, Sanchez B, Milani C, et al. Intestinal Microbiota Development in Preterm

Neonates and Effect of Perinatal Antibiotics. J Pediatr 2015;166(3):538-44.

91. Arboleya S, Sanchez B, Solis G, et al. Impact of Prematurity and Perinatal Antibiotics on

the Developing Intestinal Microbiota: A Functional Inference Study. Int 2016;17(5)

doi: http://dx.doi.org/10.3390/ijms17050649

92. Corvaglia L, Tonti G, Martini S, et al. Influence of Intrapartum Antibiotic Prophylaxis for

Group B Streptococcus on Gut Microbiota in the First Month of Life. J Pediatr

Gastroenterol Nutr 2016;62(2):304-08.

93. Jaureguy F, Carton M, Panel P, et al. Effects of intrapartum penicillin prophylaxis on

intestinal bacterial colonization in infants. Journal of Clinical Microbiology

2004;42(11):5184-88.

Page 42 of 49


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


24

94. Mazzola G, Murphy K, Ross RP, et al. Early gut microbiota perturbations following

intrapartum antibiotic prophylaxis to prevent group B streptococcal disease. PLoS

ONE 2016;11 (6) (no pagination)(e0157527) doi:

http://dx.doi.org/10.1371/journal.pone.0157527

95. Azad MB, Konya T, Persaud RR, et al. Impact of maternal intrapartum antibiotics,

method of birth and breastfeeding on gut microbiota during the first year of life: a

prospective cohort study. Bjog 2015 doi: 10.1111/1471-0528.13601

96. Keski-Nisula L, Kyynarainen HR, Karkkainen U, et al. Maternal intrapartum antibiotics

and decreased vertical transmission of Lactobacillus to neonates during birth. Acta

Paediatr 2013;102(5):480-85.

97. Cox LM, Yamanishi S, Sohn J, et al. Altering the intestinal microbiota during a critical

developmental window has lasting metabolic consequences. Cell 2014;158:705–21.

98. Francino MP. Antibiotics and the Human Gut Microbiome: Dysbioses and Accumulation

of Resistances. Frontiers in Microbiology 2015;6 doi: 10.3389/fmicb.2015.01543

99. Saari A, Virta LJ, Sankilampi U, et al. Antibiotic exposure in infancy and risk of being

overweight in the first 24 months of life. Pediatrics 2015;135(4):617-26. doi:

10.1542/peds.2014-3407

100. Li Q, Han Y, Dy ABC, et al. The Gut Microbiota and Autism Spectrum Disorders.

Frontiers in Cellular Neuroscience 2017;11 doi: 10.3389/fncel.2017.00120

101. Ashkenazi-Hoffnung L, Melamed N, Ben-Haroush A, et al. The Association of

Intrapartum Antibiotic Exposure With the Incidence and Antibiotic Resistance of

Infantile Late-Onset Serious Bacterial Infections. Clin Pediatr 2011;50(9):827-33.

102. Glasgow TS, Young PC, Wallin J, et al. Association of intrapartum antibiotic exposure

and late-onset serious bacterial infections in infants. Pediatrics 2005;116(3):696-702.

103. Gordon M, Samuels P, Shubert P, et al. A randomized, prospective study of adjunctive

ceftizoxime in preterm labor. American journal of obstetrics and gynecology

1995;172(5):1546-52. [published Online First: 1995/05/01]

104. Roca A, Oluwalana C, Bojang A, et al. Oral azithromycin given during labour decreases

bacterial carriage in the mothers and their offspring: a double-blind randomized

trial. Clin Microbiol Infect 2016;22(6) doi: 10.1016/j.cmi.2016.03.005

Page 43 of 49


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


25

105. Stoll BJ, Hansen N, Fanaroff AA, et al. Changes in pathogens causing early-onset sepsis

in very-low-birth-weight infants. New England Journal of Medicine 2002;347(4):240-

47. doi: http://dx.doi.org/10.1056/NEJMoa012657

106. Huang J, Li S, Li L, et al. Alarming regional differences in prevalence and antimicrobial

susceptibility of group B streptococci in pregnant women: A systematic review and

meta-analysis. Journal of global antimicrobial resistance 2016;7:169-77. doi:

10.1016/j.jgar.2016.08.010 [published Online First: 2016/11/13]

107. Phares CR, Lynfield R, Farley MM, et al. Epidemiology of invasive group B streptococcal

disease in the United States, 1999-2005. Jama 2008;299(17):2056-65. doi:

10.1001/jama.299.17.2056 [published Online First: 2008/05/08]

108. Kenyon S, Pike K, Jones DR, et al. Childhood outcomes after prescription of antibiotics

to pregnant women with spontaneous preterm labour: 7-year follow-up of the

ORACLE II trial. Lancet 2008;372(9646):1319-27. doi:

http://dx.doi.org/10.1016/S0140-6736(08)61203-9

109. Kenyon S, Hagberg H, Norman JE. Preterm Labour, Antibiotics, and Cerebral Palsy.

United Kingdom: RCOG, 2013.

110. Kenyon S, Pike K, Jones DR, et al. Childhood outcomes after prescription of antibiotics

to pregnant women with preterm rupture of the membranes: 7-year follow-up of

the ORACLE I trial. The Lancet 2008;372(9646):1310-18. doi: 10.1016/S0140-

6736(08)61202-7

111. Mulla ZD, Ebrahim MS, Gonzalez JL. Anaphylaxis in the obstetric patient: analysis of a

statewide hospital discharge database. Annals of allergy, asthma & immunology :

official publication of the American College of Allergy, Asthma, & Immunology

2010;104(1):55-9. doi: 10.1016/j.anai.2009.11.005 [published Online First:

2010/02/11]

112. O’Connor M, Nair M, Kurinczuk J, et al. UKOSS Annual Report 2016. Oxford: National

Perinatal Epidemiology Unit, 2016.

113. McCall SJ, Bunch KJ, Brocklehurst P, et al. The incidence, characteristics, management

and outcomes of anaphylaxis in pregnancy: a population-based descriptive study.

Bjog 2017 doi: 10.1111/1471-0528.15041 [published Online First: 2017/12/02]

Page 44 of 49


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


26

114. Lin FY, Troendle JF. Hypothesis: Neonatal respiratory distress may be related to

asymptomatic colonization with group B streptococci. Pediatr Infect Dis J

2006;25(10):884-8.

115. Dinsmoor MJ, Viloria R, Lief L, et al. Use of intrapartum antibiotics and the incidence of

postnatal maternal and neonatal yeast infections. Obstetrics & Gynecology

2005;106(1):19-22.

116. Wohl DL, Curry WJ, Mauger D, et al. Intrapartum antibiotics and childhood atopic

dermatitis. J Am Board Fam Med 2015;28(1):82-9. doi:

http://dx.doi.org/10.3122/jabfm.2015.01.140017

117. Department of Health. UK Five Year Antimicrobial Resistance Strategy 2013 to 2018,

2013.

Page 45 of 49


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


27

Conflict of interest declarations:

This research was commissioned by the UK National Screening Committee (NSC). Sian

Taylor-Phillips, Chris Stinton, Hannah Fraser and Aileen Clarke are supported by the National

Institute for Health Research CLAHRC West Midlands initiative. Anne Mackie is the Director

of the UK NSC, John Marshall is the Evidence Lead and Cristina Visintin is an Evidence

Review Manager for the UK NSC. The views expressed in this publication are those of the

authors and not necessarily those of the NHS, the National Institute for Health Research,

Public Health England, or the Department of Health. Any errors are the responsibility of the

authors.

Data sources and contributors:

The sources of information used to prepare this manuscript are from the NSC policy reviews

of 2012 and 2016, in addition to the GBS model that was developed by the UK NSC, and

studies on GBS epidemiology and screening published after the 2016 NSC review.

This piece of research and the completion of this manuscript involved a multi-disciplinary

team of information specialists, epidemiologists, infectious disease, microbiology, and

obstetrics and gynaecology consultants, screening and public health specialists, statisticians,

and reviewers. AM, the Director of the UK NSC, JM, the Evidence Lead for the UK NSC, and

CV, an Evidence Review Manager contributed to the writing of this manuscript but did not

conduct any of the review processes or the synthesis and interpretation of the original

reviews. The research team below conducted the 2016 NSC review for GBS.

FS has completed a PhD specialising in GBS screening and has previously conducted

systematic reviews, including NSC reviews; FS secured funding, co-ordinated the review

process, developed the protocol, created and applied the search strategy to collect the data,

sifted, extracted, and quality assessed 20% of the articles, and synthesised the data for the

2016 review. FS also combined the evidence from the 2012 and 2016 NSC evidence reviews

selecting the best available evidence for the purpose of this article and led the writing of

Page 46 of 49


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


28

this manuscript. JG is an expert systematic reviewer specialising in screening and test

accuracy and has previously conducted NSC reviews; JG carried out data sifting, extraction,

quality assessment, and synthesis for all of the data for the 2016 review, reviewed the

merging of data between the 2012 and 2016 evidence reviews, and reviewed the

manuscript for redrafting. OU, CS and KF are also expert systematic reviewers who have

conducted previous reviews for the NSC and health technology assessments for NICE; they

contributed to protocol development for the 2016 review and reviewed this manuscript for

redrafting. JP is a medical doctor with expertise in evidence-based medicine and systematic

reviews who has conducted NSC reviews; JP reviewed this manuscript for redrafting. NM is

an academic public health physician and epidemiologist with expertise in infectious disease

control, ER is the lead public health microbiologist for East Midlands Pubic Health England,

and CB is a consultant in infectious diseases and medical microbiology at Public Health

England; they contributed to protocol development for the 2016 review and reviewed this

manuscript for redrafting, providing expertise on infection and microbiology. BT is a clinician

scientist, consultant obstetrician & gynaecologist and RCOG accredited subspecialist in

reproductive medicine who has managed numerous patients with GBS in pregnancy; BT

contributed to protocol development for the 2016 review and reviewed this manuscript for

redrafting, providing obstetrics and gynaecology expertise. SJ is an academic support

librarian and HF has studied the Masters in Screening course; they contributed to protocol

development, search strategy development, and data collection of the 2016 review, and

reviewed this manuscript for redrafting. AC is a clinical public health academic who heads

the Division of Health Sciences at the Warwick Medical School and leads one of nine

technology assessment review teams providing systematic reviews to NICE; AC contributed

to protocol development for the 2016 review, and reviewed this manuscript for redrafting.

STP is an associate professor of screening and test evaluation with wide experience in

systematic reviews specialising in screening, including NSC reviews; STP secured the

funding, co-ordinated the review process and developed the protocol for the 2016 review,

and reviewed this manuscript for redrafting.

Acknowledgements

Page 47 of 49


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


29

We would like to thank Dr Michael Millar, Dr Arlene Reynolds, and Dr Magdalena Skrybant

for providing advice and input into this research. We would also like to thank Nick

Johnstone-Waddell for providing graphical input for the report.

Licence

“The Corresponding Author has the right to grant on behalf of all authors and does grant on

behalf of all authors, an exclusive licence (or non exclusive for government employees) on a

worldwide basis to the BMJ Publishing Group Ltd ("BMJ"), and its Licensees to permit this

article (if accepted) to be published in The BMJ's editions and any other BMJ products and

to exploit all subsidiary rights, as set out in our licence.”

Page 48 of 49


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


631,512 term pregnant women in UK

138,933 GBS positive 492,579 GBS negative

some will change GBS status during pregnancy

350 term neonates with EOGBS

deaths10

treated and recover286-310

long term disability30-54

573,448 term pregnant women in UKYear 2000

126,159 GBS positive 447,289 GBS negative

some will change GBS status during pregnancy

205 term neonates with EOGBS

deaths13

treated and recover162-175

long term disability17-30

represents 4,000 women

represents 50 neonates

represents 4,000 women

represents 50 neonates

Year 2014

Page 49 of 49


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

confidential: for review only - bmj · 2019-02-27 · confidential: for review only 2 steady in...

Documents