Group B Streptococcus is the most common bacterial cause of sepsis in neonates in the UK, affecting 0.4 per 1,000 live births. Sepsis is associated with preterm labour and prolonged rupture of membranes. Between 20% and 40% of women of child bearing age carry the Group B Streptococcus as part of their normal genital flora, and therefore during a normal vaginal delivery the baby is exposed to this bacterium.
NICE currently recommend that intrapartum antibiotics should be given to:
- Women who have had a previous baby with invasive Group B Streptococcus infection OR who have been shown to be colonised, have bacteriuria or infection with Group B Streptococcus in the current pregnancy
- Women with preterm labour (<37 weeks) and pre-labour rupture of membranes
- Women with preterm labour (<37 weeks) and prolonged rupture of membranes during labour (> 18 hours)
- Women with a temperature >38oC in labour
1st line: IV Benzylpenicillin
2nd line: IV Clindamycin
Duration: until the baby is born
The current guidance for intrapartum antibiotics in the USA is different. In the USA all pregnant women are screened for Group Streptococcus colonisation between 35 and 37 weeks and if positive given IV antibiotics in labour. Antibiotics are also given to women with the same indications as in the NICE guidelines (above).
The ORACLE study (2001) investigated whether antibiotics in labour could reduce neonatal infection. In this double-blind randomised study women in labour were given either Benzylpenicillin, Erythromycin, Co-amoxiclav or placebo, results showed no benefit from giving antibiotics in labour. Regardless of this, screening has become routine practice in the USA and has generated some evidence of a modest benefit.
The main argument put forward to justify the introduction of a screening program in the UK is that it will prevent the deaths of new-born babies at a cost of £12 per mother for screening. A noble idea, however it is important to evaluate the evidence in a pragmatic way before deciding if it is the correct thing to do.
Based on data from the Office for National Statistics UK and the calculations on reduction in Neonatal Group B Streptococcus sepsis from antimicrobial prophylaxis from the Centre for Disease Control (CDC) in the USA, the adoption of the USA screening could prevent 210 cases and 10 deaths from neonatal Group B Streptococcus infection per year from early onset Group B Streptococcus infection in the UK. However there would still be 70 cases per year and 4 deaths.
Based on published costs of screening, antibiotics and hospital care it is estimated that it would cost the NHS £361 million or approximately £36 million pounds per baby saved. Put into the context of the overall funding of the NHS this is a lot of money. A Hospital such as the one in which I work has an annual budget of about £270 million to deliver all aspects of healthcare, so the cost of saving one life equates to 7.5% of the total yearly budget for the hospital. Is this cost effective?
- Comparison of UK with USA: The incidence of neonatal Group B Streptococcus in the UK is the same as the USA AFTER screening and prophylaxis was introduced and therefore there is no assurance that the UK rate can be further reduced by introducing a national screening program.
- Effect on overall mortality: Although there is some evidence that intrapartum antibiotics reduce both the incidence and number of deaths from early onset Group B Streptococcus infection in neonates (<7 days after birth) a full review of the literature (Cochrane review) has shown that antibiotics made no difference to the overall mortality from neonatal Group B Streptococcus infection when late onset infection is included (>7 days after birth). Perinatal mortality in the USA has stayed at about 6 per 1,000 live births over the last 15 years (in the UK it has reduced from 6 to 4 per 1,000 live births during the same time frame).
- When and how to screen? There is no gold standard screening test, in the USA a combination of enrichment culture and PCR are used. PCR can take as little as 2-4 hours for a result but the more sensitive test, enrichment culture takes 48 hours (NB the time period of labour is usually up to 24 hours) Once in labour neither test would provide results fast enough to guide the need for prophylaxis. In the USA screening is performed at 35-37 weeks, but this misses women in preterm labour and women who acquire Group B Streptococcus after this time.
- Cost: The quoted figure for Group B Streptococcus screening in the media is £12 per test but I suspect this is the cost for the test itself and doesn’t factor in the cost of time for midwives to take the samples and laboratory staff to process and read the results. Even if the figure is accurate there is still a substantial cost in screening nearly 700,000 women per year in the UK (£8.4 million pounds annually).
- Unnecessary exposure to antibiotics: If all women in labour in the UK who had had a positive screen for Group B Streptococcus (approximately 20%) were given antibiotics, 140,000 courses of antibiotics would be given to try and prevent 280 cases of neonatal Group B Streptococcus infection i.e. 500 antibiotic courses to prevent 1 case or 10,000 courses to prevent 1 death.
- Antibiotics are not without risk: Part way through the ORACLE study, the investigators noticed that one group was having a much higher rate of necrotising enterocolitis (NEC) in the babies, and that part of the study had to be stopped. It was the Co-amoxiclav group, but why this predisposed to NEC was not determined. Needless to say, Obstetricians are reluctant to give Co-amoxiclav to women close to term unless they absolutely have to.
- Antibiotic resistance: Amoxicillin resistance in E. coli in women in the USA has increased considerably (up to 70%) since the introduction of Amoxicillin for intrapartum prophylaxis against Group B Streptococcus (unpublished data seen by the author at Infection and Immunity in Children 2009).
Any issue concerning childhood mortality is emotive and adds a level of difficulty to the subject of screening for Group B Streptococcus. At present there is not enough evidence to say whether it is definitely a good or bad idea; however this in itself is significant. The fact that since 1996 millions of women in the USA have been screened and offered intrapartum antibiotics without affecting overall mortality suggests that the impact of such a screening program is very small. Added to this the cost of screening, at a time when NHS budgets are constantly shrinking, may not be considered cost effective.
I’m going to sit firmly on the fence on this one and not say whether I think screening is a good idea or not. I’ll let you evaluate the evidence for yourselves…