Group B streptococcus

This issue includes an article by Hassan et al. (2011) on the prevalence of vaginal carriage during pregnancy of haemolytic streptococci, and reports that 19 of 100 had group B streptococci (GBS) isolated, 14 from low vaginal swabs, 11 from rectal swabs and 2 from throat swabs. The swabs were taken by midwives or the patient and inoculated on to selective blood agar and into a selective broth.

GBS have been reported in between 20–28% of pregnant women (Yow et al. 1980; Easmon 1984), the variation depending on the number of sites sampled, the number of times during pregnancy these sites were sampled, and the use of sensitive broth-enrichment techniques. The prevalence seems greatest in young and nulliparous patients, those who are socioeconomically deprived and varies geographically and internationally, being one-tenth less in the UK than in the USA. Approximately one-third to one-half of neonates born to mothers with GBS will themselves become colonised, but only 1–2% of these neonates will become infected with early onset features in the first week of life. The reported risks vary between 0.3–3.7/1,000 liveborn. The risk of infection is greater if the infant is born pre-term or after a prolonged labour/prolonged rupture of the membranes (MacLean and Cockburn 1995).

A study conducted in Glasgow in 1991–1992 identified 136 (14%) positive isolation from low vaginal swabs taken from 1,000 consecutive women admitted in labour. It is acknowledged that positive isolation will be less in labour compared with sampling at various sites and gestations during pregnancy. Our findings were contrary to what is described by others, i.e. carriage rate was 10.5% in women less than 30 years of age cf. 18.5% in those over 30. Carriage was 11% in nulliparous patients, 15% in para 1, 19% in para 2 and was 13.6% in those presenting in labour after 37 weeks but 16.5% if before 37 weeks. Among patients coming from a socioeconomically deprived post-code, the carriage rate was 9.5%, and among those from more affluent areas 22% (MacLean, Lucas, Carrington, unreported data). Thus, if it is not possible to predict with any certainty who is likely to be carrying GBS during their pregnancy, should all women be screened?

This question continues to be debated and has been raised again by Hassan and co-authors. The Royal College of Obstetricians and Gynaecologists advises against screening all pregnant women, whereas Group B Strep. Support, the charity that promotes awareness among pregnant women and healthcare providers, urges reconsideration. No-one has convincingly described why carriage varies from one side of the Atlantic to the other, and why only a small fraction of those neonates colonised develop life-threatening infection. As the UK becomes increasingly international we can no longer assume that the risks from GBS are the same as that reported 25 years ago (Easmon et al. 1985). This infection would make an ideal theme for a well-designed multicentre study that conducted screening for these organisms within the vagina and rectum at two occasions during the 3rd trimester, with cost analysis, to determine whether early-onset GBS disease could be prevented. If the risk remains at 0.3/1,000 livebirths, 3,000 pregnant women would need screening and between 600 and 1,000 treating to prevent one neonatal case. Such a research project would require appropriate funding and administration but would clarify the clinical problem before the inevitable progression to offering screening or treatment during labour for the individual patient because the answers remain unknown.