OBJECTIVES: To estimate the prevalence of group B streptococcus at onset of labor and to compare the accuracy of intrapartum antibiotic prophylaxis based on a risk factor strategy versus an intrapartum screening.

STUDY DESIGN: In this cross-sectional study, 700 women referred for intended vaginal delivery were tested for group B streptococcus at onset of labor by gold standard culture in vaginal and rectal swabs. The results were blinded, and intrapartum antibiotic prophylaxis was given to women with risk factors for group B streptococcus infection: previous neonatal group B streptococcus infection, group B streptococcus in urine and/or vagina in current pregnancy, gestational age <36 + 0 weeks, temperature ≥38.0 °C or rupture of membranes ≥18 h.

RESULTS: Of 642 women, who delivered within three days of specimen collection, 170 (26.5%) received benzylpenicillin as intrapartum antibiotic prophylaxis based on risk factors. The prevalence of group B streptococcus at onset of labor was 17.8% (114/642), with 62 women positive for group B streptococcus in both the vagina and rectum, 12 in the vagina only and 40 in rectum only. Intrapartum antibiotic prophylaxis was administered correctly in 31.6% (36/114) of GBS positive women at time of delivery and 78.8% (134/170) of women who received antibiotics were actually GBS negative. The positive predictive value of intrapartum antibiotic prophylaxis based on risk factors was 21.2%. The sensitivity was 31.6% and the specificity was 74.6%.

CONCLUSION: The accuracy of predicting maternal group B streptococcus infection from risk factors is low. Intrapartum group B streptococcus diagnostics is more precise and will not increase the overall use of intrapartum antibiotic prophylaxis.