Background
Early‐onset group B streptococcal (EOGBS) disease (including sepsis, meningitis, and pneumonia) causes significant morbidity and mortality in newborn infants worldwide. Antibiotic prophylaxis can prevent vertical streptococcal transmission, yet no uniform criteria exist to identify eligible women for prophylaxis. Some guidelines recommend universal GBS screening to pregnant women in their third trimester (screening‐based protocol), whereas others employ risk‐based protocols.
Objectives
To compare the effectiveness of screening‐based versus risk‐based protocols in preventing EOGBS disease.
Search strategy
Key words for the database searches included GBS, Streptococcus agalactiae, pregnancy, screening, culture‐based, risk‐based.
Selection criteria
Studies were included if they investigated EOGBS disease incidence in newborn infants and compared screening or risk‐based protocols with each other or with controls.
Data collection and analysis
Risk ratios (RR) and 95% confidence intervals (CI) were determined using Mantel‐Haenszel analyses with random effects.
Main results
Seventeen eligible studies were included. In this meta‐analysis, screening was associated with a reduced risk for EOGBS disease compared either with risk‐based protocols (ten studies, RR 0.43, 95% CI 0.32–0.56) or with no policy (four studies, RR 0.31, 95% CI 0.11–0.84). Meta‐analysis could not demonstrate a significant effect of risk‐based protocols versus no policy (seven studies, RR 0.86, 95% CI 0.61–1.20). In studies reporting on the use of antibiotics, screening was not associated with higher antibiotic administration rates (31 versus 29%).
Conclusions
Screening‐based protocols were associated with lower incidences of EOGBS disease compared with risk‐based protocols, while not clearly overexposing women to antibiotics. This information is of relevance for future policymaking.
Tweetable abstract
Meta‐analysis: general screening is associated with lower rates of early‐onset group B strep. neonatal sepsis compared with risk‐based protocols.