One of the chief preventable causes of neonatal mortality is group B streptococcal (GBS) infection. The present Centers for Disease Control and Prevention (CDC) guideline for prevention of neonatal GBS disease recommends universal prenatal screening at 35 to 37 weeks' gestation or a risk factor–based approach. Epidemiologic research cited in the guideline suggested that universal screening would likely be more effective than a risk-based strategy, but there were no confirmatory clinical studies yet available to show this. The Active Bacterial Core Surveillance Team from the CDC examined data on 629,912 births from eight different regions of the United States.

From this number, 5,144 births were randomly sampled, and 312 cases of early-onset GBS disease were revealed. The investigators then randomly selected another 4,831 births from this population-based sample for control analysis, and these were stratified by region, year of delivery, and hospital. Labor and delivery records were reviewed for GBS screening and the presence of any clinical factors that should have prompted intrapartum antibiotic use in a risk-based approach (i.e., preterm labor at less than 37 weeks, rupture of membranes at least 18 hours before delivery, intrapartum maternal fever, history of prenatal GBS bacteriuria, or a previous infant with GBS disease). Of the charts selected for review, 95 percent were successfully examined.

In the 312 cases of GBS disease that were identified, 62 percent did not have any clinical risk factors for infection. Prenatal screening for GBS varied from 24 to 70 percent, with 52 percent of the women being screened. A risk-based approach for GBS prophylaxis was more commonly employed in California and Oregon, and among Hispanic mothers.

The relative risk of neonatal GBS disease was reduced by more than one half in the group that received prenatal screening, compared with the risk-based management group (relative risk, 0.46, with a 95 percent confidence interval of 0.36 to 0.60).

In this study, intrapartum fever, maternal fever, and a previous history of an infant with GBS disease were the most potent risk factors for development of neonatal GBS infection. Of the women who were screened for GBS prenatally, 18 percent had positive cultures, and 89 percent received appropriate antibiotic prophylaxis intrapartum. In screened women with positive GBS cultures but no development of risk factors during delivery, the protective effect of prophylactic antibiotics was still robust, with 88.6 percent of predicted GBS cases prevented.

The authors concluded that prenatal screening for GBS, compared with a risk-based approach, reduced neonatal infection by more than one half.

editor's note: In an editorial accompanying this study, Eschenbach notes that at least among very-low-birth-weight infants in a recent large study, the reduction in GBS disease that occurs with intrapartum antibiotic prophylaxis was replaced by a concomitant increase in early-onset neonatal sepsis caused by Escherichia coli. He advocates immunization of pregnant women against GBS to avoid the inevitable dilemma of emerging resistance with any increase in antibiotic use.—b.z.