Vertically transmitted, early-onset infection of newborns with group B streptococci (GBS) occurs in 0.4 to 3 per 1,000 live births in populations in the United States and Canada. In infants younger than one week, GBS infection can result in sepsis, pneumonia, and meningitis, with mortality rates reported between 4.7 and 9 percent. GBS infection that becomes clinically apparent after one week of age is most likely to present as meningitis and occurs in 0.2 per 1,000 live births, with a mortality rate of 2 to 2.8 percent. The Canadian Task Force on Preventive Health Care used an evidence-based approach to provide specific recommendations for preventing perinatal GBS infection.

Premature labor, prolonged rupture of membranes, maternal fever, and previous delivery of a GBS-infected infant are among the risk factors for perinatal GBS infection. Intrapartum chemoprophylaxis with penicillin, erythromycin, or clindamycin can reduce the rate of GBS colonization in infants by 80 to 90 percent, but without treatment, approximately one-half of the infants born to GBS-colonized mothers become infected. Adequate prophylaxis is defined as at least 5 million U of penicillin intravenously at least four hours before birth, followed by 2.5 million U every four hours until delivery. Alternatives are intravenous clindamycin (900 mg) or erythromycin (500 mg) every eight hours.

Various expert groups, including the Centers for Disease Control and Prevention, the American College of Obstetricians and Gynecologists, and the American Academy of Pediatrics, have published guidelines on preventing perinatal GBS infection. These guidelines principally recommend that health-care providers use one of two strategies: screening all mothers at 35 to 37 weeks’ gestation and treating colonized patients, or selectively treating mothers based on risk factors.

The panel concluded that although no studies have directly compared outcomes in screened and unscreened mothers, there is fair evidence showing that universal screening and treatment of colonized women with risk factors reduces colonization and early-onset infection rates in infants. This is recommended as the most efficient strategy. Treating all colonized mothers and treating only colonized women with risk factors are strategies that reduce neonatal colonization rates, but clinical trials have not demonstrated a significant reduction in neonatal disease with their use. Studies do show a trend toward reduction of neonatal infection, and the authors calculate that six colonized women with risk factors need to be treated to prevent one case of early-onset neonatal infection. Preventing neonatal colonization requires treating two to three women with intrapartum antibiotics by treating either all colonized patients or colonized women who also have risk factors.

Treating all colonized mothers also is given a fair recommendation because this strategy reduces colonization of infants and may protect against infection. However, it requires treating many more women (16 to 2,059, depending on the population studied) to prevent one case of early-onset infection. This finding raises concerns about exposing mothers to adverse effects of treatment and exacerbates concerns about the increasing incidence of antibiotic resistance to penicillin, erythromycin, and clindamycin. The rates of GBS resistance to erythromycin and clindamycin are currently as high as 16 percent.

The panel found no basis for supporting treatment of unscreened mothers based on risk factors alone.