Group B streptococcus: to culture or not to culture?
OBJECTIVE: To determine if universal Group B Streptococcus (GBS) culturing and antibiotic prophylaxis of obstetric patients decreased the incidence of neonatal early-onset GBS sepsis and mortality and maternal chorioamnionitis. STUDY DESIGN: A time series observational study was conducted to compare the cohort of all obstetric patients delivering at the University of Chicago neonatal center from January 1989 through December 1993, before a GBS surveillance policy existed, with the cohort delivering January 1994 through December 1996, after initiation of a GBS policy. Included in the policy were universal GBS cultures at 28 weeks' gestation, antibiotic prophylaxis at the time of labor for all those with positive cultures and for all with risk factors of preterm delivery, preterm premature rupture of membranes, prolonged rupture of membranes greater than 18 hours, and a previous child affected by GBS or maternal fever in labor. Predictor variables were GBS culturing and antibiotic usage; outcome variables were incidence of GBS sepsis and mortality in the neonates and maternal chorioamnionitis. chi-squared and Fisher exact analyses were used with p < 0.05 being significant. RESULTS: Before the GBS policy, there were 16,272 deliveries with a 2.24/1000 deliveries rate of early-onset GBS sepsis (n = 35); after initiating the GBS policy, 9130 deliveries occurred with an early-onset GBS sepsis rate of 2.29/1000 (n = 20). Early-onset GBS sepsis case fatality rates before and after initiation of the policy were 14.3% and 0%, respectively (p = 0.09). Antibiotic use almost doubled (relative risk = 1.84; confidence interval, 1.74 to 1.93, p < 0.001) over the two time periods, and the relative risk of chorioamnionitis decreased to 0.95 (confidence interval, 0.73 to 0.99, p = 0.04). CONCLUSION: Despite universal GBS culturing and very liberal use of antibiotics in labor, we were unable to effect a statistically significant change in the rate of early-onset GBS sepsis or mortality, and there was only a slightly decreased chorioamnionitis rate.