Objective: Recommendations for the use of antenatal antibiotics in obstetrics have increased in the past few years, especially for prophylaxis against group B streptococci, for prolongation of the latency time in patients with preterm premature rupture of the membranes, and as an adjuvant treatment in preterm labor. Our objective was to determine whether the use of antenatal ampicillin affects the incidence of and resistance of early-onset neonatal sepsis with organisms other than group B streptococci.
Study design: A prospective cohort study was performed between January 1, 1991, and December 31, 1996. Every case of blood culture-proven neonatal sepsis was prospectively surveyed. The type of bacteria isolated, drug resistance, antenatal antibiotic use and treatment indication, gestational age at delivery, and other antenatal and outcome variables were gathered. Early-onset neonatal sepsis was defined as disease onset within 7 days after birth.
Results: A total of 42 cases of early-onset neonatal sepsis among 29,897 neonates delivered were found during the 6-year period. Of these, 15 cases were due to group B streptococci and 27 were the result of non-group B streptococcal organisms (21 gram-negative rods and 6 gram-positive cocci). Among the 27 non-group B streptococcal cases, 15 mothers had received antenatal ampicillin and 13 of the 15 bacterial isolates from these neonates (87%) were resistant to ampicillin, versus only 2 ampicillin-resistant isolates (17%) among the 12 cases in which no antenatal antibiotics were administered (P = .0004). Of the 15 mothers who were treated with ampicillin, 13 received more than 1 dose. In evaluating each year of the study, the overall administration of antibiotics to pregnant women in the antenatal period increased from <10% in 1991 to 16.9% in 1996. The incidence of early-onset neonatal sepsis with group B streptococci decreased during this time, whereas the incidence of early-onset sepsis with non-group B streptococcal organisms, especially Escherichia coli, increased.
Conclusions: The increased administration of antenatal ampicillin to pregnant women may be responsible for the increased incidence of early-onset neonatal sepsis with non-group B streptococcal organisms that are resistant to ampicillin. At this time penicillin G, rather than ampicillin, is therefore recommended for prophylaxis against group B streptococci. In addition, future studies are needed to determine whether alternate approaches, such as immunotherapy or vaginal washing, could be of benefit.