Neonatal early-onset Escherichia coli disease. The effect of intrapartum ampicillin.

BACKGROUND Maternal intrapartum ampicillin has been recommended for the prevention of neonatal group B streptococcal disease. OBJECTIVES To assess the effect of this practice, if any, on neonatal early-onset Escherichia coli infection and to delineate the clinical characteristics of infected neonates. PATIENTS AND METHODS All neonates with early-onset E coli infection who were born at Cook County Children's Hospital, Chicago, Ill, from January 1, 1982, through December 31, 1993, were identified from a microbiological register of all neonatal bacteremias and infections. Because intrapartum ampicillin use increased in our hospital since 1988, infection and case fatality rates from 1982 through 1987 (period 1) were compared with data from 1988 through 1993 (period 2). We studied maternal risk factors, clinical characteristics of infected neonates, and microbiological sensitivities of E coli isolates. RESULTS Early-onset E coli infection was diagnosed in 30 of 61,498 live births. The overall infection rate (0.49 per 1000 live births) did not change significantly during the 2 time periods (0.37 per 1000 live births during period 1 vs 0.62 per 1000 live births during period 2, P = .21; chi 2 test); however, there was an increase in the infection rate in neonates weighing between 1501 and 2500 g. Infected neonates had a clinical syndrome that was indistinguishable from early-onset group B streptococcal infection; respiratory distress was the single most frequent finding in 73% (22/30) infected neonates. An increase in the proportion of infections caused by ampicillin-resistant E coli was observed during period 2 (12/18) compared with period 1 (3/12, P = .03; Fisher exact test). During period 2, 61% (11/18) of mothers of infected neonates received intrapartum ampicillin compared with 17% (2/12; P = .02) during period 1. Overall, a higher proportion of neonates born to ampicillin-treated women had ampicillin-resistant infection (12/13 vs 3/17; P < .001). Mothers of 10 of 15 neonates with ampicillin-resistant infection had received more than 2 doses of intrapartum ampicillin. The difference between the prevalence of intrapartum fever in mothers with sensitive organisms (40%, or 6/15) and resistant organisms (93%, or 14/15) was also significant (P = .003). All 6 early-onset E coli-related deaths were due to ampicillin-resistant organisms; 4 of the 6 mothers received intrapartum ampicillin. CONCLUSIONS We have shown a shift of early-onset E coli infection from a less fulminant disease caused by ampicillin-sensitive organisms to a more fulminant disease caused by ampicillin-resistant organisms. Increased use of maternal intrapartum ampicillin therapy may account for these changes. In the absence of evidence for group B streptococcal disease, clinicians should consider the possibility of ampicillin-resistant E coli infection in critically ill neonates born to women with a history of intrapartum fever and treatment with intrapartum ampicillin.