Abstract 1579
Poster Session II, Sunday, 5/2 (poster 108)
Management of newborns with suspected infection is controversial. Few studies have reported data on all babies ever evaluated. We prospectively identified all "sepsis work-ups" in 6 Kaiser Permanente hospitals from 10/95 through 11/96. Among 19043 births, 2785 met study criteria (birth weight ≥ 2000gm and a sepsis evaluation). We reviewed subjects' laboratory data, maternal and neonatal records, and tracked 99.7% to 1 week post discharge. Of the 2785 infants, 2539 (91.2%) were identified as being at risk of sepsis by 12 hours of age; 853 (30.6%) received systemic antibiotics; 206 (7.4%) were ventilated; 22 (0.8%) had a positive culture (15 group B streptococcus, 5 E. coli, 2 other; 1 death in this group); 40 (1.4%) had clinically confirmed infections with negative blood/CSF cultures (3 deaths in this group); and 67 (2.4%) were rehospitalized (2 with group B streptococcus bacteremia). Among 679 babies who required supplemental oxygen, 93.1% received such therapy by 12 hours of age, while 89.8% of the 206 who were ventilated had such therapy initiated by 12 hours of age. Maternal fever, chorioamnionitis, low neonatal absolute neutrophil count (ANC) for age, and presence of neonatal clinical signs were associated with infection. There were 1217 infants whose mothers received intrapartum antibiotics and 1568 whose mothers did not. Compared to infants whose mothers were not treated, infants of treated mothers were more likely to be asymptomatic (71.5% vs. 50.9%, p = 0.001) and less likely to be critically ill within 6 hours of birth (5.4% vs. 7.5%, p = 0.038). We stratified infants according to maternal treatment status and conducted multivariate analyses. Among infants whose mothers were not treated, maternal chorioamnionitis (adjusted odds ratio [AOR] 2.40, 95% confidence interval [CI], 1.15-5.00), low ANC for age (AOR 2.84, 95% CI 1.50-5.38), and presence of meconium in the amniotic fluid (AOR 2.23, 95% CI, 1.18-4.21) were associated with an increased risk for infection, while initial asymptomatic status was associated with a decreased risk (AOR 0.26, 95% CI 0.11-0.63). Results were similar for infants whose mothers were treated except that chorioamnionitis was not a significant predictor for infection. We also found that 1) use of epidural anesthesia is associated with a 0.5°F increase in maternal temperature, even after controlling for the presence of chorioamnionitis, 2) published ANC norms misclassify almost half of babies with infections, 3) increased risk of infection is seen when rupture of membranes exceeds 12 hours, and 4) widespread practice variation exists with respect to maternal and neonatal antibiotic treatment. Current guidelines need to be revised in light of these data which highlight the protective effect of maternal intrapartum antibiotics. Evidence-based approaches should emphasize 1) careful assessment in the first 24 hours of age, 2) close attention to maternal risk factors, and 3) modification of ANC norms used to categorize infants as being at high risk for sepsis.
