Objective: In a pregnant rabbit model using intracervical inoculation of Escherichia coli with delayed antibiotic therapy, we investigated the rate of positive cultures and histologic inflammation of maternal and fetal compartments and the concentration of tumor necrosis factor-alpha in the amniotic fluid for up to 5 days.
Study design: New Zealand White rabbits at 70% gestation were inoculated intracervically with 10(3) - 10(4) colony-forming units of E coli per uterine horn. At varying intervals after inoculation (0.5 - 4.0 hours), antibiotic therapy was initiated with ampicillin-sulbactam. Primary outcomes were positive cultures and histologic inflammation score. Tumor necrosis factor-alpha levels in the amniotic fluid were determined by bioassay.
Results: A total of 60 animals were inoculated with E coli. At the endpoint, uterine cultures were positive more commonly than in the fetus or amniotic fluid (41.8% vs 27.5% vs 17.3%, respectively), which was consistent with an ascending pathway of infection. Inflammation scores were similar in uterus and placenta but lower in fetal lung and absent in fetal brain (2.8 vs 3.1 vs 0.84 vs 0.0, respectively). Comparing the durations of delay in antibiotic administration, we found a significant increase in positive uterine cultures and a significant increase in histologic inflammation score with increasing delay. The proportion of dead pups within a litter was significantly associated with the log of the tumor necrosis factor-alpha concentration in amniotic fluid and the degree of histologic inflammation in the uterus, but not with amniotic fluid or other culture positivity.
Conclusion: The administration of therapeutic doses of antibiotic does not consistently eradicate bacteria from the rabbit uterus nor, more importantly, from the fetus and the amniotic fluid. Obtaining a negative amniotic fluid culture does not exclude either infection in the decidua or the fetus or histologic inflammation with tumor necrosis factor-alpha elaboration.