Clinical trial of interconceptional antibiotics to prevent preterm birth: subgroup analyses and possible adverse antibiotic-microbial interaction.

ABSTRACT

OBJECTIVE: The purpose of this study was to explore whether endometrial microbial colonization and plasma cell endometritis are risk factors for adverse pregnancy outcomes, and whether these outcomes are influenced by interactions between interconceptional antibiotics and the micro-flora. STUDY DESIGN: Subgroup analyses of data from a double-blind, randomized, placebo-controlled trial of a course of metronidazole plus azithromycin given every 4 months to women with a prior preterm delivery to prevent recurrent preterm delivery. Endometrial cultures and histology were obtained at randomization and repeated 2 weeks after the first treatment. Fifty-nine on antibiotics versus 65 on placebo had pregnancy outcomes. Prevalence of adverse pregnancy outcomes (pregnancy loss or preterm birth < 37 weeks) was stratified by treatment group and endometrial characteristics. Subgroups were assessed and screened for potential interaction (P values for significance set a priori at < .01), prior to formal statistical testing for interaction (P values < .05). RESULTS: The prevalence of adverse pregnancy outcome was 62.7% in the presence of endometrial microbial colonization at baseline (any microbe) and 50% in the absence of colonization (RR = 1.25; 99% CI 0.42-3.7). Prevalence of adverse pregnancy outcomes was 61.9% with plasma cell endometritis, and 70.8% without; RR = 0.87 (0.50-1.5). There was a nonsignificant reduction in adverse pregnancy outcome in the absence of Gardnerella vaginalis or gram-negative rods with RR (95% CI) = 0.60 (0.3-1.2) and 0.66 (0.4-1.2), respectively. In the presence of these microbes, antibiotics appeared to increase adverse outcomes: RR = 1.5 (1.1-2.0) and 1.5 (1.1-2.1), respectively. This reversal of impact represents a crossover interaction. CONCLUSION: Neither baseline endometrial microbial colonization nor plasma cell endometritis were risk factors for adverse pregnancy outcome. However, colonization with specific microbes interacted with antibiotics to increase adverse outcomes.