Subramaniam, Akila; Waites, Ken B.; Jauk, Victoria C.; Biggio, Joseph R.; Sutton, Amelia L. M.; Szychowski, Jeff M.; Andrews, William W.; Tita, Alan T. N.
Objective To explore whether the effect of azithromycin (AZI) on postcesarean infections varied by the presence/absence of genital mycoplasmataceae placental colonization. Study Design This was a single-center substudy of multicenter double-blind C/SOAP (Cesarean Section Optimal Antibiotic Prophylaxis) trial of women randomized to AZI or placebo (+cefazolin) antibiotic prophylaxis at cesarean. Chorioamnion/placenta specimens were tested for genital mycoplasmataceae colonization by polymerase chain reaction. Primary outcome was a composite of endometritis, wound infection, or other infections up to 6 weeks postpartum. Analysis was intent-to-treat; logistic regression was used to evaluate interactions between treatment assignment (AZI/placebo) and the presence/absence of mycoplasmataceae and to quantify effects of AZI in analyses stratified by the presence/absence of these microorganisms. Results Specimens from 613 women (303 AZI and 310 placebo) were evaluated. Baseline characteristics were similar between groups, and approximately 1/3 (30.3%) had mycoplasmataceae placental/chorioamnion colonization. There was no evidence of effect modification ( p (interaction) = 0.79) between treatment assignment and the presence/absence of organisms. Stratified analyses showed fewer events in the AZI group in the presence (odds ratio [OR]: 0.42; 95% confidence interval [CI]: 0.17-1.01) and absence (OR: 0.49; 95% CI: 0.24-1) of mycoplasmataceae. Results were similar with endometritis/wound infections and with ureaplasmas/mycoplasmas considered separately. Conclusion The reduction in postcesarean infection with AZI does not vary based on the presence or absence of genital mycoplasmataceae placental colonization.