Effect of Antibiotic Treatment of Amniotic Fluid Sludge

Abstract

Background: Amniotic fluid sludge refers to the sonographic presence of echogenic, free-floating aggregates of debris located within the amniotic cavity near the internal cervical os of women with intact membranes. Clinically, it is independently associated with increased obstetric, infectious, and neonatal morbidity, including: short cervix, chorioamnionitis, and an increased risk of preterm birth. It is thought to be infectious in nature and has been described as an intrauterine bacterial biofilm. There is little evidence on the impact of treatment with antibiotics on outcome.

Objective: To determine whether outpatient antibiotics administered to women with amniotic fluid sludge would reduce preterm birth risk compared to no antibiotic treatment.

Materials and methods: This was a retrospective cohort study of all patients diagnosed with amniotic fluid sludge by transvaginal sonography between 15 and 25 weeks' gestation in the outpatient ultrasound unit at a single academic center between 2010 and 2017. Patients were segregated according to whether they were treated with oral antibiotics at the time of diagnosis. Women with multiple gestation, fetal anomalies, preterm rupture of membranes prior to initial diagnosis of amniotic fluid sludge, and active preterm labor placenta previa and/or suspected accreta were excluded from analysis. Primary outcome of preterm birth at less than 37 weeks' gestation was compared by univariate and regression analysis to control for potential co-linear and/or confounding variables. Additional outcomes were compared by univariate analysis.

Results: A total of 181 patients were initially identified, and 97 patients met inclusion criteria. Of these patients, 51 were treated with oral antibiotics (46 azithromycin and 5 moxifloxacin), and 46 were not treated. The overall incidence of preterm birth at <37 weeks was 49.4 % (48 of 97) and preterm birth <28 weeks was 22.7% (22 of 97). There was no significant difference in preterm birth, either at <37 weeks (P = .47) or <28 weeks (P = .83) between the treated and untreated women. After adjusting for race, body mass index, tobacco use, cervical length, and preterm birth history, antibiotic treatment did not reduce the risk of preterm birth (adjusted odds ratio, 1.3; confidence interval, 0.77-1.9). No differences were seen in the incidence of preterm premature rupture of membranes (P = .94) or median latency from diagnosis to delivery (P = .47). Birthweight (P = .99), sepsis (P = .53), intraventricular hemorrhage (P = .95), and neonatal intensive care unit (NICU) admission (P = .08) were not affected by antibiotic treatment. Antibiotic treatment did not affect the incidence of either clinical or histologic chorioamnionitis (P = .92 and .14, respectively) or histologic stage 2-3 maternal or fetal inflammation (P = .94 and 0.58, respectively). Sonographic resolution of amniotic fluid sludge on first subsequent scan was seen in 34% of antibiotic-treated women and 43% of untreated women (P = .42). There was no difference in latency from diagnosis to delivery or mean gestational age at delivery according to whether sludge resolved or persisted at the first subsequent scan (P = .14 for each).

Conclusion: Antibiotic treatment of amniotic fluid sludge is not associated with a reduction in premature birth. Likewise, antibiotic treatment of amniotic fluid sludge was not associated with improvement in other obstetric, neonatal, or pathologic variables. These findings suggest that the presumed infectious nature of sludge and subsequent adverse outcomes are not treated or improved by administration of azithromycin following midtrimester sonographic diagnosis.

Keywords: amniotic fluid sludge; antibiotics; azithromycin; debris; preterm birth.