Preterm prelabour rupture of membranes: the use of amniocentesis to detect intra-amniotic infection reduces maternal and neonatal duration of antibiotic exposure.

Abstract

Background: The current management of patients with preterm prelabor rupture of membranes below 32+0 weeks recommend administering antibiotics although its benefit on the short or long-term neonatal outcome is poor.

Objective: To evaluate whether the use of amniocentesis to detect intra-amniotic infection reduces maternal and neonatal duration of antibiotic exposure.

Methods: This was a retrospective observational cohort study (2014-2023) including patients diagnosed with preterm prelabor rupture of membranes below 32+0 weeks with an amniocentesis at admission to assess the presence of intra-amniotic infection. We compared two groups according to antenatal management regarding antibiotic treatment. Thus, from 2014-2019, patients were treated at least 5 days with a broad-spectrum antibiotic treatment including intravenous ampicillin and gentamicin, and a single dose of oral azithromycin, regardless of ruling out intra-amniotic infection (standard management group). Beyond 2019, gentamicin was substituted for intravenous ceftriaxone and azithromycin for oral clarithromycin. Antibiotic duration was optimized based on amniotic fluid analysis (amniocentesis-based management): if amniotic fluid glucose concentrations were greater than or equal to 14 mg/dL and Gram staining did not show the presence of bacteria, antibiotic treatment was discontinued at 48 hours. Otherwise, antibiotics were prolonged at least until microbiological amniotic fluid results. Regardless of the management group, if intra-amniotic infection was diagnosed, the type of antibiotic was individualized according to the bacteria isolated and treatment was prolonged for 7-10 days if the delivery did not occur before. There were no other differences in maternal management between the two periods.

Results: One hundred seventy-two patients diagnosed with preterm prelabor rupture of membranes below 32+0 weeks were included (122 in the standard management group and 50 in the amniocentesis-based management group). The prevalence of intra-amniotic infection was 29% in both periods, with most (61%) being due to Ureaplasma spp. There were no differences in maternal characteristics between the two groups. As expected, in the amniocentesis-based management group there was shorter maternal exposure to antibiotics (median (25th centile; 75th centile) of 2 (2;3) days (amniocentesis-based management) vs. 5 days (4;5) (standard management), p <0.0001)). In line with the reduction of the duration of antibiotics, we observed that maternal hospital stay was significantly shorter (5 (4; 9) days vs. 11 (5; 21) days, p=0.001), and outpatient management was more frequent (68% vs. 47%, p=0.011). No differences were observed in the in maternal morbidity. Similar results were found when neonatal outcomes were evaluated. Thus, in the amniocentesis-based management group, neonates received less antibiotic treatment at admission with an odds ratio (OR) 95% confidence interval (CI) of 0.31 (0.15-0.61), p<0.001 and less exposure to antibiotics during hospitalization 6 ± 14 days (amniocentesis-based management) vs. 13±18 days (standard management), p 0.023. This did not translate into worse neonatal outcomes.

Conclusions: The management of PPROM based on amniotic fluid analysis was associated with lower maternal and neonatal duration of antibiotic exposure, a shorter maternal length of hospitalization, allowed outpatient management without jeopardizing maternal or neonatal outcomes.