What evidence supports the use of azithromycin for ruptured membranes in preterm prelabor?

Background
Preterm prelabor rupture of membranes (PPROM) refers to rupture of the chorioamniotic membranes before the onset of labor prior to 37 weeks of gestation. 1 It occurs in around 2% to 3% of pregnancies in the United States and causes 25% to 30% of all preterm births.1-3 Some risk factors for PPROM include history of the condition, short cervical length, bleeding during the second and third trimester, low body mass index, low socioeconomic status, cigarette smoking, and illicit drug use.1

The pathophysiology of PPROM is not well understood but is thought to be multifaceted, with inflammation and infection being major causes.3 Infection is commonly implicated in causality, while in other cases it occurs after membrane rupture. In cases of causal infections, pathologic bacteria produce proteases, collagenases, and/or mucinases that damage membrane integrity and trigger inflammatory pathways, ultimately resulting in membrane rupture. In cases of secondary infections, bacteria travel upwards from the vagina and lead to intra-amniotic and/or fetal infection. Infections, whether causal or secondary, have been linked to adverse neonatal outcomes and maternal risks. Intra-amniotic infections in the fetus have been associated with neurologic damage, chronic lung disease, intraventricular hemorrhage, necrotizing enterocolitis, neonatal sepsis, and death. Maternal risks of antepartum intra-amniotic infection include sepsis, endometritis, postpartum hemorrhage, and even death in rare cases.

Prophylactic antibiotics (often termed “latency antibiotics”) have long been considered an integral part of PPROM management.3 The primary goal of antibiotic therapy in PPROM is to avoid fetal and maternal infections and thereby prolong pregnancy. A common pathogen that is present in intra-amniotic infections in the setting of PPROM is Ureaplasma urealyticum; other common bacteria include Mycoplasma hominis, and the streptococcal and staphylococcal species. Enteric gram-negative and anaerobes are also commonly present. Since these infections are polymicrobial in nature, antibiotics are selected to cover the wide array of organisms. Broad-spectrum antibiotics can prolong latency (defined as the time interval from membrane rupture to delivery), decrease risk of chorioamnionitis, and reduce gestational age-dependent neonatal morbidity when administered to women undergoing expectant management of PPROM.4

Guideline recommendations on the management of PPROM
The American College of Obstetricians and Gynecologists (ACOG) published a practice bulletin in 2020 on the management of PPROM.1 Management is influenced by gestational age and the presence of complicating factors. It is recommended that PPROM before 34 0/7 weeks of gestation should be managed expectantly if there are no maternal or fetal contraindications; expectant management of PPROM generally consists of hospitalization with periodic assessments for infection, abruptio placenta, umbilical cord compression, health status of the fetus, and labor. The ACOG makes specific recommendations regarding use of a single course of antenatal corticosteroids for pregnant women based on gestational age (not to be used before viability), based on evidence that found reductions in neonatal mortality, respiratory distress syndrome, intraventricular hemorrhage, and necrotizing enterocolitis. Fetal neuroprotective treatment with magnesium sulfate is recommended for women with PPROM before 32 0/7 weeks of gestation who are thought to be at risk of imminent delivery.

To reduce maternal and neonatal infections and gestational-age dependent morbidity, the ACOG recommends a 7-day course of latency antibiotics during expectant management of women with PPROM who are at less than 34 0/7 weeks of gestation.1 Specifically, a combination of intravenous (IV) ampicillin (2 g every 6 hours) and IV erythromycin (250 mg every 6 hours) for 2 days followed by oral amoxicillin (250 mg every 8 hours) and oral erythromycin (333 mg every 8 hours) for 5 days is recommended. If erythromycin is not tolerated or available, azithromycin (for example, a single oral dose of 1 g) is recommended as a suitable alternative; some centers have replaced the use of erythromycin with azithromycin in such situations. The ACOG guideline references 3 retrospective cohort studies that compared erythromycin and azithromycin for PPROM and found no differences in latency between the 2 drugs.5-7 Additionally, a cost analysis study is cited that found a substantial cost reduction when azithromycin was substituted for erythromycin in the standard antibiotic regimen for PPROM.8 The ACOG also recommends that women with PPROM and a viable fetus who are candidates for intrapartum group B streptococci (GBS) prophylaxis should receive intrapartum GBS prophylaxis regardless of earlier antibiotic treatments .1

As to benefits to the use of latency antibiotics in late preterm cases (gestation week 34 or later), there is some disagreement as evidenced in a commentary on the ACOG bulletin and a letter in response from the ACOG authors.9,10 The commentary pointed out that ACOG did not cite any references to support its recommendation against the use of latency antibiotics as part of expectant management for late PPROM.9 The commentary authors suggested that latency antibiotics should be used for patients with PPROM between 34 0/7 and 36 6/7 weeks of gestation. A letter in response from ACOG stated that although evidence supports use of latency antibiotics for patients diagnosed with PPROM prior to 34 weeks of gestation, it has not been established that the same benefit can be extrapolated to patients diagnosed with PPROM between 34 0/7 and 36 6/7 weeks of gestation.10

Some international organizations also provide recommendations on prophylactic antibiotic use for PPROM.11-13 The 2022 Society of Obstetricians and Gynecologists of Canada (SOGC) guideline on the management of PPROM has similar recommendations to ACOG.11 Antibiotics are recommended for women diagnosed with PPROM from viability to 34 weeks of gestation. The following 2 antibiotic regimens are recommended: 1) a macrolide (erythromycin, azithromycin, or clarithromycin) alone or associated with GBS coverage for 2 days (if GBS status is unknown or positive), or 2) a combination of ampicillin/amoxicillin and a macrolide regardless of GBS status. If using azithromycin, the following dosages should be used: 1 g oral single dose, OR 500 mg to 1 g oral single dose, then 250 mg oral every 24 h for 4 days. To support the use of azithromycin, the SOGC guideline references the same retrospective studies as ACOG, in addition to a 2021 meta-analysis.4-8

The Royal College of Obstetricians and Gynecologists (RCOG) and the National College of Gynecologists and Obstetricians of France (CNGOF) both have recommendations on erythromycin but not azithromycin.12,13 The 2019 RCOG guidance recommends erythromycin for 10 days following the diagnosis of PPROM, or until the woman is in established labor (whichever is sooner); gestational age is not specified.12 The 2019 CNGOF guideline recommends that antibiotic prophylaxis should be prescribed at admission for PPROM.13 Amoxicillin, a third-generation cephalosporin, and erythromycin (professional consensus) can be used individually or erythromycin and amoxicillin can be combined (professional consensus) for a period of 7 days.

Literature on azithromycin for PPROM
The 2021 meta-analysis by Seaman et al, cited in the SOGC guideline, aimed to compare the effects of erythromycin and azithromycin on the duration of latency and the rate of clinical chorioamnionitis in patients with PPROM.4 Five cohort studies with a total of 1289 women (781 treated with azithromycin and 508 with erythromycin) were included in the meta-analysis. In 4 studies (all conducted in the United States), patients received concurrent ampicillin with either azithromycin or erythromycin, followed by amoxicillin; in the remaining study (conducted in Iran), patients only received macrolide antibiotics. Patients were administered IV erythromycin (250 mg every 6 hours) for 2 days, followed by oral erythromycin (333 mg or 500 mg every 8 hours based on institutional preference) for 5 days. Azithromycin was substituted for IV erythromycin in cases of hospital shortages. Depending on the institution, one of these azithromycin regimens was used: 1 g orally given once, OR 500 mg orally on the first day followed by 250 mg orally daily for 4 days, OR 500 mg IV for 2 days followed by 500 mg orally for 5 days. Around 50% of women in the studies were given 1 g orally once. Antenatal corticosteroids for fetal lung maturity were administered in 4 studies; 1 study did not report administration of antenatal corticosteroids.

All included studies were rated as low risk of bias.4 Among the 5 studies, the mean estimated gestational age at the time of PPROM diagnosis was similar between women treated with erythromycin (29.6±2.6 weeks) and azithromycin (29.7±2.0 weeks), with a mean difference of 0.20 weeks (95% confidence interval [CI], 0.43 to 0.83; I2 [heterogeneity statistic], 84%]). The mode of delivery was also comparable between patients treated with either agent. The mean length of latency was similar between women treated with erythromycin (6.6 days) and azithromycin (6.7 days), with a mean difference of 0.07 days (95% CI, 0.45 to 0.60; I2, 0%). The median point prevalence rates of clinical chorioamnionitis in patients treated with erythromycin and azithromycin, respectively, were 25% (95% CI, 12 to 32) and 14% (95% CI, 9 to 24). The overall clinical chorioamnionitis rate in patients who received azithromycin was lower than those who received erythromycin (pooled odds ratio, 0.53; 95% CI, 0.39 to 0.71; I2, 0%). A major limitation of the meta-analysis was its inclusion of only cohort studies due to the absence of randomized trials that directly compared the 2 macrolides. Additionally, 3 of the 5 studies included were from single institutions. Although macrolide selection was based on drug availability, dosing was left to the physician’s discretion, which may have potentially confounded the results.

The meta-analysis excluded 1 retrospective study since it included patients with membrane rupture between 34 0/7 and 36 6/7 weeks of gestation (median, 34 weeks).4,14 This study compared outcomes in women who received a regimen of azithromycin plus amoxicillin (single stat dose of azithromycin 1 g orally, in addition to amoxicillin 2 g IV every 6 hours for 2 days followed by amoxicillin 250 mg orally 3 times daily for 5 days) and women in a historical control group who received erythromycin monotherapy (250 mg orally 4 times per day for 10 days). Women who received azithromycin and amoxicillin had a longer median latency from time of membrane rupture to delivery compared to those who received erythromycin monotherapy (5.5 versus 2 days; p<0.001). No differences were found in the mode of delivery or maternal high dependency unit admission.

Since the publication of the 2021 meta-analysis, 2 studies have been conducted on the use of azithromycin for PPROM.2,15 A 2022 retrospective, single-center cohort study sought to determine the most optimal broad-spectrum antibiotic for the treatment of PPROM in Taiwan.15 The study included 133 Taiwanese women diagnosed with PPROM before 34 6/7 weeks of gestational age and delivered after 24 0/7 weeks of gestational age. Patients received microbiology analysis via cervical culture. Empiric cefazolin (1 g IV every 8 hours) was administered until either delivery or no bacterial growth in the cervical culture; in cases of cefazolin allergy, clindamycin (900 mg IV every 8 hours) was administered. The mean patient age was 34.9 years, and the mean time from PPROM to delivery was 3.37 days. Of the 133 women included, 121 had positive culture results. Since >1 pathogen could be found in the cervical culture, there were 248 total culture results. The most common pathogens isolated were the Lactobacillus (27.8%), Streptococcus (12.9%), and Staphylococcus (12.09%) species. Resistance testing was performed, and ampicillin demonstrated a high resistance rate in gram-negative pathogens, indicating the need for at least a second-generation cephalosporin. Erythromycin showed a high resistance rate in gram-positive pathogens, indicating the need for ampicillin or at least a second-generation cephalosporin. The hospital did not perform a sensitivity test for azithromycin, but authors cited many studies that demonstrated the efficacy of azithromycin in the management of Ureaplasma infections. Based on the culture/resistance findings and existing literature, the authors concluded that the combination of azithromycin (1 g orally administered on admission to cover Ureaplasma and prolong pregnancy) plus an IV third-generation cephalosporin for the first 2 days followed by amoxicillin (500 mg orally every 8 hours) for an additional 5 days is the optimal regimen for PPROM. This conclusion was an extrapolation (since azithromycin was not used in the study); furthermore, the bacterial composition and resistance rates in this single-centered, Taiwanese study may not be reflective of other regions or populations, such as the U.S. population. Other study limitations include its retrospective nature and the absence of testing for Ureaplasma, one of the most common bacteria implicated in PPROM.

Lastly, a retrospective, multicenter cohort study published in 2023 included 287 women with PPROM who had received azithromycin as part of their prophylactic latency antibiotic regimen.2 The study aimed to determine whether extended azithromycin administration affects the latency time in PPROM. Patients who were 23 0/7 and 33 6/7 weeks of gestation were identified and administered either limited (<2 days) or extended (7 days) azithromycin therapy. Patients in the limited administration group received 1 of the following azithromycin regimens: 1000 mg orally once, OR 500 mg IV once, OR 500 mg orally for up to 2 days. Patients receiving extended azithromycin were administered 500 mg IV for 2 days followed by 500 mg orally for 5 days. Per the institution standard, all patients received concomitant IV ampicillin for 2 days followed by oral amoxicillin for 5 days; additionally, patients who were GBS positive in the intrapartum period were administered IV penicillin regardless of latency antibiotic regimen. The average maternal age was around 30 years, the average body mass index was approximately 30 kg/m2, and most patients were Black. There were 165 patients in the limited azithromycin group compared to 122 patients in the extended azithromycin group. The length of gestational latency (primary endpoint) was significantly longer for women who received extended azithromycin than those who received limited azithromycin; the adjusted median gestational latency in women who received limited azithromycin administration extended by >3 days (2.6 days [interquartile range, 2.2 to 3.1]) compared to 5.8 days (interquartile range, 4.8 to 6.9) for those who received extended azithromycin administration (p<0.001). No differences in chorioamnionitis and adverse neonatal outcomes were noted between the 2 groups. The study had limitations related to its retrospective nature and administration of concomitant medical treatments that may have confounded results (e.g., additional magnesium for neuroprotection, additional antibiotics for genitourinary infections).

Conclusion
Clinical practice guidelines and several studies support the use of azithromycin for ruptured membranes in preterm prelabor. The ACOG and SOGC both recommend the use of azithromycin as an alternative to erythromycin in women with PPROM.1,11 These recommendations are primarily based on retrospective studies demonstrating similar latency and neonatal outcomes between erythromycin and azithromycin, as well as better tolerance and lower cost. Reinforcing findings were also reported in a 2021 meta-analysis that compared erythromycin and azithromycin for the treatment of PPROM <34 weeks; there were similar latency and neonatal outcomes between the 2 macrolides as well as a lower rate of clinical chorioamnionitis in patients treated with azithromycin versus erythromycin.4 Further evidence on the use of azithromycin for PPROM is provided by 2 recent retrospective cohort studies— a Taiwanese study that suggested that a combination of azithromycin, a third-generation cephalosporin, and amoxicillin is the optimal regimen for PPROM, and a retrospective study that concluded that extended azithromycin administration (for 7 days) was associated with a significant increase in latency without an increase in adverse maternal or neonatal outcomes.2,15 Despite generally favorable findings, the studies had several limitations on top of the inherent disadvantage of retrospective reviews. Prospective, randomized clinical trials with a larger study population are necessary to further investigate the safety and efficacy of azithromycin compared to other agents for the treatment of PPROM.

References

  1. Prelabor rupture of membranes: ACOG practice bulletin, number 217. Obstet Gynecol. 2020;135(3):e80-e97. doi:10.1097/AOG.0000000000003700
  2. DiSciullo AJ, Hand M, Iqbal SN, Chornock RL. Outcomes after extended azithromycin administration in preterm premature rupture of membranes. AJOG Glob Rep. 2023;3(2):100206. doi:10.1016/j.xagr.2023.100206
  3. Dotters-Katz S. Antibiotics for prophylaxis in the setting of preterm prelabor rupture of membranes. Obstet Gynecol Clin North Am. 2020;47(4):595-603. doi:10.1016/j.ogc.2020.08.005
  4. Seaman RD, Kopkin RH, Turrentine MA. Erythromycin vs azithromycin for treatment of preterm prelabor rupture of membranes: a systematic review and meta-analysis. Am J Obstet Gynecol. 2022;226(6):794-801.e1. doi:10.1016/j.ajog.2021.12.262
  5. Navathe R, Schoen CN, Heidari P, et al. Azithromycin vs erythromycin for the management of preterm premature rupture of membranes. Am J Obstet Gynecol. 2019;221(2):144.e1-144.e8. doi:10.1016/j.ajog.2019.03.009
  6. Pierson RC, Gordon SS, Haas DM. A retrospective comparison of antibiotic regimens for preterm premature rupture of membranes. Obstet Gynecol. 2014;124(3):515-519. doi:10.1097/AOG.0000000000000426
  7. Finneran MM, Appiagyei A, Templin M, Mertz H. Comparison of azithromycin versus erythromycin for prolongation of latency in pregnancies complicated by preterm premature rupture of membranes. Am J Perinatol. 2017;34(11):1102-1107. doi:10.1055/s-0037-1603915
  8. Finneran MM, Smith DD, Buhimschi CS. Cost analysis of azithromycin versus erythromycin in pregnancies complicated by preterm premature rupture of membranes.Am J Perinatol. 2019;36(1):105-110. doi:10.1055/s-0038-1667369
  9. Siegler Y, Weiner Z, Solt I. ACOG practice bulletin No. 217: prelabor rupture of membranes. Obstet Gynecol. 2020;136(5):1061. doi:10.1097/AOG.0000000000004142
  10. Pettker CM, Ehsanipoor RM, Turrentine MA, Caughey AB, Kaimal AJ, Zahn CM. In reply. Obstet Gynecol. 2020;136(5):1062. doi:10.1097/AOG.0000000000004143
  11. Ronzoni S, Boucoiran I, Yudin MH, et al. Guideline No. 430: diagnosis and management of preterm prelabour rupture of membranes. J Obstet Gynaecol Can. 2022;44(11):1193-1208.e1. doi:10.1016/j.jogc.2022.08.014
  12. Thomson AJ; Royal College of Obstetricians and Gynaecologists. Care of women presenting with suspected preterm prelabour rupture of membranes from 24+0weeks of gestation: green-top guideline No. 73. BJOG. 2019;126(9):e152-e166. doi:10.1111/1471-0528.15803
  13. Schmitz T, Sentilhes L, Lorthe E, et al. Preterm premature rupture of the membranes: guidelines for clinical practice from the French College of Gynaecologists and Obstetricians (CNGOF). Eur J Obstet Gynecol Reprod Biol. 2019;236:1-6. doi:10.1016/j.ejogrb.2019.02.021
  14. Fitzgibbon A, Clooney L, Broderick D, Eogan M, McCallion N, Drew RJ. Erythromycin compared to amoxicillin and azithromycin for antimicrobial prophylaxis for preterm premature rupture of the membranes: a retrospective study. J Obstet Gynaecol. 2021;41(4):569-572. doi:10.1080/01443615.2020.1786806
  15. Chen HY, Huang KY, Lin YH, Lin SY, Lee CN. Antibiotic choice for the management of preterm premature rupture of membranes in Taiwanese women. J Formos Med Assoc. 2022;121(9):1798-1803. doi:10.1016/j.jfma.2022.03.015

Prepared by:
Jenny Radnaa
PharmD Candidate Class of 2024
University of Illinois at Chicago College of Pharmacy

Honey Joseph, PharmD, BCPS
Clinical Assistant Professor, Drug Information Specialist
University of Illinois at Chicago College of Pharmacy

January 2024

The information presented is current as November 8, 2023. This information is intended as an educational piece and should not be used as the sole source for clinical decision-making.