Premature rupture of membranes (PROM) is defined as rupture of the amniotic membrane prior to labor, which occurs in approximately 20% of all pregnancies [1]. If PROM occurs during term, immediate delivery is recommended, as it is associated with a significantly lower perinatal morbidity rate than expectant management [1234]. However, the management of women with preterm PROM (PPROM), accounting for 40% of the total preterm deliveries, is somewhat controversial. Immediate delivery may lead to complications resulting from fetal immaturity, but expectant management is associated with risks such as placenta abruptio, infection, fetal distress, and umbilical cord prolapse, causing a medical dilemma [256]. In early PPROM, defined as PROM before 34.0 weeks of gestation, expectant management is strongly recommended because of adverse neonatal outcomes from prematurity. In a study of Ekin et al. [7], although complications such as chorioamnionitis and placental abruption were increased, the overall adverse pregnancy outcomes were decreased in women managed expectantly. The optimal management of late PPROM, defined as PROM between 34.0 weeks and 36.6 weeks of gestation, remains inconclusive. Therefore, the management of late PPROM should be determined on the basis of a comprehensive acknowledgment of the risk of infection and possible complications from premature delivery. According to the 2018 American College of Obstetricians and Gynecologists (ACOG) guidelines [18], expectant management, including a combination therapy of intravenous ampicillin and erythromycin, administration of antenatal corticosteroids until 34.0 weeks of gestation and group B Streptococcus prophylaxis, is strongly recommended. The guidelines recommend prompt delivery after 34.0 weeks of gestation. However, the Cochrane review mentioned the lack of clinical evidence to support these guidelines [28] The 2018 ACOG guideline also suggests that discussion with the mother about the benefits and risks of expectant management is required if expectant management continues after 34 weeks of gestation.

Recently, several randomized controlled trials have been published that showed results contrary to those mentioned in the current guidelines. In these studies, no significant differences in the frequencies of perinatal complications were found between the expectant management and immediate delivery groups in late PPROM [9]. The incidence rate of complications of preterm delivery, such as respiratory distress syndrome, use of mechanical ventilation, and neonatal intensive care unit admission, was higher in the immediate delivery group [4]. It is expected that different management is practiced in various cases of late PPROM. Therefore, this study was conducted to investigate the current management of mothers with PPROM in Korea and to use the results as baseline data for national guidelines.

A total of 113 obstetricians participated in the survey. Table 1 showed the baseline characteristics of 112 obstetricians, excluding 1 obstetricians who did not respond to the basic information question. Among the respondents, 68.1% were female, and the mean age was 42.1 years. Overall, 31% had <5 years of experience, followed by 5–10 years (19.5%) and 10–20 years (28.3%), and 57.5% worked in tertiary hospitals. The working hospital of 93 of the 113 respondents were in Seoul or Gyeonggi Province.

In the case of PPROM occurring at <37.0 weeks of gestation, the most frequently prescribed antibiotics were cephalosporins and were used routinely by 92 (81.4%) of the responding obstetricians, whereas 44 (38.9%) of the respondents used penicillin and 44 (38.9%) used macrolides. Among them, 39 (34.5%) of the responding obstetricians reported to have prescribed cephalosporin alone and 14 (12.4%) prescribed penicillin alone (Table 2).

The prevalence of corticosteroid use and the preferred limit of gestation of corticosteroid use are presented in Table 3 and Fig. 1. The prevalence of corticosteroid use in PPROM at <37 weeks of gestation was 88.5%. According to the survey, corticosteroids were applied most frequently up to 34 weeks of gestation (50%), followed by 36 (16%) and 35 weeks (13%) of gestation (Fig. 1). The prescribed corticosteroids were betamethasone (52%) and dexamethasone (43%; Table 3).

We also investigated the preferred treatment options of the attending physician when PPROM occurred at <34.0 weeks of gestation (early PPROM) and between 34.0 and 36.0 weeks of gestation (late PPROM). The assumed conditions of the pregnant women were as follows: 1) absence of uterine contraction, 2) absence of medical/surgical disease or obstetrical problem, 3) absence of indication for cesarean section, and 4) absence of abnormal fetal conditions such as intrauterine growth restriction. When the gestational age reached 34.0 weeks in the early PPROM group, 71.3% of the responding obstetricians chose expectant management planning for delivery around term, whereas only 28.7% chose immediate delivery. In comparison for late PPROM, 57.41% preferred expectant management and 42.59% preferred immediate delivery (Table 4).

The consideration factors for each subgroup in determining immediate delivery was also studied (Table 5). The most important determinants of immediate delivery were clinical signs and symptoms of chorioamnionitis, followed by abnormal findings on amniocentesis and maternal blood test.

In this survey, 71.3% of the obstetricians favored expectant management and 28.7% chose immediate delivery when the gestational age reached 34.0 weeks in the early PPROM group. Approximately 37.5% and 58.1% of the responders who preferred expectant management in early and late PPROM, respectively, wanted to delay delivery until after 37.0 weeks of gestation. Among the responders who preferred immediate delivery, 65.6% answered that they made the decision due to concerns about ascending infection such as chorioamnionitis. These results suggested that actual decision making in clinical settings differs from the guidelines, which recommend immediate delivery in late PPROM and at 34.0 weeks in the early PPROM group.

The difference between the guidelines and our findings is quite remarkable [1]. The discrepancy between these results can be interpreted in 2 ways. First, some physicians with longer duration of clinical experience tended to prefer expectant management than immediate delivery. Second, primary physicians may still not be fully acquainted with the guidelines for PPROM. This suggests that the guidelines need to be revalidated, and the primary physicians must be educated about the verified revalidated guidelines.

The intravenous antibiotic regimen in the ACOG guideline is a combination of intravenous ampicillin and erythromycin [10111213]. However, there was variation in the choice of antibiotics. The most commonly used regimen (34.5%) in our survey was cephalosporin monotherapy, followed by macrolides or macrolides plus metronidazole combined with penicillin or cephalosporin series.

In addition, many physicians appear to conduct corticosteroid therapies based on the latest knowledge [789]. The benefits of antenatal corticosteroids have been well established in preterm neonates in reducing the risk of neonatal respiratory distress, intraventricular hemorrhage, and neonatal death [2]. According to the latest guidelines, a single course of corticosteroids is recommended for pregnant women between 24.0 and 34.7 weeks of gestation. Furthermore, recent data suggest that administration of betamethasone in the late preterm period between 34.0 and 36.6 weeks reduces respiratory morbidity in newborns [1]. Moreover, they used dexamethasone and betamethasone almost in the same proportion as that used in this study.

There is great controversy about the choice of immediate delivery or expectant management in PPROM during 34.0–36.6 weeks of gestation. Some high-evidenced studies such as randomized clinical trials and meta-analyses on PPROM management have been published lately but could not provide a definite conclusion [101112].

In the 2010 Cochrane review, it was concluded that the evidence on expectant management and immediate delivery was insufficient [14]. In 2016, the PPROMT trial published in Lancet reported that the immediate delivery group had more preterm-related complications, without significant differences in other outcomes [4]. The PPROMEXIL 1 and 2 trials of van der Ham et al. [15], reported in 2012, showed no significant differences in neonatal sepsis (2.7% at 34.0 weeks vs. 4.1% at 37.0 weeks of gestation; relative risk [RR], 0.66; 95% confidence interval [CI], 0.3–1.5) between 2 groups. However, induction of labor significantly reduced the risk of chorioamnionitis (1.6% at 34.0 weeks vs. 5.3% at 37.0 weeks of gestation; RR, 0.31; 95% CI, 0.1–0.8). The 2018 ACOG guidelines pointed out that these studies did not have sufficient statistical power to show a significant reduction in the rate of neonatal sepsis because the overall rate of sepsis was lower than anticipated. More evidence-based studies with higher statistical power are required for the new guideline that supports expectant management in PPROM during 34.0–36.6 weeks of gestation [1].

So far, there is a lack of national studies on this theme. It is important to investigate our current management of PPROM for future research. The data presented are only from a questionnaire that reflects real practices. However, this study may facilitate establishment of future policy on PPROM management in Korea. This study has few limitations despite its significance. This survey did not include all obstetricians in Korea, and most participants worked around the capital area. Therefore, the results cannot be generalized.

In conclusion, the results of the questionnaire survey were as follows: 1) The most commonly used antibiotics was cephalosporin alone, followed by cephalosporin or penicillin plus macrolide or macrolide and metronidazole. 2) The corticosteroids were applied up to 34 weeks of gestation in half of the participants. 3) The frequency of expectant management was higher than that of immediate delivery in women with PPROM after 34 weeks of gestation. This may be interpreted as an attempt to improve the perinatal outcome by prolonging the gestational age in the absence of specific findings in the symptoms and laboratory findings of the mother. 4) The factors that influenced determination of delivery in the patients with PPROM were symptoms of chorioamnionitis, abnormal findings on amniocentesis, and blood test results.

Our data showed a considerable variation in the actual management of PPROM. It is thought that more evidence, including a randomized clinical trial, will be needed to narrow the gap between the current guideline and actual clinical practice