Preterm premature rupture of membranes (PPROM) is defined as rupture of the fetal membranes <37 complete weeks gestation and before the onset of spontaneous labor. It complicates approximately 3% of all pregnancies in the United States¹ and is a significant cause of spontaneous preterm birth.² The most significant sequelae of PPROM are the complications related to prematurity, and gestational age is strongly correlated with prognosis. Umbilical cord compression or prolapse, placental abruption, nonreassuring fetal heart tracing, chorioamnionitis, and maternal or neonatal sepsis are additional complications associated with PPROM. While PPROM is associated with significant perinatal morbidity and mortality, there are many evidence-based management options available to optimize outcomes.

Premature rupture of membranes (PROM) refers to rupture of the fetal membranes prior to the onset of labor, and it occurs in approximately 8% of pregnancies at term.³ PPROM complicates approximately 3% of pregnancies and is the etiology of 25% to 30% of preterm births.^4,5

RISK FACTORS

Several risk factors for PPROM have been identified, and they are similar to the risk factors for spontaneous preterm labor. A history of PPROM in a previous pregnancy is a major risk factor, and the recurrence risk in such instances is 13.5% to 32%.^6,7 Additional risk factors include antepartum bleeding, multiple gestation, polyhydramnios, cervical shortening, positive fetal fibronectin (FFN), cerclage, low body mass index (BMI), cigarette smoking, urogenital tract infection, and illicit drug use.^8,9 Procedures that result in disruption of the amnion, such as amniocentesis or fetal surgery, have a risk of membrane rupture. However, PPROM often occurs in the absence of any identifiable risk factors.

The fetal membranes are composed of the chorion and the amnion, closely adherent layers consisting of epithelial cells, mesenchymal cells, trophoblast cells, and a collagenous matrix.⁹ During labor, mechanical forces caused by uterine contractions combined with a natural weakening of the membranes at term are believed to lead to membrane rupture.¹⁰ And prior to the onset of labor, membrane rupture can occur as a result of a variety of pathologic mechanisms. Although the pathophysiology of PPROM is not fully understood, it appears to develop (at least in part) as a result of a weakening process that is regulated at the cellular level.¹¹ Collagen remodeling, along with activation of matrix metalloproteinases and apoptosis, can contribute to this weakening.^11,12 Inflammation and intra-amniotic infection have been associated with PPROM.¹⁰ It is likely that in many cases, membrane rupture occurs as a result of a multifactorial process.

NATURAL HISTORY

Most women with PPROM will deliver within 1 week.¹³ A randomized trial comparing latency antibiotics to placebo in the setting of PPROM between 24 and 32 weeks gestation found the median time to delivery to be about 6 days. In this study, 27% of patients delivered within 48 hours, 56% within 7 days, 76% within 14 days, and 86% within 21 days.¹⁴ In general, the latency period is inversely proportional to gestational age.¹⁵ With the exception of PPROM related to amniocentesis, the membranes usually do not reseal.¹⁶

COMPLICATIONS

Although prolonging pregnancy in the setting of PPROM to achieve birth at a more advanced gestational age diminishes the complications associated with prematurity, expectant management is not without risk. It is important to be cognizant of the risks of chorioamnionitis, placental abruption, precipitous labor, nonreassuring fetal status, umbilical cord prolapse, and stillbirth. These risks should be discussed with the patient and documented in the medical record. Chorioamnionitis complicates 15% to 25% of pregnancies with PPROM. Underlying infection may be present prior to PPROM, or the breach in the barrier between the uterus and vagina may permit ascending infection. Placental abruption complicates 2% to 5% of pregnancies with PPROM. The blood loss is most often noted by vaginal bleeding, although the possibility of a concealed hematoma in the uterus should be considered.

Placental abruption appears to be more likely to complicate PPROM at earlier gestational ages and can occur either before or after PPROM.¹⁷ Malpresentation is also common following PPROM, especially at early gestational ages, and it is associated with an increased risk of umbilical cord prolapse. Umbilical cord compression due to oligohydramnios resulting in fetal heart rate (FHR) decelerations is common. Due the risk of these complications, there is an increased risk of nonreassuring fetal status requiring delivery, and approximately 1% of pregnancies complicated by PPROM result in stillbirth.¹⁸

The neonatal morbidities related to PPROM are primarily related to prematurity and include respiratory distress syndrome, intraventricular hemorrhage, and necrotizing enterocolitis. When compared to outcomes of infants with intact membranes matched for gestational age, the neonatal outcomes are generally similar.¹⁹ However, intrauterine inflammation is more common in PPROM and is associated with increased risk of neonatal central nervous system (CNS) damage and neurodevelopmental impairment. Prolonged anhydramnios, especially at <22 weeks gestation, is associated with a risk of pulmonary hypoplasia, skeletal contractures, and Potter-like facies, as discussed later in this chapter.

DIFFERENTIAL DIAGNOSIS

The differential diagnosis for women presenting with the complaint of leakage of fluid from the vagina includes PROM, leukorrhea of pregnancy, vaginal discharge related to cervicitis or vaginitis, urinary incontinence, transudate in the setting of a dilated cervix with exposed membranes, and perspiration. Many women experience a physiologic, thin, and milky vaginal discharge as early as the first trimester, and leukorrhea can increase in volume as the pregnancy progresses. A disruption in the normal vaginal flora, such as yeast vaginitis or bacterial vaginosis, can manifest as an increase in vaginal discharge. Cervicitis caused by Chlamydia trachomatis, Neisseria gonorrhoeae, or Trichomonas vaginalis may result in purulent or watery vaginal discharge that can be confused with amniotic fluid. As the gravid uterus grows and applies pressure to the maternal bladder, the development of stress urinary incontinence is common; it has been reported to occur at rates as high as 48% and 67% of nulliparous and parous women, respectively.²⁰

EVALUATION

The initial evaluation includes obtaining a detailed history and performing a physical exam; often, diagnosis of ruptured membranes can be established with a history and physical exam alone. PROM often presents as a large gush of clear fluid, followed by ongoing, involuntary passage of amniotic fluid from the vagina. However, some women only note a feeling of wetness on the perineum, which may be constant or intermittent. A sterile speculum exam allows visualization of amniotic fluid pooling in the vagina or passage of fluid through the cervix. If no fluid is observed, the patient can be asked to cough or perform the Valsalva maneuver, resulting in the passage of amniotic fluid from the uterus through the cervix. A speculum exam also enables the assessment of cervical dilation and umbilical cord prolapse as well as the collection of chlamydia, gonorrhea, and Group B Streptococcus cultures. In the setting of a dilated cervix with preterm labor or cervical insufficiency, there may be a notable amount of transudate in the vagina and the consequent risk of a false-positive diagnosis of PPROM. Unless there is concern that active labor or delivery is imminent, digital vaginal exam should be avoided in order to decrease the risk of introducing infection.^21,22,23

Determining or confirming gestational age and evaluation of fetal well-being is of primary importance. The initial evaluation should include electronic fetal monitoring and tocodynamometry to assess fetal well-being and to check for the presence of contractions. An ultrasound is indicated to determine fetal presentation, confirm gestational age, estimate fetal weight, and assess amniotic fluid volume. We routinely obtain a complete blood count (CBC), blood type and screen, rapid plasma regain (RPR) test, urinalysis, and other prenatal labs as clinically indicated; for high-risk patients, we also obtain rapid human immunodeficiency virus (HIV). If expectant management is being considered or pursued, we generally monitor the fetus continuously during the first 24 hours of labor and delivery (L&D) and then consider transferring the patient to the antepartum unit.

DIAGNOSIS

Standard ancillary tests to establish or confirm the diagnosis of ruptured membranes include assessing the pH of the vaginal fluid and evaluating for the ferning of dried vaginal fluid under a microscope. The pH of amniotic fluid is generally in the range of 7.0 to 7.3, as opposed to the more acidic environment of the vagina or urine, with a pH of 4.5 to 6.0.²⁴ The presence of blood, semen, bacterial vaginosis, or other alkaline contaminants can cause false-positive results. A microscope slide can be coated with vaginal fluid and allowed to air-dry; amniotic fluid crystallizes and branches in a typical branching pattern (see Fig. 46-1). This arborization must be differentiated from the larger, coarser ferning that results from cervical mucus (see Fig. 46-2). If the amount of fluid is scant, any of these tests may be falsely negative, and the diagnosis can be more difficult to establish. If the diagnosis is unclear upon conclusion of the initial exam, generally the patient should remain in bed and the exam should be repeated again after a period of observation.

Other complementary tests are available to help establish or exclude the diagnosis of ruptured membranes. A bedside sonogram can be performed to estimate the amniotic fluid volume. The finding of anhydramnios or oligohydramnios on ultrasound can be suggestive of ruptured membranes. However, the amniotic fluid volume can be normal with PPROM, and there are other etiologies of decreased or absent amniotic fluid. FFN testing can be helpful in select cases. While generally used to evaluate for preterm labor, a negative result is strongly suggestive of intact membranes, with the sensitivity for diagnosing PPROM being 98.2%.²⁵

Multiple commercially available tests marketed for the diagnosis of ruptured membranes are currently available (see Table 46-1). They test for various proteins, such as placental alpha microglobulin-1 (PAMG-1), insulinlike growth factor binding protein 1 (IGFBP-1), and alpha-fetoprotein (AFP). AmniSure is a point-of-care test that evaluates for the presence of PAMG-1, a protein highly concentrated in amniotic fluid. PAMG-1 immunoassay has been reported to have a sensitivity of 98.7% to 98.9%, specificity of 87.5% to 100%, positive predictive value (PPV) of 98.1% to 100%, and negative predictive value (NPV) of 91.3% to 99.1%.^26,27 However, caution needs to be used with this assay in women with symptoms of labor, as the false-positive rate in such patients with intact membranes has been found to be 13.5% to 30%.^28,29 Actim Prom is a point-of-care test to detect IGFBP-1; this test is widely available in Europe, but uncommonly used in the United States. It can be used without interference from blood, semen, urine, disinfectant solutions, medicines, or labor.^30,31

Actim Prom has a reported sensitivity of 96.8% to 99%^32,33 and a specificity of 82.7%.³⁴ The ROM Plus test identifies the dual presence of IGFBP-1 and AFP. This monoclonal/polyclonal-antibody approach has been reported to result in fewer false-negative results than other point-of-care tests (1% for IGFBP-1 and AFP combined, as opposed to 3% for IGFBP-1 alone and 4% for PAMG-1).³² The studies evaluating these commercially available kits are limited in number and typically have small sample sizes. Cost is an additional limitation of these tests. In our opinion, the use of supplemental tests should be limited to equivocal cases in which establishing the correct diagnosis is difficult.

The gold standard for diagnosing ruptured membranes traditionally has been ultrasonographically guided, transabdominal instillation of indigo carmine into the uterine cavity, followed by evaluation for leakage of blue dye from the vagina. However, at the time of this writing, there is a shortage of this dye, with no suitable substitute available.

MANAGEMENT

The treatment of PPROM depends on gestational age and the presence or absence of complications such as clinical chorioamnionitis, nonreassuring fetal testing, placental abruption, and preterm labor. Among the first decisions to be made is whether to proceed with prompt delivery or pursue expectant management. Accurate assessment of pregnancy dating is critical, and the risks of prematurity must be balanced against those of expectant management. After viability, inpatient management is recommended due to the high risk of acute complications that may require emergent delivery.

CONTRAINDICATIONS TO EXPECTANT MANAGEMENT

Not all women presenting with PPROM are acceptable candidates for expectant management. The gestational ages at which expectant management is appropriate will be further discussed later in this chapter. Stillbirth, umbilical cord prolapse, or nonreassuring FHR tracing are indications for prompt delivery. As discussed previously, chorioamnionitis is a common complication of PPROM. Due to the risk of maternal sepsis and its potential for serious morbidity and mortality, chorioamnionitis is considered an absolute indication for delivery regardless of gestational age. In the setting of PPROM with placental abruption, delivery should be considered based on fetal status, gestational age, and the severity of the abruption. When abruption occurs at a previable gestational age or it is associated with significant bleeding, severe pain, or FHR decelerations, expectant management should not be pursued. While preeclampsia, placenta previa with bleeding, fetal growth restriction, and other obstetric or medical comorbidities are not contraindications to expectant management, the threshold for delivery is lower in these situations.

EXPECTANT MANAGEMENT

The decision to proceed with expectant management of PPROM and what adjunctive treatments are utilized largely depends on gestational age. After viability, which is determined by a combination of gestational age and other factors such as the receipt of corticosteroids, birth weight, parental preferences, and institutional guidelines, we admit women to the hospital for expectant management until delivery. There is no consensus regarding the optimal method or frequency of maternal or fetal assessment. Our approach includes daily assessment of the maternal symptoms with an abdominal exam and monitoring of vital signs at least 2 to 3 times daily. Serial evaluations of white blood cell counts and markers of inflammation have not been demonstrated to be of clinical utility.³⁵ With regard to fetal assessment, after an initial 24 hours of continuous electronic fetal monitoring, we monitor the fetus for 1 hour twice a day and consider further monitoring as clinically indicated. Ultrasounds to evaluate fetal growth are performed every 3 to 4 weeks, and we perform ultrasounds to evaluate fetal presentation and amniotic fluid volume as clinically indicated.

34–37 Weeks

The management of PPROM between 34 and 37 weeks gestation has been and remains controversial, despite the attempt of three recently conducted large trials to address this issue.^36,37,38 The American College of Obstetricians and Gynecologists (ACOG) recommends delivery in the setting of PPROM at or beyond 34 weeks gestation.¹⁰ In 2012, two studies were published by the PPROM Expectant Management versus Induction of Labor (PPROMEXIL) trial group, comparing expectant management to delivery in pregnancies complicated by PPROM between 34 and 37 weeks.^36,37 The authors concluded that induction of labor was not associated with a reduction in the rate of neonatal sepsis and did not substantially improve outcomes. The only significant difference in outcomes was that delivery was associated with lower rates of chorioamnionitis.

The largest and most adequately powered trial to date is Preterm Prelabor Rupture of Membranes close to Term (PPROMT), which studied 1839 women in 65 centers across 11 countries.³⁸ This study compared induction of labor to expectant management between 34 weeks and 36 6/7 weeks, and it was the first to demonstrate that expectant management was associated with decreased rates of neonatal respiratory distress, reduced need for mechanical ventilation, shorter neonatal intensive care unit (NICU) stay, and lower risk of cesarean delivery. However, expectant management was linked to a higher rate of antepartum hemorrhage, and similar to findings in other studies, it also was associated with increased rates of intrapartum fever.³⁸ There was no significant difference in the rate of neonatal sepsis. Limitations of the PPROMT trial include that it took place over a nearly 10-year period and in many countries, so its findings are limited in terms of their applicability to contemporary practices in the United States.

Based on these results, our conclusion is that the optimal management remains unclear and is best summarized as follows: Induction of labor is associated with a reduction in the rate of chorioamnionitis and the potential to avoid other complications associated with expectant management such as stillbirth, although these risks appear to be low. Also, expectant management may be associated with reduced rates of respiratory morbidity, shorter NICU admission, and possibly a lower rate of cesarean delivery. Our general approach is consistent with the ACOG recommendations; however, expectant management at 34 to 36 weeks gestation is reasonable in select patients after appropriate counseling.

Until recently, antenatal corticosteroids were not administered to women with PPROM after 34 weeks gestation. However, a recent, multicenter randomized controlled trial (RCT) showed benefit of a single course of corticosteroids between 34 and 36 5/7 weeks gestation.³⁹ Of the 2831 women studied, 620 (21.9%) had ruptured membranes, and although outcomes in this subgroup were not specifically analyzed, it is reasonable to conclude that patients with PPROM benefit from late preterm administration of corticosteroids. Exclusion criteria in this study included having previously received corticosteroids, pregestational diabetes, and delivery expected within 12 hours. Latency antibiotics were not routinely given, and delivery was not delayed to allow completion of the course of corticosteroids. If pursuing expectant management, administration of latency antibiotics is reasonable, and this was the approach used in the PPROMT trial. We recommend against tocolysis after 34 weeks gestation.

23–34 Weeks

Expectant management is generally pursued between 23 and 34 weeks gestation unless there is an indication for immediate delivery. Adjunctive treatments that can be considered include antenatal corticosteroids, latency antibiotics, magnesium sulfate for neurologic protection, and tocolysis.

Multiple RCTs have demonstrated that latency antibiotics are associated with prolongation of pregnancy and improved maternal and neonatal outcomes.⁴⁰ Antibiotic regimens should provide coverage against the most common clinical pathogens that can cause chorioamnionitis, such as Group B Streptococcus and beta-lactamase-producing aerobes and anaerobes. While multiple regimens have shown benefits, the optimal regimen to reduce maternal and neonatal complications is unclear. The National Institute of Child Health and Human Development (NICHD) Maternal-Fetal Medicine Units Network trial on antibiotic therapy for reduction of infant morbidity after PPROM used intravenous (IV) ampicillin 2 g every 6 hours and erythromycin 250 mg every 6 hours for 48 hours, followed by oral amoxicillin 250 mg every 8 hours and erythromycin 333 mg every 8 hours for 5 days.¹⁴ This is the regimen recommended by ACOG.¹⁰

Alternatively, azithromycin can be substituted for erythromycin, resulting in fewer gastrointestinal (GI) side effects, lower cost, and similar pregnancy and neonatal outcomes.^41,42 We give azithromycin 500 mg IV daily for 48 hours, followed by 250 mg PO daily for 5 days. For women who have a non-life-threatening allergy to penicillin, we generally substitute IV cefazolin 1 g IV every 8 hours, followed by oral cephalexin 500 mg every 6 hours for 5 days for ampicillin and amoxicillin. For women with a serious, anaphylactic reaction to penicillin, we generally give clindamycin and gentamicin for broad coverage; ACOG suggests using erythromycin alone.¹⁰ Of note, all patients who are carriers of Group B Streptococcus should receive intrapartum chemoprophylaxis, regardless of any previous antibiotic therapy.¹⁰

Administration of corticosteroids to women with PPROM between 23 and 34 weeks gestation has been shown to reduce overall neonatal mortality, respiratory complications, necrotizing enterocolitis, and intraventricular hemorrhage without increasing the risk of maternal or neonatal sepsis.^43,44 The recommended regimen is 2 doses of 12 mg intramuscular (IM) betamethasone 24 hours apart, or 4 doses of 6 mg IM dexamethasone every 12 hours.⁴⁴ Currently, there is insufficient data to make a strong recommendation for or against the administration of repeat or rescue steroid courses.¹⁰