Key Abbreviations
By mouth (per os) PO
Confidence interval CI
Fetal fibronectin fFN
Group B Streptococcus GBS
Herpes simplex virus HSV
Human immunodeficiency virus HIV
Insulin-like growth factor–binding protein 1 IGFBP-1
Intramuscular IM
Intravenous IV
Intraventricular hemorrhage IVH
Lamellar body count LBC
Maternal-Fetal Medicine Unit MFMU
Matrix metalloproteinase MMP
National Institute of Child Health and Human Development NICHD
Neonatal intensive care unit NICU
Odds ratio OR
Periventricular leukomalacia PVL
Phosphatidylglycerol PG
Placental α-microglobulin 1 PAMG-1
Premature rupture of the membranes PROM
Respiratory distress syndrome RDS
Tissue inhibitors of matrix metalloproteinase TIMP
U.S. Food and Drug Administration FDA
Membrane rupture that occurs spontaneously before the onset of labor is described as premature rupture of the membranes (PROM) regardless of the gestational age at which it occurs. PROM complicates about 8% to 10% of pregnancies. Preterm PROM that occurs before 37 weeks’ gestation affects about 1% of deliveries overall, and birth certificate data suggest that it is over twofold more common in blacks. Like preterm labor and cervical insufficiency, PROM is considered a cause of spontaneous preterm birth. The relative contribution of PROM to prematurity appears to vary greatly among patient populations, affecting about 10% of preterm births in national databases but over 20% in certain high-risk populations. Its frequency appears to have declined during the past decade.
PROM at any gestational age is associated with brief latency from membrane rupture to delivery and also increased risks for perinatal infection and umbilical cord compression due to oligohydramnios. Because of this, term and preterm PROM are significant causes of perinatal morbidity and mortality. When PROM occurs at term, the risk is low for severe neonatal complications with delivery of a noninfected and nonasphyxiated infant. Clinical management should be directed toward delivery. Although complications can occur, delivery at 32 to 36 weeks’ gestation is generally associated with good infant outcomes, particularly if the fetus has documented pulmonary maturity. Given the risks of continued pregnancy and anticipated brief latency, delivery of the mature fetus is generally warranted, particularly at 34 weeks’ gestation or later. At 32 to 33 weeks’ gestation, the immature fetus may benefit from measures to accelerate fetal maturation and to prolong the pregnancy. With immediate delivery after preterm PROM at 23 to 31 weeks’ gestation, the risk is significant for newborn complications that can be reduced through adequate delay of delivery. In the absence of contraindications, management is directed toward continuing the pregnancy with attention to potential complications that include umbilical cord compression, intrauterine infection, and abruptio placentae. When PROM occurs before the limit of viability, newborn death is inevitable with immediate delivery. Although conservative management may still result in a previable delivery, some women will benefit from extended latency with delivery of a potentially viable infant. Regardless of the gestational age, the patient should be well informed regarding the potential maternal, fetal, and neonatal complications of PROM and preterm birth. These issues are discussed in detail in this chapter.
Fetal Membrane Anatomy and Physiology
The fetus develops within the amniotic sac, which is surrounded like a balloon by the fetal membranes. These membranes consist of a thin amnion layer that lines the amniotic cavity and a thicker outer chorion directly apposed to the maternal decidua. The amnion fuses to the chorion near the end of the first trimester of pregnancy, and these layers are subsequently attached by a collagen-rich connective tissue zone. For the remainder of the pregnancy, the fetal membranes include a single cuboidal amnion epithelium with subjacent compact and spongy connective tissue layers and a thicker chorion that consists of reticular and trophoblastic layers. Together, the amnion and chorion are stronger than either layer independently; individually, the amnion has greater tensile strength than the chorion.
As the pregnancy progresses, changes in collagen content and type, intercellular matrix, and cellular apoptosis result in structural weakening of the fetal membranes. Membrane remodeling is more evident near the internal cervical os and can be stimulated by thrombin-mediated increases in matrix metalloproteinases (e.g., MMP-1, MMP-2, MMP-9) and decreased levels of tissue inhibitors of matrix metalloproteinases (e.g., TIMP-1, TIMP-3) within the membranes, as well as increased poly (ADP-ribose) polymerase (PARP) cleavage. Contractions subject the amniochorionic membranes to additional physical strain that can lead to membrane rupture. Should the fetal membranes not rupture before labor, advancing cervical dilation decreases the work needed to cause membrane rupture over the internal cervical os. Preterm PROM likely results from a variety of factors that ultimately lead to accelerated membrane weakening through an increase in local cytokines and an imbalance in the interaction between MMPs and TIMPs, increased collagenase and protease activity, or other factors that cause increased intrauterine pressure (e.g., polyhydramnios).
Etiology of Premature Rupture of the Membranes
A number of risk factors have been associated with the occurrence of preterm PROM. Among these are low socioeconomic status, uterine overdistension, second- and third-trimester bleeding, low body mass index (BMI), nutritional deficiencies of copper and ascorbic acid, maternal cigarette smoking, cervical conization or cerclage, pulmonary disease in pregnancy, connective tissue disorders (e.g., Ehlers-Danlos syndrome), and preterm labor or symptomatic contractions in the current gestation. Each risk factor, individually or in concert, could lead to PROM through the mechanisms outlined above. However, the ultimate clinical cause of membrane rupture is often not apparent, and many at-risk patients will deliver at term without PROM.
Preterm PROM has also been linked to infections that involve the urogenital tract. Neisseria gonorrhoeae, Chlamydia trachomatis, and Trichomonas vaginalis have each been associated with preterm PROM. Although vaginal group B β-hemolytic Streptococcus (GBS) colonization does not appear to be associated with preterm PROM, cervical colonization may be. GBS bacteriuria is associated with preterm PROM and low-birthweight infants. Although bacterial vaginosis has been linked to spontaneous preterm births, including preterm PROM, it is unclear whether bacterial vaginosis is the inciting condition that facilitates ascent of other bacteria to the upper genital tract, or whether it is simply a marker of maternal susceptibility to abnormal genital tract colonization. Bacterial invasion can facilitate membrane rupture through direct release of proteases and also through stimulation of a host inflammatory response that results in the elaboration of local cytokines, MMPs, and prostaglandins. Histologic studies of the membranes after preterm PROM often demonstrate significant bacterial contamination along the choriodecidual interface with minimal involvement of the amnion. Further evidence that links preterm PROM and genital tract infection is that these women have a high incidence of positive amniotic fluid cultures (25% to 35%) even in the absence of clinically suspected intrauterine infection. Although some of these findings may reflect ascending infection subsequent to membrane rupture, it is probable that ascending bacterial colonization and infection are integral to the pathogenesis of preterm PROM in many cases.
Although the onset of vaginal fluid leakage is an acute event, evidence shows that the factors and events that lead to membrane rupture are sometimes subacute or even chronic. Women with a prior preterm birth (PTB), especially because of PROM, are at increased risk for PTB due to PROM in future pregnancies. Studies have also suggested associations exist between maternal inflammatory proteins, genotype, and spontaneous preterm birth (sPTB) due to preterm labor or PROM. Further, asymptomatic women with a short cervical length in the second trimester are at increased risk for preterm PROM occurring many weeks later.
Prediction and Prevention of Preterm Premature Rupture of the Membranes
Once preterm PROM occurs, delivery is often required or inevitable. Optimally, prevention of PROM would offer the best opportunity to avoid its complications. Prior PTB and especially prior preterm PROM (PPROM) have been associated with PPROM in a subsequent pregnancy. The risk of recurrence increases with decreasing gestational age of the index PTB. Those with a prior delivery near the limit of viability (23 to 27 weeks) have a 27.1% risk of subsequent PTB. Those with a prior history of PTB due to PROM have a 3.3-fold higher risk for PTB due to PROM (13.5% vs. 4.1%) and a 13.5-fold higher risk for PPROM before 28 weeks’ gestation (1.8% vs. 0.13%) in a subsequent pregnancy ( P < .01 for each). In an analysis from a prospective evaluation of PTB prediction, nulliparas and women with prior deliveries were evaluated separately because those without a prior birth lacked important historic information available to those with a prior term or preterm birth. In that study, multivariable analysis revealed medical complications (including pulmonary disease in pregnancy), work during pregnancy, recent symptomatic uterine contractions, and bacterial vaginosis to be significant markers for subsequent PTB in nulliparas when assessed at 22 to 24 weeks’ gestation ( Table 30-1 ). Among women with prior deliveries, prior PTB due to preterm labor or PROM and a positive cervicovaginal fetal fibronectin (fFN) screen were statistically significant clinical markers for subsequent PPROM after controlling for other factors. Short cervical length (<25 mm) identified by transvaginal ultrasound and low maternal BMI (<19.8 kg/m 2 ) were associated with an increased risk for subsequent PROM in both nulliparas and multiparas. Nulliparas with a positive cervicovaginal fFN and a short cervix had a 16.7% risk for PTB due to PPROM. Among multiparas, women with a prior PTB due to PROM, a short cervix on ultrasound, and positive cervicovaginal fFN screen had a 31-fold higher risk for PROM with delivery before 35 weeks’ gestation (25% vs. 2.3%) than those without risk factors ( Table 30-2 ).
| NULLIPARAS ( N = 1618) | MULTIPARAS ( N = 1711) | |
|---|---|---|
| Medical complications | 3.7 (1.5-9.0) | — |
| Work in pregnancy | 3.0 (1.5-6.1) | — |
| Symptomatic contractions within 2 weeks | 2.2 (1.2-7.5) | — |
| Bacterial vaginosis | 2.1 (1.1-4.1) | — |
| Low BMI (<19.8 kg/m 2 ) | 2.0 (1.0-4.0) | 1.8 (1.1-3.0) |
| Prior preterm birth due to PROM | — | 3.1 (1.8-5.4) |
| Prior preterm birth due to preterm labor | — | 1.8 (1.1-3.1) |
| Cervix <25 mm | 3.7 (1.8-7.7) | 2.5 (1.4-4.5) |
| Positive fetal fibronectin | — | 2.1 (1.1-4.0) |
* Results of multivariable analyses for nulliparas and multiparas (presented as odds ratios with 95% confidence intervals).
| N | <37 WEEKS (%) | <35 WEEKS (%) | |
|---|---|---|---|
| All multiparas | 1711 | 5.0 | 2.3 |
| No risk factors present | 1351 | 3.2 | 0.8 |
| Prior preterm birth due to PROM only | 124 | 10.5 | 4.8 |
| Prior preterm birth due to PROM and positive fFN * | 13 | 15.4 | 15.4 |
| Prior preterm birth due to PROM and short cervix † | 26 | 23.1 | 15.4 |
| All three risk factors present | 8 | 25.0 | 25.0 |
* Positive fFN, cervicovaginal fFN screen positive (>50 ng/mL) at 22 to 24 weeks’ gestation.
† Short cervix, cervix length <25 mm on transvaginal ultrasound at 22 to 24 weeks’ gestation.
Unfortunately, clinical risk-assessment systems identify only a small fraction of women who will ultimately deliver preterm. Although clinical and ancillary testing has increased our ability to identify women at increased risk because of potentially modifiable factors—such as cigarette smoking, poor nutrition, urinary tract and sexually transmitted infections, pulmonary disease, and severe polyhydramnios—it is unknown whether modification of these in a given patient will reduce the risk for PROM. Regardless, women at risk for PTB due to PROM based on clinical findings can be counseled regarding the symptoms of membrane rupture and contractions and can be encouraged to seek medical care should symptoms occur. Regarding ancillary testing, progesterone therapy has been recommended for prevention of PTB in asymptomatic women with a short cervix, and thus cervical length (CL) screening has potential value for both predictive and therapeutic reasons. Alternatively, routine fFN screening has similar predictive value to CL measurement, but no effective intervention can be offered based on the results. Thus routine fFN testing after PTB due to PROM is not recommended. Current evidence supports 17-α-hydroxyprogesterone caproate (17-P) treatment for women with a prior PTB due to PROM or preterm labor and also supports treatment with vaginal progesterone for asymptomatic women with a short cervical length. Data regarding the value of vitamin C supplementation in preventing PROM are conflicting and are not generally supportive. In one study, such treatment was associated with a lower risk (7.7% vs. 24.5%; P = .02). Secondary analysis of another study suggested that treatment with vitamin C and E did not reduce sPTB or late preterm birth due to PROM but was associated with less frequent PTB due to PROM before 32 weeks’ gestation. However, a review of studies in which vitamin C was given alone or in combination with other supplements suggests a negative impact on membrane strength and an increased risk for PTB. Because of these findings, vitamin C supplementation to prevent PPROM is not currently recommended.
Clinical Course after Premature Rupture of the Membranes
Maternal Risks
Hallmarks of PROM include a brief latency from membrane rupture to delivery. On average, latency increases with decreasing gestational age at membrane rupture. At term, half of expectantly managed gravidas deliver within 33 hours, and 95% delivered within 94 to 107 hours of membrane rupture. Of all women with PROM before 34 weeks, 93% deliver in less than 1 week. After excluding those who require delivery soon after admission, 50% to 60% of those conservatively managed and treated with antibiotics for pregnancy prolongation will deliver within 1 week of membrane rupture. Only a small proportion of women with membrane rupture (≤5%) can anticipate cessation of fluid leakage. About 86% of those with leakage after amniocentesis will reseal.
Risks of Premature Rupture of the Membranes
Maternal Risks
Chorioamnionitis is the most common maternal complication after PPROM. This risk increases as the duration of membrane rupture becomes more prolonged and decreases with advancing gestational age at PROM. The risks of chorioamnionitis and endometritis increase with decreasing gestational age at PROM and also in different patient populations (13% to 60% for chorioamnionitis and 2% to 13%, for endometritis). Abruptio placentae can cause PROM or can occur subsequent to membrane rupture, and it affects 4% to 12% of these pregnancies. Uncommon but serious complications of PROM managed conservatively near the limit of viability include retained placenta and hemorrhage, requiring dilation and curettage (12%); maternal sepsis (0.8%); and maternal death (0.14%).
Fetal and Neonatal Risks
Fetal complications after membrane rupture include infection and fetal distress due to umbilical cord compression or placental abruption. Umbilical cord compression due to oligohydramnios is not uncommon after PROM. Frank or occult umbilical cord prolapse can also occur, particularly with fetal malpresentation. Because of these factors, women with PROM have a higher risk for cesarean delivery for nonreassuring fetal heart rate (FHR) patterns than those with isolated preterm labor (7.9% vs. 1.5%). Fetal death complicates 1% to 2% of cases of conservatively managed PROM.
The frequency and severity of neonatal complications after PROM vary inversely with gestational age at membrane rupture and at delivery. Respiratory distress syndrome (RDS) is the most common serious newborn complication after PPROM at any gestational age. Necrotizing enterocolitis (NEC), intraventricular hemorrhage (IVH), and sepsis are common with early PTB but are relatively uncommon when PPROM and delivery occur near term. Serious perinatal morbidities with delivery remote from term can lead to long-term sequelae such as chronic lung disease (CLD), visual or hearing difficulties, intellectual disabilities, developmental and motor delay, cerebral palsy, and death. Although specific data are not available for those who deliver after PPROM, general community-based survival and morbidity data suggest that long-term morbidities and death are uncommon with delivery after about 32 weeks’ gestation. It is controversial whether gestational age–specific mortality is increased for preterm infants who deliver after PPROM.
PPROM increases the risk of neonatal sepsis twofold over that seen after PTB due to preterm labor with intact membranes. Neonatal infection can result from the same organisms present in the amniotic fluid, or from others, and it can present as acute congenital pneumonia, sepsis, or meningitis. Late-onset bacterial or fungal infections can also occur. Accumulating evidence suggests that fetal and neonatal infection and inflammation are associated with an increased risk for long-term neurologic complications. Cerebral palsy and cystic periventricular leukomalacia (PVL), as well as cognitive impairment and death and neurodevelopmental impairment in extremely preterm infants, have been linked to chorioamnionitis, which is more commonly seen after PPROM and is more likely with conservative management after membrane rupture. Elevated amniotic fluid cytokines and fetal systemic inflammation have also been associated with PPROM, PVL, and cerebral palsy. Although no data suggest that immediate delivery on admission with PROM will avert these sequelae, these findings highlight the importance of restricting conservative management after PROM to circumstances in which there is the potential to reduce neonatal morbidity through either antenatal corticosteroid administration or extended pregnancy prolongation for fetal maturation.
Pulmonary hypoplasia is a severe complication of oligohydramnios in the second trimester that results from a lack of terminal bronchiole and alveolar development during the canalicular phase of pulmonary development. It is most accurately diagnosed pathologically using radial alveolar counts and lung weights. Clinical findings, such as a small chest circumference with severe respiratory distress and persistent pulmonary hypertension in the newborn, and radiographic findings—small, well-aerated lungs with a bell-shaped chest and elevation of the diaphragm—are also supportive of the diagnosis. Whether because of fluid efflux and tracheobronchial collapse after membrane rupture or through loss of intrinsic factors within the tracheobronchial fluid, pulmonary hypoplasia develops over weeks after membrane rupture. Pulmonary hypoplasia complicated an average of about 6% of cases in series of midtrimester PROM and carries a 70% mortality rate. Its incidence is inversely correlated with gestational age at membrane rupture, and it complicates nearly 50% of cases with membrane rupture before 19 weeks and prolonged latency. The frequency of pulmonary hypoplasia can be as high as 74% to 82% with PROM at 15 to 16 weeks, persistent oligohydramnios, and a latency of 28 days. Lethal pulmonary hypoplasia rarely occurs with PROM after 26 weeks’ gestation (0% to 1.4%). However, other pulmonary complications such as pneumothorax and pneumomediastinum related to poor pulmonary compliance and high ventilatory pressures can occur with lesser degrees of this condition. Restriction deformities occur in about 1.5% of infants delivered after conservative management after midtrimester PROM but complicate up to 27% of fetuses with prolonged oligohydramnios.
Diagnosis of Premature Rupture of the Membranes
The diagnosis of PROM involves clinical history and physical examination as well as laboratory evaluation in some cases.
The diagnosis of membrane rupture is confirmed by the presence of the following findings:
- •
Visualization of amniotic fluid passing from the cervical canal or
- •
Vaginal sidewall or posterior fornix pH of more than 6.0 to 6.5 and
- •
Microscopic arborized crystals (“ferning”), owing to the interaction of amniotic fluid proteins and salts, from dried vaginal secretions obtained by swabbing the posterior fornix with a sterile swab.
False-positive pH results may occur with blood or semen contamination, alkaline antiseptics, or bacterial vaginosis. Cervical mucus can yield a false-positive ferning pattern; however, the crystals appear as more of a floral pattern. The fern pattern in samples heavily contaminated with blood is atypical and appears more “skeletonized.” Prolonged leakage with minimal residual fluid can result in a false-negative result on visual inspection or pH or ferning testing. If the diagnosis is equivocal after initial testing, the patient can be placed in a Trendelenburg position and reexamined after a few hours. The diagnosis of membrane rupture can be made unequivocally by ultrasound-guided dye amnioinfusion (1 mL indigo carmine plus 9 mL sterile saline), followed by observation for passage of dye onto a perineal pad. Although oligohydramnios without evident fetal urinary tract malformations or fetal growth restriction may be suggestive of membrane rupture, ultrasound alone is not definitive.
Noninvasive cervicovaginal markers such as fFN, alpha-fetoprotein, prolactin, human chorionic gonadotropin (hCG), placental α-microglobulin 1 (PAMG-1), and insulin-like growth factor–binding protein 1 (IGFBP-1) have been studied for their ability to confirm or exclude membrane rupture, but most are unavailable for clinical use. Such testing is unneeded when the diagnosis is confirmed clinically. Further, although membrane rupture can be confirmed by the presence of PAMG-1 in cervicovaginal secretions, and the test’s accuracy is relatively unaffected by the presence of blood, it has also been found to be present in nearly one third of laboring women and in 1 of 20 nonlaboring women without suspected membrane rupture.
Management of Premature Rupture of the Membranes
General Considerations
Management of PROM is based primarily on an individual assessment of the estimated risk for fetal and neonatal complications should conservative management or delivery be pursued. The risks for maternal morbidity should also be considered, particularly when PROM occurs before the limit of potential viability (currently 23 weeks’ gestational age). Regional factors may impact the potential risks and benefits of conservative management. In populations where the risk of intrauterine infection is high and the potential for extended latency without complications is low, the focus will tend to be on acceleration of fetal maturation, prevention of intrauterine infection, and delivery if fetal benefit from prolonged latency is not anticipated. Alternatively, in populations at low risk for intrauterine infection and with higher potential for prolonged latency, conservative management may be appropriate at a more advanced gestational age.
The diagnosis of membrane rupture is confirmed and the duration of membrane rupture is determined to assist the pediatric caregivers with subsequent management decisions. The patient is assessed for fetal presentation, contractions, findings suggestive of intrauterine infection, and evidence of fetal well-being. GBS carriage is ascertained, if available, from a recent anovaginal culture performed within the previous 5 weeks.
In general, digital cervical examinations should be avoided until it is determined that delivery is inevitable because such examination has been associated with a shortening of latency from membrane rupture to delivery. Visualization of the cervix during a sterile speculum examination offers helpful information regarding cervical dilation and effacement. Brown and colleagues found visual estimation to be within 1 cm of digitally determined cervical dilation in 64% and within 2 cm in 84% of examinations, whereas visually estimated cervical effacement was within 1 cm in 83% of cases. In addition to providing confirmatory evidence of membrane rupture, sterile speculum examination can provide the opportunity to inspect for cervicitis and to obtain appropriate cervical and vaginal cultures.
The benefit of narrow-spectrum intrapartum prophylaxis with intravenous (IV) penicillin G (5 million U and then 2.5 to 3 million U every 4 hr) or ampicillin (2 g IV, then 1 g IV every 4 hr) to prevent vertical transmission and early-onset neonatal GBS sepsis from maternal GBS carriers has been well established, and the Centers for Disease Control and Prevention (CDC) published a revised guideline for the prevention of perinatal group B streptococcal disease in November 2010. Current indications for intrapartum GBS prophylaxis and alternative antibiotic regimens for those with a penicillin allergy are discussed in Chapter 54 . Known GBS carriers with PROM at any gestation and those with PPROM and unknown GBS status should receive intrapartum prophylaxis regardless of prior antibiotic treatment. GBS carriers with PROM and chorioamnionitis should receive broad-spectrum intrapartum antibiotic therapy, including agents effective against GBS. If chorioamnionitis is not suspected clinically and there is a recent negative anovaginal culture for GBS, intrapartum antibiotics should not be administered because of the potential for selection of resistant organisms should neonatal sepsis occur.
Although practice varies regarding the management of PPROM, general consensus has been reached in regard to some issues. Gestational age should be established based on clinical history and earliest ultrasound assessment where available ( Fig. 30-1 ). Ultrasound should be performed to assess fetal growth and position, residual amniotic fluid volume, and gross fetal abnormalities that might cause polyhydramnios and PROM. Those with advanced labor, intrauterine infection, significant vaginal bleeding, or nonreassuring fetal testing are best delivered. If conservative management of PPROM is to be pursued, the patient should be admitted to a facility capable of providing emergent delivery for placental abruption, fetal malpresentation in labor, and fetal distress due to umbilical cord compression or in utero infection. The facility should also be capable of providing 24-hour neonatal resuscitation and intensive care because conservative management should generally be performed only when significant risk for neonatal morbidity and mortality is present. If the need for transfer to a tertiary care facility is anticipated, this should occur early in the course of management to avoid emergent transfer once delivery is imminent or complications arise.
