Prematurity affects multiple organ systems, resulting in both immediate neonatal and sometimes lifelong adverse health sequelae. Though some infants born late preterm do not require intensive care, the majority of preterm infants require admission to the neonatal intensive care unit (NICU) for management of prematurity. The earlier the gestational age at delivery, the longer the anticipated duration of NICU admission. The economic costs for the care of surviving preterm infants (especially very low-birthweight infants) can be enormous, and the annual US healthcare costs of medical care for preterm infants are estimated at approximately $25 billion.

Respiratory immaturity is a common complication of prematurity, with risk of respiratory distress syndrome, bronchopulmonary dysplasia, apneic episodes, and need for mechanical ventilation. Smaller brain volume, immaturity of neurologic structures, and altered sensory experiences related to NICU admission may have profound neurodevelopmental consequences in preterm infants, including cerebral palsy, intraventricular hemorrhage, retinopathy of prematurity, hearing impairment, and delayed motor and speech development. Additional neonatal sequelae of prematurity include feeding immaturity, hypoglycemia, hyperbilirubinemia, anemia, thrombocytopenia, persistent patent ductus arteriosus, pulmonary hypertension of the newborn, necrotizing enterocolitis, and heightened risk of sepsis.

The risks for severe morbidity and mortality decrease with advancing gestational age at delivery. For example, the incidence of high-grade (III or IV) intraventricular hemorrhage diminishes rapidly after 27 weeks’ gestation, rarely occuring after 32 weeks’ gestation. Likewise, neonatal morbidity from patent ductus arteriosus and necrotizing enterocolitis diminishes significantly after 32 weeks’ gestation. Lower birthweight, which is often a function of earlier gestational age at delivery, has additionally been associated with more severe neonatal morbidity. Data from the National Institute of Child Health and Development (NICHD) Neonatal Research Network sites from 1997 to 2002 indicate that survival without complications (e.g., bronchopulmonary dysplasia, severe intraventricular hemorrhage, necrotizing enterocolitis, or a combination of these disorders) ranged from 20% for infants with a birth weight of 501 to 750 g to 89% for those with a birth weight of 1251 to 1500 g.

The survival rate among neonates increases as the birth weight and gestational age at delivery increase ( Fig. 33.2 ; Table 33.3 ). Excess mortality risk for infants born preterm is concentrated in the first year of life. After data are controlled for gestational age and weight, male infants have a higher mortality risk than female infants. During the past three decades, there has been a significant improvement in the survival rate for preterm infants, with the greatest improvement occurring in the subgroup with a birth weight between 501 and 1250 g. The rate of neonatal survival is now approximately 94% for extremely preterm infants born at 28 weeks’ gestation.

Infant Mortality Rates by Gestational Age—United States, 2021.

Data from Osterman MJK, Hamilton BE, Martin JA, et al. Births: final data for 2021. Natl Vital Stat Rep . 2023;72:1–53.

Infants born at the threshold of viability (22 to 24 weeks’ gestation) continue to have the greatest risk for poor outcomes. A retrospective cohort study assessed survival rates for infants delivered between 22 and 28 weeks’ gestation from 1993 to 2012. Neonatal survival was 9% at 22 weeks; 33% at 23 weeks; and then 65%, 81%, and 87% at 24, 25, and 26 weeks’ gestation, respectively. Another retrospective study examined trends in neonatal mortality outcomes for infants born between 22 and 24 weeks’ gestation from 2000 to 2011 ( Table 33.4 ). While the overall mortality rate decreased over the study period, the mortality rate did not improve for 22-week infants. As expected, the greatest decrease in mortality was seen in the 24-week group (55% from 2000 to 2003 to 18% from 2008 to 2011).

Preterm birth is associated with multiple long-term adverse neurodevelopmental outcomes, including cerebral palsy, motor delay, cognitive impairment, speech delay, and learning disabilities. Neuropsychiatric disorders are also more prevalent among children born preterm, particularly attention-deficit hyperactivity disorder and autism spectrum disorder. As with neonatal morbidity, the risk of persistent disability increases with earlier gestational age at delivery.

The EPIPAGE cohort, a large population-based cohort study of infants born between 22 and 32 weeks’ gestation in France in 1997, showed that at 2 years of age, 20% of children born between 24 and 26 weeks’ gestation had cerebral palsy compared with only 4% of children born at 32 weeks’ gestation. While persistent disability is certainly more common among children born extremely preterm, developmental outcomes have improved over time with advancements in newborn medicine. The EPICure study group assessed the association between extreme preterm delivery and long-term physical and mental disability in a cohort of infants delivered between 22 and 25 weeks’ gestation during a 10-month period in 1995. These investigators noted rates of severe disability of 54%, 52%, and 45% among surviving infants delivered at 23, 24, and 25 weeks’ gestation, respectively. A 6-year follow-up to the EPICure study cohort reported persistent severe disability in 25%, 29%, and 18% of infants born at 23, 24, and 25 weeks’ gestation, respectively. In a later cohort of preterm infants born between 1997 and 2002, improvement was noted in the rates of severe disability, which were 33%, 21%, and 12% for infants delivered at 23, 24, and 25 weeks’ gestation, respectively.

Researchers have additionally examined the association between late preterm birth and neurocognitive performance in late adulthood; 919 Finnish men and women born preterm were evaluated at a mean age of 68.2 years. When controlling for confounders, those who were born between 34 and 37 weeks’ gestation had lower scores on tests evaluating neurocognitive performance than those born after 37 weeks’ gestation.

Multiple maternal health, socioeconomic, and obstetric factors have been identified as potential risk factors for preterm birth. Box 33.1 lists factors associated with preterm labor. The most important predictor of risk for preterm birth is a history of spontaneous preterm birth in a prior pregnancy. After one preterm birth, the risk of recurrent preterm birth in the subsequent pregnancy is approximately 35%. An individual’s risk of recurrent preterm birth increases with the number of prior preterm deliveries and earlier gestational age at delivery.

Several additional maternal health history factors are associated with an increased risk for preterm birth. Low prepregnancy body mass index (<18.5 kg/m ² ), tobacco use, and substance use are additional risk factors for spontaneous preterm delivery. Studies have repeatedly demonstrated an association between periodontal disease and preterm birth; however, randomized trials have not shown reduced preterm birth risk following treatment of periodontal disease in pregnancy. Infectious exposures during pregnancy, including urinary tract infections (UTIs) and bacterial vaginitis, may also increase preterm birth risk.

Short cervical length , as assessed by transvaginal ultrasonography and defined as cervical length < 25 mm, is associated with a greater risk for preterm delivery. In a 2006 systematic analysis, Kagan et al. concluded that shorter cervical length is associated with preterm birth in symptomatic women. Furthermore, multiple studies have shown an increased risk for preterm delivery in asymptomatic women with a shortened cervix. A Maternal-Fetal Medicine Units Network study of nearly 3000 women found that the risk for spontaneous preterm delivery is increased in women with evidence of a short cervix detected by transvaginal ultrasonography between 24 and 28 weeks’ gestation. A cervical length below the 10th percentile had a sensitivity of 37% and a specificity of 92% in predicting preterm birth before 35 weeks’ gestation, with a corresponding positive predictive value of 18% and a negative predictive value of 97%.

A history of cervical surgery, including conization and loop electrosurgical excision procedure (LEEP), has traditionally been identified as a risk factor for preterm birth because of associated cervical injury and/or shortening. However, this relationship may be related to environmental or behavioral factors that underlie the progression of cervical dysplasia rather than cervical structural changes. A systematic review and metaanalysis of patients both with and without a history of LEEP showed that while LEEP was associated with greater risk of preterm birth before 37 weeks’ gestation, the risk was equivalent to those patients with a history of cervical dysplasia without LEEP. Uterine instrumentation, such as dilation and curettage, has also been associated with an increased risk for preterm birth in some studies, possibly secondary to endometrial injury or introduction of microbes into the intrauterine space. The risk of preterm birth may also be heightened in those with altered uterocervical anatomy due to congenital uterine anomalies, including unicornuate uterus, bicornuate uterus, or uterus didelphys.

Socioeconomic factors, including economic disadvantage and lower educational attainment, are also associated with increased risk of spontaneous preterm birth. Significant racial disparities have been noted in preterm birth rates. Even when controlling for income, education level, and insurance status, there are markedly higher rates of preterm delivery among non-Hispanic Black Americans compared to their Hispanic and non-Hispanic White counterparts. The ACOG proposes that this disparity is likely a reflection of exposure of this population to systemic structural racism, and efforts to address racial inequities are an active area of focus in the obstetric community (see Chapter 14 ).

Obstetric factors associated with increased risk of spontaneous preterm birth include multifetal gestation, vaginal bleeding in pregnancy, and short interpregnancy interval (<18 months between consecutive pregnancies). In the United States, 62.1% of twins were born preterm compared to 8.76% of singletons in 2021. The high rate of preterm delivery among multiple gestations is attributable to both spontaneous preterm labor in the setting of increased uterine distention and increased risk for obstetric complications necessitating medically indicated preterm delivery. In the past three decades, there was a significant rise in the incidence of multiple gestation (see Chapter 34 ) due to a shift toward older maternal age at conception and increased utilization of assisted reproductive technology (ART). The twin birth rate rose 76% from 1980 to 2009 (from 18.9 twins per 1000 total births to 33.2 per 1000). In 2015 ART contributed to 1.7% of all infants born in the United States and 17.0% of all multiple-birth infants. Recently, rates of multifetal pregnancies have started to decline. The twin birth rate peaked at 33.9 twins per 1000 births in 2014 and has since decreased to 31.2 per 1000 births in 2021. Likewise, the triplet and higher-order multiple birth rate has fallen 59% since the 1998 peak (193.5 per 100,000 births) to a rate of 80.0 per 100,000 births in 2021. Modifications in ART may be contributing to changes in the preterm birth rate and recent declines in multiple gestation.

UTIs and infectious vaginitis have been associated with increased risk of preterm birth. While treatment of symptomatic UTIs has not been shown to reduce the risk of preterm birth, universal screening and treatment for asymptomatic bacteriuria is recommended in early pregnancy to prevent morbidity related to progression to pyelonephritis. Bacterial vaginosis (BV) is a common bacterial vaginitis caused by Gardnerella vaginalis which has been associated with increased risk of preterm birth. A systematic review and metaanalysis showed that while antibiotics effectively clear BV in pregnancy, treatment was not associated with reduced risk of preterm labor or PPROM. Universal screening and treatment for BV in pregnancy is therefore not recommended for asymptomatic patients.

Fetal fibronectin (FFN) is an extracellular matrix protein found at the uteroplacental interface that has been suggested as a biomarker for preterm labor when detected in cervicovaginal fluid. While the presence of elevated FFN levels in cervicovaginal fluid of symptomatic patients is both highly sensitive (93%) and specific (82%), this test has overall poor positive predictive value (29%) for preterm delivery within 7 days. However, negative FFN testing has a remarkably high negative predictive value (99.5%), rendering the likelihood of preterm delivery within 7 days of a negative test extremely low. Given the limited positive predictive value of FFN testing for preterm delivery, performing FFN screening is not recommended for asymptomatic patients. For symptomatic patients at high risk of preterm delivery, FFN testing with concurrent cervical length measurement may have utility in the appropriate clinical context.

Other novel proteomic biomarkers are being investigated for their ability to predict preterm birth. A 2016 study found that two serum proteins (insulin-like growth factor-binding protein 4 [IBP4] and sex hormone–binding globulin) identified asymptomatic pregnant women at risk for spontaneous preterm delivery ; these findings were replicated in a subsequent independent multicenter observational study. Proteomic analysis of cervicovaginal fluid samples obtained from high-risk asymptomatic patients has identified multiple additional biomarker candidates, including human interleukin−1 receptor antagonist protein, γ-glutamyl hydrolase, extracellular matrix protein 1, vitamin D−binding protein, metalloproteinase inhibitor 1, human laminin subunit gamma-2, and pigment epithelium-derived factor. Although promising, these markers have yet to be used in clinical practice.

Evidence suggests that progesterone therapy may be effective in reducing the rate of preterm birth in select patient populations. Multiple formulations and routes of administration of supplemental progesterone have been studied. Though vaginal progesterone has consistently been shown to reduce the incidence of preterm birth in patients diagnosed with a short cervix without a history of preterm birth, the benefit of supplemental progesterone for prevention of recurrent spontaneous preterm birth remains controversial.

Cervical cerclage placement is a surgical procedure, typically performed via a vaginal approach, in which the cervix is sutured circumferentially to provide structural support to the cervix, thereby preventing cervical shortening and dilation (see Chapter 17 ). Candidacy for cerclage placement is dependent on the patient’s obstetric history and cervical length.

History-indicated cerclage is typically performed at 12 to 14 weeks’ gestation for patients with an established history of cervical insufficiency. History of adverse obstetric outcomes consistent with cervical insufficiency include one or more mid-trimester pregnancy losses resulting from painless cervical dilation and/or need for cerclage placement in prior pregnancy. A multicenter, randomized controlled trial conducted by the Royal College of Obstetricians and Gynaecologists Working Party on Cervical Cerclage found that history-indicated cerclage placement was associated with a significantly lower risk of preterm birth before 33 weeks’ gestation with a number needed to treat of 25 patients. However, the efficacy and cost-effectiveness of history-indicated cerclage compared to serial cervical length surveillance has been less convincing in other studies, leading many providers to instead recommend a surveillance strategy for prevention of recurrent preterm birth.

Patients with a history of spontaneous preterm birth undergoing serial cervical length surveillance with sonographic cervical length < 25 mm are candidates for ultrasound-indicated cerclage . In a metaanalysis including patients with a prior preterm birth and cervical length < 25 mm before 24 weeks’ gestation, cerclage placement was associated with a 39% reduction in the risk of preterm delivery before 35 weeks. No studies have directly compared the efficacy of cerclage versus vaginal progesterone for patients with short cervical length and a history of spontaneous preterm birth. However, a metaanalysis including randomized controlled trials that compared (1) cerclage versus expectant management or (2) vaginal progesterone versus placebo suggests that vaginal progesterone therapy may be equally effective to cerclage in this population.

The benefit of ultrasound-indicated cerclage in patients with a short cervix has been consistently demonstrated only for those with prior preterm delivery. Among patients with a cervical length < 25 mm but no history of preterm birth, cerclage placement has not been shown in randomized trials to reduce the risk of preterm delivery. Some studies have, however, suggested that cerclage placement reduces the risk of preterm delivery for the subset of patients without prior preterm birth with a very short cervical length < 10 mm.

In contrast to history-indicated and ultrasound-indicated cerclages, a physical examination-indicated cerclage is placed to prolong gestation in asymptomatic patients incidentally found to have mid-trimester cervical dilation prior to 24 weeks’ gestation, regardless of past obstetric history. A metaanalysis demonstrated that, compared with expectant management, examination-indicated cerclage placement resulted in mean pregnancy prolongation of approximately 34 days and a 65% increase in neonatal survival. Contraindications include established preterm labor with impending preterm birth, chorioamnionitis, heavy vaginal bleeding, PPROM, fetal compromise, life-limiting fetal anomalies, and fetal demise. Prior to cerclage placement, amniocentesis can be performed as clinically indicated to evaluate the amniotic fluid for evidence of subclinical infection. Data are mixed regarding mandating a waiting period before placement. Some experts advocate for immediate placement, while others recommend delaying placement for up to 24 hours with monitoring for evidence of preterm labor; there is no single evidence-based standard to define best practice regarding the timing of examination-indicated cerclage placement.

When placing the cerclage, prolapsing membranes, if present, need to be replaced in the uterine cavity to reduce the risk for iatrogenic PPROM. The surgical technique typically involves the lithotomy position with steep Trendelenburg tilt and applying tension circumferentially around the external os by placing ring forceps or stay sutures to ease the membranes back into the uterus. Invasive methods may also be used, including using a Foley catheter balloon to directly push the membranes back into the uterus. However, this practice may be associated with increased risk for iatrogenic PPROM. Some evidence suggests a course of perioperative antibiotic and tocolytic administration is beneficial in patients undergoing examination-indicated cerclage placement.

Low-dose aspirin prophylaxis has been studied as a possible low-cost, readily accessible intervention for the prevention of preterm birth. The spirin in the prevention of recurrent spontaneous preterm labor (APRIL) study, a double-bind randomized controlled trial including patients with a history of spontaneous singleton preterm birth, did not observe a decreased incidence of recurrent preterm birth in patients taking aspirin compared to placebo. However, the aspirin supplementation for pregnancy indicated risk reduction in nulliparas (ASPIRIN) study, an international, double-blind, randomized controlled trial that enrolled nulliparous patients in six low- and middle-income countries, found a 25% decrease in the incidence of spontaneous preterm delivery in patients taking aspirin compared to controls. Data are currently insufficient to recommend aspirin use for the indication of spontaneous preterm birth risk reduction.

Vaginal pessaries have not been shown to decrease the risk of preterm delivery in singleton or multifetal gestations in multiple studies. Evidence does not support the administration of prophylactic antibiotics in asymptomatic women at risk for preterm labor. Likewise, evidence does not support the prophylactic use of beta-adrenergic receptor agonists to prevent preterm labor in high-risk women. Strict bed rest has not been shown to reduce the risk of preterm delivery, and may increase the risk of adverse maternal health consequences, including deconditioning, mood disturbances, and venous thromboembolism (VTE).

Determining whether a woman is in early preterm labor or in false labor is often difficult. Criteria for the diagnosis of preterm labor include gestational age between 20 0/7 and 36 6/7 weeks and regular uterine contractions accompanied by a change in cervical dilation, effacement, or both (or initial presentation with regular contractions and cervical dilation ≥ 2 cm). Compared to approximately 50% of patients with PPROM, less than 10% of women diagnosed with preterm labor progress to delivery within 7 days of presentation.

Initial assessment of the patient with possible preterm labor includes physical examination and external monitoring of contractions with a tocodynamometer (and fetal heart rate [FHR] monitoring if indicated by gestational age). Acute conditions associated with preterm labor should be considered, including intrauterine infection and placental abruption. Maternal physical examination may include a sterile speculum examination to evaluate the cervix and exclude PPROM, as well as a sterile digital cervical examination. Transabdominal ultrasonography should be performed to evaluate fetal presentation, amniotic fluid level, and estimated fetal weight. Laboratory evaluation may include complete blood count with differential, urine culture, vaginitis swabs, and vaginal/rectal culture for Group B streptococcus (GBS). In many women who have preterm uterine contractions, these contractions will cease spontaneously. In the past, clinicians assumed that intravenous hydration was a useful component of therapy. However, there is no evidence that intravenous hydration reduces the chance of preterm delivery.

Once the diagnosis of preterm labor is established, the obstetric care provider must decide whether intervention is warranted, including administration of antenatal corticosteroids for fetal lung maturation, magnesium sulfate for fetal neuroprotection, or tocolysis.

Antenatal corticosteroids are recommended for all singleton and multifetal gestations from 24 0/7 to 33 6/7 weeks’ gestation in which there is a high likelihood of preterm delivery within 7 days. Typical steroid regimens for betamethasone and dexamethasone are listed in Table 33.5 . Maximal steroid effect is reached 48 hours after the first dose, though beneficial effects for the fetus begin within a few hours of exposure.

The neonatal benefits of corticosteroid administration before preterm delivery have been clearly demonstrated in large clinical trials. The NICHD Neonatal Research Network evaluated outcomes for 11,718 preterm infants delivered after antenatal maternal corticosteroid administration between 1988 and 1992. Antenatal corticosteroid treatment significantly reduced the incidence of neonatal respiratory distress syndrome, intraventricular hemorrhage, and neonatal death in all subgroups of the population studied (including male and female infants, Black and White infants, and infants delivered before 30 weeks’ gestation). The reduction in neonatal morbidity and mortality from antenatal corticosteroid administration is additive to the reduction observed with the use of neonatal surfactant alone. Although antenatal corticosteroids had historically been reserved for pregnancies at risk for preterm delivery prior to 34 weeks’ gestation, the Antenatal Late Preterm Steroids (ALPS) trial demonstrated that these benefits extend to neonates born after 34 weeks’ gestation. A single course of corticosteroids administered to women with singleton pregnancies at risk for preterm birth between 34 and 37 weeks’ gestation resulted in a significantly lower incidence of severe neonatal respiratory morbidity for appropriately selected patients. Women with prior antenatal steroid exposure, pregestational diabetes mellitus, multifetal gestation, and anticipated delivery within 12 hours were notably excluded from the ALPS trial. Given the potential risks of steroid-induced maternal hyperglycemia in the absence of data demonstrating neonatal benefit, administration of late-preterm corticosteroids to patients with pregestational diabetes is not presently recommended.

Although there is little controversy about the efficacy of a single course of antenatal corticosteroids, there remains debate over the use of multiple courses of corticosteroids for women who remain undelivered 7 days after the initial dose of corticosteroids. A large study performed by the Maternal-Fetal-Medicine Units Network randomly assigned women at risk for preterm delivery between 23 and 32 weeks’ gestation to receive either a single course or repeated (weekly) courses of antenatal corticosteroids. Weekly corticosteroid administration did not significantly reduce the composite primary morbidity outcome, but it significantly reduced the need for neonatal surfactant, mechanical ventilation, and continuous positive airway pressure, as well as the incidence of pneumothorax. However, weekly corticosteroid administration was associated with an increase in the delivery of small for gestational age infants, and there was a significant reduction in the birth weight of the infants whose mothers received four or more courses of corticosteroids. Studies have also identified an association between repeated antenatal steroid exposure and small head circumference.

To balance the potential beneficial effects and risks of additional courses of corticosteroids, some have advocated a single “rescue” course (i.e., a second course of corticosteroids), which is administered at the time of a clinical status change with an associated high probability of preterm delivery. A large randomized placebo-controlled trial found neonatal benefit in administering an additional course of corticosteroids. Criteria for administration included absence of rupture of membranes, prior corticosteroid administration at least 2 weeks earlier, gestational age < 33 weeks, and a change in clinical scenario such that it was believed that preterm birth was likely to occur within 1 week. A metaanalysis concluded that a single rescue course of antenatal corticosteroids should be considered in women whose prior course of antenatal corticosteroids was administered at least 7 days previously and who are at acute risk for preterm delivery before 34 weeks’ gestation. The ACOG has stated that one rescue course of corticosteroids may be considered in these specific populations. However, regularly scheduled repeat courses or multiple courses of corticosteroids (>2) are not currently recommended.

The results of a large, multicenter, randomized controlled trial and a metaanalysis do not support the use of prophylactic antibiotic therapy in the management of preterm labor in patients with intact membranes as a method to reduce the likelihood of preterm birth. In fact, there was an increase in cerebral palsy in children born to mothers in preterm labor with intact membranes who received any prophylactic antibiotics versus no antibiotics (relative risk [RR] 1.82; 95% confidence interval [CI], 0.99 to 3.34; 3173 children). Accordingly, the ACOG does not recommend empirical antibiotic therapy in this population. Prophylactic antibiotic administration remains appropriate, however, in women who are GBS positive or with unknown GBS status who are thought to be in preterm labor.

Several clinical trials have provided evidence that maternal administration of magnesium sulfate provides fetal neuroprotection when administered to women at risk for imminent preterm delivery prior to 32 weeks’ gestation. It is hypothesized that magnesium decreases inflammation at the fetal blood-brain barrier and suppresses the excitotoxic activity of glutamate, although the precise neuroprotective mechanism of magnesium remains unknown.

Magnesium administered with neuroprotective intent has been studied in patients at risk for imminent preterm birth due to PPROM, preterm labor with cervical dilation between 4 and 8 cm (with or without ruptured membranes), or planned medically indicated delivery within 2 to 24 hours. In 2003 the Australasian collaborative trial of magnesium sulphate (ACTOMgSO4) study group reported the results of a multicenter, randomized, placebo-controlled study of 1062 women (1255 infants) at less than 30 weeks’ gestation in whom delivery was planned or expected within 24 hours. The investigators observed no significant difference between groups in the primary outcomes, which included total pediatric mortality, cerebral palsy, or both, at a corrected age of 2 years. However, they observed a significantly reduced rate of substantial gross-motor dysfunction, as well as a reduced combined rate of death or substantial gross-motor dysfunction, in the children exposed to magnesium sulfate in utero . Similarly, the PREterm brain protection by MAGnesium sulfate (PREMAG) study, a randomized controlled trial of magnesium sulfate administration to 573 pregnant women at less than 33 weeks’ gestation in whom delivery was planned or expected within 24 hours found that infants exposed to magnesium sulfate had a reduced rate of total neonatal mortality, severe cerebral white matter injury (which is associated with cerebral palsy), and the combination of severe white matter injury and/or death; however, the differences were not statistically significant. In the beneficial effects of antenatal magnesium sulfate (BEAM) study, the largest randomized controlled trial to date, 2241 women at imminent risk for delivery before 32 weeks’ gestation were randomized to receive magnesium sulfate or placebo. The offspring who had been exposed to magnesium sulfate in utero were significantly less likely to develop moderate/severe cerebral palsy (1.9% versus 3.5%; RR, 0.55; 95% CI, 0.32 to 0.95).

A metaanalysis of these clinical trials suggests that prenatal administration of magnesium sulfate for the indication of fetal neuroprotection reduces the occurrence of cerebral palsy (RR, 0.68; 95% CI, 0.54 to 0.87). The ACOG has stated that, based on available evidence, magnesium sulfate—given before anticipated early preterm birth—reduces the risk for cerebral palsy in surviving infants. The recently published magnesium sulphate at 30 to 34 weeks’ gestational age: neuroprotection (MAGENTA) trial suggests magnesium may not have a neuroprotective benefit in preterm infants after 30 weeks’ gestation. This randomized controlled trial of magnesium sulfate administration to 1433 patients between 30 and 34 weeks’ gestation at risk for imminent delivery demonstrated no differences in death or cerebral palsy at age 2 years between study groups.

Of note, the gestational age ranges and magnesium dosage protocols utilized in the available literature differ widely; some study protocols called for administering a single magnesium sulfate bolus (4 to 6 g given intravenously over 20 to 30 minutes), and other studies called for administering an intravenous bolus followed by intravenous infusion of varying duration (4 to 6 g given intravenously over 20 to 30 minutes followed by a 1 to 2 g per hour infusion continued for 12 to 24 hours). Because the best regimen of magnesium sulfate administration remains unclear, physicians electing to use magnesium sulfate for fetal neuroprotection should develop specific guidelines regarding inclusion criteria, treatment regimens, concurrent tocolysis, and monitoring based on the protocols of the larger published trials.

Although widely used before 34 weeks’ gestation, acute tocolytic therapy remains a source of controversy due to lack of consistent evidence that tocolysis reduces the likelihood of preterm birth or improves neonatal outcome. However, because acute tocolysis has been associated with a short (approximately 48 hours) prolongation of pregnancy, it may be beneficial to facilitate transfer of the patient from a community hospital to a tertiary care facility that can provide optimal care for the preterm neonate. Moreover, a short course of tocolytic therapy may delay delivery for 24 to 48 hours, allowing maternal administration of antenatal steroids to accelerate fetal lung maturity. Thus the ACOG supports the use of acute tocolysis to allow administration of a complete first course of antenatal corticosteroids prior to 34 weeks’ gestation but discourages the continued use of tocolysis after corticosteroid administration is complete. Furthermore, tocolysis is not recommended to facilitate administration of a rescue course of corticosteroids or late-preterm corticosteroids after 34 weeks’ gestation. Tocolytics should not be administered if there is clinical evidence of intrauterine infection or active placental abruption, as the risk of both maternal and neonatal morbidity with delayed delivery in these settings outweigh the potential fetal benefit of pregnancy prolongation ( Box 33.2 ).

Once the obstetrician has decided to begin tocolytic therapy, an appropriate agent must be selected ( Table 33.6 ). A 2003 analysis of studies that compared the four classes of tocolytic agents (i.e., beta-adrenergic receptor agonists, calcium entry–blocking agents, magnesium sulfate, and cyclooxygenase inhibitors) concluded that all are more effective than placebo in prolonging pregnancy, but the investigators found no evidence of a beneficial effect on neonatal morbidity or mortality.

Although beta-adrenergic receptor agonists (e.g., ritodrine, terbutaline), which relax smooth muscle via beta ₂ -adrenergic receptor stimulation, were used as tocolytic agents for many years, they are utilized less in current practice than other tocolytic agents that are equally efficacious with fewer side effects. A 2014 metaanalysis that compared beta-adrenergic receptor agonists with a placebo concluded that the use of beta-adrenergic receptor agonists reduced the number of women who delivered within 48 hours and decreased the number of births within 7 days. However, there was no reduction in overall preterm birth rate or in perinatal/neonatal death or neonatal morbidity. Tocolysis was significantly associated with adverse maternal side effects (see Table 33.6 ).

Given safety concerns, ritodrine is no longer available in the United States, although it remains in use as a first-line tocolytic agent in several Asian countries. In 2011 the FDA issued a warning regarding terbutaline use. Specifically, it stated that injectable terbutaline should not be used in pregnant women for prolonged treatment (beyond 48 to 72 hours) of preterm labor in either the hospital or outpatient setting because of the potential for maternal cardiac complications and death. It also noted that oral terbutaline should not be used for prevention or treatment of preterm labor because it had not been shown to be effective and was associated with similar safety concerns.

A metaanalysis has suggested that calcium entry–blocking agents (e.g., nifedipine), which block calcium inflow into cells through voltage-dependent calcium channels, have benefits over beta-adrenergic receptor agonists with respect to prolongation of pregnancy, serious neonatal morbidity, and maternal adverse effects.

Prostaglandins, which are produced by cyclooxygenase (COX) enzymes, are mediators in the final pathways of uterine contraction. They increase intracellular calcium concentrations, increase activation of MLCK, and promote gap junction formation. The nonselective COX inhibitor indomethacin is the agent in this class most often studied as a tocolytic agent. Serious maternal side effects are uncommon. Fetal concerns, particularly in the setting of prolonged use (>48 hours), include a risk for premature closure of the ductus arteriosus and oligohydramnios (caused by fetal renal dysfunction). A 2015 metaanalysis concluded that COX inhibitors have no clear benefit compared with a placebo or any other tocolytic agents.

The oxytocin receptor antagonist atosiban has also been used as a tocolytic agent. Although available in Europe, the drug was not approved by the FDA due to safety concerns following a randomized trial showing a higher rate of fetal death in patients treated with atosiban. A 2014 metaanalysis did not demonstrate superiority of oxytocin receptor antagonists (largely atosiban) compared with placebo, beta-adrenergic receptor agonists, or calcium entry–blocking agents (largely nifedipine) for pregnancy prolongation or neonatal outcomes, although treatment with oxytocin receptor antagonists was associated with fewer adverse maternal effects than treatment with the other two drug classes. A subsequent large randomized controlled trial comparing atosiban with nifedipine found that 48 hours of tocolysis with nifedipine or atosiban resulted in similar perinatal outcomes.

Historically, magnesium sulfate has also been used as a tocolytic agent in patients presenting with threatened preterm labor with cervical dilation < 4 cm. A 2014 metaanalysis of 37 heterogeneous trials that compared magnesium with placebo, no treatment, or other tocolytic drugs concluded that magnesium is ineffective in delaying or preventing preterm birth. While the benefit of magnesium for fetal neuroprotection and maternal seizure prophylaxis in preeclampsia has been established, magnesium treatment had no benefit for a range of neonatal and maternal outcomes when used solely for the purpose of tocolysis. Furthermore, its use for this indication may be associated with an increased risk for total fetal, neonatal, or infant mortality. A secondary tocolytic agent should therefore be utilized concurrently for patients who require tocolysis who are receiving magnesium for the indication of fetal neuroprotection.

The nitric oxide donor nitroglycerin has also been evaluated for its efficacy as a tocolytic. However, findings from metaanalyses indicate that nitroglycerin results in neither significantly later gestational age at delivery nor better neonatal outcome compared with a placebo or other tocolytic drugs.

The ACOG has stated that evidence supports the use of tocolytic treatment with beta-adrenergic receptor agonist therapy, calcium entry–blocking agents, or COX inhibitors for short-term prolongation of pregnancy (up to 48 hours) to allow for antenatal maternal corticosteroid administration. Multiple clinical trials have demonstrated that prolonged use of tocolytic agents (as prophylactic therapy or after completion of acute treatment) does not alter pregnancy outcome. Regardless of which tocolytic agent is chosen, the risk for side effects appears to increase when more than one tocolytic agent is administered simultaneously. It is unclear whether a combination of tocolytic drugs for preterm labor is more effective for women and/or newborns because there is a lack of large, well-designed trials that include the outcomes of interest.

PPROM is diagnosed based on the following three findings on sterile speculum examination: (1) pooling of fluid in the posterior fornix of the vagina; (2) alkaline pH of the vaginal fluid, typically assessed by blue color change of nitrazine paper; and (3) the presence of “ferning” on microscopic evaluation of the dried vaginal fluid. Digital cervical examination is generally avoided in PPROM to limit introduction of genital tract pathogens into the amniotic space. Assessment of cervical dilation is therefore typically performed visually with sterile speculum examination. As for patients presenting with preterm labor symptoms, external FHR and contraction monitoring, obstetric ultrasonography, and laboratory evaluation are recommended.

Intrauterine infection accompanies an estimated 15% to 35% of cases of PPROM. Close monitoring should therefore be performed to detect any fever, maternal and/or fetal tachycardia, tenderness of the uterine fundus, mucopurulent and/or malodorous amniotic fluid, and leukocytosis, any of which may indicate intrauterine infection.

In contrast to patients presenting with preterm labor, antibiotic therapy in patients with PPROM has been shown to increase pregnancy latency and reduce both maternal and neonatal morbidity. Thus the ACOG recommends a 7-day course of antimicrobial therapy when PPROM is diagnosed. The best antibiotic regimen is not known with certainty, although intravenous ampicillin and erythromycin (48 hours), followed by oral amoxicillin and erythromycin (5 days), is a commonly used regimen for women with PPROM who are receiving expectant management; a single dose of oral azithromycin can be used as an alternative to erythromycin. Regardless of exposure to latency antibiotics, prolonged expectant management for PPROM is a known risk for chorioamnionitis and maternal sepsis (see Chapter 36 ).

If not previously administered, a single course of antenatal steroids is recommended for patients diagnosed with PPROM between 24 0/7 and 33 6/7 weeks’ gestation; a single course of late-preterm antenatal corticosteroids can also be offered to appropriately selected patients with PPROM diagnosed between 34 0/7 and 36 6/7 weeks’ gestation in whom delivery is planned in greater than 24 hours but no more than 7 days from the time of diagnosis. The benefit of a rescue course of antenatal steroids in the setting of PPROM remains uncertain. In a 2023 double-blind, randomized controlled trial that included 386 patients with PPROM, administration of rescue steroids did not improve neonatal morbidity or mortality, but was well tolerated without increased rates of maternal chorioamnionitis or neonatal sepsis.

Magnesium is recommended for fetal neuroprotection for patients with PPROM at less than 32 weeks’ gestation in which delivery is imminent. Intrapartum antibiotics for GBS prophylaxis are additionally indicated for patients in preterm labor following PPROM.