Preterm labor is defined as the presence of uterine contractions of sufficient frequency and intensity to effect progressive effacement and dilation of the cervix prior to term gestation. Occurring at 20-37 weeks’ gestation, preterm labor precedes almost half of preterm births and is the leading cause of neonatal mortality in the United States.
Risk of preterm labor
The exact mechanisms of preterm labor are largely unknown but are believed to include the following:
Decidual hemorrhage such as abruption and mechanical factors such as uterine overdistention from multiple gestation or polyhydramnios
Cervical incompetence (eg, trauma, cone biopsy)
Uterine distortion (eg, müllerian duct abnormalities, fibroid uterus)
Cervical inflammation as a result of, for example, bacterial vaginosis (BV) or trichomonas
Maternal inflammation/fever (eg, urinary tract infection)
Hormonal changes (eg, mediated by maternal or fetal stress)
Uteroplacental insufficiency (eg, hypertension, insulin-dependent diabetes, drug abuse, smoking, alcohol consumption)
Risk assessment during pregnancy
The integrity of the cervix and the extent of any prior injury to the cervix may be assessed by speculum and digital examination. The presence of asymptomatic bacteriuria, sexually transmitted disease (STD), and symptomatic BV may be investigated
A history of prior preterm deliveries places the patient in the high-risk category. Of the predictors of preterm birth, past obstetric history may be one of the strongest predictors of recurrent preterm birth.
A short cervical length in the early or late second trimester has been associated with a markedly increased risk of preterm labor and delivery. In a study, a cervical length of 25 mm or less at 28 weeks had a 49% sensitivity for prediction of preterm delivery at less than 35 weeks. 
In patients with a history of midtrimester loss, laboratory tests for risk assessment include the following:
Rapid plasma reagin test
Gonorrheal and chlamydial screening
Vaginal pH/wet smear/whiff test
Anticardiolipin antibody (eg, anticardiolipin immunoglobulin [Ig] G and IgM, anti-beta2 microglobulin)
Lupus anticoagulant antibody
Activated partial thromboplastin time
One-hour glucose challenge test
In addition, one should consider TORCH (toxoplasmosis, other infections, rubella, cytomegalovirus infection, herpes simplex), immunoglobulin G, and immunoglobulin M screening whenever the historical or clinical suspicion is present.
Contractions of sufficient frequency and intensity to effect progressive effacement and dilation of the cervix at 24-37 weeks’ gestation are indicative of active preterm labor. If the diagnosis of preterm labor is suspected, but not confirmed, it may be prudent to first obtain a vaginal fetal fibronectin (FFN) sample before pelvic cervical examination. If the diagnosis remains in doubt after the exam, the FFN specimen can be sent to the lab for analysis.
Studies support the use of progesterone supplementation to reduce preterm birth in patients at high risk for recurrent preterm delivery.
Criteria that indicate consideration of tocolytic therapy include more than 6 contractions per hour resulting in a demonstrated cervical change or presumed prior cervical change (transvaginal cervical length < 25 mm, >50% cervical effacement, or cervical dilation ≥20 mm). If contractions are present without cervical change, management options include continued observation or therapeutic sleep for the patient (eg, morphine sulphate 10-15 mg subcutaneous).
The most common tocolytic agents used to treat preterm labor include the following:
Magnesium sulfate (MgSO4): Widely used as the primary tocolytic agent because it has similar efficacy to terbutaline (one of the previous agents of choice), with far better tolerance
Indomethacin: An appropriate first-line tocolytic for early preterm labor (< 30 wk) or preterm labor associated with polyhydramnios
Nifedipine: Despite its unlabeled status, several randomized studies have found nifedipine to be associated with a more frequent successful prolongation of pregnancy than other tocolytics
Preterm labor is defined as the presence of uterine contractions of sufficient frequency and intensity to effect progressive effacement and dilation of the cervix prior to term gestation (between 20 and 37 wk). Preterm labor precedes almost half of preterm births and preterm birth occurs in approximately 12% of pregnancies and is the leading cause of neonatal mortality in the United States. [2, 3] In addition, preterm birth accounts for 70% of neonatal morbidity, mortality, and health care dollars spent on the neonate, largely due to the 2% of American women delivering very premature infants (< 32 wk). [2, 3]
Despite the current use of material, effort, and money in perinatal medical technology, neonatal mortality rates for newborns born in the United States (5 per 1,000 babies) may rank as low as 32nd among the 33 industrialized nations, superior only to Latvia. 
Successful reduction of perinatal morbidity and mortality associated with prematurity may require the implementation of effective risk identification and behavioral modification programs for the prevention of preterm labor; these in turn require both an improved understanding of the psychosocial risk factors, etiology, and mechanisms of preterm labor and programs for accurate identification of pregnant women at risk for premature labor and delivery. In fact, recent evidence suggests that early identification of at-risk gravidas with timely referral for subspecialized obstetrical care may help identify women at risk for preterm labor and delivery and decrease the extreme prematurity (< 32 wk) rate, thereby reducing the morbidity, mortality, and expense associated with prematurity. 
Goals of management
The focus of this article is the prevention, diagnosis, and treatment of preterm labor with intact membranes. The management of preterm labor associated with ruptured membranes is reviewed in Premature Rupture of Membranes; however, the overall goals of both management schemes are similar.
Goals of obstetric patient management of preterm labor should include (1) early identification of risk factors associated with preterm birth, (2) timely diagnosis of preterm labor, (3) identifying the etiology of preterm labor, (4) evaluating fetal well-being, (5) providing prophylactic pharmacologic therapy to prolong gestation and reduce the incidence of respiratory distress syndrome (RDS) and intra-amniotic infection (IAI), (6) initiating tocolytic therapy when indicated, and (7) establishing a plan of maternal and fetal surveillance with patient/provider education to improve neonatal outcome.
Risk of Preterm Labor
The exact mechanism(s) of preterm labor is largely unknown but is believed to include decidual hemorrhage, (eg, abruption, mechanical factors such as uterine overdistension from multiple gestation or polyhydramnios), cervical incompetence (eg, trauma, cone biopsy), uterine distortion (eg, müllerian duct abnormalities, fibroid uterus), cervical inflammation (eg, resulting from bacterial vaginosis [BV], trichomonas), maternal inflammation/fever (eg, urinary tract infection), hormonal changes (eg, mediated by maternal or fetal stress), and uteroplacental insufficiency (eg, hypertension, insulin-dependent diabetes, drug abuse, smoking, alcohol consumption). [2, 3]
Although prediction of preterm delivery remains inexact, a variety of maternal and obstetric characteristics are known to increase the risk, presumably via one of these mechanisms. Finally, the fetus plays a role in the initiation of labor. In a simplistic sense, the fetus recognizes a hostile intrauterine environment and precipitates labor by premature activation of a fetal-placental parturition pathway.
Risk factors for preterm birth include demographic characteristics, behavioral factors, and aspects of obstetric history such as previous preterm birth. Demographic factors for preterm labor include nonwhite race, extremes of maternal age (< 17 y or >35 y), low socioeconomic status, and low prepregnancy weight. Preterm labor and birth can be associated with stressful life situations (eg, domestic violence; close family death; insecurity over food, home, or partner; work and home environment) either indirectly by associated risk behaviors or directly by mechanisms not completely understood. Many risk factors may manifest in the same gravida.
Methods used for predicting preterm birth include home uterine activity monitoring (HUAM), assessments of salivary estriol, fetal fibronectin (FFN), the presence of BV, and cervical length assessment.
While hospital tocodynamometry has been effective for monitoring uterine contractions to evaluate preterm labor, HUAM has not been proven valuable in detecting or preventing preterm birth and is not currently recommended for use.
The proposed use of salivary estriol measurements in detecting preterm labor was based on the belief that the adrenal gland production of dehydroepiandrosterone increases before the onset of labor, which results in an increase of maternal estriol. Unfortunately, maternal estriol levels show diurnal variation, peaking at night, and are suppressed by betamethasone administration, thereby decreasing the predictive value of salivary estriol in the detection of preterm delivery risk.
FFN is a basement membrane protein that helps bind placental membranes to the decidua. While a negative FFN is helpful in predicting women who are not destined to deliver preterm, a positive FFN has limited value in predicting women who will deliver preterm. Nevertheless, FFN has a predictive value in identifying patients who will or will not deliver within the subsequent 1-2 weeks.
While the presence of BV has been associated with the risk of preterm delivery, prospective treatment trials eradicating asymptomatic BV failed to reduce the risk of preterm delivery.
Longer term prediction of the risk of preterm delivery is achieved by cervical length measurements. A short cervical length in the early or late second trimester has been associated with a markedly increased risk of preterm labor and delivery (see discussion on cervical length). The prediction of preterm delivery may potentially be improved by combining FFN testing with measurements of cervical length.
While the risk for preterm birth in nulliparous patients is hard to determine, past obstetric experience and personal behavior may provide significant insight into future pregnancy outcome in multiparous women. Identifying at-risk patients preconceptually may allow additional treatment options. Women who seek birth control have a 30% chance of becoming pregnant in the next 2 years, suggesting that these women represent one potential opportunity for intervention. The presence of the following risk factors should be addressed prior to pregnancy.
The most common etiologies for cervical injury are elective abortion, surgeries to treat cervical dysplasia, and injury occurring at delivery. A single uncomplicated elective abortion at less than 10 weeks’ gestation does not increase the risk of midtrimester loss or preterm birth unless the cervix has been forcibly dilated to more than 10 mm at the time of the abortion. However, patients with a history of multiple first-trimester elective terminations or one or more second-trimester elective abortions may be at increased risk for preterm delivery. Cervical dilatation with laminaria or cervical ripening agents, such as misoprostol, appears to be less traumatizing to the cervix than mechanical dilation.
Cervical dysplasia should be treated appropriately whenever diagnosed. However the incidence of preterm birth and cervical incompetence may be increased 200-300% after preconceptual surgical treatment (eg, cold knife cone, cryoconization, laser cone, LEEP) of cervical intraepithelial neoplasia (CIN). The risk of subsequent preterm delivery may be proportional to the amount of cervical tissue removed during surgery. Surprisingly, the ease of performing LEEP for relatively minor abnormalities may have paradoxically led to more cervical injury than was observed with the relatively more invasive cone biopsy.
Obstetric trauma may be underestimated as a risk for midtrimester loss or preterm birth. While women may relate a history of cervical laceration, often they are unaware of the injury and the obstetric records of the previous delivery may be misleading as to the extent of the cervical injury. Therefore, visual inspection of the cervix is important to assess the degree of injury and risk. Defects that involve more than 50% of the cervical length may indicate a higher risk for midtrimester loss.  The accuracy of transvaginal ultrasonic measurements to determine risk of cervical incompetence, specifically in the presence of a history of cervical trauma, has yet to be determined.
Genital tract infection
The young gynecology patient diagnosed with gonorrhea, chlamydia, or trichomoniasis has an approximate 25% risk of reinfection during the subsequent 12 months, but a clear association between these organisms and preterm delivery has not been established. BV is a vaginal syndrome associated with an alteration of the normal vaginal flora rather than an infection specific to any one organism and a lack of vaginal inflammation is evident when compared with vaginitis. The diagnosis of BV should be suspected with a positive Gram stain result or the presence of 3 of 4 traditional diagnostic signs (homogenous gray-white discharge, >20% clue cells on saline wet smear, positive whiff test, and a vaginal pH >4.50) Patients should be treated per the US Centers for Disease Control and Prevention guidelines, with test-of-cure sampling and subsequent treatment if necessary.
Preterm labor/birth history
A history of prior preterm deliveries places the patient in the high-risk category. Of the predictors of preterm birth, past obstetric history may be one of the strongest predictors of recurrent preterm birth. Given a baseline risk of 10-12%, the risk of recurrent preterm birth after 1, 2, and 3 consecutive preterm births may be increased to approximately 15%, 30%, and 45%, respectively. Preconceptual counseling should help encourage patients to make informed decisions concerning future pregnancy in light of prematurity risk in the presence of previous preterm delivery. Often the best time to counsel the patient is at her 4- to 6-week postpartum check after a preterm delivery.
Lykke et al found that spontaneous preterm delivery, preeclampsia, or fetal growth deviation in a first singleton pregnancy predisposes women to those complications in their second pregnancy, especially if the complications were severe. In a registry-based cohort study of 536,419 Danish women, delivery between 32 and 36 weeks of gestation increased the risk of preterm delivery in the second pregnancy from 2.7% to 14.7% (odds ratio [OR] 6.12; 95% confidence interval [CI], 5.84-6.42) and increased the risk of preeclampsia from 1.1% to 1.8% (OR 1.60; 95% CI, 1.41-1.81). A first delivery before 28 weeks increased the risk of a second preterm delivery to 26.0% (OR 13.1; 95% CI, 10.8-15.9) and increased the risk of preeclampsia to 3.2% (OR 2.96; 95% CI, 1.80-4.88).
The optimal method of preventing preterm birth in multiple gestations has yet to be proven. Cervical cerclage, prophylactic bed rest, and empiric use of tocolytics have not been successful. Most recently, a randomized controlled trial by Lim et al suggests that the use of 17α-hydroxyprogesterone caproate does not prevent neonatal morbidity or preterm birth in multiple pregnancies. 
Preeclampsia in a first pregnancy with delivery between 32 and 36 weeks increased the risk of preeclampsia in a second pregnancy from 14.1% to 25.3% (OR 2.08; 95% CI, 1.87-2.31) and increased the risk for a small for gestational age infant from 3.1% to 9.6% (OR 2.82; 95% CI, 2.38-3.35). Fetal growth 2 to 3 standard deviations below the mean in a first pregnancy increased the risk of preeclampsia from 1.1% to 1.8% (OR 1.62; 95% CI, 1.34-1.96) in the second pregnancy.  See the Gestational Age from Estimated Date of Delivery calculator.
Midtrimester loss has many etiologies, including infection (eg, syphilis), antiphospholipid syndrome, diabetes, substance abuse, genetic disorders, congenital müllerian abnormalities, cervical trauma, and cervical incompetence. Unfortunately, many midtrimester losses remain unexplained. A complete workup (see History of midtrimester loss) may be of value in selected patients following a midtrimester loss.
Assessment of Risk During Pregnancy
Physical Assessment Guidelines to Establish Risk
Assessment during the first prenatal exam should include the patient's obstetric history, infection risk, and the presence of cervical or uterine abnormalities. If an evaluation for antiphospholipid syndrome is included, it should include anticardiolipin (eg, anticardiolipin IgG and IgM and anti-beta2 microglobulin) and lupus anticoagulant antibodies.
Previous preterm deliveries, including autopsy reports and medical records, if appropriate and available, should be reviewed. Social stressors (including housing and food availability), social support in the family, financial stability, domestic violence, drug abuse involving the patient or her family, and death or serious illness in a close family member should be assessed.
The integrity of the cervix and the extent of any prior injury to the cervix may be assessed by speculum and digital examination. The presence of asymptomatic bacteriuria, STD, and symptomatic BV may be investigated.
In some patients, formal cervical length assessment may be of use in risk assessment.
Cervical length during prenatal care, particularly at 24-28 weeks’ gestation, has been demonstrated to be the most sensitive prenatal predictor of preterm birth between both high- and low-risk women. In a mixed high- and low-risk population of singleton pregnancies, transvaginal ultrasound-measured cervical length at 24 weeks was highly correlated with the risk of spontaneous preterm delivery before 35 weeks.  The relative risk of preterm delivery among women with a cervix 25 mm or shorter at 24 weeks was 6.2. Furthermore, at 28 weeks, a short cervix (≤25 mm) was associated with a 9.6 relative risk of preterm delivery. Cervical length 25 mm or shorter at 28 weeks had a 49% sensitivity for prediction of preterm delivery at less than 35 weeks, a value markedly greater than that of cervical funneling.
Among high-risk women with a history of one or more spontaneous preterm births (excluding those with multiple gestation, uterine anomalies, and prior cervical surgeries), 20% of patients demonstrated a cervical length shorter than 25 mm by transvaginal ultrasonography at 22-25 weeks.  Among these patients with a short cervix and one previous preterm birth, 37.5% delivered at less than 35 weeks. In contrast, patients with a cervical length longer than 25 mm had a preterm rate (< 35 wk) of only 10.6%. Cervical length has similarly been demonstrated as the optimal predictor of preterm delivery in low-risk women. In an assessment of low-risk women, short cervical length at 24-28 weeks was detected in 8.5% of women.  These patients demonstrated a relative risk of 6.9 for preterm delivery at less than 35 weeks. As compared with fetal fibronectin or Bishop score, cervical length demonstrated the greatest sensitivity (39%), with a specificity of 92.5% and a negative predictive value of 98%.
Whereas cervical length assessment by digital exam is a semi subjective measurement, a recent study has demonstrated the value of an objective cervico-portio length measurement using Cerivlenz, an intravaginal measuring device.  These manually obtained cervical length measurements appear to be reproducible, accurate, and predictive of a short cervical length by transvaginal ultrasonography. Therefore, Cerivlenz may represent a low-cost, objective screening tool to identify at-risk patients for preterm delivery.
In addition to the 24-28 week assessment, evidence shows the value of early midtrimester cervical length measurement. Studies of Owen et al from the Maternal Fetal Medicine Units Network  demonstrate the value of cervical length measurements between 16 weeks and 23 weeks and 6 days. Serial transvaginal ultrasonographic cervical length measurements in a high-risk population demonstrated that a cervix shorter than 25 mm resulted in a relative risk of 4.5 for spontaneous preterm birth at less than 35 weeks, with a 69% sensitivity, 80% specificity, 55% positive predictive value, and 88% negative predictive value. As the NIH Maternal Fetal Medicine Units Network is initiating a study of progesterone treatment for patients with a short cervix in the early midtrimester, a program of routine cervical length screening may soon be justified.
Among patients with a short cervix, education should be provided concerning the signs and symptoms of preterm labor, especially as the pregnancy approaches potential viability. Prenatal visits/contacts may be scheduled at more frequent intervals to increase patient interaction with the care provider, especially between 20 and 34 weeks’ gestation, which may decrease the rate of extreme preterm birth. 
Management of Specific Problems
Randomized clinical trials of cerclage for sonographically suspected cervical incompetence (shortened cervical length and/or funneling) have been inconclusive with respect to prevention of preterm delivery.  However, a history of midtrimester losses with loss of cervical integrity, often results in recommendation for cerclage placement between 13 and 17 weeks’ gestation. When the patient has a history of midtrimester loss after cone or LEEP biopsy therapy, prophylactic cerclage may be considered, but consulting with a maternal fetal medicine specialist may be beneficial due to the potential risks and the controversial proven benefit.
A meta-analysis of randomized trials in women with cervical length less than 25 mm on transvaginal ultrasonography found that cerclage significantly prevents preterm birth and composite perinatal mortality and morbidity in women with previous spontaneous preterm birth and singleton gestation. 
Simcox et al conducted a randomized controlled trial in 247 patients to determine if history or ultrasonography provided a better basis for whether women at risk of preterm birth should undergo cervical cerclage. Women treated on the basis of ultrasound criteria (cervical length < 20 mm) were significantly more likely to undergo cerclage (32% vs 19%; relative risk [RR] 1.66) and to receive progesterone (39% vs 25%; RR, 1.55) than were those treated on the basis of clinician preference. However, the rate of preterm delivery between 24 and 33 weeks was 15% in both groups. The results of this study showed that ultrasonographic screening of high-risk women to determine the need of cerclage resulted in more intervention but similar outcome compared with those determined to need cerclage based on history. 
History of midtrimester loss
A history of prior midtrimester losses is carefully reviewed at the initial visit to distinguish incompetent cervix from other causes (eg, abruption, infection, intrauterine death, ruptured membranes) with review of the pathology or autopsy reports if available. Parental karyotypes are generally not helpful unless more than one midtrimester loss has occurred or a midtrimester loss has occurred in which the fetus was structurally or genetically abnormal.
Specific laboratory tests, including a rapid plasma reagin test, gonorrheal and chlamydial screening, vaginal pH/wet smear/whiff test, anticardiolipin antibody, lupus anticoagulant antibody, activated partial thromboplastin time, and a 1-hour glucose challenge test are helpful in the evaluation. In addition, one should consider TORCH (toxoplasmosis, other infections, rubella, cytomegalovirus infection, herpes simplex), immunoglobulin G, and immunoglobulin M screening whenever the historical or clinical suspicion is present. However, a random drug screen is not always recommended unless other supporting high-risk behavior exists.
A preconceptual hysterosalpingogram may be of benefit in patients with a history of 2 or more midtrimester losses. One can also attempt to pass a No. 8 Hegar dilator into the nonpregnant cervix; easy passage may be a sign of cervical incompetence. During pregnancy, whenever the suspicion of incompetent cervix exists, one should consider performing baseline transvaginal ultrasonography to assess cervical length, especially at 13-17 weeks’ gestation; abnormal findings include a length less than 2.5 cm, funneling greater than 5 mm, or dynamic changes.
A cerclage may be indicated after 2 or more midtrimester losses consistent with incompetent cervix or in which the etiology is unknown and the transvaginal ultrasonography of the cervix is abnormal. A cerclage is usually performed electively at 13-17 weeks’ gestation.
A genetic amniocentesis may be performed prior to the placement of cerclage in patients at high risk for genetic disease. Prior to an elective cerclage, sampling the patient's vagina and cervix for BV, gonorrheal, chlamydial, or trichomonal infection is also recommended, with appropriate treatment instituted. The efficacy of prophylactic antibiotics for cerclage is yet to be demonstrated.
The use of progesterone therapy to reduce preterm birth
Recent studies support the use of progesterone supplementation to reduce preterm birth in patients at high risk for recurrent preterm delivery (ie, prior preterm birth < 37 weeks' gestation, short cervical length). Weekly injections of 17 alpha-hydroxyprogesterone caproate resulted in a substantial reduction in the rate of recurrent preterm delivery among women who were at high risk for preterm delivery and reduced the likelihood of several complications in their infants.  In addition, prophylactic vaginal progesterone reduced the frequency of uterine contractions and the rate of preterm delivery in women at high risk for prematurity. 
In October 2008, the American College of Obstetricians and Gynecologists issued a Committee Opinion stating that progesterone supplementation for the prevention of recurrent preterm birth should be offered to women with a singleton pregnancy and a prior spontaneous preterm birth due to spontaneous preterm labor or premature rupture of membranes. Progesterone supplementation for asymptomatic women with an incidentally identified very short cervical length (< 15 mm) may be considered. 
On February 4, 2011, the US Food and Drug Administration (FDA) approved 17-hydroxyprogesterone (Makena) to reduce risk of preterm delivery before 37 weeks’ gestation in women with singleton pregnancy and a history of at least one spontaneous preterm birth. 17-Hydroxyprogesterone is not intended for use in women with multiple gestations or other risk factors (eg, short cervical length) for preterm birth.
The dose is 250 mg (1 mL) intramuscularly in the hip every week until 37 weeks’ gestation or delivery, whichever occurs first. Initiate between 16 weeks’ gestation and before 21 weeks’ gestation (ie, 20 wk and 6 d).
The FDA reviewed data from a multicenter, randomized, double-blind clinical trial. The study included 463 women who were pregnant with a single fetus and had a history for preterm delivery. Rates of delivery before 37 weeks' gestation were 37% in women randomized to 17-hydroxyprogesterone and 55% in controls.
A separate study evaluated children born to mothers enrolled in the controlled trial. In this study, children aged 2.5-5 years reached similar developmental targets, regardless of the mother’s treatment. The confirmatory study that is ongoing will be followed by a similar infant follow-up study to be completed about 2018.
A multicenter, randomized placebo-controlled study demonstrated that intravaginal progesterone gel effectively prevented preterm delivery in patients with a short cervix (10-20 mm). Vaginal progesterone was associated with a significant reduction in rate of preterm birth before 28, 33, and 35 weeks’ gestation; infant respiratory distress syndrome; and neonatal morbidity and mortality. Currently, applications for FDA approval are in process. Of note, the predicted clinical impact of treatment of all patients with short cervix is greater than among patients with prior spontaneous preterm birth. 
Management of Preterm Labor
Preterm labor may be difficult to diagnose and a potential exists for overtreatment of uterine irritability. Tocolytic agents, while generally safe in appropriate dosages with proper clinical monitoring, have potential morbidity and should only be used after consideration of the risks and benefits of such use. Neonatal morbidity and mortality are greatly affected by gestational age, especially when the pregnancy is less than 28 weeks’ gestation. Tocolysis should be used with caution when the fetus is previable because the expected prolongation of the pregnancy is limited, and the neonate has a minimal chance of survival at less than 23 weeks. The likelihood of survival is further reduced in the presence of significant medical complications, such as intra-amniotic infection (IAI) at these ages.
On the other hand, the risk of neonatal mortality and morbidity is low after 34 completed weeks of gestation; although a trial of acute tocolysis may be initiated, aggressive tocolytic therapy is generally not recommended beyond 34 weeks, due to potential maternal complications. Between 24 and 33 weeks’ gestation, benefits of tocolytic therapy are generally accepted to outweigh the risk of maternal and/or fetal complications and these agents should be initiated provided no contraindications exist. Although aggressive tocolysis is not typically used beyond 34 weeks’ gestation, clinicians are advised not to deliver patients at this gestation without indication because of a higher risk of neonatal morbidity in infants born at 34-36 weeks’ gestation compared with deliveries at 37-40 weeks’ gestation. 
The following table depicts survival, major short-term morbidity, and intact long-term survival by gestational age.
Table. Neonatal Morbidity and Mortality by Gestational Age (Open Table in a new window)
|Gestational Age, wk||Survival||Respiratory Distress Syndrome||Intraventricular Hemorrhage||Sepsis||Necrotizing Enterocolitis||Intact|
Tocolytic agents have not proven to be efficacious in preventing preterm birth or reducing neonatal mortality or morbidity. The primary purpose of tocolytic therapy today is to delay delivery for 48 hours to allow the maximum benefit of glucocorticoids to decrease the incidence of RDS. While tocolytics can be successful for 48 hours when membranes are intact, some clinical studies suggest that the effectiveness of tocolytics is only slightly better than bedrest and hydration, both of which have fewer adverse effects than tocolytic therapy.
A 2016 clinical study  suggested a benefit of late preterm (34 0/7 to 36 6/7 weeks) steroids in women with singleton pregnancies at imminent risk of preterm delivery within 7 days, though tocolysis should not be used in order to delay delivery. The benefits were primarily a reduction in respiratory complications, though there was a marked increase in neonatal hypoglycemia in the treatment cohort. The American Congress of Obstetricians and Gynecologists (ACOG) issued a practice advisory (Practice Advisory: Antenatal Corticosteroid Administration in the Late Preterm Period, 2016)  that indicated that “administration of betamethasone may be considered in women with a singleton pregnancy between 34 0/7 and 36 6/7 weeks gestation at imminent risk of preterm birth within 7 days.”
ACOG further clarified the bulletin as follows:
Monitoring of neonatal blood glucose is recommended for late preterm infants since late preterm birth is a risk factor for hypoglycemia; these same guidelines should be followed for infants exposed to antenatal corticosteroids administered during the late preterm period.
Late preterm antenatal corticosteroid administration should not be used in women diagnosed with chorioamnionitis (intrauterine infection).
Tocolysis should not be used in order to delay delivery to allow for administration of late preterm antenatal corticosteroids, nor should an indicated late preterm delivery (such as for preeclampsia with severe features) be postponed for steroid administration.
Administration of late preterm antenatal corticosteroids should not be given if the pregnancy was already exposed to antenatal corticosteroids.
Because the ALPS trial excluded pregnant women with diabetes, multifetal gestations, previous exposure to steroids during pregnancy, or pregnancies with major non-lethal fetal malformations, ACOG is reviewing these topics and will issue any updated clinical guidance as appropriate.
A subanalysis of the results indicates that the significant benefit was observed in the patients with a planned late preterm cesarean section. In view of the incidence of neonatal hypoglycemia and the ACOG option to “consider” betamethasone, a reasonable clinical approach may be to limit the administration of late preterm betamethasone to those patients having late preterm planned cesarean sections.
Contractions of sufficient frequency and intensity to effect progressive effacement and dilation of the cervix at 24-37 weeks’ gestation are indicative of active preterm labor. If the diagnosis of preterm labor is suspected, but not confirmed, it may be prudent to first obtain a vaginal fetal fibronectin (FFN) sample before pelvic cervical examination. If the diagnosis of preterm labor becomes obvious after the pelvic examination, the FFN specimen can be subsequently discarded. However, if the diagnosis remains in doubt, the FFN specimen can be sent to the lab for analysis.
Criteria that indicate consideration of tocolytic therapy include more than 6 contractions per hour resulting in a demonstrated cervical change or presumed prior cervical change (transvaginal cervical length < 2.5 cm, >50% cervical effacement, or cervical dilation ≥2 cm). If contractions are present without cervical change, management options include continued observation or therapeutic sleep (eg, morphine sulphate 10-15 mg subcutaneous).
When using strict criteria in women at 24 0/7 to 33 6/7 weeks’ gestation for “false preterm labor” (one contraction or less in 10 min, cervical dilation < 2 cm, and no evidence of cervical change over 2 h of observation), Chao et al demonstrated that these patients had a greater incidence of late preterm (34-36 weeks’ gestation) but not early preterm delivery (< 34 weeks’ gestation), compared with a control obstetric population.  However, those patients with cervical dilation of 1 cm were more likely to delivery prior to 34 weeks’ gestation. Although this study provides some guidance as to management, a negative FFN result and/or evidence of abated contractions may be of additional value in discharging patients with false preterm labor. In addition, measures of absolute or change in cervical length (effacement), in addition to dilation, may be of value in discriminating true versus false preterm labor.
Assessment prior to tocolytic therapy
One should always attempt to determine gestational age by first identifying the first day of the last menstrual period (LMP) and confirming it by one or more of the following:
Positive pregnancy test (home or clinic) prior to the expected date of the second missed period
Uterine size determined by bimanual examination prior to 12 weeks' gestation
Doppler fetal heart tones noted prior to 12 weeks' gestation
Ultrasonographic estimation of gestational age (ie, first trimester within 1 wk, second trimester within 2 wk, and third trimester within 3 wk)
When the LMP is not reliable, the gestational age is determined by the first ultrasonography. Following gestational age determination, assessment of fetal well-being, fetal growth, and evaluation of congenital anomaly should be conducted. Subspecialist consultation (MFM) is recommended in the presence of suspected fetal anomalies because tocolytics are generally contraindicated for any congenital anomaly incompatible with life. Tocolytics are not indicated in patients with either suspected or confirmed IAI. Use of tocolytics is relatively contraindicated when evidence of a hostile intrauterine environment exists, such as the following:
Nonreactive nonstress test results
Positive contraction stress test results
Absent or reversed diastolic flow upon Doppler examination of umbilical blood flow
Repetitive severe variable decelerations
Significant vaginal bleeding consistent with abruption, unless patient is stable and fetal well being established
Evaluate for the presence of genital tract infection
Tocolytics are contraindicated in the presence of symptomatic IAI. The definition of IAI infection (ie, chorioamnionitis) includes a temperature greater than 38.0°C (100.0°F) and 2 of the 5 following signs:
WBC count greater than 15,000 cells/mm 3
Maternal tachycardia greater than 100 beats per minute (bpm)
Fetal tachycardia greater than 160 bpm
In situations in which the diagnosis remains unclear, an amniocentesis for fluid culture (aerobic/anaerobic bacteria), Gram stain (bacteria present if Gram stain is positive or if WBC count is >50 cells/mm3), glucose level (positive if < 15 mg/dL), or leukocyte esterase evaluation may be considered. However, amniocentesis may result in a false-positive FFN test result if the FFN is performed after amniocentesis. 
A study by Romero et al indicated that IAI can be quickly and accurately diagnosed with polymerase chain reaction assay with electrospray ionization time-of-flight mass spectrometry (PCR/ESI-MS). In the study, amniotic fluid from 142 women with preterm labor with intact membranes underwent culturing and PCR/ESI-MS testing. Standard culturing techniques detected microbial invasion of the amniotic cavity in 7% of these patients, while PCR/ESI-MS detected it in 12% of them. Compared with women in whom both tests were negative, those patients who had negative cultures but positive PCR/ESI-MS results had a significantly greater incidence of intra-amniotic inflammation and acute histologic chorioamnionitis, as well as a shorter time to delivery and offspring with a greater perinatal mortality risk. 
Patients with preterm labor may be assessed for the presence or absence of lower genital tract infection.
Sterile speculum examination for ruptured membranes
Endocervical sampling for gonorrhea and chlamydia
Vaginal fluid pH
Wet smear for BV and trichomonal infection if indicated
Urinalysis and culture (if indicated)
Positive results are treated with appropriate antibiotics.
Assess for medical contraindications to tocolysis
Tocolytics should be used with considerable caution in pregnant patients with cardiac disease, especially those who require medication or have a history of congestive heart failure, cardiac surgery, significant pulmonary disease, renal failure, or maternal infection (ie, pneumonia, appendicitis, pyelonephritis). In these cases, it may be prudent to consult with an MFM specialist.
Specific tocolytic agents should not be used whenever known allergies exist. Indomethacin is contraindicated in the presence of aspirin-induced asthma, coagulopathy, or significant liver disease.
Magnesium sulfate should not be used in conjunction with select medications, such as calcium channel blockers, or when myasthenia gravis or neuromuscular disorders exist. In addition, the US Food and Drug Administration (FDA) has warned against extended magnesium sulfate injections in pregnancy. In 2013, the FDA issued a safety alert advising against the off-label administration of magnesium sulfate injections to pregnant women for more than 5-7 days as a means of stopping preterm labor, as this agent can lead to low calcium levels and bone abnormalities in the fetus. [24, 25, 26]
The warning was based in part on 18 case reports of skeletal abnormalities in infants whose mothers had received magnesium sulfate injections to stop preterm labor. In these cases, fetuses were exposed to the drug for nearly 10 weeks, on average, with neonates developing transient osteopenia and fractured bones. Epidemiologic evidence also indicates a connection between maternal administration of magnesium sulfate for more than 5-7 days and hypocalcemia and skeletal abnormalities in neonates. [24, 25, 26]
Beta-mimetics (eg, terbutaline) may be contraindicated in the presence of cardiac arrhythmia, valvular disease, and ischemic heart disease and may alter glucose homeostasis in patients with diabetes.
The administration of glucocorticoids is recommended in the absence of clinical infection whenever the gestational age is between 24 and 34 weeks. An attempt should be made to delay delivery for a minimum of 12 hours to obtain clinical benefits of antenatal steroids.
Recent data also suggest that glucocorticoids (i.e., betamethasone) may be beneficial in pregnant women at high risk of late preterm birth (within 7 days), between 34 0/7 weeks and 36 6/7 weeks of gestation who have NOT received a prior course of antenatal corticosteroids.  Administration of betamethasone led to a significant decrease in respiratory complications and the need for respiratory support, though hypoglycemia was more common in the infants exposed to betamethasone 24.0% versus 14.9% (RR, 1.61; 95% CI, 1.38–1.88). 
Importantly, administration of late preterm glucocorticoids is not indicated in women with clinical chorioamnionitis, multiple gestations, or pregestational diabetes, and tocolysis should not be used in an attempt to delay delivery in order to administer glucocorticoids in the late preterm period, nor should an indicated late preterm delivery (e.g., preeclampsia with severe features) be postponed for corticosteroid administration.
The recommended dosage of betamethasone consists of two 12 mg doses 24 hours apart while four doses of 6 mg of dexamethasone should be administered at 6-hour intervals. Whenever the following clinical conditions exist, the glucocorticoid regimen may require modification:
In the presence of insulin-dependent or gestational diabetes, the provider should be prepared for control of blood sugars.
In the event of an acutely distressed fetus, indicative of fetal hypoxia, the use of prophylactic steroids should not delay the delivery of an acutely distressed fetus.
Although the use of repeated doses of glucocorticoids remains controversial, a meta-analysis concluded that repeated doses of prenatal corticosteroids in women who remained at risk for preterm birth 7 or more days after an initial course reduced the risk of their infants developing respiratory distress syndrome and reduced serious infant outcomes (relative risk 0.83 and 0.84, respectively). Treatment with repeat doses was associated with a reduction in mean birthweight of approximately 76 g; however, no differences in growth assessments or disabilities at early childhood were noted in follow-up.  In view of these conclusions, clinicians may consider use of a single repeated dose of glucocorticoids (rescue dose) at least 7-14 days from the initial treatment if the patient remains at significant risk for preterm delivery within the next 7 days, at a gestational age less 34 weeks.
Group B streptococci prophylaxis
All patients in preterm labor should be considered at high risk for neonatal GBS sepsis. Patients in preterm labor with the potential to deliver should receive prophylactic antibiotics against GBS, unless GBS culture is negative. Prophylactic antibiotics should be administered when the diagnosis of preterm labor is made and should be continued until delivery or for a minimum of 72 hours. Patients should be re-treated if preterm labor recurs or when the patient enters labor at term depending upon culture results.
The most common tocolytic agents used for the treatment of preterm labor are magnesium sulfate (MgSO4), indomethacin, and nifedipine. In the past, beta-mimetic agents, such as terbutaline or ritodrine, were the agents of choice, but in recent years their use has been significantly curtailed due to maternal and fetal side effects, such as maternal tachycardia, hyperglycemia, and palpitations The use of these agents can lead to pulmonary edema, myocardial ischemia, and cardiac arrhythmia.
In February 2011, the US Food and Drug Administration (FDA) required the addition of a new Black Box Warning and contraindication to the terbutaline prescribing information to warn about the risk of use for preterm labor. The decision was based on reports of death and serious adverse reactions, including tachycardia, transient hyperglycemia, hypokalemia, arrhythmias, pulmonary edema, and myocardial ischemia following prolonged administration of oral or injectable terbutaline to pregnant women. The FDA concluded that the risk of serious adverse events outweighs any potential benefit to pregnant women receiving prolonged treatment with terbutaline injection (>48-72 h) or acute or prolonged treatment with oral terbutaline.
The tocolytic agents currently used to treat preterm labor appear to be equally efficacious in delaying delivery for at least 48 hours. While MgSO4 is associated with more maternal toxicity, indomethacin is associated with more fetal and neonatal toxicity.
Haas et al analyzed randomized controlled trials of tocolysis to determine the optimal first-line tocolytic agent for treatment of premature labor. Fifty-eight studies satisfied the inclusion criteria. A random-effects meta-analysis showed that all tocolytic agents were superior to placebo or control groups at delaying delivery both for at least 48 hours (53% for placebo compared with 75-93% for tocolytics) and 7 days (39% for placebo compared with 61-78% for tocolytics). No statistically significant differences were found for the other outcomes, including the neonatal outcomes of respiratory distress and neonatal survival. The decision model demonstrated that prostaglandin inhibitors provided the best combination of tolerance and delayed delivery. 
Magnesium sulfate is widely used as the primary tocolytic agent because it has similar efficacy to terbutaline with far better tolerance. Common maternal side effects include flushing, nausea, headache, drowsiness, and blurred vision. The mother should be monitored for toxic effects, such as respiratory depression or even cardiac arrest, that can occur at supratherapeutic levels. In addition, magnesium sulfate readily crosses the placenta and may lead to respiratory and motor depression of the neonate.
Although magnesium sulfate is approved to prevent seizures in preeclampsia and for control of seizures in eclampsia, its use for preterm labor is off-label. In 2013, the FDA issued a safety alert warning not to exceed 5-7 days of continuous IV magnesium sulfate when the agent is used for preterm labor. [24, 25, 26] Longer treatment duration may lead to hypocalcemia in the developing fetus and result in neonates with skeletal abnormalities related to osteopenia. Hypermagnesemia causes hypocalcemia as a result of a decrease in the secretion of parathyroid hormone. 
Several observational studies have reported an association of antenatal treatment with magnesium sulfate for preterm labor or preeclampsia with a decreased risk of cerebral palsy in low birth weight or preterm infants. [30, 31, 32] While the use of magnesium sulfate for the prevention of cerebral palsy in preterm infants has been recently suggested, it has yet to receive universal acceptance.
Antenatal magnesium sulfate should be considered for use in women at high risk of delivery before 34 weeks' gestation, mainly in those with premature rupture of membranes, active labor, and planned delivery within 24 hours. Loading and maintenance doses, and the duration of the treatment specifically for neuroprotection should not normally exceed 6 g, 1-2 g/h, and 24 hours, respectively.
The use of magnesium sulfate usually requires baseline maternal laboratory evaluation, including CBC count and serum creatinine level, urine output greater than 30 mL/h, normal vital signs, and appropriate maternal mentation. The initial recommended loading dose is 4-6 g IV over 20 minutes, followed by a maintenance dose of 1-4 g/h depending on urine output and persistence of uterine contractions.
Maintenance of magnesium sulfate therapy requires careful assessment of maternal mentation, visual symptoms, DTRs, and cardiac rate with discontinuation whenever evidence of toxicity exists. Urine output should be carefully monitored and ideally maintained at greater than 50 mL/h. Limiting intravenous intake to prevent pulmonary edema may be prudent. Oral intake can be maintained at the discretion of the provider. Serum magnesium levels may be obtained 1 hour after the loading dose and then every 6 hours and the maintenance dosage should be titrated to maintain a serum level of 4-8 mg/dL.
Since the primary therapeutic goal of tocolysis is to delay preterm delivery within 48 hours from the initiation of steroid prophylaxis, little evidence suggests that extended MgSO4 therapy is beneficial. The authors recommend discontinuing magnesium sulfate therapy after 48 hours in most patients unless the gestational age is less than 28 weeks when a gain of an additional 3-4 days may significantly reduce neonatal morbidity and mortality. Due to the risk of toxicity, consulting an MFM specialist may be beneficial if magnesium sulfate is to be continued for more than 72 hours. Since no clinical evidence suggests that oral beta-mimetics, subcutaneous terbutaline pump, or oral magnesium compounds are effective in delaying preterm birth, alternative tocolysis is not currently recommended after the discontinuation of IV MgSO4 therapy.
When acute mild toxicity exists in the presence of normal urine output, magnesium sulfate should be temporarily discontinued until the serum magnesium level and DTRs return to normal. If the toxicity symptoms are life threatening, administering 1 g of calcium gluconate by slow intravenous push and strongly considering not reinstituting magnesium sulfate despite the return to normal levels is recommended.
Indomethacin is an appropriate first-line tocolytic for the pregnant patient in early preterm labor (< 30 wk) or preterm labor associated with polyhydramnios. A more significant inflammatory response in the membranes and decidua is observed at gestational ages less than 30 weeks compared with 30-36 weeks. Indomethacin reduces prostaglandin synthesis from decidual macrophages. The fetal renal effects of indomethacin may be beneficial to reduce polyhydramnios.
Prostaglandin synthetase inhibitors, such as indomethacin, have been shown to have efficacy similar to that of terbutaline but are associated with infrequent maternal side effects. However, these agents readily cross the placenta and can cause oligohydramnios due to a decrease in fetal renal blood flow if used for more than 48 hours. The administration of indomethacin is often limited to 48 hours, and baseline labs, including CBC count and liver function tests (LFTs), should be ordered prior to initiation of therapy.
During treatment, urine output, maternal temperature, and amniotic fluid index (AFI) should be evaluated periodically. The initial recommended dose is 100 mg PR followed by 50 mg PO every 6 hours for 8 doses. If oligohydramnios occurs, the amniotic fluid usually reaccumulates when the indomethacin is stopped, but persistent fetal anuria, renal microcystic lesions, and neonatal death have been reported. Indomethacin can also cause premature closure or constriction of the ductus arteriosus. Since this effect is more common after 32 weeks' gestation, indomethacin therapy is not usually recommended after 32 weeks.
Nifedipine, a calcium channel blocker, is commonly used to treat high blood pressure and heart disease because of its ability to inhibit contractility in smooth muscle cells by reducing calcium influx into cells. Consequently, nifedipine has emerged as an effective and safe alternative tocolytic agent for the management of preterm labor. Despite its unlabeled status, several randomized studies have shown that the use of nifedipine in comparison with other tocolytics is associated with a more frequent successful prolongation of pregnancy, resulting in significantly fewer admissions of newborns to the neonatal intensive care unit, and may be associated with a lower incidence of RDS, necrotizing enterocolitis, and intraventricular hemorrhage.
A systemic review by Conde-Aqudelo found that nifedipine was associated with significantly fewer maternal adverse events than β2 -adrenergic-receptor agonists and magnesium sulfate for tocolysis in women with preterm labor. 
A recommended initial dosage of nifedipine is 20 mg orally, followed by 20 mg orally after 30 minutes. If contractions persist, therapy can be continued with 20 mg orally every 3-8 hours for 48-72 hours with a maximum dose of 160 mg/d. After 72 hours, if maintenance is still required, long-acting nifedipine 30-60 mg daily can be used.
Contraindications of nifedipine therapy include allergy to nifedipine, hypotension, hepatic dysfunction, concurrent use of beta-mimetics or MgSO4, transdermal nitrates, or other antihypertensive medication. Other commonly reported side effects of nifedipine are maternal tachycardia, palpitations, flushing, headaches, dizziness, and nausea. Continuous monitoring of the fetal heart rate is recommended as long as the patient has contractions; the patient's pulse and blood pressure should be carefully monitored. Pregnant women with liver disease should not be prescribed nifedipine.
A true episode of preterm labor becomes a powerful risk factor for recurrent preterm birth, in addition to other risk factors present prior to the current episode. The prior risk factor may have been modified; for example, infection may have been identified and treated or behavioral risk factors may have been modified. Little evidence indicates that prophylactic oral beta-mimetic, subcutaneous beta-mimetics, or oral magnesium gluconate reduce the incidence of recurrent preterm birth and therefore should not be prescribed.
Frequent contact, face-to-face or by telephone, with a knowledgeable provider appears to be as effective as home uterine activity monitoring (HUAM) or continued pharmacological therapy. Direct contact with the patient is supplemented by education and phone access to a knowledgeable, consistent provider. Some unique situations exist in which HUAM is still felt to be beneficial, including patients who are paraplegic and unable to appreciate any muscular contractions.
The goal of follow-up therapy is to maximally reduce recurrence risk and to speed the access to subspecialty care if preterm labor should recur.
Once the episode of preterm labor has been arrested, a gradual return to limited activity should be encouraged prior to hospital discharge. The following factors may influence the decision to discharge the patient:
Number of fetuses
Access to the hospital
Social support at home (transportation at all times, telephone)
The ability to maintain limited activity and pelvic rest
Good patient compliance
If the patient was referred to a subspecialty care facility, the local obstetrical and pediatric providers should be comfortable with home management. If labor should recur, they may have to manage the rapid delivery of premature infant.
The patient should be informed regarding the signs and symptoms of recurrent preterm labor. The critical signs of recurrent preterm labor include contractions greater than 4 per hour, rhythmic back or thigh pain, increasing pelvic pressure, unusual discharge, vaginal spotting/bleeding, or rupture of membranes.
The provider should have increased contact with the patient, and the patient should be directed to a specific individual to report symptoms of preterm labor or complications. The contact may be via a combination of telephone contacts and office visits. When genital tract infection may have played a role in the preterm labor; a repeat culture may be recommended 2-4 weeks after discharge.
If inpatient tocolysis was unsuccessful and the patient delivered preterm, the patient and family should receive education concerning the etiology and risk of recurrence in subsequent pregnancies. Few etiologies exist for which prediction of subsequent preterm delivery in future pregnancies is 100% accurate. Time should be spent at the postpartum visit reviewing the patient's clinical history, laboratory data, and pathology reports. Preconceptual counseling may also be critical in the decision of the patient to again become pregnant and in managing her pregnancy.
Delivery of the preterm infant involves several unique issues (avoidance of excessive trauma, caution in the use of vacuum prior to 34 weeks’ gestation, risk of breech head entrapment, etc) to optimize newborn well-being. The American College of Obstetricians and Gynecologists has endorsed a policy of delayed cord clamping in both term and preterm newborns as this practice increases newborn hemoglobin and, in preterm infants reduces rates of intraventricular hemorrhage, necrotizing enterocolitis and required transfusions. Specifically, ACOG recommends a delay in umbilical cord clamping in vigorous term and preterm infants for at least 30–60 seconds after birth, though care must be individualized and provided only to infants not requiring immediate resuscitation.