The committee on Practice Bulletins—Obstetrics of the American College of Obstetricians and Gynecologists (ACOG) has issued new recommendations on assessing risk factors for preterm birth. ACOG Practice Bulletin No. 31, which replaces Technical Bulletin No. 206 issued in June 1995, appears in the October 2001 issue of Obstetrics and Gynecology. These recommendations discuss risk factors, markers for predicting preterm birth, and clinical considerations.

Diagnosis and Predicting Risk

According to ACOG, preterm labor is defined as regular contractions associated with cervical change before 37 weeks' gestation. Spontaneous preterm birth includes preterm labor, preterm spontaneous rupture of membranes, and cervical incompetence. The pathophysiologic events that trigger preterm birth are largely unknown but may include decidual hemorrhage (abruption), uterine overdistention, and hormonal changes possibly mediated by fetal or maternal stress. Certain bacterial infections also have been associated with preterm labor, including Ureaplasma urealyticum, Mycoplasma hominis, Gardnerella vaginalis, and Peptostreptococcus and Bacteroides species. These organisms are usually of low virulence, and it is unclear whether they are etiologic or associated with an acute inflammatory response of another etiology.

Predicting risk of preterm labor is only valuable if there is an available intervention that is likely to improve the situation. Even though maternal tocolytic and steroid therapies may prolong pregnancy and decrease morbidity and mortality, they should be limited to those with true preterm labor at high risk for spontaneous preterm birth because of possible fetal and maternal consequences. Identifying women at risk allows for appropriate transfer to a facility with the needed resources, and it avoids the use of unnecessary interventions in those at low risk.

Biologic Markers

Home Uterine Activity Monitoring

Home uterine activity monitoring (HUAM) has been suggested as a method for predicting preterm birth in high-risk women. It combines telemetric recordings of uterine contractions using a tocodynamometer and daily telephone calls from a physician to offer support and advice. This method was based on the observation that some women who give birth early have an increase in uterine activity earlier in pregnancy than women who give birth at term, and that these uterine contractions may not be recognized by the patient.

Studies of HUAM vary in design, inclusion criteria, and measurements of endpoints and outcomes, making comparisons difficult. Because of study limitations, varying results, and flawed study designs, there is uncertainty about the usefulness of HUAM. The U.S. Preventive Services Task Force and the U.S. Food and Drug Administration concluded that the device was not effective. Data are insufficient to support HUAM in preventing preterm birth, and it is not recommended.

Salivary Estriol

Observational studies have shown that maternal levels of serum estradiol and salivary estriol increase before the onset of spontaneous term and preterm labor. A test using salivary estriol levels was designed to predict preterm delivery, but maternal estriol levels peak at night and may be suppressed by betamethasone administration.

The test carries a high percentage of false-positive results and can add significantly to the cost of prenatal care if used in the low-risk population. Trials with salivary estriol testing have failed to establish its usefulness for anything more than investigational purposes.

Bacterial Vaginosis

The presence of bacterial vaginosis has been associated with preterm delivery independent of other known risk factors. Bacterial vaginosis is a common alteration of the normal vaginal flora and has been found in 10 to 25 percent of patients in general gynecology and obstetric patients, with 50 percent being asymptomatic.

Trials of screening and treatment for bacterial vaginosis in pregnant women to reduce the incidence of preterm birth have been conducted in mixed populations with varying results. Some trials have shown an association with the presence of bacterial vaginosis and preterm delivery, but most results have failed to show if treatment for this condition can prevent preterm birth. There is insufficient data to support screening and treating women at low and high risk.

Fetal Fibronectin Screening

Numerous trials have shown an association with the presence of the fetal fibronectin protein and preterm birth and a decrease in the risk of preterm birth when the test result for this protein is negative.

Although a negative test result appears to be useful in ruling out preterm delivery within two weeks, the clinical implications of a positive result have not been evaluated fully because no intervention has been shown to decrease the risk of preterm delivery. The test has limited usefulness in low-risk women. For high-risk women, the following criteria should be met: intact amniotic membranes, minimal cervical dilatation, testing should be performed no earlier than 24 weeks, zero days of gestation, and no later than 34 weeks, six days of gestation, and results must be available in time to allow for decision making, ideally within 24 hours.

Cervical Ultrasonography

Transvaginal cervical ultra-sonography is a reliable and reproducible way to assess the length of the cervix. Results of a prospective trial showed that the relative risk of preterm delivery increased as the cervical length decreased.

Despite the usefulness of measuring cervical length as a predictor of preterm delivery, it is not recommended because of lack of effective treatments.

Summary of Recommendations

The following recommendation from the ACOG committee is based on good and consistent scientific evidence (Level A):

• There are no current data to support using salivary estriol, HUAM, or bacterial vaginosis screening as strategies to identify or prevent preterm birth.

• The following ACOG recommendations are based on limited or inconsistent scientific evidence (Level B):

• Screening for risk of preterm labor by means other than historic risk factors is not beneficial in the general obstetric population.

• Ultrasonography to determine cervical length, fetal fibronectin testing, or a combination of both may be useful in identifying women at high risk for preterm labor. How ever, their clinical usefulness may rest primarily with their negative predictive value given the lack of proven treatment options to prevent preterm birth.

• Fetal fibronectin testing may be useful in women with symptoms of preterm labor to identify those with negative values and a reduced risk of preterm birth, thereby avoiding unnecessary intervention.