Background: The measurement of the frequency of uterine contractions has not been useful for reducing the rate of preterm delivery in randomized trials. Nonetheless, ambulatory monitoring of contractions continues to be used in clinical practice.
Methods: We assessed the frequency of uterine contractions as a predictor of the risk of spontaneous preterm delivery before 35 weeks of gestation. We enrolled women with singleton pregnancies between 22 and 24 weeks of gestation. The women used a contraction monitor at home to record contraction frequency twice daily on 2 or more days per week from enrollment to delivery or 37 weeks of gestation.
Results: We obtained 34,908 hours of successful monitoring recordings from 306 women. Although more contractions were recorded from women who delivered before 35 weeks than from women who delivered at 35 weeks or later, we could identify no threshold frequency that effectively identified women who delivered preterm infants. The sensitivity and positive predictive value of a maximal hourly frequency of contractions of four or more between 4 p.m. and 3:59 a.m. were 9 percent and 25 percent, respectively, at 22 to 24 weeks and 28 percent and 23 percent at 27 to 28 weeks. Other proposed screening tests, such as digital and ultrasound evaluations of the cervix and assays for fetal fibronectin in cervicovaginal secretions, also had low sensitivity and positive predictive value for preterm labor.
Conclusions: Although the likelihood of preterm delivery increases with an increased frequency of uterine contractions, measurement of this frequency is not clinically useful for predicting preterm delivery.