Cervical sonography compared to manual examination as a predictor of preterm delivery

OBJECTIVE: To compare the accuracy of sonographic and manual cervical exams for the prediction of preterm delivery. STUDY DESIGN: 102 singleton pregnancies at high risk for preterm delivery were followed prospectively from 14 to 30 weeks with both serial cervical ultrasound measurements and manual examinations of the length of the cervix. The main outcome studied was GA at delivery: <35 weeks or ≥35 weeks. RESULTS: Excluding 6 induced preterm deliveries. 96 pregnancies were analyzed. The mean cervical length measured by ultrasound was 20.6 mm in pregnancies delivered preterm (n=17) and 31.6 mm in pregnancies delivered at term (n=79) (p=.003): the mean cervical lengths measured by manual Examination were 16.1 mm and 18.6 mm in the same preterm and term pregnancies, respectively (p=.06). Adjusted for gestational age, measurement of cervical length bv ultrasound predicted preterm delivery most accurately at the 16th and 20th week examinations (p<.0005). Using cutoff values which represented the 25th percentile (or the range of measurements found, the relative risk for preterm deliver was 4.8 (95% CI 2.1-11.1: p=.0004) for a sonographic cervical length of <25 mm and 2.0 (95% CI 0.5-4.7; p = .1) for a manual cervical length of <16 mm. Ultrasound examinatior had a sensitivity=59%, specificity=85%, PPV= 45%, and NPV= 91%. Manual examination of the cervix had a sensitivity=41 %, specificity=77%, PPV=28%, and NPV=86%. CONCLUSION: Cervical length measured by ultrasound is a better predictor of preterm delivery than cervical length measured by manual examination. A cervical length ≤ 25mm at 16-20 weeks is associated with a 4 fold increase in the risk of preterm birth.