Predictive value of two-dimensional and three-dimensional multiplanar ultrasound evaluation of the cervix in preterm labor.

OBJECTIVE: To compare the predictive value of conventional two-dimensional ultrasound measurement of cervical length and three-dimensional multiplanar volume assessment of the cervix for delivery at < 37 weeks of gestation among patients with preterm labor.

MATERIALS AND METHODS: This preliminary prospective study was conducted in 28 patients hospitalized for preterm labor between 24(+0) and 33(+6) weeks of gestation, defined by regular and painful uterine contractions (at least two per 10 min), intact membranes and a cervical length of < or = 26 mm measured by two-dimensional transvaginal ultrasonography at admission with a Voluson 530 (Kretz, France) machine using a 7.5-MHz transvaginal transducer. The cervical volume was then assessed by the three-dimensional triplan technique. It was automatically calculated after drawing of the cervix outlines in mid-sagittal and median axial plane images. As the border between the cervix and the lower uterine segment is virtual, it was defined by a perpendicular line to the line joining the external os and the internal os. The result of the latter measurement was not disclosed to the obstetric team. The primary outcome was the rate of deliveries at < 37 weeks. We constructed a receiver operator characteristic (ROC) curve to determine the optimal cut-off point of the cervical volume, to predict preterm delivery.

RESULTS: Sixteen (57%) patients delivered at < 37 weeks. Mean (+/- SD) gestational ages at admission and delivery were 28.2 (+/- 3.2) weeks and 35.9 (+/- 4.1) weeks, respectively. Mean cervical length and volume at admission were 20.1 (+/- 9.9) mm and 23.1 (+/- 14.6) mm3, respectively. The ROC curve showed that the optimal cut-off point was 20 mm3. The predictive values of cervical length and of cervical volume on delivery at < 37 weeks were: sensitivity, 87.5% (14/16) and 75% (12/16); specificity, 50% (6/12) and 75% (9/12); positive predictive value, 70% (14/20) and 80% (12/15); negative predictive value, 75% (6/8) and 69.2% (9/13), respectively.

CONCLUSIONS: The three-dimensional multiplanar volume assessment of the cervix probably increases the positive predictive value of cervical ultrasonography in predicting preterm delivery. Screening high-risk women could be achieved by conventional two-dimensional transvaginal ultrasound and the diagnosis of true preterm labor may be improved by three-dimensional multiplanar transvaginal ultrasound assessment of the cervix.