Objective To compare the prediction value of cervical length measured by two transvaginal ultrasonography method for preterm birth in early stage of second trimester.

Methods A total of 477 cases of pregnant women at 14 to 15 weeks of gestation were selected as the study objects. The cervical length was measured by transvaginal ultrasonography with three-point method and two-point method respectively. According to the symptoms of threatened preterm birth and pregnancy outcome during the follow-up period, the subjects were divided into full-term group, threatened preterm birth group and preterm birth group. The correlation between pregnancy outcome and cervical length was analyzed; the predictive value of the two measures for preterm birth was compared.

Results Of the 477 parturients, 51 were preterm delivery, 136 were threatened preterm delivery, and 290 were full-term delivery. The cervical length of the preterm delivery group was significantly lower than that of the threatened preterm delivery group and the term preterm delivery group, and the cervical length of threatened preterm delivery group was significantly lower than that of full-term delivery group (P < 0.05). In the parturients of same group, the cervical length measured by three-point method was significantly greater than that by two-point method (P < 0.05). The specificity, accuracy and positive predictive value of three-point method were significantly higher than those of two-point method, and the false positive rate was significantly lower than that of two-point method (P < 0.05); there was no significant difference in sensitivity and negative predictive value between three-point method and two-point method (P>0.05).

Conclusion The measurement of cervical length by transvaginal ultrasound in the middle and early pregnancy has a good predictive value for premature delivery, and the three-point method can improve the accuracy of cervical length measurement.