ijmf_a_1771553_sm5570.docx (29.9 kB)

Uterocervical angle as a predictor of spontaneous preterm birth in twin pregnancies

Download (29.9 kB)
journal contribution
posted on 29.05.2020 by Marta Benito Vielba, Carlos De Bonrostro Torralba, Andrea Espiau Romera, Montserrat Roca Arquillue, José Manuel Campillos Maza, Sergio Castán Mateo

Objective: To assess the ability of uterocervical angle (UCA) compared with cervical length (CL) to predict the risk of spontaneous preterm birth (sPTB) in twin pregnancies and its performance when it was included in a combined predictive model of clinical and ultrasonographic parameters.

Methods: We conducted a retrospective cohort study of twin pregnancies undergoing transvaginal ultrasound between 19+0–21+6 weeks to measure CL during routine second trimester scan from January 2015 through December 2016. Recorded ultrasound images of CL were reassessed to evaluate UCA. Medical and obstetric data were also collected for statistical analysis. A logistic regression model was created for predicting sPTB including UCA and other variables.

Results: A total of 177 women were included. The rates of sPTB rate below 28, 32 and 34 weeks of gestation were 4.5%, 6.8% and 12.4%, respectively. ROC curves showed a better area under the curve (AUC) for UCA at all gestational ages compared with CL (AUC for sPTB <28 weeks 0.840 (p = .005) vs 0.627 (p = .388); AUC for sPTB <32 weeks 0.706 (p = .022) vs 0.619 (p = .255); AUC for sPTB <34 weeks 0.674 (p = .008) vs 0.568 (p = .344). UCA >117 degrees was significantly associated with sPTB <28 weeks (p = .002; OR 15.3; CI 1.66–142.37; NPV, 99.2%), <32 weeks (OR 3.84; p = .031) and <34 weeks of gestation (OR 3.10; p = .016). Based on multivariate analyses, the best predictive model included uterocervical angle (p = .032), cervical length (p = .002) and maternal height (p = .001) (Nagelkerke R2 0.944).

Conclusion: In our study, an UCA > 117 degrees allowed to identify those women with twin pregnancies at risk of sPTB and performed better than CL measurement. Our combined prediction model was able to adequately predict the risk of sPTB in the twin pregnancies of our research.