EP14.23: Ultrasound screening for preterm birth in twin pregnancies: a prospective longitudinal pilot study

To compare the efficacy of available cervical parameters using transvaginal 2D ultrasound and elastography for the prediction of preterm birth in twin gestations. A population of twin pregnant women (n 36) was subjected to seriate ultrasound evaluation during II and III trimesters. The following parameters have been collected: uterocervical angle (UCA), hardness ratio (HR), elasticity contrast index (ECI), internal os strain (IOS), external os strain (EOS), and cervical length (CL). The mean values of each parameter have been compared between the women (n 24) who delivered at term (after 37 weeks for dichorionic twin pregnancies, 36 weeks for monochorionic, according to ISUOG) and those who delivered before the planned date (n 12). ROC curves for each parameter were calculated and compared to find the best predictor for preterm birth in twin gestations. No differences have been found in the comparison of baseline characteristics or neonatal outcomes between the two groups. A statistically significant difference has been pointed out for UCA (p 0,001) and IOS/EOS ratio (p 0,031), but not for the CL (p 0,128). A wider angle between the cervix and the uterus is related to a higher risk of preterm birth. UCA appeared to be the most predictive parameter for premature delivery before the due date (AUC 0,771), far better than the CL (AUC 0,354). In twin pregnancies, the CL is not the best predictor of preterm birth, unlike singleton gestations. The uterocervical angle and the cervical strain are indirect signs of the increasing tension of the uterus due to typical overdistension in twin pregnancies. Alternative ultrasound parameters, in particular UCA, can be helpful in the screening and management of twin gestations at risk of premature labour or rupture of membranes. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.