Uterocervical angle as a predictor of preterm birth on a high-risk collective between 20 and 31 weeks of gestation: A cohort analysis

Introduction The cervical length (CL) measurement is a widely used method to estimate the risk of preterm birth. Due in particular to the high false-positive rate, the establishment of markers with improved test characteristics is a great challenge. A potential predictor of preterm birth is the uterocervical angle (UCA) and this additional measurement may improve the risk assessment. It was the aim of this study to compare the test properties of CL and UCA on patients at risk for preterm birth. Material and methods 109 patients with at least one of the following signs of threatening preterm birth between 20(+0/7)and 31(+6/7) weeks were included in a prospective cohort analysis: regular (>3/30 min) or painful uterine contractions, CL below 25 mm or a history of preterm birth. Exclusion criteria were premature rupture of membranes, hypertensive disorders, vaginal bleeding, surgical cerclage, Arabin pessary or cervical dilation of more than 30 mm. The determination of the UCA was carried out in a standardized manner using the image documents captured by vaginal sonographic CL measurement. The primary endpoint was preterm birth <34 weeks, secondary endpoints were delivery The UCA was on average 103 degrees and the mean UCA in preterm and term groups did not differ significantly (P = .924). The UCA was not predictive for threatened preterm birth, even if only singletons were considered. For CL the best predictive accuracy for preterm birth <34 weeks was observed at a cut-off value of 14 mm with sensitivity 0.50, specificity 0.80, positive predictive value 0.30, negative predictive value 0.90, positive likelihood ratio 2.4, negative likelihood ratio 0.6 and an odds ratio of 3.9 (95% confidence interval 1.3-11.7,P = .016). Conclusions The assessment of UCA in patients at risk for preterm birth was not suitable to predict the probability of a threatened preterm birth. Measurement of UCA cannot be recommended in this situation.