Cervical length and quantitative fetal fibronectin in the prediction of spontaneous preterm birth in asymptomatic women with congenital uterine anomaly

BACKGROUND: Congenital uterine anomalies are associated with late miscarriage and spontaneous preterm birth. OBJECTIVE: Our aim was 1) to determine the rate of spontaneous preterm birth in each type of congenital uterine anomaly, and 2) to assess the performance of quantitative fetal fibronectin and cervical length measurement by transvaginal ultrasound in asymptomatic women with congenital uterine anomalies for the prediction of spontaneous preterm birth at <34 and <37 weeks of gestation. MATERIALS AND METHODS: This was a retrospective cohort of women with congenital uterine anomalies asymptomatic for spontaneous preterm birth, from 4 tertiary referral centers in the United Kingdom (2001-2016). Congenital uterine anomalies were categorized into fusion (unicornuate, didelphic, and bicornuate uteri) or resorption defects (septate, with or without resection, and arcuate uteri), based on pre-pregnancy diagnosis. All women underwent serial transvaginal ultrasound cervical length assessment in the second trimester (16 to 24 weeks' gestation); a subgroup underwent quantitative fetal fibronectin testing from 18 weeks' gestation. We investigated the relationship between congenital uterine anomalies and predictive test performance for spontaneous preterm birth at <34 and <37 weeks' gestation. RESULTS: A total of 319 women were identified as having congenital uterine anomalies in our high-risk population. Of the women, 7% (23/319) delivered spontaneously at <34 weeks' gestation and 18% (56/319) at <37 weeks' gestation. Rates of spontaneous preterm birth by type were as follows: 26% (7/27) for unicornuate, 21% (7/34) for didelphic, 16% (31/189) for bicornuate, 13% (7/56) for septate, and 31% (4/13) for arcuate. In all, 80% (45/56) of women who had spontaneous preterm birth at <37 weeks did not develop a short cervical length (<25 mm) during the surveillance period (16-24 weeks). The diagnostic accuracy of short cervical length had a low sensitivity (20.3) for predicting spontaneous preterm birth at <34 weeks. Cervical length had an area under the receiver operating curve of 0.56 (95% confidence interval, 0.48-0.64) and 0.59 (95% confidence interval, 0.55-0.64) for prediction of spontaneous preterm birth at <34 and <37 weeks, respectively. The area under the curve for cervical length to predict spontaneous preterm birth at <34 weeks was 0.48 for fusion defects (95% confidence interval, 0.39-0.57) but 0.78 (95% confidence interval, 0.66-0.91) for women with resorption defects. Overall quantitative fetal fibronectin had an area under the curve of 0.63 (95% confidence interval, 0.49-0.77) and 0.58 (95% confidence interval, 0.49-0.68) for prediction of spontaneous preterm birth at <34 and <37 weeks, respectively. The area under the curve for prediction of spontaneous preterm birth at <37 weeks with quantitative fetal fibronectin for fusion defects was 0.52 (95% confidence interval, 0.41-0.63) but 0.79 (95% confidence interval, 0.63-0.95) for women with resorption defects. Results were similar when women with intervention were excluded. CONCLUSION: The commonly used markers cervical length and quantitative fetal fibronectin have utility in prediction of spontaneous preterm birth in resorption congenital uterine defects but not in fusion defects. This is contrary to findings in other high-risk populations. These findings need to be accounted for when planning antenatal care, and have potential implications for predictive tests used in spontaneous preterm birth surveillance and intervention.