Evaluation of additive effect of quantitative fetal fibronectin to cervical length for prediction of spontaneous preterm birth among asymptomatic low-risk women.

Abstract

Our objective was to evaluate the possible additive effect of quantitative fetal fibronectin to transvaginal ultrasound cervical length measurement between 18(0/7) and 23(6/7) weeks for prediction of spontaneous preterm birth at <37(0/7) weeks among asymptomatic low-risk women. A prospective observational study was performed of asymptomatic women with singleton gestations between 18(0/7) and 23(6/7) weeks and no prior spontaneous preterm birth. Women with multiple gestations, rupture of membranes, vaginal bleeding, intercourse or vaginal exam within 48h of enrollment were excluded. Physicians were blinded to the quantitative fetal fibronectin levels, but the cervical length measurements were made available. The primary outcome was spontaneous preterm birth at <37(0/7) weeks. Of the 528 asymptomatic low-risk women who were prospectively enrolled, 36 (6.82%) had spontaneous preterm birth at <37(0/7) weeks. Using the receiver-operating characteristic curve, fetal fibronectin value of ≥5ng/mL was identified as the optimal cut-off for predicting spontaneous preterm birth at <37(0/7) weeks. As compared with cervical length ≥20mm alone, with the use of cervical length ≤20mm or quantitative fetal fibronectin ≥5ng/mL as screening criteria for prediction of spontaneous preterm birth at <37(0/7) weeks; sensitivity improved from 11.11 to 61.11%, specificity decreased from 99.59 to 55.08%, positive predictive value decreased from 66.67 to 9.05%, negative predictive value marginally improved from 93.87 to 95.09% and predictive accuracy decreased from 93.56 to 55.49%. Although the sensitivity improved, other predictive statistics and predictive accuracy did not improve by the addition of mid-trimester quantitative fetal fibronectin to cervical length measurement. Therefore, addition of mid-trimester quantitative fetal fibronectin to cervical length measurement cannot be recommended at this time for prediction of spontaneous preterm birth at <37(0/7) weeks in asymptomatic low-risk women.