Comparison of Cervicovaginal Fluid beta Human Chorionic Gonadotropin and Fetal Fibronectin in the Prediction of Preterm Delivery

Abstract

Background: Preterm birth is among the leading causes of perinatal morbidity and mortality globally and it’s reported to be on the increase. Identifying a reliable marker that best suggests those who may have preterm delivery would allow for adequate preparations that may limit the morbidity / mortality associated with preterm birth. To date, the search for such a predictor is an area of ongoing research. However, many substances have been tried. Objective: To compare the validity of cervicovaginal fluid beta human chorionic gonadotrophin (β-hCG) and fetal fibronectin (fFN) in the prediction of preterm delivery among at risk women. Materials and Methods: This was a longitudinal cohort study involving two hundred pregnant women between the gestational ages of 24-36 weeks. Cervicovaginal fluid samples where obtained from the posterior vaginal fornix for quantitative assessment of β-hCG and fFN respectively. The patients were recruited using purposive sampling method. After counselling and obtaining an informed consent from the patient, a sterile Cusco speculum was passed into the vagina and samples collected from the posterior vagina fornix using swab sticks. Two samples were collected from the same individual, each sample was tested for beta human chorionic gonadotrophin and fetal Fibronectin. The women were followed up till delivery. Results: One hundred and ninety-two participants were analyzed. Eight were lost to follow-up. Ninety-eight (51%) had positive β-hCG and 94(49%) had negative β-hCG while 47(24.5%) had positive fibronectin and 145(74.5%) had negative results. Thirty-five (56.5%) of preterm and 63(48.5%) of term had positive β-hCG (> 77.8mlU/ml) while 15(24.2%) of preterm and 32(24.6%) of term had positive fibronectin (>50ng/mL). Both were not statistically significant with p- value of 0.3 and 0.949 respectively. The mean cervicovaginal β-hCG level (mIU/ ml) in preterm group was 106.6±66.8 and term group was 83.5±50.1. There was statistically significant difference between preterm and term in β-hCG values, which implies that β-hCG preterm average is higher than term with (P value- 0.018). However, the mean cervicovaginal fibronectin level (ng/mL) in preterm group was 44.04 ± 56.1 and term group was 40.1±59.5. There was no statistically significant association of raised fibronectin with preterm (P value 0.696).

Quantitative cervicovaginal β-hCG test had sensitivity 56.5%, specificity 51.5%, positive predictive value 35.7%, negative predictive value 71.3% and diagnostic accuracy of 53.1% at a cut-off value of > 77.8mlU/ml in the prediction of preterm delivery. On the other hand, quantitative cervicovaginal fibronectin test had sensitivity 24.2%, specificity 75.4%, positive predictive value 31.9%, negative predictive value 67.6% and diagnostic accuracy of 58.9% at cut-off value of 50ng/mL for prediction of preterm delivery. The area under the curve shows that the two markers have similar ability in predicting PTL. In the diagram below, fibronectin has a predicting value with area under curve of 0.498 which is comparable to 0.542 of β-hCG in predicting PTL. Conclusion: Cervicovaginal β-hCG can be used as an alternative in place of fibronectin in the prediction of preterm delivery in asymptomatic high risk woman.