Risk factors and predictors of preterm delivery.

Objective The aim of this study was to assess whether a combination of risk factors and predictors for preterm delivery, such as demographic data, cervical score, serum ferritin, vaginal culture, fetal fibronectin test (fFN) and tocolytic treatment, was more useful than individual variables in identifying pregnancies with preterm delivery. Methods The study population included 94 asymptomatic women, with a gestational age of 24-27 weeks (by sonography data), receiving prenatal care in the Gynecology and Obstetrics Institute of Padova University, at risk for preterm birth. The anamnestic risk factors were limited to previous preterm delivery and second-trimester abortion. The actual risk factors were limited to multiple gestation, prophylactic cervical cerclage, congenital uterine anomaly, polyhydramnios, leiomyoma and intra uterine growth restriction. As a control group, 201 uncomplicated, symptom-free singleton pregnancies were enrolled from normal pregnant women. Statistical analysis was performed with bivariate frequency analysis (chi(2) test) and multiple logistic regression analysis. Results There was a statistically significant association between preterm birth and maternal age (p = 0.004), previous preterm delivery (p = 0.004), pregnant status (p < 0.001), pelvic score (p < 0.001) and fFN test either rapid (p < 0.001) or enzyme-linked immunosorbent assay (p < 0.001). By logistic multiple regression analysis, we identified a logistic model, using pregnant status and fFN test, predicting a probability of preterm delivery ranging from 16.1% (both predictors negative) to 96.4% (both predictors positive). Conclusion Our findings confirmed the usefulness of the fFN test in predicting preterm delivery, especially in asymptomatic patients without risk factors. A positive fFN test might represent a marker of chorionic-decidual cell activation.