Quantitative fetal fibronectin testing in combination with cervical length measurement in the prediction of spontaneous preterm delivery in symptomatic women.

Abstract

To evaluate whether in symptomatic women, the combination of quantitative fetal fibronectin (fFN) testing and cervical length (CL) improves the prediction of preterm delivery (PTD) within 7days compared with qualitative fFN and CL. Post hoc analysis of frozen fFN samples of a nationwide cohort study. Ten perinatal centres in the Netherlands. Symptomatic women between 24 and 34weeks of gestation. The risk of PTD<7days was estimated in predefined CL and fFN strata. We used logistic regression to develop a model including quantitative fFN and CL, and one including qualitative fFN (threshold 50ng/ml) and CL. We compared the models' capacity to identify women at low risk (<5%) for delivery within 7days using a reclassification table. Spontaneous delivery within 7days after study entry. We studied 350 women, of whom 69 (20%) delivered within 7days. The risk of PTD in <7days ranged from 2% in the lowest fFN group (<10ng/ml) to 71% in the highest group (>500ng/ml). Multivariable logistic regression showed an increasing risk of PTD in <7days with rising fFN concentration [10-49ng/ml: odds ratio (OR) 1.3, 95% confidence interval (95% CI) 0.23-7.0; 50-199ng/ml: OR 3.2, 95% CI 0.79-13; 200-499ng/ml: OR 9.0, 95% CI 2.3-35; >500ng/ml: OR 39, 95% CI 9.4-164] and shortening of the CL (OR 0.86 per mm, 95% CI 0.82-0.90). Use of quantitative fFN instead of qualitative fFN resulted in reclassification of 18 (5%) women from high to low risk, of whom one (6%) woman delivered within 7days. In symptomatic women, quantitative fFN testing does not improve the prediction of PTD within 7days compared with qualitative fFN testing in combination with CL measurement in terms of reclassification from high to low (<5%) risk, but it adds value across the risk range. Quantitative fFN testing adds value to qualitative fFN testing with CL measurement in the prediction of PTD.