Introduction
Quantitative fetal fibronectin (qfFN) and cervical length (CL) are reliable predictive markers for spontaneous preterm birth (sPTB) in high-risk asymptomatic women. However, it is important to consider the potential treatment paradox in this population. This study aims to investigate the impact of intervention (cerclage or progesterone) on the predictive ability of qfFN and CL.
Methods
This was a planned analysis of data from a large prospective cohort (EQUIPP, Evaluation of Fetal Fibronectin for the Prediction of Preterm Birth, n=2141). Women were grouped according to intervention (cerclage, progesterone or both). qfFN and CL measurements from the first visit between 22+0 to 27+6 were selected for analysis.
The primary outcome was sPTB <34 weeks’ gestation. Reciever Operating Characteristic (ROC) curves were generated. Area Under the Curve (AUC) was used to evaluate the predictive accuracy of each test with cerclage, progesterone or both interventions in situ.
Results
Predictive accuracy of CL and qfFN for the whole cohort [AUC 0.81 (0.76 to 0.86) and AUC 0.72 (0.64 to 0.80)] was comparable to currently published data.
In women with cerclage in situ, test performance for CL and qfFN was maintained [AUC 0.79 (0.71 to 0.86) and AUC 0.72 (0.63 to 0.80), respectively]. Predictive statistics were similar if not better in women who received both interventions (cerclage and progesterone) [AUC 0.91 (0.81 to 1.00) for CL and AUC 0.72 (0.54 to 0.89) for qfFN].
Conclusion
In spite of theoretical concerns that predictive ability will be altered by the presence of prophylactic interventions, both qfFN and CL appear to remain reliable predictors for sPTB <34 weeks’ with intervention in situ. CL may even be better in this population. This new evidence should reassure clinicians when assessing risk in women with intervention in situ.