Follistatin-free activin A is not associated with preterm birth.

Abstract

The purpose of this study was to determine whether follistatin-free activin A (an inhibin-related protein with rising serum levels before term labor) is associated with spontaneous preterm birth in an outpatient population. From 10 centers, 2929 women were enrolled in the Preterm Prediction Study. Plasma was isolated from blood samples obtained at 24 and 28 weeks of gestation and stored. A nested case-controlled study was performed; there were 197 women with spontaneous preterm birth before 37 weeks of gestation with 24-week samples and 142 cases with 28-week samples that were matched to an equal number of term control samples from each time point. A follistatin-free activin A enzyme-linked immunosorbent assay was performed, with results interpreted as positive or negative (defined as a value less than the assay limit of sensitivity). A comparison of follistatin-free activin A with other tests that were associated with spontaneous preterm birth (alpha-fetoprotein, alkaline phosphatase, cortisol, granulocyte colony-stimulating factor, fetal fibronectin, and cervical interleukin-6) was also performed. The odds ratio for associated spontaneous preterm birth at <37 weeks of gestation if the follistatin-free activin A result was positive was 1.16 at 24 weeks (95% CI, 0.62-2.17) and 1.39 at 28 weeks (95% CI, 0.68-2.84). A positive follistatin-free activin A result at either 24 or 28 weeks was not associated with a positive test for any of the other laboratory tests, except granulocyte colony-stimulating factor at 28 weeks' gestation. Follistatin-free activin A is not associated with spontaneous preterm birth in an asymptomatic outpatient population. Follistatin-free activin A may not be an appropriate screening marker for spontaneous preterm birth in the office setting.