The association between maternal biomarkers and pathways to preterm birth in twin pregnancies.

OBJECTIVE

We sought to estimate the association between cervical length (CL) and fetal fibronectin (fFN) and each pathway leading to preterm birth in twin pregnancies.

METHODS

Cohort study of 560 patients with twin pregnancies who underwent routine serial CL and fFN screening from 22 to 32 weeks in one maternal fetal medicine practice during 2005-2013. We calculated the association between a short CL (≤20 mm) or positive fFN with overall preterm birth <32 weeks, and then subdivided the analysis into preterm birth <32 weeks from preterm labor, preterm premature rupture of membranes (PPROM) and indicated causes. We excluded cases of monochorionic-monoamniotic placentation, vasa previa, twin-twin transfusion and patients with cerclage.

RESULTS

The overall rate of preterm birth <32 weeks was 6.9% (3.9% from preterm labor, 1.6% from PPROM and 1.4% indicated). A short cervix was associated with preterm birth <32 weeks arising from preterm labor (12.4% versus 2.0%, p < 0.001), but not PPROM (1.9% versus 1.3%, p = 0.651). Positive fFN was associated with preterm birth <32 weeks both from preterm labor (17.0% versus 2.4%, p < 0.001) as well as from PPROM (5.7% versus 1.0%, p = 0.034). Neither was significantly associated with preterm birth <32 weeks from indicated causes.

CONCLUSIONS

The mechanism leading toward preterm influences the accuracy of screening tests chosen to assess risk in twin pregnancies. A shortened cervical length and positive fFN is associated with spontaneous preterm labor and birth <32 weeks. However, PPROM does not appear to be preceded by a short cervix, but is preceded by a positive fFN. Neither test is associated with an indicated preterm birth.