Objectives: Our main goals were to evaluate the capability of ffDNA to increase the accuracy in prediction of preterm labour by cervical length and to explore potential mechanisms of disease associated with this pathology. Methods: Fifty-six women, with male fetus, with cervical length assessment at between 22 and 24weeks were included in the study and divided in 1) Short cervix (<15mm) delivered at term (T=20); 2) Short cervix delivered before 37weeks (PT=14); and 3) Patients who delivered at term with normal cervical length (N=22). Maternal plasma samples were collected between 22 and 24weeks of gestational age. PCR using primers against DYS14 gene were used to quantified ffDNA in plasma samples. Statistical analysis was done using ANOVA test and spearmańs correlation. Results: The median gestational age at delivery for short cervix groups was 26+1 for PT and 39+3 for T. The control group delivered at a median gestational age of 39+6weeks. ffDNA was detectable in all cases. There was no significant difference between the 3 groups. Similarly, no significant correlation was observed between ffDNA and gestational age at delivery (r=-0.23; p=0.07). Conclusions: ffDNA does not increase the accuracy of short cervix at between 22 and 24weeks for the prediction of preterm labour. © 2011 John Wiley & Sons, Ltd.