Background Fetal fibronectin (fFN) is an excellent predictor of preterm birth (PTB). However, its value at predicting PTB in primiparous women with previous cervical surgery is unknown.

Methods We evaluated the use of fFN in all women attending the Preterm Surveillance Clinic over 12 months, comparing 86 primiparous women with cervical surgery – LLETZ or cone biopsy – to 242 women tested following previous PTB (16–37 weeks). Sensitivity, specificity, PPV and NPV of fFN for predicting delivery <30, <34 and <37 weeks' in the two groups were compared. Test performance was evaluated using logistic regression.

Results Positive fFN rates were similar: 16% of primiparous women with cervical surgery (14/86) compared to 20% (49/242) of women with previous PTB. Comparison of the accuracy of fFN testing between the two groups can be seen in table 1.

Logistic regression found no significant difference between the odds ratios of a positive fFN predicting delivery before 30, 34 and 37 weeks' in the two groups.

Conclusion fFN is useful for predicting PTB in primiparous women with cervical surgery; its performance is comparable to, and may even be slightly better, to that in women with previous PTB.