Cervical length measurement and cervical phosphorylated insulin-like growth factor binding protein-1 testing in prediction of preterm birth in patients reporting uterine contractions.

OBJECTIVES

To evaluate the performance of cervical phosphorylated insulin-like growth factor binding protein-1 (phIGFBP-1) testing and cervical length measurement separately and in combination with physician's clinical judgment in prediction of preterm birth among patients with self-reported uterine contractions and intact membranes.

DESIGN

We enrolled a total of 246 women between 22 and 34 weeks of gestation.

METHODS

The initial evaluation included cervical length measurement using transvaginal ultrasonography. Short cervix was defined as <25 mm. A swab sample was obtained from the cervix for phIGFBP-1. Admission was used as a clinical marker of an increased risk of preterm delivery <or=34 weeks. The diagnostic performances of the tests and clinician's judgment, as well as likelihood ratios (LRs) were calculated.

MAIN OUTCOME MEASURES: Delivery <or=34 weeks and within 14 days.

RESULTS

The overall rate of spontaneous preterm delivery <or=34 weeks was 4.1% (10/246). Short cervix, positive phIGFBP-1 test, combination of both, and clinician's judgment were all associated with preterm delivery <or=34 weeks or within 14 days (p<0.01). The negative predictive values for delivery <or=34 weeks were 97.4, 97.6, 97.1, and 98.7%, respectively, and within 14 days 98.7, 99.0, 98.3, and 99.6%, respectively. The corresponding positive LRs for delivery <or=34 weeks were 6.8, 3.8, 75.0, 14.9, and within 14 days 9.7, 5.5, 107.3, 17.1. The negative LRs were 0.6, 0.6, 0.7, 0.3 and 0.5, 0.3, 0.6, 0.2.

CONCLUSION

The rapid phIGFBP-1-test has a high negative predictive value for preterm delivery, comparable to that of ultrasonographic cervical length measurement.