Cervical length change as a predictor of preterm birth in symptomatic patients.

BACKGROUND: Distinguishing between true and false preterm labor remains a challenge. The shortening in cervical length throughout a gestation has been theorized to be a possible predictor of spontaneous preterm birth. Although there are some studies evaluating cervical length shortening as a predictor of spontaneous preterm birth, it is not known whether the shortening in cervical length from an asymptomatic to symptomatic state, when a patient presents with preterm labor symptoms, is predictive of spontaneous preterm birth. OBJECTIVE: This study aimed to determine the utility of cervical length shortening from an asymptomatic time point (anatomic ultrasound) to when a patient presents with preterm labor symptoms as a predictor of spontaneous preterm birth. STUDY DESIGN: A prospective cohort study was performed to evaluate the use of transvaginal cervical length assessment in symptomatic women in predicting spontaneous preterm birth from January 2013 to March 2015. Women with singleton gestations who presented to our institution between 22 0/7 weeks and 33 6/7 weeks of gestation with preterm labor symptoms were included in the overall cohort. This was a planned secondary analysis to evaluate the shortening in cervical length from an asymptomatic state (anatomic ultrasound) to a symptomatic state as a predictor of preterm birth. For this analysis, inclusion criteria were known delivery status, cervical length screening performed at anatomic ultrasound, and a valid cervical length measurement at the time of preterm labor symptoms. Women with preterm rupture of membranes, cervical dilation of >2 cm, or moderate to severe bleeding were excluded. Cervical length shortening was defined as a decrease in cervical length of >10 mm from anatomic ultrasound to the time of presentation with preterm labor symptoms. The outcome evaluated was spontaneous preterm birth before 37 weeks of gestation. Chi-square test and receiver operating characteristic curves were used to evaluate the data. Multivariable logistic regression was used to calculate the odds. Test characteristics of cervical length shortening of >10 mm were determined. RESULTS: A total of 549 women were included in the original cohort, and 277 women were included in this secondary analysis. The overall rate of spontaneous preterm birth was 8.3%. There were 52 women (19%) with cervical length shortening of >10 mm. The rate of spontaneous preterm birth was significantly higher for those with cervical length shortening of >10 mm than those with cervical length shortening of ≤10 mm (21.2% vs 5.3%; P=.001). This higher risk of spontaneous preterm birth remained after adjusting for confounders including maternal age and previous spontaneous preterm birth (adjusted odds ratio, 4.71; 95% confidence interval, 1.84-12.09). Using cervical length shortening of >10 mm as a screening test had a sensitivity of 47.8%, a specificity of 83.9%, a positive predictive value of 21.2%, and a negative predictive value of 94.7%. CONCLUSION: In women presenting with preterm labor symptoms, a cervical length that is >10 mm shorter from anatomic ultrasound is associated with an increased risk of spontaneous preterm birth.

Quantitative fetal fibronectin and cervical length in symptomatic women: results from a prospective blinded cohort study.

Objectives: Our objectives were to determine whether quantitative fetal fibronectin (fFN) and cervical length (CL) screening can be used alone or in combination as prognostic tests to identify symptomatic women at the highest or lowest risk for spontaneous preterm birth (sPTB). Methods: A prospective, blinded cohort study of women presenting with a singleton gestation to our triage unit between 22-33w6d with preterm labor symptoms was performed. Women with ruptured membranes, moderate/severe bleeding, and dilation >2 cm were excluded. The primary outcome was sPTB <37 weeks. We evaluated test characteristics of quantitative fFN and CL assessment, both separately and in combination, considering traditionally reported cut-points (fFN ≥50 and CL <25), as well as cut-points above and below these measures. We found interactions between fFN >50 and CL <25 and sPTB by parity and obstetric history (p < .05) and therefore stratified results. Test characteristics are presented with positive predictive value (PPV) and negative predictive value (NPV). Results: Five hundred eighty women were enrolled and 537 women were available for analysis. Overall sPTB rate was 11.1%. Among nulliparous women, increasing levels of fFN were associated with increasing risk of sPTB, with PPV going from 26.5% at ≥20 ng/mL to 44.4% at ≥200 ng/mL. A cut-point of 20 ng/mL had higher sensitivity (69.2%) and higher NPV (96.8%) and therefore identified a "low-risk" group. fFN was not informative for multiparous women regardless of prior obstetrical history or quantitative level chosen. For all women, a shorter CL was associated with an increased sPTB risk. Among nulliparas and multiparas without a prior sPTB, a CL <20 mm optimized test characteristics (PPV 25 and 20%, NPV 95.5, and 92.7%, respectively). For multiparas with a prior sPTB, CL <25 mm was more useful. Using fFN and CL in combination for nulliparas did not improve test characteristics over using the individual fFN (p = .74) and CL (p = .31) components separately. Conclusions: This study identifies the importance of stratifying by parity and obstetrical history when using screening modalities for risk assessment in symptomatic women. For nulliparous women, either quantitative fFN or cervical length assessment can be utilized, depending on resources available, but a lower cut-point of 20 ng/mL should be used for quantitative fFN. For multiparous women, fFN is not useful and cervical length assessment should be the main screening tool utilized when there is clinical uncertainty. Regardless of parity, the PPV of fFN and CL is low and therefore the greatest clinical utility remains in its NPV.