This article is more than 5 years old. Information may no longer be current.

NIH: Common screening tests ineffective at predicting preterm birth in first-time pregnancies

Among nulliparous women with singleton pregnancies, routine universal screening using quantitative vaginal fetal fibronectin and serial transvaginal ultrasound cervical length independently and combined had low predictive accuracy for subsequent spontaneous preterm birth and should not be used routinely in such women, according to research by the NIH published in JAMA.

“Spontaneous preterm birth is a leading cause of infant mortality,” M. Sean Esplin, MD, of Intermountain Healthcare in Utah, and colleagues wrote. “Prediction, largely based on prior pregnancy outcomes, is not possible in women pregnant for the first time.”

Esplin and colleagues noted that previous studies have suggested that a short cervix, defined as less than 25 mm, and fetal fibronectin in the vagina early in pregnancy may be signs of preterm birth. In this study, they sought to determine the accuracy of transvaginal cervical length and quantitative self-collected vaginal fetal fibronectin assessments to predict spontaneous preterm birth before 37 weeks in nulliparous women. Between October 2010 and May 2014, they performed a prospective observational cohort study of 9,410 nulliparous women (median age, 27 years; 60.7% non-Hispanic white, 13.8% non-Hispanic black, 16.5% Hispanic, 4% Asian and 5.1% other) from eight clinical sites across the United States.

Of the women enrolled in the study, 474 (5%) had spontaneous preterm births, 335 (3.6%) had medically indicated preterm births and 8,601 (91.4%) had term births. Cervical length of 25 mm or less was observed in 8% and 23.3% of women with spontaneous preterm birth at 16 to 22 weeks’, and 22 to 30 weeks’ gestation, respectively. Fetal fibronectin levels of 50 ng/mL or greater was identified in 7.3% and 8.1% of women with spontaneous preterm birth at 16 to 22 weeks’ and 22 to 30 weeks’ gestation, respectively.

For screening between 22 and 30 weeks, the area under the receiver operating characteristic curve was 0.59 (95% CI, 0.56-0.62) for fetal fibronectin level alone, 0.67 (95% CI, 0.64-0.7) for transvaginal cervical length and 0.67 (95% CI, 0.64-0.7) for fetal fibronectin level and transvaginal cervical length combined. These results indicate low predictive accuracy for preterm birth for these screening tests both individually and combined.

“These methods of assessing women in their first pregnancy do not identify most of those who will later go on to have a spontaneous preterm delivery,” Uma Reddy, MD, coauthor of the study from the Pregnancy and Perinatology Branch at NIH’s Eunice Kennedy Shriver National Institute of Child Health and Human Development, said in a press release. “There is a need to develop better screening tests that can be performed early in pregnancy.”

In an accompanying editorial, Steven L. Bloom, MD, and Kenneth J. Leveno, MD, both from the department of obstetrics and gynecology at the University of Texas Southwestern Medical Center in Dallas, wrote that the findings are important because it debunks two popular screening modalities and should be used to temper their use.

In addition, the study “provides provocative insights as to how medical practice determined through consensus, rather than reproducible evidence, contributes to the narrative in the United States regarding the cost and quality of health care,” they wrote.

“While it is gratifying that such research takes place, that research should have preceded the adoption of this screening strategy into practice,” Bloom and Leveno concluded. – by Alaina Tedesco

Disclosure: Esplin and colleagues report funding from the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD), Case Western Reserve University, Columbia University, Indiana University, Magee-Women’s Hospital, Northwestern University, University of California at Irvine, University of Pennsylvania and University of Utah. Bloom and Leveno report no relevant financial disclosures.