Predicting Premature Delivery

From the WebMD Archives

Jan. 23, 2002 -- Does monitoring the frequency of uterine contractions predict whether a woman will give birth prematurely? The question has been studied for two decades with mixed findings, but new research suggests the answer is an emphatic "no."

Researchers from the National Institute of Child Health and Human Development's Network of Maternal-Fetal Medicine evaluated the usefulness of an at-home contraction-monitoring device in a group of pregnant women at high risk for premature delivery. While there were slight differences in contraction frequency between women who later gave birth prematurely and those who did not, these differences were not useful for predicting what would happen.

The findings are published in the Jan. 24 issue of The New England Journal of Medicine.

Researchers have long been frustrated in their efforts to both predict and prevent early births. The finding from this and earlier studies suggest that the strategy of detecting contractions and suppressing them is not useful, lead author Jay D. Iams, MD, tells WebMD. Iams is a professor of obstetrics and gynecology at Ohio State University.

"This two-decades-old strategy has been discredited, and we are really going back to the drawing board in our efforts to identify women who will give birth prematurely," he says. "We do have tests that allow us to reassure high-risk women with reasonable accuracy, but when it comes to predicting who will deliver early, there is really nothing out there."

Approximately one in 10 babies in the United States is born prematurely and some 5,000 infants die each year due to complications of premature birth. Risk factors for early delivery include carrying multiples (twins, triplets, etc.), having a previous child born prematurely, and uterine bleeding during the second trimester. But only about half of women who give birth prematurely are considered to be at high risk.

Iams and colleagues found a slight increase in the frequency of uterine contractions in the high-risk study participants who later delivered prematurely, but the finding had little predictive value. Measures that were more helpful included cervical length (measured with ultrasound), and presence of a substance called fetal fibronectin in the mother's vaginal secretions.

While these two tests can help determine which women will not give birth prematurely, they do not identify those who will, says Charles Lockwood, MD, who chairs the department of obstetrics and gynecology at New York University School of Medicine. Lockwood wrote an editorial accompanying the study.

"Cervical length and fetal fibronectin are clearly more predictive than monitoring uterine activity, but they still have significantly high false-positive rates," he tells WebMD. "About half of patients who are positive with these tests do not end up delivering early. But a negative test means you can avoid inappropriate interventions that can add stress and may not work anyway."

Lockwood says he believes genetics research will soon help identify effective strategies to both predict and prevent premature births.

"We have gotten pretty good at understanding the immediate causes of preterm deliveries," he says. "I think over the next five years we will develop a progressive understanding of the genetic predispositions involved. Once we have identified the genes involved with these environmental interactions, we will probably be in a better position to prevent the causes of preterm births with specific treatments that may even be recommended before a women gets pregnant."