1. Epidemiology and Global Health

Model can identify preterm births when ultrasound is unavailable

A mathematical model that uses heel pinprick or cord blood samples from newborns could accurately estimate gestational age in low-resource settings where ultrasound is not available.
Press Pack
  • Views 55
  • Annotations

Researchers have found that a model built to identify preterm births in Canada can accurately estimate the gestational age of babies born in lower-income countries.

The model, described in the open-access journal eLife, could be used in settings where ultrasound scans are not available to provide more accurate estimates of preterm birth rates. This could help researchers evaluate new programs and policies to reduce preterm birth, and in the future could also be used to guide care for individual newborns who are identified as being preterm.

Estimating the frequency of preterm births in developing countries is difficult because of the lack of ultrasound technology in many low-resource settings. While first-trimester ultrasound scans are accurate to within one week, the accuracy of measurements based on newborn examination are not as reliable. Better ways to estimate the number of true preterm births are urgently needed.

Scientists have previously developed mathematical models that combine information from newborn blood tests with clinical information such as sex and birth weight to estimate gestational age. So far, these have been based on data from North America and used in high-income settings, and it was not known whether these models would be reliable in other countries.

“We’re using metabolic fingerprints – unique patterns in specific molecules found in the blood – to help estimate gestational age,” said senior author Dr. Kumanan Wilson, an internal medicine specialist and senior scientist at The Ottawa Hospital and professor at the University of Ottawa, Canada. “This could be crucial to global efforts to reduce preterm birth and improve newborn health.”

The team collected heel pinprick and cord blood samples from 1,069 infants born in Matlab, Bangladesh. Most of the samples were from full-term infants, but some were also from infants who were identified as preterm based on ultrasound dating. They then applied existing gestational age estimation models to the Bangladesh data and compared the estimates to gestational ages validated by ultrasound.

All of the tested models were best at predicting gestational ages in infants who were born close to full term. Models that used either clinical information or blood-test information alone were not as accurate as models that used both types of information. The model containing both clinical and blood test data showed the best performance: of all the heel prick samples, 63.9% were correctly estimated to within one week of ultrasound-validated gestational age, and 94.3% within two weeks of the validated gestational age. Although cord blood samples were less accurate than heel samples, both models were able to accurately classify infants according to whether or not they were preterm (defined as less than 37 weeks’ gestational age).

“This study shows for the first time that a gestational age calculator based on a simple blood test could help identify preterm births in a low-resource country,” said Dr. Wilson. “Our next step is to test the model in other countries and explore the use of machine learning to make it even better.”

##

For more information about the study and funders, visit http://www.ohri.ca/newsroom/story/view/1099?l=en.

Media contacts

  1. Emily Packer
    eLife
    e.packer@elifesciences.org
    +441223855373

  2. Amelia Buchanan
    Ottawa Hospital Research Institute
    ambuchanan@ohri.ca
    +16137985555;ext=73687
    +16132978315

About

eLife is a non-profit organisation inspired by research funders and led by scientists. Our mission is to help scientists accelerate discovery by operating a platform for research communication that encourages and recognises the most responsible behaviours in science. We publish important research in all areas of the life and biomedical sciences, including Epidemiology and Global Health, which is selected and evaluated by working scientists and made freely available online without delay. eLife also invests in innovation through open-source tool development to accelerate research communication and discovery. Our work is guided by the communities we serve. eLife is supported by the Howard Hughes Medical Institute, the Max Planck Society, the Wellcome Trust and the Knut and Alice Wallenberg Foundation. Learn more at https://elifesciences.org/about.

To read the latest Epidemiology and Global Health research published in eLife, visit https://elifesciences.org/subjects/epidemiology-global-health.