Vaginal swabs could be used to predict likelihood of preterm births

In many cases, the cause of preterm birth is unknown, but analyzing metabolites in the vaginal microbiome could be a way to predict the risk of preterm birth.

To Grace Wade

Certain metabolites in the vagina may predict whether a baby will be born prematurely, which is one of the leading causes of neonatal death. Examination of the vaginal microbiome during pregnancy may help health care providers identify and manage high-risk pregnancies.

Tal Korem of Columbia University in New York and his colleagues evaluated 232 vaginal swab samples taken during the second trimester of pregnancy. About 75% of the samples were from identified black people, a group disproportionately affected by premature birth in the United States.

Previous studies have suggested that the vaginal microbiome plays a role in preterm birth, but unlike previous studies, Korem and his colleagues identified both microbes and metabolites in each sample. Did. Metabolites are small substances made when foods, drugs, and other chemicals break down. The team used this data to train a machine learning model to predict whether pregnancies would end prematurely, and found that specific metabolites, not microbes, had the strongest association with premature birth.

Models using data on metabolites and race had a 78% chance of correctly identifying whether a sample came from a person who was born prematurely and had a terminated pregnancy, whereas microbial, race, and other clinical The model that used the data on social factors had a 59% chance. St Chance.

Notably, many of the metabolites associated with premature birth were not created by microbes or humans, Korem said. “They come from outside sources,” he says. Some of these metabolites, such as Ethylglucoside and his EDTA, are found in certain cosmetic and hygiene products, although further analysis is needed to determine the source.

“This is far from proving,” says Korem, noting that these findings are only associations, not causation. Therefore, we cannot recommend avoiding any particular product. “It’s very tentative, but we’re kind of flagging it and saying there’s something suspicious here that needs to be investigated,” he says.

“The study did a great job of overrepresenting underrepresented populations because of its focus on recruiting black women,” says Christine Metz of the Feinstein Institute for Medical Research in New York. said. But the microbiome can change due to a variety of factors, such as environmental exposure, so models can produce different results for samples taken from the same person on different days, she says.

Sign up for our free Health Check newsletter for all the health and fitness news you need to know every Saturday