Babies’ face scans detect exposure to low amounts of alcohol in utero

Guidelines from most expert groups already suggest that alcohol and pregnancy should not mix. Nonetheless, many women choose to have a drink or a few during the course of their pregnancy (or before they realized they were pregnant). Ten percent of pregnant women in the United States reported having at least one alcoholic beverage during one month, according to a CDC survey published in 2015.

But can a small amount of drinking by an expectant mother show itself on her child’s face?

To answer that question, researchers in Australia analyzed three-dimensional images of over 400 children’s faces and heads around their first birthday. An algorithm automatically placed an array of nearly 70,000 points on those scans and looked for any substantial deviations from a standardized template made from all of the children’s scans. The children’s mothers also answered surveys about their alcohol intake several times before giving birth — data that researchers used to separate them out into groups based on when and how much they drank. Nearly 80 percent of the children in this study had some alcohol exposure as a fetus.

Currently, screening for children thought to be at-risk for fetal alcohol spectrum disorders — which includes fetal alcohol syndrome — is based in part on facial features. Health care providers often look for three classic signs: small eyes, a thin upper lip, and a flat area above the upper lip where there are typically two vertical ridges. (None of the children in this study had been diagnosed with a condition on the spectrum.)

3-D facial scanning picked up some differences in those areas. It also picked up on one otherwise difficult-to-measure change called mid-facial hypoplasia, in which the center of the face develops more slowly than the eyes, forehead, and lower jaw. Perhaps most notably, there seemed to be slight differences in the mid-face and at the tip of nose even in the children of women who had only had a little to drink and only in the first trimester of their pregnancy. (However, almost no overall changes were statistically significant.) The results were published Monday in JAMA Pediatrics.

Still, says Susan Astley, director of the Washington State Fetal Alcohol Diagnostic and Prevention Network, those findings aren’t surprising. “We’ve known for 20 years that the face [associated with FAS] is presented on a continuum,” she said. “In my opinion, they have not really contributed to the literature because they’ve made the reader think that most of the literature doesn’t exist yet.”

The paper did have one important strength. “They probably have some of the better alcohol exposure documentation,” Astley said, because researchers asked mothers about alcohol intake a few times through the study period, likely collecting more reliable data than if they had asked about it after a child was born.

“I’m a huge advocate of this 3-D analysis — I think there’s a great deal we could learn from it,” she said. However “I think the important public health questions are, even if they were finding something, will we ever be able to use this in clinic,” she said, especially since 3-D scanners are costly devices.

Reinforcing a consensus

There are a few important limitations to keep in mind. The researchers only scanned white children for this study — which can lead to false signals when researchers try to apply results to other ethnicities. (A commentary accompanying the paper noted that work is ongoing to include Aboriginal Australian children in 3-D face scanning studies.)

The American College of Obstetricians and Gynecologists states on its website that “there is no safe level of alcohol use during pregnancy.” The CDC and the American Academy of Pediatrics both concur — as does ACOG’s British equivalent, which updated its guidelines in 2016 to clarify that a woman would ideally abstain from alcohol entirely during a pregnancy.

However, some women may have been advised by an OB-GYN that some drinking during pregnancy is acceptable. Despite ACOG’s guidelines, 40 percent of OB-GYNs said in a survey that some alcohol was fine. Nearly half said there was no clear consensus on alcohol’s effect during pregnancy.

The researchers, for their part, believe that there is a consensus — and that their work further confirms it. “Prenatal alcohol exposure, even at low levels, can influence craniofacial development,” they wrote. “Although the clinical significance of these findings is yet to be determined, they support the conclusion that for women who are or may become pregnant, avoiding alcohol is the safest option.”