High Levels of Radon Exposure: Risk Factor for Gestational Diabetes?

Exposure to radon was linked to an increased risk for gestational diabetes, a population-based cohort study suggested.

Among 9,107 mothers-to-be, those living in U.S. counties with the highest radon level (≥2 pCi/L) had a 37% greater chance of developing gestational diabetes compared with those living in counties with the lowest radon level (

Even after adjustment for fine particulate matter air pollutants (PM_2.5), which may share a similar biological pathway to radon, exposure to radon was still tied with gestational diabetes with borderline statistical significance (OR 1.36, 95% CI 1.00-1.86).

The researchers explained in JAMA Network Open that radon and radon decay products emit alpha particles that can induce oxidative stress and promote inflammation, which both are implicated in chronic inflammatory disease like hypertensive disorders of pregnancy.

“Additionally, radon exposure may lead to DNA damage and epigenetic changes in placental cells as well as mitochondrial dysfunction,” they added. “Together, these combined pathways may contribute to placental vascular dysfunction by disrupting blood flow and nutrient exchange, creating hypoxic conditions that promote insulin resistance.”

“Radon is possibly the most prevalent indoor carcinogen to which human beings are exposed,” added an accompanying commentary by Alberto Ruano-Ravina, PhD, and Lucía Martín-Gisbert, MSc, both of the University of Santiago de Compostela in Spain. “Some radon-prone areas are well characterized in the U.S. and Europe, but despite this, most of the population is unaware of the potential risks, including the interaction of radon with smoking.”

Kahe’s group used data from the Nulliparous Pregnancy Outcomes Study: Monitoring Mothers-to-Be cohort, which recruited nulliparous pregnant participants from eight U.S. clinical centers from October 2010 to September 2013. Average age of the cohort was 27, and 41.6% had ever used tobacco.

A clinical diagnosis of gestational diabetes was based on a glucose tolerance test during routine clinical care. A total of 4.2% of participants developed gestational diabetes.

Radon exposure by county was divided by three categories based on the Environmental Protection Agency’s short- and long-term indoor home radon assessments: less than 1 pCi/L, 1 to less than 2 pCi/L, and 2 pCi/L or more. Average county-level radon concentration was 1.6 pCi/L. Those exposed to the highest radon concentrations (≥2 pCi/L) were more likely to be white and have an educational level of less than high school completion and were less likely to have smoked than groups exposed to levels less than 2 pCi/L.

Interestingly, there was no association between radon exposure and gestational diabetes among mothers who never smoked (OR 1.31, 95% CI 0.84-2.05) or among those with a BMI of 25 or less (OR 1.38, 95% CI 0.80-2.39).

On the other hand, the highest odds of gestational diabetes were among ever-smokers living in counties with the highest radon levels compared with never-smokers in low-radon counties (OR 2.09, 95% CI 1.41-3.11). Also, odds of gestational diabetes were elevated among those exposed to high radon and high PM_2.5 levels compared with those exposed to low radon and low PM_2.5 levels (OR 1.93, 95% CI 1.31-2.83).

Ruano-Ravina and Martín-Gisbert pointed out that the current study — much like prior studies on radon — was limited by estimating indoor radon exposure at an ecological level using county, municipal, or other geographical units, “all of which provide an aggregated radon concentration that was uniformly assigned as an exposure to participants living in the area of interest.”

They emphasized that radon exposure is well known to be highly variable. “[E]ven within a given county, residential radon concentrations may be quite different just because of a different composition of the underlying bedrock,” they said, adding that levels can even differ among neighboring townhouses.

“We need studies measuring individual radon exposure, since this is the best way to link this exposure with any given disease,” they urged. “Reliable radon measurements are inexpensive and simple, with the only limitation being that radon detectors must be placed in dwellings for at least 3 months (no personnel required).”

The researchers agreed, and said that some of the county-level radon measurements used in this study were taken during the 1990s and may not represent the radon level during the study.

Nonetheless, studies like these “are necessary as hypothesis generators,” the commentators said in calling for future studies to continue to piece together the wide health implications of radon exposure.