PFAS in Early Pregnancy May Raise Gestational Diabetes Risk, Vitamin D Status Emerges as a Key Factor

Every pregnant woman on the planet now carries a chemical legacy in her bloodstream: a family of synthetic molecules known as per- and polyfluoroalkyl substances, or PFAS, that have found their way into drinking water, food packaging, cookware and even the dust on the windowsill. These so-called forever chemicals resist degradation for decades, and a growing body of research links them to metabolic disorders. A new prospective birth cohort study conducted in Ma’anshan, China, now adds a striking nuance to that picture, suggesting that early-pregnancy exposure to multiple PFAS is associated with an elevated risk of gestational diabetes mellitus, and that a mother’s vitamin D status may influence whether those exposures translate into disease.

Gestational diabetes mellitus, or GDM, is one of the most common complications of pregnancy, affecting roughly one in six women in the studied population and considerably more in some regions. It arises when the hormonal environment of pregnancy outstrips the pancreas’s capacity to secrete enough insulin, causing blood glucose to rise to diabetic levels. The consequences ripple outward: mothers with GDM face higher risks of hypertension, cesarean delivery and later type 2 diabetes, while their children are more likely to be born large, experience birth complications and develop obesity and metabolic disease themselves. Because PFAS compounds are known endocrine disruptors, capable of interfering with hormonal signaling and lipid metabolism, researchers have long suspected they could tip vulnerable pregnancies toward glucose intolerance.

The new study, led by Cheng-Yang Hu and colleagues at Anhui Medical University, took a rigorous prospective approach. The team enrolled 420 pregnant women early in gestation and quantified fourteen distinct PFAS compounds in their serum during the first trimester, before any glucose abnormalities had appeared. Alongside the contaminant panel, the researchers measured serum 25-hydroxyvitamin D, the standard biomarker of vitamin D status, and 5-methyltetrahydrofolate, the biologically active circulating form of folate. Diagnosis of gestational diabetes was then performed with the standard 75-gram oral glucose tolerance test between 24 and 28 gestational weeks, providing a clean temporal separation between exposure assessment and outcome.

Seventy of the 420 participants, or 16.7 percent, ultimately developed GDM. Using modified Poisson regression, a technique well suited to estimating risk ratios in prospective cohorts, the investigators found that higher serum concentrations of five PFAS compounds were each associated with increased GDM risk per doubling of concentration. The implicated chemicals spanned both classic legacy contaminants and emerging alternatives: perfluorononanoic acid (PFNA), perfluorodecanoic acid (PFDA), perfluorooctane sulfonate (PFOS), perfluorooctanoic acid (PFOA), and 6:2 chlorinated polyfluorinated ether sulfonate, a replacement chemistry increasingly detected in Chinese populations. That the newer substitutes showed associations alongside the phased-out long-chain compounds is a sobering signal that industrial substitution may simply be swapping one hazard for another.

The most provocative finding, however, lay in the interaction between pollution and nutrition. When the researchers stratified their analyses by vitamin D status, the PFAS-GDM associations appeared to concentrate almost entirely among women classified as vitamin D insufficient. For PFOA, PFNA, PFDA and PFOS, significant positive associations with gestational diabetes were observed essentially only in the insufficient group, while women with adequate vitamin D showed markedly attenuated and statistically nonsignificant patterns. At the level of the chemical mixture, the contrast was stark: among vitamin D insufficient women, the weighted quantile sum mixture effect carried a mean odds ratio of 1.10 with a confidence interval excluding the null value, whereas among vitamin D sufficient women the estimate sat at a null-like 1.01. In other words, the combined burden of PFAS exposure appeared to matter mainly when vitamin D reserves were low.

To handle the reality that PFAS compounds travel together as a correlated cocktail, the team employed weighted quantile sum (WQS) regression with 100 repeated holdout validations, a mixture method that assigns weights to each chemical according to its contribution to an overall association. This mixture-level analysis, stratified by vitamin D status, provided the clearest visual evidence of effect modification, and it aligns with a plausible biological mechanism. Vitamin D participates directly in glucose homeostasis: the vitamin D receptor is expressed on pancreatic beta cells, and adequate 25(OH)D supports insulin secretion and sensitivity. Anti-inflammatory and antioxidant properties of the vitamin could likewise buffer the oxidative stress and inflammatory pathways that PFAS are thought to provoke. A mother with insufficient vitamin D may therefore have less metabolic reserve to absorb the endocrine insult of these persistent pollutants.

The story with folate followed a similar but weaker arc. Several individual PFAS were associated with increased GDM risk among women in the low-folate group, hinting that folate status, too, might shape susceptibility, perhaps through its central role in one-carbon metabolism, DNA methylation and homocysteine regulation. Yet when the mixture-level analysis was stratified by folate status, no significant between-group differences emerged. Folate may still matter biologically, the authors suggest, but the evidence in this cohort is thinner and demands replication.

Crucially, the researchers resisted the temptation to oversell their findings. Formal tests of effect modification provided only limited statistical support. Significant interaction terms were seen for just a couple of individual chemicals, perfluorohexanoic acid (PFHpA) and perfluorobutane sulfonate (PFBS), with p-values for interaction below 0.10, and the formal interaction between the overall PFAS mixture and vitamin D status was not statistically significant. The team is explicit that these exploratory results warrant confirmation in adequately powered studies before vitamin D optimization can be considered a strategy to mitigate PFAS-related metabolic risk. In an era when supplement recommendations can spread faster than evidence, that restraint is worth highlighting.

What the study does establish firmly is that early-pregnancy exposure to multiple PFAS, spanning legacy compounds and their emerging replacements, is associated with elevated risk of gestational diabetes. The implications cascade through public health. Roughly half the world’s pregnant women are estimated to be vitamin D insufficient, and PFAS contamination of drinking water has triggered regulatory battles across Europe, North America and Asia. If the interaction observed here is confirmed, women in heavily exposed communities who also lack adequate vitamin D could represent a distinctly vulnerable subgroup, and inexpensive interventions such as prenatal vitamin D screening might meaningfully reduce metabolic risk even as cleanup of the chemicals themselves proceeds slowly, given their extraordinary environmental persistence.

For now, the study stands as a vivid illustration of the gene-environment-era insight that toxic exposures rarely act in isolation. The same dose of a forever chemical may land differently in two bodies depending on nutritional context, and maternal physiology during pregnancy is uniquely sensitive to that interplay. The research team, funded by the National Natural Science Foundation of China and provincial science programs, plans the kind of larger, multi-cohort replication that will determine whether the vitamin D signal is real enough to guide clinical practice. Until then, the message for expectant mothers is familiar and uncontroversial: maintain adequate vitamin D through diet, sensible sun exposure and clinician-guided supplementation, and support the broader push to keep PFAS out of water, food and the bodies of the next generation.

Subject of Research: Early-pregnancy PFAS exposure, maternal vitamin D and folate status, and the risk of gestational diabetes mellitus in a prospective birth cohort

Article Title: Early-pregnancy per- and polyfluoroalkyl substances (PFAS), vitamin status, and risk of gestational diabetes mellitus: a prospective birth cohort study

Article References: Hu, C.-Y., Wu, X.-Y., Li, Z.-H., Dai, S.-W., Ma, Y.-B., Tao, F.-B., & Zhang, X.-J. (2026). Early-pregnancy per- and polyfluoroalkyl substances (PFAS), vitamin status, and risk of gestational diabetes mellitus: a prospective birth cohort study. Environmental Health. https://doi.org/10.1186/s12940-026-01341-4

Image Credits: AI Generated

DOI: 10.1186/s12940-026-01341-4

Keywords: PFAS, forever chemicals, gestational diabetes mellitus, vitamin D, folate, pregnancy, endocrine disruptors, environmental health, birth cohort, insulin resistance, PFOA, PFOS