Preconception Vitamin A and D Levels Linked to Thyroid Function Markers

In a cross-sectional study of 4,947 preconception women with normal thyroid-stimulating hormone levels, nutritional status of fat-soluble vitamins—specifically vitamin A and vitamin D—was significantly associated with thyroid function and thyroid autoimmunity biomarkers. The investigation revealed that circulating vitamin A levels correlated positively with thyroid-stimulating hormone concentrations, while 25-hydroxyvitamin D showed an inverse relationship with thyroid peroxidase antibody positivity that was modified by body mass index.

Fat-soluble vitamins have been established as contributors to hormonal regulation and immune function, yet their specific relevance to thyroid health during the preconception window has remained unclear. To address this uncertainty, the researchers designed the study to evaluate both individual and combined associations of fat-soluble vitamins with thyroid function markers in preconception women, with an explicit focus on assessing potential modification by body mass index.

The study cohort comprised 4,947 preconception women who exhibited normal baseline thyroid-stimulating hormone (TSH) levels. Within this study population, the researchers classified participants into low normal and high normal TSH categories, setting the threshold for high normal status at 2.5 mIU/L or higher.

To evaluate the individual and joint relationships between fat-soluble vitamins and thyroid function biomarkers, the investigators utilized generalized linear regression models, restricted cubic spline models, and Bayesian kernel machine regression models. In addition, the potential modifying role of body mass index was specifically examined through a restricted cubic spline interaction model that incorporated a dedicated vitamin-BMI interaction term.

Assessment of baseline nutritional status showed that a large majority of participants maintained sufficient concentrations of vitamin A (80.9 percent), vitamin E (94.5 percent), and vitamin K (94.6 percent), whereas 87.5 percent were deficient or severely deficient in 25-hydroxyvitamin D. In single-exposure statistical models, vitamin A demonstrated a positive association with high normal TSH status, yielding an odds ratio of 1.36 with a 95 percent confidence interval of 1.04 to 1.77. Vitamin A was also positively associated with continuous TSH levels, showing a beta coefficient of 0.10 (95 percent confidence interval: 0.04 to 0.16), with restricted cubic spline analysis confirming that these associations were linear, characterized by an overall p-value below 0.05 and a nonlinear p-value above 0.05.

In multi-vitamin analyses, vitamin A remained positively associated with TSH concentrations, while a linear inverse association emerged between 25-hydroxyvitamin D and thyroid peroxidase antibody (TPOAb) positivity. This inverse association between vitamin D and TPOAb positivity was more pronounced among participants with overweight or obesity. The authors concluded that fat-soluble vitamin status relates to preconception thyroid function and autoimmunity, with variations across BMI categories, while noting that future prospective studies are needed to determine temporal sequence and biological pathways.

Subject of Research: Associations between preconception fat-soluble vitamins, body mass index, and thyroid homeostasis

Article Title: Preconception nutritional status of fat-soluble vitamins and thyroid homeostasis: the modifying role of BMI in a cross-sectional study

Article References: Jia, G., Chen, C., Chen, C., Jiang, X., Chen, Y., Wu, X., & Zhang, Y. (2026). Preconception nutritional status of fat-soluble vitamins and thyroid homeostasis: the modifying role of BMI in a cross-sectional study. BMC Nutrition. https://doi.org/10.1186/s40795-026-01486-y

Image Credits: AI Generated

DOI: 10.1186/s40795-026-01486-y

Keywords: preconception health, fat-soluble vitamins, thyroid homeostasis, thyroid-stimulating hormone, vitamin A, vitamin D, 25-hydroxyvitamin D, vitamin E, vitamin K, body mass index, thyroid peroxidase antibodies, TPOAb positivity, restricted cubic spline models, Bayesian kernel machine regression, cross-sectional study, thyroid autoimmunity, nutritional status