Causality assessment of circulating vitamin d level on venous thromboembolism: A mendelian randomization study

Document Type

Journal Article

Publication Title

Nutrition, Metabolism and Cardiovascular Diseases





First Page


Last Page


PubMed ID





School of Medical and Health Sciences


Beijing Municipal Health System Special Funds of High-Level Medical Personnel Construction / National Key R&D Program of China / European Commission Horizon 2020 / Beijing Postdoctoral Research Foundation


Zhang, X., Sun, W., Li, N., Jian, X., Geng, T., Wu, L., . . . Zheng, D. (2023). Causality assessment of circulating vitamin d level on venous thromboembolism: A mendelian randomization study. Nutrition, Metabolism and Cardiovascular Diseases, 33(9), 1800-1807. https://doi.org/10.1016/j.numecd.2023.05.019


Background and aims: The associations of vitamin D level with venous thromboembolism (VTE) reported in observational studies, whereas these causal associations were uncertain in European population. Therefore, we used Mendelian randomization (MR) method to explore the causal associations between 25-hydroxyvitamin D (25(OH)D) concentrations and the risk of VTE and its subtypes [including deep vein thrombosis (DVT) and pulmonary embolism (PE)]. Methods and results: We used three kinds of genetic instruments to proxy the exposure of 25(OH)D, including genetic variants significantly associated with 25(OH)D, expression quantitative trait loci of 25(OH)D target genes, and genetic variants within or nearby 25(OH)D target genes. MR analyses did not provide any evidence for the associations of 25(OH)D levels with VTE and its subtypes (p > 0.05). The summary-data-based MR (SMR) analyses indicated that elevated expression of VDR was associated with decreased risk of VTE (OR = 0.81; 95% CI, 0.65–0.998; p = 0.047) and PE (OR = 0.67; 95% CI, 0.50–0.91; p = 0.011), and expression of AMDHD1 was associated with PE (OR = 0.93; 95% CI, 0.88–0.99; p = 0.027). MR analysis provided a significant causal effect of 25(OH)D level mediated by gene AMDHD1 on PE risk (OR = 0.09; 95% CI, 0.01–0.60; p = 0.012). Conclusion: Our MR analysis did not support causal association of 25(OH)D level with the risk of VTE and its subtypes. In addition, the expression of VDR and AMDHD1 involved in vitamin D metabolism showed a strong association with VTE or PE and might represent targets for these conditions.



Access Rights

subscription content