Vitamin D and Gestational Diabetes Mellitus in the IEU OpenGWAS Project: A Two-Sample Bidirectional Mendelian Randomization Study.

Abstract

Gestational diabetes mellitus (GDM) is one of the most prevalent pregnancy problems, and there is still debate over the relationship between vitamin D and GDM. Our objective is to investigate the correlation between vitamin D and GDM by employing Mendelian randomization (MR) with summary data obtained from genome-wide association studies (GWAS). Data on exposures and outcomes, namely vitamin D, vitamin D insufficiency, and GDM, were acquired from the IEU OpenGWAS Project. Bidirectional MR analysis was performed utilizing the inverse variance weighted (IVW) method as the principal analytical approach. The complementary approaches employed in this study encompassed weighted median, simple mode, weighted mode, and MR-Egger regression. A series of sensitivity analysis were conducted in order to assess the reliability of the obtained results. The data were acquired from the IEU OpenGWAS Project. Following the application of the three assumptions of MR, 13 single nucleotide polymorphisms (SNPs) were included in the MR analysis for vitamin D levels and vitamin D deficiency on GDM, and 10 and 26 SNPs were included for GDM on vitamin D levels and deficiency, respectively. The findings from the IVW analysis revealed a significant positive correlation between vitamin D levels and GDM (OR = 1.057, 95% CI: 1.011-1.104, p = 0.015). Conversely, a negative correlation was seen between vitamin D deficiency and GDM (OR = 0.979, 95% CI: 0.959-0.999, p = 0.039). The results of the reverse MR study revealed no evidence of reverse causation between GDM and vitamin D. The findings from multiple MR approaches were in line with the direction of IVW analysis. Sensitivity analysis revealed no evidence of heterogeneity, pleiotropy, or outliers, suggesting the robustness of the results. There exists a causal association between vitamin D and GDM, whereby vitamin D levels serve as a risk factor for GDM.