Abstract
Observational studies and meta-analyses have indicated a notable correlation between obesity and vitamin D deficiency, yet the causal relationship between the 2 remains contentious. This study employed a bidirectional Mendelian randomization (MR) analysis to explore the interrelation between obesity-associated body metrics: specifically body mass index (BMI), waist-to-hip ratio (WHR), body fat percentage (BFP), whole-body fat percentage (WHF), and 25-hydroxyvitamin D (25(OH)D) levels. Instrumental variables for BMI, WHR, BFP, whole body fat mass (WFM), and 25(OH)D were carefully selected based on predefined thresholds. The association between these metrics and 25(OH)D levels was assessed using the TwoSampleMR package in R 4.2.3. Analysis methods included inverse variance weighted (IVW), MR Egger regression, weighted median, weighted mode, and simple mode. Sensitivity analyses were conducted employing the TwoSampleMR and MR-PRESSO software packages in R 4.2.3 to evaluate heterogeneity and multiplicity of findings. All 4 body components exhibited statistically significant causal associations with decreased 25(OH)D levels: BMI (IVW: odds ratio [OR] = 0.912, 95% confidence interval [CI]: 0.888-0.937, P < .001), WHR (IVW: OR = 0.927, 95% CI: 0.882-0.975, P = .003), BFP (IVW: OR = 0.883, 95% CI: 0.867-0.899, P < .001), and WFM (IVW: OR = 0.850, 95% CI: 0.829-0.872, P < .001). However, no statistically significant inverse causative association was observed between these body components and 25(OH)D levels. Sensitivity analyses revealed no substantial heterogeneity or pleiotropy, ensuring robustness of the findings. This study substantiates a significant causal link between 4 obesity-related body components and decreased 25(OH)D levels, excluding reverse causality.