Mendelian randomization on the association of obesity with vitamin D: Guangzhou Biobank Cohort Study

Background Mendelian randomization (MR) analyses from the West provide evidence that obesity causes lower 25-hydroxyvitamin D [25(OH)D]. As Asian populations are prone to metabolic disorders at a lower body mass index (BMI), whether the association remains in Asian is unclear. We studied whether obesity causes vitamin D deficiency using MR analysis in Chinese. Methods We used data from the Guangzhou Biobank Cohort Study. A genetic score including seven BMI-related single-nucleotide polymorphisms (n = 15,249) was used as the instrumental variable (IV) for BMI. Two-stage least square regression and conventional multivariable linear regression in 2,036 participants with vitamin D data were used to analyze association of BMI with vitamin D. Results Proportion of variation explained by the genetic score was 0.7% and the first stage F-statistic for MR analysis was 103. MR analyses showed that each 1 kg/m(2) higher BMI was associated with lower 25(OH)D by -2.35 (95% confidence interval (CI) -4.68 to -0.02) nmol/L. In conventional multivariable linear regression, higher BMI was also associated with lower 25(OH)D (beta = -0.26 nmol/L per 1 kg/m(2) increase in BMI, 95% CI -0.46 to -0.06). Sensitivity analyses using two-sample IV analysis and leave-one-out method showed similar results. Conclusion We have first shown by MR and conventional multivariable linear regression that higher BMI causes vitamin D deficiency in Chinese. Our findings highlight the importance of weight control and suggest that vitamin D supplementation may be needed in individuals with overweight or obesity.