A Causal Effect of Serum 25(OH)D Level on Appendicular Muscle Mass: Evidence From NHANES Data and Mendelian Randomization Analyses

Abstract

Background: Low serum vitamin D status was reported to be associated with reduced muscle mass; however, it is inconclusive whether this relationship is causal. This study used data from the National Health and Nutrition Examination Survey (NHANES) and two-sample Mendelian randomization (MR) analyses to ascertain the causal relationship between serum 25-hydroxyvitamin D [25(OH)D] and appendicular muscle mass (AMM).

Methods: In the NHANES 2011-2018 dataset, 11 242 participants (5588 males and 5654 females) aged 18-59 years old were included, and multivariant linear regression was performed to assess the relationship between 25(OH)D and AMM measured by dual-energy X-ray absorptiometry. In two-sample MR analysis, 167 single nucleotide polymorphisms significantly associated with serum 25(OH)D at the genome-wide association level (p < 5 × 10^-8) were applied as instrumental variables (IVs) to assess vitamin D effects on AMM in the UK Biobank (417 580 Europeans) using univariable and multivariable MR (MVMR) models.

Results: In the NHANES dataset, serum 25(OH)D concentrations were positively associated with AMM (β = 0.013, SE = 0.001, p < 0.001) in all participants, after adjustment for age, race, season of blood collection, education, income, body mass index and physical activity. In stratification analysis by sex, males (β = 0.024, SE = 0.002, p < 0.001) showed more pronounced positive associations than females (β = 0.003, SE = 0.002, p = 0.024). In univariable MR, genetically higher serum 25(OH)D levels were positively associated with AMM in all participants (β = 0.049, SE = 0.024, p = 0.039) and males (β = 0.057, SE = 0.025, p = 0.021), but only marginally significant in females (β = 0.043, SE = 0.025, p = 0.090) based on IVW models was noticed. No significant pleiotropy effects were detected for the IVs in the two-sample MR investigations. In MVMR analysis, a positive causal effect of 25(OH)D on AMM was observed in the total population (β = 0.116, SE = 0.051, p = 0.022), males (β = 0.111, SE = 0.053, p = 0.036) and females (β = 0.124, SE = 0.054, p = 0.021).

Conclusions: Our results suggested a positive causal effect of serum 25(OH)D concentration on AMM; however, more researches are warranted to unveil the underlying biological mechanisms and evaluate the effects of vitamin D intervention on AMM.

Keywords: 25‐hydroxyvitamin D; Mendelian randomization; NHANES; appendicular muscle mass; single nucleotide polymorphism.

FIGURE 1

Working flowchart of participants selection in the cross‐sectional study. The current study included 39 156 participants who were recruited between 2011 and 2018. Participants who were pregnant (n = 247), younger than 18 years old (n = 15 331), lacking information on vitamin D concentration (n = 2446), dual‐energy X‐ray absorptiometry values (n = 9857) or body mass index (n = 33) were excluded. Finally, a total of 11 242 participants were included to evaluate the associations between vitamin D status and appendicular muscle mass.

FIGURE 2

The study assumptions of the two‐sample Mendelian randomization analysis between serum 25(OH)D and appendicular muscle mass. The assumptions include: (1) the genetic instrumental variables (IVs) should exhibit a significant association with serum 25(OH)D; (2) the genetic IVs should not associate with any other potential confounding factors; and (3) the genetic IVs must only through serum 25(OH)D but not any other confounders to influence the appendicular muscle mass. The dotted lines indicate the violate of the assumptions. 25‐hydroxyvitamin D, 25(OH)D; GIANT, Genetic Investigation of Anthropometric Traits; MRC‐IEU, MRC Integrative Epidemiology Unit; SSGAC, Social Science Genetic Association Consortium.