Sex-Dependent Association of Vitamin D With Insulin Resistance in Humans.

Abstract

Animal studies suggested that vitamin D might decrease insulin resistance. Estrogen increased insulin sensitivity and glucose tolerance in rodents. However, sex-specific association of vitamin D with insulin resistance in humans remains unclear. To investigate the sex-dependency of the association of insulin resistance and 25-hydroxyvitamin D [25(OH)D] in a large Caucasian population. Cross-sectional study from out-patients' blood samples with measurements of 25(OH)D and homeostatic model assessment of insulin resistance (HOMA-IR) drawn at exactly the same day (n = 1887). This cohort was divided into 3 groups: (1) group with vitamin D deficiency (n = 1190), (2) group with vitamin D sufficiency (n = 686), and (3) vitamin D excess groups (n = 11); the vitamin D excess group was excluded from further analysis due to the small size. Analysis of the entire study population showed that serum 25(OH)D was inversely associated with HOMA-IR [Spearman correlation coefficient (rs) = -0.19, P < 0.0001]. When considering the vitamin D status, this association was only seen in the vitamin D deficiency group but not in the vitamin D sufficient group. The correlation was sex-dependent: HOMA-IR was inversely correlated with vitamin D in women with vitamin D deficiency (rs = -0.26, P < 0.0001) but not in men with vitamin D deficiency (rs = 0.01, P = 0.714). After multivariate linear regression analysis considering confounding factors, this relationship was again only seen in women. Vitamin D was inversely and independently associated with insulin resistance only in women with vitamin D deficiency. Based on our data, we suggest that in particular vitamin D deficient women might benefit from vitamin D substitution by improving insulin resistance. This, however, needs to be proven in adequately designed double-blind placebo-controlled clinical studies.