Observational studies have shown that vitamin D insufficiency is associated with increased risk of type 2 diabetes, insulin resistance and higher plasma glucose concentrations(Reference Pittas, Lau and Hu1); however, intervention trials to date have been limited(Reference Nagpal, Pande and Bhartia2, Reference Pittas, Harris and Stark3). The effect of vitamin D supplementation (0, 5, 10 and 15 μg cholecalciferol/d) on serum glucose and insulin concentrations and the HOMA-IR was investigated in two randomised placebo-controlled double-blind 22-week intervention studies in men and women aged 20–40 years (n 215; during winter 2006–7(Reference Cashman, Hill and Lucey4)) and ≥64 years (n 215; during winter 2007–8(Reference Cashman, Wallace and Horigan5)) in Cork and Coleraine. Fasting serum levels of glucose were measured by an enzymic colorimetric assay and fasting serum insulin, intact parathyroid hormone (iPTH), and 25-hydroxyvitamin D (25(OH)D) were measured by ELISA at baseline and end point.
a,b,c,d Means with unlike subscript letters were significantly different across the four treatment groups (P<0.05).
ANOVA showed no baseline differences in serum levels of glucose, insulin, HOMA-IR or 25(OH)D between the four treatment groups. In adults aged 20–40 years linear regression analysis showed BMI as the main predictor of baseline serum glucose (β 0.344, P<0.0001), insulin (β 0.357, P<0.0001) and HOMA-IR (β 0.0.382, P<0.0001). Male gender was predictive of higher serum glucose concentration (β −0.184, P<0.005), while increasing age was predictive of lower insulin concentration (β −0.151, P<0.025). In adults aged ≥64 years BMI and iPTH were the main predictors of baseline serum glucose (β 0.184, P<0.008; β −0.204, P<0.003 respectively), whereas the main predictor for insulin and HOMA-IR was BMI (β 0.404, P<0.0001; β 0.409, P<0.0001 respectively). In both age-groups ANCOVA revealed no significant effect of the intervention on glucose, insulin or HOMA-IR concentrations across the four treatment groups, adjusting for centre, age, gender, BMI, vitamin D and calcium intakes, iPTH and 25(OH)D at baseline.
In conclusion, vitamin D supplementation had no effect on fasting serum glucose and insulin concentrations or the HOMA-IR in apparently-healthy adults aged 20–40 and ≥64 years.
We wish to acknowledge the UK Food Standards Agency and the Irish Department of Agriculture, Food & Fisheries through the Food Institutional Research Measure for their support.
