Vitamin D supplementation does not prevent diabetes

Référence

Pittas AG, Dawson-Hughes B, Sheehan P, et al, for the D2d Research Group. Vitamin D supplementation and prevention of type 2 diabetes. N Engl J Med 2019;381(6):520-530.

Sommaire

Although it is biologically plausible that vitamin D supplementation could delay the onset of diabetes ("It should work…") and there is some data from observational studies supporting this hypothesis, there is no evidence from prospective randomized trials that is actually works. The authors of this study identified adults 30 years and older with a body mass index between 24 and 42 kg/m^2 who met at least 2 of the following criteria: (1) fasting glucose level 100 to 125 mg/dL (5.6 - 6.9 mmol/L), (2) glucose level of 140 to 199 mg/dL (7.8 - 11.0 mmol/L) 2 hours after a 75-g glucose load, and (3) glycated hemoglobin of 5.7% to 6.4% (39 - 47 mmol/L). American Indians, Alaska Natives, Native Hawaiians, and Pacific Islanders were included if they were at least 25 years old. The mean age of participants was 60 years, 45% were female, the mean glycated hemoglobin was 5.9%, and the mean vitamin D level was 28 ng/mL. Approximately 260 in each group had a vitamin D level of 19 ng/mL or lower (the usual definition of normal vitamin D is greater than 20 ng/mL), approximately 440 in each group had a vitamin D level of 20 to 29 ng/mL, and the remainder had a value of 30 ng/mL or higher. Groups were balanced at the start of the study, analysis was by intention to treat, but allocation concealment was not described. All patients were advised to avoid vitamins, medications for diabetes or weight loss, and supplemental calcium. Of 7133 patients screened for the study, 2423 were randomized to receive either vitamin D 4000 IU daily or placebo. After a median follow-up of 2.5 years, 99% of participants were still accounted for (5 deaths in each group and 62 total withdrawals). At the end of the trial, there was no significant difference in the likelihood of developing T2DM between the vitamin D group and the placebo group (hazard ratio 0.88, 95% CI 0.75 - 1.04). The absolute cumulative risks of T2DM were 24.2% in the vitamin D group and 26.6% in the placebo group. The authors did a large number of prespecified subgroup analyses, with most of the hazard ratios for the association between vitamin D and T2DM less than 1.0, but none of the associations were statistically significant. There was no difference in the hazard ratios for patients with baseline vitamin D less than 20 ng/mL compared with those who had higher vitamin D levels. In a post hoc analysis, the likelihood of developing T2DM was lower in those with a very low baseline vitamin D level (< 12 ng/mL), so that group may be worth further study.