Abstract
OBJECTIVES: Determine the prevalence and identify predictors of hypovitaminosis D in patients with type 2 diabetes mellitus (T2DM); 2) correlate vitamin D levels with variables indicative of glycemic control and cardiovascular risk. RESEARCH DESIGN AND METHODS: We conducted a cross-sectional study with consecutive patients treated at a University Hospital's Endocrinology outpatient clinic located at 12°58'S latitude, between October 2012 and November 2013. Hypovitaminosis D was defined as 25-hydroxyvitamin D < 30 ng/mL (chemiluminescence). RESULTS: We evaluated 108 patients with mean duration of T2DM of 14.34 ± 8.05 years and HbA1c of 9.2 ± 2.1%. Mean age was 58.29 ± 10.34 years. Most were women (72.2%), non-white (89.8%) and had hypertension (75.9%) and dyslipidemia (76.8%). Mean BMI was 28.01 ± 4.64 kg/m(2); 75.9% were overweight. The prevalence of hypovitaminosis D was 62%. In multiple logistic regression, independent predictors of hypovitaminosis D were female gender (OR 3.10, p = 0.02), dyslipidemia (OR 6.50, p < 0.01) and obesity (OR 2.55, p = 0.07). In multiple linear regression, only total cholesterol (β = -0.36, p < 0.01) and BMI (β = -0.21, p = 0.04) remained associated with levels of 25-hydroxyvitamin D. CONCLUSIONS: Using currently recommended cutoffs, the prevalence of hypovitaminosis D in Brazilians with T2DM was as high as that of non-tropical regions. Female gender, dyslipidemia and obesity were predictors of hypovitaminosis D. Low levels of 25-hydroxyvitamin D were correlated with high cholesterol and BMI values. Future studies are needed to evaluate whether vitamin D replacement would improve these parameters and reduce hard cardiovascular outcomes.