Association between serum osteocalcin and markers of metabolic phenotype.

CONTEXT: Osteocalcin has been reported to contribute to the regulation of glucose tolerance and insulin secretion and sensitivity in experimental animals. OBJECTIVE: Our objective was to examine the association between serum osteocalcin concentration and markers of dysmetabolic phenotype using data from a completed clinical trial in adults age 65 and older [n = 380, mean age 71 yr, body mass index (BMI) 26.9 kg/m(2), 5% with diabetes]. RESEARCH DESIGN AND METHODS: In cross-sectional analyses (baseline data), we estimated the associations of serum osteocalcin and urine N-telopeptide with markers of metabolic phenotype including fasting plasma glucose (FPG) (primary outcome), fasting insulin, insulin sensitivity estimated by homeostasis model assessment for insulin resistance, plasma high-sensitivity C-reactive protein, IL-6, and measures of adiposity (BMI and body fat) (secondary outcomes) after multivariate adjustment for potential confounders. In prospective analysis (placebo arm), we estimated the associations of osteocalcin and N-telopeptide with change in the primary outcome, FPG, over a 3-yr period. RESULTS: In cross-sectional analyses, serum osteocalcin concentration was inversely associated with FPG (P = 0.01), fasting insulin (P = 0.006), homeostasis model assessment for insulin resistance (P = 0.002), high-sensitivity C-reactive protein (P = 0.01), IL-6 (P = 0.02), BMI (P < 0.001), and body fat (P < 0.001). When participants were divided into tertiles by serum osteocalcin, mean FPG was 97.1 vs. 104.8 mg/dl in the highest vs. lowest osteocalcin tertile, respectively (P < 0.01). In prospective analyses, exposure to higher osteocalcin levels during follow-up was associated with a significantly lower rise in FPG at 3 yr. Urine N-telopeptide was not associated with any marker of metabolic phenotype. CONCLUSIONS: Serum osteocalcin concentration was inversely associated with blood markers of dysmetabolic phenotype and measures of adiposity. Our findings should be considered hypothesis generating, and they need to be replicated in human studies designed to test the hypothesis that osteocalcin affects metabolism.