Experimental evidence suggests that osteocalcin is a key messenger that affects both adipocytes and insulin-producing β cells. Epidemiological cross-sectional studies have shown a negative association between plasma levels of osteocalcin and glucose. For this reason, the hypothesis that lower baseline osteocalcin plasma levels are associated with diabetes was prospectively tested.

The study population consisted of individuals at high risk for type 2 diabetes who were screened for participation in the Greek arm of a European type 2 diabetes prevention study (the DE-PLAN study). All participants were free of diabetes at baseline and underwent a second evaluation 3 years later. Diabetes status was defined according to an oral glucose tolerance test.

A total of 307 subjects were included in the present analysis. The population, including 154 men (50.3%), was middle-aged (54.4±10.2 years) and overweight (BMI: 29.5±4.9kg/m²). At baseline, mean total plasma osteocalcin was lower in those with impaired fasting glucose and/or impaired glucose tolerance compared with those with normal glucose tolerance (6.0±3.1ng/mL vs. 7.3±4.0ng/mL, respectively; P=0.01). After 3 years, 36 subjects had developed diabetes. In the prospective evaluation, there was no association between baseline osteocalcin levels and diabetes (OR: 1.04 per 1ng/mL, 95% CI: 0.93–1.15; P=0.49) on multivariable logistic regression analysis, nor was there any correlation with changes in plasma glucose after 3 years (r=0.09, P=0.38).

Our prospective results show that lower levels of circulating osteocalcin do not predict future diabetes development and, in contrast to most cross-sectional published data so far, suggest that this molecule may not be playing a major role in glucose homoeostasis in humans.