Adiponectin as predictor for diabetes among pre-diabetic groups.

Adiponectin is found to associate with diabetes in studies apart from cohort studies. This prospective cohort study is to evaluate the predictive role of adiponectin in diabetes among participants with impaired fasting glucose (IFG). A total of 42,845 participants who visited 7 health examination centers located in Seoul and Kyunggi province, South Korea, during 2004-2008 were first included. Of the 42,845 participants, 5,085 participants had IFG. IFG was categorized as stage 1 (fasting glucose 100-109 mg/dL) or stage 2 (110-125 mg/dL). The incidence rates of diabetes were followed up to December, 2011. Hazard ratios (HRs) and 95 % confidence intervals (CI) were performed by Cox proportional hazard model. Of the 5,085 participants, 652 participants developed diabetes during a mean follow-up of 4.4 years. Low adiponectin was associated with diabetes among men with stage 2 IFG (HR, 1.78; 95 %CI, 1.33-2.38) while it was associated with diabetes among women with stage 1 IFG (HR, 2.64; 95 %CI, 1.38-5.03) and stage 2 IFG (HR, 2.17; 95 %CI, 1.07-4.42). When combined men and women, the association between adiponectin and diabetes was statistically significant in stage 2 IFG with an increase of about 82 % (HR, 1.82; 95 %CI, 1.40-2.39) after adjusting for age, sex, body mass index, waist circumference, and fasting serum glucose. There was an interaction by sex and stage 1 IFG in the association between adiponectin and risk of diabetes (P < 0.001). Adiponectin was independently associated with diabetes among participants with IFG. This association was apparent in stage 2 IFG. Adiponectin may be used as a predictor of diabetes in patients having IFG.