BACKGROUND:The role of inflammation in diabetes development is not fully elucidated. The aim of this work was to investigate the independent effect of individual inflammatory markers and combinations of them on diabetes incidence and the potential mediating role of obesity. METHODS:In 2001 to 2002, a random sample of 1514 men (18-87 years old) and 1528 women (18-89 years old) was selected to participate in the ATTICA study, where Athens is a major metropolis. Interleukin-6 (IL-6), C-reactive protein (CRP), tumour necrosis factor-alpha, serum amyloid alpha, fibrinogen, and homocysteine were measured. Covariates included various clinical, demographic, and lifestyle characteristics, assessed with standard procedures. In 2012, the 10 year follow-up was performed. Diabetes diagnosis was defined according to American Diabetes Association criteria among n = 1485 participants. RESULTS:One hundred ninety-one incident cases of diabetes were documented, yielding an incidence of 12.9% (13.4% in men and 12.4% in women). After adjustments, only elevated IL-6 increased by 2.2 times the 10 year diabetes risk (third vs first tertile, 95% CI: 1.13, 4.28). After investigating combinations of inflammatory markers, combined elevated levels of CRP and IL-6 or CRP and fibrinogen (both markers ≥75th percentile vs <75th percentile) increased the risk by 1.93 times (95% CI: 1.20, 3.08) and 2.37 times (95% CI: 1.37, 4.16), respectively. Body mass index was found to significantly mediate the aggravating effect of inflammation. CONCLUSIONS:The reported results underline the significant role of individual IL-6 or combinations of CRP-IL-6 and CRP-fibrinogen in diabetes prediction. Adiposity seems to be primarily responsible for an increase in inflammatory markers, leading through this mechanism to insulin resistance and increasing diabetes risk.