Abstract

Objective– We examined whether metabolic syndrome (MetS) predicts incident type 2 diabetes (T2DM) more effectively than impaired fasting glucose (IFG) in a general Japanese population.

Research Design and Methods– A total of 1,935 non-diabetic subjects aged 40-79 years were followed-up prospectively for a mean of 11.8 years.

Results– During the follow-up, 286 subjects developed T2DM. Compared to those without MetS, the multivariate-adjusted hazard ratio (HR) for incident T2DM was significantly higher in subjects with MetS in both sexes, even after adjustment for confounding factors: age, family history of diabetes, total cholesterol, alcohol intake, smoking habits and regular exercise (men: HR=2.58; 95%CI, 1.85-3.59; women: HR=3.69; 95%CI, 2.58-5.27). The multivariate-adjusted HR of MetS for T2DM was slightly lower in men and similar in women compared to that of IFG. The multivariate-adjusted HR for T2DM rose progressively as the number of MetS components increased in both subjects with and without IFG. In stratified analysis, compared to subjects with neither MetS nor IFG, the multivariate-adjusted risk of T2DM was significantly higher in subjects with MetS alone (HR=2.37; 95%CI, 1.45-3.88) or IFG alone (HR=3.49; 95%CI, 2.57-4.74) and markedly increased in subjects with both MetS and IFG (HR=6.76; 95%CI, 4.75-9.61). Furthermore, the multivariate-adjusted risk for T2DM was also significantly higher in subjects with both MetS and IFG than in those with either one alone (both p<0.001).

Conclusions– Our findings suggest that MetS significantly increases the risk of incident T2DM, independent of IFG, and is therefore a valuable tool to identify individuals at high risk of T2DM.