Objective: The diagnostic criteria and the clinical usefulness of the metabolic syndrome (MetSy) are currently questioned. The objective was to describe the structure of MetSy and to evaluate its components for prediction of diabetes type 2 (T2DM).

Research methods and procedures: This was a case-referent study nested within a population-based health survey. Among 33,336 participants, we identified 177 initially non-diabetic individuals who developed T2DM after 0.1 to 10.5 years (mean, 5.4 years), and, for each diabetes case, two referents matched for sex, age, and year of health survey. Baseline variables included oral glucose tolerance test, BMI, blood pressure, blood lipids, adipokines, inflammatory markers, insulin resistance, and beta-cell function. Exploratory and confirmative factor analyses were applied to hypothesize the structure of the MetSy. The prediction of T2DM by the different factors was evaluated by multivariate logistic regression analysis.

Results: A hypothetical five-factor model of intercorrelated composite factors was generated. The inflammation, dyslipidemia, and blood pressure factors were predicitive only in univariate analysis. In multivariable analyses, two factors independently and significantly predicted T2DM: an obesity/insulin resistance factor and a glycemia factor. The composite factors did not improve the prediction of T2DM compared with single variables. Among the original variables, fasting glucose, proinsulin, BMI, and blood pressure values were predictive of T2DM.

Discussion: Our data support the concept of a MetSy, and we propose five separate clusters of components. The inflammation and dyslipidemia factors were not independently associated with diabetes risk. In contrast, obesity and accompanying insulin resistance and beta-cell decompensation seem to be two core perturbations promoting and predicting progression to T2DM.