The clinical value of metabolic syndrome and risks of cardiometabolic events and mortality in the elderly: the Rotterdam study.

BACKGROUND: To evaluate the clinical value of metabolic syndrome based on different definitions [American Heart Association/National Heart, Lung and Blood Institute (AHA/NHLBI), International Diabetes Federation (IDF) and European Group for the Study of Insulin Resistance (EGIR)] in middle-aged and elderly populations. METHODS: We studied 8643 participants from the Rotterdam study (1990-2012; mean age 62.7; 57.6 % female), a large prospective population-based study with predominantly elderly participants. We performed cox-proportional hazards models for different definitions, triads within definitions and each separate component for the risk of incident type 2 diabetes mellitus, coronary heart disease, stroke, cardiovascular- and all-cause mortality. RESULTS: In our population of 8643 subjects, metabolic syndrome was highly prevalent (prevalence between 19.4 and 42.4 %). Metabolic syndrome in general was associated with incident type 2 diabetes mellitus (median follow-up of 6.8 years, hazard ratios 3.13-3.78). The associations with coronary heart disease (median follow-up of 7.2 years, hazard ratios 1.08-1.32), stroke (median follow-up of 7.7 years, hazard ratios 0.98-1.32), cardiovascular mortality (median follow-up of 8.2 years, ratios 0.95-1.29) and all-cause mortality (median follow-up of 8.7 years, hazard ratios 1.05-1.10) were weaker. AHA/NHLBI- and IDF-definitions showed similar associations with clinical endpoints compared to the EGIR, which was only significantly associated with incident type 2 diabetes mellitus. All significant associations disappeared after correcting metabolic syndrome for its individual components. CONCLUSIONS: Large variability exists between and within definitions of the metabolic syndrome with respect to risk of clinical events and mortality. In a relatively old population the metabolic syndrome did not show an additional predictive value on top of its individual components. So, besides as a manner of easy identification of high risk patients, the metabolic syndrome does not seem to add any predictive value for clinical practice.