Comparison of different definitions of the metabolic syndrome in relation to coronary artery disease in high risk Serbian population

While different definitions for the diagnosis of metabolic syndrome (MetS) have been proposed, their applicability brings confusion about which criteria should be used in clinical practice. This was an observational cross-sectional study conducted during October 2008. in 3 university hospital centers in the north, midst and south of the Serbia. 1715 patients were recruited from outpatient clinical practice and primary health care offices: 37% males and 63% females, aged 34-80 years. To evaluate the impact of different criteria in discriminating high risk population for coronary artery disease (CAD) we used NCEP-ATP III, AHA/NHLBI and IDF definitions. 21,7% (373) from the patients included in the study sustained CAD. The prevalence of MetS in the CAD group was 84,7%, 86,1% and 82,0%, respectively, compared with 58,3%, 60,6% and 61,2% in the control group (p<0.0001). ROC curves ploted by the probabilities for CAD calculated in the logistic models for each definition (adjusted for age, sex, smoking and educational status) indicated that NCEP-ATP III and NHLBI-AHA definitions had a better predictive accuracy compared with IDF (p=0,006 and p=0,016, respectively). When the waist girth is introduced in NCEP-ATP III and NHLBI-AHA definitions as obligatory, this distinction was lost. The NCEP-ATP III and AHA/NHLBI definition is more suitable for discrimination of MetS diagnosis, than the later proposed IDF definition in the subjects of the given population. Inclusion of waist circumference as obligatory criteria failed to show increase in predictive accuracy for CAD.