Objective: To compare the differences of five diagnostic criteria used for metabolic syndrome (MS), issued by International Diabetes Federation (IDF), the National Cholesterol Education Program (ATP III), America-Heart-Association (AHA), Chinese Medical Association Diabetes Branch (CDS) and The Taiwan Health Bureau (TAIWAN), during a physical check-up program among population aged 35-74 years, in Taiwan.
Methods: A total number of 28 408 people who had received physical checkup program first time at the MJ centers, were recruited from 2005 to 2007. The prevalence of MS and the degree of agreement were both calculated according to the five definitions and the results of MS components. Distributions and risk factor aggregation of the results were also analyzed.
Results: According to the five definitions (1) The range of age-adjusted prevalence of MS appeared to be 10.6% (CDS) -23.6%(AHA), and were 13.4% (CDS) -27.6% (AHA) and 8.0% (CDS) -20.5% (IDF) for men and women respectively. (2) The range of five MS components were 22.5% (low-HDL-C) -39.7% (high FPG), with 22.3% of the total subjects presented at least 3 risk factors. In addition, 0% (AHA), 6.7% (TAIWAN), 6.9% (ATP III), 8.9% (IDF) and 14.9% (CDS) of the subjects diagnosed as MS-free, by the five criterions, also appeared of having > or = 3 risk factors. (3) Among all the MS subjects, the proportions of clinical symptom complex, having 5, 4 and 3 MS components were 8.0%, 29.5% and 62.5% respectively. The most common clinical symptoms complex of MS were obesity, hypertension and high FPG. (4) The MS diagnostic criteria of ATP III, AHA and TAIWAN were in good accordance with Kappa index, showing 0.81-0.98 for the three criteria. CDS and IDF were in relatively weak agreement when comparing with other definitions with Kappa index showed as 0.35 and 0.62.
Conclusion: Our findings revealed big differences in the prevalence and aggregation of risk components on MS, when using the five definitions. We suggested that prospective cohort studies be planned to investigate the impact on cardiovascular disease morbidity and mortality so as to verify whites criterion might be suitable to the population in Taiwan, considering the possible bias.