In our analysis of postmenopausal women with angiographically documented CAD, there was very high agreement between the ATP III and IDF definitions for metabolic syndrome. Overall, the two definitions were concordant in 92.2% of these women, and 38 of 45 clinical events occurred among women who met both definitions of the metabolic syndrome. Only 29 of the 372 women met criteria for one definition of the metabolic syndrome but not the other. The lack of a significant difference in prevalence of the metabolic syndrome between the two definitions and lack of difference in prognostic utility appear to result from the overall high prevalence of the metabolic syndrome (68% by both definitions) and the high prevalence of pronounced central obesity in this cohort. The mean BMI was 30.5, and 74% of these women met the more stringent ATP III criterion for elevated waist circumference. As a result, the less stringent IDF waist circumference criterion did not result in a substantial increase in the number of women meeting criteria for the metabolic syndrome.
In previous comparisons of the ATP III and IDF definitions in relatively healthy populations, the IDF definition has defined a slightly larger population of people as having the metabolic syndrome.6–14
These additional individuals categorized by the IDF and not the ATP III tended to be younger and more obese and to have less prevalent cardiovascular disease. Additionally, this increase in prevalence associated with the IDF definition was more pronounced in men than in women. The results of our analysis seem to align with these previously published reports. We studied an older female population with documented coronary heart disease. Given the very high prevalence of obesity and the metabolic syndrome in this population, there is very little variation in these two definitions.
Since the publication of the IDF definition for the metabolic syndrome in 2005, there has been a great deal of speculation about whether the IDF or ATP III definition would be better at predicting cardiovascular events. With the exception of one report in which the IDF was a better predictor of carotid atherosclerosis in women,20
the ATP III definition has been associated with a higher prevalence of cardiovascular disease and has been a better predictor of incident cardiovascular events than the IDF definition.21–24
In populations such as ours with a high prevalence of obesity, including pronounced central obesity, and with underlying coronary heart disease, the majority of individuals will meet criteria for both metabolic syndrome definitions, and the majority of events will occur among individuals who meet criteria for both definitions.
Our study has a number of limitations. First, we only studied postmenopausal women with angiographically documented CAD. Although this limits our ability to extrapolate these findings to other populations, it represents a unique population that has not been studied previously with respect to this clinical question. Second, as only 29 of the 372 women (7.8%) were differentially classified by the two definitions and there were few cardiovascular events overall, we are unable to draw firm conclusions about differences in the clinical characteristics or risk for cardiovascular clinical outcomes between these two definitions in this population, but our analyses suggest that either definition can be used for classification and prognostication. Third, the public use dataset does not specify the proportion of nonwhite women who were of African American, Hispanic, or Asian origin. As the IDF recommends a waist circumference criterion of
cm for women of all ethnicities, our inability to analyze different non-white ethnicities does not result in misclassification.