Recently, the International Diabetes Federation (IDF) consensus (1) proposed a new definition for diagnosing metabolic syndrome. The new IDF definition includes a lower waist circumference than the National Cholesterol Education Program Adult Treatment Panel III (ATP III) criteria (2) for diagnosing abdominal obesity.
Since the most frequently used definitions for metabolic syndrome involve different criteria for diagnosis of obesity, and because differences in the prevalence of metabolic syndrome seem to reproduce differences in the prevalence of adiposity (3), we determined the concordance between the 2005 IDF definition for metabolic syndrome with the ATP III and World Health Organization (WHO) (4) definitions in a population from northern Mexico.
This is a report of a population-based study of apparently healthy men and nonpregnant women aged 30–64 years from Durango City in northern Mexico who were selected through a randomized two-stage cluster sampling procedure.
The cutoff value we used for abdominal obesity was recommended for the IDF consensus for ethnic South and Central Americans (≥90 cm in men and ≥80 cm in women) and corresponds to the upper quartile in our population. To assess the degree of agreement between different metabolic syndrome definitions, we used the weighted κ test.
A total of 472 (67.4%) women and 228 (32.6%) men were studied. The mean age was 44.7 ± 11.8 years, and the mean BMI was 29.1 ± 5.3 kg/m2. The prevalence of metabolic syndrome was 22.3, 22.6, and 15.4% according to the IDF, ATP III, and WHO definitions, respectively.
The IDF definition failed to detect 7.6% of ATP III patients with metabolic syndrome, whereas 5.2% of the participants who were classified as normal by the ATP III definition had metabolic syndrome according to IDF criteria (sensitivity and specificity of 92.4 and 94.8%). The κ statistic for the agreement between IDF and ATP III definitions was 0.873.
The IDF definition failed to detect 15.7% of the WHO subjects with metabolic syndrome, whereas 26.8% of the participants who were classified as normal by the WHO definition had metabolic syndrome by IDF criteria (sensitivity and specificity of 84.3 and 73.1%). The κ statistic for the agreement between the IDF and WHO definition was 0.511.
The IDF criteria detected a higher prevalence of obesity (33.7%) than the ATP III and WHO criteria (23.7 and 22.6%, respectively). The WHO definition detected significantly fewer subjects with high blood pressure (5.7 vs. 12.3% of ATP III and IDF definitions) and low HDL cholesterol (22.9 vs. 34.1% of ATP III and IDF criteria). The prevalence of hyperglycemia and hypertriglyceridemia was 26.0 and 22.4%.
The lower cutoff point for abdominal obesity, according to the IDF definition, included 53.6 and 7.1% of the overweight and lean subjects (according to BMI) in the sample. Therefore, lowering the cutoff for abdominal obesity has the benefit of an early recognition of subjects at risk and the possibility of early lifestyle intervention.
In the population from northern Mexico, the IDF definition for metabolic syndrome has a high concordance with the ATP III definition, identifying similar proportions of subjects with metabolic syndrome and a low concordance with the WHO definition.
