Analysis of criteria for metabolic syndrome in a population-based study of Japanese-Brazilians

Abstract

Objective: Criteria for metabolic syndrome (MS) differ particularly regarding the definition of central obesity and consequently, there could be differences in the assessment of cardiovascular risk. We estimated the prevalence of metabolic syndrome, compared the agreement of the World Health Organization (WHO) criteria with the standard and a modified National Cholesterol Education Program (NCEP) criterion and investigated whether additional factors were associated with the diagnosis of the syndrome in a Japanese descendant population.Methods: In this cross-sectional, population-based survey, 1166 Japanese-Brazilians (533 men, 633 women) aged 57.4 +/- 12.4 years with mean body mass index (BMI) and waist of 25.2 +/- 4.0 kg/m(2) and 84.5 +/- 10.6 cm, respectively, were included. McNemar and kappa statistics were used to assess the concordance between WHO criteria with the standard and a modified NCEP criteria (waist of 90 and 80 cm, for men and women, respectively). in logistic regression analysis, a number of metabolic variables and albumin-to-creatinine ratio were included to test independent associations with metabolic syndrome defined by the modified NCEP criteria.Results: According to WHO, 55.4% (95% Cl 52.5-58.2%) of the subjects had MS and to NCEP 47.4% (95% Cl 44.6-50.0%). WHO criterion detected 48.3% of central obese subjects while NCEP only 14.0%. Kappa statistics showed a good strength of agreement (k = 0.67, p < 0.01) between WHO and NCEP standard definitions of MS. Using the modified NCEP criterion for Asians, more subjects with metabolic syndrome were identified (58%) and agreement with WHO was improved (k = 0.72, p < 0.001). However, similar Framingham risk scores were attributed to the subsets of subjects classified by any of the three criteria. Areas under the receiver operating characteristic curves, obtained for the modified waist values to diagnose metabolic syndrome according to WHO, were > 0.80 and corresponded, respectively, to sensitivity and specificity of 63 and 83% for men and 77 and 72% for women. In final logistic regression model, age, male sex, BMI and homeostasis model assessment-insulin resistance but not with albumin-to-creatinine ratio (ACR) were independently associated with the syndrome.Conclusions: High prevalence of MS, independent of the criterion considered, was found in this Japanese-Brazilian population. The replacement of waist cutoff by those proposed by WHO for Asians lead to this diagnosis in a higher number of subjects with elevated cardiovascular risk. Our data did not support that ACR should be included in the classical definition of MS in Japanese descendants as previously suggested by WHO.