Glycosylated Hemoglobin and Prevalent Metabolic Syndrome in Nondiabetic Multiethnic U.S. Adults

Abstract

Metabolic syndrome poses a significant risk for cardiovascular disease. Recently, glycosylated hemoglobin (HbA1c) has been included in the diagnostic criteria for diabetes mellitus and prediabetes. We sought to determine if HbA1c is associated with prevalent metabolic syndrome in nondiabetic U.S. adults. A total of 9,022 nondiabetic participants of National Health and Nutrition Examination Surveys 1999-2008 (age, 47.5 ± 18.3 years, 51% females) were divided into quintiles (Q) of HbA1c: Q1 (reference), ≤5%; Q2, 5.1%-5.3%; Q3, 5.4%-5.5%; Q4, 5.6%-5.7%; and Q5, ≥5.8%. Modified National Cholesterol Education Program Adult Treatment Panel III criteria were used to identify metabolic syndrome (n=2,821; 31.3%). Unadjusted and adjusted multivariate logistic regression analysis was performed to assess the risk of metabolic syndrome. A graded increase in odds of having prevalent metabolic syndrome with increase from each quintile of HbA1c compared to Q1 was observed after adjusting for age, sex, race, body mass index (BMI), total cholesterol, lipid-lowering therapy, current smoking, family history of diabetes, C-reactive protein, and fasting insulin. Stratified analysis based on gender, ethnicity, and BMI showed similar results. The HbA1c value of ≥5.4% remained appropriate cutoff for predicting metabolic syndrome in Caucasians and Hispanics, whereas ≥5.6% provided the best accuracy for African Americans based on receiver operating characteristics analysis. HbA1c much below the level for prediabetes was associated with prevalence of the metabolic syndrome in a cohort of nondiabetic U.S. adults. HbA1c can be considered as a surrogate marker for metabolic syndrome in nondiabetics.