Comparative associations of diabetes risk factors with five measures of hyperglycemia

OBJECTIVE: To compare the associations of diabetes mellitus risk factors with nontraditional markers of hyperglycemia (glycated albumin, fructosamine, 1,5-anhydroglucitol (1,5-AG)) to those observed with traditional markers (fasting glucose, hemoglobin A1c (HbA1c)). DESIGN: Cross-sectional study. SETTING: The community-based Atherosclerosis Risk in Communities (ARIC) Study cohort. PARTICIPANTS: A subsample of 1764 participants (309 with diagnosed diabetes and 1455 without diagnosed diabetes) from the ARIC Study who attended a clinic visit in 2005–2006. MAIN OUTCOME MEASURES: Elevated levels of glycated albumin, fructosamine, 1,5-AG, fasting glucose, and HbA1c in persons with and without a diagnosis of diabetes. RESULTS: The mean age of participants was 70 years (SD, 6), 43% were men, and 20% were African-American. Black race and family history of diabetes were generally positively associated with elevated levels of all biomarkers of hyperglycemia except 1,5-AG, which showed inverse but weaker associations with the risk factors examined. In general, patterns of risk factor associations observed for fasting glucose and HbA1c were similar to those observed for the nontraditional biomarkers of hyperglycemia but with one clear exception: body mass index (BMI). In persons without a diagnosis of diabetes, BMI was positively associated with fasting glucose and HbA1c, but the associations of BMI with glycated albumin and fructosamine were inverse, with high values of these markers at low levels of BMI. 1,5-AG, which is lowered in the setting of hyperglycemia, was positively associated with BMI. CONCLUSIONS: Traditional diabetes risk factors have similar associations with glycated albumin and fructosamine as those for fasting glucose and HbA1c, with the exception of BMI. Risk factor associations with 1,5-AG were mostly inverse. The inverse associations of BMI with glycated albumin and fructosamine, and positive associations with 1,5-AG, may reflect pathways independent of glucose metabolism and merit further examination.