Associations of glycated albumin and fructosamine with glycaemic status in urban black South Africans

Background: The associations of glycated albumin (GA) and fructosamine (FA) as alternate tests for diabetes screening in South Africans were examined to overcome shortcomings with oral glucose tolerance tests (OGTTs). Methods: Based on OGTTs, glycaemic status included 1) normal glucose, 2) dysglycaemia (impaired fasting glucose, impaired glucose tolerance, newly diagnosed diabetes), and 3) known diabetes. Results: Among 1092 participants, 21 years, mean GA (16.9%) and FA (230.2 mu mol/l) increased significantly by age and worsening glycaemic status and were significantly higher in women vs. men and BMI (kg/m2) >= 30 vs. <30. For dysglycaemia, correlations of GA and FA with fasting and 2-hour glucose levels were higher in obese (0.576 to 0.688) vs. non-obese (-0.010 to 0.522). Optimal GA threshold to identify dysglycaemia was 15.35% and comparable, but with lower sensitivity (0.54) and specificity (0.55), to Asian studies. For FA, the optimal cutpoint of 227.0 mu mol/l approximated that described in the literature. Dysglycaemia and known diabetes were associated with GA, while only known diabetes was related to FA, in models adjusted for age, gender and obesity. Conclusions: Potential exists for GA and/or FA as alternative measures of dysglycaemia in clinical practice in Africans, but longitudinal studies are required to clearly elucidate their utility.