Evaluation of agreement between hemoglobin A1c, fasting glucose, and fructosamine in Senegalese individuals with and without sickle-cell trait

Abstract

Fasting glucose (FG) and glycated hemoglobin A1c (HbA1c) perform sub-optimally in people of African origin, especially in individuals with sickle-cell trait (SCT). The purpose of this study was to compare the relationships between HbA1c, FG, and fructosamine in individuals from Senegal with and without SCT. HbA1c, FG, and fructosamine were measured in 203 adults from Senegal (100 control: 45 with type 2 diabetes (T2D); 103 SCT: 51 with T2D). Significant, positive correlations were observed between HbA1c and FG, fructosamine and FG, and fructosamine and HbA1c in both groups. The limits of agreement were inappropriately large in both groups for the Bland-Altman plots of HbA1c and FG (control: -95.97 to 83.97%; SCT: -115.9 to 91.52%), fructosamine and FG (control: -100.6 to 99.89%; SCT: -105.6 to 100.6%), and fructosamine and HbA1c (control: -52.03 to 38.98%; SCT: -88.04 to 71.41%). In both groups, the greatest proportion of subjects were considered above the clinical cut-point for hyperglycemia when fructosamine was used as the criterion (control: 33%; SCT: 44.6%), and the lowest percentage of subjects were classified as over the clinical cut-point when HbA1c was used as the criterion (control: 21%; SCT: 27.7%).Substantial disparities between HbA1c, FG, and fructosamine were observed in both groups, and these differences were exaggerated in the SCT group. Therefore, these three biomarkers should not be considered to be interchangeable measures of glycemic control. These biomarkers should be used thoughtfully, and special care should be taken when using them in individuals with SCT.