A dosagem da frutosamina no diagnóstico do Diabetes Mellitus – contribuição do estudo longitudinal de Saúde do Adulto (ELSA-BRASIL)

Abstract

Fructosamine is the result of non-enzymatic glycation of glucose to proteins, notably albumin. Its use has historically been associated with monitoring of diabetes mellitus, and more recently as a diagnostic test. Fructosamine provides information on the control of diabetes in the last 2 to 4 weeks, a period closely related to the half-life of albumin. One of the objectives of this sstudyy is the evaluation of its performance against other established diagnostic markers, fasting plasma glucose (FPG), 2 h glucose after oral glucose overload (2-hr PG) and glycated hemoglobin (A1c) may contribute to the definition of its role in the diagnosis of diabetes. Another aim is the generation of a specific reference interval for the Brazilian population, not yet available. This is a cross-sectional study with baseline data (2008-2010) of 2228 individuals from the Longitudinal Study of Adult Health (ELSA-Brazil). The establishment of the reference interval followed the protocol suggested by the Clinical and Laboratory Standards Institute (CLSI). The levels of fructosamine were determined by the colorimetric method (Nitroblue Terrazolium - NBT). Associations between the fructosamine and the other markers were defined by Spearman's correlation coefficient. The evaluation between the diagnostic performances was performed by the Area under the Curve (AUC – ROC curve). Fructosamine presented significant correlations with FPG, r = 0.26 (p <0.001), 2-hr PG, r = 0.10 (p <0.001) and A1c r = 0.22 (p <0.001). The correlations between FPG and A1c were similar to each other, but higher than the correlation with 2-h PG. The AUC of the fructosamine (0.741), when using criteria based on FPG, was similar to A1c (0.782) for diabetes diagnosis (p = 0.207). When 2h PG was used as the diagnostic criterion, the area of A1c (0.745) was similar to the area of fructosamine (0.678) with borderline significance (p = 0.053). Fasting glucose and 2-h PG had AUCs (0.844 and 0.801, respectively) higher than that of fructosamine (0.700) when diabetes was defined by A1c (p = 0.005 and p = 0.064, respectively). A total of 466 individuals were selected for the establishment of the reference interval. Those individuals known to be diabetic, individuals with alterations in the classic tests for hyperglycemia and those with alterations in the tests of microalbuminuria and creatinine were excluded. The reference interval was 186 to 248 μmol / L for women and 196 to 269 μmol/L for men. The levels of fructosamine were higher in males than in females (p = 0.006) and in the non-white population (p = 0.034). In addition, it presented a negative correlation with body mass index (r = -0.117; p = 0.011). Frutosamine is associated with FPG, 2h-PG and A1c, potentially useful in the diagnosis in additional to its recognized role in monitoring of diabetes. Additional studies suggesting appropriate cutoff values for diabetes screening, should be considered. The reference range created in this study will allow a more adequate interpretation of the fructosamine test, especially in situations where glycated hemoglobin cannot be used.