An Assessment of the Utility of Serum Fructosamine in the Diagnosis and Monitoring of Diabetes Mellitus.

backgroundFructosamine (FA) has gained importance as a new biomarker for hyperglycemia in the past decade and may be of indispensable use in certain conditions where hemoglobin A1c (HbA1c) falls short of utility such as disorders of red blood cells, patients with rapid glycemic excursions requiring more short-term monitoring, pregnancy, chronic kidney disease, etc. Methods: The present study was a hospital-based observational cross-sectional study conducted in the Department of Biochemistry, All India Institute of Medical Sciences (AIIMS), Nagpur, India. Serum HbA1c, FA, albumin-corrected fructosamine (AlbF), total protein-corrected FA (PrF), hemoglobin (Hb), and hematocrit (Hct) were estimated in 32 controls (Group I) and 32 cases of diabetes mellitus (DM) (Group II). The clinical data and lab results were presented as mean±SD/±standard error (SE) of the mean. Student's t-test and ANOVA were used to compare various parameters between the groups. Pearson correlation analysis was performed to assess the correlation between different diagnostic parameters. The receiver operating characteristic (ROC) curve was plotted to assess the diagnostic significance and cut-off value for FA, AlbF, and PrF.

resultsThe controls and cases were matched for age and gender distribution. Serum HbA1c (p<0.0001), serum FA (p<0.0001), fasting blood sugar (p=0.001), postprandial blood sugar (p<0.0001), random blood sugar (p=0.001), hematocrit (p=0.002), AlbF (p<0.0001), and PrF (p<0.0001) were found to be significantly higher in known diabetic subjects compared to controls. The case group was further subdivided into pre-diabetic and diabetic groups. On correlation analysis of HbA1c with various parameters, a moderate correlation of HbA1c was noted with FA (r=0.522, p<0.0001) and AlbF (r=0.375, p=0.002) in all subjects. Additionally, a moderate correlation of FA (r=0.479, p=0.033), AlbF (r=0.444, p=0.050), and PrF (r=0.441, p=0.065) with HbA1c was also found in subjects with diabetic range glycemia. No such correlation was noted in the pre-diabetic group. No significant correlation was noted between FA and its corrected values in any range of glycemia. None of the parameters assessing glycemia were found to be significantly affected by hemoglobin status. On ROC curve analysis, HbA1c was found to be the best parameter (area under the curve (AUC) =83%, p<0.0001) followed by AlbF (AUC= 80.5%, p<0.0001) and uncorrected FA (AUC=80.5%, p<0.0001) to diagnose DM.

conclusionSerum FA should be considered a valid diagnostic biomarker and of indispensable use in special populations where HbA1c falls short of utility such as patients with red blood cell disorders or those showing rapid glycemic excursions such as those on corticosteroid therapy or insulin therapy, etc. It exhibits additional advantages over HbA1c with respect to lower reagent cost and easy automation on any conventional laboratory instruments based on simple colorimetry.