Comparison of 2 hour plasma glucose and serum fructosamine concentrations as screening tests for diabetes mellitus

ABSTRACT

To evaluate the clinical usefulness of fructosamine as a mean for diagnosing Diabetes mellitus, serum fructosamine and plasma glucose levels 2 h following 75 g Oral glucose load were compared in 158 subjects participating in a community Health survey. The subjects were classified as having normal, impaired or diabetic responses using WHO criteria of 135 subjects with normal and 7 subjects with impaired glucose tolerance. 130 and 5 subjects respectively had fructosamine within the ninty-fifth percentile. Of 16 subjects with diabetic response, 13 presented with high fructosmine. Fructosmaine assay yielded 7 false positive and 3 false negative results. Consequently, the sensitivity, specificity, predictive value positive, predictive value negative and efficiency calculated were 81.3%, 95.0%, 65.0%, 97.8% and 94.0% respectively. A highly significant correlation was observed between 2 h plasma glucose and measured fructosamine [r= 0.76, P<0.001] and between 2 h plasma glucose and corrected fructosamine [r= 0.77, P<0.001] in the studied population. A significant correlation was also observed between serum albumin and measured fructosamine [r= 0.33. P<0.001] in the non-diabetic subjects, and between measured and corrected fructosamine [r= 0.92, P<0.001] in the studied population. The results suggest that serum fructosamine provides an additional useful screening method for diabetes mellitus that may complement glucose measurement and correction of its level for albumine concentration may be unnecessary in subjects with normal protein metabolism