Selective determination of non-enzymatic glycosylated serum albumin as a medium term index of diabetic control.

Serum proteins are non-enzymatically glycosylated dependent on the concentration of free glucose and measurements of their concentration are used to control diabetic carbohydrate metabolism. Eight patients with insulin-dependent diabetes mellitus (IDDM) and 8 patients with non-insulin-dependent diabetes mellitus (NIDDM) with glycosylated hemoglobin levels of at least 10.5% were studied during a 6-week period of antidiabetic therapy. Glycosylated serum albumin (GSA) and glycosylated total serum proteins (GSP) were measured weekly using an affinity chromatography procedure. The fructosamine test (FA) and the measurement of mean blood glucose (MBG) were also carried out weekly. Glycosylated hemoglobin and its glucose adduct HbA1c were determined at 14-day intervals (HPLC-method). All measured parameters decreased during the period of the study. The correlation coefficients for the glycosylated proteins versus the MBG determined one week earlier were highest for GSA [IDDM: r(GSA/MBG-1) = 0.726, p < 0.001 for the single values and 0.984, p < 0.001 for the mean values; NIDDM: r (GSA/MBG-1) = 0.636, p < 0.001 for the single values and 0.986, p < 0.001 for the mean values]. The differences between the IDDM and NIDDM group probably occurred because 6 NIDDM patients were taking glibenclamide (7.0-10.5 mg/day) which is known to inhibit the glycosylation reaction of albumin. The fructosamine test is more prone to interferences than the selective determination of GSA. GSA determination therefore, gives precise data in medium term diabetic control.