Serum fructosamine in patients with diabetes mellitus

Serum fructosamine was compared with other measures of blood glucose control in 11 non-diabetic volunteers, 14 type 1 and 14 type 2 diabetic patients. Estimates of mean plasma glucose concentrations for the 28 diabetic patients were made by nine physicians, based on their interpretation of historical data, home capillary blood glucose profiles, fasting and random plasma glucose and plasma lipid levels. Significant differences between estimated and measured mean glucose levels were apparent with a tendency for physicians to underestimate mean blood glucose in the hyperglycaemic range (glucose greater than 11 mmol/l). Fructosamine results on the same patients correlated linearly both with mean plasma glucose concentrations (r = 0.86, p less than 0.001) and with glycosylated haemoglobin (HbA1c) levels (r = 0.93, p less than 0.001) and correctly classified diabetes control in most patients. Despite marked fluctuations of plasma glucose concentration, serum fructosamine levels measured at different times of the day did not alter significantly. We conclude that a random serum sample analysed for fructosamine provides a simple and reliable means to measure the efficacy of therapy and often provides information superior to clinical assessment of diabetic control.