[Diagnostic value of fasting glucose, fructosamine, and glycated haemoglobin HbA(1c) with regard to ADA 1997 and who 1998 criteria for detecting diabetes and other glucose tolerance abnormalities].Ann Acad Med Stetin. 2004; 50(1):131-8.AA
New diagnostic criteria for diabetes mellitus proposed by the American Diabetes Association in 1997 and the World Heath Organization Consultation Report in 1998 recommend lowering of the fasting plasma glucose (FPG) to 7.0 mmol/L. This change in the diagnostic FPG cut-off point was based on the results of well-documented epidemiological studies showing that increased risk of microangiopathy starts at values closer to 7.0 than 7.8 mmol/L used in the past. To facilitate the diagnosis, ADA Expert Committee recommends using FPG as the main diagnostic tool and eliminating OGTT from routine clinical practice. In contrast to ADA, WHO Consultation Group strongly recommended keeping OGTT in routine use. Due to the inconvenience, poor reproducibility, non-physiological character and labour-intensiveness of OGTT, an alternative test has been sought. The aim of this study was to determine whether fasting capillary glucose (FCG) along with fructosamine and glycated haemoglobin (HbA(1c)) perform better for the detection of glucose tolerance abnormalities than FCG alone. OGTT was performed in 1528 patients. Serum fructosamine was determined in 480 and glycated haemoglobin in 234 of these patients. To assess the value of FCG, fructosamine and glycated haemoglobin in predicting post-load glycaemia and detecting glucose tolerance abnormalities, multiple linear regression analysis and Receiver Operating Characteristics analysis were done. Fructosamine correlated stronger with 2h-postload glucose concentrations than with fasting glucose. HbA(1c) correlated stronger with FCG than with 2h-postload glucose. Combined use of fructosamine and FCG predicted 2h-postload glucose better than combined use of FCG and HbA(1c). Receiver Operating Characteristics curve analysis showed that FCG was the best criterion in discriminating diabetes. Combined use of FCG and fructosamine slightly improved the ability to discriminate glucose tolerance abnormalities from normal glucose tolerance. The following conclusions were drawn: (1) FCG is the most effective predictor of 2h-postload glucose and the best criterion for discriminating diabetes and other glucose tolerance abnormalities from normal glucose tolerance. (2) Because of the limited sensitivity and specificity of fasting glucose, fructosamine and glycated haemoglobin tests, OGTT is irreplaceable in the identification of patients with glucose tolerance abnormalities. Nevertheless, fructosamine is a potentially useful post-load glycaemia index.