A method for determination of serum glycated albumin by high-performance liquid chromatography is presented. The system involves anion exchange chromatography to separate albumin and consecutive boronate affinity chromatography to separate glycated and nonglycated albumin. The method is rapid (20 min), precise (coefficient of variation, 0.7-4.9%), requires only a small sample (5 microliters), and can be automated. Assay of glycated albumin by this method is not influenced by the protein concentration of the sample or the presence of glucose. The variation in glycated albumin values in consecutive samples obtained within a day from diabetic patients (coefficient of variation, 2.02 +/- 0.65%) was significantly smaller (p less than 0.001) than that of values for fructosamine (coefficient of variation, 4.33 +/- 2.0%). The values of glycated albumin in normal subjects (20.2 +/- 1.6%) were clearly less than those in diabetic patients [39.6 +/- 5.4% in 40 Type 1 (insulin-dependent) and 39.4 +/- 5.9% in 25 Type 2 (non-insulin-dependent) patients]. The serum glycated albumin level was well correlated with HbA1c in 65 diabetic patients (r = 0.60). Because the life span of albumin in the circulation is short, measurement of glycated albumin should be useful as a short-term index of glycaemic control.

No abstract

No abstract

SUMMARY.Age-related changes in the blood concentrations of glycated proteins during childhood and adolescence were examined by measuring the blood fructosamine (FA), glycated albumin (GA), glucose levels and total serum protein levels in 611 healthy children (325 boys and 286 girls) aged 0-18 years old. These levels were compared with those of 130 healthy adults. GA was measured by a newly developed, highly precise method of high-performance liquid chromatography (inter-assay coefficient of variation, 4,9%). The FA values of all groups of up to 15 years old (grouped in two year age brackets) were significantly lower than those of adults. The mean serum concentrations ofGA in infants up to four years old were also significantly lower than in adults. A slight but statistically significant positive correlation was found between age and values of FA (r = O' 36, P < 0,001), but not of GA (r = 0'04). Levels of total serum proteins showed a positive correlation with those of FA (r=0-47, P

Additional key phrases: diabetes mellitus; haemoglobin Ale; glycated protein; fructosamineMany studies have been shown that fetal malformation and abnormal growth are significantly increased in pregnancies with poorly-controlled diabetes mellitus and that intensive glycaemic control is required in diabetic pregnancy. 1 Various indices including haemoglobin Ale, glycated protein and fructosamine are reported to be useful for assessing glycaemic control during pregnancy.t-' Glycated protein and fructosamine reflect glycaemic control levels over a shorter period, and so presumably more rapidly, than HbA1c" Measurement of glycated albumin seems more suitable than that of glycated protein because metabolism of serum albumin is better understood than that of other serum proteins. There are many reports of the usefulness of glycated albumin for assessing glycaemic control levels in non-pregnant diabetic patients. However, as there seem to be no reports on its value during pregnancy, we examined the changes of glycated albumin during pregnancy in comparison with those of other measurements. SUBJECTS AND METHODThe subjects studied were 249 pregnant women with normal glucose tolerance before pregnancy. Their fasting plasma glucose, serum total protein (T'P), glycated albumin, fructosarnine, HbA le and urinary glucose concentrations were examined several times between week 10 and 37 of pregnancy. The women were classified into three groups: a normal group, 225 women showed neither glycoCorrespondence: Dr Kenji Shima.198 suria nor hyperglycaemia (fasting glucosẽ 6'1 mmollL) throughout pregnancy; glycosuria group, 13 women who showed glycosuria at least once but not hyperglycaemia; and a hyperglycaernia group, 11 women who showed a fasting glucose concentration of > 6· 1 mmollL at least once. Glycated albumin was measured by column affinity chromatography by the HPLC method described previously. 4 Fructosamine was measured by the nitroblue tetrazolium method with an assay kit from Hoffmann-La Roche Co. (Basle, Switzerland). HbA\c was measured by an HPLC method, and plasma glucose by the international standard method with glucose dehydrogenase. Serum total protein was measured by the biuret method. The coefficients of variation of these assays for glycated albumin, fructosamine, HbA le, plasma glucose and total protein were 3'6,1-7,2-6,2'3 and 0-6l1Jo, respectively. The mean glycated albumin in 153 non-pregnant women aged 18 to 30 (mean 20' 3, SD 2· 5 years) with normal glucose tolerance was 15'6, SD 0·6%.For analysis, data were grouped into three gestational periods (weeks 10-19, 20-29, and 30-37 of gestation) named early, middle and late pregnancy, respectively. Data are given as means (SD). Statistical significance was analysed by ttest for paired and unpaired data or by analysis of variance. RESULTSThe results are summarized in Table 1. In normal pregnant women, fasting plasma glucose and glycated albumin did not change significantly as pregnancy progressed, total protein and fructosamine concentrations were significant...