Abstract
We studied the screening method for glucose tolerance without glucose drinking and with the use of a combination of fasting glucose (FPG) and hemoglobin A1c (HbA1c), which have recently been analyzed as routine tests in Japan. FPG analysis was limited to the enzymatic method and HbA1c was analyzed by four methods with two different HbA1c analyzers based on the principle of high performance liquid chromatography (HPLC). Although the accuracy and precision of glucose determination were guaranteed by the Expert Committee on Quality Control in JAHTS (Chairman: H. Kiyose), the interlaboratory differences in HbA1c determination were remarkable. HbA1c levels were then recalculated by the SDI method using each average and standard deviation (X, SD). These Xs and SDs were calculated from the HbA1c levels of examinees judged as normal by the 75 g oral glucose tolerance test (O-GTT) based on the recommendation of the Japan Diabetes Association (JDA). We specified the screening regions as follows: 1) FPG≥140 mg/dl, 2) HbA1c≥X+2 SD, 3) FPG≥110 mg/dl-<140 mg/dl and HbA1c≥X+1 SD. These conditions had to be satisfied. By using these criteria, the screening ratio was calculated as 17.1%, sensitivity 38.2%, specificity 42.2%, false positive 2.6% and false negative 25.5%. However, as the screening ratio was obtained from selected subjects, this ratio had to be corrected. Finally, the screening ratio was recalculated as 12.7% by the population of 75g O-GTT in 11, 569 randomized subjects.
