Abstract
Since the hemoglobin A1C (HbA1C) measurement is to standardized internationally using references proposed by the International Federation of Clinical Chemistry and Laboratory Medicine (IFCC), we determined reference interval of HbA1C (IFCC). At the same time, we also determined reference interval of glycoalbumin (GA). The reference population (n=699 (men: 363, women: 336), age distribution: 23-91 years old) consisted of participants of the Funagata Diabetes Study and 3 other health checkup programs in Japan. All participants were subjected to a 75 g oral glucose tolerance test. HbA1C (IFCC) levels were measured with two types of routine HPLC and two types of routine immunoassay calibrated with JDS Lot 2 (assigned IFCC). Since no significant variation was found among the four measurements, the mean of these measurements was used as the representative value. GA levels in subjects were measured with an enzyme assay. Reference individuals for HbA1C (IFCC) and GA reference intervals were primarily selected based on National Committee of Clinical Laboratory Standards (NCCLS) recommendation. Multiple regression analysis showed that neither HbA1C (IFCC) or GA was associated with age, gender, questionnaire items suspected of influencing test results such as “Do you think you are healthy?,” physical examination results, or routine laboratory tests, e.g., hemoglobin, albumin, and creatinine. Reference individuals were selected from the reference population, with normal glucose tolerance, i.e., fasting plasma glucose<110 mg/dl and plasma glucose<140 mg/dl 2 hours after glucose loading. Reference individuals were further narrowed down using latent abnormal value exclusion method. Finally, using power transformation based on the parametric method, HbA1C (IFCC) and GA reference intervals were determined to be from 3.0 to 4.1%, and from 12.3 to 16.9%.
