[A comparison of 2 new rapid methods for determination of HbA1C concentration in patients with diabetes mellitus].

We have evaluated a new immunoturbidimetric assay (DCA 2000 HbA1c system, Bayer, Denmark) for determination of HbA1c. The aim of the study was to evaluate accuracy, precision and feasibility for the DCA 2000 method when employed in a diabetes centre by a technical assistant and at a general practitioner's by non lab staff. The results were compared with a high performance liquid chromatographic method (HPLC, AUTO A1C, Kyoto Daiichi Kagaku Co., Kyoto, Japan) which is the current laboratory method, and therefore used as reference. Assay time for the DCA 2000 method was nine minutes, while the HbA1c result was displayed within four minutes by HPLC. Blood samples were drawn after informed consent from 118 patients during a period of two months at the out-patient clinic of the Dept. of Paediatrics, Glostrup Hospital (n = 67) and at a general practitioner's (n = 51). Each sample was analyzed twice by each method on two consecutive days. In the HbA1c range from four to 14% (n = 67) the average within-assay precision (SD) for the HPLC method was 0.13%, whilst it was 0.23% for the DCA 2000 method (p < 0.001). The within-assay precision was low and acceptable, and for both methods it was independent of the current HbA1c concentration. For the DCA method precision was almost similar (p > 0.07) when carried out by a technical assistant (SD: 0.20%) and by non lab staff (SD: 0.25%). Interbatch variations for HbA1c results investigated with two different batches of reagents within a month were SD 0.30% (HbA1c range: 4.9-5.9%, n = 30) and SD 0.44% (HbA1c range: 10.5-12.1%, n = 30) for these two preparations.(ABSTRACT TRUNCATED AT 250 WORDS)