Evaluation of Bio-Rad D-10® and arkray adams HA-8160® HPLC analyzers in HbA1c measurement

Glycosylated hemoglobin A1c (HbA1c) is an important parameter used for the assessment of time-dependent glycemic status and the diagnosis and follow-up of diabetes. This study aimed to evaluate the analytical performance of two HPLC analyzers, the Bio-Rad D-10®, and the Arkray Adams HA-8160®. Accuracy and imprecision studies were conducted, and a method comparison study was performed with 105 samples. Samples were collected on five consecutive days and measured on both analyzers within two hours. Bland-Altman and regression analyses were used for statistical evaluation of the data. Low and high-level within-day CV values were calculated as 1.22% and 0.60% for the Arkray analyzer and 1.2% and 0.30% for the Bio-Rad analyzer, respectively. They were calculated as 1.27% and 2.52% for the Arkray analyzer and 2.32% and 3.44% for the Bio-Rad analyzer, respectively, between days. The within-day CV values for both analyzers were below the limit of 2.5% specified by the International Federation of Clinical Chemistry (IFCC). The bias average for three months was 3.2% for the Arkray analyzer and 1.2% for the Bio-Rad analyzer. The Spearman correlation coefficient (r) was 0.973 (p<0.0001) between the two measurements, and the slope and intercept values calculated in the regression analyses showed the linear relationship and harmony between the two analyzers [R2 = 0.967, (y = 0.860x + 0.604)]. Both HPLC analyzers are shown to be reproducible and accurate for routine biochemistry laboratory use.