BACKGROUND: HbA1c measurement is currently routinely used to predict long term outcome of diabetes, thus playing a fundamental role in the management of diabetes. The relationship between HbA1c value and long term diabetic complications has been established by a randomised control Diabetes Control and Complications Trial (DCCT) which used high performance liquid chromatography (HPLC) as a reference method for HbA1c assay. To ensure that HbA1c results from a variety HbA1c assay methods are similar to the DCCT values, the American Diabetes Association (ADA) recommended that all laboratories should use methods certified by the National Glycohemoglobin Standardization Programme (NGSP) with interassay coefficient variation (CV) of < 5% (ideally < 3%). The International Federation of Clinical Chemistry (IFCC) working group on HbA1c standardisation has set a CV < 2.5% as a criteria for its reference laboratories. OBJECTIVES: To evaluate the performance of Arkray Adams HA-8160 HbA1c analyser which uses a cation exchange HPLC method and its correlation to HbA1c assay on Cobas Integra 800 which is an immunoturbidimetric method. METHODS: For the imprecision study, patient samples and control material of two levels were analysed on HA-8160 analyser 20 times in a single run (within-run imprecision) and twice a day on five consecutive days (between-run imprecision). For the recovery study, two samples each with high and low values were selected and mixed in ratios of 1:3, 1:1 and 3:1, and were analysed by HA-8160. Sixty samples were analysed by both Cobas Integra 800 and HA-8160 for method comparison study. Ten uraemic samples and ten thalassaemic samples were assayed on Cobas Integra 800 and HA 8160 for interference study. RESULTS: Within-run CVs were 0.6% and 0.7% for medium and high value samples respectively, 0.6% and 0.7% for low and high level controls respectively. Between-run CVs were 0.5% and 0.4% for medium and high value samples respectively, 0.5% and 0.6% for low and high level controls respectively. The mean recovery was 100.1%. A good correlation between the 2 methods (Adams = 1.00 Cobas - 0.11, r = 0.98) was observed. CONCLUSIONS: The Akray Adams HA-8160 HbA1c analyser performed within the target CV of < 2.5% and showed a good correlation with the Cobas Integra 800.
Glycosylated hemoglobin A ; Sjogren's syndrome B antibody ; Adams ; Performance ; cyclophosphamide/etoposide