HbA1c analysis by capillary electrophoresis - comparison with chromatography and an immunological method

Background: Haemoglobin A1c (HbA1c) has become an even more important analyte for clinical laboratories during recent years with the introduction of its diagnostic use for diabetes mellitus. Several different analytical principles can be used, each with their advantages and disadvantages.Aim: We wanted to compare Sebia Capillarys 2 Flex Piercing (Capillarys) with our routine HbA1c methods, which were an HPLC method (Tosoh G7) and an immunoassay (Tina-Quant on Roche Modular P) by analysing a large clinical material. Furthermore, we investigated sample stability.Methods: HbA1c analysis was performed in parallel by all three methods for more than 600 patient samples including common and some rare haemoglobin variants, as well as for several controls, some with set target values. Sample stability at room temperature and refrigerated was assessed for up to seven days.Results: Capillarys produced generally somewhat lower HbA1c values than both comparison methods, apparently due to positive bias for the comparison methods. Leaving out samples with haemoglobin variants, we found a mean bias (95% CI) for Capillarys compared to Tosoh G7 (without factorization) and Modular of -0.39 (-0.40 to -0.38) and -0.16 (-0.17 to -0.14) % HbA1c, respectively. HbA1c results were similar between instruments for samples from dialysis patients and for samples with heterozygous common haemoglobin variants, except that Tosoh G7 reported too low results in the presence of Hb E. For heterozygous Hb Raleigh, Capillarys and the immunoassay gave similar results.Conclusion: Capillarys is a convenient instrument for routine HbA1c analysis.