Comparability of different methods of glycated hemoglobin measurement for samples of patients with variant and non-variant hemoglobin

BACKGROUNDGlycated hemoglobin (HbA1c) is measured to monitor patients with diabetes. However, the measurement results can vary according to the analysis method and presence of variant hemoglobin. Thus, we compared HbA1c results between liquid chromatography-tandem mass spectrometry (LC-MS/MS) as the reference method and matrix assisted laser desorption/ionization time of flight mass spectrometry (MALDI-TOF MS). METHODS%HbA1c were measured using the two methods in 45 non-variant and 73 heterozygous variant samples. Precision was calculated; the results were compared using Passing-Bablok regression and the concordance correlation coefficient (CCC). The average bias between methods was compared with the lowest bias of 2.3% for biological variation. RESULTSThe precision of the two methods was < 2%. The R2 for the non-variant samples were 0.986 and the CCC was 0.99. Based on α- and β-chain, the variant samples were divided into four groups: α-chain, α-chain negligible, β-chain, and β-chain negligible variants. The R2 between the two methods of the four groups were ＞0.95; However, the average biases of α-chain and β-chain variants were above the minimum bias. CONCLUSIONLC-MS/MS and MALDI-TOF MS had good comparability in the measurement of HbA1c in non-variant samples, but the existence of variant hemoglobin caused discrepancies.