Further evaluation of the world health organization international reference reagent for Haemoglobin A₂ measurement

Abstract

Introduction The accurate measurement of HbA2 is essential for the detection of β‐thalassaemia carriers and as no single calibrant is used by the various manufacturers of analysers, differences are seen in results obtained. The World Health Organization International Reference Reagent for HbA2 (WHO IRR 89/666) was made available to diagnostic laboratories in the 1980s and remains the only international reference material available. A previous study (2015) demonstrated that the WHO IRR remained suitable for use as an HbA2 standard as tested by 52 participants in the UK NEQAS Haematology Abnormal Haemoglobins Programme. This study was undertaken to include simultaneous analysis of three whole blood specimens over a range of HbA2 values with the WHO IRR and to include participants from laboratories outside of the UK. Method Three whole blood specimens with HbA2 levels ranging from 2.4% to 5.7% and the WHO IRR were distributed to 56 laboratories located in 14 different countries. Participants were requested to test the specimens at defined intervals and return results accompanied by chromatograms or electropherograms produced. Results Differences found in results from different analyser groups reflect the bias found in the 2015 study in that bias is seen according to the methodology used and also varies in relation to the level of analyte being measured. Conclusion Results of measurements from whole blood specimens and the lyophilized WHO IRR standard did not show any deterioration of the IRR, and it remains suitable for use. Linearity and calibration of analysers remain a problem.

The accuracy of haemoglobin A2 measurements in the presence and absence of haemoglobin S

“…This reflects our earlier results when we found A 2 by capillary electrophoresis to have a median of 2.65% in normal subjects while HPLC had a median of 2.77% ( p < 0.001). It also reflects the results observed by others for samples with a normal A 2 including a comparison of BioRad Variant II HPLC with Sebia Capillarys 2 capillary electrophoresis [ 4 ] and a recent study using three capillary electrophoresis methods and five HPLC methods [ 1 ]. However interestingly the first of these two studies found a lower haemoglobin A 2 by capillary electrophoresis when the haemoglobin A 2 was elevated as a result of β thalassaemia heterozygosity while the latter found no difference between these two technologies.…”

“…The diagnosis of β thalassaemia heterozygosity requires that quantification of haemoglobin A 2 is both precise (reproducible) and accurate (close to ‘truth’). The difference in mean values with different technologies that have previously been reported [ 1 ] and the difference in median values of normal samples observed in this study indicate that accuracy has not yet been achieved despite attempts at standardisation.…”

“…The Variant II HPLC with Sebia Capillarys 2 capillary electrophoresis [4] and a recent study using three capillary electrophoresis methods and five HPLC methods [1]. However interestingly the first of these two studies found a lower haemoglobin A 2 by capillary electrophoresis when the haemoglobin A 2 was elevated as a result of β thalassaemia heterozygosity while the latter found no difference between these two technologies.…”

“…The quantification of haemoglobin A 2 is crucial in the diagnosis of β thalassaemia heterozygosity with small differences in the percentage being diagnostically important. However, problems have been noted with there being significant differences in measurements on different instruments, sometimes even with different instruments from the same manufacturer; in one study the mean values observed on nine instruments ranged from 2.2% to 2.8% [ 1 ]. In addition to problems with calibration and standardisation of methods, further complexity is introduced by the fact that high‐performance liquid chromatography (HPLC) separates what is termed haemoglobin A 0 from two post‐translationally modified fractions, glycated and glutathionylated haemoglobin A.…”

Erythrocyte Indices MCV and/or MCH as First Round Screening Followed by Hb-analysis for β-thalassemia Carrier State

“…(14) Furthermore, HbA2 <3.5% could have potentially been erroneously attributed to iron deficiency. (32) Low value of SLI or MI together with normal to borderline HbA2 levels might suggest an IDA or otherwise possible an α-thalassemia carrier. The possible α-thalassemia defect might be deletional or non-deletional α-globin, and therefore, need to be further analysed by molecular analysis.…”