P34

Are hemoglobin variants eluting in “variant window” on Bio-Rad Variant Turbo 2.0 affecting the HbA1c results?

Britta Landin1,*, Lena Fredriksson1

1Clinical Chemistry, Karolinska University Laboratory, Stockholm, Sweden

Introduction: HbA1c methods are usually evaluated for interference by common hemoglobin (Hb) variants (HbS, E, C and D). Many rare Hb variants have similar chromatographic characteristics.

Objectives: We wanted to compare HbA1c results obtained with ion-exchange HPLC vs. immunoassay for all samples with heterozygous Hb variants eluting in the ”Variant Window”, i. e. where HbS, HbE and HbD elute.

Methods: Only one sample from each patient was studied. The HbA1c methods (HPLC=Bio-Rad Variant Turbo 2.0; immunoassay=Siemens DCA Vantage) were regularly compared using normal samples. DCA results were assumed to be unaffected by the Hb variants found in this study. Halfway through the study Bio-Rad changed the HPLC resin and adjusted the integration.

Results: 2-3 ‰ of all samples showed a peak in the Variant Window. 9 % of those samples contained other Hb variants than HbS, HbE or HbD. We found 37 patients with 19-21 different types of rare Hb variants (2 samples remain to be identified).  The mean ratio between individual HPLC and DCA results for each diagnosis group were compared with the corresponding ratio obtained for normal samples. The HPLC method initially overestimated HbA1c by approx. 8 % when HbE was present. During the second phase of the study this problem was eliminated and HbS, HbE and HbD all affected the HPLC results by less than ±5%. The method change also somewhat improved the integration for other Hb variants eluting in the Variant Window, although several variants displayed method discrepancies exceeding 10% both before and after the introduction of Resin D.

Image:

Conclusion: Experiences regarding interference from Hb variants gained with one type of a chromatographic HbA1c method cannot be transferred to another version of the method, even if  the method designation remains unchanged. For the time being Hb variants presenting in the Variant Window of Bio-Rad Variant Turbo 2.0 seem to be a negligible problem. However, other Hb variants - clearly separated or not - might cause significant errors in HbA1c results.

Keywords: Biochemistry, Blood, Method