Mathew P Estey1, Karina Rodriguez-Capote2, Tolulope Adelowokan3, Trefor Higgins4. 1. DynaLIFE(Dx), Edmonton, Canada; Department of Laboratory Medicine and Pathology, University of Alberta, Canada. Electronic address: mathew.estey@dynalifedx.com. 2. DynaLIFE(Dx), Edmonton, Canada; Department of Laboratory Medicine and Pathology, University of Alberta, Canada. 3. Synergy Medical Clinic, Sherwood Park, Canada. 4. DynaLIFE(Dx), Edmonton, Canada.
Abstract
OBJECTIVES: To investigate the underlying cause of unexpectedly low HbA1c results (3.3-3.5%) obtained by HPLC in three siblings undergoing routine screening for type 2 diabetes mellitus. DESIGN AND METHODS: HbA1c was measured using an alternate method based on a different analytical principle (the Siemens DCA 2000+ immunoassay). Hemoglobin fractionation was performed by HPLC on the BioRad Variant II, gel electrophoresis at acid and alkaline pH on the Sebia Hydrasys 2, and capillary electrophoresis on the Sebia Capillarys 2. Sequencing of the beta globin gene was also conducted. RESULTS: HbA1c analysis by immunoassay gave significantly higher results, ranging from 5.2-5.5%. Hemoglobin fractionation by HPLC showed an abnormal peak comprising approximately 43% of total hemoglobin, suggesting the presence of a beta chain hemoglobin variant. Gel electrophoresis at alkaline pH revealed a very unusual pattern, with 3 abnormal bands migrating with Hb F, between Hb F and Hb S, and slightly cathodal to Hb S. A single band in the Hb A position was seen on gel electrophoresis at acid pH. Capillary electrophoresis revealed two abnormal peaks, comprising 42% and 5% of total hemoglobin. Sequencing of the beta globin gene showed heterozygosity for Hb Hirose (beta 37(C3) Trp>Ser), an extremely rare variant with a substitution at the α1β2 interface. CONCLUSIONS: We describe the chromatographic and electrophoretic properties of Hb Hirose, and demonstrate that this rare variant causes falsely low HbA1c results on the BioRad variant II Turbo 2.0. Recognition of this interference is crucial in order to prevent reporting erroneous results.
OBJECTIVES: To investigate the underlying cause of unexpectedly low HbA1c results (3.3-3.5%) obtained by HPLC in three siblings undergoing routine screening for type 2 diabetes mellitus. DESIGN AND METHODS: HbA1c was measured using an alternate method based on a different analytical principle (the Siemens DCA 2000+ immunoassay). Hemoglobin fractionation was performed by HPLC on the BioRad Variant II, gel electrophoresis at acid and alkaline pH on the Sebia Hydrasys 2, and capillary electrophoresis on the Sebia Capillarys 2. Sequencing of the beta globin gene was also conducted. RESULTS: HbA1c analysis by immunoassay gave significantly higher results, ranging from 5.2-5.5%. Hemoglobin fractionation by HPLC showed an abnormal peak comprising approximately 43% of total hemoglobin, suggesting the presence of a beta chain hemoglobin variant. Gel electrophoresis at alkaline pH revealed a very unusual pattern, with 3 abnormal bands migrating with Hb F, between Hb F and Hb S, and slightly cathodal to Hb S. A single band in the Hb A position was seen on gel electrophoresis at acid pH. Capillary electrophoresis revealed two abnormal peaks, comprising 42% and 5% of total hemoglobin. Sequencing of the beta globin gene showed heterozygosity for Hb Hirose (beta 37(C3) Trp>Ser), an extremely rare variant with a substitution at the α1β2 interface. CONCLUSIONS: We describe the chromatographic and electrophoretic properties of Hb Hirose, and demonstrate that this rare variant causes falsely low HbA1c results on the BioRad variant II Turbo 2.0. Recognition of this interference is crucial in order to prevent reporting erroneous results.