AbstractBackgroundEvaluation of glycated hemoglobin determination methods in patients with clinically silent hemoglobin variants.MethodsHbA1c results were determined with various methods, including a new enzymatic assay, a boronate affinity HPLC, immunoassays and ion‐exchange HPLC in patients with the clinically silent hemoglobin variants Hb Graz, Hb Sherwood Forest, Hb D and Hb O Padova.ResultsThe effect of hemoglobin variants on glycated hemoglobin determination was method‐dependent. The enzymatic and boronate affinity HPLC method did not interfere with any of the variants evaluated. In contrast, Hb Graz interfered with all immunoassay and ion‐exchange HPLC methods evaluated. The Tosoh ion‐exchange HPLC method HLC‐723 did not detect the late migrating Hb O Padova in the chromatogram, but this hemoglobin variant still interfered causing artificially low HbA1c results.ConclusionsOur study underscores the need for clinical laboratories and physicians to be aware of the limitations of their HbA1c assay methods as well as the importance of visual inspection of ion‐exchange chromatograms to detect abnormalities caused by the hemoglobin variants. Samples with clinically silent Hb variants should be analyzed by a second method with a different assay principle, preferably a boronate affinity HPLC or an enzymatic assay. Copyright © 2004 John Wiley & Sons, Ltd.