Effects of hemoglobin S variants on the measurement of glycosylated hemoglobin A1c by four analytical methods

Glycosylated hemoglobin (HbA1c) identifies the average plasma glucose concentration over 120 days. HbA1c test measurements can lead to false outcomes in patients with inherited hemoglobin variants as hemoglobin S, which is prevalent in the Middle Eastern population. Our aim was to evaluate four different methods for measuring the blood HbA1c levels in diabetic patients with sickle cell anemia (SCA) and sickle cell trait (SCT). Blood samples were collected in EDTA tubes from 38 diabetic patients with either SCA or SCT. HbA1c levels were measured on all samples at three different local hospitals using three different analyzers: the Cobas Integra700 (Roche, SZ), the Dimension RXL (Siemens, USA), and the DCA2000 (Siemens, USA). All methods were compared to the standard HPLC method Variant (Bio-Rad, USA). There was a good correlation (R ²) between HbA1c levels measured by the HPLC Variant II and immunoassay analyzers and between the immunoassay analyzers themselves in SCT, but not in SCA, patients. However, an excellent correlation was observed in the SCA group between the Dimension RXL and Cobas Integra analyzers (R ² = 0.9918). Significant differences were observed in all the three immunoassay methods compared to the Variant HPLC (P <0.00001) and between the DCA2000+ and the Dimension RXL (P = 0.0144), as well as between the DCA2000+ and the Cobas Integra (P = 0.0247). We concluded that while HbA1c levels were undetected by some methods, other methods were able to measure HgbA1c with a high degree of accuracy. Therefore, we recommend that any affected HbA1c samples should be measured by two different methods.