ABSTRACT
This study aims to compare the HbA1c test results obtained by widely used methods using samples with various lipemia levels and Hb variants, and to determine whether it is possible to correct the lipemia effect in the identical samples. Out of the laboratory information system (LIS), 48 patients with various HbA1c results were identified including patients with and without Hb variants. After the baseline measurements, all samples were spiked with intralipid solution and treated by a subsequent 0.9% saline replacement procedure. HbA1c values were measured four times sequentially with enzymatic and capillary electrophoresis (CE) methods for each sample, and the measurements were categorized as follows: Baseline; Spiked, 5g/L; Spiked, 20g/L; Post-saline replacement. Sequential HbA1c measurements using the CE method did not show a significant difference, but samples containing 20 g/L triglycerides and samples treated with 0.9% saline replacement showed a significant difference when compared to baseline measurements in both patients with and without Hb variants using the enzymatic method (p < 0.001). The correlation between the two methods was strong at baseline measurements (r = 0.977), declined with lipemia (r = 0.968 and r = 0.737 for 5 g/L and 20 g/L triglycerides, respectively), and then increased with 0.9% saline replacement (r = 0.962) in patients without Hb variants. This study revealed that the enzymatic method, but not CE was susceptible to lipemia interference both in patients with and without Hb variants. Lipemia interference could be partially eliminated with 0.9% saline replacement, but enzymatic measurements were still somewhat affected.