B-075 A solution for peak migration in capillary electrophoresis hemoglobin A1c readings in patients with hyperleukocytosis

Abstract Background Hyperleukocytosis can cause a peak shift toward the cathode in capillary electrophoresis hemoglobin A1c (HbA1c) measurements. The mechanisms of the shift are not fully understood, and HbA1c values cannot be determined here because no Hb peaks are within the expected zones. This study aims to develop a simple solution for the problem. Methods HbA1c measurements were performed using Sebia Capillarys 3 Tera instrumentation and procedures on whole blood samples with a white blood cell count of at least 100,000 cells/µL. Samples were first analyzed without any treatment to confirm the presence of cathodal shifts of HbA1c, A0, and A2. The samples were then frozen/thawed one to three times at -70 degrees Celsius as needed with several samples also diluted with water or saline followed by repeat HbA1c measurements. Results Twenty-two untreated samples showed cathodal shift and x-axis compression for all peaks present. Fifteen of these cases (68%) only required freeze/thawing to correct peak migration, whereas seven cases (32%) also required dilution with water or saline. Interestingly, some of the migrated peaks zoned to a position normal for another peak (i.e. HbA1c peak zoning to HbA0 location). With the exception of one sample, only one or two freeze/thaw cycles were used to return the peaks to normal positions. After two freeze/thaw cycles, two samples were diluted with 100 µL of deionized water, two samples were diluted with 500 µL of deionized water, two samples were diluted with 500 µL of saline, two samples underwent a 50:50 dilution with deionized water, and one sample underwent a 25:75 dilution with saline followed by an additional freeze/thaw cycle to achieve peak correction. The HbA1c results of three of 22 samples with peak changes would not be reportable due to the low optical density of the measurements. Relative change in HbA1c peak areas after correction was 0.15%±0.14% (mean± SD, n=22). There was a weak correlation between white blood cell count (>100,000 cells/µL) and change in HbA1c peak area (R2 = 0.269, P = 0.013), while no correlation was observed between cell lineage with change in HbA1c peak area or x-axis position. Conclusion Freeze/thawing with dilution is an effective method for the correction of peak shift and compression caused by hyperleukocytosis. Results are reportable clinically for 86 % of samples. The mechanisms of peak migration in patients with hyperleukocytosis and peak correction with freeze/thawing and dilution remain to be studied.