ABSTRACT
BACKGROUND: In the Bio-Rad D-100TM (Bio-Rad, Hercules, CA) HPLC system for hemoglobin A1c (HbA1c) measurement, 7 peaks elute: HbA1a, HbA1b, HbF, LA1c, HbA1c, P3, and HbA0. HbA1c is calculated from the ratio of the HbA1c peak area to the total area, excluding HbF and peaks after HbA0, if present. A P3 peak >10% flags for potential interferences. METHODS: We investigated 26 samples with elevated P3 peaks to determine the presence of hemoglobin variants, the effect of prolonged specimen storage in the P3 peak. The relationship between the P3 peak and the HbA1c concentration were also investigated. RESULTS: No hemoglobin variants were identified when the P3 peak was <14% (n = 14). Hemoglobin variants were detected in 7 of 12 with a P3 peak between 17.0% and 28.2%. Sample storage at room temperature had minimum impact on the P3 peak area (n = 20); the average P3 bias was -0.5 (-8.1% bias) after 3 days and 0.6 (12.2% bias) after 5 days. P3 increased with increasing HbA1c concentrations in samples with P3 < 10%. Most samples with P3 above 10 and up to 14% had marked HbA1c elevations. CONCLUSIONS: Minor elevations of the P3 peak were due only in part to hemoglobin variants, particularly in samples with P3 above 17% (below 28.2%). These elevations caused a decrease in HbA1c, whether hemoglobin variants are detected or not. Prolonged storage at room temperature did not cause P3 peaks to increase above 10%.
