Lipid Interference Flagging and Haemoglobin Correction in Lipemic Samples With a Hematology Analyzer

BACKGROUND: Lipemia can interfere with haemoglobin (HGB) levels, which is critical for evaluating anemia and polycythemia. This study aims to develop and evaluate the performance of a lipid flagging and haemoglobin correction system. METHODS: Lipemic samples were simulated by the addition of a fat emulsion to 79 whole-blood samples from healthy participants. Additional blood samples were collected from 152 patients with lipemia. Plasma exchange was performed to eliminate lipid interference and establish reference HGB values. A lipid interference flagging method was developed, showing the "Lipid Particles?" flag on the hematology analyzer. Lipid interference was defined as samples showing a percentage change in HGB (ΔHGB%) before and after plasma exchange exceeding 4%. The accuracy of the optical haemoglobin (HGB-O) value was assessed in the samples with confirmed lipid interference. RESULTS: A flagging method was developed based on the changes in routine blood parameters and blood cell scattergram. Among the 152 clinical lipemic samples, 27 were identified as lipid interference. While the "Lipid Particles?" flag was triggered for 28 samples, yielding a sensitivity of 92.59%, specificity of 97.60%, positive predictive value (PPV) of 89.29%, and negative predictive value (NPV) of 98.39%. Additionally, among the 27 lipid-interfered samples, the HGB-O value showed excellent correlation (Pearson's r = 0.935) and an average deviation of 0.72% (95% CI, -1.02% to 2.47%) with the reference value. CONCLUSIONS: These findings highlight the clinical value of the "Lipid Particles?" flag in identifying lipid interference, demonstrating the accuracy of HGB-O in providing a reliable haemoglobin measurement.