Evaluation of the Effect of Hemolysis on Quantitative Chemiluminescent Immunoassay Results for 10 Analytes

Background: The aim is to evaluate the effect of hemolysis on the quantitative chemiluminescent immunoassay results of 10 analytes and to provide a basis for formulating specific sample rejection criteria and reviewing report results. Methods: Hemolysis based on the clinical hemolysis index, hemolysis 1+, 2+, and 3+ samples and matched normal samples were collected. The quantitative chemiluminescent immunoassay results of 10 analytes from the two samples (hemolysis and normal) were determined and differences between the results obtained from samples with different degrees of hemolysis and those obtained from normal samples were evaluated. Results: A total of 34 pairs of samples were collected, including 10 pairs of 1+ hemolysis samples, 10 pairs of 2+ hemolysis samples, and 14 pairs of 3+ hemolysis samples. The quantitative chemiluminescence immunoassay detection results for the 10 analytes showed that regardless of the degree of hemolysis, the differences in alpha fetoprotein (AFP), carcinoembryonic antigen (CEA), carbohydrate antigen (CA19-9), luteinizing hormone ( LH), follicle-stimulating hormone (FSH), and ferritin (FER) between the hemolysis and normal samples were all lower than the total allowable error (TEa) based on biological variation; there were no statistically significant differences between the samples. However, the results for insulin (INS) began to decrease significantly at a hemolytic index of 1+, folic acid (FOL) showed an increase at a hemolytic index of 2+, and there was a significant difference at a hemolytic index of 3+. Conclusions: This research identified the analytes that are susceptible to hemolysis interference in chemiluminescent immunoassays. The influence of hemolysis on hemolytic clinical laboratory tests was closely related to the assay system used; thus, laboratories should evaluate the effect of hemolysis on their own analysis systems and define assay-specific hemolysis warning indices.