Optimization of hemolysis index thresholds for biochemical and immunoassays.
Dupuy Anne Marie AM, Tabone Delphine D, Coulon Caroline C, Morena Marion M et al.
Our study aimed at establishing analyte-specific hemolysis-index thresholds on 18 chemistry assays to avoid re-sampling on Atellica© Solution (Siemens Healthcare Diagnostics, Los Angeles, CA, USA). We focused on the analytes which showed the highest degree of hemolysis interference at one or two baseline concentrations for the following 18 analytes: ALT (alanine aminotransferase), ammonium, AST (aspartate aminotransferase), CK (creatine kinase), creatinine, folates, FSH (follicle stimulating hormone), GGT (γ-glutamyltransferase), Hp (haptoglobin), iron, insulin, LD (lactate dehydrogenase), lipase, potassium, urea, uric acid, vitamins B12, and D. Interference was quantified as a percentage difference from the Hb-free baseline sample. Each calculated bias was compared to the four cutoffs: 10%Δ, analytical change limit, total change limit, or reference change limit. Analytes that did not exceed the predefined allowable-bias criterion for hemolysis included creatinine, FSH, GGT, iron, urea, uric acid, and vitamin B12. Among the affected analytes, the largest positive and negative biases were observed for AST, folates, LD, and potassium. The principal quantitative optimized thresholds for the hemolysis-sensitive analytes were AST, folates, iron, potassium, and LD tests. Our study allowed us to supplement the data provided by the manufacturer and to optimize the thresholds for interference with hemolysis management of hemolyzed samples.