Concentration dependent impact of hemolysis on lipase result: Does it clinically matter?

The hemolysis index (HI) interference threshold for the Roche lipase assay was recently decreased in the package insert from 1000 (10 g/L hemoglobin) to 100 (1 g/L hemoglobin), but limited data was provided to support the change. A decrease in the threshold could increase re-collection rates, disrupt laboratory workflow and change result interpretation. We performed interference studies to verify the new manufacturer HI claim at clinically relevant lipase concentrations.

Five lipase concentrations in lithium heparin plasma (36-227 U/L; n = 3-5) were spiked with hemolysates of varying hemoglobin concentrations (0, 0.5, 1, 2, 3, 4, 5, 8, 10, 12 g/L). Lipase concentrations and HI were measured on the Roche Cobas c503 analyzer in triplicate and singleton, respectively, and means and standard deviations of replicates were calculated. Interference was quantified as the absolute and percent differences from the 0 g/L control, then compared against several published total allowable error (TEa) thresholds.

At 1 g/L hemoglobin, the HI was 95 ± 4 and yielded a 0.4-1.0 % difference from baseline across all lipase concentrations. Depending on the TEa criteria used, the lowest lipase concentration group (38 ± 2 U/L) either exhibited significantly (p < 0.05) elevated results starting at 3 g/L hemoglobin (HI = 289 ± 18) or no difference up to 12 g/L hemoglobin (HI = 1171 ± 22). All observed differences were within TEa limits for other lipase concentrations.

Hemolysis affected lipase results in a concentration-dependent manner with an analytically significant impact only at lipase concentrations <40 U/L. This study stresses the importance of verifying new manufacturer claims and the value of assessing clinical change impact.