Abstract Background Spectrophotometric-based assays can be susceptible to interference from sample hemolysis, icterus, and lipemia. Our standard practice is to verify the manufacturer’s claims on the limits above which these potential substances can cause interference to avoid reporting biased results. Recently, we verified the Roche colorimetric lipase test and sought to determine the impact of hemolysis interference on several different lipase concentrations. Typically, only one or two analyte concentrations are used when verifying interference limits; however, it is conceivable that certain analyte concentrations may be more prone to erroneous results in the presence of an interferent. To test this hypothesis, we prepared four sample pools with varying levels of lipase spanning the analytical measuring range and then measured their lipase concentrations at different levels of hemolysis. Methods We prepared four plasma specimen pools with average lipase values of 29, 60, 112, and 235 U/L, each having an initial hemoglobin concentration of less than 19 mg/dL. Hemolysate was freshly prepared by freezing and thawing red blood cells. The hemolysate and isotonic saline were used to spike the plasma pools to achieve hemoglobin concentrations at approximately 250, 500, 750, and 1000 mg/dL. Additionally, a sample with no hemolysate was prepared to serve as a baseline. Lipase and hemolysis were measured in triplicate for all samples and then averaged. The percent recovery was determined by dividing the lipase concentration observed in hemolyzed samples by the baseline lipase measurement. Results The plasma pool with a lipase concentration of 29 U/L exceeded our standard +/- 10% acceptability criteria for bias with an over-recovery of 114% of the original lipase concentration at a hemoglobin concentration of 287 mg/dL, the lowest hemolysis level tested. This trend continued upward, to 133% recovery of lipase concentration at a hemoglobin concentration of 1,112 mg/dL. The 60 U/L pool crossed this threshold at 569 mg/dL hemoglobin, reaching a maximum recovery of 111%. However, the pools with lipase values of 112 and 235 U/L remained within our +/-10% recovery limit at all hemoglobin levels tested reaching a maximum of 106% and 100% recovery, respectively. Conclusion When verifying interference limits, analyte concentration can impact the cutoff value used for interferents as shown by our data. At two different lipase concentrations within our institution’s reference range of 10-65 U/L, we found different cutoff values for hemoglobin concentration when using lipase concentrations of 29 U/L compared to 60 U/L. Additionally, we observed an overall trend that, for this colorimetric method, an increase in lipase concentration yielded resistance towards interference from hemolysis with only +/- 1% variance in lipase concentration at all hemolysis levels tested for the 235 U/L lipase pool. When verifying manufacturers’ interference limits, caution should be taken when choosing which analyte concentration to use. When possible, multiple concentrations should be used to determine an appropriate limit.
Ogorek et al. (Wed,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: