Bi-variate approach to negative interference of bilirubin towards an acetaminophen assay.

It is well known that enzymatic assays for acetaminophen are positively interfered by bilirubin. The effect on acetaminophen not only depends on the concentration of bilirubin but also on that of acetaminophen. We demonstrated a negative interference instead of a positive one in a commonly used routine analyzer and investigated the recovery of acetaminophen in an enzymatic assay by a bi-variate regression. Commercially available blank serum specimens were spiked with acetaminophen and bilirubin at various concentrations, and were analyzed in the Beckman Coulter AU5822 analyzer. The specimens were run in duplicates. The results were then analyzed by least-square analysis and was built into a bi-variate quadratic model. The recovery of acetaminophen in this experiment ranged from 38.9% to 100% throughout a range of 23 μmol/L to 2052 μmol/L (for acetaminophen) and 19 μmol/L to 570 μmol/L (for bilirubin). A contour map, as well as a bi-variate equation was established, describing the relationship between acetaminophen recovery, acetaminophen concentration, and bilirubin concentration. It was shown that the degree of bilirubin interference in a commercially available acetaminophen assay is dependent on both bilirubin and acetaminophen concentrations. There was a decrease in the apparent acetaminophen concentration by an average of 30% at a bilirubin concentration of 420 μmol/L in the Beckman Coulter AU5822 analyzer. The complex relationship can be modeled by mathematical means. This allows the laboratory staff to estimate the recovery of acetaminophen when bilirubin level is concurrently measured.