Bovine Serum–Based Bilirubin Calibrators Are Inappropriate for Some Diazo Methods

Abstract

In 2003, the College of American Pathologists (CAP)1 included in the Neonatal Bilirubin Surveys a human serum–based sample enriched solely with unconjugated bilirubin (UBIL). This sample, shipped frozen in a cold pack, resembled closely a clinical sample from a healthy neonate and was commutable, as expected. The mean of all results submitted to CAP (CAP All Data) for bilirubin (199.4 mg/L) was close to that of the reference method (194.4 mg/L)(1). In subsequent years, the bias has increased(2). Most recently (2009, neonatal bilirubin NB-01), the difference between the mean value for CAP All Data and the reference method value was 9 mg/L, and the differences for the Olympus, Ortho Vitros, and Siemens ADVIA assays were closer to 25 mg/L. Because ditaurobilirubin (DTB), which is present in most commercial calibrators, and bovine serum, which is used instead of human serum for preparing calibrators, have been shown to interfere with the measurement of bilirubin(3), we investigated the effects of DTB and bovine serum on the measurement of total bilirubin by diazo methods. UBIL and DTB were purchased from Lee Biosolutions. Human serum pools (nos. 1 and 2) were obtained from volunteers from the laboratory personnel of the Children’s Hospital of Wisconsin. Bovine sera and donor calf serum were purchased from Sigma–Aldrich and MP Biomedicals. Fresh bovine serum was obtained from a local abattoir in Milwaukee, Wisconsin. Stock solutions of UBIL and DTB were prepared in all of the above-described protein matrices. The concentrations of UBIL (210 mg/L) …