Reference measurement procedure for total bilirubin in serum re-evaluated and measurement uncertainty determined.

Background: For the determination of total bilirubin in serum the candidate reference method developed by Doumas et al. has international recognition. The primary standard SRM 916a (NIST) was recommended for use as the primary reference material for calibration. Nowadays, no primary standard is anymore commercially available. Further, a description of uncertainty components was missing.

Methods: Two reference laboratories have re-investigated the candidate reference measurement procedure. Beside minor modifications, mainly the use of a molar absorption coefficient instead of calibration by use of bilirubin standard solutions has facilitated the operating, and improved the analytical performance. All relevant sources of measurement uncertainty were investigated.

Results: A measurement range of 5-525 μmol/L and a CV of 0.5% to 1.4% (long term imprecision) were determined. Excellent agreement was obtained comparing to Doumas procedure (r = 0.9999) and during a two laboratory comparison participating at IFCC RELA ring trials (mean deviation: 0.6%). The combined expanded measurement uncertainty (probability 95%) for bilirubin concentrations >30 μmol/L was estimated as 2.2%.

Conclusion: A reference system for total bilirubin based on the described reference procedure shall enable metrological traceability and optimized standardization of the values obtained in clinical routine laboratories.