Evaluation of Total Bilirubin in a National Quality Program

Accurate quantification of neonatal bilirubin remains a perennial challenge. To monitor analytical performance of total bilirubin measurements across hospitals in the Netherlands, a dedicated neonatal external quality assessment (EQA) program is in place.

We retrospectively analyzed Dutch EQA data from 2013 until 2025. Each year, the 81 participating laboratories received 24 pooled adult blood serum samples spiked with unconjugated bilirubin (range 100-400 µmol/L or 5.85-23.40 mg/dL). Bias from the target value determined by the reference measurement procedure (RMP) and the coefficient of variation (CV) were established for both unfactored and factored measurements.

In total, 25 983 bilirubin measurements were analyzed. Three analytical principles were used on 12 platforms from 6 manufacturers. Diazo-based methods exhibited the lowest median bias (+0.4% in 2024), alongside spectrophotometric methods (-3.8% in 2024). Vanadate-based methods showed a bias of +19.5% in 2024 that was reduced to +3.0% for factored measurements. Substantial differences in bias were observed among manufacturers and platforms, with some exhibiting concentration dependency. Comparable patterns were seen for CVs, which generally remained below 5%.

EQA results demonstrated temporal variability in bias across all analytical principles, manufacturers, and platforms used for neonatal bilirubin assays. Bias was found to vary with bilirubin concentration. Standardization of neonatal bilirubin assays is urgently needed to ensure more reliable and accurate measurements across different systems. Meanwhile, EQA programs are important in monitoring future assay adjustments to ensure appropriate and consistent management of neonatal hyperbilirubinemia.