M. Peake et al., Bilirubin measured on a blood gas analyser: a suitable alternative for near-patient assessment of neonatal jaundice?, ANN CLIN BI, 38, 2001, pp. 533-540
The reliability of a recently released total bilirubin assay for a blood ga
s analyser was assessed in two Australian hospital laboratories. The instru
ment computes total bilirubin concentration from multi-wavelength absorbanc
e measurements of undiluted whole blood or plasma. Performance of the Radio
meter ABL 735 blood gas analyser bilirubin method (software version 3.6) wa
s compared with a proven Roche diazo method for Hitachi analysers, calibrat
ed using primary standards prepared from NIST SRM 916a bilirubin. Acceptabl
e bilirubin results were found over a wide concentration range for most neo
natal samples of whole blood or plasma. For adult specimens, bilirubin resu
lts were approximately 10% lower on the blood gas analyser. Within-run impr
ecision (whole blood) was <2.5%, between-day imprecision (synthetic control
s) <1.0%, and the bilirubin assay for both whole blood and plasma was linea
r to 1000 mu mol/L. Using sampling options from 35 muL to 195 muL, bilirubi
n results differed by less than 3%, with a 95 muL syringe option producing
the highest results. We conclude that the Radiometer ABL 735 bilirubin assa
y is suitable for near-patient assessment of neonatal jaundice using whole
blood, thus eliminating the need for sample centrifugation. Verification us
ing laboratory methods can be used when required. A positive correction of
approximately 10% is required for adult specimens to conform with Hitachi r
esults (SRM 916a calibration), possibly due to the optical characteristics
of the higher proportion of conjugated bilirubin and other substances prese
nt in most adult samples.