112 - Quantification of bilirubin in whole blood with benchtop point-of-care instruments using external quality assessment samples

Saturday, April 29, 2023

3:30 PM – 6:00 PM ET

Poster Number: 112
Publication Number: 112.238

Lauren EH. Westenberg, Erasmus MC, Rotterdam, Zuid-Holland, Netherlands; Miranda van Berkel, Radboud umc, Nijmegen, Gelderland, Netherlands; Andrei Tintu, Erasmus MC, Rotterdam, Zuid-Holland, Netherlands; Jolande Y.. Vis, Erasmus MC, ROTTERDAM, Zuid-Holland, Netherlands; Eric Steegers, Erasmus MC, Rotterdam, Zuid-Holland, Netherlands; Jasper V. Been, Erasmus MC Sophia Children's Hospital, Rotterdam, Zuid-Holland, Netherlands; Christian V. Hulzebos, Beatrix Children's Hospital, University Medical Center Groningen, Groningen, Groningen, Netherlands

Presenting Author(s)

Lauren EH Westenberg, MD (she/her/hers)

PhD-Candidate
Erasmus MC
Rotterdam, Zuid-Holland, Netherlands

Background: Accurate measurement of total serum bilirubin (TSB) in newborns is essential for the diagnosis and treatment of neonatal hyperbilirubinemia. Variation in terms of bias and imprecision in TSB measurements affect clinical decision making. Harmonization of TSB results is endorsed by external quality assessment (EQA) programs. Benchtop point-of-care (POC) instruments using whole blood samples are increasingly being used for bilirubin quantification, but data on analytical performance of POC devices in EQA programs are currently not available.

Objective: We aim to evaluate analytical performance of four commonly used point-of-care instruments for quantifying bilirubin in whole blood.

Design/Methods: We set up an EQA program with participation of twenty-three Dutch hospital laboratories evaluating four commonly used POC methods (Radiometer ABL 90, Siemens Rapid Point 500, IL GEM Premier 4000 and 5000) for measurement of whole blood bilirubin. The EQA consisted of two samples with a bilirubin concentration between 100 μmol/L - 500 μmol/L (5.8 mg/dL - 29.2 mg/dL, resp.). The EQA samples were prepared by a reference laboratory: heparinized adult blood was spiked with different concentrations of synthetic bilirubin (B4126, Sigma-Aldrich, St.Louis, USA). The EQA samples were sent three-monthly to participating hospitals, i.e., six times, from January 2021 until June 2022.

Results:

The data contained 202 bilirubin measurements on four POC devices: 152 using Radiometer ABL 90; 32 using Siemens Rapid Point; 8 using IL GEM Premier 4000 and 10 using IL GEM 5000, respectively. The POC bilirubin concentrations were compared to the target bilirubin concentrations (Figure 1). Mean POC bilirubin concentrations were higher than the target bilirubin concentration (+55% at 100 μmol/L; +5% at 200 μmol/L; +6% at 300 μmol/L and +24% at 400μmol/L) or lower (-3% at 500 μmol/L). Coefficients of variation (CVs) varied significantly between the four POC methods. Mean CVs of POC bilirubin measurements were similar at all target bilirubin concentrations: 5.5% at 100 μmol/L, 6.6% at 200 μmol/L, 5.4% at 300 μmol/L, 4.7% at 400μmol/L, and 6.9% at 500μmol/L, respectively.

Conclusion(s): This EQA pilot study demonstrates considerable bias and imprecision for measurements of whole blood bilirubin on benchtop POC devices. Further research is needed to study (pre)analytical factors, potentially involved herein