A new Murdoch Children's Research Institute study has highlighted the challenges of consistently diagnosing child blood samples and sourcing healthy samples for research. Published in the Clinical Chemistry journal, the research paper, 'Pediatric reference intervals across 5 analyzers', used the blood samples of 616 healthy children and found that results from different laboratories were not always uniform.

MCRI's co-group leader of haematology research and the University of Melbourne Associate Professor Vera Ignjatovic says laboratories test whether levels of blood biomarkers (like sodium and potassium) are within a 'normal' range but defining 'normal' can be difficult, especially when it comes to children.

"To define a normal range you need to test the blood of a large group of people," says A/Prof Ignjatovic. 

"Getting access to the blood of healthy people can be problematic, especially when it comes to children and newborns. While parents are generally not reluctant to get their children to donate blood for research, children can only donate small amounts of blood, a factor that can create difficulties."

 A/Prof Ignjatovic said that variations in results were either due to differences in how the blood was tested or because specific blood reference values for children are needed.

"Better reference intervals (which define the 'normal' range) could help improve clinical decisions in children around the world. Clinical laboratory test results are critical to evidence-based medicine, with nearly 80 per cent of physicians' decisions based on information provided by laboratory reports." 

The study examined the results of tests for 30 blood biomarkers produced by five different automated blood-analyzing machines.

"Results from the five analysers were mostly within the 'allowable error' margin, however, there were some variation," A/Prof Ignjatovic said.

"For creatinine, a waste product produced by muscles, levels varied by 11.9 per cent between machines. Whether patient samples tested at different laboratories, using different analyzing machines, can be directly compared is of global importance.

A/Prof Ignjatovic says that the lack of 'regulatory incentives' for manufacturers of lab machines may be contributing to the lack of data on paediatric-specific reference intervals and that changes to regulations could facilitate improvements in the availability of age-specific reference ranges.

"This would help improve clinical result interpretation and decision-making," she says. 

Lead author of the paper, MCRI researcher, Monsurul Hoq, says the need for consistent blood diagnosis is even more crucial when a child's care is shared by a specialist, central hospital and local health care providers.

For the study, blood was taken from 616 child participants aged a few hours old to 17 years. Of the 30 blood biomarkers assessed, results varied statistically and clinically for six of the 30.
A/Prof Ignjatovic and Mr Hoq have called for more research in the area.

Until now no study has compared the reference values obtained by testing samples from the same individuals on five analyser machines using blood samples from healthy newborns and children.

The child participants are part of the Harmonising Age Pathology Parameters in Kids (HAPPI Kids) study – a prospective, cross-sectional study of paediatric blood samples for commonly requested biomarkers.

Available for interview

  • Associate Professor Vera Ignjatovic, Co-lead of Haematology Research

Media contact for Murdoch Children's


This study was primarily funded by The Royal Children's Hospital Foundation, with supplementary research funding from Ortho Clinical Diagnostics. In addition, the study received an in-kind reagent supply from Medical and Scientific Affairs at Roche Diagnostics International. Funding bodies did not have any input in the design of the study, sample collection, analysis and interpretation of data or in preparation of the manuscript.