Page 82 - 16Neonatal Jaundice_compressed

4 Early prediction of serious

hyperbilirubinaemia

Introduction
This chapter builds on the work that has been done in recognition and risk factor assessment for
neonatal hyperbilirubinaemia. A tool or test that could be used to distinguish babies who were
likely to develop significant, potentially serious hyperbilirubinaemia from those whose jaundice
would only ever be mild would be extremely valuable in clinical practice, particularly in the
modern era of very short hospital stays. Tests that have been reported as potentially useful in
this area in the past include markers of haemolysis and early timed bilirubin measurements. The
evidence has been systematically evaluated.

Clinical questions
What is the accuracy of the following tests in predicting neonatal hyperbilirubinaemia?
i) umbilical cord blood bilirubin levels
ii) timed serum bilirubin levels
iii) transcutaneous bilirubin levels
iv) end-tidal CO levels nomograms
v) risk assessment
vi) Coombs’ test
What is the effectiveness (clinical and cost) of various tests in predicting hyperbilirubinaemia
and preventing morbidity/mortality?

Since the tests routinely used for recognising/detecting jaundice have also been studied to
predict hyperbilirubinaemia at a later age, it was decided to conduct a combined systematic
literature search to answer two questions:
1. the diagnostic accuracy of tests in recognising jaundice
2. the prediction of hyperbilirubinaemia at a later age.

Primary screening of 2840 titles and abstracts from the database led to the retrieval of 148
papers.
Altogether, 22 studies 9;12-14;26;26-42 were selected for inclusion in this prediction chapter. Four
studies each were included for evaluating the predictive accuracy of umbilical cord blood
bilirubin levels 26;29-31 and serum bilirubin levels measured within the first 24 hours of age, 13;26-28
respectively. End-tidal carbon monoxide (CO) levels were assessed in two studies 32;33 with
different population characteristics and threshold values. Eight studies were grouped together
under ‘Pre-discharge risk assessment’ since they all evaluated different strategies (serum
bilirubin, transcutaneous bilirubin or clinical risk factors) during the pre-discharge period, to
predict subsequent hyperbilirubinaemia. DAT tests were assessed in five studies. 38-42
Furthermore, two or more strategies were compared in three of these studies.

Regarding the effectiveness of these tests, nine studies 43-51 have been included: four evaluated
transcutaneous bilirubin measurement, 43-46 two 47;48 evaluated pre-discharge bilirubin estimation
and three 49-51 evaluated DAT. No studies were identified that evaluated the effectiveness of
clinical assessment, risk index scoring, umbilical cord blood bilirubin or corrected end-tidal
carbon monoxide (ETCOc) measurement.

53